Parallel Lives Revisited: Mediterranean Guest Workers and their Families at Work and in the Neighbourhood, 1960-1980 9781785337796

Originally coined in 2001 in a report on racial tensions in the United Kingdom, the concept of “parallel lives” has beco

181 49 4MB

English Pages 220 Year 2018

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
Illustrations
Foreword
Preface
Acknowledgements
Introduction. Parallel Lives and Segregation
Chapter 1 Postwar Migration to the City of Ghent
Chapter 2 Integration Processes of Immigrants in the Local Labour Market and the Workplace
Chapter 3 Immigrant Workers’ Relations with Colleagues and Employers
Chapter 4 Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood
Chapter 5 Immigrants’ Social Relations with Neighbours
Conclusion
Quantitative Appendix
Interviews
Bibliography
Index
Recommend Papers

Parallel Lives Revisited: Mediterranean Guest Workers and their Families at Work and in the Neighbourhood, 1960-1980
 9781785337796

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Parallel Lives Revisited

Parallel Lives Revisited Mediterranean Guest Workers and their Families at Work and in the Neighbourhood, 1960–1980

Jozefien De Bock

berghahn NEW YORK • OXFORD www.berghahnbooks.com

First published in 2018 by Berghahn Books www.berghahnbooks.com © 2018 Jozefien De Bock All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data Names: Bock, Jozefien De, author. Title: Parallel lives revisited : Mediterranean guest workers and their families at work and in the neighbourhood, 1960-1980 / Jozefien De Bock. Description: New York : Berghahn Books, 2018. | Includes bibliographical references and index. Identifiers: LCCN 2017051492 (print) | LCCN 2017051787 (ebook) | ISBN 9781785337796 (eBook) | ISBN 9781785337789 (hardback : alk. paper) Subjects: LCSH: Foreign workers, Mediterranean--Belgium--Ghent--Social conditions--20th century. | Immigrants--Belgium--Ghent--Social conditions--20th century. | Belgium--Emigration and immigration--Social aspects--20th century. | Mediterranean Region--Emigration and immigration--Social aspects--20th century. | Ghent (Belgium)--Ethnic relations--History--20th century. | Ghent (Belgium)--Social conditions--20th century. | Foreign workers, Mediterranean--Interviews. Classification: LCC HD8378.5.M43 (ebook) | LCC HD8378.5.M43 .B63 2018 (print) | DDC 331.6/21822049314209046--dc23 LC record available at https://lccn.loc.gov/2017051492 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-1-78533-778-9 hardback ISBN 978-1-78533-779-6 ebook

For Thomas, who should have been here when this book came out



Contents

List of Illustrations

viii

Forewordx   Leo Lucassen Prefacexii Acknowledgementsxiii Introduction1 Chapter 1  Postwar Migration to the City of Ghent

17

Chapter 2 Integration Processes of Immigrants in the Local Labour Market and the Workplace

41

Chapter 3 Immigrant Workers’ Relations with Colleagues and Employers

72

Chapter 4 Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood

100

Chapter 5  Immigrants’ Social Relations with Neighbours

131

Conclusion162 Quantitative Appendix

176

List of Interviews

182

Bibliography187 Index198

Illustrations

Figures Figure 1.1 Slum-like housing in Ghent in the 1970s, example of a beluik.22 Figure 1.2 Faruk Köse and his friends, who were among the first Turkish immigrants to arrive in Ghent in 1965. 26 Figure 1.3 Celebrating the wedding of Italian immigrant Angelo Cocca and his Belgian wife Huguette, early 1960s. 33 Figure 2.1 Turkish woman working in the Dutch fish factory Diepvries Breskens, 1980. 45 Figure 2.2 Immigrant man at work in the textile industry, UCO Gent, 1980. 57 Figure 2.3 Turkish immigrant Faruk Köse with some of his Belgian and Turkish colleagues at UCO Ltd, 1965. 63 Figure 3.1 The football team of the UCO textile factory, where Belgian and immigrant workers played together, 1963. 82 Figure 3.2 The young men from restaurant Al Parma with owner Gianni Bombini in the middle, early 1970s. 87 Figure 3.3 Weekly gathering of Spanish immigrants in Ghent at the Hogar Español, 1960s. 91 Figure 4.1 Article by A. De Keuleneir, decrying the bad housing conditions of immigrants living in lodging houses, Vooruit, 28 February 1964. 102 Figure 4.2 Protest in front of the city hall for better housing, 1978.111 Figure 5.1 Zohra and her ‘adoptive’ grandmother Maria, early 1970s.136 Figure 5.2 Living very close together in the beluiken, 1970s. 145 Figure 5.3 Local and immigrant children play together in the Brugse Poort neighbourhood, 1970s. 149

Illustrations • ix

Tables, Charts and Maps Table 1.1 Percentage of independent immigrants taking up a first job in textiles, construction or domestic service, 1960–1980. 28 Chart 2.1 Employment episodes at UCO Rooigem, 1963–1991.52 Map 4.1 Spatial dispersion of Mediterranean immigrants in Ghent, 1965. 117 Map 4.2 Spatial dispersion of Mediterranean immigrants in Ghent, 1970. 118 Map 4.3 Spatial dispersion of Mediterranean immigrants in Ghent, 1975. 119

Foreword Leo Lucassen

This book on the settlement of Mediterranean labour migrants in the Belgian city of Ghent shows the importance of historical analyses when we want to understand the present-day position of migrants and their offspring. Parallel Lives Revisited questions in particular widely shared assumptions about ‘parallel societies’ and ethnic minorities who would ‘hunker down’. Both concepts have gained wide acceptance and foster the idea that migrants, especially with a Muslim background, prefer their own kind and deliberately isolate themselves from the mainstream. Parallel Lives Revisited puts these assumptions to the test and provides a much more nuanced and layered understanding of the growing social and ethnic segregation in Western European cities. Most importantly, it shows that the root causes of this segregation, which rapidly established itself in the 1980s and 1990s, are not primarily ethnic (and/or religious) in nature, but are largely explained by specific socio-economic and spatial developments. With the benefit of hindsight we can conclude that the timing of the immigration of Turkish and North African former ‘guest workers’ was rather unfortunate. In particular, the family reunification that gained momentum at the end of the 1970s coincided with a protracted economic recession, which pushed the former guest workers, who were concentrated in ‘dying industries’ such as textile mills, out of their jobs. This marginalization led to ethnic concentrations in cheap and derelict neighborhoods and the decrease of interethnic contacts, especially among the first generation migrants. By highlighting the importance of socio-economic factors, and their spatial expression, Parallel Lives Revisited is an important contribution to the current debate on integration in Western Europe. Although cultural factors and ethic preference do play a role in the building of social networks by migrants and their descendants, socio-economic determinants appear at least as important. This conclusion is in line with research on current patterns of educational, spatial and social mobilities among the children and grandchildren of labour migrants, as these show a clear correlation between social mobility and the nature and composition of social networks. Parallel Lives Revisited is therefore

Foreword • xi

not only a very rich case study on a particular Belgian industrial city, but also shows why such longer term analyses of the settlement process of ‘guest workers’ are crucial for general insights in the inter-generational integration process. As such it embodies a highly desirable antidote to essentialist – and sometimes outright racist – interpretations that tend to prevail in current political debates on immigration and its effects. Leo Lucassen is Director of Research of the International Institute of Social History in Amsterdam and professor of social history at the University of Leiden.

Preface

When I was twenty-two, I worked for a small NGO on a project dealing with the history of Moroccan migration to my hometown Ghent. Interviewing Moroccan men and women who had come to the city in the 1960s and 1970s, I discovered a whole new way of looking at the place where I had studied and lived for many years. I discovered a tale of hope, dreams, successes and bitter disappointments. I found out that I was not the only one who loathed the Belgian weather, or preferred couscous over potatoes. In short, I learned about the lives of people with a very different background from my own, who shared with me a love for the city of Ghent. One side of the stories these courageous men and women told me struck me enormously: the many tales of friendships with Belgian colleagues and neighbours during the first years of their stay, friendships that were real and long-lasting in spite of the many differences between them. These stories not only took me by surprise, but also the children and grandchildren of these people, who often attended the interviews. In a context where immigrants and their descendants are constantly accused of living ‘parallel lives’, withdrawing in immigrant neighbourhoods, creating their own ethnic economies, attending ‘black schools’, etc., the stories of these friendships across ethnic boundaries got lost. With this book I wanted to tell them again, whilst not romanticizing them or exaggerating their importance. Over the course of the past fifty years, things have changed and immigrants and locals have undoubtedly grown further apart. How and why this has happened are crucial questions to which I hope this book ­provides at least a partial answer.

Acknowledgements

Most of the credit for this book goes to the people who so generously told me the story of their lives. They are too many to be summed up here, but you can find a list of them at the back of the book. It is not a small thing to lay out your story for someone you do not know. It is only because of their trust and belief in my project that I have managed to bring this book to a good end. To honour these people, I have tried to let them speak for themselves throughout the text by providing extensive quotations written in a language that comes close to the spoken word. Any interpretation of what they have said is, of course, my responsibility. Also worthy of praise are my academic teachers, Professor Frank Caestecker and Professor Gerhard Haupt, who have been my guides throughout the research that lies at the heart of this book. This research has been made ­possible by a doctoral grant from the European University Institute and a postdoctoral research grant from Ghent University. Further, I am very grateful to the people of the Ghent population archive; the Amsab-Institute of Social History; the Integration Office and other city administrations; and the many libraries and archives I worked at. Many thanks go to my mum, dad and brother, who have supported me throughout, in work and life. It hasn’t been an easy journey and they have been there all the way. Thank you to the friends and family that I knew when I was in Florence, and to my new friends and family back in Ghent. To the godfather of my child, thank you for reading through the whole manuscript when I really couldn’t. Any mistakes that remain are entirely my own. Finally, to Sam, who was there when the idea for this book was born and is still there today: third time’s the charm!

Map of the Neighbourhoods of Ghent during the period under study © Samuel Standaert, based on a map from Schmit (1972) A: City centre B: Watersportbaan-Ekkergem C: Sint Pieters square D: Ghent South E: Rabot

F: Brugse Poort G: Sint Macharius H: Sluizeken-Muide I: Sint Pieters railway station

Introduction Parallel Lives and Segregation



In 1996, German sociologist Wilhelm Heitmeyer coined the term ‘Parallelgesellschaft’ in a newspaper article describing the results of an enquiry into the lifeworlds of young Turks in Germany. He used the term as a warning: if things continued the way they were, certain religious-political groups active among Turkish youngsters might go on to develop an inscrutable ‘parallel society’, separated from the majority (Heitmeyer 1996). The term lay dormant for a while, but reappeared in the aftermath of the murder of Dutch filmmaker Theo Van Gogh in 2004. Meanwhile in the United Kingdom, the Community Cohesion Review Team – set up after the riots that swept across the country in 2001 – had coined a similar concept, that of ‘parallel lives’. The team’s report expressed a concern about the extent to which ‘the physical segregation of housing estates and inner city areas … [was] compounded by so many other aspects of our daily lives’, for example in ‘separate educational arrangements, community and voluntary bodies, employment, places of worship, language, social and cultural networks’. From their observations, the team concluded ‘that many communities operate on the basis of a series of parallel lives. These lives often do not seem to touch at any point, let alone overlap and promote any meaningful interchanges’ (Cantle 2001). Since then, the terms ‘parallel lives’ and ‘parallel societies’ have come to be part of standard vocabulary. Both in political discussions and in the public debate, the idea that some ethnic minorities are in fact living separately from the majority has become a basic assumption, backed up by a certain amount of empirical data. Decades of research producing different kinds of

2  •  Parallel Lives Revisited

‘­segregation indices’ (Saltman 1991, 1–2) have shown occasionally high rates of residential concentration for specific immigrant groups in specific parts of European cities, and research mapping segregation and mixing at the level of individual social networks has shown a low occurrence of close relationships across ethnic boundaries (Leibold, Kühnel and Heitmeyer 2006). This kind of research however generally focuses on situations that are characterized by extreme levels of concentration and segregation, paying a lot less attention to more commonly occurring patterns of dispersion and mixing. Such a focus is related to the policy-orientedness of migration research, which means that it is mostly interested in what is perceived as problematic. Indeed, the whole discourse of ‘parallel lives’ is far from neutral. In general, only the  lives of those populations that are politically constructed as troublesome (such as Pakistani in the United Kingdom, Algerians in France, or Turks in Germany) tend to be framed in these terms. Groups that sometimes exhibit equally high degrees of segregation, such as expatriates in metropoles around the world, are hardly ever accused of living in ‘parallel societies’. Clearly, the issue of class is at stake here, as patterns of concentration and segregation of high-income groups are not problematized, whereas those of low-income groups are. Further, it is remarkable that the research hardly ever focuses on the majority group. The extent to which ‘indigenous populations’ allow for people with a different ethnic background to live, work and go to school amongst them and become part of their social lives is much less the object of scrutiny than the other way around (Martinovic 2013). In the political and the public discourse, this translates into the fact that the blame for ‘parallel lives’ is clearly put on the ‘others’, the immigrants and their descendants. However, research that looks at diversity in practice from different angles comes to very different conclusions than the research cited above. It shows how ‘despite the alarming talk about immigrants and minorities concentrating in ethnically diverse neighbourhoods and related worries about social cohesion … people of diverse ethnic backgrounds do get along in shared urban spaces’ (Pratsinakis et al. 2017, 103–104). The results of such empirical research looking at the social relations of people ‘on the ground’ also go against the grain of the popular theory claiming that ethnic diversity an sich should be seen as a cause for the loss of social cohesion in Western societies. Popularized by American sociologist Robert Putnam as the ‘hunkering down-thesis’ (Putnam 2007), this theory has strengthened many policy makers in their belief that social problems in diverse neighbourhoods need to be tackled first and foremost in terms of (ethnic) diversity. However, as different scholars – amongst whom Putnam himself in his 2007 paper – have indicated, the premises on which this theory is based have a number of shortcomings. One of those is that the data used by Putnam (as by many other scholars) provide a static picture of reality, and do not allow

Introduction • 3

for a dynamic interpretation of what are essentially processes over time. The importance of this is demonstrated, for example, by Gesthuizen et al., whose research refuted the effect of ethnic diversity as such, but did find proof for the impact of a change in the degree of diversity through recent immigration (Gesthuizen, van der Meer and Scheepers 2009, 1­ 31–33). Another problem with the hunkering down-thesis is that the concepts it uses – ethnic diversity and social capital – are so broad that they in fact become all-encompassing, which makes it difficult to understand  the exact meaning of the relations between them. There is a need to break them down into more specific concepts, that can inform us better about what is actually going on (Gijsberts, van der Meer and Dagevos 2012). By making a detailed reconstruction of how newcomers from different countries of origin  – some problematized, some not – found their place at work and in the neighbourhood, what kind of social capital they developed there, and how all of this changed over time, this book hopes to contribute to both the parallel lives- and the social cohesion-debate.

A Time-Related Thwarting of our Knowledge The policy-orientedness of migration-related research is not only reflected in its subject matter, but also in its history, with the bulk of research a­ ppearing after the mid-1970s. With a few exceptions,1 postwar labour migrants in Europe initially aroused little interest from the receiving states, and thus from the social scientists working there. Only with the outbreak of the ­economic crisis did these immigrants and their descendants become the centre of attention (Cottaar, Bouras and Laouikili 2009, 17–20). Therefore, not only is our knowledge of their integration patterns limited in time; it is essentially l­imited to a context of crisis and economic hardship. What we think happened during the first thirty years of postwar labour migration is more inspired by our imagination than based on actual research. For example, in his excellent work on immigrant integration in a Dutch city, Peter Reinsch states ‘When I consider the historical background of much European immigration, rooted in the demand for unskilled labourers to do menial work that indigenous Europeans were unwilling to do, an image arises of oppressed immigrants  populating factory production lines occasionally interspersed with an indigenous overseer’. However, he immediately concedes, ‘no local statistics are available from the “guest worker era” of the 1960s and 1970s that would corroborate the image of more secluded laborers in the past’ (Reinsch 2001, 197). The provenance of such ideas is certainly related to the approach taken by a number of iconic studies on postwar labour migration, such as the ­classic Immigrant Workers and Class Structure in Western Europe by Castles

4  •  Parallel Lives Revisited

and Kosack. Basing their work on a Marxist interpretation of macro-scale developments, these authors described the position of foreign workers as follows: Immigrant workers in France, Germany, Switzerland, and Britain are usually employed in occupations rejected by indigenous workers. In a situation of full employment, the nationals of the countries concerned have taken advantage of opportunities for moving into better-paying, more pleasant jobs, usually in the white collar or skilled sectors. The immigrants have been left with the jobs deserted by the others. (Castles and Kosack 1985, 112)

Although they did not explicitly address the issue of spatial or social mixing across ethnic boundaries, such descriptions did feed the idea of a high degree of segregation between immigrant and local workers. More recently, a new group of scholars have come to take an interest in the issue of postwar labour migration to Europe. Migration historians have retraced the history of this migration to its initial years, approaching the subject from different angles – including bottom-up and micro-scale ­perspectives – and thereby uncovering a more nuanced picture of this crucial period. Some studies have analysed the political and legislative frameworks encompassing this migration; others have looked at the reactions of receiving societies to the arrival of newcomers; and yet others have focused on the integration of the latter in their new environment (Oltmer, Kreienbrink and Sanz Diaz 2012). Offering such a historical perspective on the integration processes of postwar labour migrants is exactly what this book will do. By going back to the beginning and following these processes over a period of twenty years, it aims to help us better understand the trajectory postwar immigrant populations have covered since their arrival and the position they find themselves in today. As such, the book hopes to provide a much needed historical background to present-day discussions on ‘parallel lives’.

The Subject of this Book Broadly speaking, this book studies the spatial and social integration of immigrants in the receiving society. The notion of integration is stripped of its normative connotations: we do not judge integration processes in terms of success or failure, nor do we speak of ‘more’ or ‘less’ integrated immigrants. Following Lucassen (2005), we use an open and functionalist definition of integration, conceived as ‘the general sociological mechanism that describes the way in which all people find their place in society’. Further, we focus on the structural and social aspects of integration, rather than on what has

Introduction • 5

been called ‘identificational integration’, referring to issues of ascribed and self-ascribed identity (Lucassen 2005, 18–20). Some very good historical work on this subject has been done. For example, in her excellent book on the identificational integration patterns of Moroccan immigrants in the Netherlands, based on a detailed study of changing dress codes, Dutch historian Aniek Smit has shown how such integration cannot be seen as a one-way road to either more or less separateness. Whereas certain aspects of difference have faded over time and generations, others have regained importance, and all of these changes have had a different meaning to different individuals and groups (Smit 2011). Studying the identificational integration processes of Polish and Italian immigrants and their Belgian colleagues and neighbours in a miners’ town, her Belgian colleague Leen Beyers has shown how, for newcomers, the boundaries between ‘natives’ and ‘foreigners’ continuously remained strong. It was only over the course of the following generation(s) that these ‘outsiders’ could become real ‘insiders’, and this only because conditions were right (Beyers 2008). Clearly, a thorough understanding of such identificational integration processes requires a study over many decades and across generations. This does not fall within the scope of this book, which only looks at the settlement processes of first generation immigrants during the first twenty years in their new ‘home lands’. Because of this, the book does not discuss the extent to which immigrants applied for and acquired the nationality of the receiving society. In part, this is also because in the period under study naturalization rates of first generation immigrants were very low, not least because of restrictive legislation and complex, expensive procedures. Only when this was changed, from the early 1990s onwards, did naturalization become an attainable option for many and do naturalization rates tell us more about immigrants’ (identificational) integration processes (Caestecker et al. 2016). For the same reason, the book does not address the issue of interethnic family formation, even though it is at the heart of the debate on segregation and parallel lives (Caestecker 2005). We argue that at least for first generation immigrants, patterns of partnerchoice have been heavily related to the independent variables of sex-ratio of the immigrant group at the time of arrival, and of age and marital status at migration. However, we do recognize the impact of partner-choice on the further development of immigrants’ social and human capital in the receiving society, showing how mixed marriages endowed immigrants with a wider social network and more bridging social capital – the other side of a causal relationship between upward social mobility and a higher degree of mixed marriages as described in the literature (Lucassen and Laarman 2009, 55). Therefore, throughout the analysis, being married across or within ethnic boundaries is brought to the fore as an important explanatory variable. In the conclusion to the book, where we briefly discuss the integration processes of

6  •  Parallel Lives Revisited

immigrants and especially of their descendants from the 1980s onwards, we will pay more attention to the issue of partner-choice. Even though the book does not focus on the identificational aspects of integration, it does make use of the concept of ethnicity. However, rather than making this concept the subject of study, looking at its changing meaning to different individuals and groups of people, this book uses it in a more descriptive way, as a proxy for what could also be labelled national, regional, linguistic, etc. identity. Ethnicity is understood as a social variable referring to shared origin and culture, that is self-ascribed and/or ascribed by others and can be activated in order to mobilize social capital, to discriminate, to obtain economic gain, etc. Throughout the book, it comes to the fore as only one of many different social variables that have impacted upon the positions and trajectories of Mediterranean immigrants in the receiving society. Its precise meaning is analysed more thoroughly when specific phenomena, coined in terms of ‘ethnicity’ – such as ‘ethnic workplaces’ or ‘ethnic ­neighbourhoods’ – are discussed. In this book, it makes sense to use ‘ethnicity’ in such a descriptive way, even though this does imply that we lose track of the constant negotiations of ethnic boundaries between and within groups that are at the heart of this concept. The structural and social aspects of immigrants’ integration processes are studied through the lens of two spheres of integration: work and housing. Even though in practice, these spheres are strongly interconnected and overlapping, here they are separated for the sake of the analysis. Work and housing of course are not the only domains where immigrants found their place in the receiving society. Lucassen’s definition of integration covers almost every aspect of human life. The choice for work and housing relates to the fact that they are seen as particularly useful indicators of integration, as they span a large part of the daily lives of postwar labour migrants and their families. The importance of work in the integration processes of these immigrants can hardly be overstated, even if it is often treated as secondary in the public debate. As work was at the very core of their migration project,  its nature and context need to be more closely examined in order to allow us to fully understand their trajectories in the receiving societies (Sontz 1987). Housing on the other hand is at the centre of attention in the public debate, as it is the socio-spatial concentration of immigrants in so-called ‘ethnic neighbourhoods’ that most clearly seems to prove their segregation and the development of ‘parallel societies’. Apart from this, residential location is pivotal to the integration processes of immigrants, as it is ‘a factor which not only reflects social distance and acts as a symbol of status, but which also determines, to a large extent, access to services and therefore to life chances’ (Robinson 1999, 415). Finally, these two spheres were chosen for their high degree of comparability, allowing us to confront the roles of immigrants as

Introduction • 7

‘replacement workers’ and ‘replacement dwellers’ in the secondary segments of both the labour and housing markets. Clearly, this book sets out to answer many of the same questions asked by others who tackle the issues of integration, segregation and parallel lives. Where do immigrants work and live? What kind of jobs do they do? What do the houses and neighbourhoods they live in look like? What are their opportunities for social contact across ethnic boundaries? To what extent are these opportunities translated into actual contacts? And what kind of relationships come out of these contacts? It differs however from most of the literature by combining two subjects that are often developed separately: the structural positioning of immigrants, and the social relations they develop. The focus on the latter has recently gained a much stronger foothold in the social sciences due to the popularity of the paradigm of ‘everyday multiculturalism’ (Wise and Velayutham 2009); however, it is still underdeveloped in comparison to the former. Combining the structural with the social seems crucial for an understanding of the whole picture of the integration process, as ‘the behaviour of individuals and the interactions between groups cannot be simply clarified by differences in position or orientations’ and vice versa (Reinsch 2001, 21). By looking at immigrants’ positions in the labour and housing markets on the one hand and their actual social behaviour in the workplace and the neighbourhood on the other hand, this book aims to provide a more thorough understanding of both. As already indicated above, this book tackles a period that has only recently become the subject of research: the 1960s and 1970s, when the bulk of postwar labour migrants and their families arrived in the receiving societies. Understanding the historical circumstances of their arrival as well as the initial position allocated to them is necessary to fully comprehend their subsequent trajectories and those of their offspring. As this book will show, it is not only the amount of time immigrants have spent in the receiving society but also the exact moment at which they arrived that are important factors influencing the positions they occupy and the kinds of social relations they engage in. Therefore, we cannot understand the current position of immigrants if we do not look back at the moment in which they first arrived. This has also been shown for more recent newcomers in Europe today (Pratsinakis et al. 2017, 112–113). Finally, the book stands out from the current literature in that it goes beyond a snapshot in time and instead follows the integration processes of these immigrants, their degree of concentration and the extent to which they were segregated from others over a period of twenty years. As befits a historical study, the book analyses and explains these historical developments within an ever changing context, paying attention to both changes

8  •  Parallel Lives Revisited

in individual and group behaviour and in the political, social and economic opportunity structures surrounding it. As such, the book aims to offer a more nuanced understanding of the mechanisms underlying immigrants’ integration processes in the past as well as today.

Methodology Understanding immigrants’ integration processes and their changes over time asks for a balanced perspective on the interplay of many different factors. It means understanding the complex relations between causes and consequences, and between choice and constraint. Such complexity is absent from much of the discourse on ‘parallel lives’ and ‘parallel societies’, where the blame for the perceived division in society is squarely put with the immigrants and their descendants – especially if they are Muslim – whereas the impact of the attitudes and behaviour of members of the majority is not taken into account (Phillips 2006). In the realm of the social sciences, the matter of choice vs. constraint translates into the ever present question of agency vs. structure. Even though never as black-and-white as in the public debate, scholarly analyses of the patterns of concentration/segregation of immigrant populations tend to position themselves towards one end of the spectrum between structural determination and individual agency. When it comes to the study of postwar labour migrants, structural explanations have long prevailed, presenting them as victims of the economic and political powers that structured their migration to and integration in Western European societies. More recently, there has been an increase in studies that pay closer attention to the agency of these immigrants, portraying them not as victims but as active agents, adventurers even, carving out their own trajectories and shaping their own lives (von Oswald, Schönwälder and Sonnenberger 2003, 31). At times however, these studies seem to go too far, presenting individual immigrants’ trajectories as unique experiences and neglecting the structural contexts that frame them and the collective stories in which they are embedded (Fernandez Asperilla 2004, 195). In order to combine the insights of both stances, this book takes what Nancy Green has called a poststructural structuralist approach (Green 1997b, 71–72), examining the history of immigrant integration as an interplay between the agency of individuals and groups on the one hand, and the structures surrounding their settlement and integration processes on the other hand. Within this poststructural structuralist framework, the book adopts a multidimensional perspective on the integration processes it studies, as opposed to the ethno-focal perspective that prevails not only in the public debate

Introduction • 9

but also in a lot of research. Indeed, many integration studies are marked by what we could call (paraphrasing Rogers Brubaker) ‘ethnic groupism’: the tendency to see ‘ethnic groups’ as ‘fundamental units of analysis and basic constituents of the social world’ (Brubaker 2004, 2). From this perspective, the position and behaviour of immigrants are often interpreted as and even reduced to a consequence of their ‘ethnic identity’ (Baumann 1996, 1–6). As early as the 1990s, Nancy Green pointed out the ‘risk in “ethnicizing” relations beyond what the historical record can bear’ (Green 1997a, 209), and one decade later, Lucassen et al. claimed that the ‘uncritical assumption that immigrants comprised homogeneous national or ethnic groups often stands in the way of a nuanced understanding of the integration process’ (Lucassen, Feldman and Oltmer 2006, 15). Despite these and other critiques, many studies continue to be framed by an ethnic lens that obscures the diversity of immigrants’ experiences and trajectories (Glick Schiller, Çağlar and Guldbrandsen 2006, 613). Many others, including the advocates of the relatively recent ‘super-diversity’-paradigm,2 have called for a multidimensional perspective to replace the ethno-focal one. Such a perspective allows for a clearer delineation of the differences and similarities within immigrant populations that cross ethnic boundaries, and for a clearer understanding of how these affect the ways in which immigrants construct their lives in the receiving societies.3 This book applies such a multidimensional perspective, looking at a wide range of social variables beyond ethnicity. These include not only socio-demographic characteristics such as age, gender and marital status, but also migration-related factors such as moment of arrival, period of residence and administrative status, as well as socio-cultural attributes such as language proficiency and education. In order to analyse the interplay between structure and agency whilst paying attention to all of these different variables, three specific methodological choices were made. One, the focus of the research was not on one ‘ethnic group’, but on a broader category of immigrants not defined by their ethnicity. The subjects of this book therefore are not ‘Turkish’ or ‘Italian’ immigrants, but what I have called ‘Mediterranean immigrants’, postwar labour migrants and their families, coming from the Mediterranean periphery to the European core during the period of economic boom. Two, the research methodology combined quantitative and qualitative methods, providing both a numerical analysis of the research population and its behaviour as a group as well as a more narrative analysis of the opportunity structures as well as the choices and motivations of individual immigrants. Three, the research was carried out by means of a local case study. The immigrant population that was to be the subject of study was delineated based on their move to a particular locality, and the integration processes that would be studied were constricted to those that occurred within the context of this locality. This

10  •  Parallel Lives Revisited

focus at the local level allows the book to fully grasp the complex interplay between the behaviour and experiences of immigrants in their everyday lives and the highly specific and ever changing opportunity structures in which these occurred (Ruble, Hanley and Garland 2008, 8). Finally, for the analysis of social relations both at the workplace and in the neighbourhood, we have made use of a classification system developed by urban sociologist Talja Blokland (2003a). Even though Blokland elaborated this system specifically for neighbour relations, it can easily be applied to workplace-based relations as well. According to Blokland, this kind of social relations are at the very least ‘interdependencies’, abstract connections that become clear when, for example, the behaviour of one individual becomes a nuisance to the others. When people start doing each other small favours that benefit both parties, Blokland speaks of ‘transactional relations’. Neighbours and colleagues who value good social relations and are willing to put in more effort can develop ‘attachments’ – they greet each other and regularly chat, and are willing to help when they are needed, without however really getting to know the other as a specific individual. When the uniqueness of individuals does matter and social relations develop into friendships, or when neighbours or colleagues are also relatives, their relations are described by Blokland as ‘bonds’ (Blokland 2003a, 78–85). These concepts will come back in chapters 3 and 5 of this book.

Case Study and Research Population The case study upon which this book is based focuses on the integration processes of labour migrants and their families coming from six countries on both sides of the Mediterranean (Spain, Italy, Turkey, Tunisia, Algeria and Morocco) who settled in the Belgian city of Ghent over the course of the 1960s and 1970s. The choice of immigrants from these six countries was based on their numerical importance. Each country was represented in the city by at least 450 citizens during the period under study. Other nationalities who could also fall under the category of Mediterranean immigrants, such as Portuguese, Yugoslavs or Greeks, were left out as their numbers never reached more than a couple of dozens.4 The period 1960–1980 was chosen because those twenty years witnessed major changes in the (local) opportunity structures as well as the immigrant population itself. Whereas up until the late 1950s, the city had been almost exclusively dependent on its textile industry, the following decades were characterized by a process of large-scale economic diversification. At the same time, the size of the non-Belgian population in the city skyrocketed, and its

Introduction • 11

origin changed from mostly neighbouring countries to the Mediterranean. In addition, against a wider backdrop of international developments, studying these two decades allows for an interesting comparison between a period of economic growth, coinciding with a relatively open-door policy towards labour migrants, and a period of economic crisis, marked by a more restrictive immigration policy. Finally, the city of Ghent was chosen as the focus of this book as a historical context that differed from the one that is more common in this kind of studies: a metropolis such as Paris, London or Berlin; a smaller city with long-term established immigrant communities, like Birmingham, Antwerp or Lyon; or a (semi-)rural town around a site of industrial activity, such as the mining towns in the north of France, the south of Belgium or the German Ruhr. Such a different context leads to different outcomes in the integration process. As a middle-sized provincial capital with a diversifying economy, Ghent offered a setting that allowed for a relatively large degree of agency in the choice of a job and house, but not quite the unlimited opportunities of the metropolis. It provided a context big enough to sustain the development of ‘ethnic economies’ and ‘ethnic neighbourhoods’, but too small to contain a fully developed ‘ethnic infrastructure’. Finally, in this city, postwar ‘guest workers’ and their families were the first immigrants to arrive and settle in large numbers. The city had hardly seen any international migration in the Interbellum or the immediate postwar period. Ghent was not alone in this. At precisely this moment, all over northwestern Europe, middle-sized urban communities such as Ghent were confronted for the first time with a massive influx of immigrants coming from afar. As Sarah Hackett, who studied the immigrant populations in two such communities, Newcastle and Bremen, declared: ‘further research on [their] numerically smaller and often neglected ethnic minority populations … would offer a new sphere to the study of integration’ (Hackett 2013, 225). This book therefore holds value beyond its case study, and thus we contend that its main findings can be extrapolated to other, similar cities, which up until now have remained largely understudied.

Source Material Paying attention to both structure and agency through a mix of quantitative and qualitative methods requires a combination of different kinds of sources. The bulk of the material for the quantitative research for this book was a sample of individual immigrant files, stored in the archives of the city’s registrar’s office. Most of the qualitative research was based on (the transcriptions of ) oral history interviews.

12  •  Parallel Lives Revisited

The archive of the registrar’s office of the city of Ghent contains a collection of individual files that were drawn up by the administration for all immigrants at their arrival in the city, and this from the mid nineteenth century until the end of the period under study (Caestecker, Strubbe and Tallier 2009). These files include the personal data of all members of the household: name, place and date of birth, family history, migration history, professional data, information about residence, labour and professional permits, list of addresses in the city, etc. Since no individual files were drawn up for immigrants who already had a permanent residence permit when they arrived in the city, these immigrants were traced by a search in the so-called ‘registers of arrival’ and in the population registers. A sample was taken from these archives, providing data on some 1,600 individuals. It is likely to approach 10 per cent of the total immigrant population, but since there are no data on the total number of arrivals, we cannot be more precise. More information on this sample and the way in which it was used can be found in the quantitative appendix at the end of the book. An additional source of quantitative data was the personnel register of the Union Cotonnière, later UCO Ltd, then the biggest employer of immigrant labour in the city. From this register, a sample of 570 immigrant employees was taken, selected in the same way as the sample described above. The quantitative processing of these data allowed for a better insight in the employment trajectories of the immigrants working in the city’s textile industry. Data for the qualitative part of the research came from a variety of sources, most importantly the yearly reports of the city of Ghent; a number of contemporary BA and MA theses; the minutes of the Ghent City Council; and different institutional and private archives, most importantly those of the local unions, the city’s Consultative Commission for Guest Workers, and of Maurice Maréchal, the city’s first integration officer. However, the most important sources for the qualitative analysis were interviews conducted with first generation immigrants and, to a lesser extent, with so-called ‘privileged witnesses’ (employers, integration workers, Belgian partners of immigrants, etc.). Some of these interviews were carried out specifically for this research, whereas others had been generated in the course of other projects, either by the author herself or by others. The forty-nine interviews that were conducted specifically for this research were carried out over the period 2009–2011. The first contacts with interviewees were made by visiting key spaces such as mosques, shops and cafés; by contacting companies that had employed many immigrants; and through the author’s personal network. From there, subsequent interviewees were found through the method of snowballing. Interviewing continued until a qualitatively representative sample was reached. Specifically, we ensured that the sample covered both men and women of different nationalities; recruited

Introduction • 13

and spontaneous, pioneer and chain migrants; independent labour migrants and family migrants; immigrants working in different economic sectors; those who were single at arrival and those who were married; etc. Interviews were conducted mostly at the homes of the people concerned, sometimes at the workplace or at one of the above-mentioned key spaces, and sometimes in a ‘neutral’ place like a coffee house, the public library, a park, etc. The method used was that of life story interviewing, opening up the interview to include all aspects of the interviewee’s life, but centred around the topic of migration to and integration in Ghent. No questionnaires were used, but the interviewer did try to let the interviewee touch upon the themes that were specific to the research. The interviews lasted anywhere between one and three hours, and some respondents were interviewed more than once. Most of the interviews with first generation immigrants were conducted in what we could call an ‘in-between language’ (mostly French). In some cases, the interview was done in Dutch, often in the presence of a son or daughter of the interviewee, who functioned as an ad hoc-interpreter. Only the Italian respondents were interviewed in their native language. In addition, the research made use of seventy-seven oral history interviews carried out in the framework of other projects that took place over the course of the 1990s and early 2000s, fourteen of which by the author herself, the others by students at the university of Ghent or by volunteers of local immigrant organizations. The sampling methods, circumstances of the interviews and languages used are widely divergent from one project to another, and, within the same project, from one interviewer to another. The specific issues some of these interviews suffer from will be brought up where they are quoted in the text. Suffice it to say here that – in the case of students – most of the interviews were carried out through the mediation of a (voluntary) interpreter, and some of the interviewers were clearly not properly prepared to do this kind of work, which negatively influenced the quality of the interviews. A third set of nine interviews were not historical but contemporary, carried out over the years 1970–1971 by Robin Roeck, a student at the University of Ghent. These interviews are particularly interesting as they give additional information compared to the oral history interviews, showing, for example, the high degree of mobility and the importance of return in the earlier years of labour migration. Whereas our quantitative research clearly indicated a large percentage of immigrants leaving the city after a very short period of time (see chapter 1), the oral history interviews did not say much about this group of passers-by. Their social relations at the workplace and in the neighbourhood probably remained superficial, as was also the case for permanent immigrants in the beginning of their stay in Ghent. When reading the analysis in this book, it is important to keep in mind that many newcomers

14  •  Parallel Lives Revisited

never made it past the initial stage of settling in, and that the description of social relations as they developed over time was only valid for a small part of  the  total number of immigrants arriving in the city during the period under study. By combining these interviews with different kinds of contemporary source materials, we have tried to make up for the biases that invariably enter all sources that were created post factum. It is not within the scope of this introduction to repeat the debate on these and other methodological issues that arise when using these sources. This has been done time and again, and the interested reader can consult a plethora of publications on the topic.5

The Scope of the Book The book starts with an introductory chapter in which we briefly introduce the changes that took place in the labour and housing markets of the city of Ghent during the period under study. In the second chapter, we explore the position taken by immigrants in the local labour market, looking at the different economic sectors in which they were concentrated and the kind of jobs they did. To get a better idea of the social contact these jobs allowed for, we briefly describe their spatial and temporal characteristics. We then focus on the mobility of immigrant workers in the labour market and look at the extent to which immigrants worked with Belgian colleagues and other immigrants across ethnic boundaries. In the third chapter, we look at the actual development of social relations between immigrant workers, their colleagues and their employers in the micro-cosmos of the workplace. Distinguishing between different kinds of workplaces, we try to explain the nature of these relations and their changes over time. The fourth chapter focuses on the position of immigrants in the local housing market and their spatial dispersion across the city. First, we analyse the kind of housing immigrants and their families came to occupy. Further, we sketch some of the characteristics of the neighbourhoods they ended up in. Finally, we take a look at their residential mobility and discuss the extent to which immigrants shared their streets and neighbourhoods with Belgians and other immigrants across ethnic boundaries, and how this changed throughout the period under study. The fifth chapter of the book analyses the development of social relations in the neighbourhoods where immigrants and their families came to live. After a short description of the specific situation in temporary housing, we first take a look at the relations immigrant families developed with co-ethnic neighbours, then at those with neighbours across ethnic boundaries. In the

Introduction • 15

last part of this chapter, we discuss the ways in which these neighbourly relations changed over the course of the 1970s, in a climate marked by economic recession and heightened racism. The book concludes with a summary of the mechanisms underlying the ways in which immigrants’ structural positions and social relations at work and in the neighbourhood have changed over time. What happened in the 1960s and 1970s had a major impact on what was to follow. The second part of the conclusion touches upon the more recent past, providing a brief discussion of how the immigrants that are the subject of this book and their descendants have fared over time. On a final note, the conclusion revisits the notions of ‘parallel lives’ and ‘hunkering down’, looking at what this book has added to the ongoing debate.

Notes 1. E.g. the Algerian population in France which, at the time of the War of Independence, 1954–1962, came to be considered a threat to national security and thus worthy of surveillance (Blanc-Chaléard 2006, 52). 2. See Vertovec 2007 for a comprehensive overview of the meanings and development of the term. 3. For a discussion of the limitations of an ‘ethnic lens’ in migration research, see, amongst others, Kloosterman and Rath 2003, 6; Glick Schiller, Çağlar and Guldbrandsen 2006, 613; Ratcliffe 2009, 446–47; Vertovec 2007. 4. In 1980, the final year of the period under study here, only 51 Portuguese, 32 Yugoslavs and 66 Greeks were registered in Ghent. Yearly report of the city, 1980. 5. A thorough discussion of this and other issues that come to the fore in oral history practice can be found in, amongst others, Tonkin 1992; Perks and Thomson 1998; Charlton, Myers and Sharpless 2008.

Chapter 1

Postwar Migration to the City of Ghent 

Ghent in the Postwar Period: a Dual Labour Market in a Diversified Economy From the very start of the Industrial Revolution, the city of Ghent had been a hub of textile activity, so much so that it came to be known as ‘the Manchester of the Continent’. Throughout the nineteenth century and up until World War I, the textile industry dominated the economy of the city. This changed from the late 1950s onwards, when Ghent witnessed the development of new economic sectors and industries, bringing about a diversification of the economy that had not been there before (De Kinder 1977, 3; 1972, 3–6). Large government investments in infrastructural works and public housing, as well as a growing demand for private housing, linked to the rising average income of the population, greatly stimulated the development of the construction industry (Voordeckers 1966, 102–4). Government investments also played an important role in the attraction of new industrial activities, especially to the Ghent Canal Zone, where in the second half of the 1960s several multinational companies, such as steel giant SIDMAR, the VOLVO automobile factory and the TEXACO oil refinery, opened their doors.1 However, as elsewhere in Western Europe, the most important expansion and diversification took place in the service sector,2 which saw an impressive development of trading companies, banks, research activities, public and private services, hotels and restaurants, etc. (Allaert 1994, 29).

18  •  Parallel Lives Revisited

This diversification of ­economic sectors was reflected in a growing diversity of labour market opportunities, with manual wage labour in textiles losing its relative importance to manual jobs in construction, metal, chemistry, food, catering, etc., and more and more also to non-manual, white-collar jobs in the public and private services, banking and trade (Voordeckers 1966, 63–111). Also in absolute numbers, employment in textiles lost much of its importance, shrinking from 21,395 in 1947 to a mere 7,558 in 1972 (De Wilde 2007, 91, 97). This quantitative expansion of the labour market resulted in an almost complete absorption of the available labour force. Whereas in 1958, 9.15 per cent of the active population of the city had been unemployed, by 1964 this percentage had gone down to a historical low of 2.4 per cent. Throughout the following years and up until the crisis year of 1974, local unemployment figures remained very low, oscillating around a yearly average of 1,500 unemployed.3 Many of the jobs that were newly created, especially those for whitecollar workers, but also blue-collar jobs, particularly those in the SIDMAR, VOLVO and TEXACO factories, came with a higher wage, better working conditions and a higher social status than most of the jobs that had traditionally been available. We can say that these jobs belonged to the primary sector of the labour market. Most of them were allocated to men (Lambrechts 1979). Many, if not all, of the unskilled and semiskilled jobs in the textile industry, in confection and in other manufacturing industries, in domestic services, cleaning and catering, and in the construction sector belonged to what is called the secondary sector of the labour market. This secondary sector was also divided by gender, with jobs in confection, domestic services and cleaning being done almost exclusively by women, and jobs in construction by men. Other sectors and industries, such as textiles and catering, had a more mixed workforce, employing both men and women (Voordeckers 1966, 67–111). Throughout the period under study, unskilled and semiskilled work in textiles, confection and other manufacturing industries such as food-processing remained very labour-intensive, despite structural reforms in these sectors. Further, these industries were in decline. This downward spiral and lack of a long-term perspective contributed to the unpopularity and low status of these jobs (Bade 2003, 229; De Kinder 1977; Pauwels 1986; De Wilde 2007; Penninx and Roosblad 2000, 5). For jobs in domestic services, cleaning and catering, the picture was even bleaker. Throughout the period under study, these jobs remained completely unskilled, and often belonged to the black labour market. Further, the degree of unionization was extremely low, which left the workers in these sectors almost completely unprotected (Bade 2003, 229; Anderson 1999, 131). Finally, also in construction, most of the jobs were unattractive. In the era of Taylorism (up until the crisis of the mid1970s), apart from a small core of fixed workers, there was a large group of

Postwar Migration to the City of Ghent  •  19

‘interim labourers’, lowly skilled workers who were easily hired and fired according to the temporary needs of the company. The economic crisis of the mid-1970s negatively affected the size of the building companies, which led to the loss of many unskilled jobs and a re-valorization of skilled building work. Still, in a number of construction activities, like ground works, unskilled jobs characterized by hard work and low pay abounded (Bade 2003, 229; Hollifield 1992, 152; Duc 2000, 68–71). Over the course of the Golden Sixties, the position of many wage earners in the local labour market improved drastically (De Kinder 1972, 23; Bernabé 1968, 52–53). Due to the generalization of higher education and the development of the welfare state, throughout Western Europe the schooling level and the economic aspirations of the new generations increased, meaning that more and more people aimed for those jobs that were higher skilled and better paid, situated in the primary sector of the labour market. This caused jobs in the secondary sector of the labour market to be deserted, as there were not enough local (male) labourers available or willing to fill them (Heijke 1995, 12; Hollifield 1992, 113; Moulaert and Martens 1982, 82). The only workers that were prepared to carry out this heavy and dangerous work in insecure positions were those people who, in dual labour market-theories, are called ‘target earners’. These workers are not looking for long-term, stable employment, but only aiming to earn as much money as possible within as short a period of time as possible. For them, work is essentially an a-social activity, no more than a means to an end. As opposed to most other workers, their job is not central to their identity formation. In postwar Europe, women were generally counted among this group, as were adolescents living at home, small-scale farmers who combined their farm work with paid employment, and immigrant workers: all of them working only to add some income to their main activity (looking after the household, working on the land) or to achieve short-term goals (typically the consumption or purchase of specific goods or properties, or the setup of an independent activity) (Piore 1979, 54, 87–89).4 However, at exactly the same time as employers in Western Europe experienced an increased need for target earners, female participation in the labour force decreased. During the first half of the twentieth century, women working outdoors had become highly stigmatized, and the model of the housewife had become the role model for most Western European women. Economic changes, cultural changes and policy measures worked together to gradually push women out of the labour market. In Belgium, by the 1950s, female labour market participation was at its lowest level of all times. At the same time, more adolescents stayed at school for longer, due to the democratization of higher education. Small-scale farmers were already rare, as farming had long lost its importance as a major economic sector (Caestecker and

20  •  Parallel Lives Revisited

Vanhaute 2011, 99, 101–5). Therefore, in the local labour market, there were fewer target earners to be found. In this context, many employers, not only in Ghent but everywhere in Western Europe, tried to solve their labour shortages by attracting a foreign workforce (Heijke 1995, 13; Bade 2003, 229; Deslé 1990, 445; Lambrechts 1979, 209, 219–20).5 In Ghent, the labour shortages in the secondary sector of the labour market were most acutely felt during the economic boom of the mid-1960s, first by the building companies that were working in the Canal Zone,6 then by other sectors and industries, especially the city’s textile industry.7 But even after that, when the economy fell back to a slower growth rate and the extraordinary number of job offers of the 1960s had all but disappeared, many employers offering jobs in the secondary sector of the labour market continued to struggle to attract local labourers. Turning to immigrant labour then became a longterm solution, rather than the temporary patching up it had seemed at first (De Kinder 1972, 23). With the first oil shock of 1973, the period of economic growth came to an abrupt end. The years 1974–1975 saw the eruption of a worldwide economic crisis that hit Western Europe particularly hard. This was not just another temporary dip in the business cycle. A period of economic restructuring set in, characterized by the ever growing importance of the service economy and the displacement of industry and thus of industrial employment from the core to the periphery of the global economic system, or from  the countries of the global north to those of the global south (Caestecker and Vanhaute 2011, 107–8). In Ghent, the economic crisis of the mid-1970s brought about the final downfall of the textile industry. As European textiles lost out to industrial production in the Soviet Union and in Third World countries, the Ghent textile factories had to close their doors, one by one. The construction sector also suffered enormously from the crisis. The effect was slightly delayed, but from the late 1970s on, building activity diminished considerably, as both private and public spending on construction were severely curtailed. Luckily, by then, the secondary sector of the local economy had become much more diversified, including many industries that withstood the crisis better. More importantly, Ghent had already begun its transformation from an industrial city to a commercial, educational, leisure and administrative hub, allowing for a significant expansion of the tertiary sector. However, the increasing availability of manual work in the tertiary sector could only partially compensate for the loss of many un- or semi-skilled jobs in the secondary sector, with soaring rates of unemployment as a consequence. In the first crisis year, the number of unemployed almost doubled, from 6,613 in 1974 to 12,103 in 1975. After that, it continued to rise for ten years, reaching a peak of 24,279 ­unemployed in 1985.8

Postwar Migration to the City of Ghent  •  21

Ghent in the Postwar Period: City Flight and Housing Problems Whereas the city of Ghent managed to take advantage of the Golden Sixties when it came to economic growth and employment, the Sixties were no Golden Age in every sense of the word. Long before the outbreak of the economic crisis, the city had fallen prey to a demographic decline with farreaching consequences. Like other cities in Western Europe – and particularly in Belgium – Ghent experienced a severe case of ‘suburban flight’, seeing a large percentage of its population leave for the surrounding countryside (Claeys and Van Hove 1987, 23–26). The urban population went down from 166,096 just after the war to 111,158 by the end of the millennium. Only in 1999 was this downward trend reversed.9 As elsewhere in Belgium, it was mostly young, well-to-do families that left the city. A general impoverishment of the city and a quickly ageing population were among the most serious consequences of this phenomenon. More directly visible in the urban landscape, it also resulted in widespread vacancies and in a growing number of derelict buildings (Kesteloot and Meert 2000, 54–56; Claeys and Van Hove 1987, 31–33). This negative evolution was mitigated by the communal merger of 1977, whereby ten smaller – and mostly richer – suburban communes became part of the administrative unit of Ghent, but even that did not halt the decline of the urban population, as suburban settlers continued to move further away from the city. The suburban flight created a situation of housing surplus, which stood in stark contrast with the situation immediately after World War II, when the city was confronted with a severe housing shortage. As the war ended, the local authorities immediately started construction works, first on singlefamily dwellings, and from the mid-1950s onwards on multi-storey dwellings. Once the most pressing housing needs were alleviated, the city of Ghent turned its attention to its slum problem. However, its renewed ‘anti-slum crusade’ did not deliver the expected results. Especially with the demolition of the typical Ghent beluiken, small closes of workmen’s houses that had been built during the nineteenth and early twentieth century, the city made a lot less progress than had been foreseen. Between 1950 and 1969, only 110 beluiken were pulled down or rebuilt, so that by the end of the decade, the city still counted 279 beluiken, comprising 2,921 houses of which 30 per cent were uninhabited.10 In theory, the continuous decrease of the urban population should have solved the housing problem. Over the course of the 1960s, thousands of dwellings were registered as uninhabited.11 However, from the 1950s onwards, it was the quality rather than the quantity of housing that posed the main problem. Housing in Ghent was bad, and did not improve much over the

22  •  Parallel Lives Revisited

period under study. In general, the housing stock of the city was old. Half of it dated from before World War I, another quarter from the interwar period. Even if by 1970 most houses were connected to the water supply (96  per cent), the sewerage system (92 per cent) and the gas mains (79 per cent) and almost two thirds had a private toilet with flushing system (63 per cent), still only less than half had a private bathroom or shower (45 per cent), and only about a quarter had central heating (27 per cent). Many of the empty dwellings were not only uninhabited, but also uninhabitable, and many of those that were inhabited did not meet modern housing standards. Further, by 1970, most families in Ghent still rented their house rather than owning it,12 whereas for the whole of Belgium, the opposite was the case. The first half of the 1970s was a period of important building activity in Ghent. The era was characterized by high-rise building. Hundreds of social dwellings, mostly flats, were added to the city’s housing stock, all of which were swiftly rented out. However, during the second half of the decade, and especially from 1977 on, the economic crisis hit the building sector at full force and building activity diminished considerably. In this climate of crisis, the city began to change its attitude towards the hovels and beluiken it had been aiming to annihilate for so long. Renovation and restoration came to be seen as a valid alternative to demolition. Not only economic but also social considerations influenced this change of perspective. By then, the negative

Figure 1.1  Slum-like housing in Ghent in the 1970s, example of a beluik. AmsabInstitute of Social History, collection Haenebalcke

Postwar Migration to the City of Ghent  •  23

social impact of high-rise apartment blocks had become clear, and there was a growing willingness to go back to single-family dwellings as the preferred form of housing. In 1976, as an experiment, the city began a renovation project in two beluiken that was successfully concluded by early 1980. With these and other projects, renovation became the keyword in the first half of the 1980s, not only in social housing, but also in the private sector, where funding was given to individuals and families who wanted to renovate their houses. The crisis caused a real impoverishment of the local population, which had an immediate effect on their housing situation (Claeys and Van Hove 1987, 38). Notwithstanding a continuous decrease in population, from 148,860 in 1970 to 132,534 in 1979, and an increase in the number of social dwellings, reaching more than 9,000 by 1980 (Werkgroep Huisvesting Gent 1984, 8, 10), the waiting lists for social housing were enormous.13 Even those people who were forced to leave their houses in the framework of rehabilitation and city renewal projects often could not be rehoused in social dwellings. More than once, they ended up in hovels and beluiken just like the ones they had been made to move out of.14 In fact, a large part of the population could not even afford social housing, as the newly constructed housing projects boasted rental prices that were many times higher than the rent these people were used to pay and could afford – even when they were granted relocation and rental allowances.15 Independent of income restrictions, larger families (with more than two children) had almost no chance of being rehoused in social accommodation, as most of the social housing in the city consisted of apartments with three bedrooms or less.16 These people were restricted to the private sector of the rental market. By the 1980s, the quality of this private rental housing was still far from satisfactory, although it had improved over the course of the 1970s. For the whole agglomeration, 65 per cent of dwellings were now equipped with a private toilet with flushing system and a private bathroom or shower, and more than half of these also had central heating. However, such ‘fully equipped’ houses were more common in what used to be the neighbouring villages of Ghent than in the city itself. In the working-class quarters to the north and east of the city centre, they were rare (Claeys and Van Hove 1987, 64–66). Even in 1981, more than half of the housing stock still dated from before World War II, with the oldest dwellings in the city centre and the newest ones in the former villages (Claeys and Van Hove 1987, 51). There was a clear division of the market, with a more modern part, with all the necessary comfort, and an aged part, where modern comfort was rare. In accordance with what we have seen in the labour market, we could speak of a ‘dual housing market’, divided into a primary and a secondary sector. Both on the social and on the private rental market, the demand for good quality, affordable single-family dwellings greatly exceeded the a­ vailability.

24  •  Parallel Lives Revisited

This allowed for bad quality housing to remain relatively competitive, blocking the incentive for renovation and causing the further dilapidation of houses, streets and even complete neighbourhoods (Kesteloot, De Decker and Manço 1997, 74). The lower income groups in society, such as the elderly, the disabled, immigrant and non-conventional families, found themselves relegated to this kind of housing.17 In other words, these people were ‘replacement dwellers’, restricted to the secondary sector of the housing market, that offered rental houses and apartments of bad quality at relatively low prices.18

A History of Mediterranean Migration to Ghent In 2014, Ghent celebrated its status as a ‘super-diverse city’ in the framework of the commemoration activities of fifty years of Moroccan and Turkish  migration to Belgium. That year, an abundance of publications, exhibitions and politicians’ speeches called Ghent ‘the New York of Belgium’, mentioning that almost 30 per cent of its inhabitants now were ‘of foreign origin’ – including children with at least one parent of foreign nationality – and that there were more than 150 different nationalities present in the city.19 However, for scholars dealing with diverse societies, these numbers seem low, compared to cities like New York where around 60 per cent of inhabitants are immigrants and children born to immigrant mothers20 and where – talking in terms of ‘racial diversity’ – not one single ‘racial group’ makes up more than a third of the population.21 Yet, for the people of Ghent, the idea that nearly a third of their fellow citizens have foreign roots was and is something extraordinary, as the city only relatively recently started to attract larger numbers of international immigrants. In fact, until the 1960s, the share of non-Belgian inhabitants in Ghent (including the children of immigrants) was negligible, oscillating around 2 per cent. In 1935, the first year for which we have data diversifying by nationality, the city was home to some 3,800 non-Belgians, representing about 2.3 per cent of its 165,269 population. The large majority of these were immigrants coming from Belgium’s neighbouring countries, especially the Netherlands and France. That year, they represented 70 per cent of all non-Belgians in Ghent, compared to only 7 per cent for the immigrants from the countries around the Mediterranean. The latter had become more numerous by 1938, when 244 Italians, 149 Spanish, 20 Greeks, 9 Portuguese and 8 Turks were registered in the city’s population records. There was also a small Algerian presence of around ten people; however, as they were counted as French citizens, their Algerian roots remained invisible in the official population statistics.22 Italian immigrants had long been present in Ghent, working as

Postwar Migration to the City of Ghent  •  25

labourers in various sectors of the local economy, but more often as skilled craftsmen, hawkers or independent shopkeepers, or as owners or employees in the hotel and catering industry.23 The significant Spanish presence however was recent, related to the arrival of more than two hundred niños de la guerra, refugee children from the Spanish civil war (Pee 2001, 66–75). During World War II, the number of non-Belgians went down to about 3,300 or 2 per cent of the total population. In 1945, the Mediterranean presence in Ghent consisted of only 215 Italians, 74 Spanish, 19 Greeks, 16 Portuguese and 1 Turk and approximately 8 musulmans français. The immediate postwar period did not see their numbers rise significantly. During the 1950s, Mediterranean migration to Ghent almost came to a standstill, and if anything, the number of immigrants went down. This however would change considerably in the decades that followed. From the mid-1960s onwards, immigrants from the Mediterranean came to dominate the city’s foreign population. While between 1945 and the early 1960s these immigrants and their descendants made up a mere 10 per cent of the total non-Belgian population, this suddenly increased to over 25 per cent in 1965 and 65 per cent in 1999. Their numbers skyrocketed from 325 in the immediate postwar period to more than 1,000 in 1965 and over 11,300 in 1999. The effect of this new migration on the outlook of the urban population was magnified by the simultaneously ongoing suburban flight, that only concerned Belgian families. However, by 1999, they still made up only 5 per cent of the city’s population as a whole.24 After that date, the numbers are no longer reliable due to a sudden increase in naturalizations, which meant that many immigrants and their descendants disappeared from the statistics.25 Whereas in the beginning of the period under study, labour migration to Ghent consisted mostly of Italian and Spanish immigrants, from the mid1960s onwards more newcomers from different countries of origin arrived in the city. First, it looked like Ghent was going to have a big Algerian community, as the number of Algerians went up from 87 in 1963 to 179 in 1964 and 200 in 1965, after a recruitment effort among the Algerian community in northern France by the Union Cotonnière (De Bock 2012). However, as it turned out, it was the Turkish immigrants and their descendants who came to dominate the immigrant presence in Ghent. In 1964, when the first Turkish immigrants arrived, they represented barely 2 per cent of the population under study. By 1971, they had surpassed all other Mediterranean immigrants, and in 1980, more than 60 per cent of the city’s Mediterranean population was Turkish. By then, 5,690 Turkish people resided in the city. Today, the Turkish community still is by far the largest minority group in Ghent, with an estimated 8.8 per cent of the city’s population having Turkish roots.26 Moroccan immigrants, who were to become the second largest group, were equally badly represented in the beginning of the 1960s. Moroccan migra-

26  •  Parallel Lives Revisited

Figure 1.2  Faruk Köse and his friends, who were among the first Turkish immigrants to arrive in Ghent in 1965. Amsab-Institute of Social History, collection Köse

tion to Ghent only took off in 1964, but by 1971 the number of Moroccan residents surpassed all but that of the Turkish. In 1980, 1,084 people with Moroccan nationality lived in Ghent, representing more than 10 per cent of its Mediterranean population. Throughout the period under study, immigrants of other nationalities remained present in much lower numbers. For the dynamics of Turkish migration to Ghent, it is important to know that a very high percentage of immigrants came from the same place of origin. Over half of the Turkish independent immigrants from our research sample were born in or around the town of Emirdağ (province of Afyon, Central Anatolia).27 No other single town, city or even district around the Mediterranean has accounted for such a large number of immigrants arriving in Ghent. Therefore, the Turks from Emirdağ constitute the only ‘transplanted community’ in the city (Den Exter and Kutlu 1993, 26; Ostergren 1979, 10). The presence of such a transplanted community in Ghent has greatly facilitated the continuous arrival of new immigrants from Emirdağ, even after the borders were closed to labour migrants in 1974 (Reniers 1997, 10).

The Arrival of Postwar Labour Migrants and Their Families in the City The arrival of postwar labour migrants, or ‘guest workers’, and their families in the city was not constant over time, but concentrated in peaks, ­alternated

Postwar Migration to the City of Ghent  •  27

with moments of very low numbers of arrival. Despite the specificities that characterize each nationality, we can discern three moments of arrival that were especially important. First, what we could call the Golden Sixties (the period 1963–1967); second, the years of economic revival (1969–1971); and third, the initial crisis years (1974–1975). For all ‘guest workers’, the initial period of arrival in Ghent can be situated in the same short time span, the height of the Golden Sixties, from 1963 to 1967. These first years were probably the most important ones in the history of postwar migration to Ghent, as they saw the arrival of large numbers of immigrants who had no previous network in the city, and who would establish the roots for a new system of chain migration. Most immigrants in this period were men, except for the Spanish, half of whom were women. The large majority (more than 80 per cent of Turkish and more than 90 per cent of other nationalities) were registered in Ghent as independent labour migrants, whose right to reside depended on a regular employment contract.28 The arrival of these first labour migrants coincided with a period of strong economic growth that demanded the employment of a far bigger labour force than the local labour market could provide. The history of Mediterranean migration to Ghent is not exceptional in this. Indeed, for the whole of Western Europe, it was the booming economy of the postwar period and the ensuing demand for labour that made it possible for millions of Mediterranean immigrants to come and live in Europe’s industrial regions and towns. In the city of Ghent, three economic sectors accounted for the attraction of the majority of labour migrants in the period 1960–1980 (Table 1). Most importantly, there was the textile industry, in constant need of labourers since the beginning of the 1950s, especially for night and shift work. In the years of high economic activity, 1963–1966, this need became a necessity in order for the industry to fulfil its competitive potential, causing it to start attracting foreign labour (De Wilde 2007). Secondly, there was the construction industry and in particular the sudden and massive development of the Ghent Canal Zone, where, by the mid-1960s, an army of ­building contractors was constructing a gigantic industrial complex where several international companies would soon set up production. With these building contractors, who came from Antwerp, Brussels, Liège and also from abroad, considerable numbers of foreign workers of different ­nationalities arrived in and around Ghent. At the same time, new job opportunities were created in a local environment where labour reserves were already short, a situation that attracted other immigrant workers (De Kinder 1972, 23). Thirdly,  a non-negligible demand for foreign labour found its origins in the Ghent bourgeoisie, which was looking to recruit foreign domestic servants for its large households in and around the city.

28  •  Parallel Lives Revisited

Table 1.1  Percentage of independent immigrants taking up a first job in textiles, ­construction or domestic service, 1960–1980 (n=567). For more detail, see the quantitative appendix, chapter one (2). % economic sector Algerians Italians Turks Moroccans Spanish Tunisians

Textiles

Construction

Domestic service

60% 2% 53% 69% 5% 59%

11% 58% 21% 5% 3% 10%

0% 3% 0% 1% 64% 4%

Through this channel, many female immigrants, together with a number of men, found their way to Ghent. These three sectors had a different importance for immigrants of different nationalities. Whereas in the mid-1960s Italian immigration to Ghent was linked almost exclusively to the building works along the Canal Zone, the arrival of Spanish immigrants was closely related to the recruitment of domestic labour, and the immigration of Tunisians, Moroccans, Turks and especially Algerians was largely attracted by the textile industry. This differentiated economic profile was caused by the recruitment policies of employers, the historical presence of immigrant intermediaries in the city and last but not least, the random arrival of individual pioneers who set up new migration chains, aimed at a specific industry or economic sector (De Bock 2017). Due to an economic downturn in the last months of 1966, UCO Ltd, the largest employer in the Ghent textile industry, stopped the recruitment of labourers, forced its labour force into a situation of partial unemployment and even began to fire those labourers who had only recently been hired (De Bruyn 1966, 32). On a national scale, legislative measures were taken to further prevent the arrival of new labour migrants into the country by imposing the strict application of articles 1 and 2 of the Royal Decree of 31 March 1936, allowing the immigration of foreign workers only after a double, preliminary authorization had been granted to both the employer and the employee concerned.29 During the years of economic growth, the stipulations of this article had been largely ignored to the benefit of those industries that were in urgent need of foreign labour. Between 1962 and 1966, immigrants who had arrived on a tourist visa and had found work were almost automatically granted a regularization of their residential status (Martens 1976, 139–42; Haex, Martens and Wolf 1976, 39–41). This practice now came to an end. However, new labour migrants kept arriving in Belgium, and not all of them effectively left the country when they were refused a

Postwar Migration to the City of Ghent  •  29

residence permit. Unintentionally, the restrictive application of immigration policy during the economic downturn of 1967 caused the coming into being of a group of ‘illegal immigrants’, people who resided in Belgium but did not have a legal right to live or work there (Surkyn 1993, 4; Tinnemans 1994, 86–87).30 The sudden drop in the availability of jobs in the local labour market, combined with the tightening of immigration policy at the national level, had an instant effect on the arrival (or at least the official registration) of new immigrants in Ghent.31 The year 1967 saw a drop in registrations for residency. Those who still registered had arrived either to work in domestic services, or with one of the large building contractors in the Ghent canal zone, both economic niches that were not directly influenced by the downward business cycle. Otherwise, they had come to Ghent as ‘dependent immigrants’, whose right to residency was based on the residency status of their spouse or parent. Already in 1968, the economic downturn was overturned and a new period of economic growth began. Unemployment figures started going down as of 1969, a trend that would continue in 1970, while in 1971, the number of unemployed remained more or less the same.32 However, this upward business cycle could not prevent the failure of a number of local companies, among whom some larger and smaller textile enterprises.33 Neither could it avert the regular outbreak of spontaneous strikes in all sectors of the economy, but especially in the building, textile and metal industries.34 In these tumultuous years, new immigrants continued to arrive in the city. Men still made up the majority of arrivals, but women represented a much bigger part than in the mid-1960s. In fact, more than half of Turkish immigrants, who by then had become the most numerous group were women. Most of these Turkish women were also dependent partner-migrants, arriving in Ghent to join their spouse. Among other nationalities, however, independent labour migrants continued to make up the large majority.35 As in the previous period, most of them started work in the textile industry. The textile giant UCO Ltd, founded in 1967 through the fusion of several big textile companies in and around Ghent, employed the majority of these new textile workers, whose arrival coincided directly with the years of falling unemployment 1969–1970, and was almost immediately halted as of 1971. Also in this period, the construction sector remained the second biggest employer for newly arriving immigrants. Even though the construction works in the Canal Zone had come to an end, the building sector in Ghent was dynamic enough to keep attracting immigrant workers from other places in Belgium and from abroad. For newly arriving female labour migrants, domestic service remained an important sector of employment in this period, as did the entertainment industry, where women worked as ­dancers or hostesses in nightclubs.36

30  •  Parallel Lives Revisited

After a short period of slackening economic growth, the upward business cycle that had started in 1968 regained strength and up until the summer of 1974, the economy continued to grow. However, this economic growth did not succeed in absorbing the growing number of excess labour in the local labour market. Most enterprises had carried out intense rationalizations in their production process, sustaining their economic growth without employing extra labourers. The second half of 1974 then witnessed a sudden downturn in the business cycle, that caused unemployment in the region around Ghent to almost double in the span of one year (see above). With a delay of some months, the oil crisis of 1973 set off the worst economic crisis the West had seen since World War II. This major change in the economic situation pushed many Western European governments to call an official halt to the further entry of labour migrants (Hollifield 1992, 73–74; Messina 2007, 29–30). In Belgium, a ‘migration stop’ was announced on 1 August 1974. However, it is precisely at this moment of crisis that we see the arrival of a new batch of immigrants in Ghent. More so than in the period 1969–1971, this group of arrivals was made up of men and independent labour migrants (respectively 58 and 60 per cent of the largest group of arrivals, the Turks). Rather than being due to any shortage in the labour market, the arrival of these immigrants, or more accurately, their appearance in the population statistics was very likely related to the regularization campaign organized by the Belgian government. Like many other European countries, Belgium offered immigrants who had entered the country before the migration stop, but who had not succeeded in regularizing their status, the chance to legalize their stay. Between August 1974 and August 1976, the campaign resulted in the regularization of 7,470 ‘illegal immigrants’ (Rea 2000, 321–29).37 After the migration stop, non-EEC migrants could only enter Belgium as labour migrants in very specific cases. However, they could still legally migrate to Belgium as students, political refugees or family members of immigrants who already resided in the country (Aerts and Martens 1978, 36). Many Moroccan, Algerian and Tunisian immigrants now arrived in Ghent as scientific researchers, students or interns, whereas more than 60 per cent of Turkish immigrants arrived as spouses in the framework of a family reunification or family formation.38 Only the Italians, as EEC citizens, continued to arrive as labour migrants, most of them attracted by work in the building and catering sectors. During this period, Spanish migration to the city of Ghent, as to Western Europe in general, all but came to a halt, reaching its lowest levels between 1973 and 1978. The outbreak of the worldwide economic crisis and the ongoing economic development in Spain resulted in a significant reduction of the economic and income differences between the two areas, thus making migration a less attractive option (Ródenas Calatayud 1994, 139–43).

Postwar Migration to the City of Ghent  •  31

Postwar Mediterranean Migration to Ghent: Socio-demographic Profiles Apart from the difference in size, immigrant populations of different nationalities also had varying socio-demographic profiles, related to their diverse migration trajectories, to the differing political and economic frameworks in which their migration took place, and, to a large extent, to historical coincidence as well. In Ghent, for example, Tunisian immigrants were mostly young, single men, as those were the recruitment criteria of the textile industry that brought them to the city. Among Spanish immigrants on the other hand, there were many women, as most Spaniards came to Ghent to work in domestic services, a largely female sector (De Bock 2017). When we divide the Mediterranean population of the city in groups of adults and children, an interesting variation between nationalities comes to the fore. For Spanish nationals, the adults/children ratio remained more or less stable at around two to one throughout the period under study. Among Algerian, Italian, Tunisian, Moroccan and especially Turkish nationals, however, children gradually came to make up an ever larger part. For Algerians and Italians, this evolution towards a younger population was reinforced by a gradual stabilization and even decrease of the numbers of adults. For Tunisians, Moroccans and Turks, it was entirely due to the fast growing number of children, who either arrived as immigrants or, increasingly over the course of the period under study, were born in the city. By 1971, more than 25 per cent of people with Italian and Tunisian nationality, around 40 per cent of people with Algerian nationality and nearly 50 per cent of those with Moroccan and Turkish nationality living in Ghent were below fifteen years of age.39 Also among adults, ages differed considerably from one nationality to another. The Turkish group can generally be considered the ‘older one’, with 26 per cent of immigrants older than thirty-five at arrival, and only 27.5 per cent younger than twenty-five. The percentages of immigrants over thirtyfive were also high among Spanish (23.5 per cent) and Italian (24 per cent) arrivals, but these groups had higher numbers of young immigrants as well, 33.5 per cent and 42 per cent respectively. The youngest group by far were the Tunisians, where the under twenty-five category made up 53 per cent of the total number of arrivals, and the over thirty-five category only 7.5 per cent. Among Algerians and Moroccans, the under twenty-five group made up over one third, and the middle group (twenty-five to thirty-five) just under half of the sample.40 Dividing the Mediterranean population of the city according to sex, we notice that there was a clear gender imbalance among Algerians, Tunisians and Italians, with only around one third of women in 1973. Over the course

32  •  Parallel Lives Revisited

of the 1970s, for the Algerians and Tunisians, this imbalance was slightly alleviated, whereas for the Italians it became more pronounced. Men and women were more equally represented among Moroccan and especially Turkish and Spanish nationals. The first two groups underwent a feminization over the course of the period under study, resulting in a female presence of 46 and 49 per cent respectively in 1980. Spanish women on the other hand gradually lost their majority position, but still represented about 48 per cent of the population by 1980.41 These diverging gender balances are often explained in function of the differences in gender roles in the countries of origin. Countries where it is more common for women to work outdoors or to function as the main breadwinner of the family will see more women leave as labour migrants, whereas countries where women are on average more bound to a role within the household will experience less female out-migration or will see women migrate as accompanying partners of male migrants. The immigration policies of receiving countries also have an impact on the gender balance of migratory flows. During the era of guest worker migration, some countries, such as Germany, purposefully recruited independent female labourers, leading to higher numbers of young, single women arriving in the country at an early stage of the guest worker programme. Others such as Belgium and France encouraged family reunifying migration, so that most women arrived at a later stage (Erdem and Mattes 2003; Martens 1976). The marital status of immigrants when they arrive in a specific receiving society also goes a long way to explain differences in the gender balance between immigrant groups – especially differences that remained strong over time. Together with other social and cultural factors, immigrants’ marital status at arrival directly relates to the potential for family reunification, which in the period of guest worker migration was mostly a migration of female rather than male partners. Further, this variable greatly influenced the number of mixed relationships that developed in the receiving society. Looking at the marital status of Mediterranean immigrants arriving in Ghent, we see that 72 per cent of Turkish independent immigrants were married to a partner from the home country, while this was only the case for 42 per cent of Spanish, 38 per cent of Moroccan, 29 per cent of Italian, 18 per cent of Tunisian and 16 per cent of Algerian independent immigrants. A large-scale survey in Germany in 1968 indicated a similarly high percentage of married individuals among the Turkish population, with 82 per cent of Turkish men and 71 per cent of Turkish women being married at the time of the survey. Erdem and Mattes relate this to the traditionally young age of marriage in Turkey (Erdem and Mattes 2003, 172) – although we saw that Turkish migrants in Ghent were also, on average, older at arrival than immigrants of other nationalities. 3 per cent of Moroccan, 7 per cent of Tunisian,

Postwar Migration to the City of Ghent  •  33

8 per cent of Italian and Spanish and even 11 per cent of Algerian independent immigrants were already married to a local (or, in some cases, thirdcountry) partner at the time of their arrival in Ghent. This greatly diminished the chances for a partner from the home country to come and join them at a later stage.42 Further, Turkish immigrants who were single upon arrival in Ghent almost exclusively married partners from the home country. This was the case for the entirety of our sample. Marriages with locals occurred more frequently among the Moroccan population, where about 33 per cent of unmarried immigrants married a local or third-country partner during their stay in Ghent. For Algerian and Tunisian immigrants who were unmarried at arrival, the percentage of mixed marriages was as high as 68 per cent.43 As our sample sizes for these groups are small, we have to be careful with the interpretation of these numbers. However, it is clear that they are much higher not only in comparison to Turkish migrants in Ghent, but also to the descendants of guest worker migrants today, especially the non-European ones (Lievens 1999). Data for the period 1992–2003 show, for example, an intermarriage rate of 14 per cent for second generation Moroccan men in the Netherlands and 30 per cent for second generation Algerian men in France (Lucassen and Laarman 2009, 59–60).

Figure 1.3  Celebrating the wedding of Italian immigrant Angelo Cocca and his Belgian wife Huguette, early 1960s. Amsab-Institute of Social History, collection Leon

34  •  Parallel Lives Revisited

This difference between first and second generations can largely be explained by the highly unbalanced sex ratio confronting the first ­generation, which we discussed above. The differences between nationalities on the other  hand have to be explained another way. One important factor is the degree to which newcomers were already married when they arrived in Ghent  – as we have seen, Tunisians and Algerians were overwhelmingly single (75 per cent and 73 per cent of independent newcomers respectively), especially when compared to Turks (27 per cent). Further, we can also state that many of them were adventurers, who had migrated not only  to work, but also to experience life in Western European societies. North Africans had often been steeped in French culture in their home countries, and the impact of French colonialism on the degree of intermarriage between Algerians and French is well established (Lucassen and Laarman 2009, 60). The fact that many of them spoke French also enabled better communication with locals – as we will explain further, French was an important language in Ghent in the era when guest workers arrived. Further, more North Africans than Turks had previously lived in big cities, either in their home countries or elsewhere in Western Europe, so that their way of life was more compatible with that of locals than that of rural migrants coming straight from the countryside (De Bock 2015). Finally, the numbers of Algerian and Tunisian immigrants  in Ghent were so low that they were confronted with less social pressure from co-ethnics than those who lived in large, transplanted communities, as was the case for Turkish immigrants (Beyers 2008). It is important to note that not all immigrants arriving in Ghent stayed in the city. Much the opposite is true: for many of them, Ghent was only a temporary stop-off in their migration project. Around 40 per cent of Algerian, Italian and Spanish immigrants did not stay in the city for more than one year; this was also the case for around 25 per cent of Turkish and Moroccan and around 20 per cent of Tunisian immigrants.44 In the literature, these short-term sojourns are often neglected, as they are not easy to measure. Von Oswald et al. (2003) report evidence from Darmstadt (Germany), indicating the importance of short-term stays, as almost one third of immigrants arriving there between 1955 and 1967 stayed for less than eleven months, with most of them leaving already within the first six months (von Oswald, Schönwälder and Sonnenberger 2003, 29–30). Those immigrants who were the first to arrive in Ghent – Algerian, Italian and Spanish immigrants – were also most likely to leave after a very short period. Possible explanations are the inherent mobility of the jobs in the building sector (Salah 1973) where most Italians worked, as well as in the domestic service sector (De Keyzer 1995, 34), to which most Spanish labour migrants were attracted, and the presence of a large Algerian com-

Postwar Migration to the City of Ghent  •  35

munity in Northern France, attracting Algerian immigrants with an ‘ethnic ­infrastructure’ that did not exist in Ghent.45 With such a large percentage of immigrants staying for very short periods of time, especially in the early years of migration to Ghent, the Mediterranean population in the city was marked by a significant turnover. These passersby probably did not develop many lasting ties to the locality. However, even during their short stay, they lived in the city, rented a room or a house somewhere, worked, met up with others, etc. As such, they formed an integral part of the local immigrant population, albeit with a very specific profile. The large majority of these short-term dwellers were men. Women were more likely to stay in the city for a longer period, and they were relatively overrepresented among those staying for five years or longer – within the Turkish population, for example, they made up 48 per cent of such long-term immigrants, compared to only 23 per cent of those staying for less than one year. Related to this, most short-term dwellers were independent immigrants. Dependent migrants were much more likely to settle in a more permanent way. Among Turkish immigrants again, only 42 per cent of independent migrants settled in Ghent for five years or more, whereas this was the case for 59 per cent of dependent immigrants. This clearly indicates that the majority of settlers were couples or families while the majority of short-term dwellers were single men.46 Whereas these single male passers-by are fully included in those parts of our analysis that are based on contemporary data, when it comes to the ­analysis based on oral history interviews, they are much less represented. Nearly all interviews used in this research were carried out with immigrants who ended up settling permanently (or at least for a very long time) in Ghent. Therefore, this book – as most of the research on postwar migration – will be slightly biased towards the experiences of families of long-term settlers.

Conclusion As a medium-sized provincial capital with a diversifying labour market and a housing market characterized by extensive vacancy, the city of Ghent offered a lot of opportunities for postwar labour migrants and their families, who were the first international immigrants to arrive in the city in large numbers. For a long time, Ghent could be classified as a mono-industrial textile town. This changed during the period under study, with the development of many new industries and services that worked together to create a more diversified economic fabric. These developments coincided with a quantitative and qualitative expansion of the local labour market, and with a more pronounced differentiation between primary and secondary sector jobs. The

36  •  Parallel Lives Revisited

latter in particular were confronted with a growing shortage of labour, which necessitated the recruitment of foreign labour. Despite its diversifying economy, Ghent suffered a serious case of suburban flight, seeing its population decrease by nearly one-third over the second half of the twentieth century. The impoverishment this entailed showed itself first and foremost in the city’s housing situation, which could without exaggeration be called disastrous. Thousands of dwellings were uninhabited, and many of those that were inhabited were actually uninhabitable. Throughout the period under study, the quality of the Ghent housing stock remained low, and thousands of families were forced to live in terrible circumstances. Postwar labour migration to the city of Ghent only really started to take off in the mid-1960s, in a context of economic boom and acute shortages on the labour market. Whereas at first, there was a mix of immigrants from different Mediterranean countries, soon Turkish immigrants came to outnumber all other nationalities and dominated the immigrant population of  the city. The arrival of immigrants was not constant over time, but occurred in peaks, related to changing economic and political opportunities and constraints. Immigrants of different nationalities arrived at different moments, to work in different economic sectors. Nationality groups had different age, gender, family and mobility profiles. Whereas Turkish migration and, to a lesser extent, Moroccan migration was more centred on families, including many women and children who came to join male heads of family, Spanish migration was more balanced in terms of gender, and Italian, Tunisian and Algerian ­migration was characterized by single men, many of whom left the city relatively quickly or, alternatively, married Belgian women and settled permanently. To a great extent, these different profiles explain the differences in structural and social integration in the labour and housing markets, as well as at the workplace and in the neighbourhoods, as will come to the fore in the ­following chapters.

Notes   1. Yearly reports of the city of Ghent, 1965–1970.  2. During the 1960s, all Western European countries saw their strong economic ­expansion  coinciding with a fast growing employment in the tertiary sector (Heijke 1995, 12).   3. The numbers are for registered unemployment. Yearly reports of the city of Ghent.   4. In fact, it was common for young, single immigrant workers to behave in the same way as other adolescents, moving from one job to another, earning money just to finance their

Postwar Migration to the City of Ghent  •  37

  5.

  6.   7.

  8.   9. 10. 11.

12.

13.

14. 15.

leisure-time activities, or saving up to buy specific material goods, such as a motorbike, a car, fancy clothes, etc.; immigrant housewives and immigrant peasants often also had the same goals as local housewives and peasants. Lambrechts shows how this solution was preferred over the engagement of female workers, who formed a major labour reserve but were initially not seen as suitable to take up jobs that were traditionally ‘male’, as employers did not want to break through the gender barriers that characterized the labour market. She indicates however that this hesitation towards female employment changed when the ‘social cost’ of foreign labour migration became clear, towards the beginning of the 1970s, and the ‘National Committee for Economic Expansion’ even tried to slow down the immigration of foreign workers by recommending female employment. Where at the height of building activity, in 1966, some 5,500 to 6,500 mostly temporary new jobs had been created (Voordeckers 1966, 104). The textile industry had been confronted with a continuous outward stream of local labourers during the whole postwar period, but up until then had managed to make do, as at the same time, job opportunities in textiles kept on shrinking, and as there was the possibility of recruiting workers in the surrounding countryside, where more and more farmer families could no longer make a living off the land alone. However, the economic boom of the mid-1960s, which created a sudden rise in the demand for textiles, caused a much more acute labour shortage that had to be solved as quickly as possible if the machines were to keep running to their full potential. There was a shortage of male as well as female textile workers; the professions that had the most trouble finding workers were spinner, spinster, bobiner, bobinster and weaver (De Bruyn 1966, 8). Yearly reports of the city of Ghent, 1968–1975. Yearly reports of the city of Ghent. The data concerns the territory of the city of Ghent as it existed before the fusion in 1977. Yearly reports of the city of Ghent. Over the course of the 1960s, between 1,200 to 1,800 dwellings were registered as uninhabited. By 1970, vacancy had become an even more urgent problem, with 4,299 or nearly 7 per cent of the total number of private houses registered as uninhabited. Yearly reports of the city of Ghent, 1960–1970; National Population Census 1961. In 1970, 64 per cent of dwellings in Ghent were not inhabited by the owner, but by tenants. This percentage was much higher in the city centre, and probably even higher in reality, as the population census that produced these data did not take into account dwellings that were inhabited by ‘temporary inhabitants’ (Claeys and Van Hove 1987, 62). Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, consultation meeting of 8 February 1979. In December 1980, 748 people were registered as candidate-tenants at the Ghent Housing Association in: ibidem, document s.d. ‘Algemene beschouwing aangaande de huisvestingsproblematiek in Gent’. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, ‘Dossier huisvesting’, 1995, 1.2.2.2 De stadswoningen. In 1982, for a new social apartment with one or two bedrooms, one had to pay between 7,000 and 9,000 francs per month (costs included), whereas on the private rental market, one could rent a small house or a flat with two or three bedrooms for 3,000 to 4,500 francs per month. City Archive Ghent (De Zwarte Doos), series XXXIII sociale affairs and housing, deposit 1994, number 72 decisions of the municipal executive and city council, 1980–1981, letter from G. Casteleyn, director of the third direction, to D. Van de Wynckel, city councillor of social affairs, 27/11/1981. Amsab-ISG, archive Maurice

38  •  Parallel Lives Revisited

16. 17. 18.

19. 20. 21. 22.

23.

24. 25.

26. 27.

28. 29.

Maréchal, archive Woonfonds vzw, journal Woonfonds, 28/01/1982, 14/04/1982, 18/05/1982. This was a problem not only in Ghent, but everywhere in Belgium (Cumoli 2012, 123). In 1978, some 78 per cent of social dwellings in the city were apartments (Werkgroep Huisvesting Gent 1984, 8). Claeys and Van Hove 1987, 68; Werkgroep Huisvesting Gent 1984, 14; and AmsabISG, Archive Maurice Maréchal, Archive Woonfonds vzw, document s.d. ‘Renoveren, saneren en slopen ten bate van wie?’. In December 1980, e.g. you could rent a small house in the Voorjaarsstraat for 1,750 francs per month, and an apartment in the Rooigemlaan for 1,800 francs per month. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, journal Woonfonds 26/12/80. Lokale inburgerings- en integratiemonitor Gent, editie 2015, http://aps.vlaanderen.be/ lokaal/pdf/integratiemonitor/Gent.pdf, accessed 1 March 2016. NYC Planning, The Newest New Yorkers’, December 2013, http://www.nyc.gov/html/ dcp/pdf/census/nny2013/nny_2013.pdf, accessed 1 March 2016. 2014 American Community Survey, http://factfinder.census.gov/faces/tableservices/jsf/ pages/productview.xhtml?src=CF, accessed 1 March 2016. Their presence could be approximated because in a preliminary stage of the research, I made a reconstruction of the total Algerian population of the city of Ghent based on other data than the numbers from the yearly reports of the city (the individual alien files, the registers of arrival and the population registers of the city). From the registers of arrival, the population registers and the individual alien files, kept in the population archive of the city of Ghent, I took a sample of Italian immigrants arriving in the city of Ghent roughly between 1925 and 1985. Of my sample, those Italian immigrants who did not arrive as partners or children of immigrants who were present before them, who arrived in the city between 1925 and 1938 and who were still present when WWII broke out (n=25), 2 started work in a local factory, 2 were employed in the catering sector, 3 were hotel or pub owners, 4 were independent shopkeepers, 4 were craftsmen (1 carpenter and 3 mosaic workers) and 6 were hawkers. Further, 1 was an engineer, 1 a maid, 1 a seamstress and 1 had no profession. Yearly reports of the city of Ghent, 1945–1999. The reform of the Belgian nationality code in 1999 turned the acquisition of Belgian nationality into almost a formality, and greatly expanded the categories of people who automatically received Belgian nationality at birth. Many immigrants and their descendants took up this opportunity and became Belgian. Data for 2013 e.g. indicates the presence of 7,633 ‘Mediterraneans’ of foreign nationality in the city; including those who were born as non-Belgians or of whom one of the parents was born as non-­ Belgian, the number is more than four times as high, at 32,484. Lokale Inburgerings- en Integratiemonitor 2015, Gent, http://aps.vlaanderen.be/lokaal/pdf/integratiemonitor/ Gent.pdf, accessed 1 March 2016. Lokale inburgerings- en integratiemonitor 2015, Gent. Of my sample of Turkish independent immigrants arriving in Ghent between 1960 and 1980 for whom I have information as to their place of birth (n=170), 58 per cent were born in the town of Emirdağ or in one of the villages surrounding it (e.g. Karacalar, Kuruca, Yenikapı, etc.). See the quantitative appendix, chapter 1 (1). It was with the Royal Decree of 20 July 1967 that the regulation concerning the issuance of work permits to foreign labourers was reworked and tightened so as to limit as much

Postwar Migration to the City of Ghent  •  39

30. 31.

32. 33.

34. 35. 36.

37. 38. 39. 40. 41.

42. 43.

as possible the new arrival of foreign workers in an economy that was confronted with a sudden rise in unemployment. This legislative initiative followed an administrative practice that had slowly begun in February 1967, when a Ministerial Decree with the same content had been distributed to the local administrations. This break in immigration policy is often ignored in the literature dealing with postwar migration legislation, which often treats the period until 1975 as one uninterrupted whole (Martens 1976, 53–55; De Bruyn 1966, 22; von Oswald, Schönwälder and Sonnenberger 2003, 20; Khoojinian 2014). The same happened in the Netherlands, after the so-called ‘Roolvinkstop’ of 1 June 1968, a law that heavily restricted the admission to the country of foreign workers: from then on, many immigrants became ‘illegal’. When interpreting the drop in numbers of arrival in 1967, the role of this tightening immigration policy has to be taken into account, as the numbers I have gathered only reflect the ‘official’ arrival of immigrants, and do not include those immigrants who did not register. However, my numbers for 1967 already show a sudden decrease of arrivals from the month of January, long before the Royal Decree was promulgated in July, which would suggest that the crisis in the local labour market played at least as important a role in this decrease. Yearly reports of the city of Ghent, 1968–1974. Already in 1967, the company ‘Verenigde Gentse Spinnerijen’, employing 280 textile workers, had to close its books. One year later, the same thing happened to the Dierman Company, as well as the Belgian Textile Printing Company, both of whom had previously employed a number of foreign workers. Also in 1968, one of the branches of the newly created UCO Ltd, the spinning mill UCO-De Hemptinne, was shut down. This was followed by the failure of a number of other textile companies in 1970. Yearly reports of the city of Ghent 1967–1971. Yearly reports of the city of Ghent, 1967–1971. See the quantitative appendix, chapter 1 (3). Of a sample of 74 independent, professionally active female migrants arriving in Ghent between 1960 and 1980, some 54 per cent started working in domestic service, whereas some 35 per cent found employment in the entertainment industry. In this chapter, I treat those women working as entertainers as employees, even though they were officially registered as self-employed. The clubs that were mentioned by name are Cabaret Maxim’s, Cabaret Sexy, Cabaret Love, Cabaret Topless and Venus. Unfortunately, I found no information as to the numbers of regularizations that took place in Ghent. See the quantitative appendix, chapter 1 (4). Yearly reports of the city of Ghent, 1960–1971. There are no data for 1972. From 1973 onwards, the data are split by sex and no longer by age. For more information about the sample, see the quantitative appendix, chapter 1 (5). When interpreting these numbers, we have to take into account that they present aggregate data for non-Belgians of all age groups. For most nationalities, the gender imbalance would have been much larger if the data had only taken into account the adult population, which would have largely been made up of actual migrants and contained only a small percentage of people born in Ghent. For more information on the sample, see the quantitative appendix, chapter 1 (6). We did not withhold the data for Spanish and Italian immigrants, as the sample sizes were insignificant (below 10 individuals). For more information on the sample, see the quantitative appendix, chapter 1 (7).

40  •  Parallel Lives Revisited

44. For more information on the sample, see the quantitative appendix, chapter 1 (8). 45. Kamel, a first generation Algerian immigrant who arrived in Ghent in 1963, suggested that the language might also have had something to do with it: most Algerians spoke at least some French, and many were scared off by the prospect of having to learn yet another language if they were to stay in Ghent. Interview with Kamel. 46. For more information on the sample, see the quantitative appendix, chapter 1 (9).

Chapter 2

Integration Processes of Immigrants in the Local Labour Market and the Workplace 

As we have seen in the previous chapter, the arrival of Mediterranean immigrants in Ghent in the postwar period has to be situated within the framework of guest worker migration, triggered by the economic boom and the ensuing job creation. The large majority of immigrants arrived in Ghent either as independent labour migrants or as their dependent family members. Most of them started their professional activity as wage earners in the local labour market. Therefore, the workplace has to be seen as one of the main domains providing immigrants with the opportunity to develop meaningful social relations in the receiving society. In this chapter, we will analyse the position of immigrants in the labour market and in the workplace, paying special attention as to whether and to what extent the opportunity to develop such relations actually presented itself.

Mediterranean Immigrants as Wage Earners in the Local Labour Market Economically Active versus Inactive Immigrants: A Gender Bias Just because most immigrants came to Ghent in order to work, does not mean that all immigrants arriving in the city were (immediately) economically active. Many immigrants were registered by the administration as economically inactive. Apart from a number of students, these were mostly partner-migrants, coming to join their spouse who was already legally ­

42  •  Parallel Lives Revisited

resident. Most of these partner-migrants were women. Percentages of initially inactive immigrants were much higher among Turkish (36 per cent), Moroccan (28 per cent) and Tunisian (21 per cent) than among Italian (14 per cent), Algerian (12 per cent) and Spanish (10 per cent) immigrants.1 This coincides with the higher percentages of partner-migrants (Turkish, Moroccan) and students (Tunisian) respectively within these migrant populations. However, some of these ‘economically inactive’ immigrants sooner or later found work or set up a business. Their economic activities have very likely remained underreported in the administrative files, as they were not important for their residential status, which depended entirely on their spouse. For this and other reasons – for example the fact that many women worked in jobs that remained undeclared, such as cleaning in private households – most of the quantitative data I will present in this chapter cannot escape a gendered bias, privileging male over female employment. It is true however that many partner-migrants never went out working during their stay in Ghent. These economically inactive immigrants were mostly women who already had children at the time of arrival. When asked about their economic activity, these women explained they had no time to combine their tasks as a housewife with an out-house job.2 At the start of the 1960s, having children was indeed a major reason to stay at home, also for Belgian women. Apart from the difficulties of combining family life with working outdoors, women, and especially married women, still encountered a lot of discrimination on the labour market.3 However, during the period under study, most discriminatory laws and practices that had reigned in the immediate postwar period were being abandoned, and measures were taken to encourage women to go out to work. The ‘working woman’ thus gradually came to be seen in a more positive light (Osaer 1991, 386–96). This was not always the case though, and also in many immigrant families, women’s work remained negatively evaluated. In her 1978 study of Moroccan families in the Netherlands, Lotty van den Berg-Eldering found that besides the presence of small children in the family, a negative attitude of the male head of the family concerning female labour also prevented many women from going out to work. The opposition against working outdoors existed in all the families that were part of her study, but was stronger in more ‘conservative’ families coming from the Moroccan countryside. Those women in her sample group who did go out to work came from the big cities, from what she called more ‘modern’ families (van den Berg-Eldering 1978, 122). Likewise, in my interview sample, women who said they were forbidden by their husbands to go out to work (or whose husbands told the interviewer they had forbidden their wives to work) mostly came from more conservative rural areas such as the Moroccan Rif or Turkish central Anatolia, and referred to their family’s expectations concerning women’s roles as mothers and housekeepers.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  43

A third reason why immigrant women are reported to have had lower rates of economic activity than their men has to be situated at the level of legislation. In a large-scale survey among Turkish women in Germany in 1980, the most important explanation women gave for not having a job was their lack of a work permit. In the wake of the halt in recruitment of 1973, the German authorities had made it very difficult for newly arriving immigrants to obtain such a permit (Erdem and Mattes 2003, 176). From the point of view of the Belgian authorities, the arrival of partners, de facto female partners, was never meant to bring in new labourers, but rather to ensure the reproduction of the labour force in the long term (Ouali 2004, 26–27). Immigrant women were first and foremost seen as wives and mothers. Therefore, their right to residency was inextricably linked to that of their husband, and in times of economic crisis, their right to work came under threat. Still, even though the introduction of restrictions on the granting of labour permits to partner migrants was discussed in Belgium, it was never implemented and is therefore less pertinent for the analysis of our case study. A lack of opportunities can be considered a fourth reason why immigrant women were less economically active. This was very much the case for example in mining regions, where there were hardly any jobs available for them,4 but much less so in Ghent. With its diversified economic fabric and its textile industry, traditionally employing high percentages of women, female workers were in high demand. Still, in those years, there was a large labour reserve among local women (not only in Ghent but in the whole of Belgium), so that employers did not need to recruit foreign women to the same extent they had to recruit foreign men (Lambrechts 1979, 55, 60–62). This is yet another factor that helps to explain the lower degree of economic activity for immigrant women in Ghent. Clearly, the division between economically active and inactive immigrants was very much a division between immigrant men and women. Here, we can find one of the most important gender-related distinctions in the experiences of immigrants from the postwar period in the receiving society, not only in Ghent but everywhere in Western Europe. Byron and Condon, in their study of the integration processes of Caribbean immigrants in Great Britain and France, have noticed how ‘a broad distinction between the life courses of women and men’ has been ‘based primarily upon the relationship each group has maintained with the labour market’ (Byron and Condon 2004, 11–12). We need to keep this distinction in mind for the analysis in this and the following chapter, as it will be biased towards male experiences – with the exception of those paragraphs dealing with domestic service, an overwhelmingly female economic sector (see chapter 1). Chapters 4 and 5, dealing with the position of Mediterranean immigrants in the housing market and their

44  •  Parallel Lives Revisited

social contacts in the neighbourhood, on the other hand, will be more biased towards women’s experiences. A Large Majority of Manual Workers Postwar immigrants in Western Europe generally integrated into the economic fabric as wage labourers rather than as self-employed, and as manual labourers rather than as non-manual employees. Data on France, Switzerland, Germany and Britain in the 1960s, as well as data on France, Germany and the US in the 1970s, show an overwhelming majority of manual wage labourers among the immigrant newcomers who were economically active (Castles and Kosack 1985, 62–87; Piore 1979, 18). Also in Ghent, manual wage labour was by far the most important path for Mediterranean immigrants, accounting for around 85 per cent of Spanish and Italian, 89 per cent of Algerian and Tunisian, 91 per cent of Moroccan and 96 per cent of Turkish immigrants in our sample.5 Among those few immigrant employees beginning their career in Ghent as white-collar workers, we find ­secretaries, engineers, social workers and laboratory technicians, as well as clergymen and teachers of religion, language and culture, sent by the immigrants’ home countries to provide for their specific religious and cultural needs.6 Those few immigrants that started out as self-employed were mostly active in the entertainment sector, as freelance dancers, singers and musicians, or were small business owners, most of them inn-keepers or shopkeepers. The majority of these initially self-employed immigrants were women.7 But again, the ­self-employed and white-collar employees made up very low percentages of the total of economically active immigrants arriving in Ghent. From this point onwards, the analysis will be focused on those that started work as blue collar wage earners.8

The Concentration of Immigrant Workers in Specific Economic Sectors, Jobs and Companies As is commonly known, postwar immigrant labourers in Western Europe generally performed work that was mainly unskilled, unhealthy and/or badly paid, and that was situated in the secondary sector of the labour market. They have often been identified as replacement workers, taking up those jobs that were rejected by indigenous labourers, even though a non-negligible number of them supplemented rather than replaced local workers (Bade 2003, 2­ 29–31). Still, this does not change the fact that the majority of Mediterranean immigrants took up those jobs that were the least ­attractive. Even though such jobs can be found in nearly all sectors of the labour market, they do tend to be concentrated in specific economic sectors at specific moments in time.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  45

In Ghent, there were three economic sectors that attracted the large majority of immigrant wage earners at their arrival: the textile industry, the construction industry and the domestic service sector (see chapter 1). As time went by, immigrants found jobs in a wider range of economic sectors, such as catering, metal works, cardboard, chemistry and food processing. Still, throughout the period under study, textiles and construction remained the main employers for immigrant labourers in the city.9 Only domestic services were surpassed in importance by the catering sector, where mostly Italian immigrants were employed.10 Men and women were distributed differently over these sectors. Whereas immigrant men worked mostly in construction, women were largely concentrated in domestic services. Both men and women worked in the textile sector, albeit in different roles. From the mid1970s on, many Turkish women also started work in Diepvries Breskens Ltd, a Dutch fish processing company which had begun to recruit in the Ghent labour market.11 Immigrant labourers in Ghent followed the same male-female employment patterns that characterized immigrant labour and manual wage labour in general.12 An exception to this is the fact that in Ghent some immigrant men also started work in domestic services, a more typically female sector of employment. Most of these men were employed together with their wives, as part of a couple of housekeepers.13

Figure 2.1  Turkish woman working in the Dutch fish factory Diepvries Breskens, 1980. Amsab-ISG © Lieve Colruyt

46  •  Parallel Lives Revisited

The uneven distribution of immigrant workers in the Ghent labour market, however, was not only related to their concentration in specific economic sectors doing specific kinds of jobs. It was also caused by their uneven dispersion over different companies and employers. Whereas some companies employed large numbers and/or high percentages of immigrant workers, others – within the same sector, offering the same kind of jobs – employed hardly any or none. Out of thirty-one textile companies in Ghent in 1967, the highest percentage of immigrant workers could be found in a small company called Texico, where four out of seven workers were foreigners. The highest number of immigrant workers was to be found in textile giant UCO Ltd. This company employed 245 non-Belgians, accounting for around 10 per cent of the workforce (totalling 2,530 employees). At the other end of the spectrum, there were large companies that employed hardly any immigrant workers, such as Hanus or Fabelta Ltd, where only around 1 per cent of the workforce was foreign (9 out of 1,035 and 16 out of 1,415 respectively) (De Bruyn 1966, 7, 13). Here, rather than the functioning of dual labour markets, it is the personal and professional attitude of employers towards immigrant workers that explains these differences. In the region around Ghent, employing immigrant workers was not the monopoly of a handful of big companies. Mediterranean immigrants living in Ghent were hired by a broad range of employers, from private households and small- and medium-sized businesses to large enterprises. Over the period 1965–1967, for example, the local employment office counted fifty-three different employers for Turkish immigrants alone, not including private households (Bernabé 1968, 54; De Bruyn 1966, 7, 13). Most of these employers were local. However, immigrants who had settled in Ghent also worked for companies and private employers that were established elsewhere: in the surrounding villages, or further away in small towns such as Zele, Laarne and Lokeren, or in bigger cities like Brussels and Antwerp.14 During the mid-1960s, a significant number of immigrant labourers worked for foreign or multinational firms that were involved in the construction works around the Canal Zone. Most of these were temporary labourers, being hired on a day-to-day basis until the works were completed. Some however were part of a mobile workforce, travelling from one construction site to another together with the companies that hired them. These immigrants were skilled professionals, such as carpenters, ironworkers, welders, mechanics, etc.15 Only some of them ended up staying in Ghent for a longer period of time, abandoning their position and seeking work that was locally based.16 Apart from these, only a small minority of immigrant workers living in Ghent worked for non-Belgian employers. These employers however were not large-scale firms, but small-scale family businesses, run by people who

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  47

had migrated to Ghent themselves. Due to the historical presence of several Italian entrepreneurs in the city (see chapter 1), many of these immigrant employers were Italians who hired a mostly Italian workforce. This phenomenon was so important that from 1967 onwards, when the works in the Canal Zone (that had attracted a lot of Italian labourers) were finished, the majority of Italian newcomers went to work for a local Italian employer. Apart from that, there were also Spanish employers hiring mostly Spanish workers. Only later, from the 1980s onwards, would Turkish entrepreneurs in the city create businesses that were big enough to hire personnel, most of whom were also Turkish (De Bock 2014). The large majority of immigrant workers worked for Belgian-owned companies and Belgian employers. At first sight, this might not seem very surprising, as there were of course many more Belgian than foreign employers in and around Ghent. However, in the period under study, it was not self-evident for Belgian employers to hire immigrant workers. Studies carried out in Ghent in the late 1960s show that many employers were rather negative about foreign workers, and only turned to immigrant labour when there were absolutely no Belgian candidates available (Bernabé 1968, 63; Vanoutryve 1968, 58; De Bruyn 1966, 9). One of the reasons for this hesitation was certainly the extra effort that hiring immigrants entailed. Employers had to apply for a special work permit, they were obliged to provide housing, there were language problems, etc. However, once employers had experienced working with foreign labourers, they often became more nuanced in their preferences, up to the point where some of them actually began to prefer immigrant workers over Belgian candidates. The former were perceived as having retained a certain work ethos their Belgian workforce no longer possessed. One employer, when asked why he employed Turkish labourers, explained how the Belgian workers in his firm were abusing the system that guaranteed them one week’s pay in case of illness, whereas his Turkish employees did not take part in this (Bernabé 1968, 62). Another employer, when asked about his Spanish workforce, told the interviewer that his Spanish workers never stole anything from the company, unlike the Belgian ones, for whom this was a current practice (Vanoutryve 1968, 58). Yet another one reported his Turkish female workers refusing to leave their workplace even when they were ill – which, from his point of view, was seen as a positive work attitude (Bernabé 1968, 71). Whether this change of preference was a general occurrence, and to what extent it was actually put into practice in the hiring (and firing) policies of local employers, remains to be seen. It is common knowledge that immigrant workers are generally discriminated against rather than privileged (Pitti 2005, 69; Heijke 1995, 46–51). Further, even in those cases where the director or personnel manager of a firm preferred foreign workers, this feeling was not

48  •  Parallel Lives Revisited

necessarily shared by everyone in the company. In several cases, when an employer took measures that appeared to favour foreign workers over their Belgian colleagues, the latter, with the support of the local unions, heavily protested against such ‘preferential treatment’. Such attitudes and their consequences for the construction of social relations in the workplace will be further discussed in chapter 3.

Immigrant Wage Earners’ Professional Careers: Fragmentary and Highly Mobile Some of the immigrants interviewed for this book, who have been professionally active as wage labourers, worked for the same employer for the whole or most of their professional life in Ghent. Most of these immigrants had been employed by textile giant UCO Ltd. Even though UCO encountered serious economic problems from the 1970s onwards, every time one of its factories was closed down, the company managed to redistribute most of its labourers over its remaining factories, rather than lay them off.17 For example, Charif, an Algerian immigrant who worked for UCO Ltd from the moment he arrived in Ghent in 1965 until he took his pension in 2003, started work in UCO Rooigem, was moved to UCO Texas, and ended his career in UCO Braun, the last remaining UCO factory.18 Of the three sectors that attracted the majority of foreign labour in Ghent, only the textile industry, and especially UCO Ltd, offered immigrants a real chance for an uninterrupted long-term career. Still, of my sample of Mediterranean immigrants who started work at UCO between 1960 and 1980, hardly anyone worked for this company for fifteen years or more before taking their pension.19 Others who managed to complete a more or less uninterrupted career were those few immigrants who had made their way into the primary sector of the labour market, and worked for companies such as Volvo and EBES, where long-term professional stability was relatively common.20 The majority of immigrant workers however had professional trajectories of a more fragmentary nature. During their economically active life in Ghent, immigrants generally worked for different employers, doing different kinds of jobs, for short or long stretches of time, often with periods of unemployment in between. A typical ‘interrupted trajectory’ is that of Ali G., a Turkish immigrant from Emirdağ who arrived in Ghent in 1974 with a contract to work for a building company, Maes Ltd. After only one year, he was laid off. He then started work with a company called Aquavia, laying cables for water, gas and electricity, where he stayed for eight years, until he was made redundant there as well. For the following three years, Ali was on the dole, looking for but not finding work. Then he heard about a

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  49

­ osition in the Fabelta ­textile company in Ninove, a small town some 40 km p from Ghent. He applied and got the job. Five years later, he got fired again. Luckily, he was able to quickly start in another Fabelta factory, this time in Zwijnaarde, only 5 km from the centre of Ghent, where he was still working at the time of the interview in 1994.21 This is of course only one story, but it seems representative for the experience of many immigrant workers in Ghent – especially men. Professional Mobility Postwar labour migrants to Europe in general were indeed a highly mobile workforce. They frequently moved from one job to another, between employers, sectors, etc.22 Of the same sample of Mediterranean immigrants who worked for UCO Ltd (see above), more than 30 per cent of Moroccans, Tunisians and Turks, 50 per cent of Spanish, and more than 65 per cent of Italians and Algerians stayed there for less than one year. To a large extent, this high degree of job mobility was inherent to the kind of work immigrants did, and as such was forced upon them. The jobs they were hired to do were largely unskilled or semi-skilled and could therefore be done by any ablebodied worker with a minimum of training or without any training at all. This meant that labourers could easily be laid off when no longer needed, as long as there were enough others willing to take their place when new labour shortages arose (Castles and Kosack 1985, 79–86; Salah 1973, 45). However, immigrant workers were not just passive pawns, willingly moved around by employers’ decisions and economic circumstances. Just like local labourers, they were active agents who sought to better their situation and find work that responded as closely as possible to their own goals and aspirations.23 For many of them, especially in the first years of their stay in Western Europe, this meant work that would allow them to save up as much money as possible, so that they could provide for the ones they had left behind and would soon be able to return back home to construct a better life for themselves and their families (Bade 2003, 229). As target-earners (see chapter 1), these immigrants tended to judge the desirability of a job first and foremost on its financial aspect. If more money could be made elsewhere, immigrant workers generally did not hesitate to change jobs, even if this meant they had to breach their contract (Salah 1973, 44; Yücel 1987, 127; von Oswald, Schönwälder and Sonnenberger 2003, 21, 31; De Muynck 1972, 62–63). Their limited affinity with the company they worked for also added to the ease with which they changed employers.24 This was a characteristic they shared with other target earners. In fact, research into the turnover of immigrant workers in four local companies in Ghent (two in the metal industry and two textile factories) showed that local workers with the same profile as

50  •  Parallel Lives Revisited

immigrant workers (with similar socio-demographic characteristics, doing the same kind of job, in the same shift system) had a very similar turnover rate to that of their foreign colleagues (De Muynck 1972, 82–83). As opposed to local workers, however, Mediterranean immigrants were also confronted with legal restrictions in their professional careers. On the one hand, these limited their access to the labour market, thus excluding them from many jobs that could have allowed them to be even more mobile. On the other hand, they forced them into a system of temporality – at least for the first years of their stay – and thus of heightened job mobility. In Belgium, during the period under study, the majority of Mediterranean immigrants were subject to a system of work permits. These not only determined their right to reside in the country – that is, if this right was not dependent on that of their spouse or parent – but also severely limited their right to work wherever they saw fit (Martens 1976, 36; Romero 1991, 15). After an ‘inquiry into the state of the labour market’, carried out by the local employment office (RVA),25 immigrant workers whose presence was considered valuable could be granted a ‘B-permit’ by the Ministry of Labour. This was a work permit for foreigners that was valid for one, or at most two years, and that restricted the immigrant’s access to the labour market to the specific job, professional category, employer and workplace he or she had arrived to work in. As of 1965, after one year, the B-permit became valid for all employers within a given sector.26 Up until 1962, most immigrants had to resubmit their demand for a B-permit every year, and this for a period of ten years.27 From 1962 onwards, this period of yearly renewal changed to every five years for those living on their own and three for those living with their families as well as for Spanish, Moroccan and Turkish citizens.28 These regulations thwarted the regular access of immigrant workers to the labour market, and undoubtedly had an important impact on the fragmentation of their careers and their high degree of labour mobility. However, most immigrant labourers experienced these limitations only during the first years after their arrival. Later, they received a work permit A, valid for an unlimited amount of time and allowing the permit holder to look for a job in all sectors of the labour market. In addition, from 1964 onwards, all EEC citizens (for the purposes of our case study, only the Italians) were immediately granted a special EEC work permit, equal to an A-permit,29 and from 1968 onwards, they were no longer subjected to the system of work permits (Romero 1991, 107). Whereas the professional mobility of immigrant workers in Ghent was rather high over the course of the 1960s, it significantly decreased throughout the period under study. Two main reasons for this can be put to the fore: the changing characteristics of the immigrant population itself, and the ­shifting economic and legal opportunity structures.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  51

On the one hand, much can be explained by pointing at the changes in the composition of the immigrant population, in which settled families came to make up an ever growing part. This was especially the case after the 1974 migration stop, which put an official end to the arrival of new independent labour migrants, those that were the most mobile (Villey 1981, 47). Indeed, immigrant workers who establish a dependent family in the host society tend to become a much more stable workforce, not only geographically but also professionally (Salah 1973, 43). They may remain ‘target earners’ in thought, saving up to go back to their home country one day, but the presence of their family prevents them from taking financial risks such as quitting a badly paid but stable job for a more profitable, short-term contract. The care for this family requires a constant income, and one well above that which is needed to provide for a single worker or a working couple. Employers and governments knew this very well. Already from the early 1960s onwards, the Belgian government greatly encouraged family reunification by installing provincial reception centres for immigrant families and by intervening in the transportation costs of spouses and children (Martens 1976, 131–33). On the other hand, immigrant workers’ diminishing mobility was equally influenced by changes in the local economic opportunity structure. The extraordinary demand for labour in the mid-1960s, which had made it relatively easy for immigrant workers to change jobs, did not last. Already in 1966–1967, a short-term economic recession caused a drop in the number of available jobs. However, this recession was only temporary, and during the late 1960s and early 1970s, the demand for labour rose again, which allowed new immigrants to come and work in Ghent and move about from one job to another. The following recession, which started in 1974, had a much more permanent impact on their economic integration. In Ghent, as elsewhere in Western Europe, the consequences of the oil crisis forced many companies to close their doors or to drastically reduce their workforce, causing thousands of people to lose their jobs. The post-1974 period, characterized by a non-stop rise in unemployment, did not lend itself to a high degree of professional mobility. Voluntarily moving from one job to another became difficult, especially for immigrant workers, as there simply were fewer jobs available and losing their job had more severe consequences for them than it had for locals, entailing for example the risk of losing their residence permit (Bollen 1978, 66–68). Further, forced mobility became less prevalent in this period. The ‘migration stops’ that had been put in place by most Western European governments, in the face of rising unemployment, pushed employers to manage their immigrant workforce more carefully. As the continuous inflow of easy-to-hireand-fire labourers seemed to have come to an end, those employers who

52  •  Parallel Lives Revisited

Chart 2.1  Employment episodes at UCO Rooigem, 1963–1991

had a structural need for immigrant labour began to put in more effort to keep them (Hollifield 1992, 80). The fact that, at least in the textile industry, the days of extreme short-term employment were over, is illustrated by the figure above, which represents the length of employment of immigrant workers at one of the UCO factories. On the horizontal axis, we can see the start date of each individual employment episode; on the vertical axis, the total duration (in number of days) of these employment episodes. Whereas the period before the migration stop is clearly characterized by a continuous coming and going of immigrant workers, after 1974, short-term employments become the exception rather than the rule. Short-Term Careers As explained above, the post-1974 crisis did not only cause a decrease in job mobility. First and foremost, it caused a serious and continuous rise of unemployment, which hit immigrants harder than the rest of the population, especially in the first crisis years.30 In 1975, Belgian as well as foreign unemployment skyrocketed. At the national level, the number of foreigners entitled to unemployment benefits almost doubled. In Ghent, it rose even more spectacularly, from 70 in 1974 to 453 in 1975, after which it increased more slowly, reaching 847 in 1980. Over the period 1974–1980, about one quarter of unemployed foreigners in Ghent were women.31 Turkish immigrants seem to have been more affected by unemployment than immigrants

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  53

of other nationalities. Their share among the foreign unemployed rose from 39 per cent in 1974 to 54 per cent in 1980, while their share in the total foreign population only grew from 37 per cent to 40 per cent. Italian immigrants were also slightly overrepresented: 6 per cent as compared to 4 per cent of the total foreign population in 1974, and 5 per cent as compared to 3 per cent in 1980. This was different or at least less pronounced for Tunisian, Moroccan and especially for Spanish immigrants, the latter making up 0 per cent as compared to 4 per cent in 1974, and 2 per cent as compared to 3 per cent in 1980.32 Unemployment cut short the careers of many immigrants who had previously been active as wage earners. Some were made redundant long before they reached pensionable age, and never managed to find another job.33 One of them was Ali D., a Turkish immigrant who arrived in Belgium in 1973 and could not find work after he became unemployed in 1979, at age thirtynine. This excerpt from his interview also shows the fragmentary nature of his short career: First I worked in Zele. We dug for all kinds of cables and repaired cables when the roads were broken up. … I worked there for three years. Then I was on the dole for six months. Then I found work in Ghent in the same sector. I worked there for two and a half years. When that work was finished, I had to go on the dole again. Then I found a job with a company that was carrying out restoration works at the railway station Brussels North. I didn’t want to keep changing jobs, you know, but every time the works were finished, they sent me to the unemployment office. In Brussels, the job only lasted for four months, and after that, everyone was fired. After a while on the dole, I found work with a building firm in Ghent. I worked there for three months and since then, I am unemployed. … That was in 1979.34

Many immigrant workers were forced to take up early retirement when their company went into liquidation.35 Some of them saw this as a blessing in disguise. Others did not, for financial reasons, but also because they had difficulties adapting to a life without work, like Ahmed B., an Algerian immigrant who went into pre-pension when the UCO factory where he worked closed down and was very bitter about this. After twenty years at UCO Ltd without a day of sick leave or unemployment, he experienced his forced prepension as a blemish on his spotless trajectory.36 Economic inactivity had an important impact on the social integration of these immigrants in the receiving society. It often caused serious psychological and physical (psychosomatic) problems, especially when the period of unemployment dragged on and was not filled with other activities. Immigrant men often had a harder time dealing with unemployment than immigrant women. Women who stayed at home seem to have succeeded

54  •  Parallel Lives Revisited

in ­overcoming their initial loneliness by creating social networks with kin, friends and neighbours, whom they visited frequently. Men on the other hand were more confronted with feelings of worthlessness, and often had to create a brand new social network years after their arrival in the receiving society. Most of them eventually did succeed in doing so, spending their free time in cafés, clubhouses, mosques, etc. There, they got to know mostly co-ethnics and as such managed to extend their ethnic social networks. Unemployment however did mean they lost many if not most of their ties to locals, as the workplace generally had been their most important locus for interethnic contact (De Bock 2013, 208–11; Byron and Condon 2004, 103; Bouras 2012, 216). It is hard to calculate what percentage of immigrant workers had a continuous, long-term career and what percentage had a heavily fragmented career or only worked for a short time. However, based on the evidence we have, the latter situation seems to have prevailed. This means that for most immigrant workers, the time span for developing social relations based in the workplace was limited, and opportunities to create that special bond that can develop between long-term colleagues scarce. This has to be kept in mind when we discuss the nature of workplace-related social contacts in chapter 3.

Spatial and Temporal Characteristics of Immigrants’ Jobs Apart from the characteristics of immigrants’ professional careers – often highly mobile both from a temporal and a spatial perspective – the features of the actual work they did are also important indicators for us to understand the (social) integration of immigrant wage earners in the workplace. In what follows, we will discuss the social, temporal and spatial specificities of work in the three most important economic sectors for immigrant employment during the period under study. Construction Most jobs in construction were characterized by a high degree of spatial mobility, as the workplace itself regularly moved (Martini 2009, 597–98). The spatial dimensions of working in construction further depended on the size of the company and the geographical scope of its activities. Small, local firms generally moved over shorter distances than large, international firms. If the construction site was located near the town or city where the workers lived, they had to make their own way to work. If it was far away, the employer often arranged for a van or bus to pick them up at a central meeting

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  55

point. For such short-term displacements, workers generally returned home every evening.37 For work that was too far away to commute, workers tended to stay the week and only came home at weekends. Some even stayed away for several weeks or months at a time (Deslé 1990, 443).38 In this case, workers were housed together in temporary accommodation, near or even at the construction site. In Ghent, a group of Italian immigrants working for one of the big building companies in the Canal Zone between 1965 and 1967 were living in what is described as a ‘barracks camp’ on the building site of the SIDMAR steel factory.39 Such a setup was typical for immigrants working on large building sites.40 Not taking into account the sometimes long-term sojourns away from home, construction work came close to the ‘ideal’ timeframe of a nine-to-five working day and five or five-and-a-half day working week, or at least it did in theory (Hilton 1968, 164–65). In practice, however, working days and weeks could stretch out much longer, depending on the urgency of the work and on the inclination of the employer to stick to official regulations or not.41 Still, work in the construction industry in this era almost never included night shifts. Construction workers worked during the day – for practical reasons (the need for daylight) more than anything else. Most construction firms were small- or medium-sized, but even in big firms, employees tended to work in small task groups. Talking to each other during work was sometimes possible, but not always, depending on the physical distance between each worker and the kind of work performed. In excavation works, for example, the noise could be deafening, making any conversation impossible. On large building sites, people sometimes worked so far away from each other that they could not talk to anyone during working hours. Miloud, an Algerian immigrant who came to Ghent in the early 1960s, remembers not being able to converse during work hours at the building site where he was first employed, and having to wait until the working day was over before he could properly talk to his nearest colleague, who later became his best friend: So I was employed, and we had to climb to a great height. I have vertigo, so it took me a lot of time to climb up. We were at six metres, or even more, ten metres, sitting like that with a brush. As I have vertigo, I didn’t move too much, carefully, quietly … Out of the blue, I heard someone, facing me, who was singing these songs by Enrico Macias, an Algerian singer who came to France … So he was singing, and I asked ‘Hey, how do you know Enrico Macias?’ And he said ‘I’m from Algiers’. And I am from Oran, the second largest city in Algeria. And that’s how we got to talk. So I told him, ‘Listen, we can talk later on the ground, not right now, I have to finish working before I come down’, because I could not go down and climb up again, because of the vertigo. And that is how we met.42

56  •  Parallel Lives Revisited

However, as construction work is generally teamwork, it does require a certain level of communication between the members of a team (Martini 2008, 374; Duc 2000, 154–62), even though in big firms, on large building sites, this level of cooperation and thereby communication could sometimes be limited to a minimum (Duc 2000, 69, 122, 154–62).43 Workers also got the chance to spend some time together during breaks, smoking cigarettes or having lunch. Further, if they travelled to work together, social contact was possible during that time. Contact with superiors on the other hand seems to have remained limited to their immediate supervisors, who gave orders. In small- and medium-sized companies, this would include the employer, whereas in larger companies, most workers only came into contact with their foreman (Martini 2008, 380). Whereas skilled workers were often tied to a specific job in a specific company and moved with the company from one building site to another, unskilled workers – especially in large companies – tended to be hired on a day-to-day basis, working on the construction site only until the specific job they were hired to do was finished. Most immigrant workers in construction were unskilled workers. For them, the social contacts they could establish at work were necessarily fleeting, as the composition of the workforce at any one site changed from one day to another. In general, there was also a large turnover in construction, which made the creation of long-lasting ties between colleagues a difficult feat (Hilton 1968, 17–18). Textiles In Ghent, working in textiles generally meant working for one of the big textile companies of the city,44 the most important of which was UCO Ltd. During the period under study, UCO had many different factories in Ghent, spread out over the territory of the city. Most of them were situated in the city’s nineteenth-century belt, north of the historical centre (see chapter 4). For some time, UCO Ltd and Fabelta Ltd sent out company buses to pick up workers from the villages around the city, going as far north as the Dutch border, 35 km away.45 Those living in Ghent had to make their own way to the factories. However, the relatively small size of the city permitted workers living in Ghent to arrive at their workplaces on foot, by bike or by public transport, rather than by car. It also meant that the time to commute remained short – crossing the entire city by bicycle takes about half an hour. Textile factories dictated a very specific work rhythm to their workforce. Work in the production line was done in shifts. During the 1960s, most factories seem to have worked with three different systems. Some employees (but rarely those on the production line) only worked during the day, others at night, and yet others worked in a rotating system, that changed weekly

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  57

between an early morning shift (from 8 am to 1.15 pm) and an afternoon shift (from 1.15 pm to 9.30 pm). Women only ever worked the daytime shift or the two-shifts system, as they were legally forbidden to work at night. Men worked in all systems.46 From the late 1960s onwards, and especially over the course of the 1970s, the two largest textile companies, UCO Ltd and Fabelta Ltd, experimented with other systems. Fabelta Ltd adopted a continuous system, whereby workers (only male workers) had to work the afternoon shift for seven days, got one day of rest, worked the early morning shift for seven days, followed by four days of rest, then did the night shift for seven days, followed by two days of rest.47 This very peculiar system greatly impacted on the lives of the workers. It was very different from the nine-to-five working day and five-day working week which by 1960 had become the national standard, especially in the ever expanding tertiary sector.48 UCO Ltd introduced a variation on this system, the semi-continuous system, where workers had to work three forty-eight-hour weeks in a row, after which they got to stay home for one week. Fixed night shifts were maintained throughout, and at one point, weekend shifts were introduced, where workers worked two twelvehour days at the weekend for a full week’s wages.49 The introduction of these systems was part of a programme of rationalization, aimed at increasing productivity. But rationalization schemes not

Figure 2.2  Immigrant man at work in the textile industry, UCO Gent, 1980. ©Lieve Colruyt

58  •  Parallel Lives Revisited

only changed the temporal dimensions of work in textiles, they also had an ­enormous impact on the nature of the work itself. At the beginning of the 1960s, work in textiles was still very much manual labour. When the threads on a spinning machine broke and became entangled, they had to be untied  and reattached by hand; raw materials and finished products had to be loaded onto carts by hand and manually wheeled around the factory floor; the dust produced by the cards had to be cleaned up by workers who had no other job than continuously sweeping the factory floors, etc.50 Over the course of the 1960s, and even more so in the following decades, a lot of the work came to be done by machines. Working in the textile industry became less of a manual job, but not an easier one. Under the supervision of so-called ‘labour analysts’, the pressure was continuously increased. Labourers had to work harder over the span of a shorter working day. Many employers put into place a system of piece-wages, resulting in fierce competition among the workers and an ever increasing work pace (De Wilde 1997).51 A job at the production line in textiles was never a sociable job. Even though task groups were relatively small in size (in the Rooigem factory in 1972, for example, task groups were made up of between three and twenty people)52 and workers generally worked with the same colleagues, in the same shifts, the work environment did not allow for much social contact during working time. Contact between men and women was limited, as task groups were generally separated along gender lines, and different tasks took place in different parts of the factory.53 In the spinning mill, each employee worked at the specific machines that were assigned to him or her, which were placed at some distance from one another. This distance increased over the course of the period under study, as each individual worker had to manage ever more machines (De Wilde 1997, 187). In the weaving mill, the noise of the looms was deafening, which made any conversation during work hours impossible (Bernabé 1968, 64–65). On the factory floor, there was no time for chitchat, only for conversation that was directly relevant to the work itself (explanations, orders and remarks). However, the breaks that interrupted a textile worker’s day did leave some time for social contact, especially the midday break, when workers had both time and space to sit down together and talk to each other.54 The communication between workers and their supervisors took place at a different level. In UCO Rooigem, for example, on their first day at work, newcomers were welcomed to the factory by the director or his substitute. They were informed about the rules and regulations by the personnel manager, and shown around the factory by an experienced worker, the moniteur, who was assigned this position for his or her knowledge of the work and the factory. Finally, they were introduced to their job by their direct superior,

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  59

the foreman of their team. The latter remained an important person in the employee’s working environment, with whom he or she came into contact on a daily basis (De Bruyn 1966). Personal contact with the other supervisors remained limited to specific moments in time such as payday, or when there were problems which could not be solved through the mediation of the foreman alone.55 Domestic Services Arguably the most important division among jobs in the domestic service sector is that between work as a live-in domestic and work as a domestic who is living out (Momsen 1999, 13–14). The question of living in or living out is not only relevant to the housing situation of the employee, but has farreaching consequences for the spatial, temporal and social dimensions of his or her daily life. As, in the period under study, only rich bourgeois and noble families in Ghent employed live-in domestics, these domestics were living in the most expensive neighbourhoods of the city (Vanoutryve 1968, 49). Many domestics also lived in the villages surrounding the city, especially those to the south, where rich families had large mansions and small castles.56 Domestic  workers who were living out generally resided in much poorer quarters, in a small house or apartment they could afford with the low wages they earned. Every workday, they had to make the trip from home to work. As Ghent is a small city, this was generally not too far away.57 Both types of domestics’ working space essentially consisted of their employer’s house. Part-time domestics, mostly cleaning ladies, sometimes had more than one employer, therefore working in more than one house. Live-in domestics were limited to the house where they lived. However, some tasks, such as ­shopping, also brought them to other places in the city during their work time. From a temporal perspective, the work of a live-in domestic did not resemble the standard nine-to-five working day or five-day working week. Live-in domestics worked every day, all day, except for Wednesday and Sunday afternoons, which were fixed (half ) days-off for domestic servants in the city of Ghent.58 Working days generally surpassed twelve hours. José Luis and Jesusa, who worked as live-in domestics in the 1960s and 1970s, described their working day as starting around 6 am and ending when they had finished cleaning up after dinner, which started at 7 pm.59 Their work, however, was not equally intense all day long. There were moments of relative rest and moments of hard work, such as the preparation and serving of the main meals. Still, even when they just had to ‘be present’, live-in domestics could not dispose of their own time, as they could

60  •  Parallel Lives Revisited

always be called upon by their employers. This complete subordination of one’s own time to that of one’s employers has been identified as exactly what accounts  for much of the bind of live-in domestic work (Anderson 1999,  124). Here, the d ­ ifference with domestics living out becomes very clear, the latter only working for any one employer for a couple of hours per day, often not even every day. Out-house domestics had flexible workinghours, determined in mutual agreement with their employer. They rarely worked late in the evening, or at weekends. Rather, they worked when their employers were out working themselves, so as not to disturb them during their time off.60 In the period under study, those employers who opted for live-in servants generally only hired one domestic worker at a time, or a couple of husband and wife. By then, the era where one family employed a whole hierarchy of live-in domestics had already come to an end (Pasleau and Schopp 2001, 243–44). This means that single live-in domestics, mostly women, worked alone, and only got to see their ‘colleagues’ (others who did the same kind of work) if they went out to find them on their time off. Loneliness and  isolation were regular problems, inherent to the particular situation of the domestic servant: living in a family without really being part of it, and working almost non-stop (König 2004, 106). Even part-time domestics could be confronted with these problems, as they also worked on their own.  However, they ­generally had more opportunities for social contact after working hours. On the other hand, the contact between domestic workers and their employers was often intense. Unlike in most other economic sectors, in domestic services the relationship between employer and employee  is a highly personalized  relationship, not in the least because of its location in the private domestic space of the employer (Narula  1999, 160). However, it is not an equal relationship: employer and employee are separated from  each other by a strict social division, based on class and often ethnic differences, in which gender also plays an ­important role.61 As the jobs described here represent the labour market experiences of the large majority of immigrant workers in Ghent, the different social, spatial and temporal characteristics of these jobs are important for an understanding of the social relations immigrants developed at work, and should be kept in mind for the analysis presented in the following chapter. However, they are not only valid for immigrants, but for all labourers working in these jobs and economic sectors. In order to assess not only the opportunities immigrants had to develop workplace-based relations in general, but also the relations that crossed ethnic boundaries in particular, we also need to know the extent to which these workplaces were ethnically mixed or segregated. This is what the last part of this chapter is about.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  61

Segregation of Immigrant Workers? Postwar labour migrants’ workplaces are often imagined as segregated workplaces, where immigrants worked together only with other immigrants, separated from local workers. In his excellent book, Measuring Immigrant Integration, Peter Reinsch describes his idea of immigrants’ workplaces in the postwar period as ‘an image … of oppressed immigrants populating factory production lines occasionally interspersed with an indigenous overseer’ (Reinsch 2001, 197). He admits however that, due to a lack of empirical data, this image is completely hypothetical. Also for Ghent, the data on Mediterranean immigrants’ workplaces in the 1960s and 1970s are scarce, but what is available shows a different picture. It indicates that immigrant wage earners generally worked in ‘mixed’ workplaces, together with a majority of Belgian colleagues. In 1972, Marc De Muynck, a Masters student who researched the position of immigrants on the factory floor in four local companies, noted that ‘it is a given fact that the foreigner is assigned the least-valued jobs, in mostly unhealthy circumstances, but it is also necessary to point to the fact that in each company, there are still Belgians doing the same jobs under the same circumstances’ (De Muynck 1972, 53). In this final part of the chapter, we will analyse the actual composition of the labour force in the companies and workplaces where immigrants worked, and how this changed throughout the period under study. The prevalence of mixed workplaces, as found by De Muynck, can be explained, firstly, as a simple consequence of numbers. Over the course of the period under study, immigrant workers made up a minority of the workforce in most companies, even in UCO Ltd, by far the most important employer of immigrant labourers in Ghent. In 1967, four years after the company had started recruiting foreign labourers, they made up less than 10 per cent of its workforce.62 However, as we have seen, some textile companies employed higher percentages of foreign workers than others. Even within UCO Ltd, there were significant differences between factories, often dependent on the kind of work that was carried out. In two of the spinning mills, for example, 25 per cent and 20 per cent respectively of the workforce were non-Belgian, whereas in two of the weaving mills, this was only around 1 per cent. There was also a considerable difference between male and female workers. Within UCO, immigrants made up 12.5 per cent of the male but only 1.5 per cent of the female employees. Comparing all textile factories, non-Belgians only accounted for a significant percentage (10 per cent) of the female workforce in one of UCO’s weaving plants, whereas they made up 10 per cent or more of the male workforce in eighteen out of the thirty-one factories.63 Taking this gender division into account, we cannot see a direct

62  •  Parallel Lives Revisited

link between the size of a company and the percentage of immigrant workers it employed: it was not the case that only small companies employed high percentages of foreign workers or vice versa. Immigrants often only accounted for a small percentage of the workforce not only in the textile industry but also in other economic sectors. In 1967, an enquiry among ten companies employing Turkish immigrants revealed  that foreign workers on average only made up around 10 per cent  of the total number of employees, with a minimum of 0.8 per cent and a maximum of 20 per cent (Bernabé 1968, 56). The interviews sketch a similar picture. Several immigrant workers remember being the only foreigner amongst a group of Belgians. This is especially the case for those that  worked for a small-scale, local employer,64 or in an economic sector that  did not attract many immigrant workers, where they were hired to carry out the dangerous or dirty tasks for which no local candidates could be found.65 Also, those  few immigrants who managed to acquire a job in the primary sector of the labour market generally only worked with Belgian colleagues.66 There were however a small number of companies where immigrant workers made up if not all, than at least the majority of the workforce. The same 1967 enquiry revealed a construction firm, where 81 per cent or 210 out of 260 workers were foreign. This firm was a foreign company, with a mobile workforce it had hired abroad (Bernabé 1968, 56; Vanoutryve 1968, 56). It was probably one of the firms carrying out the works in the Canal Zone, who left the city once these were finished (see chapter 1). However, such concentrations seem to have been the exception rather than the rule. Over the course of the 1970s, when more immigrant workers arrived in the city, and more immigrants who had originally arrived as partners or children of labour migrants accessed the labour market, in some companies the numbers and percentages of immigrant workers went up considerably. Still, they tended to remain a minority. In 1971, the percentages in four companies, selected precisely because they employed a large number of foreign workers, were 1.6 per cent, 9.28 per cent, 11.32 per cent and 21.3 per cent respectively (De Muynck 1972, 70–71). In 1972, in the twining department of the UCO-Rooigem factory, 24 per cent of the employees were foreign; in the spinning-department, there were no immigrant workers in the day shift (which existed solely of female workers), whereas five of the thirteen workers in the night shift (an all-male shift) were foreign.67 In 1975, in the UCO factory La Louisiana, three of the fifteen workers in an all-female task group in the spinning department were foreign.68 However, in 1978, in the spinning department of another company, only two of the thirty-nine workers were immigrant workers, whereas there were no immigrants at all in maintenance and in preparation.69 Further, there is

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  63

Figure 2.3  Turkish immigrant Faruk Köse with some of his Belgian and Turkish colleagues at UCO Ltd, 1965. Amsab-Institute of Social History, collection Köse

evidence of employers purposefully separating immigrant workers (at least those speaking the same language) from each other, so as to hinder strong group formation and to prevent casual conversation among co-workers, in order to speed up production.70 Therefore, the evidence allows us to safely state that throughout the 1970s, immigrants clearly still worked with mostly Belgian colleagues. On the other hand, immigrant workers had fewer occasions to work with other immigrants of different origins. Even though in general the immigrant workforce of Belgian employers was a diverse workforce, very often concentrations of immigrant workers of one specific nationality or region  of origin did occur. This was especially the case in small firms, or when there  were only  few immigrant workers in one company. However, also in larger companies with a more numerous immigrant workforce, most immigrants tended to come from the same country or even place of origin.71 Such ‘ethnic concentrations’ did not usually derive from some kind of ‘ethnic preference’ on the part of the employer (although this did sometimes play a role, especially when the employers were immigrants themselves – see below). Rather, they were largely unintended, the consequence of a general preference of both employers and employees for using informal networks to recruit or find a job (Granovetter 1995). Further, and especially from the

64  •  Parallel Lives Revisited

early 1970s onwards, concentrations of Turkish workers were more likely to occur than  concentrations of other nationalities simply because there were many more Turkish immigrants working in the city. What Andreas Wimmer (2004) pointed out for immigrants’ social networks in general also holds true for immigrants’ social networks at work, namely that ‘the relative homogeneity or heterogeneity of (these) social networks should be seen in relation to the demographic size of different ethnic-national groups, because the statistical chances of relating with somebody from a larger group are obviously higher than with a member of a small group or category’ (Wimmer 2004, 16). Whereas most workplaces in Ghent where immigrants worked were ‘mixed’, there is one exception to this rule. These are the small-scale companies run by immigrant employers, which we mentioned in part two of this chapter. Stereotypically, these employers recruited their labour force by mobilizing direct connections to the ‘ethnic community’ from which they emigrated, both in the sending and in the receiving country.72 Therefore, they generally employed if not only, than at least mostly immigrant workers of the same ethnic origin and thus created what we could call ‘ethnic workplaces’. One of these companies was a photo laboratory run by Spanish immigrant Edmundo Arnedo Martinez. In 1967, all of his staff was Spanish. Later, Belgian employees were hired too, but Spanish immigrants and their descendants still made up the majority of his workforce.73 Another example is the family business of Italian construction worker Antonio Bearzatto jr., who recruited a great deal of his staff directly from his native region of Friuli, through ties with old acquaintances, family and friends. Bearzatto only ever hired Belgians as apprentices, not as full-time, paid employees.74 More recent immigrant entrepreneurs such as Italian restaurant owners Giovanni Bombini and Mauro di Pilato also turned towards their region of origin in order to recruit their staff, although they did not turn away Italian immigrants from elsewhere who were already living in Ghent.75 However, some immigrant entrepreneurs were more inclined to (also) hire immigrant workers across ethnic boundaries. For the Italian restaurant business, these others seem to have been especially North Africans, with whom the Italian padroni could communicate in French and who could be ‘disguised’ as Italians, so as not to upset the local clientele (De Bock 2014). Other times, immigrant entrepreneurs purposefully hired Belgian staff, in order to ‘disguise’ their business as Belgian. To avoid negative reactions from his clientele in an era of heightened racism, Algerian restaurant owner Rachid decided to hire an exclusively Belgian staff when setting up his first restaurant in the early 1980s. Only after five years did he feel secure enough to also hire immigrant personnel.76 These different hiring policies created ‘ethnic’ and ‘mixed’ workplaces respectively.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  65

Conclusion In this chapter, we have seen that not all Mediterranean immigrants coming to Ghent in the period under study were immediately economically active. In particular, women who had arrived as partner-migrants only rarely started work straight away, and many of them never went out to work. Such lasting inactivity was related to their family status, to cultural preferences, and partly also to the economic opportunities available in the city. However, many other partner-migrants did go out to work, albeit often only for a short period, or for several short periods over time. Such interrupted or short-term careers were not specific to partnermigrants. Many labour migrants did not manage to maintain a steady job. Because of their initial inclination to change jobs frequently, but also because of the insecurity of the kind of jobs they worked in, immigrants’ careers were highly fragmented. The economic downturn of the second half of the 1970s and the ensuing increase in unemployment forced many immigrant workers out of a job long before they had reached pensionable age. A high degree of labour mobility and prolonged periods of economic inactivity therefore were common for these workers, and have to be taken into account when assessing the importance of the workplace for their social integration into the receiving society. Because of dual labour market mechanisms, but also because of individual employers’ personal preferences, immigrant labourers in Ghent were concentrated in specific economic sectors and even specific companies. The idea however that postwar labour migrants invariably ended up in ‘ethnically segregated workplaces’ has to be readjusted. In Ghent, such ‘ethnic workplaces’ did exist, but they were few, and only employed a small minority of the immigrant workforce. During the period under study, the large majority of Mediterranean immigrants only made up a small part of the total ­workforce of any one company, and worked with Belgian colleagues on a day-to-day basis. A mix of immigrants of different ethnic backgrounds however was more rare, and the majority of immigrants working for the same company tended to come from the same country or even region of origin. Apart from jobs in the domestic service sector, most jobs allowed for the development of workplace relations between colleagues, as they required people to cooperate, and provided time and space to talk during breaks. More than in other sectors, however, in domestic services, as well as in immigrant businesses, immigrant workers were in a position that privileged the development of personal relations with their employers. The following chapter will deal with these workplace relations, both across and within ethnic boundaries, between colleagues and with employers, in different kinds of workplaces.

66  •  Parallel Lives Revisited

Notes   1. See the quantitative appendix, chapter 2 (1).   2. The wives of first generation Turkish immigrants Celal and Gazi both had five children when they arrived in Belgium and said they did not really consider finding a job; the same was true for first generation Moroccan immigrant Yamina.   3. In Belgium, many women who wanted or were forced to give up work when they had children could not re-enter the labour market later on in their lives. Some companies had the policy of firing women once they got married, or once they got older and started to earn more, in order to replace them with younger, cheaper employees (Lambrechts 1979, 62–63).   4. After Brussels, it was the district of Ghent that saw the highest rate of female economic activity of all provincial capitals in Belgium. In 1961, 22.3 per cent of the female population of the district of Ghent was economically active; in Brussels, the numbers were higher, at 25.4 per cent, but in all other provincial capitals, women’s degree of economic activity was much lower, with 20.1 per cent in Liège, 19 per cent in Antwerp, 18 per cent in Bruges, 16.2 per cent in Namur, 15.9 per cent in Charleroi, 13.4 per cent in Hasselt and a mere 11.8 per cent in Arlon (Lambrechts 1979, 63–65).   5. See the quantitative appendix, chapter 2 (2).   6. In Ghent, most of these clergymen and teachers were Turkish, which seems logical as Turkish immigrants quickly became the most numerous nationality group in the city. Further, only Turkish teachers were officially engaged by the local authorities to teach Islam and Turkish language and culture in the city’s schools, in the framework of the ‘education in the own language and culture’. Clergymen and teachers of other nationalities were associated with immigrants’ self-organizations and religious institutions, or with the consular services of the home country. The Italian organization ‘Bel-Italia’, the Spanish ‘Asociación de padres de familia’, and the Algerian and Moroccan ‘Amicales’ offered language courses to immigrants’ children (occasionally also to Belgian spouses), with the financial help of their respective consular services (based on interviews with Clelia, Francisco and Mohamed Le., and informal conversation with Tahar). Moroccan Koran teachers and imams were sometimes brought in by the immigrants themselves, sometimes sent through the Moroccan authorities; Spanish and Italian priests were sent to Belgium as pastors to provide religious services in the immigrants’ own language.   7. Out of my sample of forty-four Mediterranean immigrants, whose first registered economic activity in the city was self-employed, twenty-eight, or 64 per cent, were women.   8. This changed from the mid-1970s onwards, when the self-employed became a much more important category among Mediterranean immigrants. For a further discussion of immigrant entrepreneurship in Ghent in the 1960s and 1970s, see De Bock 2014.   9. Even in the long run, this uneven division over economic sectors did not change very much, despite the enormous changes in the local economic fabric (see chapter 1). A research carried out in 1991 revealed that immigrant workers were still largely concentrated in textiles, and that those who were unemployed at the time of the research had been working in textiles, in the construction industry, the food processing industry and in private services (Rosvelds et al. 1993, 17, 36). 10. See the quantitative appendix, chapter 2 (3). 11. The company is mentioned as an important employer for Turkish women in an interview with Johan de Vreese. I also found two Turkish women starting work there in 1975 in the individual immigrant files. Further, for a folder with information on the working

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  67

12. 13. 14.

15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25.

26. 27. 28.

c­onditions of Turkish women in Diepvries Breskens, see personal archive Dr Ri de Ridder, folder ‘Diepvriesvisfabrieken’, 1980. Over the whole of Belgium, in 1970, 50 per cent of female immigrants worked in trade, catering, cleaning and personal services. In 1977, this percentage had gone up to 60 per cent, and female migrants made up 44.5 per cent of migrants employed in these sectors (Aerts 1979, 15). Of my sample of fourteen male immigrant workers arriving in Ghent between 1960 and 1980 and starting work as domestic servants, only 2 (14 per cent) were not part of a couple that had been recruited to work together in the same household. In particular, immigrants working as domestic servants often went to work in the villages surrounding the city, as that is where the big country houses of the local bourgeoisie were to be found. List of employers from database individual immigrant files; Document ‘Lijst van werkgevers’ (list of employers), s.d., found in the private archive of Maurice Maréchal, Amsab-ISG. See the quantitative appendix, chapter 2 (4). See the quantitative appendix, chapter 2 (5). Interview with Willem. Interview with Charif. This was the case for none of the Italian (n=3), 1 per cent of Algerian and Tunisian (n=97 and n=67), 7 per cent of Turkish (n=176), 9 per cent of Moroccan (n=128) and 17 per cent of Spanish (n=6) textile workers. In the interviews, those who spent more than twenty years working for UCO Ltd. are overrepresented, especially among Moroccan workers, as many of them resulted from projects in which immigrant workers were contacted on the basis of personnel lists of UCO Ltd. Among them are Amor, Moh, Abdelkader, Mohammed L., M’hamed and Mohamed T. Interviews with Mohamed J. and Mohamed H. (Volvo) and Kazim and Yakup (EBES). Interview with Ali G. See e.g. Khoojinian 2014, 300ff for an extensive analysis of the mobility of foreign miners in Europe in the postwar period. For an extensive discussion of the strategies used by immigrant labourers to reach their professional goals, see De Bock 2013, 181–83. The same limited affinity with their company has been noted for female labourers, who, during the period under study, were also characterized by a larger turnover than their local male counterparts (Lambrechts 1979, 224). During the economic boom in the years 1963–1965, the pre-emptive inquiry into the situation on the labour market was suspended, as the need for foreign labour was so urgent employers could not wait for the procedure to be completed before putting their foreign employees to work. Yearly Reports of the National Service for Work Exchange and Unemployment (RVA), 1963, 1966. Martens 1976, 54; Moulaert 1981, 16–19; Yearly Reports of the National Service for Work Exchange and Unemployment (RVA), 1960, 1965. Foreigners who were born in Belgium and could opt for Belgian nationality, and others ‘who could not be refused a permit’, such as political refugees, generally received an A-permit straight away (Martens 1976, 54). From 1962, Spanish, Moroccan, Turkish, British, Greek, Maltese and Swiss workers could apply for an A-permit after four years of regular employment or five years of regular residence in Belgium, and from 1967 on, after three years of regular employment. All others (except for EEC citizens, who were immediately granted a work permit A from 1964 onwards) had to have worked regularly in Belgium for six years, or for four years from 1967, or had to have resided regularly in Belgium for ten years; these periods came

68  •  Parallel Lives Revisited

down to respectively three and five years of regular work for the immigrant worker who lived in Belgium with his or her family (before 1967) (Martens 1976, 126). 29. Yearly Reports of the National Service for Work Exchange and Unemployment (RVA), 1964. 30. In 1975, non-Belgians made up 15 per cent of the country’s unemployed, whereas they represented only 10 per cent of its active population (Onthaalcomité voor Gastarbeiders van Brabant 1977, 9–10). In France, between 1973 and 1976, the employment of foreign labour decreased by 11 per cent; in Germany, over that same time span, it decreased by 26 per cent (Lebon 1984, 165; Körner 1984, 177). 31. National Employment Office (RVA), Collectie Economische en Sociale Studies – ­jaarlijkse telling van de volledig werklozen, 1974–1980, Archive RVA. 32. I found no detailed data for Algerian immigrants. National Employment Office (RVA), Collectie Economische en Sociale Studies – jaarlijkse telling van de volledig werklozen, 1974–1980, Archive RVA. 33. Some examples are: Hakan P., a Turkish immigrant who became unemployed in 1983 at age 38; Mehmet S., a Turkish immigrant who became unemployed in 1991 at age 46; and Omer K., a Turkish immigrant who became unemployed in 1984 at age 38. 34. Interview with Ali D. 35. Examples of immigrants who had to take their pre-pension are Celyn, a Turkish immigrant who worked for UCO for 17 years and was pre-pensioned at age 51; Mohammed L., a Moroccan immigrant who worked for UCO for 38 years and was pre-pensioned at age 53; Moh, a Moroccan immigrant who worked for UCO for 22 years and was prepensioned at age 53; and Amor, a Tunisian immigrant who worked for UCO for 22 years and was pre-pensioned at age 57. 36. Interview with Ahmed B. 37. In 1971, Mustafa and his Belgian and Turkish colleagues were picked up every morning at the Dampoort station to go to work in Mechelen, some 100 km from Ghent, and dropped off on the same place every evening. Interview with Mustafa. Mohammed Y. worked all over Belgium for a construction firm, but returned to Ghent every night. Informal conversation with Mohammed Y. 38. Kazim, who worked for EBES, went to Norway with his firm. Interview with Kazim. Working for his second employer, Enzo went to work all over Belgium and often spent the whole week away from home. Interview with Enzo and Christiane. Mohamed Abdellah regularly had to work in Vlissingen (the Netherlands) for the Antwerp firm Fuster Boudon (file number 237 Abdellah Mohamed, number 154, archive Provinciale Metaalbewerkersbond, Archive AMSAB-ISG). Bernabé reports on how the skilled workforce of an international construction company that was working in the Ghent Canal Zone travelled with the company everywhere it went (Bernabé 1968, 57). 39. Of my sample of Mediterranean immigrants arriving in Ghent between 1960 and 1980, 14 immigrants were registered as living in ‘Winkel Rostijne – Barrakkenkamp’ (Winkel refers to a village, part of which had become part of the city of Ghent with the annexation of the Canal Zone in 1965, Rostijne refers to a hamlet in that village). 40. Castles and Kosack 1985, 252, 293. 41. Interview with Ahmed B., who talked about work on a granary, for which they had to work fourteen hours a day and also at the weekend. 42. Interview with Kamel and Miloud. 43. Cooperation and thus communication on the building site became less important during the era of Taylorization, after World War II. However, after the crisis of the mid-1970s, it regained its value.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  69

44. Out of the thirty-one textile factories that employed foreign workers in 1967, only five employed fewer than 100 people (De Bruyn 1966, 7). 45. Interviews with Willy and Amand. 46. Interview with Willem, who worked for UCO Ltd from 1954 until 1996, from 1964 as technical director and from 1982 as personnel manager of all UCO factories; interview with Willy, who worked for UCO Ltd from 1960 until 2002 in wages administration. 47. Interview with Amand; interview with Rabah, a first generation Algerian immigrant who worked for Fabelta for 22 years. 48. Throughout the course of the 1980s, however, shift work, part-time work and new kinds of labour regulations became more widespread, and the ‘normal’ nine-to-five working day became the exception again (Mampuys 1991, 238, 264; Caestecker and Vanhaute 2011, 102). 49. In 1972, the direction of UCO Rooigem proposed a system whereby one would work for three weeks straight, followed by one week of rest. Report of the committee meeting of 16/11/1972, box 3, committee meetings ‘Eendracht maakt macht’ 1971–1973, part I Reports, archive number 147 TACB Ghent, archive Amsab-ISG. See also De Wilde 1997, 186–87. 50. Interview with Willy, who worked for UCO Ltd from 1960 until 2002 in wages administration. 51. In textile factory Ernest De Porre & Fils, in 1971, all spinners and weavers were paid according to their production; other employees were paid an hourly wage. Document 12/08/71, Ernest De Porre & Fils, Nieuwe uurlonen vanaf 01.08.71, box 14, nr. 1107 Ernest De Porre S.A. spinnerij Gentbrugge 1948–1978, part II: dossiers per bedrijf, archive nr. 147 TACB Ghent, Amsab-ISG. 52. Document 10/5/72, Rooigem, Ancienniteit op de twijnderij & aktuele plaats, box 15, nr. 1118 UCO Rooigem 1956-1973, part II: dossiers per bedrijf, archive nr. 147 TACB Ghent, Amsab-ISG. 53. In UCO Rooigem, 1972, in the reeling hall, there were three shifts; the early and afternoon shift were entirely made up of women, the night shift was composed of all men. In the twisting mill, the two continuous teams were entirely made up of men, there was one female and one male montage team, two all-male teams of ‘aftrekkers’, two all-male teams of ‘spoelvoerders’ and an all-female team of wire winders. The only mixed team was a team of cleaners; however, they were not supposed to work together, but went through the factory, on their own, to keep it dust-free. Document 10/5/72, Rooigem, Ancienniteit op de twijnderij & aktuele plaats, box 15, nr. 1118 UCO Rooigem 1956–1973, part II: dossiers per bedrijf, archive nr. 147 TACB Ghent, Amsab-ISG. 54. For those working the day shift, the midday break came in the middle of their working day, while it came after the working day for those working the morning shift, and before for those working the afternoon shift; those in the night shift had a mid-shift break, when they could eat the food they brought from home. The place where most workers spent this break was the cafeteria; Fabelta Ltd had one big cafeteria, where the workers could eat a cooked meal, and several small ones, where they could go to eat their own food. Interview with Amand, who worked in Fabelta Ltd between 1964 and 1982 as personnel manager. 55. Interviews with Willem, Paul, Amand and Willy. 56. Information from the individual files of immigrants who worked as domestic servants. 57. When Maria de la Luz, after four years of work as a live-in domestic, got married and  went to live with her husband in their own apartment, she became a part-time

70  •  Parallel Lives Revisited

c­leaning lady, working in the mornings for a family who had previously employed another Spanish cleaning lady, who had re-migrated to Spain for good. She was asked to continue working at her last address as a live-in servant, but refused as it was in Sint Martens Latem, a village some 15 km from Ghent, which she found too far. Interview with Maria de la Luz. 58. Interview with Maria de la Luz; interview with Lucien. 59. Informal conversation with José Luis and Jesusa. 60. Interview with Maria de la Luz. 61. This comes clearly to the fore in several articles of the edited volume by Momsen (1999). 62. As this percentage was calculated based on numbers that included white-collar as well as blue-collar employees, and immigrant workers were hardly present in that first category of employees, the part they took in the actual blue-collar workforce of UCO was probably a little higher than 10 per cent. Still, the difference will be minimal, as the large majority of employees were blue-collar workers (83 per cent in 1972) (De Bruyn 1966, 13–14; De Wilde 2007, 97). 63. Own calculations, based on data found in De Bruyn (1966, 7). 64. Dahmane left the textile industry in 1965 for a job in a small Belgian firm, installing insulation, where he was the only immigrant among Belgian colleagues. Interview with Dahmane. Somewhere in the late 1960s, early 1970s, Mustafa X. worked for a Belgian company at the sluice between Ghent and Terneuzen, with all Belgian colleagues. Interview with Mustafa X. 65. Hüseyim Y., who worked for metal giant ACEC, was the only immigrant in the section where he worked. Interview with Hüseyim Y. 66. Such as Kazim, who was the only foreigner in his team at EBES, and Mohamed J., who was the only foreigner in his team at Volvo. Interviews with Kazim and Mohamed J. 67. Document 10/5/72, Rooigem, Ancienniteit op de twijnderij & aktuele plaats, box 15, number 1118 UCO Rooigem 1956–1973, part II: dossiers per bedrijf, archive number 147 TACB Ghent, Amsab-ISG. 68. Document 09/03/1975, La Louisiana, document ‘Personeel Lessona’, box 15, number 1118 UCO Rooigem 1956–1973, part II: dossiers per bedrijf, archive number 147 TACB Ghent, Amsab-ISG. 69. Document s.d. (1978), Ernest De Porre S.A., list of personnel, box 14, number 1107 Ernest De Porre S.A. spinnerij Gentbrugge 1948–1978, part II: dossiers per bedrijf, archive number 147 TACB Ghent, Amsab-ISG. 70. Interview with Amand. This strategy is also mentioned as a common employers’ strategy in Martini (2008, 382). 71. In the 1967 enquiries, Turkish workers always made up a large part of the total of foreign workers in the companies where they were employed, between 60 and 100 per cent; Spanish workers only formed a very small part of the total number of staff in the companies where they were employed, but did make up the majority of the total number of immigrant workers (Bernabé 1968, 57; Vanoutryve 1968, 56; Martini 2008, 384). 72. This has been attested for small-scale immigrant entrepreneurs in general, across time and space (Waldinger, Aldrich and Ward 1990b, 38). 73. Interview with Jose Luis and Jesusa. 74. Interview with Clelia, daughter of Antonio; interview with Enzo, who was recruited by Antonio in 1960, at twenty years old. He did however, through the mediation of Enzo, hire a Spanish man at some point.

Integration Processes of Immigrants in the Local Labour Market and the Workplace  •  71

75. For Giovanni’s first restaurant, ‘All Parma’, at Woodrow Wilsonsquare 8, I found 9 members of staff enrolled in the population register, 8 of whom were from Bisceglie. For his second restaurant, ‘All Piccolo Parma’, 4 out of 4 registered staff were from Bisceglie. For a total of 7 Italian restaurants, I found 25 members of staff enrolled in the population register, 21 of whom were from Bisceglie. 76. Interview with Rachid.

Chapter 3

Immigrant Workers’ Relations with Colleagues and Employers 

During the period under study, most Mediterranean immigrants in Ghent worked in ‘mixed workplaces’, with a majority of Belgian colleagues. Further, there was a small number of immigrants who worked in ‘ethnic workplaces’, mostly run by employers who were immigrants themselves. Finally, there was a significant number of immigrants working in private homes, as domestic servants. These different kinds of workplaces led immigrants to construct different kinds of social relations with their colleagues and employers. It is these relations that this chapter will discuss. The focus will be on ‘mixed workplaces’, as these represent the experience of the majority of immigrants in Ghent during the period under study, including many of those who at one point were employed in ethnic workplaces or in domestic services, as this was often only a temporary situation. The dynamics of social relations developed at the workplace form an integral part of immigrant workers’ everyday experiences. Studying them is essential in order to fully understand their social integration in the receiving society. However, whereas this subject has generated a vast amount of literature within the social sciences, it has not yet attracted the attention of many historians (Martini 2008, 373). One of the major obstacles for studying this kind of relationships in the past lies in the limitation of the source material. Whereas sociologists and anthropologists can resort to participant observation and surveys among workers and employers (Estlund 2005; Ogbonna and Harris 2006), such methods are not available to historians. Still, much can be learned from historical sources, such as reports of the

Immigrant Workers’ Relations with Colleagues and Employers  •  73

local unions and life story interviews. These are the main sources that were used here.

Immigrants Working in Mixed Workplaces for Belgian Employers Immigrants’ Relations with Colleagues In collective memory, immigrants’ workplace relations with locals are often remembered as tainted by conflict: locals protesting against the employment of immigrant labourers, immigrants being used as strike-breakers, immigrants protesting against bad working conditions, etc. Throughout Western Europe during the period of guest worker migration, workplace-related conflicts between immigrants and their local colleagues, between different groups of immigrants, and even more so between immigrants and employers were indeed manifold. For example, in the Belgian mining regions, incidents involving groups of foreign miners were commonplace. In January 1964, a quarrel between a Belgian supervisor and a Turkish miner in the Flemish village of Waterschei resulted in a collective action of Belgian supervisors demanding the removal of all Turkish miners from the locality. In March and April of that same year, groups of Turkish miners went on strike to demand a rise in wages. In the same region again, in 1976, striking Turkish miners prevented their Belgian colleagues from going to work. Also in Germany, there  were many examples of strikes and protests involving immigrants. These were caused by a variety of factors: the sacking of a co-national, bad canteen food, insufficient medical support, or discontent about the difference between the promised wages (before tax) and the amount of money actually paid out (Khoojinian 2006, 98–101, 113; von Oswald, Schönwälder and Sonnenberger 2003, 31–32). In Ghent however, for the whole period 1960–1980, there are no reports of major workplace conflicts or incidents involving immigrant workers as a group, or specific groups of immigrant workers, nor any reports of strikes organized by immigrants themselves, or organized protest actions of locals against immigrants. Immigrant workers in Ghent do not seem to have actively profiled themselves as a distinctive group, claiming specific rights, at least not in the context of the workplace. In interviews, immigrants described the atmosphere and social contacts with Belgian colleagues in general as ‘good’, ‘positive’, ‘appreciative’ and ‘congenial’. This image of ‘peace and quiet’ however does not withstand a closer look at the interethnic relations between colleagues on a more day-to-day basis. On the one hand, individual local workers were certainly not always happy with the arrival of immigrants at their workplace (De Muynck 1972, 60–61).

74  •  Parallel Lives Revisited

In a number of instances, they explicitly vented their discontent about this particular issue through their union delegates. This antagonism towards foreign labourers seems to have been stirred by a feeling of discrimination, a fear that immigrant workers were or would be treated better than ‘their own people’.1 The following excerpt from a meeting of the works council of the UCO-Braun textile factory sheds light on this kind of reasoning, which apparently was already present at a very early stage of labour migration to Ghent (the meeting took place in February 1965). It is interesting to note how several of the delegates in their discourse equate ‘the staff’ with the Belgian employees of the factory, clearly excluding ‘the foreigners’, as if they were not a legitimate part of the personnel: Mr De Craemer [union delegate] then talks about the transfer of foreigners, who used to be employed in the de Hemptinne spinning mill, to this factory and asks for the precise reasoning behind this measure. He expresses the discontentment of the staff and posits this as a discussion of principle in order to prevent difficulties in the future. Of course, it would not be accepted that Belgian employees would become unemployed while foreigners remain at work. … Mr Rooyackers [union delegate] asks what would happen with the Belgian employees who would not be able to claim unemployment benefits because they haven’t worked long enough, when a system of alternate unemployment is introduced, and the foreigners who would keep on working because of their contract. This would definitely lead to serious discussions. Further, Mr Vercammen [union delegate] expresses the discontentment of the staff about the fact that during the holiday period of 1964 cleaning work has been especially created in order to employ the foreigners because they couldn’t claim paid holiday … whereas nothing has been done for the Belgian employees who also couldn’t enjoy the legislation concerning paid holiday.2 [author’s emphasis]

Other complaints that were voiced by Belgian workers through the union regarded the latter’s perception that immigrants were working too slow, getting fewer negative remarks, taking up too much holiday, etc.3 Sometimes immigrant workers’ divergent behaviour, due to different cultural habits, caused a negative reaction from local workers. A study from 1967 reported Turkish workers not wanting to make use of their companies’ canteen facilities because they feared they would end up eating pork without knowing it. The refusal of these immigrants to eat ‘the same as everybody else’ reportedly caused frictions between them and their Belgian colleagues (Bernabé 1968, 72–73). Especially Islamic practices and rituals carried out at the workplace made Belgian workers frown at their immigrant colleagues. In some factories, Muslims obtained permission from their supervisors to pray during work hours. When Mohamed J. started saying his prayers in the Volvo factory, where he worked as one of the very few immigrant employees, his colleagues

Immigrant Workers’ Relations with Colleagues and Employers  •  75

did not appreciate this. However, later, as Mohamed showed them that he was a dedicated worker, he believes that they came to accept the fact that he practised his religion at the workplace: At first, they needed to get used to it. They tried to provoke me. I didn’t react to it. But those people, Alhamdulillah [‘praise be upon Allah’, an exclamation of gratitude for the way things were], nothing to remark upon them. They saw that I was working in a serious way. I didn’t take any leave of absence for illness, I didn’t do that, I swear. I also never came late. I didn’t wander around during work hours, no, I always stayed at my place. But when it was time to pray, I hurried. I carried out my prayers, and I returned to my place.4

Feelings of being discriminated against were also experienced by immigrant workers. Many of them felt that they were not treated equally to their Belgian colleagues, especially by their immediate foremen and direct employers. A small-scale questionnaire, carried out in 1971–1972 among twentyfive male immigrant workers in four local companies, showed that 65 per cent of those questioned were convinced that they were assigned less pleasant work than their Belgian colleagues.5 The same enquiry also showed that this was not just an impression, but that discriminatory behaviour from direct supervisors and foremen was widespread and posed real problems for the successful integration of immigrant workers in the company. The researcher concluded that such middlemen actually stand with one foot in the middle of the ordinary workers and as such, share the same feelings and prejudices [as the latter], while they stand with the other foot on the first steps of the hierarchy and as such, are expected to have an objective outlook and a more sophisticated approach. More than once, these two extremes clash with each other, and result in an aggressive and disgruntled attitude towards the foreigner, the underlying motivations of which are xenophobic inclinations, which are sometimes only subconsciously present. (De Muynck 1972, 53–54)

However, in the archives of the local unions, there is no mention of immigrant workers declaring such discrimination to their union delegates, even though at that time, many of them were unionized. This is probably related to the fact that immigrants found themselves in a subordinate position – as newcomers, as a minority, and legally – making it difficult for them to vent their criticism on employers and supervisors, or if they did, to push for the unions to represent them in this endeavour. During the period under study, there were only few immigrant delegates who could have spoken about their own experiences or whose position would have made it easier for immigrants to voice these specific complaints (De Bock 2013, 184–89). Finally, the

76  •  Parallel Lives Revisited

above-mentioned examples of anti-immigrant statements by union delegates testify to the inclination of the trade unions to prioritize the rights of Belgian labourers over those of ‘foreigners’. Tensions between local workers and newcomers tend to become more palpable as more immigrants begin to arrive, threatening the locals in their majority position (Sleegers and Vermeulen 2007, 124). The same happens in times of crisis, when people are laid off or are threatened with lay-offs, and immigrant workers come to be seen as competitors rather than as those who do the work nobody else wants to do.6 This was also the case in Ghent, where moments of crisis heightened the protest of locals against the hiring (or not firing) of immigrant workers. When people were being made redundant at the United Spinning Mills of Ghent during the short economic recession of 1967, union delegates immediately protested against the fact that five Turkish labourers were kept at work.7 When Fabelta Ltd brought in fifteen Tunisian labourers in the crisis year of 1974, one of the delegates called for an immediate halt to immigrant recruitment.8 Again, these protests seem mostly inspired by Belgian workers’ fear of being disadvantaged compared to immigrant workers when it came to hiring and firing. They were directed towards employers rather than towards immigrant colleagues themselves. It is hard to tell to what degree immigrant workers were aware of the many protests Belgian workers voiced against them through their union delegates. In interviews, immigrant workers are generally positive about their contacts with Belgian co-workers.9 Their answer to the question ‘How is/was your relationship with your Belgian colleagues?’ tends to describe a good relationship, with Belgian colleagues helping out when there were difficulties and Belgians and immigrants sharing their meals together in the company’s canteen.10 Immigrant workers from the UCO Rooigem textile factory, interviewed in 1967, did mention tensions with workers of other nationalities. Spanish, Moroccan and Algerian immigrants told the interviewer they did not like to work together with Turkish immigrants ‘because they are very competitive and very short-tempered’, whereas a Turkish immigrant reported not wanting to work with his Algerian colleagues, ‘because the relationship with them is not good’ (De Bruyn 1966, 56–57, 62–63, 66–67). When asked about their relationships with Belgian colleagues, they reported only good relationships. However, when they were asked more explicitly whether they had encountered any racism or discrimination, several immigrant workers did report incidents which had taken place on the shop floor. Still, they seem to have interpreted these as exceptions (‘just one annoying man’), rather than as a sign of latent racism in the workplace. Mohamed H., another Moroccan immigrant working for Volvo Ltd, remembers getting along fine with everybody except with one man:

Immigrant Workers’ Relations with Colleagues and Employers  •  77

I do remember a colleague who always annoyed me, his name is [anonymous] … I never reacted even though I often had problems with this man. I used to go to the bathroom to calm down. Later we became friends, even though his remarks didn’t stop. Once, in the canteen, he asked me, when I was drinking my soup, if I had brought it with me from Morocco!11

This ‘desire to downplay the effects of racism within the workplace’ was also noted by Joanna Herbert in her study of interethnic relations in Leicester, especially with her male respondents. She interprets this kind of statement as evidence of the acceptance of racism, with interviewees claiming that they ‘just took it’ (Herbert 2008, 154). The refusal of immigrant interviewees to refer to these kinds of incidents in terms of racism, and their tendency to individualize problems with Belgian workers rather than attribute them to a broader problem of widespread xenophobia, point to tensions on the shop floor being present but remaining inexplicit. Discrimination and conflict were probably more widespread than these testimonies let us believe, something that is related to the specificities of this kind of source material (see introduction). Studies for other groups in other places show much more overt racism in the workplace.12 Further, even if most of the workplace relations between immigrants and locals in Ghent during the period under study did not include open conflict and were based on regular, friendly contact, they hardly deserve adjectives such as ‘positive’, ‘appreciative’ and ‘congenial’, which are the ones used by immigrant interviewees. Mohamed H.’s interpretation of ‘good contacts at work’ (‘we said hello and talked a bit, nothing more’) is a clear example of what most immigrant workers seem to imply when they talk about the ‘good relationships’ they maintained with (Belgian) colleagues. These relationships generally do not seem to have surpassed the stage of what Blokland calls ‘transactional relations’, including minimal cooperation on the shop floor and friendly politeness during breaks. This interpretation is far from what labour sociologists – maybe in a somewhat naive way – describe as ‘good workplace relationships’. These are characterized not only by cooperation, but also by intense socializing at different moments throughout the workday – the kind of relationships Blokland calls ‘attachments’. Such ‘good workplace relationships’ or ‘attachments’ can result, over time, in colleagues getting to know each other as individuals, developing feelings of affection, empathy, sympathy and loyalty towards each other, and as such, a sense of belonging and often even real friendships or ‘bonds’, in Blokland’s terms (Estlund 2005, 82, 85). For immigrant workers in Ghent during the period under study, more meaningful workplace relations and especially real bonds do seem to have developed within ethnic boundaries. This was also noted by contemporary researchers. In her study of Moroccan families in the Netherlands (see

78  •  Parallel Lives Revisited

c­ hapter 2), Lotty van den Berg-Eldering found that the Moroccan women of her sample group who worked outdoors did not have very intense contacts with Dutch women: ‘In the company, the Moroccan women acted as a group. They lunched together and hardly ever joined the Dutch women. They also didn’t feel the need to invite these Dutch women to their homes. This group formation reinforced mutual social control’ (van den Berg-Eldering 1978, 103). Ersan Yücel, studying Turkish migrant workers in Germany as a participant observer, noticed that The Turks usually had their lunch break at the same time and ate together. The food each person brought was put on a table and shared. If some of the Turks did not stop working while others were eating or drinking tea or coffee, they were always invited to share the food and beverages. This invitation did not usually extend to workers of other nationalities, symbolizing the limitation of intimate relationships to their own group, thereby heightening the consciousness of solidarity within the group. (Yücel 1987, 139)

The differences between contacts with Belgian colleagues and contacts with co-ethnics clearly come to the fore in several interviews with immigrants from Ghent, such as this one with Mehmet S., a Turkish immigrant who told the interviewer about the time he worked in a meat factory, between 1974 and 1979: Interviewer: Did you have contacts with your Belgian colleagues? Mehmet: We said hello to each other and sometimes we talked to each other, but of course we didn’t have much time. Interviewer: What about your Turkish colleagues? Mehmet: We visited each other. Also, our children visited each other. If anything needed to be done, they helped me. Interviewer: How is the contact with your ex-colleagues? Mehmet: The contact is still good. Interviewer: Are you still in contact with your Belgian colleagues? Mehmet: No. Interviewer: With your Turkish colleagues? Mehmet: With a few.13

Even though the terms in which this interviewer asked his questions clearly pushed the interviewee to distinguish between categories (Belgian vs. Turkish) that he himself might not have chosen, the factual information we get from these interviews does point to a more superficial working relationship between immigrants and locals than between co-ethnics. This does not mean however that the latter necessarily became friends. Friendship requires more than just a good workplace relationship. For Yusuf E., for example,

Immigrant Workers’ Relations with Colleagues and Employers  •  79

a Turkish ­immigrant from Piribeyli who came to Ghent in 1974, his real friends were the people from his village, whom he had known for a long time and who were living nearby in Ghent as well.14 The shallowness of immigrant workers’ contacts with most of their colleagues who were not also neighbours, old friends or fellow villagers can be explained in different ways. First, as all target earners, immigrant workers were not very involved with the non-financial aspects of work, such as sociability with fellow workers. Striving to earn as much money as possible in as short a time span as possible, they generally saw their work purely as a means to an end, not as an essential part of their social and personal selves (Piore 1979, 54). Further, as their stay was meant to be temporary, there was no reason for them to put much energy in getting to know their fellow workers, as they would soon be leaving. The high professional mobility many of them exhibited during the initial years of their stay (see chapter 2) also prevented many workers from getting to know their colleagues, as they simply did not get to work together for very long. However, this mobility as well as their temporary commitment as target earners and ensuing attitudes to work did not remain as strong during the whole of their professional career. Those who became more stable certainly had more incentives to try and make contact with fellow workers, and they did (see further). Immigrants who worked for many years in the same workplace developed better relationships with their colleagues than those whose careers were highly fragmented. But meaningful work relationships within rather than across ethnic boundaries seem to have remained much more frequent. A clue as to why this might have been is provided by Ali G., a Turkish immigrant who had arrived in Ghent in 1974 and was working for Fabelta at the time of the interview, in the mid-1990s. He pointed out how a lack of time to talk at work and the urge to go home immediately after did not provide him with the opportunity to make friends with his Flemish colleagues. On the other hand, he could meet his Turkish colleagues outside of work, in the neighbourhood, the Turkish café or the mosque. And, importantly, they ‘spoke the same language and had the same culture’.15 For many immigrants, language differences were and indeed remain the most important barrier against the development of workplace-based attachments and bonds across ethnic boundaries.16 The idea is that colleagues get to know each other through spontaneous conversations, talking about work, current events, sports, family and other ‘stuff of daily life’ (Estlund 2005, 82). From this perspective, the ability to communicate effectively is essential to go beyond the most superficial of contacts. If there is no communication, there can be no ‘good workplace relations’. In Ghent in the period under study, language was indeed an important factor for the development of social relations in the workplace, s­eparating

80  •  Parallel Lives Revisited

people who did not speak the same language and uniting those who did (De Muynck 1972, 46–47). However, linguistic differences were not absolute. Many immigrant workers and their colleagues managed to converse in a ‘third language’, which was native to neither of them but which they both understood. The most important of these was French. Up until the mid-1970s, in Ghent, French was commonly spoken among the higher and middle classes and frequently used in administrative, political and judicial contexts. In particular those immigrants who had migrated from North Africa, had previously lived in France or Wallonia, or whose mother tongue belonged to the Romance language family, tended to speak French upon arrival or learnt it during their stay in Ghent. This allowed them to communicate with their employers and with some of their Belgian colleagues. ‘There were some who spoke French, but not all of them’, remembers Dahmane, an Algerian immigrant who started work at one of the UCO factories in 1963, as part of the first group of immigrant workers recruited from Northern France.17 The widespread acceptance of French as a lingua franca in Ghent initially put Turkish workers at a disadvantage, as they only rarely spoke this language. For them, learning Dutch was the only way to ensure communication with their Belgian colleagues and supervisors, if not necessarily with other immigrants. However, not many immigrant workers managed to do this (De Bock 2013, 161–66). Most of them learned enough of the local language to be able to do their job properly, but not enough to keep a casual conversation going. For those immigrants, workplace contacts across ethnic boundaries do not seem to have developed beyond transactional relations. Hüseyim Y., who worked for steel giant ACEC Ltd for twenty-two years, explained this as follows: ‘I didn’t speak the language. I understood what the chefs wanted me to do, but apart from that, I had a hard time making contacts. It is difficult to trust someone with whom you have difficulties making contact.’18 Even years of working together clearly did not suffice to turn these relations into actual bonds. Celyn C., who worked in the same UCO factory for seventeen years, kept referring to his Belgian colleagues as ‘colleagues, no real friends’, with whom he ‘didn’t have that much contact’. His Turkish colleagues however were friends, as ‘with them he could talk more easily and they have the same culture’. He was still in contact with some of them three years after he lost his job.19 Towards the end of the period under study, the language barriers changed. As the ongoing federalization of the Belgian state resulted in a more pronounced ‘pro-Dutch’ policy, French lost its privileged status.20 This had a major impact on those immigrants who, until then, had got along just fine with their knowledge of French. Civil servants, police officers, bus drivers as well as colleagues and employers started to change their minds about speaking French, sometimes bluntly refusing to do so. For many of the

Immigrant Workers’ Relations with Colleagues and Employers  •  81

i­mmigrants of the first generation, this change came as a shock. As they were not emotionally involved in the tensions between the two major languages of the Belgian state, they could not understand why, after all these years, they suddenly were required to speak the other official language, even though they knew that most of the Flemish people they came into contact with understood French perfectly well.21 Other workers however did overcome the language barrier, creating closer interpersonal workplace relations across ethnic boundaries despite linguistic differences. Amar S., an Algerian immigrant who came to Ghent to be with his Belgian girlfriend and who worked at Fabelta between 1970 and 1982, does not remember any of his close colleagues being able to speak French, so he tried to speak Dutch to them: ‘When I worked at Fabelta, there, they didn’t understand French. So, they were all young people, who came for example from Nazareth-Eke, Eeklo and so on (communes near Ghent, ed.), so I tried to speak Flemish with them, and so, we tried: correct, not correct, correct, not correct.’22 As opposed to Celin and Hüseyim, Amar also saw quite a lot of his Belgian colleagues outside of work hours. As ‘they were all young people’, they simply had more time to spend with each other. They also had more time and were more inclined to learn to speak the language, which only improved their contacts with Belgian colleagues (Bernabé 1968, 76). Here, age and life cycle were important factors deciding whether and how workplace-based relations could develop. Young people, especially those who were unmarried and had no family to look after – the overwhelming majority of whom were men, as in Ghent, there were hardly any single immigrant women working in mixed workplaces – were more open than older immigrants to contacts with colleagues of different backgrounds on and off the shop floor. As twenty-somethings, they all ‘just wanted to have fun’, which was common ground enough to overcome or simply ignore linguistic and cultural differences. Some of the young immigrant men at UCO Ltd signed up for the UCO football team, where they played together with Belgian colleagues and immigrants of different nationalities. Others regularly went out with their Belgian colleagues after work or at the weekend.23 Charif B. even spent his weekends in the villages outside the city, as that is where his colleagues in the textile factory came from.24 This way of spending time with colleagues outside work hours and off the shop floor ensured the development of closer and deeper workplace-based relations.25 As the years went by and these young men grew older, married and got children, they generally dropped out of the football team and stopped going out at weekends. Even when they stayed at work in the same factory, with the same colleagues, their workplace relations became (again) restricted to the workplace itself, and the contacts with their colleagues fell back to a more superficial level.

82  •  Parallel Lives Revisited

Figure 3.1  The football team of the UCO textile factory, where Belgian and immigrant workers played together, 1963. MIAT-Museum about industry, labour and textile

Apart from migration motives, length of employment, language and life cycle, it was the work itself that impacted on immigrants’ chances and motivations to develop more meaningful relations with their colleagues. Work that required intense cooperation automatically brought workers closer together than work that was more solitary; work that was carried out in difficult circumstances allowed for less contact than work that was carried out in a quiet, clean and less stressful environment. The size of the company also mattered. In small building companies run by locals, for example, immigrant workers generally picked up more of the majority language than they did in large companies, as they needed it to carry out their job. Even during breaks, working for a large company often meant that there was more of a choice as to whom one spent one’s break with, whereas in a small company, colleagues generally all stayed together during breaks.26 Still, even though a lot of the workplace-based relations immigrant workers constructed across ethnic boundaries remained at a superficial level, these relations were far from meaningless. Their importance became clear when mass-scale unemployment hit immigrant workers from the second half of the 1970s onwards. Apart from the physical and psychological problems many of the long-term unemployed, especially men, had to deal with, unemployment also meant a complete loss of interethnic contact for many. The

Immigrant Workers’ Relations with Colleagues and Employers  •  83

same thing has been shown by Bouras for Moroccan immigrant men in the Netherlands (Bouras 2012, 156). The complete absence of such contacts had an important impact on the social and human capital of these immigrants. In their stories, many express this change in terms of a gradual loss of Dutchlanguage proficiency. Such is the case, for example, for Celyn C., who, when confronted (in quite a rude manner) with his limited knowledge of Dutch by a Flemish interviewer, explained himself by saying he used to be able to speak the language a lot better when he was still working in the factory: Interviewer: Was language a problem in your contacts with your [Belgian] colleagues? Celyn: I learnt some Flemish in the factory and we spoke Flemish to each other, but I forgot a lot since then. Interviewer: But you don’t know that many words. How do you communicate with your neighbours? Celyn: I used to know a lot more than now, and there was no problem between us.27

Immigrants’ Relations with Employers As with the relationships between immigrant workers and their colleagues, the available source material indicates no large-scale, group-based disputes between immigrant workers and their employers. From their side, employers seem to have been much less discriminatory towards immigrant workers than their immediate supervisors. Research carried out in 1971–1972 in four local companies employing a large number of immigrant workers showed that higher staff members, as a rule, behaved in a correct way towards their immigrant workforce: ‘The higher management normally ­creates no problems. Most of these functionaries have a broad outlook on the situation, they are aware of the necessity to hire foreign workers and are, because of that, more likely to treat them on the same basis as local employees’ (De Muynck 1972, 53). In interviews, employers and higher supervisors also firmly state that they did not behave differently towards their immigrant workforce than towards their local staff. For them as patrons, they claim, the good progress of the work was what mattered most, independent of who the workmen were.28 Some employers put in an extra effort towards their foreign employees, by trying to learn a bit of their language, or undertaking a ‘study-trip’ to companies that were already used to working with immigrant labourers (Bernabé 1968, 66). In addition, employers and high-rank ­supervisors did not tolerate any discriminatory behaviour towards their foreign workforce, as it disturbed the workers’ productivity. Especially when lower-ranked supervisors exhibited such behaviour, employers and ­managerial staff tended to intervene.29

84  •  Parallel Lives Revisited

Among employers and higher supervisors, racism and prejudice more often found their expression in a paternalistic attitude rather than an outright discriminatory one. Paternalism seems to have been the guiding principle for their behaviour towards their immigrant workforce, much more so than towards their local workers. Often, employers saw their foreign workers as an undifferentiated group of poor, illiterate farmers, who had to be taught how to behave in a modern society. This is clearly what former personnel manager Amand thought of the Tunisian workers Fabelta had recruited: If you talk about those Tunisians who arrived here. I remember it, yes, those people had to have all kinds of documents, work permits, all those things had to be filled in for them. … They were here, on our expense, for one year, you know? So the state said, ‘You can go and get them, but you have to look after them for one year’. No board, but housing, accommodation, all that was needed. It’s incredible what that entailed, you know? Because, when you have those chaps [In Dutch, he uses the dialect word ‘kadeeën’, which normally refers to young children] here, you have to explain to them – those people come from the edges of the desert, you know? – and in the supermarket I told them: ‘Look, “promo”, that means that it is a bit cheaper’, etc. Yes, afterwards they had to go and buy their food for themselves, you know?30

Amand really did believe these Tunisian workers came from ‘the edges of the desert’. In the interview, he repeated this several times. What he clearly  did not know was that actually, most of them came from one of the major  Tunisian cities, had finished secondary education, and several of  them  had worked as government officials or in trade (De Bock 2013, 64-65, 139). The immigrant workers themselves often saw their relations to their supervisors and employers as individualized relations, a characteristic that is typical for target earners (Piore 1979, 27). This manifested itself in their behaviour towards them. Sometimes, a house call, simply meant to pick up an employee to go to work, could turn into a ceremony. Paul, former personnel manager at UCO, clearly remembers one such house call: I have to say, it was pouring down with rain, and as I arrived she says ‘come in’, it was his wife, ‘come in, come in’. So I went in, I couldn’t keep waiting in the rain, and I was shown into that little living room and I had to sit down in the sofa there, and I waited and I waited. I say, damned, that man is somewhere in the house, or in the bath, or I don’t know. And eventually, both of them came in, completely, well, dressed up in traditional costume, with a tea tray, with cakes, with boiling hot tea, in those glass cups, and then pouring it, a whole ceremony like that. But well, those people, they were very hospitable I have to say, and also … they appreciated it of course, when you say, look, I will share a meal with them … but of course I couldn’t always do that.31

Immigrant Workers’ Relations with Colleagues and Employers  •  85

Many immigrant workers also brought their supervisors gifts when they returned from their holidays in the home country. Paul was presented with such gifts several times, as was his colleague Willy, who also worked for UCO Ltd.32 Belgian employers and supervisors often did not know how to behave when they were confronted with gifts or invitations to immigrants’ homes. On the one hand, they were reluctant to accept them, as they did not want to be accused of favouritism – something Belgian workers were very anxious about. On the other hand, they did not want to insult the people who offered these gifts and invitations. Clearly, they were not  used to or intending to take part in the individualized relationship their  immigrant  workers  were trying to construct with them. The latter, for their part, saw things  differently: ‘My boss loved me’, ‘I was like a son  to  him’, etc. are recurrent themes in the interviews with immigrant workers.33 Such an individualized perspective on the employer-employee relationship seems to have been largely influenced by the ability of workers to communicate directly, without the interference of an interpreter, with their supervisors, and therefore, in practice, by their knowledge of Dutch, French or even German. Those who were unable to do so always needed the help of an interpreter, which interfered with the personal character of the encounter. In small companies, or even in larger companies if there were only a few immigrant workers of a specific language group, communication between employers or supervisors and immigrant workers who did not speak their language happened ad hoc. Whenever there was a specific problem, an immigrant worker who could understand both parties was called to translate on the spot. Only in larger companies, for large groups of immigrant workers who spoke the same language and if the company management considered it necessary, was a position of interpreter created. Generally, companies did not hire professional interpreters but appealed to those employees who had learned one of the languages of the receiving country. These ‘amateur interpreters’ were paid as ordinary workers, but they often also received a special remuneration for their efforts (Bernabé 1968, 76–77; Vanoutryve 1968, 59). In Ghent, most (but not all) of these interpreters were Turkish, as many Turkish workers only understood their mother tongue, while workers of other nationalities generally had at least some notion of French or sometimes Dutch. For newcomers who did not speak the language, these interpreters became important figures. Often, they were the ones who introduced them to the factory and who explained their work to them, who communicated the management’s decisions to them, and whom workers had to go to whenever they had a problem.34 This way, they gained a lot of power over their co-ethnic colleagues. Sometimes, they overstepped their mark, and started behaving like superiors (De Bruyn 1966, 23), or found ways

86  •  Parallel Lives Revisited

to make money out of their compatriots directly.35 Their position, based on their knowledge of the language, procured them a special status, which often reached beyond the shop floor into other domains of life. Kazim, who worked as an interpreter in one of the UCO factories for a couple of years during the 1960s, later became an important middleman between the local authorities  and ‘the Turkish community of Ghent’. When the authorities needed to  know the opinion of ‘the Turks’, he was one of the few people they turned to.36 Also Ömer, who worked as an interpreter in several UCO factories over the course of the 1960s, managed to make use of his skills elsewhere, working for the local police and in court cases.37 As such, the social reality of mixed workplaces created a new category of people who were important mediators between locals and immigrants, enhancing and limiting contacts between the two groups at the same time. Things were very different in what we have called ‘ethnic workplaces’, where such mediators were not needed, as the main obstacle for contact – language – was not a problem there.

Immigrants Working in ‘Ethnic Workplaces’ for ‘Ethnic Employers’ Immigrants’ Relations with ‘Co-ethnic’ Colleagues Relations between colleagues in what we have called ‘ethnic workplaces’ (workplaces where all or the large majority of workers as well as the employer are immigrants sharing an ascribed ethnicity) were much more like what Estlund (2005) and other social scientists have described as ‘ordinary workplace relations’ than those between colleagues of different ethnic backgrounds in ‘mixed workplaces’. One very important reason for this is that colleagues in ‘ethnic workplaces’ all spoke the same language (or at least a variation of it) and were therefore not confronted with what in mixed workplaces seemed to be the single most important barrier for social contact. As such, everybody in the workplace could engage in casual conversation with everybody else, get to know them, and possibly become ‘good colleagues’, obtaining that ‘real sense of belonging’ that is supposed to be more present among co-workers ‘than among any group other than family or friends’ (Estlund 2005, 82). In addition, colleagues in ‘ethnic workplaces’ tended to have more in common than just a shared language or nationality. Due to the very specific ways in which many ‘ethnic employers’ recruited their workforce, they often came from the same region of origin or even the same village or town, and spoke the same dialect.38 Some of them were family or had already been friends back home.39

Immigrant Workers’ Relations with Colleagues and Employers  •  87

As opposed to interethnic relations between colleagues in mixed workplaces, that, as we have seen, were mostly single-stranded (limited to the shop floor or its immediate surroundings), the relations between colleagues in ‘ethnic workplaces’ were generally multi-stranded, containing more than one focus of interaction.40 Not only did they work together, they often also lived together (in the house of their employer or at their workplace, like many of the Italian young men who came to Ghent in the 1970s and lived above the restaurants where they worked); they saw each other in church (in the special masses that were held for Spanish and Italian immigrants), in (immigrant) associations, clubs or cafés; and, when they were young, they went out together after work or during the weekend. As such, their ties to each other can be described as attachments or even bonds, as opposed to the transactional relationships that characterized interethnic workplaces. Girolamo, who arrived in Ghent at age fifteen to work in an Italian restaurant, had a very close relationship with his immediate colleagues, and also with the staff of other Italian restaurants, most of whom came from his hometown Bisceglie (De Bock 2013, 226–28). As young lads, in between and after shifts, they went partying at the Kuiperskaai, next to where they worked. On their days off, they all went for dinner at ‘Bisceglie’, an authentic Italian restaurant which ‘made all the more Italian stuff, casareccia, things like that, beans, peas.’ Rather than a mere restaurant, the place was ‘like a family’ to them.41

Figure 3.2  The young men from restaurant Al Parma with owner Gianni Bombini in the middle, early 1970s. Amsab-Institute of Social History, collection Salerno

88  •  Parallel Lives Revisited

Immigrants’ Relations with ‘Co-ethnic’ Employers Employer-employee relationships are not the same in ‘ethnic firms’ as in ‘mixed workplaces’, a difference that has been explained by Waldinger et al. by referring to the ways in which both parties in an ‘ethnic firm’ make use of their ‘ethnic commonality’ to structure and organize the employment relationship. This happens, for example, by grounding authority on ethnic allegiances rather than discipline, and by using a repertoire of ethnic symbols and customs to underline similarity in the face of potential conflict (Waldinger, Aldrich and Ward 1990a, 38). Over the following paragraphs I will argue that it is not only this abstract notion of ‘ethnic commonality’, but even more the existence of mutual social ties, both in the home country and in the receiving society, that accounts for many of the specificities of the employer-employee relationship in such ‘ethnic firms’, at least the ones encountered in Ghent. In Ghent, ‘ethnic employers’ tended to share a number of old world ties with their employees. In the home country, their families knew each other, they had lived in the same neighbourhood, or they had friends or acquaintances in common. Often, they were also related by kinship (De Bock 2013, 226–28). The extension of their relationship into the home country undoubtedly imposed a number of constraints, as their behaviour towards each other could be reported back home. There, it could have repercussions on social relations within a family, a neighbourhood or even a whole town. Further, even if there were no old world ties linking immigrant employers to their employees, as was the case for example in the photo-­ laboratory of Arnedo Martinez (see chapter 2), their relationship in the receiving society was often characterized by a degree of multi-strandedness that did not exist between Belgian employers and their immigrant workers. ‘Ethnic employers’ went to the same churches as their employees. Often, they belonged to the same clubs and associations and went to the same tea-rooms and cafés.42 Some employers even housed their employees in their family dwelling, especially if they themselves had recruited them from abroad.43 By offering them a place in their own house, these employers made the lives of their workmen easier. Not only did they not have to look for accommodation, they actually lived with their employer’s family, and as such did not have to cook their own food or wash their own clothes.44 Often, they were really made to feel ‘at home’. This was the case for Girolamo, who slept in the same room as the sons of his employer, Mauro di Pilato.45 Enzo, who was recruited by mosaic maker Antonio Bearzatto at age twenty, was also housed in the home of his employer. When talking about it, he ­remembers the following anecdote:

Immigrant Workers’ Relations with Colleagues and Employers  •  89

At that moment, my roommate was Giovanni, also a Friulian, and we differed one year in age. The room was divided in two, with two beds, next to each other, like one bed but separate. … In the summer, I left the window open and the door … That ‘nice lady’ [the sister of his employer] had put up a frame in my room with two nails here and one on top. And then there was a gust of wind! [making the frame fall on his head] Luckily I was lying like this [showing how he lay with his arms above his head]. If not, that glass frame of the Madonna would have ruined my face! So at two in the morning, I took the frame, I opened the door of the room in the hallway, papapam, everybody awake! I told her ‘If you hang another Madonna or Christ above my head one more time …!’ I told her I would come and take her outside! I had woken up everybody, you know!46

From Enzo’s anecdote, we get an interesting glimpse into the relations within such a household that was made up of an employer, his family and his employees. Whereas many employees who were housed with their employer, Enzo included, stress the benefits of being able to live ‘as in a family’ (for example, the workmen at Antonio Bearzatto’s house always ate dinner with the family, at the big dinner table), this ‘family atmosphere’ also had its drawbacks. In Enzo’s story, the insistence of the sister of the padrone on hanging up religious tokens in the workmen’s bedrooms is clearly seen as an intrusion in their private space. Notwithstanding their being ‘part of the household’, the power balance between family members and workers was not even. In the mosaic and restaurant businesses, the employer-employee relationship was all the more complex because some of the employers used to work together with their employees, as colleagues. Of course, they could not just set aside their previous, much more egalitarian relationship, even though they had now become padrone.47 All of these ties that bound employers and employees together in and outside the workplace made their relationship more personal, creating a number of expectations as well as obligations from both sides. Personal loyalty as well as the kind of ‘ethnic solidarity’ referred to above turn an ‘ethnic workforce’, made up of relatives and co-ethnics who are tied together by bonds or at least real attachments, into a more stable and reliable workforce, prepared to work under conditions that ‘outsiders’ find unacceptable (Waldinger, Aldrich and Ward 1990b, 38; Boissevain et al. 1990, 142). In Ghent, more than one former employee remembers the long working days, the unpaid extra hours, the unannounced jobs at weekends and in the evenings, the relatively low wages they were paid, etc.48 On the other hand, these same ties and feelings also limit the leeway of employers, obligating them to look after their workforce in ways other employers did not have to, for example by providing them with a welcoming home, or by helping them out with administrative or other issues (Waldinger, Aldrich and Ward

90  •  Parallel Lives Revisited

1990b, 38; Boissevain et al. 1990, 143). The following testimony of Enzo is telling. When he first talks about his ‘foster family’, his memories are very fond, but as the interview moves through the years he spent with them, they become less positive: But I have been lucky; first, I worked for a firm that was a bit like family, and Friulian, we spoke Friulian. … As I arrived in Ghent, I had the good luck to have my brother, and the family. I arrived at the station, with two suitcases, one with my tools and one with clothes, and he [his brother] was there, he waited for me to arrive, he came with the tram to pick me up. I lived there, I  ate, I slept in their house, where they had rooms. … There was only this problem, when a client came, you had to go work another half hour extra! Ah, yes, but those things … we were also a bit of a family. Via my brother. Because he [his employer] was his father-in-law, so … You know, us, Italians! On Saturday, in those days, you had to work four hours. For those four hours, you didn’t receive any money, the boss paid you your fiscal taxes. Social security. But it happened now and then, that you had to work as well after lunch. When there was something special. So I would do that. … At a certain point I told her [his wife]: ‘I can’t live there anymore. I am not, after work, I am not free. Because it is a bit of a family, you can’t refuse, as with us, Italian types, the family comes first. … It happened that I was already dressed up, white shirt, to go out, and that the truck would arrive, with the material. So you had to unload it, take another shower, wash the shirt again, but I had to pay for that, you know, for the shirt? It had to be sent to the drycleaners!49

This mixture of nostalgia for the ‘family atmosphere’ that made them feel at home in a strange country and bitterness for the way they were exploited is characteristic of many of the interviewees who have worked for ‘ethnic employers’.50

Immigrants Working in the Domestic Service Sector Immigrants’ Relations with ‘Colleagues’ During the period under study, in Ghent, domestic servants were generally employed on their own, or sometimes together with their spouse. As such, they had no ‘real’ colleagues, working for the same employer at the same workplace. However, in their free time, immigrant domestic servants tended to meet up with others who did the same kind of work. These others were often, but not exclusively, people they had known before migration (family, friends, neighbours or fellow villagers) and whom they met i­nformally in cafés, public places, or even in the house of one of their employers. Aurora, a Spanish immigrant who arrived in Ghent in the early 1960s in order to

Immigrant Workers’ Relations with Colleagues and Employers  •  91

work as a housemaid, remembers regularly meeting up with a group of Spanish domestics in the kitchen of the house where her parents worked.51 It also helped that most domestic servants in Ghent were Spanish. The mutually reinforcing system of a Spanish intermediary and recruitment through the personal networks of (Spanish) domestics who were already there had resulted in a practical ‘monopoly’ of Spanish immigrant men and women in the local domestic service sector (De Bock 2012, 53–54). From 1965 onwards, Spanish domestics also met each other (and other Spanish immigrants) on Sunday afternoons in the chapel of the Crombeen-Institute, a Franciscan school and convent where a weekly mass was held in Spanish by a Spanish priest. On Sunday evenings and Wednesday afternoons, they gathered in the Hogar Español, a club house sponsored by the Spanish consulate and supported by the local section of the Christian trade union ACV, at the initiative of their Spanish employee Miguel Echeverria.52 The hours of the mass and the opening times of the Hogar Español were in fact geared especially at domestic servants, who only had Wednesday and Sunday afternoons off (see chapter 2). Later, the Hogar also opened on Friday and Saturday evenings, for those who did not work in domestic service.53 From the early 1970s onwards, there was also ‘Las Tapas’, a bar run by a Spanish couple, Maria del Pilar and Agustin. In ‘Las Tapas’, Spanish women spent

Figure 3.3  Weekly gathering of Spanish immigrants in Ghent at the Hogar Español, 1960s. Amsab-Institute of Social History, collection Ayestaran

92  •  Parallel Lives Revisited

their Sunday a­ fternoons at the big table, drinking coffee, while their men set at the bar, having a couple of beers.54 Immigrants’ Relations with Employers The working relationship between domestic workers and their employers is a very special relationship, highly personalized but at the same time very hierarchic and unequal. In the case of immigrant domestic workers, class  differences combine with ethnic differences to create a significant social  distance between employers and employees, in which gender often plays an important role as well (Momsen 1999; Cox 1999). The foreignness of these domestic servants effectively hides the unequal power relationship by setting it outside the social classification system of the society of which the employer  household is part (Harzig 2007, 54). Employers do not only  hire immigrant  domestics because they can no longer find local workers to fill  in  these positions, but also because it is easier for them to see immigrants as servants, whose lives are completely subordinate to their wishes and needs. The attitude of Ghent employers towards their immigrant domestics does not seem to have been openly or consciously racist, but was clearly characterized by a high degree of paternalism. Ghent employers perceived their (Spanish) domestics as people who were unspoilt by modern society and its norms and legislations – especially those that were meant to protect the workforce. A student who, in 1968, conducted research into the lives of Spanish immigrants in the city, reported the following incident, which sheds an interesting light onto the mind-set of these employers: In one case, the permission to conduct an interview [with Spanish employees] was refused [by the employer], in a diplomatic way. As a reason, this person  indicated the very primitive development of her domestic staff. Our  questions would probably trigger a certain awareness or shock, possibly causing her employees to become more demanding: ‘Once they know something, they want more’. Actually, we heard this statement more than  once,  without however causing a refusal [to interview]. (Vanoutryve 1968, 60)

The language differences between employers and their staff were often mitigated as the higher classes in Ghent generally spoke fluent French, and many domestics had a notion of this language, having acquired it in their home country or whilst working in France, before they came to Ghent.55 Sometimes employers had a notion of Spanish, often through their experience with ­previous (Spanish) domestics.56 For newcomers who could not communicate

Immigrant Workers’ Relations with Colleagues and Employers  •  93

with their employers, however, the first months in service could be challenging. Language and other difficulties are remembered in some of the stories people tell and laugh about heartily today: She [his wife, who was a domestic herself ] had an aunt here … she died now you know, she died, an aunt of hers, and that was of course ‘français hein, français’, but of course those people didn’t understand any French, you know? And there was a party, and they had to serve, and there you go, there was a piece of cake left, and Madame said ‘Carmen, il faut mettre le gâteau dans le frigo’. ‘Le gâteau’, a cake, but the ‘gatto’, in Spanish, that is a cat. So she put the cat in the fridge! Like Luis, Encarnita’s mum and dad, they arrived here, and they got a  room in the attic, well installed, a room in the attic, and the first night, they were embarrassed to go and ask their employer, … and because of that, she said, ‘I have to go to the toilet, but I don’t dare, and there’s no toilet upstairs’, so they had to open the roof window, and she had to put her bum through the window, and he had to hold her. That really happened you know! Really!57

These stories however are also testimony to the feelings of insecurity and helplessness domestic servants were confronted with when they first arrived, and shed light on the inequality of the work relationship they found themselves in. On the other hand, the stories of many former domestics today also speak of a comfortable work environment, with friendly, caring employers. Indeed, a good working relationship was the main criterion according to which domestic servants assessed their jobs. Bad working relationships usually resulted in them leaving their position for another one (De Bock 2013, 177–78). Furthermore, good relationships could have an enormous impact on these immigrants’ lives – and that of their immediate social circle – far beyond the workplace. This was the case for Maria del Pilar, former owner of ‘Las Tapas’. The employers of her sister-in-law Maxima found jobs and a place to live for her and her husband. Later, when Maria del Pilar wanted to set up her business and had to apply for a professional card, they sorted out her paperwork. They had an even closer relationship with Maxima, as they became the godparents of her child.58 The inequality inherent in their work relationship apparently did not prevent good, long-lasting relations from developing between domestic servants and their employers. Further, good relationships with their employers seem to have provided at least some domestic servants with an important stack of social capital, as (former) employers offered them a hand with job-hunting, ambitions to set up as independent business owners, or the search for a house. At the same time, these personal relations did not inspire

94  •  Parallel Lives Revisited

immigrant workers to consider their employers as equals. Even today, they still speak about them in terms of ‘Monsieur’ and ‘Madame’, rather than ‘my boss’ or ‘my employer’, describing them as ‘grand people’ and referring to them by using their professional titles such as ‘magistrate’, ‘doctor’ and ‘colonel’.

Conclusion In order to understand the workplace-based relations constructed and maintained by immigrant workers, we have distinguished three different types of workplaces, as they offered different contexts for the formation of such relations. Immigrants working in domestic services had no real colleagues, as they generally worked alone or with their spouse. The ‘colleagues’ they met outside of work hours were people they had known before migration or others they met at immigrant clubs or religious services. Their relations with their Belgian employers were very much personalized, while at the same time highly unequal. Employers treated their staff in a paternalistic way, and mobilized ethnicity as a tool to justify this inequality. Still, sometimes Belgian employers became a permanent part of the social network of their (former) domestics, a part that could be mobilized later on in life, long after the immigrants had left their employment. For immigrants working in ‘ethnic workplaces’, where (almost) everyone shared the same ascribed ethnicity, relationships between colleagues could develop in a more ‘normal’ way. As everyone at work spoke the same language, communication was easy. Contacts between co-workers in these workplaces were generally also multi-stranded and, because of the way in which the staff was recruited, many colleagues shared old world and even kinship ties. Further, immigrant workers’ relations with employers were not only characterized by a kind of ‘ethnic solidarity’ which ensured loyalty and stability, they were also multi-stranded and highly personal, with immigrants living in their family homes, sharing old world ties, meeting in clubs and at religious gatherings, etc. During the period under study, however, the large majority of immigrant workers worked in mixed workplaces. Here, they formed a minority, working together with a majority of local workers and – albeit to a lesser extent – with immigrants of different ethnic backgrounds. In these workplaces, they came into contact with colleagues across ethnic boundaries on a daily basis. At the inter-group level, these contacts were riddled with tension. Only local workers managed to voice these negative sentiments openly; immigrant workers did not. Throughout the period under study, in Ghent, tensions

Immigrant Workers’ Relations with Colleagues and Employers  •  95

did  however remain low key and non-violent. There were no outbreaks of large-scale conflicts between groups, only small-scale incidents between individuals. In theory therefore, the general climate did allow for the development of more meaningful workplace relations across ethnic boundaries, and especially between locals and immigrants. However, in practice, these relations only rarely surpassed the level of what Blokland has called interdependencies or transactional relations. The specific motivations of immigrants as target earners as well as their often high degree of job mobility hindered the development of workplace relations that were lasting and meaningful, as did language barriers and the  single-strandedness of most of these relations. Language also played a major  role in the development of relations between immigrant workers and their Belgian employers. If they could communicate directly with their employers, immigrants tried to construct personalized relationships with them. These attempts however were not reciprocated by the employers. High-ranking officials regarded their immigrant employees in a paternalistic way, and among low-ranking supervisors, racist and discriminatory behaviour was a serious problem. The most meaningful social relations, attachments  or even bonds developed by immigrant workers were with co-ethnics. Here too, a shared language was key, as it allowed them to have the kind of conversations colleagues need to have to really get to know each other. But the fact that these relations were often multi-stranded, overlapping with old world ties (including kinship), shared neighbourhoods, club membership, etc., made it easier for them to become more meaningful – and vice versa. Still, this does not mean that working in a mixed environment had no impact at all on immigrant workers’ social integration in the receiving society. Those immigrants who did speak the language or who managed to communicate with their Belgian colleagues in a third language did manage to construct more meaningful interethnic relations, not only within but also outside the workplace. The development of such relations also depended on the age and stage in the life cycle of the immigrants, as young and single workers had more time available to spend with their colleagues than older and married ones. Finally, even the superficial contact that characterized the workplace-based relations between immigrants and locals was not without impact on the lives of the former. The importance of these relationships however only became clear when many of them lost their job from the second half of the 1970s onwards. For those immigrants, especially the men, this resulted in a complete loss of contact with locals, exemplified by the total obliteration of the little bit of Dutch all of them had picked up while working.

96  •  Parallel Lives Revisited

Notes   1. From a 1983–1984 questionnaire with Dutch trade union members, the same idea of locals being discriminated against came to the fore, not only on but also off the shop floor (Lucassen and Lucassen 2011, 81).   2. Meeting of the works council, 24 February 1965, number 1128 UCO Braun reports union delegation 1958–1976, box 16, part II files per company, archive number 147 TACB Ghent section – my emphasis.   3. See e.g. Meeting of the union delegation, 30 October 1969, number 1128 UCO Braun reports union delegation 1958–1976, box 16, part II files per company, archive number 147 TACB Ghent section; Report of the committee meeting, 21 February 1974, box number 4 committee meeting ‘Eendracht maakt macht’, 1973–1975, part I report books etc., archive number 147 TACB Ghent section.   4. Interview with Mohamed J. Not all practicing Muslims found it necessary to pray during work hours. Mohamed E. e.g. thought it was not correct to take time off work for praying, saying ‘One is paid to work, not to pray’. Interview with Mohamed E.   5. However, the research concluded that this was not the consequence of a discriminatory policy on the part of the employers, but of the fact that these immigrants had been hired to fill in those jobs that were avoided by local employees (De Muynck 1972, 52).   6. In times of crisis, the scarcity of jobs causes rivalries between immigrant workers and local workers, as the positions taken by immigrants become more similar to those taken by the locals and they become competitors on the labour market. Khoojinian (2006, 116), looking at Turkish workers in Belgium from the mid-1970s on, describes exactly this evolution towards more competition between immigrants and local workers of the lower classes.   7. Report of the committee meeting, 23 March 1967, box number 2 committee meeting ‘Eendracht maakt macht’, 1965–1967, part I report books etc., archive number 147 TACB Ghent section.   8. Report of the committee meeting 5 September 1974, box number 4 committee meeting ‘Eendracht maakt macht’, 1973–1975, part I report books etc., archive number 147 TACB Ghent section.   9. This is the case in interviews that were carried out by Belgian interviewers, but also in those that were carried out by ‘co-ethnic’ interviewers. 10. Interview with anonymous Moroccan man from Doukala (near El Jadida) who arrived in Ghent in 1966 and started work at UCO Ltd; interview with Ahmed T, a Moroccan immigrant who arrived in Ghent in 1969 and started work at UCO Ltd; interview with Mustafa, a Turkish immigrant who arrived in Ghent in 1967 and worked at Fabelta in 1971; interview with Hüseyim, a Turkish immigrant who arrived in Ghent in 1971 and started work at ACEC Ltd; interview with Halil (Hallil in the list), a Turkish immigrant who worked in the cardboard factory Pol Madou in 1971; interview with Soleiman, a Turkish immigrant who worked for Fabelta in 1971; interview with Ahmed X. and his wife, both Turkish immigrants who worked for Fabelta and UCO-Desmet in 1971; interview with Hassan, a Turkish immigrant who worked for an UCO factory in Lokeren in 1971; interview with Duz, a Turkish immigrant who worked for UCO-Galveston in 1971; interview with Mustafa (2), a Turkish immigrant who worked for a small building company in 1971; interview with Rabah, an Algerian immigrant who arrived in 1973 and started work at Fabelta Ltd. 11. Interview with Mohamed H.

Immigrant Workers’ Relations with Colleagues and Employers  •  97

12. For Antillian immigrants in France in the 1960s and 1970s, e.g., Anselin noted a virulent racism at the workplace (Anselin 1979, 122–25). 13. Interview with Mehmet S. 14. Interview with Yusuf E. 15. Interview with Ali G. 16. In a recent study into the dynamics of workplace relationships within a highly diverse workforce, Ogbonna and Harris point to language as one of the two most important manifestations of diversity on the shop floor, together with religion (Ogbonna and Harris 2006, 392–94). 17. Interview with Dahmane. 18. Interview with Hüseyim Y. Again, these interviews were carried out in rather blunt terms, with questions pushing the interviewees in certain directions and limiting their own input in the interview. Still, the gist of the interview does give an idea about the impact of language on the construction of social relations at work. 19. Interview with Celyn C. 20. For an autobiographical essay on the importance and eventual disappearance of the French language in Flanders and Ghent, see Beyer de Ryke (2002). 21. Most of the incidents people recalled during the interviews were related to civil servants, police agents, doctors etc., not to their direct colleagues, employers or neighbours. Interviews with Charif, Mohamed E. and Club Belgo-Tunisien. 22. Interview with Amar S. 23. Mohamed A., a Moroccan immigrant who arrived in Ghent in 1964, at age twentytwo, remained unmarried until 1966, and worked for UCO during the whole of his professional life in Ghent, also reports going out with his colleagues. Interview with Mohamed A. So does Mustafa, a Turkish immigrant who arrived in Ghent in 1967, at age twenty-five, first unmarried and then married to a local girl; when he worked for a small building firm for some time, his co-workers took him with them when they went out. Interview with Mustafa. 24. Interview with Charif, an Algerian immigrant who arrived in Ghent in 1965, at age twenty-seven, and remained unmarried until 1969; he started work at UCO and worked there until his pension. 25. Mohamed S., a Moroccan immigrant who arrived in Ghent in 1966 at age twentyfive, and remained unmarried until 1970, played amateur football in the UCO team; he worked at UCO for five years, until 1971. Ahmed T., a Moroccan immigrant who arrived in Ghent in 1969 at age twenty-three, and remained unmarried until 1976, also played in the UCO team, together with Belgians and immigrants of different backgrounds. Interviews with Mohamed S. and Ahmed T. 26. Martini (2008, 381–82) clearly demonstrated these differences between small and big companies in the building sector. 27. Interview with Celyn C. Ömer K., a Turkish immigrant from Piribeyli who arrived in Ghent in 1975, explains his lack of knowledge of Dutch by the fact that he has not worked in Flanders: first, he worked in Marche-en-Famenne (Wallonia) for one year, where he picked up some French; then he moved to Ghent, but went to work in Berlare and Brussels, where he kept speaking French even with his Flemish colleagues. Ten years after his arrival in Belgium, he became unemployed, and he hadn’t worked since then when he was interviewed in 1993: ‘I worked in Wallonia and there, I had to learn French. Here, I am without work, and I have no contacts anymore with Belgians, so I cannot learn the language.’ Interview with Ömer K.

98  •  Parallel Lives Revisited

28. Interview with former head of personnel at UCO Ltd, Willem: ‘Also a reason that we, we never, never, once, but that was already a lot later, I was already head of personnel, and I had a meeting, in Galveston, with the works council, and there was somebody who, blast him, dared say “well, we are discriminated against here”, and I said, “look, I don’t want to hear that anymore, because it isn’t true”. I said, “if it is true, go to your director, and if you still think you are not treated in all fairness, you can come to me immediately”. I never saw anyone.’ Interviewer: ‘That was a foreigner?’ Willem: ‘Yes, a foreigner who said that, but well, that was … that must have been in the 1990s. Before, I had never heard that. I never heard anything about it afterwards anymore.’ From the interviews with Willy, Paul and Amand, I also got the impression that they did not think they treated their foreign employees differently than their Belgian employees, even though they did not stress this as much as Willem did. 29. When Ahmed B. was bullied by his foreman while working for Fabelta, the personnel manager of the company quickly stepped in, ensuring that the bullying was stopped immediately. Interview with Ahmed B. 30. Interview with Amand. 31. Interview with Paul. 32. ‘And what they also did sometimes, was bringing something from their country, a copper bowl or that kind of thing. Normally I don’t accept that, but well, they are standing there, what do you do then hey? It is also, I think, if you don’t accept that, it is an affront for them hey? Then they say, “well, we aren’t good enough” or “the gift isn’t good enough”.’ Interview with Paul. ‘Well, of course they wanted to go to their home countries during the holidays, which was logical. And when they came back, they sometimes brought something with them. Once they brought me a pair of shoes, those with the curls … that was a Turk. And a bottle, one or another alcoholic drink, but without label. That one I gave away, because … (laughs) … That happened a couple of times.’ Interview with Willy. 33. Interview with Ömer; interview with Charif; interview with Mohamed T. 34. Interview with Yusuf E., who had been working as an ‘interpreter’ in the same company for nearly twenty years at the time of the interview; interview with Kazim, who worked as an ‘interpreter’ in the UCO De Smet-Guequier factory for some years in the 1960s; interview with Ömer, who worked as an interpreter in different textile factories for some years in the 1960s. 35. This was certainly the case for Ömer, who developed side activities such as recruiting new workers and making them pay for his services while he was working as an ‘interpreter’. Interview with Ömer. 36. Interview with Kazim. 37. Interview with Ömer. 38. The workforce of mosaic and terrazzo maker Antonio Bearzatto (see chapter 2) did not speak Italian to each other at work, but Friulian. They even sang it. Interview with Enzo. 39. Interview with Enzo, who worked for Antonio Bearzatto with his brother Angelo; interview with Girolamo, who worked with a number of friends from his home town; interview with Elio, who worked with a friend of his, with whom he had worked in Italy; interview with Piergiorgio, who worked with a colleague who was indirectly related to him (his cousin was the godfather of Giorgio’s little brother). 40. The concepts ‘single-stranded’ and ‘multi-stranded’ relationships, meaning relationships that contain one or more than one focus of interaction (e.g. kinship, friendship, common religious belief, economic obligation, etc.), come from Mitchell (1969, 22). 41. Interview with Girolamo.

Immigrant Workers’ Relations with Colleagues and Employers  •  99

42. Elio’s employer, one of the Tondats, went to the Italian mass and to the Dante Alighieri club, where Elio himself also went. Interview with Elio; in one of his anecdotes, he also mentions meeting the sister of his employer in a tea-room where he went. 43. Girolamo, Enzo, Elio and Piergiorgio all lived in the house of their employer (Elio in the house of his employer’s mother) for several years. 44. Interview with Enzo; interview with Clelia; interview with Piergiorgio. 45. Interview with Girolamo. Piergiorgio also really felt ‘at home’ in his employers’ house: ‘It was my family, they were my parents, because, you know, for me, they were my second parents, and actually, if you want, we can even say my first parents, as I spent more time with them than with my real parents.’ Interview with Piergiorgio. 46. Interview with Enzo. 47. This was the case for Girolamo, who worked with his current employer until the latter took over the business; it was also the case for Vito, who started work at Italia Grill for an employer who had worked together with him in ‘All Parma da Gianni’, the restaurant of his cousin. Informal conversation with Vito. 48. Interview with Enzo, who worked for and lived with the Bearzatto’s; interview with Piergiorgio, who worked for and lived with the Zangrando’s; interview with Elio, who worked for the Tondat family, another family of mosaic makers; interview with Isabel, who worked for Supercolor; interview with Jose Luis and his wife, who also worked for Supercolor. 49. Interview with Enzo. 50. Girolamo, who said he was treated like a son by his first employer, also sounds a lot less positive when talking about his working hours. Officially, he had to work from 10 am to 3 pm and from 6 pm to 10.30 pm; however, he says, these hours were never really respected, and they had to stay at work until all the customers had left. Interview with Girolamo. Others tell similar stories. Interview with Elio; interview with José Luis and Jesusa; interview with Maria Isabel. 51. Interview with Aurora. 52. ‘Crombeen’s oude en nieuwe wereld’, unpublished journal of the Franciscan Sisters of the Crombeen, September 1967: 31–32. 53. Interview with Lucien. 54. Interview with Maria del Pilar. 55. Interviews with Maria de la Luz (had worked in France) and with José Luis and Jesusa (had worked in France). 56. Vanoutryve (1968, 61) found that three out of six employers she interviewed could speak Spanish, and that the others could make themselves understood in French (implying that their staff understood French). The number of Spanish-speaking employers however seems out of proportion, especially when compared to the information I got from the interviews with former domestic servants. 57. Interview with Lucien. 58. Interview with Maria del Pilar.

Chapter 4

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood 

Up until now, the analysis has focused on the integration processes of immigrants at work. Working as a wage earner was an experience that many Mediterranean immigrants shared. However, the workplace was not equally important to all. Especially for women, the neighbourhood tended to be a much more relevant site of integration. Therefore, in the next two chapters, more attention will be paid to the experiences of women. The first part of this chapter however is an exception to this. It looks at the initial housing experiences of immigrant workers arriving in Ghent on their own. These first housing experiences are treated separately, as they often had a far-reaching impact on the further spatial and social integration of newcomers in the city. This was certainly also the case for immigrants arriving in Ghent to work as domestics. However, their housing experiences were very particular, inextricably linked to their working situation. Therefore, they will be treated separately later on. As few female immigrant workers made their way to Ghent independently outside the context of domestic services, the focus here is exclusively on the experiences of men. To a certain extent, this chapter mirrors chapter two, moving the focus from labour to housing and residence. However, as the spatial component is more crucial for housing, it pays special attention to the issue of spatiality, building on the spatial analysis of our housing data as well as on contemporary studies dealing with (immigrant) housing in Ghent. For the qualitative analysis, the chapter makes use of immigrants’ descriptions of their housing experiences.

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  101

Mediterranean Immigrants’ ‘First Housing’ in Ghent When arriving in Ghent, the very first thing immigrants had to secure for themselves was a place to spend the night and, if their stay was to be prolonged, to live. This ‘first housing’ of immigrant newcomers rarely receives special attention in the literature. However, like their first job – and often strongly intertwined with it – immigrants’ first housing could have an important impact on their ensuing trajectories in the receiving society. The kind of housing they initially lived in and its spatial positioning in the city impacted upon the (kind of ) people they initially met, the first impressions they got of the city and its inhabitants, and even the places they went to live in at a later stage. At a group level, this is very clear for the Tunisian population in Ghent, many of whom were first housed in an upper-middle class white neighbourhood and ended up staying there, often marrying a Belgian partner (De Bock 2013, 303–33). Immigrants who were what we have called ‘dependent partner migrants’ generally had their partner arrange their first housing for them, and simply moved in with them. The same holds true for those independent labour migrants coming to join family or friends. Their housing situation therefore was similar to that of more settled immigrants, which will be analysed further on. Here, we will focus on those immigrants who were not housed by relatives or friends, but ended up in a different sector of the housing market, either on their own or through their employers. Those labour migrants who arrived in the city without a previously existing network and without a job contract in hand generally had to find a place to sleep for themselves. These immigrants tended to end up in lodging houses. In Ghent, during the period under study, there were a number of privately owned lodging houses, whose small rooms at (relatively) low prices attracted people who were planning to stay only temporarily. These lodging houses were spread over the urban territory, with a concentration in the city centre. They were especially important in the mid-1960s, when many pioneers arrived in Ghent. Apart from lodging houses offering accommodation to large numbers of lodgers, immigrant newcomers also rented rooms above cafés and snack bars.1 Initially, these rooms were let by local publicans, men and women, but quickly, immigrant newcomers who had set up their own cafés also started letting out rooms, like the Turkish owner of café Schoonzicht in the ‘immigrant’ neighbourhood Sluizeken-Muide.2 In these lodging houses, immigrants slept with two or three people in the same room (or more, if the rooms were dormitories), sharing kitchen and bathroom facilities. Immigrants who were recruited by local employers often ended up in this kind of housing as well. For those who had been officially recruited,

102  •  Parallel Lives Revisited

Figure 4.1  Article by A. De Keuleneir, decrying the bad housing conditions of immigrants living in lodging houses, Vooruit, 28 February 1964. Amsab-Institute of Social History, collection Vooruit

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  103

e­mployers were legally obligated to provide them with housing for the duration of their contract, most commonly one year.3 For those who were not officially recruited, no such obligation existed. In practice however, many employers also provided these workers with housing, in an attempt  to  stabilize this highly mobile workforce or as a way of gaining a competitive  advantage when the demand for labour was high. In the early period of Mediterranean labour migration to Ghent, newly arriving immigrants who started working for textile giant Union Cotonnière (later UCO Ltd) were generally offered one of the old workman’s or beluik houses (see below) that were part of the company’s housing stock. In the mid1960s, the company offered this kind of housing not only to the immigrant workers it had recruited abroad, but also to those who had arrived in the city on their  own initiative. This was the case for Ömer, who arrived in Ghent  together with a friend in 1964. When they went to look for work at one of the company’s factories, they  were not only offered a job, but also housing, furniture and everything they needed to settle in.4 Two years later, in 1966, Ali Gh., a Tunisian immigrant who was recruited in Tunisia by the Union Cotonnière, was offered housing in a beluik just next to the factory where he was due to start working. However, as opposed to those immigrants who had arrived  spontaneously and were happy to find any accommodation, Ali and his compatriots did not accept the housing they were offered. They demanded to be rehoused immediately. This demand was met by the Union Cotonnière, as it was bound by contract to provide ‘decent housing’ to its recruits, and probably  realized it could not easily appease these immigrants, who were not illiterate farmers but well-educated white-collar workers: So they showed us where they wanted us to live, and it was not pleasant, that’s the first thing. We found a room below and a room above, the toilets outside, the water outside, well, these were the houses of the textile workers in the past. … In any case, I was not happy, as I had never lived in such circumstances. So I gave the director an ultimatum, I said ‘Here is an ultimatum, if you don’t find us suitable housing, where we are comfortable, we are going back to Tunisia.’ Because I am not a slave, I have always made a decent living, I come from a respectable family, I have a nice house where we live well. But here, we are housed like slaves, really like slaves! And so, after a while, two, three months, they gave us a house in the Vrijheidslaan, a fancy street, where only rich people live (laughs).5

From then on, UCO Ltd would offer the accommodation in beluiken and old workmen’s houses mostly to newcomers who arrived on their own initiative. Those who had been recruited from abroad, as well as a number of Algerian interns who came to Ghent to be trained in the UCO factories in

104  •  Parallel Lives Revisited

the late 1970s and early 1980s, were lodged in modern (apartment) buildings in uptown residential neighbourhoods.6 Companies that did not have their own housing stock did not generally provide accommodation for their immigrant workers, unless they had been officially recruited. Fabelta Ltd rented housing especially for its Tunisian recruits, but not for other immigrant workers.7 To save expenses, for the sake of both the company and the immigrants (who were required to pay their own rent), bedrooms were shared between two or three men. As such, there was little space or privacy in these company houses.8 Still, even though it was certainly not luxury housing, Fabelta’s company housing was decent, and situated in the better residential neighbourhoods of the city. The situation of a group of Italian immigrants who had arrived in Ghent with one of the big building companies working in the Canal Zone was very different. They were living in what can only be described as barracks on the building site of the SIDMAR steel factory, between 1965 and 1967. Their living conditions were abominable. Such a setup was typical for immigrants working on large building sites, and has also been seen elsewhere in Europe in that era (Castles and Kosack 1985, 252, 293; Vogel 2005, 125 ff.). A final and very different instance of employer-provided housing was the situation of immigrants who were housed in the private home of their employer. This was the case for a number of immigrants working for small immigrant-led businesses, who had been recruited directly from their home country, such as the Friulian mosaic makers working for Antonio Bearzatto and some of the staff of the Italian restaurants that were popping up everywhere over the course of the 1970s (see chapter two).9 Such housing in the employer’s home has also been attested elsewhere, for example among Italian terrazzieri, figuristi and gelatieri in the Interbellum in Amsterdam and Rotterdam.10 In a different way, it was also the case for immigrants who worked as live-in domestic servants. These very specific housing situations had a particular impact on the integration processes of immigrants in the receiving society. To some extent, this impact has been discussed in the previous chapter, from the point of view of employer-employee relationships. It will be further developed in the next one. This initial stage of ‘first housing’ could last as little as a couple of days, but could also span a longer period, depending on the kind of housing and the needs of the immigrants living there. Lodging houses only served as a short-term solution for most immigrants. Dahmane, an Algerian immigrant who had been recruited by the Union Cotonnière in 1964, stayed for only about one week in the lodging house his employer had referred him to, and about two months in a room above a café.11 Kamel, another Algerian immigrant, reckons he spent two to three weeks in the lodging house provided by his employer.12 Mustafa, a Turkish immigrant who was a pioneer in Ghent,

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  105

stayed in a lodging house for five months before he moved out.13 The data from the individual immigrant files confirm that such short to medium stays were the norm.14 This situation stands in contrast to what was going on in France, where lodging houses became semi-permanent dwellings for many immigrant workers.15 This is probably related to the fact that in those years, in France, the housing opportunities for newcomers were much more limited than in Belgium. The only immigrants who seem to have stayed for a considerable time in their ‘first housing’ in Ghent are those who were offered a year-long contract by their employer, as well as those who were housed in the home of their employer. The latter – especially if they were still young boys – often remained there for quite some time, until they were old enough to go and live on their own or had met someone with whom they moved in. Girolamo, an Italian immigrant from Bisceglie, left the house of his employer only after four years, at age nineteen. Enzo, also Italian but from Friuli, lived with his employer’s family for five years and moved out when he was twenty.16 For these young men, leaving the house of their employer was not directly related to changing jobs. This was very different for the immigrant women and men working as live-in domestics, as for them, employment and housing were obviously inextricably linked. An enquiry among Spanish domestics in 1968 showed that these immigrants, and especially the single women among them, chose to change workplace and thus place of residence quite regularly. Others quickly gave up work in domestic services and changed to another economic sector in order to gain more personal freedom. Having one’s own place to live was of paramount importance in making the decision to leave (Vanoutryve 1968, 65, 67, 97–98, 102–4).

Immigrants as Replacement Dwellers in the Secondary Sector of the Private Rental Market Individual Immigrants At a certain point in time, those immigrants who decided to stay had to move out of their ‘first housing’ and find a more permanent place to live. At this stage, the large majority of immigrants were in Ghent on their own without a partner or children, and with the intention of going back soon. In correlation with their objective to earn as much money as possible as quickly as possible, when looking for housing, most of these target earners were only interested in limiting their expenses to a minimum. Because of this, they ended up renting those dwellings that belonged to what we could call the ‘secondary sector’ of the housing market, where housing was at its cheapest, but also at its worst. There, they fulfilled the role of ‘replacement dwellers’, taking up those

106  •  Parallel Lives Revisited

dwellings that had been vacated by their previous inhabitants, and for which local tenants were hard to find. This analogy between the roles of immigrants in the labour and housing markets is very helpful for an understanding of their integration processes. It has also been made by Martens and Wolf in their work on the position of foreigners on the Brussels’ housing market in the early 1970s (Martens and Wolf 1974, 5). The phenomenon of ‘replacement dwellers’ was not unique to Belgium, but could be found all over Western Europe, as the better-off sections of the indigenous working classes moved out of their houses to council housing or to suburban estates, and only the most marginalized groups in society were left behind to compete for dwellings in the residual rented sector (Bovenkerk et al. 1985, 33–35). In Ghent, many immigrants were housed in one of the city’s decrepit nineteenth-century beluiken. In 1967, for example, forty-two out of fiftyfive streets where Turkish immigrants were registered upon receiving their first work permit were beluiken. These were a kind of cul-de-sac constructed by factory owners and speculators in the nineteenth and early twentieth centuries on what used to be courtyards or gardens of other houses. The houses in these beluiken were generally very small, mostly made up of only two rooms of approximately four by four metres each, one on the ground floor and one above. They were heated with coal. The shared toilets and water supply were outside in the courtyard (‘Fotobijlage “Gent 1980 Goddomme!”’ 1980). However, not all labour migrants lived in this kind of housing. In the mid-1960s, a considerable number of immigrant labourers rented a room in a newly constructed high-rise apartment building with all modern comforts, where they lived together with Belgian students (Bernabé 1968, 93). Others succeeded in renting a furnished room in a private house. This was the case for Kamel, an Algerian immigrant who arrived in Ghent in 1963, who rented a garret in the house of an old Flemish lady.17 Yet others found an apartment, which they typically shared with other immigrants, like Moh, a Moroccan immigrant who arrived in Ghent in 1966 and shared a one-bedroom flat with a colleague.18 Still, even though these places were generally more decent than the beluik houses, they could hardly be considered part of the ‘primary sector’ of the rental market. They were often very small (for example, the rooms in the new apartment block measured only 8.75 m2), and facilities such as toilets and showers (if there were any) were generally to be shared with other inhabitants of the house or building (Bernabé 1968, 93). However, at this point in their migration project, renting cheap housing, independent of its quality, was the option that best suited the wishes and needs of most of these labour migrants. Flavia Cumoli (2012), in her ­comparative study of Italian workers in Belgium and northern Italy,

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  107

has shown how their bad housing conditions in Belgium were intrinsically related to the idea of an imminent return and the will to reduce living costs to a minimum. She contrasts this situation with the housing of Italian immigrants in the suburbs of Milan, whose migration project was much more definitive and who were therefore prepared to invest more in housing (Cumoli 2012, 30). Indeed, as their migration project changed from a short-term to a longer-term endeavour, Mediterranean immigrants in Ghent also tried to find a better place to live.19 Immigrant Families As we have seen, this change was generally first signalled and then hastened by the arrival or establishment of the family, which caused a modification of the immigrants’ housing needs and objectives. However, it only rarely meant that immigrants managed to leave the secondary sector of the housing market. A study by the University of Ghent, carried out in 1978, found that many ‘guest worker families’ were still living in beluik houses. At the same time, the study concluded that about one third of the city’s beluiken needed extensive, structural renovation or rebuilding in order to become inhabitable (Balthazar 1978).20 In the reports of the plenary meetings of ‘Woonfonds Gent’ – a local tenants’ association that was set up in 1978 – we find further indications of how problematic the situation really was: December 1980: the Akgul family: seven people, live in the neighbourhood Sluizeken-Muide and wish to stay there. Their current residence is in a bad condition: the roof is broken, the house is humid. February 1981: the Coban family: arrived in Ghent recently, live together with the husband’s brother, as they cannot find an appropriate place to live. The family has searched for a house for a long time, but is always refused. April 1982: Arife Palit: They live with two families (five people) in a onebedroom house that is due for demolition. January 1983: Abdil Kiran: Family of five people, their house in the beluik Ossenstraat has been declared unfit for habitation.21

During the period under study, most immigrant families remained stuck in bad quality rental properties. Their housing trajectories were shaped not only by the realities of demand and supply, but also by an increasingly discriminatory attitude towards them, on the part of home owners as well as (semi-)public officials. Like elsewhere in Western Europe, immigrant families in Ghent collided with racist discriminatory behaviour in their search for a house – much more so than in their search for a job (Castles and Kosack 1985, 267–68; Byron and Condon 2004, 135–38). In Ghent, this d ­ iscrimination

108  •  Parallel Lives Revisited

got worse over the course of the period under study. According to the immigrants who arrived in the city in the early and mid-1960s, finding a place to live was initially not that difficult.22 Some pioneers explicitly say they found the housing situation in Ghent rather good, compared to other places in Europe. Gazi, a Turkish immigrant who arrived in 1966, recalls how he eventually moved back to Ghent after having spent some time in the Netherlands, because he could not find a place to live there.23 For Algerian immigrants coming from France in the period just after Algerian independence, the housing situation in Ghent was infinitely better, as Charif explains: So, in France, they did not want to provide lodging. To work, it was good, to make money, it was good, but for housing … They did not want to give it, especially to foreigners. Especially because, after the Algerian independence, we became Algerian. Before, when we were French, everything went fine, but in 1962 we became Algerian, and this made it hard to find a house in France. That is the main reason why I came to Ghent.24

From the late 1960s onwards, this ‘openness’ of the rental sector towards immigrants seems to have changed dramatically, turning into a situation of extreme discrimination by the beginning of the 1970s. Miloud and Kamel, two Algerian immigrants who arrived in Ghent in the early 1960s, situate this change over the period 1966–1968. During those years, they had returned to Algeria. They found that when they came back to Ghent, things were no longer as before.25 The Reasons for Discrimination It is possible that the arrival in the city of larger numbers of immigrants from the mid-1960s onwards had an immediate negative impact on owners’ attitudes towards immigrants. All over Western Europe, houses in areas that became known as ‘immigrant neighbourhoods’ tended to depreciate, as local tenants and buyers perceived such areas as less appealing and began to avoid them. Therefore, many owners, especially those with several properties in the same neighbourhood, tried to keep immigrants out in order to preserve the value of their real estate (Castles and Kosack 1985, 283). The increasing reluctance of Ghent landlords to let their place to foreigners could also have been due to a worsening image of the housing habits of immigrant families. Robin Roeck, a sociology student who studied Turkish migration to Ghent, noted in his 1968 Masters thesis that ‘practically only the owners of hovels want to let their houses to Turks. The others are too afraid of the lack of hygiene and upkeep from the side of these foreigners, so that they prefer to leave their property vacant’ (Bernabé 1968, 93). The reputation of immigrant

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  109

tenants did not get any better over the following years. From a number of interventions in the assembly of the city council in 1977, we get a startling picture of how at least some – and probably the majority – of locals regarded the housing habits of the immigrant population. Reacting to a suggestion made by another Deputy to solve the housing problems of certain groups by letting the city of Ghent act as a mediator between them and the owners, city councillor Herman-Michielsen addressed the assembly as follows: In short, it boils down to the fact that the municipality – so in fact society at large – would be responsible towards the owners of those buildings for which the tenants fail to fulfil their obligations. In this context, the word ‘guest workers’ seems to have come up, precisely because in that group of people, many failures to fulfil their rental obligations have been noted, and are only rarely sanctioned, which might be specific for this group. No one will deny that in our city – as in many other cities – a rehabilitation of the housing stock is necessary. Some dwellings that fall below the current norms of modern conveniences have to be rebuilt or pulled down. … In the group of people who rent this kind of housing, guest workers take up an important percentage. However, we have to establish that, as inhabitants of these dwellings, they do not submit to the urban regulations that, I believe, are obligatory for every citizen. A lack of comfort of course doesn’t stimulate care, but stacking garbage in cellars, throwing dirt in the sewers causing them to block, and destroying door and window frames is unacceptable.26

Stories of Turkish tenants stripping their house of its woodwork and burning it to keep warm in winter, butchering sheep and chickens in their courtyard and ditching the offal down the drain, and keeping their garbage stacked away for months instead of putting it out to be collected, are a recurrent theme. They appear not only in the interventions of city councillors, but also in interviews with former social workers, employers, and even some immigrants, the latter either non-Turks or Turkish people from the cities, accusing ‘the people from Emirdağ’ (the cluster of villages where the majority of Turkish people in Ghent came from) for their peasant-like behaviour.27 Whether these stories are real or fictitious is less important. More important is the fact that they probably caused serious damage to the reputation of immigrant tenants, leading Belgian owners to change their attitude towards them. Turkish immigrants in particular seem to have got a bad name, but the refusal to rent to immigrants applied to all those who sounded or looked different from the average Belgian (De Bock 2013, 355–56). An important third element contributing to the increase in discrimination was the change in the kind of housing immigrants were looking for. Whereas at first most individual immigrants settled for dwellings that could no longer attract local candidates, from the mid-1960s onwards, more and

110  •  Parallel Lives Revisited

more ­immigrant families were looking for good quality housing, which put them in direct competition with Belgian families. In fact, bad quality housing did remain accessible to immigrant families throughout the period under study. Habib and Ahmed, two Tunisians who had a lot of trouble finding a decent place to live, saw renting such housing as a last resort and yet often the only available solution: Habib: We had to make do with something that wasn’t really good, a dwelling that did not attract any Belgians, for example, so that the owner had no other choice than to rent it to whomever wanted it. Ahmed: So we would find a house, for example where all the living units shared the same toilet, the same bathroom, etc. That kind of housing you could find.28

Finding a house in the primary sector of the rental market was already difficult, as demand far exceeded what was available. For immigrant families however, it was almost impossible, because of the enormous problem of discrimination. Alternative ways of moving up the housing ladder were to become a home owner or gain access to social housing. For immigrant families during the period under study, these were however far from self-evident. Social Housing and Home Ownership From the mid-1960s, the Belgian National Housing Association had been in favour of allowing immigrants access to social housing, and from 1 January 1974 onwards, equal access was stipulated by law (Hubeau 1994, 748, 871).29 In practice however, during the period under study, immigrants were severely underrepresented in social housing, not only when their weaker socio-economic position was taken into account, but even compared to their share in the Belgian population in general. Data for Ghent are available only from 1975 onwards. Even then, they show that only a fraction of the Ghent social housing stock was let out to non-Belgians, who made up nearly 7 per cent of the local population. That year, the most important social housing association in the city, the Ghent Housing Association, rented only 14 out of 3.287 or 0.4 per cent of its housing to non-Belgian families. The other housing associations did only slightly better. Where nationalities were mentioned in the sources, no Turkish immigrants appear, even though by then they formed the largest immigrant group in the city.30 Five years later, the Ghent Housing Association still only housed 38 immigrant families in a total of 3,217 dwellings, just over 1 per cent of its housing stock31 – a very low percentage compared to the 6 per cent non-Belgians in the total population.32 This situation was characteristic of the whole of Western Europe (Castles and Kosack 1985, 304–9). The reasons for it are not ­straightforward, but

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  111

here, also, discrimination clearly played a role. In Ghent, it took a long time before this could be proven. Officially, the near-absence of immigrants in the social housing stock was thought to find its roots in their lack of information. In November 1972, all providers of social housing assured the city councillor of social affairs that they made no distinction between locals and immigrants in the allocation of dwellings.33 However, when, from the late 1970s onwards, the tenants’ association Woonfonds tried to enrol its immigrant members on the housing associations’ waiting lists, the latter did not respond positively. Finally, suspicions of widespread discrimination against immigrant candidates were confirmed when in 1982 the governing board of ‘De Goede Werkmanswoning’ decided to no longer grant housing to the non-Belgian candidates on its waiting lists or even allow new immigrant candidates to enrol.34 Other associations did not go that far, but did not really try to become more accessible to immigrant families either. Also striking is the refusal of the city authorities to make use of a grant they had received from the EEC in order to renew part of the city’s housing stock, as the conditions of the grant stipulated that a large part of the newly created housing had to be allocated to immigrant families, and this in a ­neighbourhood where many immigrants were already living.35

Figure 4.2  Protest in front of the city hall for better housing, 1978. Amsab-Institute of Social History, collection Devreese

112  •  Parallel Lives Revisited

Social housing in Belgium however was scarce, especially when compared to other Western European countries (De Decker 2001, 20). Becoming a home owner was a much more common way to move out of the private rental sector. By 1961, the majority of Belgian families owned their own house, and levels of home ownership increased throughout the period under study (De Decker 2001, 19; Watson 1971, 5–6). Even in the cities, where owner occupancy was less important and social housing was more prevalent, the owner-occupied sector remained the biggest one. In 1970, it accounted for 36 per cent of occupied houses in Ghent.36 Buying a house and becoming a home owner has been reported as a residential strategy for immigrants in many European countries (de Villanova 1997, 109–10, 115–17; Van Kempen 1997, 180, 183; Byron and Condon 2004, 150–51; Glebe 1997, 144). In Ghent, this strategy only became important from the mid-1980s onwards, when most immigrants had come to accept that they had become permanent residents, and that their return project would not be realized anytime soon.37 By 1994, some 80 per cent of the Turkish population in the city was estimated to be owner-occupiers, and this compared to only about 37 per cent (for the city centre) to 48 per cent (for the whole territory) of its Belgian population in 1991.38 Over the course of the 1960s and 1970s however, home ownership among immigrant families remained extremely rare.

Immigrants’ Housing Careers: From Mobile Movers to Stable Home Owners Starting Out with a High Residential Mobility The fact that throughout the 1960s and 1970s immigrants did not manage to move out of the secondary sector of the housing market does not mean that they remained stuck in the dwellings they initially moved into. As was the case in the labour market, immigrants tended to be quite mobile. The literature points to a higher than average residential mobility among this population.39 Especially during the first years of their stay in Ghent, immigrants had a relatively high degree of intra-urban mobility. Our data show how one third of independent Mediterranean immigrants who stayed in Ghent for five years or longer moved house more than three times within their first five years; 14 per cent moved three times, 15 per cent two times and 18 per cent one time.40 These data probably underestimate the average mobility of immigrants in Ghent, as the more mobile among them, who stayed less than five years, were not included. A 1967–1968 survey on Turkish immigration to Ghent showed that 62 per cent of Turkish households had moved house three times or more during their stay in Ghent – which for 92

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  113

per cent of them was equal to or less than two years (Bernabé 1968, 110, 127). A 1972 survey, carried out among the ‘guest worker population’ of the province of East-Flanders,41 showed that one in three immigrant families had moved house once, one in three had moved two or three times and about one in ten had moved four or more times since their arrival in Ghent. The average length of residence of these people was 5.29 years. On average, Algerian immigrants had moved more often than Moroccan immigrants, who in turn were more mobile than Tunisians and Turks. The authors of the publication following the survey explained these differences in mobility by referring to the differing lengths of stay of these groups. On average, those who had arrived earlier had moved house more frequently than those who had arrived later, with the exception of the Tunisians who, even though they had arrived last, had moved more often than the Turks (Janssens 1973, 14, 16, 27). To a large extent, immigrants’ highly mobile residential profile during the first years of their stay was inherent to their position as tenants in the residual rented sector of the housing market. Immigrants often became the victims of the arbitrary behaviour of their landlords and landladies, or had to move because the dwellings they were living in were simply too derelict. Yet, the decision to move was not always forced upon them. Often, they themselves decided to leave and move into housing that better suited their needs and wishes. As they generally started off in the very lowest strata of the housing market, the chance that they would want to find better housing was quite high. Moving house as a positive choice to seek more living space and comfort seems to have been a decision that was more common to immigrant families than to single immigrants. On the one hand, this was because families simply needed more space. For Pakistani families in the United Kingdom, for example, Bowes et al. found that the chief factor influencing their decision to move was space, related to changes in household size (Bowes et al. 2000, 172). This is in fact the most common reason for all people to move house, whether they are immigrants or not (Clark 2012, 66–67). On the other hand, such decisions seem to have been pushed by the female rather than the male partner in the household. Many immigrant women, who came to join their husbands in Ghent, were indeed much less ready to accept the bad living conditions the latter had become used to – possibly because they generally had to spend a lot more time in the house. The reaction of the mother of Fatma K., a Turkish immigrant who arrived in Ghent in 1971, is one encountered in many stories: I was very disappointed. I hadn’t expected it to be that bad. The toilet was outside and we had to share it with other houses. The cockroaches crawled over the walls of the toilet. There was one room and that was the living room. We

114  •  Parallel Lives Revisited

had almost no running water for the whole house. Behind the door, there was a tap without a sink, where you could take water with a bucket. All of us slept in the same room. My mother cried for days.42

According to Fatma, it was the disappointment of her mother that instigated their move to a bigger and more modern property three months after their arrival. Maria del Pilar, a Spanish immigrant who arrived in Ghent together with her husband in 1964, was not so patient. When she saw the beluik house her sister-in-law had rented for her – with the wooden toilet and the water outside in the street – she simply refused it. She was not prepared to live in such bad conditions, even if it meant losing the deposit.43 In another story, Yamina, a Moroccan immigrant who came to Ghent to join her husband in the late 1960s, recounted how she was the one who took matters in hand when she thought the house was getting too small, and asked around among her Belgian acquaintances in order to find a bigger house.44 Towards a More Stable Position Over Time As was the case for their professional mobility, Mediterranean immigrants’ residential mobility decreased significantly throughout the period under study. Similarly, the reasons for this can be situated on two levels: that of the characteristics of the immigrant population itself, and that of the changing opportunity structure. Over the course of the 1960s and 1970s, an ever larger part of the Mediterranean population in Ghent came to consist of immigrant families (see chapter one). This greatly stabilized the immigrant population, not only in their mobility between different localities, but also in their intra-urban mobility. Immigrants living on their own more frequently had very short housing episodes than immigrants living together with their family. Nearly half of the housing episodes of ‘single immigrants’ lasted for less than six months, whereas such short stays at one address made up less than one quarter of immigrant families’ and one fifth of mixed Belgian-immigrant families’ housing episodes respectively. Furthermore, housing episodes of more than five years were much more frequent for immigrant families (10 per cent) and especially mixed families (24 per cent) than for single immigrants (4 per cent).45 Indeed, the formation or arrival of the family in the receiving society is widely considered to have a stabilizing effect. The arrival of children in particular is seen as a major inhibitor for residential mobility, even though at first family expansion often leads to the decision to move (Clark 2012, 71). This ‘familyeffect’ interacted with a more general decrease in mobility as immigrants moved from one place to another, leaving their initial housing – which

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  115

they had found in a period of transition – behind. As we have seen above, independent immigrants’ first address in Ghent was often only meant as a temporary place to live. The data from the individual files show that 10 per cent of independent immigrants stayed at their first (registered) address for less than one month; another 40.5 per cent between one and six months; and only 29.5 per cent for more than one year. Once immigrants had moved away from that first place, their mobility decreased quickly. At the second address, only 4 per cent of immigrants stayed for less than one month; 35 per cent between one and six months; and 37.5 per cent for more than one year. This trend continued, with percentages of 2.5, 30 and 44 per cent respectively for the third address, and 3.5, 22.5 and 50 per cent respectively for the fourth.46 There also seems to have been a change over time in the residential mobility of immigrants independent of the number of places they lived before, and less directly related to the arrival and settlement of families. Immigrants arriving in the period 1976–1980 were generally much less mobile than those arriving in the period 1960–1965, with only 7 per cent compared to 43.5 per cent moving more than three times during the first five years of their stay.47 Apart from the difference in demographic profile between these cohorts, and their different migration trajectories (more chain migrants vs. more pioneers), it was the local opportunity structure that impacted upon their mobility. By the mid-1970s, there was a tightening of the housing market, with fewer houses available (due to slum clearances, etc.) and more  discrimination towards immigrant families looking for a house. Further, with the crisis in the labour market and the drop in professional mobility, some of the reasons for residential mobility (moving to a different job, getting a higher income, etc.) became less prevalent, and more so among the immigrant population than among the population in general. Immigrants’ (initially) high residential mobility was not without consequences for their social integration in the city. Every move from one place to another implies a change in social network, especially if the move entails a  change of housing block or neighbourhood. Studies have shown that intra-local mobility has an important impact on the social, identificational and political integration processes of immigrants in the receiving society.48 As long as they were mobile, it was much harder for immigrants to create long-lasting ties to the people and places surrounding them. Only a more settled housing situation, which mostly coincided with the transition from individual immigrant to family, offered them a better chance to create such ties. However, the extent to which immigrants managed to construct meaningful social relations with neighbours did not only depend on their length of stay at any one place, but also on the kind of neighbourhood they inhabited.

116  •  Parallel Lives Revisited

Immigrant Neighbourhoods, Their Populations and Changes over Time The fact that immigrants were relegated to the secondary sector of the housing market had consequences not only for the living conditions in their homes, but also for their residential dispersion in the city, the analysis of which makes up the remainder of this chapter. Research into immigrants’ spatial dispersion usually makes use of data that reveal concentration patterns per (cluster of ) statistical sector(s). The data that were used here, however, indicate the exact location of a sample of individual, adult immigrants and their families. This kind of mapping allows us to better grasp the patterns of residential dispersion that remain hidden in sectorial data, which automatically draw the attention to areas of concentration and away from those where no or less concentration occurred.49 It also allows us to see ­residential concentrations below the level of the statistical sector, for example on the level of the housing block or street segment. It is only by looking at these concentrations that we can tell whether immigrant families were evenly distributed within the neighbourhoods where they lived, or whether they formed small clusters by concentrating in specific parts (Sule Özüekren and van Kempen 2003, 173; Musterd and Deurloo 2002).50 As we can see on the first map, in 1965 Mediterranean immigrants lived dispersed over the entire territory of the city. This dispersion coincides with the spatial dispersion of the different kinds of ‘first housing’ newcomers occupied in this era, existing mostly of lodging houses and employer-provided housing (see above). The pattern holds true for immigrants of all nationalities, although we can see a stronger presence of Moroccan immigrants in the northern half and of Italian and Spanish immigrants in the southern half of the city. Apart from these general trends, the map clearly shows the existence of several small pockets of residential concentration as mentioned above. In this initial stage of Mediterranean migration, these concentrations were to be found in a number of beluiken, and, at the level of the single address, in lodging houses or employer-provided housing. Over the second half of the 1960s, the Mediterranean immigrant population as a whole remained dispersed over the entire territory of the city, albeit somewhat less in the southeast and to the west of the ring road. However, there was a growing concentration of immigrant residence to the north and east of the city centre. This area is generally known as the city’s nineteenthcentury belt, characterized by nineteenth-century industrial development, composed of factory buildings and workmen’s housing. During the period under study, the whole area was in decay, with several abandoned industrial terrains and high rates of vacancy.51 The housing quality was bad, with most

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  117

Map 4.1  Spatial dispersion of Mediterranean immigrants in Ghent, 1965. Created with GPS Visualizer ©2003–2016 Adam Schneider on an OpenStreetMap © OpenStreetMap contributors

dwellings dating from before WWI and lacking modern comforts such as city gas, central heating or a private bathroom.52 It is here that the majority of the beluiken were to be found (Balthazar 1978, 25–26). The inhabitants of the area were mostly working-class families and lower income groups (such as the elderly), living in the cheap rental properties that were on offer (Claeys and Van Hove 1987, 40; Vanneste 1981, 166–68). The nineteenth-century belt was divided into four districts: Brugse Poort, Rabot, Sluizeken-Muide and Sint Macharius (see the general map of Ghent in the front matter). Over the second half of the 1960s, it was SluizekenMuide that began to manifest itself as the main upcoming area for immigrant settlement. At the same time, the Brugse Poort neighbourhood also came to attract more immigrant families. The families that came to be concentrated in these neighbourhoods were mostly Turkish and Moroccan. Spanish, Italian and Algerian households remained more dispersed.

118  •  Parallel Lives Revisited

Map 4.2  Spatial dispersion of Mediterranean immigrants in Ghent, 1970. Created with GPS Visualizer ©2003–2016 Adam Schneider on an OpenStreetMap © OpenStreetMap contributors

Many Tunisians went to live in the southern part of the city. This relates to the fact that those who were recruited by UCO Ltd and Fabelta were first housed there. This area, from the southern entertainment district down into the so-called ‘Million’s quarter’, functioned as a residential area and remained so during the period under study.53 In 1970, more than half of the population here were registered as white-collar workers (Vanneste 1981, 166). The expansion of the city in this direction was largely characterized by a suburbanization movement of the well-to-do, turning it into an upper-class corridor into the countryside (Claeys and Van Hove 1987, 40; Vanneste 1981, 168). In the first half of the 1970s, the presence of Tunisian immigrants in this neighbourhood became even more pronounced, as can be seen on the 1975 map. During that same period, the northern edge of the city centre, and especially the Patershol neighbourhood, also became an important area of immigrant concentration. This neighbourhood is part of the medieval nucleus of

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  119

Map 4.3  Spatial dispersion of Mediterranean immigrants in Ghent, 1975. Created with GPS Visualizer ©2003–2016 Adam Schneider on an OpenStreetMap © OpenStreetMap contributors

the city. In the period under study, it was in decay and would remain so until at least the mid-1980s. Its inhabitants belonged to the lower socio-economic strata of the population: the elderly, the unemployed, single parent families, but also students and artists (Oosterlynck and Debruyne 2010, 30–31). As was the case for the centre as a whole, this neighbourhood was heavily affected by the phenomenon of city flight. Between 1970 and 1981, its population dropped by 37 per cent, and by 1984, the number of vacancies stood at 16 per cent (Claeys and Van Hove 1987, 28, 72). By 1975, the neighbourhoods of Brugse Poort, Rabot and SluizekenMuide had become home to the majority of Mediterranean immigrants in the city. Moroccan immigrants were slightly more concentrated in Brugse Poort, Turkish immigrants in Sluizeken-Muide. At the lower-scale level, we can still clearly see some concentrations in specific beluiken. This general trend of moving to the nineteenth-century belt (and to Patershol) also comes to the

120  •  Parallel Lives Revisited

fore when we look at a number of studies dealing specifically with Turkish immigrants’ residential patterning in the city in 1967, 1971 and 1981.54 In 1967, 29 per cent of Turkish workers lived in the neighbourhood SluizekenMuide, and 24 per cent in the city centre (especially in Patershol) (Bernabé 1968). By 1971, 39 per cent of all Turkish citizens lived in Sluizeken-Muide, 15 per cent in the city centre, 15 per cent in Rabot and 12 per cent in Brugse Poort (Schmit 1972). Over the course of these four years, the residential patterning of the Turkish population clearly had become more concentrated in the nineteenth-century belt. During the 1970s, this trend continued. By 1981, an overwhelming majority of Turkish citizens were housed there (Vanneste 1985), now also in the Sint Macharius neighbourhood, a new area of settlement where in 1978 a Turkish mosque – the first one in Belgium to be owned outright by the immigrant community – was founded (De Gendt 2014, 158–59). Like the Patershol neighbourhood, over the course of the period under study, the nineteenth-century belt gradually became deserted by its middle and upper working-class inhabitants. The locals who stayed behind were those who could not afford to move, such as elderly people living on a small pension, unskilled workers, small shopkeepers, etc. (Kesteloot, De Decker and Manço 1997, 76). As the years went by, this Belgian population aged considerably. Its age structure was exactly the opposite to that of the Mediterranean newcomers. In the Sluizeken-Muide neighbourhood, for example, for the year 1975, children made up half of the ‘guest worker’ population but only one sixth of the Belgian population; adults aged between fifteen and sixty-four made up half of the newcomers compared to nearly two thirds of the Belgians; and among immigrants, the number of elderly aged sixty-five and over was negligible, whereas they counted for one fifth of the locals (Debbaut et al. s.d., 18–19).55 Only among the very young did immigrants (or more accurately, people who did not have Belgian nationality, as most of them were probably born in Ghent) make up an important part of the population. Among working-age adults and especially among the elderly, they remained a small minority during the entire period under study.

Immigrant Segregation in the Neighbourhood? In order to understand the social integration of immigrants in the neighbourhoods where they live, it is important to look not only at their patterns of concentration, but also at their level of segregation (here understood as the spatial isolation of one group from others, as in Kesteloot 1987, 228). These two questions are not the same. For example, when 60 per cent of an immigrant population in a city lives in one specific neighbourhood, we can

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  121

say that they are concentrated in this neighbourhood. However, they might only make up 10 per cent of the population of that neighbourhood, so that, at least at the neighbourhood level, we cannot say that they live isolated from other groups. In this final part of the chapter, we will take a look at the degree to which Mediterranean immigrants in Ghent lived in neighbourhoods that were ‘ethnically segregated’, and how this changed over time. Within the Home Let us however begin at the level of the individual house, by looking at the public lodging houses and employer-provided housing where many independent immigrants initially ended up. In theory, the lodging houses of the city were open to anyone who could afford to stay there, without distinction of class, age or ethnicity. The same lodging houses were used by labourers, white-collar workers and students and by immigrants of different nationalities. One lodging house, which functioned throughout the period under study, provided shelter to French, Italian and Portuguese builders in 1966– 1967 (at the height of the building works in the Canal Zone), to Moroccan students in 1971–1972, to a Spanish domestic servant in 1975, and to a Moroccan workman in 1980.56 However, these people only rarely occupied the same lodging houses at the same time.57 Generally, at any one point in time, their inhabitants were a more or less homogenous group.58 The same thing happened in other forms of employer-provided housing. Employers tended to house workers from different countries of origin separately: if not in different buildings, then at least in different flats.59 This practice was also common in other countries.60 Grouping people by country of origin was probably not only linked to their arrival at different moments in time, but also based on a conscious decision of the employers, meant to facilitate their employees living together. At least these people would be able to communicate with each other and understand each other’s habits, which might not be the case if housemates were ‘of mixed origin’. Still, this did not mean that these immigrants could not come into contact with others across ethnic boundaries and more specifically with locals. With the possible exception of those living on the construction sites in the Ghent Canal Zone, employer-housed foreign workers lived in regular, often densely populated neighbourhoods. They had many neighbours nearby with whom they could potentially meet. At Street Level Moving from the level of the house to that of the street, we have seen that concentrations of immigrants of the same nationality in one p ­ articular

122  •  Parallel Lives Revisited

street were common. Sometimes, these led to a situation of quasi-segregation at street level. In 1967, one study found that, over the whole territory of the  city, there were seven streets – all of them beluiken – that harboured more than twenty Turkish labour migrants (Bernabé 1968, 43). The same  year, another study found one beluik in the Brugse Poort inhabited  by fourteen (and, unofficially, even more) Spanish households (Vanoutryve 1968). In 1971, yet another study discovered that in the three neighbourhoods where most Turkish families were concentrated, there was one street with thirteen; one  with fifteen; one with twenty-one; and one with twenty-four Turkish  families (Schmit 1972). By 1979, the highest percentage of Turkish inhabitants in one street was 73 per cent. The whole street had been declared uninhabitable and was eventually demolished in 1983 (Lekens and Lagrou  1996, 44). However, there were still only ten streets in Ghent where more than 50 per cent of the inhabitants was Turkish, even though there were more than 5,000 Turks living in the city by then. Further, these streets were dispersed over the entire northern part of the city (Vanneste 1981, 165, 173). Clearly, these ‘street-level segregations’ were the exception rather than the rule. In addition, they did not exclude immigrants from meeting others, across ethnic boundaries, in the neighbourhoods where they lived. At Neighbourhood Level Indeed, throughout the entire period under study, at the neighbourhood level, residential segregation in the city of Ghent remained very low. By 1977, in the city centre, non-EECers made up 8.7 per cent of the total population. In Brugse Poort, the number of non-EECers was relatively high –1,934 – but their percentage of the quarter’s much larger population was only 5.7 per cent. In Sluizeken-Muide, the neighbourhood with the highest concentration of immigrant households, only 16 per cent of inhabitants were nonEEC citizens.61 Still, the immigrant settlement in these neighbourhoods, and especially in Sluizeken-Muide, provoked strong reactions from the locals. By the mid-1970s, Sluizeken-Muide had become known as a ‘Turkish neighbourhood’, and its Turkish inhabitants were accused of ‘huddling together’ in ‘ghetto-like’ residential patterns. A 1977 city council debate on immigrant housing clearly reflects this idea: When some 70 to 100 years ago in America and Canada many immigrants from different countries arrived … whole neighbourhoods were founded there with Portoricans, whole neighbourhoods with Italians, and last but not least, go and have a look in Isawaka, there the Belgians still live like they lived here,  with their own shops, own bakeries, own community. All sociological

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  123

research, Mister Mayor, Ladies and Gentlemen, has shown that a concentration of ­foreigners in a homogeneous population is not healthy. Also in the United States and Puerto Rico this has been found. … I would like to remind you of what happened in the Middle Ages. When people got fed up with the situation, others were simply massacred. We cannot let things get that far!62

It is true that over the course of the 1970s, more immigrant families came to live in relatively more segregated neighbourhoods. Whereas in 1970 there was only one statistical sector (at the level of the housing block) in the city of Ghent with a foreign population of more than 10 per cent, by 1981, there were many sectors that far exceeded this level, going from 14 to 31 per cent. They were situated in the city’s nineteenth-century belt and in the Patershol quarter. However, in none of these statistical sectors did non-Belgians come to dominate the local population (Claeys and Van Hove 1987, 45). Further, at no point during the period under study were there areas in the city inhabited by a majority of immigrants of one nationality. Throughout the period under study, the ‘Turkish’ Sluizeken-Muide was also populated by Algerian, Moroccan, Tunisian, Italian and Spanish immigrants. Even in the beluiken, ‘guest worker’ families were generally still a minority, occupying only 6 per cent of the inhabited beluik houses in the city centre, 11 per cent in Brugse Poort, 11.5 per cent in Rabot, 11.5 per cent in Sint Macharius and 26 per cent in Sluizeken-Muide – and much less in other neighbourhoods (Balthazar 1978, 24–26). This spatial pattern of concentration without segregation at the neighbourhood level was common to all Belgian cities. It was similar to what was observed in other Western European countries, with the exception of situations of forced segregation through specific social housing projects, employer’s housing or bidonvilles (Kesteloot 1987, 228). Even though in some countries, segregation figures were certainly higher than in others, real ghettoes did not and do not exist (O’Loughlin 1987, 22–23; Sule Özüekren and van Kempen 2003, 163). Still, in public opinion, the ‘ghettoization’ of immigrant populations is and has been a recurrent argument, not in the least in the ‘parallel lives’ debate. Since the early 1970s, several European countries have allowed communal authorities to implement quota systems, with the intention of putting a halt to the further settlement of ‘ethnic minorities’ in so-called ‘concentration neighbourhoods’ in order to ‘enhance their social integration into society’ (Özüekren and Ergoz-Karahan 2010, 363; Bolt, Özüekren and Phillips 2010, 177). In the Netherlands, the city of Rotterdam was the first to install such a system, allowing a maximum of 5 per cent non-Dutch inhabitants per neighbourhood, as a reaction to the riots between locals and immigrants that took place in 1972 (Lucassen and Lucassen 2011, 79). In Germany, the Federal Government

124  •  Parallel Lives Revisited

introduced a ‘foreign residence quota’ in 1975, giving city authorities the right to refuse the further ­settlement of foreigners once a threshold of 12  per  cent had been  passed (Leitner 1987, 80–81). In Belgium, such a quota system was implemented for the first time in the Brussels commune of Schaarbeek in 1981.  These quotas, limiting the right of immigrants to live where they saw fit, were heavily contested everywhere. As they clearly violated the European Convention of Human Rights, they were eventually struck down by higher courts or abandoned by the governments that had installed them. In Ghent, in the framework of discussions about the city’s housing situation in the late 1970s, action committees asked for a ‘dispersion’ of the immigrants and a ‘maximum percentage’ of immigrants per area. This demand was not met by the city authorities.63 It does however indicate that the relations between locals and immigrants – at least at group level – were not all good. But how did this come to be? And how were the relations between immigrants and their Belgian neighbours at the individual level? These and other questions will be answered in the fifth and last chapter of this book.

Conclusion In this chapter, we have seen how independent immigrants who arrived in Ghent on their own, without a pre-existing network, often found their first housing in lodging houses or in private dwellings provided by their employer. Here, they lived together with immigrants coming from the same country and often even the same place of origin. The spatial positioning of this first housing in the city also impacted on their later residential concentration. This is most clearly demonstrated by the Tunisians, the only group studied here who came to be concentrated in the richer residential area south of the Sint-Pieters railway station, where they had been housed by UCO Ltd and Fabelta Ltd upon their arrival in the city. Moving out of this temporary housing, immigrants typically found a place to live in the residual rental market. The quality of this housing was bad, but as it was also mostly cheap, it coincided with their initial goal as target earners: to save as much money as possible in as short a time span as possible. As time went by, this goal changed, especially when the immigrants began to live together with their family. However, even then, they only rarely managed to climb up the housing ladder. Among the many reasons for this, a very important one was the high degree of discrimination on the rental market, both private and social. Access to social housing, as well as buying a house, only became viable options for immigrant families from the second half of the 1980s onwards.

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  125

This does not mean that immigrants were passive pawns when it came to finding a place to live. As was the case for their professional careers, ­immigrant households showcased a great deal of residential mobility, with many short-term moves from one house to another. This mobility was partially forced upon them, partially chosen; it could lead to saving money on rent or to micro-improvements in their living conditions. However, immigrants’ mobility diminished over time, related to the changing composition of the immigrant population, their prolonged stay in the city and the outbreak of the economic crisis. Those who stayed in Ghent for a longer time generally settled down in a more permanent way. The relegation of immigrant households to a specific sector of the housing market, together with a host of other factors, resulted in an ongoing concentration of immigrants in specific streets and neighbourhoods over the course of the period under study. From a very dispersed residential pattern, immigrants in Ghent became more concentrated in the northern part of the city centre and especially in the nineteenth-century belt, an area to the north and east of the centre, consisting of workmen’s houses and abandoned industrial grounds. This was especially the case for Moroccan and Turkish immigrants, who were also the most numerous. Italian, Spanish and Algerian immigrants remained more dispersed, whereas Tunisian immigrants were concentrated in the area to the south of the city centre. However, this growing residential concentration did not result in a segregated urban landscape. Only at the level of the individual house, in lodging houses and other employer-provided housing, did immigrants live separate from locals and from other immigrants of different ethnic origin. During the period under study, there were some streets with high levels of immigrants, but these were the exception rather than the rule. On a wider geographical scale, the degree of segregation remained very low. Only over the course of the 1980s did streets and statistical sectors where over half of the population were non-Belgians become more common. Even then, only around 15 per cent of the immigrant population in Ghent lived in such areas. The large majority of immigrant households continued to live in streets and neighbourhoods where more than half of the inhabitants were Belgian. Also, when we take into account the difference between the age pyramids of the immigrant and Belgian populations in these neighbourhoods, we have to conclude that it was only among the young that immigrants made up an important part of the population. Among adults, they remained a small minority, sharing their neighbourhoods with a majority of Belgians. In theory therefore, immigrants in Ghent, even those living in the most s­egregated neighbourhoods, had lots of opportunities to develop social relations across ethnic boundaries with their neighbours. The fifth and final chapter of this book will discuss how these opportunities played out in practice.

126  •  Parallel Lives Revisited

Notes   1. Interviews with Dahmane, Kamel, Miloud and Willy.   2. Interview with Mustafa (2) and Ismail X., both Turkish immigrants from Emirdağ, who both lived in a room above this café in 1971.   3. That the employer was responsible for the provision of housing for single employees was stipulated in the bilateral agreements between Belgium and the sending countries, i.e. Spain, Greece, Morocco, Turkey, Italy, Tunisia and Algeria. For example, article XIII of the Belgian-Tunisian bilateral agreement states that: ‘The employer commits himself to provide the single employee, who asks for it, for a rent which is regular for the region, decent housing, which minimally corresponds to the demands concerning hygiene that have been determined by Belgian legislation.’ Loi du 13 décembre 1976 portant approbation des accords bilatéraux relatifs à l’emploi en Belgique de travailleurs étrangers, Moniteur Belge, 17 June 1977.   4. Interview with Ömer.   5. Interview with Ali Gh.   6. L’Arbi, a Moroccan immigrant who was recruited by UCO Ltd in 1971, was housed in an apartment in the Stropstraat, near the Uyttendaele plant, where the rooms were neat and clean, modern amenities such as gas and running water were present, each apartment had a big kitchen, etc. Interview with L’Arbi. The apartment buildings where the Algerian interns were housed were located in Kunstlaan 9 and Muinkkaai 20. Data from individual immigrant files.   7. Interview with Amand, head of personnel at Fabelta at that time.   8. Group interview with Ali L., Abdelham, Nahib and Habib, four Tunisian immigrants who had arrived in Ghent in the 1970s.   9. Interviews with Piergiorgio, Enzo and Girolamo. 10. Chotkowski, Vijftien ladders en een dambord, cited in Vogel 2005, 104–10. 11. Interview with Dahmane. 12. Interview with Kamel. 13. Interview with Mustafa (2). 14. Of my sample of six immigrants who lived in the Cataloniëstraat, three were registered there for less than two months, the other three between two and six months; of those living in the Raas van Gaverestraat, eight were registered there for less than six months, four between six months and one year; in the Sint-Bernadettestraat, one was registered there for less than two months and two others between two and six months; in the Coupure, two for less than one month, and two others for just over seven and nine months respectively. As the departure date of immigrants who left their address without warning the city administration was equated to the date when they were struck off the alien or population registers, which could be months later, the data from the individual files most likely overestimates the duration of stay at these immigrants’ first address. 15. Castles and Kosack (1985, 252) quote ‘L’immigration portugaise’ in Hommes et Migrations Etudes 1967: ‘of the Portuguese living in hostels, more than one third had lived in them between one and three years and 17% more than three years’. 16. Interviews with Girolamo and Enzo. 17. Interview with Kamel. 18. Interview with Moh.

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  127

19. This change in behaviour is something that is common to guest workers of all nationalities everywhere in Europe. It has also been attested for example for Portuguese i­mmigrants in France by Callier Boisvert (1987, 66). 20. Between 1960 and 1974, however, 125 beluiken (many of them still inhabited) had been demolished and were as such not included in this study. 21. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds, journal Woonfonds 1980– 1983. 22. Both Charif and Miloud, both Algerian immigrants who came to Ghent in the first half of the 1960s, say that it was easy to find a place to live back then. 23. Interview with Gazi. 24. Interview with Charif. 25. Interview with Kamel and Miloud. 26. Intervention of Mrs Herman-Michielsen ‘about the newspaper report in which the plan of deputy Monsaert arranging for the municipality to become the ‘landlord’ of a great number of guest workers is announced’, minutes of the Ghent city council, 19 December 1977, pp. 2355–56. Other interventions in which this image of Turkish (and other foreign) tenants comes to the fore are, amongst others, the ones of Mr Plasschaert, city councillor and president of the local branch of the socialist textile union, minutes of the Ghent city council, 24 May 1977, pp. 867–69 (‘They only have one endeavour: to live as cheaply as possible, they crawl into everything that’s vacant, with our without permission’) and of Deputy Tackaert-Foket, minutes of the Ghent city council, 16 June 1977, pp. 1398–1401 (‘[it is] probably the poorest of the poor who come here to us, and that they originate from small villages. Many of them probably even were cave-dwellers, so that they brought that way of living here, with all the ensuing problems’ and ‘A Turkish family moves in somewhere, two days later, two families are living in the same house, the demolition begins, the house next door is let out, and so on’). 27. When asked for the reasons why, according to him, the attitude of the locals towards the Turkish immigrants had changed, Adem, a Turkish immigrant from Istanbul, told the interviewer: ‘It was the period when most people from Emirdağ came. They lived outside, they washed outside. Their wool was drying on the guardrails along the canal. That’s how the Belgians’ ideas about them changed.’ Interview with Adem. Indeed, Miloud, an Algerian immigrant from Oran who had arrived in Ghent in 1964, was convinced that certain immigrants could not be trusted to live in European houses: ‘I do not say that it was not justified. Sometimes, there were foreigners who possibly behaved badly. Because they came from the countryside, and they were not used to living in apartments, and they had the habit to live far away from one another. They were used to having space, chickens and things like that. So as to those people, it is possible that they have not always been very delicate towards the owners.’ Interview with Miloud. 28. Interview with Trahib, Habib and Ahmed L. from the club Belgo-Tunisien. 29. Hubeau does not make it entirely clear whether before 1974 there existed a formal discrimination based on nationality in the allocation of social housing; however, he does quote a circular letter from the National Housing Association, dated 25 October 1966, calling for equal treatment of immigrants in the allocation of social housing. 30. City Archive Ghent, series XXXIII Social Affairs and Housing, box number 112 ‘Consultative Commission Migrants’ (reports), letter from the Gentse Maatschappij voor de Huisvesting to the city council, third direction, 13 October 1975; letter from the S.M. Volkshaard to the city council, 21 November 1975; letter of the Goede Werkmanswoning to Madam Foket A., city councillor, 3 October 1975; handwritten note concerning the

128  •  Parallel Lives Revisited

city’s social housing stock and its experiences with immigrant tenants, 15 December 1975. 31. De Witte (1997, 78–79) citing the results of a questionnaire, carried out by the ‘Werkgroep Huisvesting Gent’ in 1980. 32. Yearly report city of Ghent, 1980. The percentage of non-Belgians in the city’s total population had gone down compared to 1975 because of the merging of communes in 1977, whereby the villages around Ghent, that counted a lot less immigrants, were merged with the actual city. 33. City Archive Ghent, series XXXIII Social Affairs and Housing, box number 112 ‘Consultative Commission Migrants’ (reports), Report of the meeting of Thursday 30 November 1972. 34. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, journal Woonfonds vzw, 83/02; in a document entitled ‘consultation meeting 8 February 1979’, Woonfonds Ghent claims that ‘renting houses is very hard because … the housing associations have enormous waiting lists and are almost inaccessible to foreigners’; in the report of the General Assembly of 28 March 1981, we read the following: ‘For the moment, the situation is that most social housing associations house very few migrant families. Several associations are also reluctant to do so. However, they argue their position by saying very few migrants are registered with them’. 35. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, Dossier Huisvesting, 1995, 1.2.2.2 De stadswoningen. 36. In reality, the percentage of owner-occupied houses was probably lower still, as the census did not take into account dwellings that were inhabited by ‘temporary inhabitants’ such as students, who in practice were never owners (Claeys and Van Hove 1987, 62; White 1984, 36). 37. Kesteloot et al. suggested this change of mind as the most important reason for the Turkish immigrants they studied to make the move to home ownership (Kesteloot, De Decker and Manço 1997, 86). 38. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, ‘Dossier huisvesting’, 1995, table 8. 39. On the situation in German cities, see Clark and Drever (2000, 833; White 1984, 136). Clark and Drever refer to studies that contend the opposite; studies based on city-level statistical data however tend to show a higher degree of mobility and seem more reliable than studies based on surveys, interview panels, etc. which show a less mobile picture – as we have found in our own research, high degrees of mobility are often only temporary, and as such less prevalent among more settled immigrants who would probably make up the main target group of such qualitative research. 40. For more information on the samples used, see the quantitative appendix, chapter 4 (1). 41. For this survey, students of the Catholic Education Centre for Social Work in Ghent visited 383 Moroccan, Turkish, Algerian, Tunisian and Yugoslav families in order to examine their housing conditions and experiences. 42. Interview with Fatma K. 43. Interview with Maria del Pilar. 44. Interviews with Yamina. 45. See the quantitative appendix, chapter 4 (2). 46. See the quantitative appendix, chapter 4 (3). 47. See the quantitative appendix, chapter 4 (4). 48. One historical study that pays attention to precisely this issue is the PhD dissertation of Flavia Cumoli, who aptly showed the high degree of intra-local mobility of Italian

Integration Processes of Immigrants in the Local Housing Market and the Neighbourhood  •  129

migrants in both a Belgian mining context and an Italian suburban metropolitan context (Cumoli 2012, 31). 49. For more detailed information on the making of the maps, see the quantitative appendix (5). 50. Musterd and Deurloo used micro-level data for their analysis of the residential patterning of immigrant newcomers in the city of Amsterdam, thereby showing residential patterns to be much more dynamic than had generally been assumed. 51. These reached 22 per cent in parts of Sluizeken-Muide and 16 per cent in Sint Macharius in 1984 (Claeys and Van Hove 1987, 72). 52. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds, document ‘Algemene beschouwing aangaande de huisvestingsproblematiek in Gent’, s.d. 53. In a statistical survey from 1979, the southwest of the city clearly came to the fore as an area of high quality housing (Vanneste 1981, 168–69). 54. As each of these studies used a different kind of dataset to reach its conclusions, they are not directly comparable. Still, their methods and purposes were similar enough to allow for a rough comparison. 55. Similar differences in population structure could be found in the Brugse Poort neighbourhood. A demographical analysis of a part of the neighbourhood from 1986 shows a similar population structure, with 49 per cent of the non-Belgian population aged twenty-one or younger, 48.5 per cent aged between twenty-one and sixty-five, and 2.5 per cent aged over 65, compared to 21, 61 and 18 per cent of the Belgian population respectively. Archive Amsab-ISG, archive Maurice Maréchal, yearly report Neighbourhood Centre Brugse Poort, 1985–1986, pp. 10-11 – I calculated the percentages myself. 56. Information from individual immigrant files and from the population register, Hoogstraat 60, for the census periods 1960–1971 and 1971–1980. 57. With the exception of the builders working in the Canal Zone, who were very likely employed by the same building company and probably had their lodging arranged for them by their employer. 58. In one lodging house in 1963–1964, our sample showed only Algerian immigrants; in three other houses in 1966, only Turkish lodgers were living. Data from the individual immigrant files and from Bernabé (1968, 43). These lodging houses were also mentioned in the interviews with Kamel, an Algerian immigrant who arrived in Ghent in 1963; with Cemil, a Turkish immigrant (from Posof ) who arrived in Ghent in 1967; and with Mustafa (2) and Ramazan, Turkish immigrants from Emirdağ who arrived in Ghent in 1966. 59. In 1963, the Algerian recruits of the Union Cotonnière were all housed together. A group of Moroccans who were recruited in 1964 were all housed together in a dormitory at one of the UCO factories, and no immigrants of other nationalities were lodged there. Turkish newcomer Ömer, who came to Ghent in 1964, was housed together with four other Turkish newcomers, all of whom were working for the Union Cotonnière. Ali Gh., recruited by the same company in 1966, was housed together with the other recruits, who were all Tunisian; so was Abdelham, who arrived with a contingent of Tunisian workers in 1970. Interviews with Kamel, M’hamed, Ömer, Ali Gh. and Abdelham. Data from the individual immigrant files. 60. It has been attested for example for the housing of ‘guest workers’ by employers in the Netherlands (Vogel 2005, 92). 61. Archive Amsab-ISG, archive Maurice Maréchal, archive Woonfonds vzw, Dokumentatiemap Woonfonds Initiatieven, 1978.

130  •  Parallel Lives Revisited

62. City Archive Gent, Gemeenteblad Gent 1977 (minutes of the city council), assembly of 24 May 1977. 63. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds, Woonfonds Initiatieven 1978, pp. 4, 7.

Chapter 5

Immigrants’ Social Relations with Neighbours 

In Ghent, immigrants’ residential patterning provided them with many opportunities to develop neighbourly relations across and, as time went by ever more so, within ethnic boundaries. In this chapter, we will focus on the extent to which these opportunities materialized in social reality. As with our analysis of workplace-based relations, we will make use of the framework constructed by urban sociologist Talja Blokland, differentiating between interdependencies, transactional relations, attachments and bonds (see introduction). As most of the content of this chapter draws from the life stories of immigrants, neighbourhood relations are described from their point of view, rather than that of their local counterparts. Also, the category ‘neighbours’ reflects the ways in which immigrants themselves understood this concept. It does not refer to a spatially bound group of people, but rather to those that were defined as such by each individual interviewee. Finally, this chapter pays more attention to the experiences of women than to those of men. During the period under study, immigrant women made more use of the neighbourhood than men and therefore developed more and different kinds of neighbourly relations. While women would more commonly stay at home for longer periods of their lives and thus spend more time in the neighbourhood, there seem to be other factors involved as well, as those women who had long-term, continuous careers would also stress the importance of their neighbourhood relations in interviews. This was not the case for the men, whose narratives were more centred on their work. Other researchers in

132  •  Parallel Lives Revisited

other contexts, such as Byron and Condon looking at Caribbean migrants in Britain and France (Byron and Condon 2004), and Herbert looking at South Asians in Leicester (Herbert 2008), have noted the same gender divide in the narratives of their respondents.

Immigrant Newcomers in Temporary Housing Within the Home When they first arrived in Ghent, independent immigrants generally went to live somewhere temporarily. They often ended up in lodging houses or housing provided by their employer. The relations these newcomers developed there often remained important throughout their lives in Ghent. Therefore, we need to include them in our analysis. The tenants in a lodging house, café or inn mostly lived together with other immigrants, most of whom came from the same country or even place of origin. They also came into contact with the owners of those places. In particular during the first years of Mediterranean migration to Ghent, these were locals – often among the very first locals immigrant workers came into contact with. In the immigrants’ stories, these people generally come across as friendly and willing to help with administrative problems, the search for a job, etc.1 Only one interviewee, the Algerian Kamel, mentioned the concierge of a lodging house expressing racist feelings towards his foreign lodgers. He connected this attitude to the fact that the man was an old colon, who had only recently returned from the former Belgian colony of Congo.2 Similar situations have been documented in France, where the relations between immigrant workers and lodging house owners at that time were often thwarted because the former were pieds-noirs who harboured racist ideas about immigrants in general and Algerian immigrants in specific (Castles and Kosack 1985, 262). In lodging houses, people generally lived in close proximity to one another. This nurtured the development of close social contacts. Studying hostels in early postwar Melbourne, O’Hanlon (2005) found that for many newcomers, living in a hostel or guest house ‘gave them instant access to a circle of friends and acquaintances they might not have met had they been required to find accommodation on their own’ (O’Hanlon 2005, 88.11). The same observations can be made for immigrants that were housed together by their employer in other types of accommodation. Even if they did not live together for long, fellow lodgers could still form lasting ties, which were maintained elsewhere after they had moved out, for example in cafés and other meeting places. For instance, Kamel moved out of the lodging house where he had been housed by UCO Ltd after only two or three weeks. However, he did

Immigrants’ Social Relations with Neighbours  •  133

remain in contact with the other Algerians he had met there, having become their ‘letter-writer’.3 These housemates not only lived together, but often also worked together (or at least did the same kind of job, for the same employer), and went out together in their free time. This nurtured the creation of multi-stranded relations between them and thus – in Blokland’s terms – the development of attachments and even bonds.4 The connections they made could last for life. Even Nahib, a Tunisian immigrant who only lived in the flat provided by his employer Fabelta for three or four months in 1974, remains in contact with his former housemates today.5 In the Neighbourhood Apart from these in-house relations, immigrant newcomers also came into contact with the neighbourhood at large. Such contacts were not necessarily initiated by the immigrants themselves. Those that arrived in Ghent in the early and mid-1960s were often received with great interest by their Belgian neighbours, as they were (among) the first ‘foreigners’ these locals had ever met. Adem, a Turkish immigrant from Istanbul who arrived in Ghent in 1965, remembers this very well: ‘When we just arrived, we were a kind of curiosity. Every morning, we had something standing at our door. We received plates, cups, etc. The people thought: “They have migrated so they probably don’t have anything”. The people were very friendly. The neighbours helped.’6 Indeed, some of the locals took it upon themselves to offer material help to these newcomers, bringing them food, furniture, household effects, etc. This also happened to Ömer, a Turkish immigrant from Istanbul who arrived in Ghent in 1964: ‘We received help from the neighbours. Some brought bread, some brought things, others a pot of food. Those were locals, because apart from ourselves, there were no Turks in the neighbourhood. They knew nothing about Turks. Some of us dressed very poorly and they pitied us – “poor foreigners” … They helped us a lot.’7 Apart from arousing curiosity, immigrant newcomers clearly inspired some of the locals to acts of charity. These were carried out in a paternalistic way, aimed at helping those ‘poor foreigners’, as Ömer phrased it. Sometimes, such charity did not even involve any human contact, as locals simply left things on their doorstep, as in the story of Adem. Indeed, this kind of charity did not arise from a wish to get to know the newcomers personally, but rather from a more general feeling of righteousness or a will to ‘do good’. As such, it did not necessarily give rise to reciprocal relations between newcomers and locals.8 If such relations did develop further, they were often unequal, with newcomers being treated as inferior to those who helped them. They only rarely surpassed the stage of transactional relations. Still, most immigrants

134  •  Parallel Lives Revisited

seem to have appreciated this help, especially when they had just arrived. This was the case for Faruk, a Turkish immigrant from Emirdağ who arrived in Ghent in 1965: The people were very friendly. They really helped us a lot. We rented a house somewhere and the neighbours came to ask us what we ate and drank. They went with us to all the shops because we didn’t know where to buy things. They showed us everything, they discussed with the house-owners, they really helped us a lot. They also made food for us. That was really great.9

Besides their direct neighbours, local publicans and shopkeepers would also offer immigrant newcomers a helping hand. These small entrepreneurs were among the first people in the neighbourhood whom immigrants established social relations with. They regularly visited their cafés and shops, especially during the early years of Mediterranean migration to Ghent, when there were hardly any ethnic entrepreneurs. These initially interdependent entrepreneur-client relationships could develop into more meaningful, transactional relations, as was the case for Enzo and his friends and their local newsagent’s: We had a small shop there, down the steps, which sold cigarettes and so on. There was a lady there and we called her ‘la bionda esplosiva’ [the explosive blonde]. To us, Italians, she was very friendly! (laughs) And we used to say ‘let’s go to la bionda esplosiva’. And with us, she laughed, she joked, and we went to buy our cigarettes there.10

The initial contact of immigrant newcomers with people in the neighbourhood could also be directly mediated by their employers. For Ömer and his companions, for example, the Union Cotonnière had arranged a place where they could relax after work and come into contact with neighbours: the local club house for the elderly.11 However, this kind of arrangement seems to have been rare and limited to the pioneers. Unlike in the mining regions, employers in Ghent did not get involved too much with how their employees spent their spare time.12 One group of immigrants for whom it was more difficult to construct meaningful social relations in the neighbourhood was those working as domestic servants. As the houses where they worked and lived were invariably situated in the more expensive neighbourhoods, there was an enormous class difference between them and their neighbours. This made the development of ordinary neighbourly relations unlikely. Domestic servants sometimes got to know their (or rather their employers’) neighbours. However, these relationships more resembled those between employers and employees – at least,

Immigrants’ Social Relations with Neighbours  •  135

this is the idea we get from whatever little information the sources convey on this subject. The following anecdote (set during the holidays at the seaside rather than in Ghent) recounted by José Luis, who worked as a domestic with his wife Jesusa during the first years of his stay in Ghent, is telling: ‘In the summer, we went to the sea, to Le Coq sur Mer, with the family. There, we lived in a villa, just next to the sea, with a large garden. We met many important people there: Ministers, a General, etc. Mr General even explained to me how I had to look after the plants in the garden.’13 The fact that he refers to this ‘neighbour’ as ‘Mr General’ clearly indicates a rather hierarchical relationship between the two men, not one of ordinary neighbourliness.

The Arrival of the Families: More Curiosity and Charity First presented by the sociologists of the Chicago school, the ‘succession model’ shows how the first individuals of an immigrant group to establish themselves in a specific neighbourhood generally only arouse the curiosity of the locals, who see them as a new, exotic element in the neighbourhood. Due to the small numbers of immigrants, locals do not feel threatened by their presence. The model also shows how the interest of locals in their immigrant neighbours decreases and their attitude becomes more hostile as more immigrants come to live in ‘their’ neighbourhood (Bovenkerk et al. 1985, 23). For Leicester, Joanna Herbert registered a radical shift in white attitudes towards South Asian immigrants over the course of the 1970s, from a general curiosity about their ‘exotic’ neighbours and a keen interest in establishing contact with them, to an increasing hostility and an upsurge in racial harassment and even violence (Herbert 2008: 112–17). Something similar seems to have happened in Ghent, although at a slightly different pace, with the change taking place over the course of the late 1970s to early 1980s, and involving very few violent incidents. With the arrival of the first immigrant families, however, the interest of charitable locals (mostly women) in their foreign neighbours, was swiftly revived. In many cases, Belgian neighbours were closely involved in the preparation of their immigrant neighbours’ family reunification. Those immigrants who only had a small ethnic network in Ghent in particular asked for their assistance. Several interviewees recall their Belgian neighbours insisting on helping. When Moh, a Moroccan immigrant who arrived in Ghent in 1966, told his neighbours that his family was coming to Belgium, they asked that he let them know when they arrived, so they could drive him to the airport to pick them up – which they did.14 The story of Zohra, a Moroccan immigrant who came to join her father in Ghent in 1971, together with her mother and brother, also testifies to this help. In

136  •  Parallel Lives Revisited

this case, the contact between her father and his neighbours seems to have been even more intense than that reported by Moh: My father had prepared for our coming. And how did he do that? Well, in the Brugse Poort neighbourhood, he lived in the Abrikoosstraat. … In the Eksterstraat [Klapeksterstraat, around the corner of the Abrikoosstraat] lived the Haemelinck family, Philemon, Marie-Jeanne. … And they knew my father. That was a family he got to know. And Maria … she had a greengrocer’s opposite the apartment building where my father lived. … The men – there were other Moroccan men living there too – and my father regularly came into her shop, which she opened up after her divorce. The friendship with my father was so close that every time, he told her about us and how it was, and how hard it was [that they were left behind in Morocco]. And then, because of her and of the Haemelinck family [saying] ‘You have to bring those children’ – as he always talked about us, they became curious of course. So, our coming had actually been arranged.15

Once the immigrant family had arrived, and especially during the initial period after their arrival, immigrants continued to receive charity from Belgian neighbours, consisting of food, second hand clothing, old furniture, etc. Yamina, a Moroccan immigrant who had come to join her husband in the late 1960s together with her five children, still laughs when she ­remembers

Figure 5.1  Zohra and her ‘adoptive’ grandmother Maria, early 1970s. Amsab-Institute of Social History, collection Boucharafat

Immigrants’ Social Relations with Neighbours  •  137

how her neighbours used to bring them chocolate and candy.16 Van den Berg-Eldering (1978) described this kind of relations as highly unequal, and the motivations of ‘natives’ in entering into them as driven by feelings of superiority and often a firm intention of assimilating the immigrant family into ‘Dutch culture’. She found that such relations often ended in conflict, when the Dutch neighbours became disillusioned, seeing that the immigrant family was not going to abandon its own cultural habits, or when the immigrant family got fed up with the unrequested interference of the Dutch neighbours in their private sphere (van den Berg-Eldering 1978, 190 ff). In our interviews, we did not find any examples of such relationships turning into conflicts. That does not mean however it did not happen. The interviewees might simply have been disinclined to recollect such conflicts. From some of the interviews, it becomes clear that the friendliness of Belgian neighbours somehow depended on immigrants’ willingness to accept their paternalism. As immigrants contradicted locals by voicing different beliefs and habits, the reactions of the latter could become negative. Zineb, a Moroccan woman who came to Ghent to join her husband in 1971, also remembers a negative side to her neighbour relations, despite the assumptions of the interviewer to the contrary: Zineb: We had our neighbours, westerners, and we visited them … and they came to see us. Interviewer: And how were they, compared to people today, because today, people are no longer …? Zineb: They were difficult, because no matter what you told them, they said no. … When we asked questions about giving birth, and we said that it is God who gives you your child, they said no, it is the man who makes sure we give birth to our children. Interviewer: You mean according to the religion? Zineb: Yes, according to the religion. Interviewer: But you got along? There was no unkindness? Zineb: No, they were friendly. … When they saw us with our djellabas, they had problems with that. They asked ‘What on earth is this?’17

Mixed Belgian-immigrant families do not seem to have experienced this kind of interference and caritas, even though they sometimes could have used some (material) help as well. This reinforces the idea that the assistance of Belgian neighbours was inspired by their curiosity with regard to the  exotic and by feelings of superiority towards those ‘poor foreigners’, rather than by a sense of good neighbourliness. Immigrants who were married to Belgian spouses frequently had the impression that their Belgian neighbours saw them as intruders, rather than as a curiosity.18 Still, the arrival of the immigrant family, and especially the presence of children, triggered

138  •  Parallel Lives Revisited

locals to become (more) involved with their foreign neighbours, whether in a positive or a negative way. However, some immigrants, and especially those women who had no previous experience living among people with different cultural habits, tried to avoid all contact with their local neighbours, out of fear or shame. This was the case for Halice, a Turkish immigrant from a village near Emirdağ who came to Ghent to join her husband in 1968: ‘Actually I didn’t want to come here. I told myself “What will I do there with those strangers, with those heathens?” Frankly, I was scared. At home, we had a small window and from there, I watched the heathens with fear. I wondered what they were like.’19 Hassenia, a Moroccan woman who came to Ghent to join her husband in 1968, initially also attempted to avoid her Belgian neighbours. However, as she and her family lived in a beluik with shared facilities, that was not easy: When he [her husband] brought us here, we lived in a beluik, where the toilets were outside. We were ashamed to show ourselves to the Flemish people  when we went outside. We were ignorant. … I was afraid, when I saw  the Flemish, I was afraid. I went outside and then immediately went back  inside (laughs). … We were ignorant. … on Mondays, we went to take  the garbage out, … and  when they came to collect the garbage, I went outside and took my bucket. The Flemish man [her neighbour] winked at me, they kept on winking at me and I asked myself, why the winking? … I left the bucket  where it was and  ran off. I went inside and closed the door (laughs). I went inside, closed the door and when el Haj [referring to her husband] came home, I told him what the Flemish man had done. He assured me that that was their way of greeting. Nothing to worry about. (laughs) I was scared then, I told you, no? … It felt like monsters. (laughs) Like monsters.20

After a while, when they had become accustomed to their neighbours and to life in a European city, such extreme feelings of insecurity and fear disappeared. However, during the first years after their arrival, there were still serious obstacles to overcome before immigrants could develop meaningful relationships with their neighbours. To begin with, as their stay in Ghent was meant to be temporary, they did not have much incentive to get to know them. Second, they tended to move frequently within the city, which made it hard to develop long-term neighbourly relations. However, as temporality turned into a more settled state, many immigrant families did enter into meaningful social relations with their neighbours, both within and across ethnic boundaries. In the remainder of this chapter, we will look at the ways in which these neighbourhood-based relations developed from interdependencies and transactional relations into attachments and ­sometimes bonds.

Immigrants’ Social Relations with Neighbours  •  139

Neighbour Relations Within Ethnic Boundaries During the initial stages of settlement, immigrant families are generally thought to ‘rely on kin and co-ethnic friends and acquaintances in the local residential area for social interaction’ (Ray and Preston 2009, 221). However, whereas men were generally not immediately interested in the exact spatial location of these social contacts, this was different for women. In her study of Moroccan immigrants in the Netherlands in the 1970s, van den BergEldering (1978) found that especially for those immigrant women who were housebound and were not allowed by their husbands to move freely around the city (mostly Muslim women from rural backgrounds), it was important to get to know other women who were physically nearby. This way, they could socialize with them on their own initiative. Because of cultural and linguistic barriers, the women that ‘qualified’ for their neighbourhood networks were initially only kin or old-world relations – people they had known before migrating. In the absence thereof, and if their husband permitted it, women from the same place or region of origin, speaking the same language and adhering to the same religious and cultural prescriptions, could also become part of these networks. At least during the first years after their migration, the rural Riffian women in van den Berg-Eldering’s sample never worked outdoors and had their daily contacts limited to family members. If there were no family members or other ‘suitable’ contacts living nearby, these women lived very isolated lives (van den Berg-Eldering 1978, 164–67). A similar if not identical picture can be seen in the interviews with Riffian women who migrated to Ghent over the course of the 1960s and 1970s.21 Moumna, a Berber woman from Nador who arrived in Ghent in 1967 in order to join her husband, only had social contact with her sister-in-law, who lived in the same street, and with her Berber neighbour. She did not know whether there were other Berber women in the city: In those days, there were not that many Moroccans. I did know my sisterin-law who now lives outside Ghent. There was also my neighbour at the Dampoort [railway station], where we used to live. There were three of us, as far as I know. In Brussels I didn’t know anybody, but they said there were many Berbers living there. So as far as I know, there were three of us. We even lived in the same street.22

Zineb, a Moroccan woman from Beni Sidel (also in the Rif ) who came to join her husband in Ghent in 1971, spent the first period of her stay in Ghent alone at home. She was not allowed to go out to work and had no family members living nearby. Later, when she got to know the other

140  •  Parallel Lives Revisited

Moroccan women in the neighbourhood, she began to construct new social relations.23 Getting to know these other women, who were not family members or with whom there were no old-world ties, was no sinecure for women who were housebound. Sometimes, they were the wives of friends or colleagues of their husbands, who were only introduced to them when the latter invited them over to their house or when they were invited by the former. Other times, the women spotted other women when they passed their house. This is how Hadda, also a Berber woman from the Rif, got to know Hassenia, who was Arabic-speaking: One day, your mother [Hadda is speaking to Hassenia’s daughter] came by and I watched her go by, and I thought to myself, ‘She is Moroccan, not Turkish!’ You were still little … Sometime later I saw Mimouna, whom I didn’t recognize [as being Moroccan] because she was dressed completely differently, like a Turkish woman. … I was so bored because I was all alone here, until I met your mother and Mimouna and Loubna, then I said ‘thank you God!’ I was bored in the beginning, and then I told them ‘I came here from Morocco, and God brought me to this place all alone, I don’t have any brothers or sisters here, I have no one [family]’, and still today I have no one here.

The other women that came to belong to Hadda’s neighbourhood network, Mimouna and Loubna, were introduced to her in a different way. Hadda recognized Hassenia as Moroccan – and thus as a possible c­ ontact – by her clothing. Mimouna on the other hand was dressed ‘like a Turkish woman’ and so Hadda initially did not think it fitting to contact her. Apparently, the boundaries within which Hadda felt she could look for friends were not limited to kin and village of origin, as has been reported for other Berber women in Europe at the time (van den Berg-Eldering 1978; Cammaert 1985), but to the wider category of ‘Moroccans’. Hassenia had married a Berber man and spoke a little Berber, and she taught Hadda some Arabic, so they could communicate with each other. Mimouna, whom Hadda had not recognized as Moroccan, was Berber, like Hadda. When the three of them met, they spoke mostly Berber to each other. Hadda also became close to Mohamed, Hassenia’s husband, who spoke Berber to her. In this case, it was language that brought them together. Hassenia and Hadda regularly met at each other’s house, went to childcare consultations together, cooked for each other, etc. If they had the opportunity to slaughter a sheep according to the Muslim ritual, their families put their money together to buy the animal and divided the meat amongst each other. They had a strong bond that still lasts today.24 In her study of working-class Portuguese families in a French provincial town, Colette Callier-Boisvert found that these families had a good knowledge of who the other Portuguese in the neighbourhood were, even if they

Immigrants’ Social Relations with Neighbours  •  141

were not regularly in contact with them. They met each other through chance encounters in the street or in the marketplace (Callier Boisvert 1987, 70). A similar observation can be made for Ghent, where immigrant women did not necessarily become close to all co-ethnic women in their ­immediate vicinity, but generally did know where these women lived and who they were. Mimount, a Berber woman from Tafersit, shared this knowledge with her interviewer, who was looking for other Berber women to talk to: There are some [women] here, but I don’t know their house numbers. I do know where they live, but I don’t remember the house numbers. But on the other side of the street there are only people from Ait Waryagher and Ait Temsamane [Riffian villages other than her own], near the mosque. Here there is no one, no foreigners, but a bit further down, there are a number of Ikerhiyen.25

To a certain extent, this preference for co-ethnics in the development of neighbourhood-based relations had a self-fulfilling aspect. Immigrants not only made contacts with co-ethnics in the neighbourhoods where they came to live; they also purposefully went to live in those neighbourhoods where they had close co-ethnic contacts. Donna Gabaccia, in her study of Sicilian immigrants in New York’s East Side in the beginning of the t­ wentieth century, noticed how these immigrants ‘responded to high ­mobility by attempts  – often successful – to make their kin into their real neighbours’ (Gabaccia 1984, 104). By moving house, immigrants had a say in whom their neighbours were. To a certain extent, they ‘constructed’ their n ­ eighbourhoods themselves. As such, neighbours were often also kin, old- or new-world ­contacts by choice rather than by chance.

Neighbour Relations Across Ethnic Boundaries Things were different when it came to the relations immigrant families maintained with their Belgian neighbours. Initially, as we have seen, the power balance in these relationships was highly unequal, as they were characterized by charity, with immigrant neighbours at the receiving and Belgian neighbours at the giving end. However, over time, this imbalance seems to have sorted itself out, and neighbourly relations between locals and newcomers became increasingly egalitarian and reciprocal, as has been shown by Callier-Boisvert in her study of Portuguese immigrants in France (Callier Boisvert 1987, 71). First however, linguistic and cultural hurdles had to be overcome. Depending on how big the gap was, even the most basic attempts at neighbourly interaction across ethnic boundaries could sometimes fail. This becomes clear from

142  •  Parallel Lives Revisited

Hassenia’s description of her walks through the neighbourhood with her Moroccan friends: After a while, we got used to it, bit by bit, and we began to leave the house. There were three Moroccan women, only three. Only Hadda, Mimouna and I. We began to go outside, we took our children to school and picked them up. We had become used to it. We heard the people say ‘Hi’ and we wondered what this ‘hi’ meant. (laughs) We thought it was funny. We heard nothing else but ‘hi’. Funny, ‘What is this word that people keep saying to us?’26

From ‘Good Neighbourly Relations’ to More Profound Relationships In many interviews, especially those that remain at a superficial level, immigrants call the relations they constructed with their Belgian neighbours ‘good neighbourly relations’. However, when we take a closer look at what these relations entailed, they come across as what Blokland calls ‘interdependencies’. Rather than actual interchanges, most neighbourhood-based relations across ethnic boundaries seem to have been highly impersonal, not going beyond mutual greetings and superficial politeness. When asked about his contact with ‘the other foreigners and the people in the neighbourhood’, Elomar, a Moroccan immigrant who came to Ghent in 1970, answered that it was ‘good’ but that he did not remember much about it, and that ‘the people in the neighbourhood were very friendly’.27 To the same question, Abdelkader, another Moroccan immigrant who also arrived in Ghent that same year and later married a Belgian woman, answered that with the ‘other foreigners’, ‘the contact was good. We had good relations and chatted a bit’ and with the Belgians, he had ‘always had good relations, no problems’.28 ‘Good neighbours’ indeed were often seen as neighbours with whom there were no problems, especially by immigrant men, who spent relatively little time in the neighbourhood. Ahmed M., a Moroccan immigrant from Nador who arrived in Ghent in 1970, voluntarily offered this information at the end of his interview: In Arabic we say: ‘Who doesn’t bother anyone will also not be bothered’. Allah gave me a good place here in Belgium. I have a good life here in Belgium. I never had a bad neighbour. I always had good neighbours. … I never had a problem with a neighbour. No neighbour ever complained that we made too much noise. I never had any problems. Not with the authorities, not with the police. Nobody ever came knocking on my door to ask why we were being annoying. I never did anything.29

However, not all neighbour relations across ethnic boundaries remained stuck at the stage of interdependency. When neighbours began

Immigrants’ Social Relations with Neighbours  •  143

to exchange small services benefitting both parties, neighbour relations became what Blokland calls ‘transactions’. These could gradually change into ­‘attachments’, meaning that neighbours made a commitment to good neighbourliness, if not necessarily to each other as individuals. These are the kind of neighbourly relations Maria de la Luz and Francisco, a Spanish couple that moved into a neighbourhood with mostly Belgian inhabitants, remember: Well, they were very friendly, very lovable, yes … We knew everyone here, we said ‘good morning’, we still knew how to chat … There were never any problems. He [her husband] speaks Dutch with them, I speak French with them, we never have problems. We have always gotten along well with everybody. If somebody needed help, we helped each other, yes, we really were a community, yes.30

Barriers to Neighbourly Relations Maria points to her and her husband’s language skills as a sort of explanation as to why there were never any problems in the neighbourhood. Other immigrants also stressed the possibility of communicating with each other as an important factor. Not being able to understand each other was a serious barrier to the further development of neighbourly relations between immigrants and locals.31 Ahmed M., who did not speak much Dutch, found that this limited the kind of things he could talk about with his neighbours: ‘Ah, with the neighbours I really don’t have any trouble, with talking. I talk to them. But as I said, the language of the street … that’s what I know. If a Flemish person asks me ‘Where are you going? Where have you been? What is your name?’, ok. But when it becomes more difficult …’.32 Immigrants who did not speak Dutch or French, mainly Turkish immigrants and Maghrebi women with a rural background, had problems making  themselves understood and understanding what their neighbours wanted to say to them – as became clear also from the story of Hassenia, Mimouna and Hadda (see above). These problems however were mitigated when their children began to go to school and started to pick up the language. From then on, these children served as interpreters,33 which enhanced the opportunities for more profound social relations with Belgian neighbours. The development of strong social relations across ethnic boundaries could be inhibited not only by language but also by other cultural differences. These could be related, for example, to religious prescriptions such as the prohibition of eating pork or drinking alcohol.34 Mehmet K., a Turkish teacher who spoke fluent Dutch, felt that the cultural differences between

144  •  Parallel Lives Revisited

him and his Belgian neighbours imposed a certain limit as to how close they could get: ‘My neighbours ask me “What will you drink?” I drink water. After a while, they get a little bit drunk and pleasant, but I abstain. When we are making jokes, they don’t understand mine and vice versa. Our jokes differ from each other. They eat food that I cannot eat. I always feel a barrier.’35 These barriers did not stand in the way of ‘good neighbourliness’. They just impeded the development of what Blokland calls ‘bonds’, that is longlasting personal relations between specific individuals in which emotions play an important role. However, the difficulties many immigrants experienced in developing real bonds with their Belgian neighbours were not specific to neighbourly relations across ethnic boundaries. Also in general, the development of real bonds between neighbours had become much less frequent in postwar urban Europe than it had been in an earlier era, when neighbourly relations and friendship relations more frequently overlapped (Blokland 2003b, 14–15). The Development of Attachments and Bonds Further, it is not the case that more profound relationships between immigrants and their Belgian neighbours did not occur. When they did, such relationships often developed in the context of the old workmen’s neighbourhoods, especially in the beluiken, where facilities such as running water and toilets were shared, and neighbours had to work together to ensure the viability of their living environment. Because of the close physical proximity between neighbours, this kind of neighbourhood promoted the development of neighbourhood relations that went beyond ‘interdependencies’. This was the case, for example, for Mohamed E., who lived in a beluik in the Brugse Poort neighbourhood. The relationship he constructed with the old ladies in the beluik clearly resembles a real attachment: Next door, there lived an old lady, an old lady who really liked me, she was a very friendly old lady. She did everything for me, really, she was like my mother. I respected her a lot. There were two really great ladies. When they saw me, they called out ‘hey, Mohamed!’. They were very friendly people, like I said. I have not forgotten about them, really (claps his hands with joy)! They were very friendly ladies.36

When Mohamed moved to a street with small workmen’s houses in the Rabot neighbourhood, he once again developed close connections to his neighbours. He even provided some financial help to one of them, an old

Immigrants’ Social Relations with Neighbours  •  145

Figure 5.2  Living very close together in the beluiken, 1970s. Amsab-Institute of Social History, collection Haenebalcke

man who lived on a small pension. In his interview, he recalls how his ­neighbours liked him very much, and were really friendly: There was a man here, and his wife, and he always said to me: ‘I would really like it, Mohamed, if you would live in this house. I wouldn’t like anybody else to come and live in this house. And I told him ‘If God wishes it’. And I always loved him. He was somebody who had worked a lot in textiles as well, someone who had worked a lot. Unfortunately, he had a wife who was always ill. She had an illness. When my first child was born, she said, ‘Please, Mohamed … I would like to buy a present for you, for your daughter.’ My first daughter. And I said ‘No, no!’. Because she couldn’t walk, you know! Really! And she said: ‘Bring the car, I will go with you’. So I went with her, she went into the shop, she said ‘Stay here’. And very slowly [Mohamed shows how the woman walked, bent over and very slowly], she went to buy a lot of clothes for my daughter, for my first daughter.37

These kinds of meaningful relationships between neighbours across ethnic boundaries do not figure widely in our imagery of the 1960s and 1970s. Today, we find it hard to envisage that the immigrants of the first generation, who come across as rather closed and withdrawn in their own ‘parallel societies’, could have harboured real friendships with locals despite the considerable cultural and linguistic hurdles they had to overcome. In reality, however, such friendships were not that rare. At least in Ghent, where the

146  •  Parallel Lives Revisited

opportunities for cross-ethnic contacts in the neighbourhoods were good, many immigrant families – and especially women – created bonds with their Belgian neighbours. Interestingly enough, when immigrant and Belgian neighbours developed actual friendships, language does not seem to have been that important. For immigrants, speaking the local language seems to have been a consequence of (rather than a condition for) these friendships. In their study of interethnic contact in six European cities today, Pratsinakis et al. also noted that, for the development of intimate contacts at the neighbourhood level, ‘linguistic skills do not constitute a significant barrier … even if they do play a role in interethnicties overall’ (Pratsinakis et al. 2017, 112). Often, these relations started off because one or both parties were in need of company. In particular, elderly people and women who did not go out to work had a greater need for neighbourly contacts than men and women who were professionally active. This was the case for Khadija, a Moroccan immigrant from Casablanca who came to Ghent to join her husband in 1977, and her neighbour, an elderly Belgian woman who did not have any children. The elderly lady dropped by Khadija’s place every morning and evening for a cup of coffee.38 Hadda, who, when asked about her relationship with her Belgian neighbours, said that she had no contact with any of them, actually did have a relatively close attachment to an elderly lady in the neighbourhood, Julia. Both women spent most of their time at home and benefited from each other’s company, although apparently, Julia’s attachment to Hadda was greater than Hadda’s to Julia – something Hadda only realized when Julia was leaving the neighbourhood: Well, I lived my days with my children, with the television. I learned French from an old lady then … Julia, she has died now. She dropped by and we went out together. She called me to her house when she had problems with something, for example when the TV didn’t work anymore or the cooker didn’t work, I helped her with everything. She was alone and she came to see me every day, she came when I was cooking and sniffed and said ‘It smells nice here!’ and I said ‘come in’ and I opened the door – with us, they say that when a stranger from another faith knocks at your door and you give them something to eat, that is considered a good deed – so she came in and I served her some food and she started eating it, she wasn’t afraid of burning her mouth, ‘smack, smack’, she had no problems with eating the hot food. I said ‘Julia, Julia, watch out, you will end up burning your mouth!’ … So one day I got a note, and she said she was going to the hospital to be treated and afterwards to a care home and she said ‘Everything that there is in this room, you can take it, you are like a daughter to me’. I said ‘Julia, no, I can’t do that’ and she said ‘Yes, it’s all for you’. So I didn’t go and see her, she asked me to come and I said no, you’re crazy!39

Immigrants’ Social Relations with Neighbours  •  147

An example of such a relationship that developed into a real bond, is the story of Hassenia and Liliane. As we have seen above, Hassenia initially only looked for contact with other Moroccan women and had a superficial relationship with her Belgian neighbours. This changed however when she met Liliane to whom, as time went by, she became very close: Hassenia: She was like a sister to me, me and her, we were like sisters. … And that’s it, that’s the way it was, she came to see me, I went to see her, we visited each other. Interviewer: And you trusted her? And so did El Haj [referring to Hassenia’s husband]? Hassenia: She liked him, she liked El Haj. She liked his whole family, she liked my whole family. And she knew everyone personally, the ones in Holland, the ones in France, she knew everyone personally! She was very happy when they came to visit, she came to say hello. … And her daughter was brought up together with my girls. Interviewer: What was her daughter’s name? Hassenia: Tania. … Liliane came round a lot. … She always came, this was  her place. She drank coffee with me. In the evening, she was with me. When she became ill, she came to me, she called me Hassenia, my real name is Hassenia, she said, ‘Hassenia, I’m ill. … I’m not going to live, I’m very ill’. … She talked to me, but I couldn’t say a word, I let her tell me everything. Interviewer: How did you understand her? Hassenia: I understood her, in the meanwhile I had learned it [the language]  … We understood each other. Before we didn’t, when we had just arrived here. … She asked me to look after her daughter … I promised that I would look after Tania as I would after my own daughters. She grew up here, and I looked after her; was she to have a child, I would look after it as well.

Hassenia had only recently met Liliane when she came to help with an emergency situation, involving one of Hassenia’s children: One of my children burnt himself once, when we were still living in the beluik, and she phoned an ambulance for me, El Haj was at work. The ambulance came and undressed the child, the child was burnt on his upper body, his chest. … Liliane took care of everything, Liliane. I didn’t know what to do. … I knew nothing back then, no ambulance, no telephone, I also didn’t have a telephone. She saw that the child was burnt and phoned the ambulance. The ambulance took the child, who was burnt by the hot water. If she hadn’t been there, I would have waited until his father came home, around 18h. I knew nothing. The Moroccan women who lived nearby also didn’t know what to do.

148  •  Parallel Lives Revisited

From then on, there was a strong bond between them. An important part of their friendship was directed towards each other’s children, who grew up together: She took my girls and went for a walk with them, she brought them back and took them to dye their hair, she cut their hair, she took care of their hair. (laughs) She did their hair and brought them back, she took them to the seaside and brought them back, but I didn’t come with her. (laughs) I stayed at home with El Haj. My girls went to the seaside with her, the girls. To the seaside, they went with her, they walked with her. Tania [Liliane’s daughter]: I knew Rahmouna’s family [Rahmouna is one of Hassenia’s daughters] very well, I went there every day as a child, in the morning to get my hair braided and after school to play. … Later, I also went to Rahmouna’s house a lot because her sister helped me with French, maths and other subjects. I often stayed over at Rahmouna’s house, and I also often came to family feasts with them. … When I was a little girl, I could understand and speak Moroccan, now I only understand it.40

The Role of Children in Neighbour Relations Whereas Liliane took the children out for treats, Hassenia watched over her daughter Tania before and after school, when Liliane had to work. In this case, both women were about the same age. More often, when the friendship involved a young immigrant woman and an elderly Belgian woman, it was the Belgian woman who provided child care support for the immigrant woman, especially when the latter did go out to work. This was the case for example for Fatima, a first generation Moroccan immigrant who arrived in Ghent to join her husband in 1976. When she first arrived, her Belgian neighbour, Margriet, helped her to find her way in Belgian society. When Fatima had children, Margriet looked after them when she went out to work: She brought the children to school for me … And she went to get them without me. They grew up with her, from the moment they opened their eyes,  they called only for her, Mémé [a dialect word for grandmother]. … As if she was their mother, yes, because she spent more time with them than I did. I went to work, she went to get them. When they were ill, she looked after them at home. She came to my house and stayed with them until I came back from work. And the next day she said I could get ready at ease, have breakfast and go to work. I didn’t have to worry about the children, she would take care of them. … Even today, I haven’t forgotten about her! As God is my witness, I haven’t forgotten about her! Because she did everything for me.41

Immigrants’ Social Relations with Neighbours  •  149

Children do seem to have played a key role in the establishment of contacts between immigrants and their Belgian neighbours. They often remained important throughout the relationship. It was not rare, like in the story of Hassenia and Liliane, for the neighbours’ children to come to be seen as part of the family, and for them to consider the family as their own. Kamel, an Algerian immigrant from Oran, and his Belgian wife first left their own children in the care of their next-door neighbour, a Belgian woman, and later looked after the children of another neighbour themselves: We have raised their children as well, our neighbour’s children have been raised with us for five, six years. And their mother, she came to get them in the evenings. We did it for free, voluntarily … [about a specific girl:] And her mother, when she came to get her in the evening, she saw that we were eating red hot chili peppers, and the girl was eating with us. Her mother asked ‘What is she eating?’, and we said ‘You see, here we eat Algerian food.’ Very very strong red peppers and those kind of things. And Nina [the girl], she grew up like that, she grew up together with us’.42

However, when it came to neighbourhood relations, children were not always ties that brought neighbours closer together. A lot of incidents or bad feelings between neighbours were also related to their children, and more specifically to quarrels between them. For Hassenia, seeing the neigh-

Figure 5.3  Local and immigrant children play together in the Brugse Poort neighbourhood, 1970s. Amsab-Institute of Social History, collection De Meyer

150  •  Parallel Lives Revisited

bours’ children fighting with her son was a traumatic experience, as she did not have the (linguistic) skills to stop them: ‘They beat our children, but we didn’t know how to respond to them. … André’s children [one of the neighbours’ children] beat him … They beat Hassan [her son], and I didn’t know how to talk to them, like “don’t hit him”, or something like that. They hit him, and he came home crying, and I was silent. I didn’t know how to speak up.’43 In a white environment that was not only indifferent, but sometimes also hostile to these immigrant women, the few friendships they developed with Belgian women were even more important to them. Therefore, it is not surprising that they figure so prominently in their life stories. According to Joanna Herbert, these relationships were an important source of socioemotional support for immigrant women, ‘conferring affection, respect and social acceptance and fulfilling a fundamental psychological need’. The fact that they are recollected so fondly suggests that the women ‘preferred to remember these relationships because they signalled a positive experience and perhaps helped to justify staying’ (Herbert 2008, 129–30).

Changes over Time Relations across Ethnic Boundaries When the children who had brought immigrant and local families into contact, for better or for worse, grew up and left their parents’ house, the relations between these families tended to diminish in strength. Not only did immigrant families lose the connection they had made through their children; they also lost their interpreters. This is expressed clearly by Ali D., a Turkish immigrant from Piribeyli who arrived in Ghent in 1976: There are both Turks and Flemish people living in my street. The contact with my Flemish neighbours is good, but we only say ‘good day’. We don’t visit each other. … When they want to say something to me, I don’t understand it and when I want to say something, they don’t understand me. I would like some contact, but I don’t know the language and so contact is impossible. … I used to live in the Gerststraat [a beluik]. There, the neighbours visited us more often. They were mostly elderly people. They gave presents to our children and we had a good time there. My children translated for me and the contact was good. Now my children have left the house and I can’t get any contact anymore.44

His story also indicates another important factor. In the neighbourhood where Ali lived before, his neighbours were mostly elderly people. As we

Immigrants’ Social Relations with Neighbours  •  151

have seen, this was the case for all of the neighbourhoods in the nineteenthcentury belt around the centre, where most immigrants went to live. Over the course of the period under study, many of these elderly locals either passed away or were taken into care. As they disappeared, so did the neighbourhood-based relationships the immigrants had developed over the years, although those that had turned into real bonds sometimes remained strong. One example is the bond between the family of Mohamed T., a Moroccan immigrant who came to Ghent in 1973, and their old Belgian neighbour. Their relationship did not only consist of drinking tea at each other’s house, keeping spare keys, etc., but also extended to the old lady joining the family on their yearly trip to Morocco twice.45 When the family moved house, this bond did not diminish in strength. Mohamed kept visiting the old lady every Sunday, for years, until she passed away.46 Indeed, when real bonds between neighbours developed, these bonds could be long-lasting and were often maintained even when one of them had moved out of the neighbourhood, albeit mostly through letters, cards and phone calls at special occasions.47 As the Belgian neighbours they had known for so long were replaced by new ones (Belgian or immigrant), immigrant families had to construct new neighbourhood relations in a changing context, marked by economic crisis and racism (Deslé 1995, 543, 554–55). In Ghent, the general mood towards foreigners began to change from the early 1970s onwards, as indicated by the appearance of signs reading ‘interdit aux étrangers’ (forbidden for foreigners) or ‘interdit aux nord-africains’ (forbidden for North Africans) in the windows  of bars and cafés.48 This certainly was not conducive to the development of cross-ethnic social ties in the neighbourhood. Also, as more immigrant families came to live in neighbourhoods with higher spatial ­concentrations of ­co-ethnics (see chapter 4), the opportunity for interethnic contacts diminished. As Beyers and Vervoort have shown, social ties with locals are less likely to develop in ethnically concentrated than in ethnically mixed residential areas (Vervoort 2012; Beyers 2007a). Over time, neighbourly relations between local and immigrant families not only seem to have become less prevalent, they also seem to have grown less friendly. Some of the interviewees report neighbours behaving badly towards them only because they were foreigners, not because of bad personal experiences: [interpreter translating the answer of the interviewee] Previously, now he is gone, one of his neighbours was against Moroccans, actually against all migrants. But he left him alone. In fact, they left each other alone. They said hello, but that was all. For the rest, he has no problems here with the neighbours.49

152  •  Parallel Lives Revisited

This is a very good neighbourhood. I do have a neighbour with whom I used to have some problems. The children of this neighbour used to tell my children their dad did not like Turks, as his wife had run off with a Turk. After a couple of months, however, he got to know me a bit better and then, things went ok.50

In a similar way to how they narrated the problems they encountered at the workplace, immigrants tended to frame these negative encounters with Belgian neighbours as exceptional cases rather than as signifiers of ­widespread racism in the neighbourhood. Herbert (2008) noticed this as well, stating that her South Asian respondents tended to view racial incidents in the neighbourhood as isolated and self-contained (Herbert 2008, 114–15). Individual incidents between immigrants and their Belgian neighbours had occurred ever since they first arrived in Ghent. However, as the general attitude towards (non-European) immigrants became more negative, such incidents came to be framed more often in terms of ‘culture’ or ‘ethnicity’. For instance, quarrels between children offered locals an occasion to vent their negative feelings about the presence of ‘others’ in their neighbourhood in general, telling them to ‘go back to their country’. This is illustrated by the following incident, related by Robin Roeck, a Master student studying the integration processes of Turkish immigrants in Ghent in the early 1970s: In a small park in the neighbourhood, a group of Turkish and a group of local children started fighting, because the former had walked across the area where the latter were playing football, and as such had disturbed them in their game. The Turkish children shouted that the park belonged to everyone and that they didn’t have the right to reserve such a large area for themselves. … In the end, the heads of each group started fighting with each other. At that moment, an old man, who had been observing the whole scene and had been uttering threats to the Turkish children, interfered. He grabbed the Turkish child, pulled him away from the Belgian one and shouted ‘Leave these boys alone, go and play elsewhere but leave these boys to play football here! The park isn’t yours alone!’ In the meanwhile, a lady had approached the man, saying that he shouldn’t behave in such a brutal way towards these children. The man replied to her ‘They don’t understand anything else. Fighting is all they know, the older ones as well as the little ones, you saw it for yourself ’, to which the lady replied ‘They are only children, and children fight sometimes, surely that isn’t that bad?’. This really enraged the man, who then shouted, ‘Yes, but these here are Turks, they should stick to their own kind! We don’t need them here. They come here to take advantage of our jobs, of the child allowances we pay, of the social security, and now their children are chasing our children out of the park. Who do they think they are?!’ (Roeck 1971, 4–6)

It is not surprising that this incident happened in a park at the edge of Sluizeken-Muide, the neighbourhood with the highest percentage of

Immigrants’ Social Relations with Neighbours  •  153

i­mmigrants in the city (see chapter 4). As Smets and den Uyl (2008) have shown, incidents between local inhabitants and immigrant newcomers in such neighbourhoods can, under certain circumstances, develop into deeper ‘ethnic conflicts’, not only between individuals but also between groups of residents (Smets and den Uyl 2008, 1455–57). According to ethnic competition theories, locals perceive immigrants as more threatening in neighbourhoods with a high concentration of immigrants (see Blalock 1967 cited in Fenton 2010, 94; Vervoort 2012, 900). In Sluizeken-Muide, the deterioration of group relations between locals and immigrants reached a first peak in 1978, when some locals set up the Action Committee Sluizeken-Muide, demanding not only a revaluation of the neighbourhood, but first and foremost a ‘dispersal of the guest workers’ according to a ‘fixed norm of a maximum allowed percentage’.51 Reports on the situation in the neighbourhood from the late 1970s and early 1980s mention how local residents reacted negatively and even aggressively to the presence of immigrants in their midst: The housing stock in our neighbourhood has aged, that’s why many guest workers come here. This happened in a relatively short time, in around three years. The concentration quickly became noticeable, the difficulties quickly emerged. The original neighbourhood residents do not take it anymore, also want to assert themselves. Everywhere we meet the people [local inhabitants], we hear the same complaints.52 Many residents of Sluizeken-Muide don’t feel good about this s­ituation. The neighbourhood used to be theirs, now they feel threatened, they feel ­powerless. … It is not clear for the neighbourhood residents why the migrants are here, how long they will stay here, how the situation will evolve.  … Therefore, many locals react aggressively towards the migrants.53

However, in Ghent, it has never come to actual riots between locals and immigrants like the ones that took place for example in the Dutch cities of Rotterdam and Schiedam, in 1972 and 1976 respectively, or in the London neighbourhood of Brixton in 1981 (Bovenkerk 1999). Even when in 1982, a Turkish café was the object of a racist attack whereby one Turkish young man was shot, there was no large-scale rioting, only a peaceful protest action of the Belgian-led organization Anti-Fascist Front.54 The immigrant population in Ghent never closed ranks against the locals as some locals did against them. During the period under study, we found no evidence of immigrant newcomers responding as a group to the (verbal) aggression of local action groups such as the Action Committee Sluizeken-Muide. The only reaction we came across was that of immigrants going out of their way to be well-liked and respected by their Belgian neighbours. Some of them really had the intention to show the locals ‘a good image’ of immigrants. This desire to be on friendly terms with Belgian neighbours – a desire which many Muslims referred to as

154  •  Parallel Lives Revisited

a duty prescribed by the Koran – was sometimes thwarted by the reluctance of these neighbours to establish a friendly relationship. Mehmet B., a Turkish immigrant from Eskisehir who had arrived in Ghent at age eighteen to come and join his father, was really bothered by the indifference of his Belgian neighbours. He did his best to change this: Before, there were two old people of at least 70 living next door. They didn’t talk to me. One day, I went to see them and I told my neighbour that Belgium was her country and the world was Allah’s, but that the house next to hers was mine as I had bought it. ‘You have to get along with me’, I said, but they refused. Two years later, they had a problem with their roof. A drain pipe had broken and to fix it – the roofs were very high – the plumber had to pass through my house along my roof. My neighbour phoned me to ask permission. I immediately ran down the stairs and invited them over. My wife made coffee and cookies. Later, both of them returned home crying. They kept on weeping and saying sorry. From then on, we got along with them very well.55

Mehmet K., a Turkish immigrant who arrived in Ghent in 1976, also attached great importance to his relations with his neighbours. From his position as a teacher and later as a social worker, he tried to convince other Turkish people to maintain good relations with their neighbours, and with the Belgian population in general, so that there would be a better mutual understanding: In the summer, I am always together with my neighbours. We eat together. We talk in the garden for hours. When I go to Turkey, I bring a lot of presents with me. I give my house keys to my neighbour so that he feels I trust him. I always tell the Turks they have to give something to the Belgians when it’s Eid al-Fitr [the feast at the end of Ramadan]. I tell them they have to talk about the celebrations they have. I advise them to take a Belgian boy to Turkey, so that the boy can tell his parents what it was like in Turkey. They have to get to know each other.56

Relations within Ethnic Boundaries It was not only the relationships between immigrants and their Belgian neighbours, but also those between neighbours who were ­perceived as ­co-ethnics that changed over time. For many immigrants, over the course of the period under study, these relationships became more important. First of all, this was a consequence of the fact that ever more immigrants were coming to live in the city, and immigrants with the same ethnic background tended to concentrate in specific neighbourhoods. Immigrants who live in neighbourhoods with a higher concentration of co-ethnics tend to develop more relations

Immigrants’ Social Relations with Neighbours  •  155

within than across ethnic boundaries (Vervoort, Flap and Dagevos 2011; Beyers 2007a, 94–99, 208–10). Further, as during the period under study, many immigrants (especially men) lost their job and became permanently unemployed, they began to develop more relations in the neighbourhood, spending more time there than before. Going to local bars and cafés, which by now were run by co-ethnics, they mostly developed relations with people from the same country or even place of origin. These relations did not only increase in frequency, they also changed qualitatively, especially for people living in ‘ethnic neighbourhoods’ such as Sluizeken-Muide and later also the other neighbourhoods in the nineteenth-century belt. Van den Berg-Eldering (1978) found that ‘living separately’, away from locals, not only limited the contacts immigrants maintained across ethnic boundaries, but also rendered the social control that immigrant groups exercised over the behaviour of individual group members much more efficient (van den Berg-Eldering 1978, 184). Zohra, a Moroccan immigrant who arrived in Ghent in 1971 at the age of seven and went to live in the Brugse Poort neighbourhood, remembers how this social control was already very strong in the late 1970s: Zohra: I remember, I was fourteen and we were playing outside, and it was Ramadan. I had an apple in my hand, and my father just arrived. He saw me and grabbed me by the collar, and dragged me inside. The mosque wasn’t far from our house, and the people of the mosque passed there, and he said to me ‘you know it is Ramadan!’. If those people would happen to see me eat during Ramadan, that would have shamed my father. I didn’t give that kind of thing any thought. Interviewer: But your father said ‘If you must, then eat it inside?’ Zohra: Eat it inside. But don’t show it outside. He was mostly preoccupied with what the others would say of it.57

Avoiding this kind of social control was a major reason for some immigrant families to move out of these neighbourhoods, and into ‘whiter’ neighbourhoods, or even to one of the villages surrounding the city. Since the turn of the millennium, this has become a growing trend (Verhaeghe, Van der Bracht and Van de Putte 2012, 74–76).

Conclusion Together with the workplace, the neighbourhood is one of the most important sites for the social integration of newcomers, especially for the development of social contacts beyond old-world ties and ethnic boundaries. Still, when it comes to the study of the daily lives of immigrants, especially in

156  •  Parallel Lives Revisited

the past, social relations in the neighbourhood have been relatively underresearched. Those few historical studies that have paid explicit attention to immigrants’ neighbourly relations show how they were essential for the ways in which immigrants constructed their lives in the receiving society (for example, Gabaccia 1984; Beyers 2007b). Initially, individual immigrants without a network in the city got to know other immigrants from the same country or even place of origin in the lodging houses and other types of temporary housing where they typically ended up. As living together in these places was quite intense, the contacts immigrants made there generally developed into meaningful social relations that continued even after the parties involved were no longer housemates. When the first immigrants arrived in a neighbourhood, they were welcomed by the locals with great curiosity. At the same time, they were the object of charity for neighbours who thought – often correctly – that they were in dire need of help. This mixture of curiosity and charity became less important as more immigrants came to live in the neighbourhood, but tended to resurface when the first families arrived. Immigrant families’ chances of developing meaningful relations with their neighbours were initially thwarted by the fact that they were not intending to stay in Ghent for long. Their relatively high residential mobility within the city also blocked the development of neighbourly relations, as every move entailed a change of (immediate) neighbours. Apart from kin and other old-world ties, the first social contacts immigrant families made in the neighbourhood were generally other immigrants, coming from the same region or country of origin. Especially for women who were housebound, these neighbourhood-based social contacts were very important, as they provided support in their daily lives. The relations with these co-ethnics tended to be multi-stranded, overlapping with kinship or other old-world ties, workplace-based relations, etc. Despite the many barriers to the development of close neighbourly relations across ethnic boundaries, especially at a linguistic and cultural level, many immigrant families did move past the stage of ‘interdependencies’ with at least some of their Belgian neighbours. This was easiest for those who spoke French or Dutch, but immigrants with a limited knowledge of the language could also manage to get to know their Belgian neighbours. They either learnt to communicate through practice, or they had children who could function as interpreters. When immigrant families lived in the same n ­ eighbourhood for a longer time, curiosity and charity no longer dominated their relations with locals. A more general sense of good neighbourliness became more important, from both sides, as they now realized they were going to be neighbours for a while. Especially in the beluiken, where neighbours had to share the facilities, it was impossible to maintain purely functionalist ­relations. Also, in many cases, the need for company was an important moti-

Immigrants’ Social Relations with Neighbours  •  157

vation for neighbourly relations to develop, especially for women who were housebound, both Belgians and immigrants. On the one hand, interethnic neighbour relations that could be qualified as transactional relations and even as attachments were not at all rare. The development of real bonds between immigrants and their Belgian neighbours on the other hand was not that common. Such bonds did develop, especially between women. That special relationship with one Belgian neighbour, often an elderly woman, was a recurrent narrative among immigrant families in Ghent. During the period under study, the general climate towards (non-­ European) immigrants deteriorated in the whole of Western Europe, also in Ghent. Anti-immigrant feelings became more widespread and were politically translated through the success of extreme-rightist parties. Neighbourhood relations across ethnic boundaries, between immigrants and locals, were influenced by this deteriorating climate. Conflicts between neighbours came to be framed more in terms of ethnicity and culture. In those neighbourhoods where many immigrants lived, the local population began to feel threatened. In Ghent, an Action Committee was set up in the ‘Turkish’ Sluizeken-Muide neighbourhood; it asked the authorities to limit the number of immigrants in the neighbourhood and disperse them over the entire territory of the city, following the idea of a ‘tolerance threshold’. Immigrants in Ghent however do not seem to have reacted as a group against this changing climate. The only reactions we found were individual ones, aimed at spreading a positive image to the Belgian population in general or being respected by their neighbours as individuals. It was not only interethnic neighbour relations that changed during the period under study; relations with co-ethnics in the neighbourhood changed as well. Many immigrant families lived in neighbourhoods where the number of co-ethnics became ever larger, so that neighbourly relations within ethnic boundaries became more frequent. However, more co-ethnics in the neighbourhood generally also meant more social control, something that was not appreciated by all immigrant families.

Notes   1. Mustafa, a Turkish immigrant from Emirdağ, had only been in Belgium for one week when the police came round to tell him he was staying in the country illegally and had to leave. The landlady of the lodging house where he was staying (together with many other Turkish immigrants) found him a job in a frame making factory owned by one of her family members. The owner of the factory applied for a work permit for Mustafa, so that his status could be regularized. Mustafa also mentioned that he found his landlady

158  •  Parallel Lives Revisited

  2.   3.  4.

  5.   6.   7.   8.

  9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21.

always prepared to help whenever any of her foreign lodgers had problems. Interview with Mustafa (2). Dahmane, an Algerian immigrant who had been recruited by the Union Cotonnière, was helped by the landlord of the café where he rented a room. He accompanied Dahamane and his friend to the population office, in order to regulate their paperwork. Interview with Dahmane. Interview with Kamel. Interview with Kamel. M’hamed, who was recruited with seventeen other Moroccan men and housed in a dormitory in the Opgeëistenlaan, says that these seventeen were also regularly in contact in their free time. (Interpreter: ‘They were young. In their free time, they were always together’.) Interview with M’hamed. Interview with Nahib. Interview with Adem. Interview with Ömer. Talja Blokland observed an unequal status in the attachments of native Dutch to their immigrant neighbours, and noted that ‘self-righteousness and confidence that their efforts for the foreign residents of their neighbourhood were worthwhile made them feel superior’ (Blokland 2003b, 170–71). Interview with Faruk. Interview with Enzo. Ismail, a Turkish immigrant from Emirdağ who arrived in Ghent in 1966, remembers how the lady who owned the local café always treated him and his friends very nicely, and how she also helped to furnish their house. Interview with Ismail. Süleyman remembers the help from the people in their local café: ‘We have been received well, also in the café opposite our house. They knew we were foreigners but still they helped us. We didn’t understand everything and they tried until we had understood. They helped us on all grounds, materially and morally. They asked us what we needed, tables or chairs for example.’ Interview with Süleyman A. Interview with Ömer. For a very interesting study discussing the impact of the mining company on immigrants’ and locals’ spare time and housing, and as such on their neighbour relations, see Beyers 2007a. Interview with José Luis and Jesusa. Interview with Moh. Interview with Zohra. Interview with Yamina. Interview with Zineb. Interview with Habib. Interview with Halice, wife of Suleyman A. Mimount, a Berber woman who came to Ghent in 1970, was also scared of the Belgian neighbours: ‘No, my child, I will tell you. The women who have been fetched by their husbands, stayed indoors. We didn’t know anyone, only the locals. We didn’t talk to them. We were afraid of the locals and kept the doors closed.’ Interview with Mimount. Interview with Hassenia. As the interviews with Moroccan Riffi women gave more detailed information on this subject than interviews with other women, the following analysis of immigrant women’s neighbourhood relations within ethnic boundaries are biased towards their experiences. Still, these experiences can probably be generalized for many rural women, depending on the cultural values of their family and the extent of their old-world network in the city of Ghent.

Immigrants’ Social Relations with Neighbours  •  159

22. Interview with Moumna. 23. Interview with Zineb. Animzade, a Turkish immigrant from Posof who came to join her husband in Ghent in the early 1970s, and who was not permitted by her husband to go out to work, only visited her immediate neighbours, Göcmen Turks (whom she calls ‘Albanians’ to indicate the difference between them); there were two other women from Posof in the neighbourhood, but as she did not know them, she did not go to visit them: ‘In the neighbourhood were two women from Posof, but I didn’t know them. Next to us however there were Albanians and we visited each other.’ Interview with Animzade. Ömer’s wife, who came from Istanbul to join her husband in 1965, started work quite quickly after her arrival but stopped again in 1966, when her first child was born. Her neighbourhood contacts were with (unspecified, but most likely Turkish) women who lived nearby: ‘I wasn’t scared at all. There were a couple of women in the neighbourhood whom I could visit. … We had good neighbours. The men went to a café and the women regularly gathered in the house of one of us. Later we moved. Actually I never felt alone, only a little bit in the beginning.’ Interview with Ömer’s wife. 24. Interviews with Hassenia and Hadda. 25. Interview with Mimount. 26. Interview with Hassenia. Hadda’s neighbourly relations could also be characterized as polite but indifferent: ‘I didn’t really talk to the Belgians, I didn’t understand them. They passed by my door and said “Hi Madam”, and I let my daughter play in front of the door, and then they would come closer, “Oh, beautiful baby, oh beautiful baby”, and I just kept smiling at them. (laughs) Yes, that’s the way it went, I didn’t know what to say.’ Interview with Hadda. 27. Interview with Elomar. 28. Interview with Abdelkader A. 29. Interview with Ahmed M. For Moh, a Moroccan immigrant who lived in Algeria and came to Ghent in 1966, the quiet respect neighbours show for each other is one of the great things about living in Belgium: ‘I like everything here. I don’t dislike anything. No one ever bothered me. No one ever attacked me on the street, even if they were drunk. No one ever fought with me. Not at home, not the neighbours, men or women. Nothing. You saw how many people live here. Do you hear one sound or noise? Nothing! Is there anywhere else in the world something like this peace? Imagine, someone lives above you. You come home from your work at night to rest. And you hear a noise. Someone turns on the radio. Another watches TV very loudly. You can’t sleep. You keep the neighbours up. That isn’t good. Bad! It makes you angry, no? And now, you come home with pleasure. And you fall asleep quietly. Just as if you were living on your own.’ Interview with Moh. 30. Interview with Maria de la Luz. L’Arbi, a Moroccan immigrant from Kenitra who arrived in Ghent in 1971, also reminisces about a period of good neighbourliness: ‘We had a good understanding. All Belgians were friendly. Some of them came to visit you and asked if you needed anything, there were people who came to help you. I used to address people in French then. Everyone answered. You could always go see them if you needed anything. It was a time when there were people who still talked, if you asked something in French, they answered.’ Interview with L’Arbi. 31. Beyers also found language barriers to be an important impediment to the development of interethnic neighbour relations (Beyers 2007a, 95–97, 211–13). 32. Interview with Ahmed M. 33. The role of children as mediators in interethnic neighbour relations is also attested by Beyers. She distinguishes between two roles that children fulfilled, that of interpreter and that of informer. Here, I focus on children’s role as interpreter (Beyers 2007a, 99).

160  •  Parallel Lives Revisited

34. These cultural prescriptions were attested by van den Berg-Eldering as important ­inhibitors for informal contacts between Dutch and Moroccans (van den Berg-Eldering 1978, 183). 35. Interview with Mehmet K. Kamel and Miloud also remarked that they cannot maintain close friendships with Belgians, as there are cultural differences that inhibit the formation of such friendships. Kamel: ‘My best friends, I have Algerian friend, not that many however … but I have a lot of Belgian mates. Many many, it’s true, who like me a lot, and whom I like a lot as well, you know? But I know that there are, we are friends up to a certain point.’ Miloud: ‘There are limits that need to be respected, e.g. that you cannot visit at whatever time you choose.’ Interviewer: ‘And that is not the case with Algerian friends?’ Kamel: ‘No, with Algerian friends no. … For the moment, there are three or four of us, including Tahar, who meet up regularly, amongst ourselves. And we talk, we talk about Algerian politics, of course, about the situation of the country, what is going on … And that is what friends are, you can come ring the doorbell any time, or if there’s a problem, you phone, and they are one hundred percent ready, without thinking about it.’ Interview with Kamel and Miloud. 36. Interview with Mohamed E. 37. Interview with Mohamed E. 38. Interview with Khadija. 39. Interview with Hadda. 40. Interview with Tania. 41. Interview with Fatima B. 42. Interview with Kamel. Zohra, a Moroccan immigrant who arrived in Ghent in 1971, at age seven, together with her mother and her brother, had a ‘mémé’, Maria, a Yugoslavian woman who had lived in Ghent for a long time: ‘So every Sunday, we went to see mémé, well, Maria, across the street, in her house … So she looked after us. She liked children, but never had any herself. My mum mostly cleaned the house and didn’t like it that much to have to drag the children around, that was more mémé’s job. Actually, we had somebody who did fun stuff with us: go to Maaltepark, to the Farmer’s House. She was strict but righteous.’ Interview with Zohra. 43. Interview with Hassenia. 44. Interview with Ali D. 45. Indeed, Belgian neighbours who had become close to an immigrant family were sometimes invited to come along when the family returned to their home country for the holidays. Zohra’s family also took their adoptive grandmother with them when they went to Tanger for the summer. Interview with Zohra. 46. Interview with Mohamed T. In fact, the house the family moved to was bought especially for them by their neighbour. On the ways in which immigrant families were supported by their Belgian contacts in their search for better housing, see De Bock 2013. 47. Moh is still in contact with his neighbours from the 1970s, Suzanne and Bernard, who came with him to pick up his family at the airport; he sees them now and then. Interview with Moh. Fatima still receives phone calls and visits from the children of Margriet, her old neighbour from the 1970s, who has already passed away. Interview with Fatima. 48. This has been attested by several interviewees and is documented in a documentary from 1972 (Buyens 1972). 49. Interview with anonymous Moroccan man from Mont Arruit (Rif ), who arrived in Ghent in 1969. 50. Interview with Yusuf.

Immigrants’ Social Relations with Neighbours  •  161

51. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, Dokumentatiemap Woonfonds Initiatieven, August 1978, p. 7. 52. Amsab-ISG, Archive Maurice Maréchal, Archive Woonfonds vzw, Dokumentatiemap Woonfonds Initiatieven, August 1978, p. 7. 53. ‘Wanneer Een Migrantenbeleid? Beleidsvoorstellen van V.z.w. De Buurt, De Sleep, Woonfonds Uit de Wijk Sluizeken-Muide Te Gent.’ 1982, 8–9 54. Archive Amsab-ISG, Archive Maurice Maréchal, Journal of the association SGOV, volume 3, number 9, 28 October 1982. 55. Interview with Mehmet B. 56. Interview with Mehmet K. 57. Interview with Zohra.

Conclusion 

The analysis presented in this book has shown that for the early years of postwar labour migration, the idea that immigrants were leading ‘parallel lives’ holds no ground. Contrary to the images of segregated workplaces and residential ghettoes that dominate our representation of the past, in the middlesized cities and towns of Western Europe, immigrants formed only a small minority living and working among a majority of locals. Most of them had ample opportunity to make social contacts across ethnic boundaries. This opportunity also translated itself into the development of actual interethnic relations, and this despite the many barriers that existed between newcomers and established, not least their language differences. Furthermore, even though most of these interethnic relations did not surpass the level of polite friendliness, many immigrants did develop a limited number of profound relations, even close friendships, with locals. However, throughout the period under study, a steadily growing fraction of immigrants came to work and live in relatively more segregated spaces. As a consequence, their opportunity for social contact across ethnic boundaries diminished, while that for contact with co-ethnics increased. This situation of growing concentration and segregation, combined with other factors, resulted in interethnic relations becoming a less important feature of immigrants’ lives. At the same time, immigrants’ social relations within ethnic boundaries became more important and more exclusive. The following paragraphs provide a summary of the arguments presented in this book, first looking at the structural position of immigrants in the

Conclusion • 163

labour and housing markets, then at the social relations they developed there. The focus here is on change over time and how this change can be explained.

The Structural Position of Mediterranean Immigrants in the Labour and Housing Markets In the domain of work, postwar labour migrants have always been concentrated in specific economic sectors, carrying out specific kinds of jobs. These jobs belonged to the secondary sector of the labour market. They were heavy, unhealthy, badly paid and low-status, often in industries where the availability of work was insecure. Immigrants as a group remained blocked in these secondary jobs for a variety of reasons, such as the singularities of the economic opportunity structure, the objectives of immigrants as target earners, their cultural capital and the discriminatory effects of immigration legislation. In a period of economic growth, this resulted in a high degree of horizontal job mobility – moving between jobs without realizing an actual improvement of one’s work situation – that was both chosen and forced upon them. After 1974, when the labour market began to shrink, this mobility quickly came to a halt. Even within the secondary sector of the labour market, immigrants tended to be concentrated in specific companies. Such concentrations were related to the attitudes of employers towards immigrant labour, as well as to the mechanisms of network-based job searching. The latter in particular resulted in ethnicity-specific concentrations on the shop floor, although these could also have been due to the personal preferences of employers. As a result, when immigrants worked with other immigrants, these tended to come from the same country or even region of origin. Nevertheless, real ‘ethnic workplaces’ where the majority of employees were co-ethnics were rare. In Ghent, only immigrant employers were used to hiring mostly or exclusively co-ethnics. During the period under study, such employers were not numerous and typically employed only a few workers. The large majority of immigrants worked in what we have called ‘mixed workplaces’, surrounded by mostly Belgian colleagues. In the 1960s, immigrants only made up a small percentage of any company’s workforce in the city of Ghent – up to around 20 per cent in those companies that were renowned as immigrant employers. While these percentages rose over the course of the 1970s, immigrants continued to work with a majority of Belgian colleagues. On the one hand, this was because most companies still employed many Belgians. On the other hand, it was because Belgian employers tended to purposefully separate their immigrant workers in order to avoid the formation of cliques along ethnic lines. However, in those companies

164  •  Parallel Lives Revisited

where work was lowly valued or in unpopular shifts (such as the night shift in the textile industry), percentages of immigrant workers did reach 30 to 40 per cent by the end of the decade. Even if ‘mixed workplaces’ remained the rule throughout the period under study, many immigrant workers did get cut off from interethnic or in fact any kind of workplace-based social contact during those years. The reason for this was that they lost their job. The post-industrial transition that started in the 1970s was particularly detrimental to jobs in the secondary sector of the labour market. In the aftermath of the 1974 crisis, many immigrant workers were pushed into unemployment for long periods, sometimes even for the remainder of their active years. All in all, it appears that unemployment rather than an increase in ethnic segregation was the main cause for immigrant labourers’ diminished opportunities for interethnic contact in this period. In the domain of housing, newly arrived independent labour migrants tended to live in specific forms of temporary housing, such as lodging houses or employer-provided housing. While this resulted in their initial dispersion over the city, they were still rather segregated within these homes. When they moved into more permanent forms of housing, immigrants typically ended up in cheap, low-quality housing that belonged to the secondary sector of the housing market. Initially, renting this kind of housing coincided with the immigrants’ goals as target earners – trying to spend as little money as ­possible – and was compatible with their basic needs as single workers. Moreover, such housing was readily available due to the ongoing suburban flight during the period under study. Here, immigrants functioned as ‘replacement dwellers’, as there were no other candidates to rent these houses. This position also led them to move house frequently, continuously looking for lower rents or minor improvements in their living conditions. However, when their housing needs and objectives changed – mostly through forming a family – they found themselves stuck in this residual rental sector, with little chance of moving into a decent dwelling. The main cause of their blocked upward mobility appears to have been the high degree of discrimination, not only among private landlords but also among the city’s social housing associations. The spatial arrangement of immigrant families in Ghent reflected the dispersion of this low-quality housing. Immigrant families came to be concentrated in those neighbourhoods that boasted a high degree of vacancy and derelict buildings, especially the nineteenth-century belt to the north and east of the city centre. The spatial concentration of immigrants in specific streets within these neighbourhoods can be related to discrimination and bad housing on the one hand – visible for example in the high percentages of immigrant families in some of the beluiken – and to the effect of social networks and the preferences of immigrants themselves on the other hand

Conclusion • 165

(De Bock 2013, 419–36). Nevertheless, throughout the 1960s their numbers were still very low and they only made up a small minority in the neighbourhoods where they lived. From the 1970s onwards, the concentration of immigrants in specific neighbourhoods became more pronounced. New immigrants came to replace the Belgian inhabitants of these neighbourhoods, whose large-scale departure to the suburbs had come to take the dimensions of an actual white flight. Increasing discrimination in the housing market relegated immigrant families to those areas that were already inhabited by immigrants. There, landlords did not have much to gain by wielding a racist letting policy. The decade also saw the birth of the first ‘ethnic neighbourhoods’ in the city, starting with Sluizeken-Muide, an area in the nineteenth-century belt that came to be perceived as the first ‘Turkish quarter’. For a variety of reasons, for example the provision of specialized services catering to the needs of immigrant communities, these neighbourhoods attracted ever more newcomers. However, even in these areas, the immigrant population as a whole – let alone one specific ethnic group – still made up only a minority of the total population. In 1981, the housing block with the highest percentage of immigrants city-wide accommodated 31 per cent non-Belgians. Clearly, throughout the period of study, immigrants in Ghent continued to live together with a majority of Belgian neighbours.

The Social Integration of Mediterranean Immigrants in the Workplace and the Neighbourhood During the period under study, both the workplace and the neighbourhood offered ample opportunities for immigrants to come into contact with locals. One exception to this was the situation of people working in ethnic workplaces, who only rarely had Belgian colleagues. Another one was that of live-in domestics, for whom only highly unequal relationships with Belgian employers and neighbours of the upper classes were within reach. Despite the opportunities to meet locals, most immigrants initially turned to co-ethnics for social contact, both in the workplace and in the neighbourhood. The fact that they spoke the same language was an especially important facilitator in this, as was their shared cultural understanding. Furthermore, colleagues, employers and neighbours who were co-ethnics tended to be bound to each other through multi-stranded ties, combining roles of kin- and friendship, a shared place of origin, and membership of religious or other associations to relationships at work or in the neighbourhood. Unsurprisingly, these contacts frequently developed into meaningful social relations.

166  •  Parallel Lives Revisited

For many immigrants, the construction of such meaningful relations across ethnic boundaries was less obvious. Even though they were in contact with Belgian colleagues and neighbours on a daily basis, linguistic and cultural differences impeded the advancement of social relations beyond the transactional stage. Others, who were able to effectively communicate with locals and had already become accustomed to living in a European(ized) urban environment, found it easier to get to know their Belgian colleagues and neighbours better, and develop more meaningful social relations with them. Initially, speaking French was a major advantage in this. This changed however from the mid-1970s onwards, as the language fell into disfavour among many of the locals. Still, even for those who found themselves confronted with linguistic and cultural barriers, the development of attachments or even bonds across ethnic boundaries was not impossible. The degree to which immigrants were able and inclined to construct such relationships was strongly related to specific social variables. In the domain of work, age and marital status were key in the development of cross-ethnic social relations. When colleagues were young and unmarried, they had more time to spend together outside of work. Language problems resolved themselves as this kind of social contact allowed immigrants to pick up the language from their colleagues. As they grew older, got married and had children, such free time diminished greatly, and so did the intensity of workplace-based social relations. The opposite was true for relationships in the neighbourhood: relations with Belgian neighbours intensified as individual labour migrants were joined by their families. Children in particular turned out to be ‘bridge builders’, at first as a common point of interest and later, as they learned the local language, as readily available interpreters. However, more than the presence of children, it was the amount of time spent at home that nurtured the development of close neighbourly relations across ethnic boundaries. In particular, immigrant women who did not go out to work tended to create bonds with (mostly elderly) Belgian women in the neighbourhood. Such relations seem to have developed based on a mutual need for company, despite sometimes high linguistic and cultural barriers. These female bonds often extended to include the families on both sides of the friendship, and tended to be long-lasting, continuing even after one or both families moved away from the neighbourhood. From the mid-1970s onwards, as unemployment among immigrant workers soared, those who no longer had a job also lost their daily contact with Belgian colleagues. The cafés, club houses and places of prayer where they came to spend a lot of their time hardly offered opportunities for such interethnic contact. At the same time, the first ‘ethnic neighbourhoods’ came into being, and even outside of those, more immigrants came to live in areas where many other immigrants lived. All of this changed the dynamics

Conclusion • 167

of co-ethnic relations, which became invested with an element of social control. At the same time, the decrease in opportunities for cross-ethnic social contact resulted in a decrease of social relations across ethnic boundaries. As the elderly neighbours with whom they had created close attachments over the years passed away or moved into care homes, and the few new neighbours were harder to reach in a climate of economic crisis and more overt racism, interethnic contacts in the neighbourhood became rare.

What Happened Next? Of course, the history of guest worker migration does not stop in the late 1970s, even though many Western governments would have preferred it so. Despite sometimes surprisingly high rates of return, many immigrants did settle permanently in the receiving societies. Their children grew up there, had children themselves, and today, we have arrived at the fourth generation, the great-grandchildren of the pioneering guest workers. Over the course of the 1980s, the lasting presence of these populations became an undeniable fact – one that receiving states were going to have to deal with, whether they liked it or not. In Ghent, the numbers of people of Mediterranean origin went up continuously from 8,804 or 3.7 per cent of the population in 1980 to 12,397 or 5.4 per cent in 1993, after which they stabilized for a while. Since the year 2000, due to a wave of naturalizations, the numbers provided by the population office are unfortunately no longer reliable. Continuous legislative changes, culminating in an important reform of the Belgian nationality code in 1999, made the acquisition of the Belgian nationality almost a formality. They also greatly expanded the categories of people who automatically received Belgian nationality at birth. As a result, many immigrants and their descendants became registered as Belgian nationals, and their foreign roots no longer showed in the official statistics (Verschueren 1995; Lesthaeghe 2000, 6). Only the data recently produced by the Flemish Migration and Integration Monitor allow us to get a more accurate picture. According to these data, the Mediterranean population of Ghent counted 32,484 people in 2013, representing 19 per cent of the total urban population.1 As the number of Belgian nationals among the Mediterranean population grew, they no longer came to be seen as ‘guest workers’. Rather, from the 1980s onwards, official sources started to refer to them as ‘migrants’. A decade later, this terminology was replaced by the word ‘allochtonous’ (not from this ground), differentiating them from the ‘autochtonous locals’ by their vaguely defined ‘foreign roots’, as opposed to their place of birth (Verhaeghe 2010). Despite their having grown up or being born in the receiving country, many

168  •  Parallel Lives Revisited

of the children of the postwar immigrants followed in the footsteps of their parents. The trend towards increased separateness that was started in the aftermath of the crisis, continued throughout the 1980s and 1990s. At a structural level, this finding is valid for immigrants and their descendants of all nationalities, albeit to different degrees; however, in the public and the political discourse, it was only those people with roots outside the European Union that came to be perceived as leading ‘parallel lives’ (Martiniello and Rath 2010, 246–49). From the 1980s onwards, relatively segregated workplaces became more common in Ghent. Immigrant workers themselves indicate that in companies such as UCO and Fabelta, the percentages of ‘ethnic minorities’ increased significantly over the course of that decade. Ahmed B., an Algerian immigrant who worked for UCO Ltd from 1984 until 2004, estimated that some 80 per cent of the workforce in the factory was foreign, especially in the night and weekend shifts.2 Rabah, also Algerian, who worked for Fabelta Ltd from 1973 until 1995, remembered that after the company was taken over by the Beaulieu group in 1982, immigrants suddenly came to make up a large majority of the workforce.3 As the textile industry went downhill rapidly, the sector continued to suffer from ‘white flight’. This resulted in an ever higher percentage of jobs being taken over by immigrant children, who thereby became stuck in the same kind of jobs their parents had initially arrived to work in. This phenomenon was not particular to Ghent. All over Western Europe, an ethno-stratification of the labour market was emerging, where people of immigrant background remained concentrated in secondary sectors, got lower wages, more often worked part-time and more often were unemployed (Verhoeven and Martens 2000; Verhoeven 2001, 226–27). This was equally valid for the so-called ‘second generation’, who remained overrepresented in non-skilled labour categories and underrepresented in the top segment of the labour market. Data for Belgium from 1991 show how only 4.1 per cent and 8.3 per cent of employed Turkish and Moroccan men aged eighteen to twenty-nine took up managerial positions, whereas this rate was more than 25 per cent among their Belgian peers. Non-skilled labour on the other hand employed 33.8 per cent of Turkish and 35.9 per cent of Moroccan youngsters, whereas this was the case for only 19.4 per cent of Belgian youngsters (Neels and Stoop 2000, 282–84). As more Belgians left unskilled jobs and more immigrant children entered them, there was an increase in ethnic ­segregation at the workplace, which resulted in decreasing opportunities for the second generation to come into contact with Belgian colleagues. However, even more problematic was the fact that huge numbers of immigrant children did not manage to find a job at all. In 1991, 33.9 per cent of Turkish and 34.4 per cent of Moroccan youngsters were unable to find work,

Conclusion • 169

compared to only 9.6 per cent of their Belgian peers (Neels and Stoop 2000, 282–83). The total absence of workplace-based relations further diminished the chances for those who were unemployed to create social contacts across ethnic boundaries, inevitably leading to more separateness. One of the consequences of these high levels of unemployment was the rise of immigrant entrepreneurship. While up until the 1980s, the level of entrepreneurship among immigrants in Ghent remained low, from that decade onwards, the city experienced a veritable boom of small-scale enterprises. These ranged from shops and cafés to restaurants and travel agencies. Especially among the Turkish population, the development of these enterprises, together with the rise of ethnic associations, religious institutions and Turkish professionals providing services to an (almost) exclusively Turkish clientele, resulted in the setup of what has been called an ‘ethnic enclave’. This type of social organization is characterized by social-economic disadvantage, the existence of an ‘ethnic economy’, a strong institutionalization and diversification of services, as well as a degree of spatial concentration (Verhaeghe, Van der Bracht and Van de Putte 2012, 16–21). Indeed, in the field of housing, the increasing residential concentration and growing percentages of immigrants in certain neighbourhoods that had started in the 1970s, continued in the following decades. However, in Ghent, the housing situation of immigrant families never came close to the ghettoes earlier observers had feared. A street level analysis of the Brugse Poort area in 1986 showed only three streets where more than 50 per cent of the inhabitants were non-Belgian. Two thirds of the non-Belgian population still lived in streets where 75 per cent of their neighbours were Belgian.4 In 1990, a similar analysis within the Sluizeken-Muide neighbourhood revealed a more segregated picture. Nearly half of the non-Belgian population lived in streets where more than 50 per cent of the inhabitants were non-Belgian5; only one in ten lived together with 75 per cent or more locals.6 Clearly, the opportunity for immigrants to meet locals in these neighbourhoods had considerably decreased over time. Still, even in this area, boasting the highest concentration of immigrants in the entire city (37 per cent of the population as a whole), non-Belgians continued to share their streets with (some) Belgian neighbours. In Sluizeken-Muide, but also in other areas of the city’s nineteenth-century belt, economic and spatial changes resulted in the development of ‘ethnic enclaves’, at least for those ethnic groups that were numerous enough, i.e. the Turks and to a lesser extent also the Moroccans. As these enclaves expanded, both in inhabitants and in the offer of goods and services they provided, their existence itself contributed to a growing separateness between immigrants and locals from a social as well as an identificational point of view (Verhaeghe, Van der Bracht and Van de Putte 2012, 19–20). Other factors

170  •  Parallel Lives Revisited

that are thought to have encouraged the development of (relatively) ‘parallel lives’ were, for example, the growing prevalence of ‘black schools’ and the boom in communication and travel technologies that allowed for a more continuous and direct contact with the country of origin. Everyday contacts with ethnic Belgians became more limited and ethnic identities came to take on a more important part in people’s daily lives (De Gendt 2014, 181–85). This is sometimes seen as one of the causes of the following boom of immigration marriages, whereby people who were born and raised in Belgium went to look for a partner in the country of origin. For first generation postwar labour migrants, the prevalence of such immigration marriages is rarely questioned, as the gender balance was such that their choice was almost exclusively between Belgian partners and partners still living in the home country. The option of marrying a partner from abroad became much more contentious for the second and following generations. While marriages within the ‘own’ ethnic group were already interpreted as ‘evidence of distance from the majority population’ (Sürig and Wilmes 2015, 169), marriages that brought in new immigrants tended to be seen as a way of circumventing immigration regulations, and as a choice against ­assimilation (González Ferrer 2006, 172). As a result, they were the subject of considerable suspicion and increasing scrutiny from the side of the receiving society (Caestecker 2005). Of the old immigrant groups, it was Turkish immigrants that practiced such marriage migration most keenly. At the beginning of the 1990s, second and third generation Turkish men and women in Belgium were more likely than their first generation counterparts to ‘import’ a partner from the country of origin.7 The prevalence of this practice was such that the rates of interethnic marriage started to decrease compared to previous generations. Whereas in 1991, 4.3 per cent of first generation Turkish men in Ghent were married to a Belgian partner, this was only the case for 1 per cent of the second generation. For Turkish women, the percentages went down from 2 per cent to none at all.8 Over the course of the 1980s and 1990s, growing rates of unemployment among ethnic minorities, decreasing numbers of Belgian colleagues in the workplace, the development of an ethnic economy and ‘parallel institutions’, an increasing degree of residential segregation, and a more ethnically oriented marriage pattern resulted in the postwar immigrant communities being accused of leading ‘parallel lives’. However, when this idea was put to the test by empirical research, it invariably led to the conclusion that – at least as far as Europe was concerned – ‘parallel societies’ in the strict sense of the word did not exist. Immigrants and their descendants were never completely separated from other members of society, neither from a spatial nor from a social point of view (Halm and Sauer 2006). Furthermore, as we have

Conclusion • 171

tried to prove in this book, the existence of a certain degree of separateness is something that developed in a specific historical context of economic crisis and a socio-political backlash against immigration. It is not something that automatically came forth from the preferences of immigrants themselves. Today, there are indications that the 1980s–1990s trends towards more separateness are being reversed. Whereas the perception of (non-European) Mediterranean immigrants and their descendants as leading parallel lives has not diminished, their economic position is very slowly improving, they are becoming more spatially dispersed, and there are less immigration marriages and more mixed couples. These three factors are considered crucial in breaking down the barriers to interethnic contact. Data for 2008 show how the second generation is slowly catching up with its Belgian peers when it comes to labour market participation. Whereas the activity rate of first generation immigrants between 25 and 49 years old, born outside the EU, was at 62.4 per cent, it had risen to 69.9 per cent for those of the second generation – although the gap with Belgians (whose activity rate lay at 90.3 per cent) remained high (Djait, Boussé and Herremans 2011). Also, second and third generation youngsters still have much more trouble than Belgians in finding a first job – it takes them longer and many more of them are still unemployed one year after graduation. Even when they do find work, the kind of jobs they find continue to be qualitatively inferior to those found by their Belgian peers. This is more often the case for youngsters of Turkish and North African descent than for those whose parents came from Southern Europe (Glorieux, Laurijssen and Van Dorsselaer 2009, 95–96). In these matters, as in the education of immigrant children, Belgium is among the worst performers within Europe. Still, even in Belgium a middle class of immigrant origin is slowly emerging and more and more descendants of postwar guest workers are making it to the top. Today, Ghent counts a lot of entrepreneurs, doctors, lawyers and politicians whose parents or grandparents had come there with the idea to save up some money and go back home (Verhaeghe, Van der Bracht and Van de Putte 2012, 40). At another level, previously notoriously white employers have opened their doors to employees with a foreign background. Whereas in the 1970s, it was highly exceptional for immigrants to find a job in Volvo Cars Ghent, in 2013 almost one in five of its 3,700 employees had foreign roots (Rasking 2013). As these companies, as well as government institutions such as universities and municipalities, become more open to ‘allochtonous’ workers, new types of workplaces become mixed and more people get the chance to come into contact with others across ethnic boundaries. Even more so than their disadvantaged socio-economic position, the residential concentration of immigrants and their descendants in the ‘ethnic enclaves’ of Western Europe seems to have reached a crucial turning point.

172  •  Parallel Lives Revisited

In Ghent, the increasing spatial segregation of Turkish and North African immigrants came to a halt around the turn of the millennium. Between 2001 and 2011, the segregation index (indicating which percentage of people should move to evenly spread out the population over the entire city) of the Turkish population fell from 65 to 51 per cent, and that of the Moroccan population from 53 to 42 per cent. The highest percentages of (Turkish) immigrants in any neighbourhood were measured in 2001 (44 per cent) and 2003 (52 per cent), but these have since gone down considerably. Further, over the course of the 2000s, the concentration of Turkish immigrants in those neighbourhoods that had come to be seen as ‘ethnic enclaves’ dropped significantly, from 19 to 16 per cent, and even more spectacularly in the very core of the enclave, from 33 to 25 per cent. Finally, those people who left the immigrant neighbourhoods in the nineteenth-century belt went to live in the twentieth-century suburbs, that were until recently almost exclusively white (Verhaeghe 2012, 61–63, 68–73, 74–76). Of course, these data do not tell us anything about the development of actual neighbourly relations across ethnic boundaries. However, they do indicate that the opportunity for such relations has increased considerably, not only for the former immigrants and their descendants, but also for a large group of ‘locals’ who previously lived in all-white neighbourhoods. Finally, over the past two decades, the partner choices of immigrants and their descendants have also changed in a way that suggests a move towards more openness. On the one hand, immigration marriages, involving partners from the country of origin, have become much less important. Whereas in the late 1990s, scholars were still saying that ‘the proportion [of immigrants and their descendants] married to a partner from the country of origin has certainly not declined in the most recent period’ (Lievens 1999, 740), things have changed considerably since then. From the turn of the millennium onwards, the arrival of partners from the countries of origin of the guest worker communities – and especially from Turkey and Morocco – has seen a constant and substantial decline (Caestecker et al. 2013, 11–15). Between 2001 and 2008, the percentage of immigration marriages for Turkish and Moroccan immigrants and their descendants in Ghent has gone down from 72 to 49 per cent and from 61 to 38 per cent respectively. Among the second generation, immigration marriages are swiftly decreasing, counting for just over 40 per cent for Turkish and 27 per cent for Moroccan youth in 2008 (Verhaeghe, Van der Bracht and Van de Putte 2012, 107–11). Interethnic marriages on the other hand are on the rise, and this is the case all over Europe. Research in France, for example, has clearly shown how the second generation more often marries with a French national born in France than the first generation, with the exception of Turkish men and Portuguese women (Safi 2010, 101–3). Similarly, Germany has witnessed a

Conclusion • 173

general growing tendency from one immigrant generation to the other to marry across ethnic boundaries (González Ferrer 2006, 177–78). In Belgium, the percentages of interethnic marriages differ between groups. However, for the Turkish and Moroccan populations of the second generation, they are approaching similar levels, reaching 14 and 16.5 per cent respectively by 2008 (Caestecker et al. 2013, 11–13). In Flanders (data for Ghent are missing), the relative increase in mixed marriages over the period 2002–2007 is impressive: 79.3 and 112.8 per cent for second generation Turkish men and women, and 22.9 and 43.5 per cent for second generation Moroccan men and women. The trend towards ever more interethnic marriages is therefore undeniable, even if the total numbers remain small (Verhaeghe, Van der Bracht and Van de Putte 2012, 112–14).

Parallel Lives and the Question of Choice The mechanisms behind the ‘move towards more separateness’ that was visible between the mid-1970s and the new millennium are manifold. It has been the aim of this book to provide a better understanding of these. Contrary to the accusations of immigrants’ presumed unwillingness to integrate that often accompany the ‘parallel lives’ discourse, this book has told a more complex story, bringing to the fore elements that otherwise tend to be treated as secondary. First, it has shown how the second-rate position to which immigrants were assigned when they first arrived, together with the mechanisms that blocked their upward mobility, have nourished the creation of ethnic concentrations at the workplace and in the neighbourhood. Second, it has pointed towards the economic crisis and the job losses it entailed as having a major impact on the structural position of immigrants and their descendants and on their opportunities to mix. Third, it has made a strong case for the influence that individuals and institutions of the receiving society had on the advancement of segregated workplaces and neighbourhoods and the dwindling opportunities for meaningful social relations across ethnic boundaries, expressed for example through discriminatory hiring and renting practices as well as different forms of ‘white flight’. The complex interplay of these and other factors in the way things have developed is a strong counter-argument to the idea that ‘parallel lives’ can be put down to a simple choice or preference of the immigrants concerned. It is also an argument for revising the belief that a higher degree of ethnic heterogeneity as such causes a decrease in social capital (Putnam 2007). The research in this book questions the general validity of this thesis, showing how the relationship between ‘more diversity’ and ‘less interethnic contact’ (as one of the many measurements of social capital) is not straightforward

174  •  Parallel Lives Revisited

at all. It is related to changes in opportunity structures, in the characteristics of local and immigrant populations and in the behaviour of individuals and groups, both immigrants and locals. The move towards higher degrees of concentration and relative segregation at the workplace and in the neighbourhood, as described in this book, happened in this specific instance, but it is not bound to happen. The decrease of interethnic social relations and the growing importance of social contact within ethnic boundaries did occur, but it does not necessarily occur at all times and in all places. These changes were the result of a specific historical process, which has played out differently across Western Europe, even though there are many similarities to be found as well. The ideas that the presence of diversity on its own can be the direct cause of people ‘hunkering down’, and that immigrants choose to be separated from others in the societies where they live, are attractive in their simplicity. This book has tried to deconstruct this simplicity by carefully reconstructing a specific historical reality and its evolution over time – an approach that has been advocated by Putnam himself in his famous 2007 article. This kind of historical analysis should allow us to draw lessons about what went wrong, how we can prevent such alienation from happening, and how we can foster mutual understanding within the highly diverse populations that characterize our societies today and will continue do so in the future.

Notes 1. http://regionalestatistieken.vlaanderen.be/sites/default/files/docs/LIIM-Gent.pdf, accessed 22 January 2017. In comparison, the data of the population register, counting only nonBelgians, only counted 7,633 Mediterranean citizens, or 3 per cent of the population. Data yearly reports of the city of Ghent. 2. Interview with Ahmed B. 3. Interview with Rabah. 4. Calculated from data in Archive Amsab-ISG, archive Maurice Maréchal, yearly report Neighbourhood Centre Brugse Poort, 1985–1986, pp. 4–9. 5. The highest concentrations were to be found in the Logstraat, the Stukwerkstraat and the Erasmusstraat, with 25 out of 29 or 86 per cent, 17 out of 22 or 77 per cent, and 76 out of 104 or 73 per cent non-Belgians respectively. 6. A further 45.53 per cent of the foreign population in the area lived together in another seventeen streets with concentrations of between 25 per cent and 50 per cent n ­ on-Belgians. Only 8.36 per cent of the non-Belgian population in the area lived in streets with between 10 per cent and 25 per cent non-Belgian inhabitants, and a mere 1.98 per cent lived in streets where less than 10 per cent of the inhabitants were foreign. De Poort-Beraber vzw, ‘Migranten in Gent’, 7–8.

Conclusion • 175

7. The numbers are 72.9 per cent (1st generation), 79.5 per cent (1.5th generation) and 76 per cent (2nd generation) for Turkish men (n=647), and 50 per cent (1st generation), 74.2 per cent (1.5th generation) and 71.6 per cent (2nd generation) for Turkish women (n=484) respectively. Data: Database partner choice of residents of Turkish and Moroccan origins in Ghent, Population census 1991, with special thanks to Professor John Lievens (UGent) for providing these data and calculating the percentages. 8. Database partner choice of residents of Turkish and Moroccan origins in Ghent, Population census 1991.

Quantitative Appendix 

In order to collect the bulk of the quantitative data for this book, I selected a sample from the individual immigrant files stored in the archives of the population administration of the city of Ghent, and supplemented it with entries from the register of arrival and the population register. The selection was based on nationality (Spanish, Italian, Turkish, Tunisian, Algerian, Moroccan) and on the first letter of the surname of the head of family, starting from A and continuing until a number that was deemed high enough to be quantitatively representative was reached: 115 heads-of-family for the Spanish, 150 for Italians, Tunisians and Algerians, 200 for Moroccans and 300 for Turks. This number was related to the size of the immigrant population in the city – the larger the population, the higher the number. Nonetheless, its exact size was an arbitrary decision. The only available published data are static, showing how many immigrants of a specific nationality were present in Ghent at a certain point in time, whereas the data I used pertained to arrivals, including those who stayed in the city for a very short period. Each file or entry in the register referred to a ‘family’. This could be a one-person household, but also a couple with seven children. While collecting the sample, I allocated a separate identification number to each individual, indicating whether their residential status was related to their professional activity (A), or dependent on that of their spouse (B), parent (C, D, E, etc.) or sometimes child (M, V). I also collected the data of Belgian or ­third-country spouses of immigrants (Z). This way, the sample was expanded

Quantitative Appendix • 177

to 131 entries under ‘Spanish’, 165 under ‘Italian’, 185 under ‘Tunisian’, 163 under ‘Algerian’, 284 under ‘Moroccan’ and 563 under ‘Turkish’. For each of these entries in our database, I collected as much information as possible: nationality, sex, date and place of birth, dates of arrival and departure, place they were arriving from and departing to, last known residence abroad, profession, employer, marital status at arrival and changes in marital status, last known number of children, naturalization, and a history of addresses in Ghent with dates of registration. In order to get a full overview of the immigration history of the city, I  did  not limit the sample to those immigrants who arrived during the period under study. The earliest arrivals in the sample date from 1925, the latest from 1994. For the actual quantitative research, only those immigrants  who arrived between 1960 and 1980 were included, and for most of it, only those who arrived as either independent (A) or spouse (B). This resulted in a working sample of 88 Spanish, 80 Italian, 145 Tunisian, 131 Algerian, 200 Moroccan and 381 Turkish individuals – 1,025 people in total. Of course, not all entries in our database provided all the information I wanted to collect. Sometimes, a profession was missing; often, there was no information as to the date or place of departure, etc. Because of these missing values, the number of individuals (n=) upon which graphs and calculations are based differs depending on the content of the query. Throughout the manuscript, I have chosen to present only the total number for each calculation, so as not to overburden the text. Here, the reader can find a more detailed overview of the sample sizes used.

Chapter 1 1) Socio-demographic data on Mediterranean arrivals in Ghent, 1963– 1967. Sample sizes n=70 (Algerian), n=38 (Italian), n=110 (Turkish), n=58 (Moroccan), n=44 (Spanish), n=31 (Tunisian) 2) Table 1: Percentage of independent immigrants taking up a first job in textiles, construction or domestic service, 1960–1980 (excluding people doing an internship or members of the clergy, but including professionals). Sample sizes n=93 (Algerian), n=60 (Italian), n=146 (Turkish), n=101 (Moroccan), n=73 (Spanish), n=94 (Tunisian) 3) Socio-demographic data on Mediterranean arrivals in Ghent, 1969– 1971. Sample sizes n=6 (Algerian), n=10 (Italian), n=60 (Turkish), n=50 (Moroccan), n=17 (Spanish), n=32 (Tunisian)

178 • Quantitative Appendix

4) Socio-demographic data on Mediterranean arrivals in Ghent, 1976– 1980. Sample sizes n=37 (Algerian), n=16 (Italian), n=57 (Turkish), n=39 (Moroccan), n=1 (Spanish), n=21 (Tunisian) 5) Age at arrival of independent immigrants in Ghent, 1960–1980. Sample sizes: n=120 (Algerian), n=74 (Italian), n=218 (Turkish), n=147 (Moroccan), n=80 (Spanish), n=121 (Tunisian) 6) Marital status of independent immigrants at arrival in Ghent, 1960–1980. Sample sizes: n=122 (Algerian), n=73 (Italian), n=220 (Turkish), n=149 (Moroccan), n=65 (Spanish), n=122 (Tunisian). To best represent the potential for partner-migration per nationality group, I counted those couples that both arrived as independent immigrants as one entry, but this only made a significant difference for the Spanish group. 7) Percentage of mixed marriages for immigrants who were unmarried at arrival, 1960–1980, and got married during their stay in Ghent. Sample sizes n=19 (Algerian), n=6 (Italian), n=20 (Turkish), n=30 (Moroccan), n=7 (Spanish), n=38 (Tunisian) 8) The mobility of independent and dependent immigrants arriving in Ghent, 1960–1980. Sample sizes n=131 (Algerian), n=80 (Italian), n=381 (Turkish), n=199 (Moroccan), n=88 (Spanish), n=145 (Tunisian). The date of arrival I used is the date on the front page of the file, which is the date of the inscription in the alien register; the date of departure is also the date on the front page of the file, and this is either the date the immigrant left (or had said he or she would leave), or, in case there was no declaration, the date the immigrant was officially cancelled from the alien register by the city administration – which could be quite some time after his or her actual departure. Still, I made use of this date as there was no other date available. Because of this, the period of stay of Mediterranean immigrants was probably even shorter than these statistics suggest. 9) Gender and status profile of short-term versus long-term Turkish immigrants, 1960–1980. Short-term immigrants (less than one year): n=100; long-term immigrants (more than five years): n=188; independent immigrants: n=221; dependent immigrants: n=160.

Chapter 2 1) Economic status (active or inactive) of independent and dependent immigrants in Ghent, 1960–1980. Sample sizes n=130 (Algerian), n=79 (Italian),

Quantitative Appendix • 179

n=368 (Turkish), n=192 (Moroccan), n=88 (Spanish), n=143 (Tunisian). Out of 251 immigrants initially enrolled as economically inactive, 14 per cent were students, while 81 per cent were partner-migrants, of whom 97 per cent were women. The economic status of the other 5 per cent was undetermined. 2) Immigrants enrolled as economically active and for which I had information as to their economic status, 1960–1980. Sample sizes n=115 (Algerian), n=68 (Italian), n=236 (Turkish), n=138 (Moroccan), n=79 (Spanish) and n=113 (Tunisian). 3) Occupational sector of labour migrants arriving in Ghent between 1974 and 1980, n=122, 47.5 per cent worked in textiles, 17 per cent in construction and 10.5 per cent in catering, whereas only 5 per cent were registered as starting work in domestic services. These numbers should however only be taken as an indication, as they do not differentiate between nationalities (the total numbers were too low for that). 8 out of 13 Italians who arrived in this period were employed in catering. 4) Skilled labourers among immigrants working in the building sector, 1963–1967. Of my sample of immigrant builders who were first enrolled as working for one of the large building companies in the Ghent Canal Zone (n=36), 72  per cent were classified as having worked as a skilled labourer abroad, before coming to Belgium. In comparison, of those immigrant builders, arriving in the same period, who did not arrive together with these companies and for whom I have information as to their previous employment (n=12), only 17 per cent could be classified as skilled labourers. 5) Period of stay among immigrants working in the building sector, 1963–1967. Of my sample of immigrant builders who were first enrolled as working for  one of the large building companies in the Ghent Canal Zone (n=36), 64 per cent stayed in the city for less than one year, 31 per cent stayed between  one and five years, and only 5 per cent stayed for longer than five  years. In comparison, of those immigrant builders, arriving in the same  period, who did not arrive together with these companies (n=15), only 33 per cent stayed in the city for less than one year, 27 per cent stayed between one and five years, and 40 per cent stayed for longer than five years.

180 • Quantitative Appendix

Chapter 4 1) Residential mobility of independent immigrants arriving in Ghent over the period 1960–1980, staying for a minimum of five years, during the first five years of their stay: n=134 (all nationalities). In order to calculate this mobility, only those immigrants whose individual files have been consulted have been taken into account. In the population register, the addresses were incomplete, only starting from the moment the immigrant was registered in the population register. On the one hand, this might have skewed the data in the sense that those immigrants who were immediately enrolled in the population register were probably less mobile than those that were enrolled in the alien register. On the other hand, the data also underestimate the residential mobility of the immigrants enrolled in the alien register, as they only represent those relocations that were officially registered at the city’s alien office. Both the unofficial transfers of registered immigrants as well as all mobility of unregistered, undocumented immigrants are not captured by these data. Furthermore, the data do not take into account those immigrants who stayed in the city for less than five years, who were probably much more mobile. 2) ‘Housing episodes’ (stay at one single address) of different length for independent Mediterranean immigrants arriving in Ghent, 1960–1980. Sample sizes: single immigrants (n=1018), immigrant families (n=262) and mixed families (n=91). Sometimes, when the family moved into the same house where the ‘single immigrant’ had been living, the calculated housing episode was cut short. The reason for this was that I calculated the time before the arrival of the family as one housing episode for a ‘single immigrant’ and the time after the arrival as one housing episode for the family, even though in reality, there was no moving house. Further, I only took into account those immigrants for whom I could consult their individual files, as the addresses were incomplete in the population register. 3) Total duration of independent immigrants’ stay at their first, second, third and fourth address, 1960–1980. Sample sizes: n=592 (first address), n=285 (second address), n=151 (third address) and n=116 (fourth address). Only those immigrants whose individual files have been consulted have been taken into account, as the addresses were incomplete in the population register. Remarkably enough, the percentages for stays between six months and one year remained relatively stable over the consecutive addresses. 4) Residential mobility of independent immigrants who arrived in Ghent, 1961–1965 (n=23), 1966–1970 (n=50), 1971–1975 (n=47) and 1976–1980

Quantitative Appendix • 181

(n=14) and stayed in Ghent for five years or more, and this during the first five years of their stay. In the data, the change in addresses for the period 1976–1980 might be under-reported as during that period, the paper files were gradually transferred to digital files, and we cannot be sure that the address changes were systematically added to the paper files as well. However, whilst working with the files, I did get the impression that most paper files had been kept up-to-date until the 1990s. 5) About the maps in chapter 4: a number of issues have to be taken into account when interpreting maps 3, 4 and 6. First of all, they project data from the period 1960–1980 on an OpenStreetMap of today. Streets that no longer exist were replaced with current streets approximating the location of the old ones, e.g. addresses in the ‘Kerkstraat’, a beluik in the neighbourhood Rabot that was demolished in the early 1970s, were changed to ‘Guldenvliesstraat’, a street that overlaps with part of what used to be the Kerkstraat. In these cases, but also for addresses in streets where house numbers have changed over the years, the location of the address on the map is not entirely correct. Still, this should not prove to be a major problem, as it does not impact much upon patterns of dispersion or concentration in general. Further, some remarks have to be made relating to the sampling method. As the archive from which the sample was taken was sorted alphabetically rather than chronologically, there was no way of knowing whether the sample size of each ‘nationality group’ (or, for that matter, of each regional group) was proportionate with that of the others. Therefore, even though I chose to represent all six nationalities together on each map, quantitatively, their spatial patterning is not directly comparable. Still, this is only a minor drawback, as the maps are not intended to measure percentages of residential concentration but to visualize the different trends and changes in the spatial patterning of immigrant households in the city during the period under study. In addition, the sampling only reflects the situation in the city of Ghent itself, as it existed before the merger of 1977. Therefore, the maps do not indicate whether there was a movement from the city towards the – often highly urbanized – villages surrounding it that came to be part of the city in that year. Finally, as I wanted to focus on immigrant housing in the city, and not on employers’ housing, I have excluded from the maps all data pertaining to housing in the house of employers, as well as all addresses indicating cabarets, restaurants and student homes.

Interviews 

Interviews carried out by the author (transcribed and in the possession of the author) Abdelham, Trahib, Habib and Ahmed L., first generation Tunisian immigrants, 2 September 2010, Nieuw Gent Abdeslam S., first generation Moroccan immigrant, worked for the Arabic section of the Christian Union, 14-29 November 2007, Willebroek Ahmed B., first generation Algerian immigrant, 29-30 April 2008, Ghent Ahmed T., first generation Moroccan immigrant, 2 January 2008, Ghent Ali Gh., first generation Tunisian immigrant, 13 October 2010, Ghent Amand, Belgian, former head of personnel Fabelta Ltd, 9 March 2011, Drongen Amar S., first generation Algerian immigrant, 27 May 2008, Ghent Amor and his wife, first generation Tunisian immigrants, 11 February 2008, Ghent Antonio, first generation Italian immigrant, 1 February 2011, Ghent Bader, first generation Tunisian immigrant, 14 April 2008, Ghent (interview not recorded) Carlos and Annie, 1.5 generation Spanish immigrant and his Belgian wife, 18 August 2010, Ghent Cecile, Belgian, former integration worker, 26 November 2007, Ghent Charif, first generation Algerian immigrant, 7 March 2008, Ghent Clelia, second/third generation Italian (born in Ghent), 15 February 2011, Ghent Dahmane and Jenny, first generation Algerian immigrant and his Belgian wife, 26 October 2009 and 10 November 2009, Wondelgem Elio and Erika, first generation Italian immigrant and his 1.5 generation Hungarian wife (she arrived in Belgium as a child), 17 March 2011, Destelbergen Fatma K., 1.5 generation Turkish immigrant, 22 March 2011, Ghent Fatma S., first generation Tunisian immigrant, 13 January 2011, Ghent

Interviews • 183

Filadelfio, first generation Italian immigrant, 14 September 2010, Ghent Francisco and Maria de la Luz, 1.5 generation Spanish immigrant and first generation Spanish immigrant, 15 March 2011, Ghent Giovanni, first generation Italian immigrant, 1 February 2011, Ghent Girolamo, first generation Italian immigrant, 17 August 2010, Ghent Italo, first generation Italian immigrant, 8 February 2011 (interview via Skype) José Luis and Jesusa, first generation Spanish immigrants, 6 September 2010, Ghent José, first generation Spanish immigrant, 27 August 2010, Evergem Juan, 1.5 generation Spanish immigrant, 26 August 2010, Ghent Kamel and Miloud, first generation Algerian immigrants, 30 October 2009, Ghent Kamel, first generation Algerian immigrant, 22 October 2009, Ghent Khalifa, first generation Tunisian immigrant, 8 July 2010, Brugge Lorenzo/Enzo and Christiane, first generation Italian immigrant and his Belgian wife, 31 July 2010, Ghent Lucien, Belgian married to a first generation Spanish immigrant, 23 July 2010, Ghent Marcel, Belgian, priest, 17 October and 16 November 2007, Ghent Maria del Pilar, first generation Spanish immigrant, 18 January 2011, Drongen Maria Isabel and Aurora, 1.5 generation Spanish immigrant and first generation Spanish immigrant, 23 February 2011, Ledeberg Mohamed E., first generation Moroccan immigrant, 12 November 2007, Ghent Mohamed T., first generation Moroccan immigrant, 9 February 2008, Ghent Naima, first generation Tunisian immigrant, 24 November 2010, Ghent Paul, Belgian, former employee UCO Ltd, 18 October 2007, Ghent Piergiorgio/Giorgio, first generation Italian immigrant, 9 August 2010, Mariakerke Rabah and Lieve, first generation Algerian immigrant and his Belgian partner, 29 July 2010, Ghent Recep, first generation Turkish immigrant, 28 March 2011, Ghent son of Ernest Gratz, Belgian, 26 August 2010, Ghent Taieb, first generation Moroccan immigrant, 21 July 2010, Ghent Tania, Belgian, grew up together with a Moroccan family, 14 April 2008, Lochristi Valère, Belgian, priest, 23 March 2011, Oostakker Willem, Belgian, former head of personnel UCO Ltd, 2 February 2011, Wondelgem Willy, Belgian, former employee UCO Ltd, 21 November 2007, Wondelgem

Informal conversations carried out by the author Adnan, second generation Turkish immigrant, 1 September 2010, Ghent Ahmed Bo., first generation Algerian immigrant, 9 July 2010, Ghent Ahmed R., first generation Tunisian immigrant, 8 March 2011, Ghent Albert, Belgian, former union delegate Socialist Union, 2 February 2011, Oostakker Artemio, first generation Spanish immigrant, 11 August 2010 son of ‘Berbat’, second generation Turkish immigrant, 26 March 2011, Ghent El Arras, second generation Moroccan immigrant, 15 July 2010, Ghent Lakhdar, first generation Algerian immigrant, 22 July 2010, Melle Mohammed X., first generation Moroccan immigrant, 12 July 2010, Ghent Mohammed Y., first generation Algerian immigrant, 2 July 2010, Ghent Omar, first generation Moroccan immigrant, 15 July 2010, Ghent Rachid, first generation Algerian immigrant, 8 July 2010, Ghent

184 • Interviews

Tahar, first generation Algerian immigrant, 10 October 2009, Ghent Vito, first generation Italian immigrant, 26 August 2010, Ghent shopkeeper Vits, Belgian, 2 July 2010, Ghent Jan Zwaenepoel, Belgian, former butcher ‘Zwaenepoel’, 24 July 2010

Transcripts from Interviews by Emma Tytgadt from E. Tytgadt, ‘Het migratieverhaal van de eerste generatie Marokkaanse mannen in Gent’, Masters thesis, University of Ghent, 2002 Abdelhakim, first generation Moroccan immigrant, from Oran (Algeria) Abdelkader A., first generation Moroccan immigrant, from Temsamane (Rif, Morocco) Abdelmajid, first generation Moroccan immigrant, from Mehdya-plage (Kenitra, Morocco) Abderrahman, first generation Moroccan immigrant, from Al Hoceima (Rif, Morocco) Ahmed S., first generation Moroccan immigrant, from Mjat (Sousse, Morocco) Anonymous man, first generation Moroccan immigrant, from Doukala (Al Jadida, Morocco) Anonymous man, first generation Moroccan immigrant, from Mont Arruit (Rif, Morocco) Elomar, first generation Moroccan immigrant, from Aklmi (Algeria) L’Arbi, first generation Moroccan immigrant, from Kenitra (Morocco) M’Hamed A., first generation Moroccan immigrant, from Tafersit (Rif, Morocco) Mohamed B., first generation Moroccan immigrant, from Casablanca (Morocco) Mohamed H., first generation Moroccan immigrant, from Casablanca (Morocco) Mohamed S., first generation Moroccan immigrant, from Al Hoceima (Rif, Morocco) and his wife Mohammed L., first generation Moroccan immigrant, from Timsitame (Rif, Morocco)

Interviews by Luc Vandezande from L. Vandezande, ‘De positie van de eerste generatie Turkse mannen in Gent’, Masters thesis, University of Ghent, 1994 Ali D., first generation Turkish immigrant, from Piribeyli (Konya, Turkey) Ali G., first generation Turkish immigrant, from Emirdağ (Turkey) Celyn C., first generation Turkish immigrant, from Emirdağ (Turkey) Faruk K., first generation Turkish immigrant, from Emirdağ (Turkey) Hakan P., first generation Turkish immigrant, from Emirdağ (Turkey) Hüseyim Y., first generation Turkish immigrant, from Emirdağ (Turkey) Mehmet K., first generation Turkish immigrant, from Rize (Turkey) Mehmet S., first generation Turkish immigrant, from Piribeyli (Konya, Turkey) Ömer K., first generation Turkish immigrant, from Piribeyli (Kony, Turkey) Yusuf E., first generation Turkish immigrant, from Piribeyli (Konya, Turkey)

Interviews • 185

Interviews by Robin Mertens from R. Mertens, ‘Minder-heden of beter verleden? Turkse migratie in eigen beeldvorming. Het herinneringslandschap van een lokale gemeenschap te Gent’, Masters thesis, University of Ghent, 2000 Adem and Halise, first generation Turkish immigrant, from Stip (former Yugoslavia), later Istanbul Aydin, first generation Turkish immigrant, from Stip (former Yugoslavia), later Istanbul Bayran (probably Abdilkadir), first generation Turkish immigrant, from Emirdağ (Turkey) Cekkör, first generation Turkish immigrant, from Turkmeinhim (Turkey) Harm (Harun), first generation Turkish immigrant, from Ordu (Turkey) Johan de Vreese, Belgian, former integration worker Kamil, first generation Turkish immigrant, from Bursa (Emirdağ, Turkey) Kazim, first generation Turkish immigrant, from Istanbul (Turkey) Mehmet B., first generation Turkish immigrant, from Emirdağ (Turkey) Mehmet Ko., first generation Turkish immigrant, from Emirdağ (Turkey) Yakup, first generation Turkish immigrant, from Istanbul (Turkey)

Interviews by Robin Roeck from R. Roeck, ‘Psycho-sociale problematiek rond de immigratie van buitenlandse arbeiders naar België. Een exploratief onderzoek bij de Turkse gemeenschap in Gent’, Masters thesis, University of Ghent, 1971 Ahmed X., first generation Turkish immigrant, from Emirdağ (Turkey) Duz, first generation Turkish immigrant, from Nïgde (Turkey) Hallil (probably Halil), first generation Turkish immigrant, from Emirdağ (Turkey) Hassan, first generation Turkish immigrant, from Emirdağ (Turkey) Ismael X., first generation Turkish immigrant, from Emirdağ (Turkey) Mustafa (2), first generation Turkish immigrant, from Emirdağ (Turkey) Mustafa X., first generation Turkish immigrant, from Korria (Turkey) Ramazan, first generation Turkish immigrant, from Emirdağ (Turkey) Soleiman (probably the same as Süleyman A.), first generation Turkish immigrant, from Emirdağ (Turkey)

Interviews from Freek Neirynck, Sylvain Van Labeke and Ayse Isçi (ed.), Herinneringen/Hatıralar, Ghent: Provincie Oost-Vlaanderen, 2002 Celal and his wife, first generation Turkish immigrants, from Emirdağ (Turkey) Cemil and his wife (Animzade), first generation Turkish immigrants, from Posof (Kars, Turkey) Feyyaz, first generation Turkish immigrant, from Binbaşi Eminbey (Turkey) Gazi and his wife, first generation Turkish immigrants, from Emirdağ (Turkey) Ismail Ç. and his wife, first generation Turkish immigrants, from Emirdağ (Turkey) Mehmet Ali, first generation Turkish immigrant, from Denizli (Turkey) Ömer and his wife, first generation Turkish immigrants, from former Yugoslavia, later Istanbul Süleyman A. and his wife (Hatice), first generation Turkish immigrants, from Emirdağ (Turkey)

186 • Interviews

Interviews carried out in the framework of the project ‘Gentse Gasten’ (Nakhla vzw, 2007), stored by Amsab-ISG, www.amsab.be/gentsegasten Dsouhdia, first generation Moroccan immigrant, from Ksar el Kebir (Morocco) Fatima, first generation Moroccan immigrant, from Fahs (Tangier, Morocco) Hadda, first generation Moroccan immigrant, from Nador (Rif, Morocco) Hassenia, first generation Moroccan immigrant, from Béchar (Oran, Algeria) Khadija, first generation Moroccan immigrant, from Casablanca (Morocco) Mimount, first generation Moroccan immigrant, from Tafersit (Rif, Morocco) Moh, first generation Moroccan immigrant, from Nador (Rif, Morocco) Mohamed J., first generation Moroccan immigrant, from Oudzem (Morocco) Mohamed Le., first generation Moroccan immigrant, from Taza (Morocco) Moumna, first generation Moroccan immigrant, from Beni Sidel (Rif, Morocco) Yamina, first generation Moroccan immigrant, from Hammam – Bou Hadjar Zineb, first generation Moroccan immigrant, from Beni Sidel (Morocco) Zohra, 1.5 generation Moroccan immigrant, from Tangier (Morocco)

Interviews (excerpts from transcripts in Dutch) by Tina De Gendt from T. De Gendt, Turkije aan de Leie: Vijftig jaar Turkse migratie in Gent, Tielt: Lannoo, 2014 Faruk, first generation Turkish immigrant, from Emirdağ (Turkey) Fatma, 1.5 generation Turkish immigrant, from Istanbul (Turkey) Jozefien D., Belgian, worked for the Union Cotonnière from 1951 to 1964 Meral, 1.5 generation Turkish immigrant, from Karaagac (Emirdağ, Turkey) Nazim, first generation Turkish immigrant, from Posof (Turkey) Osman, first generation Turkish immigrant, from Isparta (Turkey) Rükiye, first generation Turkish immigrant, from Emirdağ (Turkey)

Bibliography

Aerts, M. 1979. ‘Dossier migrantenvrouw en arbeidsmarkt’, Bareel: Tijdschrift voor wie ten dienste staat van buitenlanders 2(7): 9–17. Aerts, M., and A. Martens. 1978. Gastarbeider: Lotgenoot en landgenoot. Leuven: Kritak. Allaert, Georges. 1994. ‘Ghent: The Western Node of the Metropolitan Network of Flanders’, in P. Kooij and P. Pellenbarg (eds), Regional Capitals: Past, Present, Prospects. Ghent, Groningen, Münster, Norwich, Odense, Rennes. Assen: Van Gorcum, pp. 19–35. Anderson, Bridget. 1999. ‘Overseas Domestic Workers in the European Union: Invisible Women’, in J.H. Momsen (ed.), Gender, Migration and Domestic Service. New York: Routledge, pp. 117–33. Anselin, A. 1979. L’émigration antillaise en France: Du ghetto au bantoustan. Paris: Anthropos. Bade, Klaus J. 2003. Migration in European History. The Making of Europe. Malden, MA: Blackwell Pub. Balthazar, H. (ed.). 1978. ‘Onderzoek naar de Gentse beluiken: Bouwfysische, Sociologische, Historische en Kunsthistorische Evaluatie’. W.I.A.R.U.G. - Seminarie voor Nieuwste Geschiedenis. Baumann, Gerd. 1996. Contesting Culture: Discourses of Identity in Multi-Ethnic London. Cambridge Studies in Social and Cultural Anthropology 100. Cambridge: Cambridge University Press. Berg-Eldering, Lotty van den. 1978. Marokkaanse gezinnen in Nederland. Alphen aan de Rijn: Samsom Uitgeverij. Bernabé, J. 1968. ‘Immigratieproblemen: Een sociografische studie van Turkse arbeiders te Gent’. Licentiaatsverhandeling sociale wetenschappen, Ghent: RUGent. Beyer de Ryke, Luc. 2002. Les Lys de Flandre: Vie et Mort des Francophones de Flandre (13022002). Combats Pour La Liberté de L’esprit. Paris: Guibert. Beyers, Leen. 2007a. Iedereen zwart: het samenleven van nieuwkomers en gevestigden in de mijncité Zwartberg, 1930–1990. Amsterdam: Aksant. ———. 2007b. ‘Unfolding Urban Memories and Ethnic Identities: Narratives of Ethnic Diversity in Limburg, Belgium’, in R. Rodger and J. Herbert (eds), Testimonies of the City: Identity, Community and Change in a Contemporary Urban World. Historical Urban Studies. Aldershot: Ashgate Publishing Limited, pp, 119–37. ———. 2008. ‘From Class to Culture: Immigration, Recession, and Daily Ethnic Boundaries in Belgium, 1940s–1990s’. International Review of Social History 53: 37–61. Blanc-Chaléard, Marie-Claude. 2006. ‘Old and New Migrants in France: Italians and Algerians’, in L. Lucassen, D. Feldman and J. Oltmer (eds), Paths of Integration: Migrants in Western Europe (1880–2004). IMISCOE Research. Amsterdam: Amsterdam University Press, pp. 46–62.

188 • Bibliography

Blokland, Talja. 2003a. Urban Bonds: Social Relationships in an Inner City Neighbourhood. Cambridge: Polity Press. ———. 2003b. Urban Bonds: Social Relationships in an Inner City Neighbourhood. Translated by Lee K. Mitzman. Cambridge: Polity Press. Boissevain, Jeremy, Jochen Blaschke, Hanneke Grotenberg, Isaac Joseph, Ivan Hubert Light, Marlene Sway, Roger David Waldinger and Pnina Werbner. 1990. ‘Ethnic Entrepreneurs and Ethnic Strategies’, in R. Waldinger, H. Aldrich and R. Ward (eds), Ethnic Entrepreneurs: Immigrant Business in Industrial Societies. Newbury Park: Sage Publications Ltd, pp. ­131–56. Bollen, R. 1978. Explorerend onderzoek naar de rechtspositie van de buitenlandse werknemers inzake sociale zekerheid. Leuven: Instituut voor Sociaal Zekerheidsrecht. Bolt, Gideon, Sule Özüekren and Deborah Phillips. 2010. ‘Linking Integration and Residential Segregation’, Journal of Ethnic and Migration Studies 36(2): 169–86. Bouras, Nadia. 2012. Het Land van Herkomst, Perspectieven op verbondenheid met Marokko, 1960-2010. Hilversum: Uitgeverij Verloren. Bovenkerk, F., K. Bruin, L. Brunt and H. Wouters. 1985. Vreemd volk, gemengde gevoelens: Etnische verhoudingen in een grote stad. Meppel/Amsterdam: Boom. Bovenkerk, Frank. 1999. ‘Rassenrellen of Criminaliteit?’, Migrantenstudies 15(4): 255–70. Bowes, Alison, Naira Dar and Duncan Sim. 2000. ‘Housing Preferences and Strategies: An Exploration of Pakistani Experiences in Glasgow’, in F.W. Boal (ed.), Ethnicity and Housing: Accomodating Differences. London: Ashgate, pp. 170–178. Brubaker, Rogers. 2004. Ethnicity Without Groups. Cambridge, MA: Harvard University Press. Buyens, Frans. 1972. Over Gastarbeiders. Documentary. Byron, Margaret and Stéphanie Condon. 2004. Migration in Comparative Perspective: Caribbean Communities in Britain and France. Routledge Research in Population and Migration. London: Routledge. Caestecker, Frank (ed.). 2005. Huwelijksmigratie, Een zaak voor de overheid? Minderheden in de Samenleving, nr. 17. Leuven: Acco. Caestecker, Frank, John Lievens, Bart Van de Putte and Koen Van der Bracht. 2013. ‘Trends in partnerkeuze van eerste en tweedegeneratiemigranten in België: Partnerkeuze van personen woonachtig in België met een migratieachtergrond, 2001–2008, met bijzondere aandacht voor de bevolkingsgroep van Marokkaanse en Turkse herkomst in België’. Ghent: Ghent University, Department of Sociology. Caestecker, Frank, N. Perrin, T. Eggerickx and B. Renaud. 2016. Belg Worden: De Geschiedenis van de Belgische Nationaliteitsverwerving Sinds 1830. Mechelen: Wolters Kluwer. Caestecker, Frank, Filip Strubbe and Pierre-Alain Tallier. 2009. ‘De Individuele Vreemdelingendossiers Afkomstig van de Openbare Veiligheid (Vreemdelingenpolitie) 1935–1943’. Zoekwijzer 15. Available at https://biblio.ugent.be/publication/888110/f​i​l​e​ /​9​1​9​7​5​8​. Caestecker, Frank and Eric Vanhaute. 2011. ‘Leven en werken in het industriële België, 1850–1975’, in A. Grauwels, D. Luyten, F. Caestecker, P. Scholliers and E. Vanhaute (eds), Hedendaagse economische geschiedenis van België: een inleiding. Ghent: Academia Press, pp. 81–113. Callier Boisvert, Colette. 1987. ‘Working-Class Portuguese Families in a French Provincial Town: Adaptive Strategies’, in H. C. Buechler; J. M. Buechler (eds), Migrants in Europe: The Role of Family, Labor, and Politics. New York, Westport, CT, London: Greenwood Press, pp. 61–76. Cammaert, M.F. 1985. Migranten en thuisblijvers: een confrontatie: De leefwereld van Marokkaanse Berbervrouwen. Leuven: Universitaire Pers.

Bibliography • 189

Cantle, Ted (ed.). 2001. ‘Community Cohesion: A Report of the Independent Review Team’. Home Office. Castles, Stephen and Godula Kosack. 1985. Immigrant Workers and Class Structure in Western Europe. 2nd edn. Oxford: Oxford University Press. Charlton, Thomas L., Lois E. Myers and Rebecca Sharpless (eds). 2008. Thinking About Oral History: Theories and Applications. Lanham, MD: AltaMira Press. Claeys, Danny and Erik Van Hove. 1987. Sociale structuuratlas van de stad Gent. LeuvenAmersfoort: Acco. Clark, W. and A. Drever. 2000. ‘Residential Mobility in a Constrained Housing Market:  Implications for Ethnic Population in Germany’, Environment and Planning 32(5): 833–46. Clark, William A.V. 2012. ‘Residential Mobility and the Housing Market’, in D.F. Clapham, W. A. V. Clark and K. Gibb (eds), The SAGE Handbook of Housing Studies. London: SAGE, pp. 66–83. Cottaar, Annemarie, Nadia Bouras and Fatiha Laouikili. 2009. Marokkanen in Nederland: De pioniers vertellen. Amsterdam: J.M. Meulenhoff. Cox, Rosie. 1999. ‘The Role of Ethnicity in Shaping the Domestic Sector in Britain’, in J.H. Momsen (ed.), Gender, Migration and Domestic Service. New York: Routledge, pp. 134–47. Cumoli, Flavia. 2012. Un tetto a chi lavora: mondi operai e migrazioni italiane nell’Europa degli anni Cinquanta. Milan: Guerini e associati. De Bock, Jozefien. 2012. ‘“Alle Wegen Leiden Naar Gent”: Trajecten van Mediterrane Migranten Naar de Arteveldestad, 1960-1980’, Brood & Rozen 3: 47–75. ———. 2013. ‘“We Have Made Our Whole Lives Here”: Immigration, Settlement and Integration Processes of Mediterranean Immigrants in Ghent, 1960–1980’. PhD dissertation, Florence: European University Institute. ———. 2014. ‘“Guest Workers”, Entrepreneurs and the Role of Ethnicity in Immigrant Business: A Critical History of “Ethnic Entrepreneurship” in Ghent, 1960–1980’, Journal of Belgian History XLIV(4): 36–66. ———. 2015. ‘Not All the Same after All? Superdiversity as a Lens for the Study of Past Migrations’, Ethnic & Racial Studies 38(4): 583–95. ———. 2017. ‘Of Employers, Uncles and Interpreters: The Diverse Trajectories of Guest Workers to the Belgian City of Ghent, 1960-1975’, Journal of Ethnic and Migration Studies (published online), 10 April 2017. De Bruyn, Jean. 1966. ‘Een onderzoek naar de adaptatie van de gastarbeiders, in het textielbedrijf N.V. U.C.O. Rooigem te Gent’. Verhandeling hoger onderwijs, Arbeidershogeschool, Hoger Instituut voor Maatschappelijk Dienstbetoon. De Decker, Pascal. 2001. ‘Jammed between Housing and Property Rights: Belgian Private Renting in Perspective’, European Journal of Housing Policy 1: 17–39. De Gendt, Tina. 2014. Turkije Aan de Leie: 50 Jaar Migratie in Gent. Tielt: Lannoo Uitgeverij. De Keyzer, Diane. 1995. Madame est servie: Leven in dienst van adel en burgerij. Antwerp: Van Halewyck. De Kinder, Roger. 1972. ‘De economische situatie in Oost-Vlaanderen en een van haar facetten, de gastarbeiders: Rede naar aanleiding van de opening van de gewone zitting van de provinciale raad uitgesproken op 2 oktober 1972 door de heer Roger De Kinder, Gouverneur van de provincie’. Ghent: Provinciale raad van Oost-Vlaanderen. ———. 1977. ‘De textielnijverheid, proefterrein van de internationale werkloosheidsverdeling: Rede uitgesproken door de heer Roger De Kinder, Gouverneur van de provincie, naar aanleiding van de opening van de gewone zitting van de provinciale raad op 1 oktober 1977’. Ghent: Provincieraad van Oost-Vlaanderen.

190 • Bibliography

De Muynck, Marc. 1972. ‘De integratie van de gastarbeider in zijn werkmilieu, een rendementsonderzoek’. Licentiaatsverhandeling psychologische en pedagogische wetenschappen. Ghent: Universiteit Gent. De Wilde, Bart. 1997. Witte boorden, blauwe kielen: patroons en arbeiders in de Belgische textielnijverheid in de 19e en 20e eeuw. Ghent: Ludion - Amsab - Profortex. ———. 2007. Gent/Rabot: De teloorgang van de textielnijverheid. Ghent: Gent Cultuurstad vzw. De Witte, E. 1997. ‘Ruimtelijk-economische analyse van etnisch ondernemerschap in Gent: Met toepassing op de Turkse gemeenschap’. Masters thesis, Leuven: KULeuven. Debbaut, H.L., P. De Mey, N. Hostyn, J. Mestdagh and D. Rygole. s.d. Onderzoek naar bewoners en bewoning in het gebied gelegen tussen Sluizeken en Muidebrug in de stad Gent. Ghent: Rijksuniversiteit, Seminarie Toegepaste Economie. Den Exter, J. and E. Kutlu. 1993. ‘Emirdag: Over de effecten van migratie op een Turks district’, Migrantenstudies 9(4): 23–33. Deslé, Els. 1990. ‘Bouwen en wonen te Brussel (1945-1958): De moeizame uitbouw van de keynesiaanse welvaartsstaat en de rol van de mediterrane gastarbeiders’, Belgisch Tijdschrift voor Nieuwste Geschiedenis 21(3–4): 413–82. ———. 1995. ‘Racism and the “Foreign Labour System”: An Exploration of Their Mutual Relationship Based on the Experience of Belgium in the Immediate Post-War Period’, in M. van der Linden and J. Lucassen (eds), Racism and the Labour Market: Historical Studies. Berne: Peter Lang, pp. 537–61. Djait, Faiza, Dave Boussé and Wim Herremans. 2011. ‘De arbeidsmarktsituatie van migranten en hun nakomelingen in vlaams en europees perspectief ’. WSE-rapport 2011–6. Leuven: Steunpunt Werk en Sociale Economie. Duc, Marcelle. 2000. Le travail en chantier. Toulouse: Octares. Dumont, M.E. 1951. Gent, een stedenaardrijkskundige studie. Bruges: De Tempel. Erdem, Esra and Monika Mattes. 2003. ‘Gendered Policies – Gendered Patterns: Female Labour Migration from Turkey to Germany from the 1960s to the 1990s’, in K. Schönwälder, R. Ohliger and T. Triadafilopoulos (eds), European Encounters: Migrants, Migration and European Societies since 1945. Aldershot: Ashgate, pp. 167–85. Estlund, Cynthia. 2005. ‘Working Together: Crossing Color Lines at Work’, Labor History 46(1): 79–98. Fenton, Steve. 2010. Ethnicity. Second edn. Cambridge: Polity. Fernandez Asperilla, Ana. 2004. ‘Endoctrinement ou formation? Les émigrantes espagnoles, le travail et la qualification professionnelle vers le milieu du XXe siècle’, in A. Morelli and E.  Gubin (eds), Pour une histoire européenne des femmes migrantes: Sources et méthodes. Sextant, revue du Groupe interdisciplinaire d’Études sur les Femmes 21–22. Brussels: Universite Libre de Bruxelles, pp. 185–220. ‘Fotobijlage “Gent 1980 Goddomme!”’. 1980. Gramschap, no. 11. Gabaccia, Donna R. 1984. From Sicily to Elizabeth Street: Housing and Social Change Among Italian Immigrants, 1880-1930. SUNY Series in American Social History. Albany, NY: State University of New York Press. Gesthuizen, Maurice, Tom van der Meer and Peer Scheepers. 2009. ‘Ethnic Diversity and Social Capital in Europe: Tests of Putnam’s Thesis in European Countries’, Scandinavian Political Studies 32(2): 121–42. Gijsberts, Mérove, Tom van der Meer and Jaco Dagevos. 2012. ‘“Hunkering Down” in Multi-Ethnic Neighbourhoods? The Effects of Ethnic Diversity on Dimensions of Social Cohesion’, European Sociological Review 28(4): 527–37. Glebe, Günther. 1997. ‘Housing and Segregation of Turks in Germany’, in R. van Kempen and S. Özüekren (eds), Turks in European Cities: Housing and Urban Segregation. Comparative

Bibliography • 191

Studies in Migration and Ethnic Relations 4. Utrecht: European Research Centre on Migration and Ethnic Relations, Utrecht University, pp. 122–57. Glick Schiller, Nina, Ayşe Çağlar and Thaddeus C. Guldbrandsen. 2006. ‘Beyond the Ethnic Lens: Locality, Globality, and Born-Again Incorporation’, American Ethnologist 33(4): 612–33. Glorieux, Ignace, Ilse Laurijssen and Yolis Van Dorsselaer. 2009. Zwart op wit – de intrede van allochtonen op de arbeidsmarkt. Antwerpen: Garant. González Ferrer, Amparo. 2006. ‘Who Do Immigrants Marry? Partner Choice Among Single Immigrants in Germany’, European Sociological Review 22(2): 171–85. Granovetter, Mark S. 1995. Getting a Job: A Study of Contacts and Careers. 2nd edn. Chicago, IL: University of Chicago Press. Green, Nancy L. 1997a. Ready-to-Wear and Ready-to-Work: A Century of Industry and Immigrants in Paris and New York. Comparative and International Working-Class History. Durham, NC: Duke University Press. ———. 1997b. ‘The Comparative Method and Poststructural Structuralism: New Perspectives for Migration Studies’, in J. Lucassen and L. Lucassen (eds), Migration, Migration History, History: Old Paradigms And New Perspectives. Rev. 2nd edn. International and Comparative Social History 4. Bern: Peter Lang, pp. 57–72. Hackett, Sarah. 2013. Foreigners, Minorities, Integration: The Muslim Immigrant Experience in Britain & Germany. Manchester: Manchester University Press. Haex, Jos, Albert Martens and Sylvain Wolf. 1976. Arbeidsmarkt, discriminatie, gastarbeid. Leuven: KULeuven. Halm, Dirk and Steffen Sauer. 2006. ‘Parallelgesellschaft Und Etnische Schichtung’, Aus Politik Und Zeitgeschichte 1–2: 18–24. Harzig, Christiane. 2007. ‘Domestics of the World (Unite?): Labor Migration Systems and Personal Trajectories of Household Workers in Historical and Global Perspective’, in Immigration, Incorporation & Transnationalism. New Brunswick: Transaction Publishers, pp. 39–62. Heijke, Hans. 1995. Economische aspecten van migratie uit het Middellandse–Zeegebied. ’s Hertogenbosch: KPC. Heitmeyer, Wilhelm. 1996. ‘Für Türkische Jugendliche in Deutschland Spielt Der Islam Eine Wichtige Rolle’, Die Zeit, 23 August. Herbert, Joanna. 2008. Negotiating Boundaries in the City: Migration, Ethnicity, and Gender in Britain. Aldershot: Ashgate Publishing Limited. Hilton, W.S. 1968. Industrial Relations in Construction. London: Pergamon Press. Hollifield, James Frank. 1992. Immigrants, Markets, and States: The Political Economy of Postwar Europe. Cambridge, MA: Harvard University Press. Hubeau, Bernard. 1994. Het huisvestingsrecht in gewestelijk, gemeenschaps- en federaal perspectief. 2 vols. Bruges: Die Keure. Janssens, J. 1973. ‘Huisvesting van de gastarbeiders woonachtig in Oost–Vlaanderen’. Ghent: Dienst voor sociale aangelegenheden, Provinciebestuur Oost–Vlaanderen. Kesteloot, Christian. 1987. ‘The Residential Location of Immigrant Workers in Belgian Cities: An Ethnic or a Socio-Economic Phenomenon?’, in G. Glebe and J. O’Loughlin (eds), Foreign Minorities in Continental European Cities. Erdkundliches Wissen, Heft 84. Stuttgart: Steiner, pp. 223–39. Kesteloot, Christian, Pascal De Decker and Altay Manço. 1997. ‘Turks and Housing in Belgium, with Special Reference to Brussels, Ghent and Visé’, in R. van Kempen and S. Özüekren (eds), Turks in European Cities: Housing and Urban Segregation. Comparative Studies in Migration and Ethnic Relations 4. Utrecht: European Research Centre on Migration and Ethnic Relations, Utrecht University, pp. 67–97.

192 • Bibliography

Kesteloot, Christian and Henk Meert. 2000. ‘Segregation and Economic Integration of Immigrants in Brussels’, in Minorities in European Cities: The Dynamics of Social Integration and Social Exclusion at the Neighbourhood Level. Migration, Minorities, and Citizenship. Basingstoke, Hampshire and New York: Macmillan, St. Martin’s Press, pp. 54–72. Khoojinian, Mazyar. 2006. ‘L’acceuil et la stabilisation des travailleurs immigrés turcs en Belgique, 1963-1980’, Cahiers d’Histoire du Temps présent 17: 73–116. ———. 2014. ‘L’immigration, une main-d’oeuvre d’appoint temporaire? Marché du travail, politiques étatiques et trajectoires des travailleurs turcs recrutés pour l’industrie charbonnière belge (1956–1980)’. PhD thesis, Brussels: Université Libre de Bruxelles. Kloosterman, Robert and Jan Rath (eds). 2003. Immigrant Entrepreneurs: Venturing Abroad in the Age of Globalization. Oxford: Berg Publishers. König, Mareike. 2004. ‘Itinéraires de domestiques allemandes à paris vers 1900: sources, méthodes et interprétations’, in A. Morelli and E. Gubin (eds), Pour Une Histoire Européenne Des Femmes Migrantes: Sources et Méthodes. Sextant, Revue Du Groupe Interdisciplinaire d’Études Sur Les Femmes 21–22. Brussels: Universite Libre de Bruxelles, pp. 83–115. Körner, Heiko. 1984. ‘Return Migration from the Federal Republic of Germany’, in D. Kubat (ed.), The Politics of Return: International Return Migration in Europe. Rome and New York: Centro Studi Emigrazione, Center for Migration Studies, pp. 175–86. Lambrechts, E. 1979. Vrouwenarbeid in België: Het tewerkstellingsbeleid inzake vrouwelijke arbeidskrachten: 1930–1972. Centrum voor Bevolkings-en Gezinsstudiën, Ministerie van Volksgezondheid en van het Gezin, Brussels. Lebon, André. 1984. ‘Return Migration from France: Policies and Data’, in D. Kubat (ed.), The Politics of Return: International Return Migration in Europe. Rome and New York: Centro Studi Emigrazione, Center for Migration Studies, pp. 153–69. Leibold, Jürgen, Steffen Kühnel and Wilhelm Heitmeyer. 2006. ‘Abschottung von Muslimen Durch Generalisierte Islamkritik?’, Aus Politik Und Zeitgeschichte 1–2: 3–10. Leitner, Helga. 1987. ‘Regulating Migrants’ Lives: The Dialect of Migrant Labour and the Contradictions of Regulatory and Integration Policies in the Federal Republic of Germany’, in G. Glebe and J. O’Loughlin (eds), Foreign Minorities in Continental European Cities. Erdkundliches Wissen, Heft 84. Stuttgart: Steiner, pp. 71–89. Lekens, Luc and Jacques Lagrou. 1996. Gentse Wijken van Vroeger: Deel Twee: De Muide. Nazareth: De Klaroen. Lesthaeghe, Ron (ed.). 2000. Communities and Generations: Turkish and Moroccan Populations in Belgium. Brussels: VUB University Press. Lievens, John. 1999. ‘Family-Forming Migration from Turkey and Morocco to Belgium: The Demand for Marriage Partners from the Countries of Origin’, International Migration Review 33(3): 717–44. Lucassen, Leo. 2005. The Immigrant Threat: The Integration of Old and New Migrants in Western Europe Since 1850. Studies of World Migrations. Urbana, IL: University of Illinois Press. Lucassen, Leo, David Feldman and Jochen Oltmer (eds.). 2006. Paths of Integration: Migrants in Western Europe (1880–2004). IMISCOE Research. Amsterdam: Amsterdam University Press. Lucassen, Leo and Charlotte Laarman. 2009. ‘Immigration, Intermarriage and the Changing Face of Europe in the Post War Period’, The History of the Family 14(1): 52–68. Lucassen, Leo and Jan Lucassen. 2011. Winnaars en verliezers: een nuchtere balans van vijfhonderd jaar immigratie. Amsterdam: Bert Bakker. Mampuys, Jozef. 1991. ‘Hoofdstuk III: De Christelijke Vakbeweging’, in E. Gerard (ed.), De Christelijke Arbeidersbeweging in België, 1891–1991. Deel 2. Leuven: Universitaire Pers, pp. 146–271.

Bibliography • 193

Martens, A. and S. Wolf. 1974. ‘Buitenlandse werknemers op de huisvestingsmarkt’. Leuven: Katholieke Universiteit Leuven – Sociologisch Onderzoeksinstituut. Martens, Albert. 1976. Les immigrés: Flux et reflux d’une main d’oeuvre d’appoint. Louvain: Presses Universitaire de Louvain – editions vie ouvrière. Martini, Manuela. 2008. ‘Entre étrangers: Transmission du métier et échanges au travail dans le BTP (fin XIXe-première moitié du XXe siècle)’, in P. González-Bernaldo, M. Martini and M.-L. Pelus-Kaplan (eds), Étrangers et sociétés: représentations, coexistences, interactions dans la longue durée. Rennes: Presses universitaires de Rennes, pp. 373–84. ———. 2009. ‘Médecins du travail et travailleurs immigrés dans le Bâtiment en France (1946–1975)’, Revue Historique 651(July): 595–619. Martiniello, Marco and Jan Rath (eds). 2010. Selected Studies in International Migration and Immigrant Incorporation. Amsterdam: Amsterdam University Press. Martinovic, B. 2013. ‘The Interethnic Contacts of Immigrants and Natives in the Netherlands: A Two-Sided Perspective’, Journal of Ethnic and Migration Studies 39(1): 69–85. Messina, Anthony W. 2007. The Logics and Politics of Post-WWII Migration to Western Europe. Cambridge: Cambridge University Press. Mitchell, James Clyde. 1969. Social Networks in Urban Situations: Analyses of Personal Relationships in Central African Towns. Manchester: Manchester University Press. Momsen, Janet Henshall. 1999. Gender, Migration and Domestic Service. New York: Routledge. Moulaert, Frank (ed.). 1981. ‘Dossier Economische Crisis en Immigratie’, Bareel: Tijdschrift voor wie ten dienste staat van buitenlanders 4(14): 7–19. Moulaert, Frank and Albert Martens. 1982. ‘Arbeidsproces, sectorale dynamiek en gastarbeid in België (1970-1977)’, in J. M. M. van Amersfoort and H. B. Entzinger (eds), Immigrant en samenleving. Mens en maatschappij, 57 (1982). Deventer: Van Loghum Slaterus, pp. 77–98. Musterd, Sako, and Rinus Deurloo. 2002. ‘Unstable Immigrant Concentrations in Amsterdam: Spatial Segregation and Integration of Newcomers’, Housing Studies 17(3): 487–503. Narula, Rekha. 1999. ‘Cinderella Need Not Apply: A Study of Paid Domestic Work in Paris’, in J.H. Momsen (ed.), Gender, Migration and Domestic Service. New York: Routledge, pp. 148–63. Neels, Karel and Reinhard Stoop. 2000. ‘Reassessing the Ethnic Gap: Employment of Younger  Turks and Moroccans in Belgium’, in R.J. Lesthaege (ed.), Communities and Generations: Turkish and Moroccan Populations in Belgium. Brussels: VUB University Press, pp. 279–319. Ogbonna, Emmanuel and Lloyd C. Harris. 2006. ‘The Dynamics of Employee Relationships in an Ethnically Diverse Workforce’, Human Relations 59: 379–407. O’Hanlon, Seamus. 2005. ‘Full Board and Lodging: Hostels for Migrant Workers in Early Postwar Melbourne’, History Australia 2(3): 88.1–88.15. O’Loughlin, John V. 1987. ‘Introduction: Foreign Minorities in Continental European Cities’, in G. Glebe and J. O’Loughlin (eds), Foreign Minorities in Continental European Cities. Erdkundliches Wissen, Heft 84. Stuttgart: Steiner, pp. 9–29. Oltmer, Jochen, Axel Kreienbrink and Carlos Sanz Diaz (eds). 2012. Das ‘Gastarbeiter’System: Arbeitsmigration und ihre Folgen in der Bundesrepublik Deutschland und Westeuropa. Munich: Oldenbourg Verlag. Onthaalcomité voor Gastarbeiders van Brabant. 1977. ‘Werkloosheid en Gastarbeid’: Nationaal Centrum voor Ontwikkelingssamenwerking. Oosterlynck, Stijn and Pascal Debruyne. 2010. ‘De strijd voor een stad op mensenmaat in Gent: Het Pandinistisch Verblijvingsfront en de herwaardering van de stad als woonomgeving’, Brood & Rozen 15(3): 24–41.

194 • Bibliography

Osaer, Antoon (ed.). 1991. ‘Hoofdstuk V: De Christelijke Arbeidersvrouwenbeweging’, in E. Gerard (ed.), De Christelijke Arbeidersbeweging in België, 1891–1991. Deel 2. Leuven: Universitaire Pers, pp. 316–411. Ostergren, Robert C. 1979. ‘A Community Transplanted: The Formative Experience of a Swedish Immigrant Community in the Upper Middle West’, Journal of Historical Geography 5(2): 189–212. Oswald, Anne von, Karen Schönwälder and Barbara Sonnenberger. 2003. ‘Einwanderungsland Deutschland: A New Look at Its Post-War History’, in K. Schönwälder,  R. Ohliger and  T.  Triadafilopoulos (eds), European Encounters: Migrants, Migration and European Societies since 1945. Aldershot: Ashgate, pp. 19–37. Ouali, Nouria. 2004. ‘Quarante ans de présence marocaine en Belgique’, in N. Ouali (ed.), Trajectoires et dynamiques migratoires de l’immigration marocaine en Belgique. Carrefours 4. Louvain-la-Neuve: Academia–Bruylant, pp. 19–61. Özüekren, Sule and Ebru Ergoz-Karahan. 2010. ‘Housing Experiences of Turkish (Im) migrants in Berlin and Istanbul: Internal Differentiation and Segregation’, Journal of Ethnic and Migration Studies 36(2): 355–72. Özüekren, Sule and Ronald van Kempen. 2003. ‘Special Issue Editors’ Introduction: Dynamics and Diversity: Housing Careers and Segregation of Minority Ethnic Groups’, Housing, Theory and Society 20(12): 162–71. Pasleau, Suzy and Isabelle Schopp. 2001. ‘La Domesticité En Belgique de 1947 À L’aube Du 21e Siècle: Un Essai de Définition’, Sextant - Revue Du Groupe Interdisciplinaire d’Etudes Sur Les Femmes - Domesticité 15–16: 235–277. Pauwels, L. 1986. Recht en plicht: 100 jaar christelijk syndicalisme in de textiel 1886–1986. Leuven: KADOC. Pee, Kathleen. 2001. ‘Pensando en volver: migratiebeleving- en geschiedenis van Spaanse migranten in Gent sinds 1930’. Licentiaatsverhandeling geschiedenis, UGent. Penninx, Rinus and Judith Roosblad. 2000. ‘Introduction’, in R. Penninx and J. Roosblad (eds), Trade Unions, Immigration, and Immigrants in Europe, 1960–1993: A Comparative  Study of the Attitudes and Actions of Trade Unions in Seven West European Countries. International Studies in Social History, vol. 1. New York: Berghahn Books, pp. 1–19. Perks, Robert and Alistair Thomson (eds.). 1998. The Oral History Reader. Ebrary. London and New York: Routledge. Phillips, D. 2006. ‘Parallel Lives? Challenging Discourses of British Muslim Self-Segregation’, Environment and Planning D: Society and Space 24(1): 25–40. Piore, Michael J. 1979. Birds of Passage: Migrant Labor and Industrial Societies. Cambridge: Cambridge University Press. Pitti, Laure. 2005. ‘Catégorisations ethniques au travail: Un instrument de gestion différenciée de la main-d’oeuvre’, Histoire & Mesure 20(3/4): 69–101. Pratsinakis, Manolis, Panos Hatziprokopiou, Lois Labrianidis and Nikos Vogiatzis. 2017. ‘Living Together in Multi-Ethnic Cities: People of Migrant Background, Their Interethnic Friendships and the Neighbourhood’, Urban Studies 54(1): 102–118. Putnam, Robert D. 2007. ‘E Pluribus Unum: Diversity and Community in the TwentyFirst Century The 2006 Johan Skytte Prize Lecture’, Scandinavian Political Studies 30(2): 137–74. Rasking, Johan. 2013. ‘Doe niet alsof alle werknemers gelijk zijn’, De Standaard, 29 October. Ratcliffe, Peter. 2009. ‘Re-Evaluating the Links between “Race” and Residence’, Housing Studies 24(4): 433–50.

Bibliography • 195

Ray, Brian and Valerie Preston. 2009. ‘Are Immigrants Socially Isolated? An Assessment of Neighbors and Neighboring in Canadian Cities’, International Migration & Integration 10: 217–44. Rea, Andrea. 2000. ‘Immigration, État et Citoyenneté: La formation de la politique d’intégration des immigrés de la belgique’. Thèse de doctorat en sociologie, Brussels: Université Libre de Bruxelles. Reinsch, Peter. 2001. Measuring Immigrant Integration: Diversity in a European City. Aldershot: Ashgate. Reniers, Georges. 1997. ‘On the Selectivity and Internal Dynamics of Labour Migration Processes: A Cross-cultural Analysis of Turkish and Moroccan Migration to Belgium’, IPD Working Paper no. 1997-7, p. 25. Robinson, Vaughan. 1999. ‘Asians in Britain: A Study in Encapsulation and Marginality’, in S. Vertovec (ed.), Migration and Social Cohesion. Cheltenham and Northampton: Edward Elgar Publishing Ltd, pp. 411–37. Ródenas Calatayud, Carmen. 1994. Emigración y economía en España, 1960–1990. 1st edn. Biblioteca Civitas economía y empresa 11. [Alicante]: Universidad de Alicante; Madrid: Editorial Civitas. Roeck, R. 1971. ‘Psycho-sociale problematiek rond de immigratie van buitenlandse arbeiders naar België: Een exploratief onderzoek bij de Turkse gemeenschap in Gent’. Licentiaatsverhandeling psychologie en pedagogische wetenschappen, UGent. Romero, Federico. 1991. Emigrazione e integrazione europea, 1945–1973. Studi di storia 20. Rome: Edizioni Lavoro. Rosvelds, S., A. Martens, Y. Ben Abdeljelil and P. Arryn. 1993. Zelfde zweet, ander brood: Onderzoek naar de arbeidsmarktpositie van Belgen en migranten op twee lokale arbeidsmarkten: Antwerpen en Gent. Diensten voor programmatie van het wetenschapsbeleid. Programma maatschappelijk onderzoek. 2: Migrantenstudies. Brussels: Diensten voor programmatie van het wetenschapsbeleid. Ruble, Blair A, Lisa M. Hanley and Allison M. Garland. 2008. ‘Introduction: Renegotiating the City’, in L.M. Hanley, B.A. Ruble and A. Garland (eds), Immigration and Integration in Urban Communities: Renegotiating the City. Washington, DC: Woodrow Wilson Center Press, pp. 1–16. Safi, Mirna. 2010. ‘Patterns of Immigrant Intermarriage in France: Intergenerational Marital Assimilation?’, Zeitschrift Für Familienforschung 22(1): 89–108. Salah, Ali. 1973. La communauté algérienne dans le département du Nord. Lille: Editions Universitaires. Saltman, Juliet. 1991. ‘Theoretical Orientation: Residential Segregation’, in E. Huttman, J. Saltman and W. Blauw (eds), Urban Housing Segregation of Minorities in Western Europe and the United States. Durham, NC: Duke University Press, pp. 1–17. Schmit, Fernand. 1972. ‘De invloed van de immigratie op het familieleven: Een sociografische studie van Turkse arbeiders in Gent’. Licentiaatsverhandeling sociale wetenschappen, Ghent: RUGent. Sleegers, F. and F. Vermeulen. 2007. ‘Mengen in de schaduw: de werkvloer’, in L. Veldboer, J. W. Duyvendak and C Bouw (eds), De mixfactor: integratie of segregatie in Nederland. Amsterdam: Boom, pp. 117–26. Smets, Peer and Marion den Uyl. 2008. ‘The Complex Role of Ethnicity in Urban Mixing: A Study of Two Deprived Neighbourhoods in Amsterdam’, Urban Studies 45(7): 1439–60. Smit, Aniek. 2011. ‘Mijn vader had een afro!’ hoe marokkaanse migranten in nederland zich kleden sinds de jaren zestig. Amsterdam: Amsterdam University Press.

196 • Bibliography

Sontz, Ann H.L. 1987. ‘Factory and Community in a West German Immigrant Neighborhood’, in H.C. Buechler and J.M. Buechler (eds) Migrants in Europe: The Role of Family, Labor, and Politics. New York, Westport, CT, London: Greenwood Press, pp. 189–206. Sürig, Inken and Maren Wilmes. 2015. The Integration of the Second Generation in Germany: Results of the TIES Survey on the Descendants of Turkish and Yugoslavian Migrants. IMISCOE Research. Amsterdam: Amsterdam University Press. Surkyn, Johan. 1993. ‘Migratiegeschiedenis en regionale herkomstverschillen bij Turkse vrouwen in Vlaanderen en Brussel’. Working Papers Etnische Minderheden in België 3. Tinnemans, W. 1994. Een gouden armband: Een geschiedenis van Mediterrane immigranten in Nederland (1945–1994). Utrecht: Nederlands Centrum Buitenlanders. Tonkin, Elizabeth. 1992. Narrating Our Pasts: The Social Construction of Oral History. Cambridge Studies in Oral and Literate Culture 22. Cambridge: Cambridge University Press. Van Kempen, Ronald. 1997. ‘Turks in the Netherlands: Housing Conditions and Segregation in a Developed Welfare State’, in R. van Kempen and S. Özüekren (eds), Turks in European Cities: Housing and Urban Segregation. Comparative Studies in Migration and Ethnic Relations 4. Utrecht: European Research Centre on Migration and Ethnic Relations, Utrecht University, pp. 158–90. Vanneste, D. 1981. ‘Sociaal-ruimtelijke patronen in een stad; betekenis van de indicator “gastarbeiders”: een onderzoek van de Gentse situatie’, Tijdschrift van de Belgische Vereniging voor Aardrijkskundige Studies, 161–74. ———. 1985. Gent, een geografische gids: sociale en economische structuren in stad en haven. Acta Geografica Lovaniensia 25. Leuven: Instituut voor Sociale en Economische Geografie. Vanoutryve, Sonia. 1968. ‘Immigratieproblemen: een sociografische studie van Spaanse werknemers te Gent’. Ghent: Universiteit Gent. BIB. Digital copy on hard drive. Verhaeghe, Pieter Paul. 2010. ‘De discursieve kracht van het begrip “allochtoon”’. Poliargus, 18 February. Verhaeghe, Pieter-Paul, Koen Van der Bracht and Bart Van de Putte. 2012. Migrant zkt toekomst: gent op een keerpunt tussen oude en nieuwe migratie. Antwerp: Garant. Verhoeven, H. 2001. ‘Arbeid en werkloosheid’, in J. Vranken and K. van der Heyden (eds), Komende generaties: wat weten we (niet) over allochtonen in Vlaanderen? Minderheden in de Samenleving 10. Leuven: Acco, pp. 225–36. Verhoeven, H. and A. Martens. 2000. De vreemde eend in de bijt: arbeidsmarkt en diversiteit. Leuven: Steunpunt Werkgelegenheid, Arbeid en Vorming. Verschueren, Herwig. 1995. ‘De ontwikkelingen in de Belgische nationaliteitswetgeving belicht vanuit het Europese en Internationale recht’, in K. De Feyter, M.-C. Foblets and B. Hubeau (eds), Migratie- en migrantenrecht: recente ontwikkelingen (deel i). Bruges: Die Keure, pp. 240–63. Vertovec, Steven. 2007. ‘Super-Diversity and Its Implications’, Ethnic and Racial Studies 30(6): 1024–54. Vervoort, Miranda. 2012. ‘Ethnic Concentration in the Neighbourhood and Ethnic Minorities: Social Integration Weak and Strong Social Ties Examined’, Urban Studies 49(4): 897–915. Vervoort, Miranda, Henk Flap and Jaco Dagevos. 2011. ‘The Ethnic Composition of the Neighbourhood and Ethnic Minorities’ Social Contacts: Three Unresolved Issues’, European Sociological Review 27(5): 586–605. Villanova, Christian de. 1997. ‘Turkish Housing Conditions in France: From Tenant to Owner’, in R. van Kempen and S. Özüekren (eds), Turks in European Cities: Housing and Urban Segregation. Comparative Studies in Migration and Ethnic Relations 4. Utrecht:

Bibliography • 197

European Research Centre on Migration and Ethnic Relations, Utrecht University, pp. 96–121. Villey, Olivier. 1981. ‘Le redéploiement actuel de la main-d’oeuvre étrangère passé le premier choc de la crise’, Travail et Emploi 8: 47–55. Vogel, Jaap. 2005. Nabije vreemden: een eeuw wonen en samenleven. Cultuur en migratie in Nederland, d. 4. The Hague: Sdu Uitgevers. Voordeckers, Annie. 1966. De Arbeidsmarkt in Oost-Vlaanderen 1947–1971. Ghent: Economische Raad voor Oost-Vlaanderen. Waldinger, Roger David, Howard E. Aldrich and Robin Ward (eds). 1990a. Ethnic Entrepreneurs: Immigrant Business in Industrial Societies. Newbury Park, CA: Sage Publications Ltd. ———. 1990b. ‘Opportunities, Group Characteristics, and Strategies’, in R. Waldinger, H.  Aldrich and R. Ward (eds), Ethnic Entrepreneurs: Immigrant Business in Industrial Societies. Newbury Park, CA: Sage Publications Ltd, pp. 13–48. ‘Wanneer Een Migrantenbeleid? Beleidsvoorstellen van V.z.w. De Buurt, De Sleep, Woonfonds Uit de Wijk Sluizeken-Muide Te Gent.’ 1982. Watson, C.J. 1971. Social Housing Policy in Belgium. Occasional Papers: University of Birmingham, Centre for Urban and Regional Studies: 19. Birmingham: University of Birmingham. Werkgroep Huisvesting Gent. 1984. ‘Huis te huur: Huisvestingsproblemen van bepaalde bevolkingsgroepen in het Gentse’. Woonfonds vzw. White, Paul. 1984. The West European City: A Social Geography. London and New York: Longman. Wimmer, Andreas. 2004. ‘Does Ethnicity Matter? Everyday Group Formation in Three Swiss Immigrant Neighbourhoods’, Ethnic and Racial Studies 27(1): 1–36. Wise, A. and S. Velayutham. 2009. Everyday Multiculturalism. Houndmills: Palgrave Macmillan. Yücel, A. Ersan. 1987. ‘Turkish Migrant Workers in the Federal Republic of Germany: A Case Study’, in H.C. Buechler; J.M. Buechler (eds), Migrants in Europe: The Role of Family, Labor, and Politics. Contributions in Family Studies, no. 12. New York: Greenwood Press, pp. 117–48.

Index

Algerian immigrants, 24–25, 31–35, 176–77 Ahmed B. on UCO Ltd, 53, 98n29, 168 Amar S. on workplace relations, 81 Charif, 48, 81, 108 Dahmane, 80, 104, 158n1 employment of, 42, 44, 76 in France, 2, 15n1, 25, 33, 34–35, 108, 132 history of, 24–25 housing of, 103, 108, 113, 132 Kamel, 40n45, 104, 106, 108, 129n58–59, 132, 149, 160n35 in lodging houses, 102, 104, 132–33 Miloud, 55, 108, 127n27, 160n35 mixed marriages, 32–34 Rabah, 168 Rachid, 64 spatial dispersion of, 117, 123, 125 textile industry and, 28, 49 allochtonous, 167–68, 171 attachments. See social relations, categories beluik (workmen’s houses in closes) 21–23, 106–7, 144–45 demolition of, 21, 22 immigrants in, 106, 107, 113–14, 122–23, 123, 138, 144 pictures of, 22, 145 renovation of, 22–23 residential concentration and, 116, 119, 164 segregation at street-level and, 121–22 UCO Ltd and, 103

Blokland, Talja, 10, 131 social relations, categories of, 10, 77, 87, 89, 95, 133–34, 142–43, 144–48, 156–57, 165–67 Bonds. See social relations, categories Brugse Poort, neighbourhood, xiv, 117, 119, 122, 129n55, 149, 155, 169 building sector. See construction sector Castles, Stephen and Godula Kosack, 3–4 chain migration, 27–8. See also transplanted community charity, neighbour relations inspired by, 133–34, 135–36, 141 children of immigrants, 24, 31,167 cross-ethnic neighbour relations and, 135–36, 137–38, 142, 143, 147, 148–50, 152, 159n33, 166 economic inactivity and, 42, 66n3 diminished mobility and, 114–5 pictures of, 22, 136, 145, 149 city flight, 21–24, 119. See also suburban flight clergymen, 44, 66n6. See also religious institutions concentration, neighbourhoods, 6, 116–17,118, 119, 120, 125, 141, 164–65 without segregation, 123, 125 concentration, workplaces, 46, 62–64, 65, 163 construction sector, 17–19, 20, 22, 29, 45, 46, 54–56. See also Ghent Canal Zone economic crisis and, 19, 20, 22

Index • 199

immigrant labour in, 27–28, 29, 30, 55, 62, 64, 82, 104 mobility in, 34, 48–49, 53, 54–55 spatial and temporal characteristics of, 54–56 crisis. See economic crisis; oil crisis discrimination and racism colleagues and, 74–77 crisis and, 15, 151, 167 employers and, 83–84, 95, 98n28 immigrants dealing with, 64, 74–75, 76–77, 111–12, 153–54 on the housing market, 107–10, 111,124, 132, 164, 165 on the labour market, 42, 47–48 neighbours and, 135, 151–52, 153, 157 social housing and, 110–11, 124, 126n29 unions and, 75–76 women and, 42, 66n3 domestic services, 18 economic crisis and, 29 housing experiences in, 100, 105 immigrant labour in, 27–28, 29, 45 men in, 67n13 mobility in, 34 neighbour relations in, 134–35 Spanish immigrants and, 27–28, 31, 90–92, 105 spatial and temporal characteristics of, 59–60 women in, 18, 29, 31, 39n36, 43, 45, 54 workplace relations in, 60, 72, 90–94 dual housing market, 23–24 primary sector of the, 106, 110 residential patterning and, 116–17 secondary sector of the, 24, 105–7 dual labour market, 18–20, 44 primary sector of the, 18, 19, 48, 62 secondary sector of the, 18, 19, 20, 44, 163–64, 168 economic crisis, 20. See also oil crisis construction and, 18, 19, 20, 22–23 immigration (policy) and, 11, 27, 28–29, 30, 39n31, 43, 115 mobility and, 51, 115

social relations and, 76, 96n6, 151, 164, 167, 171 textile industry and, 20 unemployment and, 18, 19, 20, 51, 52–53, 65 economic growth, 11, 19–20, 27, 29, 30 immigration (policy) and, 11, 20, 27, 28, 163 Emirdağ, 26, 38n27, 109, 127n27 employers, co-ethnic, 46–47, 163. See also ethnic entrepreneurs; ethnic workplace exploitation and, 89–90 housing and, 88–89, 104–5, 181 Italian, 47, 64, 87–90, 104–5 recruitment by, 47, 63–64 relations with employees, 88–90, 94, 165 Spanish, 47, 64, 88 employers, of domestic workers housing and, 105, 181 recruitment by, 27–28 relations with employees, 60, 92–94, 165 paternalism and, 92, 94 employers and supervisors, cross-ethnic, 47, 86, 163. See also UCO Ltd Amand on Tunisian workers in Fabelta, 84 discrimination and racism and, 83–84, 95, 98n28 gifts for, 85, 98n32 housing and, 101–4, 121, 124, 126n3, 132–33 paternalism of, 84, 95 Paul on gifts from immigrant workers, 84–85, 98n32 recruitment by, 28, 46, 47–48, 51, 51–52, 63–64 relations with employees, 56, 58–59, 83–86, 95 Willem on discrimination at UCO, 98n28 Willy on gifts from immigrant workers, 85, 98n32 ethnic enclave, 169, 171–72 ethnic entrepreneurs, 44, 47, 64, 66n8, 87, 91, 101, 169, 171 ethnic groupism, (Brubaker), 9

200 • Index

ethnic neighbourhood. See neighbourhood, ethnic; Sluizeken-Muide ethnic workplace. See workplace, ethnic ethnicity, 6, 9 everyday multiculturalism, 7 Fabelta, 46, 49, 56, 57, 69n54, 76, 81, 84, 104, 118, 133, 168 families, 35, 36 housing and, 23, 24, 107–8, 109–10, 113 mobility and, 35, 51, 114–15 reunification of, 29, 30, 32–33, 41, 51 social relations with neighbours and colleagues and, 81, 135–38 female. See women first housing, 100, 101–5, 116, 117, 124. See also beluik (workmen’s houses) for dependent partner migrants, 101, 113–14 as employer home, 59, 88, 104, 105 lodging houses as, 101–2, 104–5, 116, 121, 124, 125, 129n58, 132, 164 gender, 18, 31–32, 41–44, 58, 60, 92, 100, 131–32, 170. See also women Ghent. See also specific topics case study, choice of, 11 ethnic diversity in, today, 24, 167 map of, xiv migration history to, 24–26 Ghent Canal Zone, 17, 20, 27, 29, 46, 121 Italian immigration and, 28, 47, 55, 104 Golden Sixties, 19, 21, 27, 37n7 Green, Nancy, 8, 9 guest workers. See labour migration home ownership, 22, 112, 124 housing, 6–7, 21–24, 36, 100–30, 164–65, 169–70. See also beluik (workmen’s houses); ethnic neighbourhood; first housing; home ownership; housing associations; housing market, secondary sector of; mobility, residential; social housing conditions of, 21–24, 102, 103–04, 105–7, 107, 113–14

employers and, 47, 88–89, 101–05, 121, 124, 126n3, 132–33, 181 ethnic workplaces and, 87, 88–89, 104–05, 181 protest for better, 111 suburban flight and, 21, 25, 36, 164 temporary, 55, 101–5, 124, 132–35, 164 UCO Ltd and, 102–4, 118, 124, 126n6, 129n59 women and conditions of, 113–14 housing associations, 37n13, 107, 110–11, 128n34, 164. See also Woonfonds Gent housing market, secondary sector of, 23–24, 105, 116, 164. See also beluik (workmen’s houses) and nineteenthcentury belt discrimination and, 107–10, 164 families and, 107 individual immigrants in, 105–7 replacement dwellers and, 24, 105–6, 164 residual rental market and, 124 hunkering down, (Putnam), 2–3, 174 identificational integration, 4–5 illegal immigration, migration policy and, 28–29, 30, 38n29, 39n30, 39n31 immigration marriages, 170, 172 integration, 4–5, 6–9, 11 defining, 4–5 interethnic families. See mixed marriages interdependencies. See social relations, categories intermarriage. See mixed marriages Italian immigrants, 5, 13, 24–25, 31–35, 42, 104, 106–7, 176–77 co-ethnic employers and, 47, 64, 88–90, 104–5 employment of, 28, 44, 47, 49, 53, 55 Enzo, 68n38, 70n74, 88–89, 90, 98n38, 105, 134 ethnic workplace relations for, 87 Girolamo on working in Italian restaurants, 87, 88, 99n50, 105 housing of, 106–7 migration stop and, 30 pictures of, 33, 87 spatial dispersion of, 116, 117, 123, 125

Index • 201

labour. See also construction sector; domestic services; employers; mobility, professional; social relations, workplaces; textile industry; unemployment; women, in labour force dependent partner migrants and, 41–42, 43 entrepreneurship and, 44,169 short-term careers and, 48–49, 54, 65 labour market, secondary sector of, 1, 17–20, 35–36, 44 gender and, 18, 45 immigrant labour and, 19–20 replacement workers and, 6–7, 44 second generation immigrants and, 168 target earners in, 19–20, 49–50, 79 unemployment and, 20 labour migration, postwar, 7, 9, 10 economic boom and, 20, 27, 51–52 in the literature, 3, 4, 8 legislation, 10–11, 28–29, 30 to Ghent, 25–30, 36 work permits and, 43, 50, 67n28 language, 165–166 courses and teachers, 44, 66n6 French in Ghent, 34, 40n45, 80–81, 166 interpreters and, 85–86, 150, 166 interviews and, 13 mixed marriages and, 34 neighbourhood relations and, 139, 140, 141–42, 143, 146, 150, 156, 166 unemployment and, 82–83, 95 workplace relations and, 47, 62–63, 79–81, 82, 83, 85, 86, 92–93, 94–95, 166 lodging houses, 101–2, 104–5, 116, 121, 124, 125, 129n58, 132, 164 conditions of, 102 in France, 105 social relations in, 132–33, 156, 157n1 Lucassen, Leo, x–xi, 5, 9 integration defined by, 4–5, 6 marriages, 34. See also mixed marriages immigration, 170, 171, 172 migration age and, 32–33

marital status, immigrants at arrival, 32–33 Mediterranean immigrants, definition, 9, 10 men. See also gender labour and, 18, 43, 45, 49, 58, 87 neighbourhood relations and, 100, 131–32 ratio of arrivals and immigration stock, 27, 29, 30, 31–32 short-term dwellers as single, 35 unemployment and, 53–54, 82–83, 95 workplace relations and, 81, 131–32 migration stop, 30, 51, 52 mixed marriages, 5–6, 32–34, 36, 114, 137–38, 170, 172–73 second generation, 33–34, 170–73 mobility, professional, 48–52, 65, 163 construction sector and, 34, 46, 54–55 decrease of, 50–52 domestic services and, 34, 105 employers and, 103 permits and, 50, 67n28 reasons for, 49–50 social relations and, 79 textile industry and, 48, 49 mobility, residential, 112–15, 125, 128n39, 141, 156, 164 decrease of, 114–15, 125 in between cities, 13–14, 34–35 reasons for, 113–14 social relations and, 115, 138, 141, 156 Moroccan immigrants, xii, 10, 24–26, 30, 31–35, 36, 158n21, 176–77. See also van den Berg-Eldering Abdelkader on neighbourhood relations, 142 Ahmed M. on neighborhoods relations, 142, 143 Elomar on neighbourhood relations, 142 employment of, 28, 42, 44, 49, 53, 76 Fatima on her relation with a Belgian neighbour, 148, 160n47 Hadda on neighbourhood relations with Moroccans and Belgians, 140, 146, 159n26 Hassenia on neighbourhood relations with Moroccans and Belgians, 138, 141–42, 147–48, 149–50

202 • Index

Moroccan immigrants (cont.) housing of, 113 Khadija on her relation with a Belgian neighbour, 146 Moh, 106, 135, 159n29, 160n47 Mohamed E., 96n4, 144–45 Mohamed H. on racism at the workplace, 76–77 Mohamed J. on praying at the workplace, 74–75 Mohamed T. on his relation with a Belgian neighbour, 151, 160n46 Moumna on co-ethnic neighbours, 139 pictures of, 136 second generation, 168, 172–73 spatial dispersion of, 116, 117, 119, 123, 125 Yamina, 114, 136–37 Zineb on neighbourhood relations with Belgians and Moroccans, 137, 139–40 Zohra, 135–36, 155, 160n42, 160n45 multi-stranded relations, 87, 88, 94–95, 133, 156, 165. See also single-stranded relations naturalization, 5, 25, 167 Belgian nationality legislation, 5, 38n25, 167 neighbour relations. See social relations, neighbourhood neighbourhood, ethnic, 6, 11, 108, 122, 123, 154–55, 157, 165, 166–67. See also segregation, neighbourhood; social relations, neighbourhood, co-ethnic and cross-ethnic; Sluizeken-Muide Brugse Poort as, xiv, 117, 119, 122, 129n55, 149, 155, 169 city flight and, 119, 120, 165 locals’ reactions to, 122–23, 152–53, 157 nineteenth-century belt and, 117, 119–20, 123, 125, 151, 155, 164 quota systems and, 123–24, 157 Sluizeken-Muide as, xiv, 101, 117, 119, 120, 122, 123, 152–53, 155, 157, 165, 169 social control and, 155, 157, 166–67

Turkish immigrants and, 119, 120, 122, 123, 165 nineteenth-century belt, 21, 56, 106, 117, 119–20, 123, 125, 150–51, 155, 164, 169 definition, 118 oil crisis. See also economic crisis economic crisis and, 20, 30 unemployment and, 51 Parallelgesellschaft. See parallel lives parallel lives, 1–2, 145, 162, 168, 169–70, 173–74 class and, 2 origins of term, 1 political and public debate and, 1–2, 5, 6, 8, 123, 168, 170, 171 parallel societies. See parallel lives partner migrants, dependent, 32, 41, 43, 65, 170, 172 housing for, 101, 113 labour of, 41–42 ratio of, 27, 29, 30, 32–33 paternalism, 84, 92, 94, 95, 133, 137 Patershol, neighbourhood, 118–19, 120, 123 poststructural structuralism, 8–9 Putnam, Robert, 2–3, 173–74 hunkering down-thesis of, 2 racism. See discrimination and racism regularization, 28–29, 30, 157n1 Reinsch, Peter, 3, 61 religious institutions, 54, 66n6, 79, 87, 88, 91, 94, 120, 155 replacement dwellers, 24, 105–6, 164 replacement workers, 6–7, 44 second generation entrepreneurship of, 169, 171 labour market segregation and, 168, 171 partner choice of, 33–34, 170, 172–73 residential segregation and, 169, 171–72 unemployment of, 168–69, 171 secondary sector of the housing market. See housing market, secondary sector of

Index • 203

secondary sector of the labour market. See labour market, secondary sector of segregation, 1, 5, 6, 162 segregation, workplaces, 3–4, 61–64, 163–64, 168, 171. See also concentration, workplaces ethnic concentrations and, 63–64, 163 ethnic workplaces and, 64, 65 mixed workplaces and, 61–64, 65 segregation, neighbourhoods, 1–2, 120–24, 165, 169, 171–72. See also concentration, neighbourhoods; neighbourhood, ethnic age groups and, 120 beluiken and, 121–22, 123 city council debate on, 122–23 ethnic concentration and, 123, 125, 141 lodging houses and, 121, 164 quotas and, 123–24 Turkish immigrants and, 120, 122, 165, 172 short-term careers, 48, 52–54, 65 short-term dwellers, 13, 34–35, 104–5, 114–15 SIDMAR steel factory, 17, 18, 55, 104 single-stranded relations, 87, 85. See also multi-stranded relations definition, 98n40 Sluizeken-Muide, neighbourhood, xiv, 101, 117, 119, 120, 122, 123, 152–53, 155, 157, 165, 169 social housing, 22, 23 discrimination and, 111, 124, 164 immigrants and, 110–11, 124, 126n29, 128n34 social relations, categories (Talja Blokland), 10 attachments, 10, 77, 79, 87, 89, 95,133,144–48 bonds, 10, 77, 87, 89, 95, 133, 144–48 interdependencies, 10, 95, 142, 144, 156 transactional relations, 10, 77, 95, 133–34, 143, 166 social relations, co-ethnic, in the neighbourhood, 14, 139–41, 154–55, 156, 157, 165, 167 changing, 154–55, 156

language and, 139–40 in lodging houses, 132–33, 156 social control and, 155 women and, 131–32, 139–41, 158n21, 159n23 social relations, cross-ethnic, in the neighbourhood, 2, 133, 141–50, 150–54, 156, 157n1,158n10, 166, 172 barriers to, 143–44 changes over time, 135, 150–54, 157, 167 charity and curiosity and, 133–34, 136–37, 156 children and, 135, 143, 147, 148–50 cultural differences and, 137, 143–44, 160n35 difficulties in, 132, 137, 138, 149–50, 151–53, 157 discrimination and racism and, 151–53, 157 language and, 141–42, 143, 146, 150, 156, 166 women and, 135, 138, 146–48, 150, 157 social relations in domestic services, 72, 90–94 colleagues and, 60, 90–92 discrimination and racism and, 92 employers and, 60, 92–94 language and, 92–93 social relations in ethnic workplaces, 72, 86–90 bonds as, 87, 89 co-ethic colleagues and, 86–87 employers and, 88–90 exploitation in, 89–90 housing and, 87, 88 language and, 86 multi-stranded relations as, 87, 88 social relations in mixed workplaces, 6–7, 73, 94, 95, 166 co-ethnic colleagues, 77–79, 80 conflicts in, 73–75, 76 cross-ethnic colleagues, 76, 77, 78, 79, 81 cultural and religious habits and, 74–75 discrimination and racism, 74, 75, 76–77, 83 employers and, 83–86

204 • Index

social relations in mixed workplaces (cont.) interpreters and, 85–86 language and, 62–3, 79–81, 82–83, 85–86, 166 mobility and, 79 at UCO Ltd, 74, 80, 81, 82 unemployment and, 82–83 union discrimination and, 74, 75–76 social relations, language and, 34, 62–63, 79–81, 82–83, 85–86, 92–93, 94–95 social relations, neighbourhood. See also, social relations, co-ethnic, in the neighbourhood; social relations, crossethnic, in the neighbourhood Blokland framework for, 10, 131, 133, 138, 142–43, 144 domestic servants and, 134–35 employers and, 134 men and, 131–32 women and, 131–32, 138, 146–48, 166 social relations, workplace, 72, 165–66. See also, social relations in domestic services; social relations in ethnic workplaces; social relations in mixed workplaces Blokland, framework for, 10, 77, 87, 95 unemployment and, 82–83, 164, 166–67 source material, 11–14 interviews as, 12–13, 182 population archive as, 12, 176–77 Spanish immigrants, 24–25, 30, 31–35, 90–91, 176–77 Aurora on meetings of Spanish domestics, 91 domestic services and, 27–28, 31, 90–92, 105 employment of, 28, 42, 44, 47, 49, 53, 87, 105 housing of, 93, 114 José Luis and Jesusa, 59, 135 Maria de la Luz, 69n57, 143 Maria del Pilar, 91, 93, 114 pictures of, 91 Spanish employers and, 47, 64 spatial dispersion of, 116, 117, 122, 123, 125 women as, 27, 31–32

spatial and temporal characteristics of construction sector, 54–56 of domestic services, 59–60 of textile industry, 56–59 suburban flight, 21, 25, 36, 119, 164, 165 housing surplus and, 21, 119 immigrant population percentages and, 25 super-diversity city, 24 super-diversity-paradigm, 9 target-earners, 19–20, 49–50, 79 housing and, 105, 106–7, 124 mobility and, 49, 51 workplace relations and, 79, 84, 95 textile industry, 10, 17, 18, 57. See also UCO Ltd economic downfall of, x, 18, 20, 28, 29, 48, 168 gender roles in, 18, 63, 45, 57, 69n53 housing and, 102, 103 immigrant labour in, 20, 27–28, 29, 46, 48, 57, 61–62, 63, 66n9, 82, 163–64 mobility in, 48, 49, 52 segregation in, 61–62, 163–64, 168 social relationships with colleagues in, 58, 63, 74, 76, 80, 81, 82 spatial and temporal characteristics of, 56–59 supervisor and worker relationships in, 58–59 unions in, 74 transactional relations. See social relations, categories transplanted community, 26, 34. See also Emirdağ Turkish immigration and, 26 Tunisian immigrants, 28, 30, 31–35, 36, 76, 84, 176–77 Ali Gh. on bad quality housing by UCO Ltd, 103 employment of, 28, 42, 44, 49, 53 Habib and Ahmed on bad quality housing, 110 housing of, 84, 101, 103–4, 110, 113, 126n3

Index • 205

mixed marriages and, 32–34 single men as, 31, 36, 133 spatial dispersion of, 118, 123, 124, 125 textile industry and, 49, 76, 84, 118 Turkish immigrants, 24–26, 27, 29, 30, 31–35, 62, 63–64, 70n71, 119–20, 122, 125, 165, 169, 172, 176–77. See also Emirdağ Ali D., 53, 150 Ali G. as, 48–49, 79 Celyn C. on workplace relations in mixed workplaces, 80, 83 employment of, 28, 42, 43, 44, 45, 46, 47, 62 entrepreneurship of, 47, 169 Faruk, 26, 63, 134 Fatma K. on housing experiences, 113–14 housing habits of, stories about, 108–9, 127n26, 127n27 housing of, 106, 110, 112–13 language and, 79, 80, 85–86, 143, 150 Mehmet B. on neighbour relations, 154 Mehmet K. on cultural differences and neighbour relations, 144–44, 154 Mehmet S. on social relations at work, 78 spatial mobility and, 34–35, 112–13 partner choice of, 32–33, 170, 172–73 pictures of, 26, 46, 63 second generation, 1, 168, 169, 170, 171, 172–73 social relations in mixed workplaces for, 63, 74, 76, 78–79 social relations with neighbours and, 133, 143–44, 150, 152, 153–54, 159n23 spatial dispersion of, 117, 119–20, 122, 123, 125 transplanted community and, 26, 34 unemployment for, 52–53, 168–69 women as, 29, 32, 42–43, 45, 46, 138, 159n23 UCO Ltd, 12, 28, 29, 56–57, 63, 67n19, 96n10. See also employers, Paul; Willem; Willy football team at, 81, 82

housing by, 102–4, 118, 124, 126n6, 129n59 immigrant labour data at, 12, 28, 46, 48, 49, 52, 61, 62, 168 language and interpreters at, 80, 86 long-term employment at, 48, 52, 53 pictures of, 57, 63, 82 social relations at, 58, 74, 76, 80, 81, 84–85 unemployment, 18, 20, 29, 30 economic crisis and, 20, 51, 52–53, 65 for immigrant workers, 52–54, 65, 68n30, 68n33 language and, 82–83, 95, 97n27 men and, 54, 82–83 second generation and, 168–69, 171 social relations in mixed workplaces and, 54, 74, 82–83, 164 social relations in the neighbourhood and, 155 textile industry and, 28, 29 women and, 52, 53–54 Union Cotonnière. See UCO Ltd unions immigrant workers and, 74, 75–76, 127n26 secondary sector of the labour market and, 18 van den Berg-Eldering, Lotty on ethnic neighbourhoods, 155 on female labour, 42, 77–78 on neighbour relations, 137, 139 VOLVO car factory, 17, 18, 48, 74, 76, 171 white flight, 165, 168, 173. See also suburban flight textile industry and, 168 women, 100, 131. See also women, in labour force Belgian women and their relations with immigrants, 134, 144, 145, 146–49, 150, 151, 157, 166 housebound, 19, 42, 43, 65, 139–40, 156–57 housing conditions and, 113–14 as long-term immigrants, 35

206 • Index

women (cont.) Moroccan, 32, 36, 42, 77–78, 139–40, 143, 158n21 migration status of, 32, 42, 65 neighbour relations for, 131–32, 137, 138, 139–41, 143, 145–50, 156–57, 158n21, 166 ratio of, 27, 29, 31–32 Spanish, 27, 31, 32, 36, 90–92 Turkish, 29, 32, 35, 36, 42, 45, 46, 143, 159n23, 170 women, in labour force, 18, 19–20, 41–44, 45, 66n4, 67n12, 78–79, 81 discrimination of, 42, 66n3 domestic services, 29, 31, 39n36, 42, 43, 45, 54, 60, 91, 105

entertainment, 29, 39n36, 44 legislation and, 42–43 textile industry, 43, 45, 57, 58, 61–62 unemployment and, 52, 53–54 Woonfonds Gent, 107, 111 workmen’s houses. See beluik work. See labour work permits, 38n29, 43, 50, 67n28, 84 workplaces, 6–7, 41, 54, 61. See also construction; domestic services; social relations, workplace; textile industry descriptions of immigrant, 54 domestic services, 59–60, 90–94 ethnic, 6, 64, 65, 72, 86–90, 94, 163, 165 mixed, 61–64, 73–86