Selected Studies In: International Migration and Immigrant Incorporation

imiscoe
TEXTBOOKS
martiniello & rath (eds.)

Migration and ethnic studies are on the rise. A body of literature has rapidly
grown and, within it, a European research area is emerging. Yet, the scholarship
is still highly fragmented, being largely orientated towards the United States
and other countries with longer, older narratives of immigration. Unlike
people, theories and concepts do not travel easily, meaning we cannot take for
granted that research results are equally applicable on all continents. The first
Selected Studies in
volume of the imiscoe Textbooks Series answers the pressing need for a
European perspective on migration. Assembling for the first time in a single
binding are 25 classic papers that have had a lasting impact on studies of
international migration and immigrant integration in Europe. Not only is
Selected Studies in International Migration and Immigrant Incorporation

International Migration
this book a body of knowledge drawing together complementary expertise
developed in the field thus far, it is a launch pad for cross-national comparisons
around the globe.
and Immigrant
Marco Martiniello is research director of the National Fund for Scientific
Research (frs-fnrs) in Belgium and a professor of sociology and politics at
the University of Liège, where he also serves as director of the Center for
Ethnic and Migration Studies (cedem). Jan Rath is a professor of urban
sociology at the University of Amsterdam, where he also serves as director of
the Institute for Migration and Ethnic Studies (imes).
Incorporation
“The editors have selected from both the grounding classics and the best new work to show marco martiniello & jan rath (eds.)
how migration is transforming the rich democracies.”
Professor John Mollenkopf, The City University of New York
“A collection of must-read, though sometimes hard-to-find, pieces that any scholar or

student interested in immigration to Europe and its consequences will want to consult.”
Professor Roger Waldinger, University of California, Los Angeles
“A must not only for courses focused on Europe, but also a most useful tool for shedding new
light on North American migration by casting it in an often neglected comparative context.”
Professor Aristide Zolberg, The New School for Social Research
isbn 978 90 8964 160 1
amsterdam university press · www.aup.nl Amsterdam University Press

1
2
3
4
5
6
7
8
9
10
Selected Studies in International Migration 11
and Immigrant Incorporation 12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
migration en ethnic deel 1.indd 1 04-03-10 15:56

1 IMISCOE
2 International Migration, Integration and Social Cohesion in Europe
3 The IMISCOE Research Network unites researchers from, at present, 25 institutes
4 specialising in studies of international migration, integration and social cohesion
5 in Europe. What began in 2004 as a Network of Excellence sponsored by the Sixth
Framework Programme of the European Commission has become, as of April 2009, an
6 independent self-funding endeavour. From the start, IMISCOE has promoted integrated,
7 multidisciplinary and globally comparative research led by scholars from various branches
8 of the economic and social sciences, the humanities and law. The Network furthers existing
studies and pioneers new scholarship on migration and migrant integration. Encouraging
9 innovative lines of inquiry key to European policymaking and governance is also a priority.
10
11 The IMISCOE-Amsterdam University Press Series makes the Network’s findings and
results available to researchers, policymakers and practitioners, the media and other
12 interested stakeholders. High-quality manuscripts authored by Network members and
13 cooperating partners are evaluated by external peer reviews and the IMISCOE Editorial
14 Committee. The Committee comprises the following members:
15 Christina Boswell, School of Social and Political Science, University of Edinburgh,
16 United Kingdom
17
Tiziana Caponio, Department of Political Studies, University of Turin / Forum for
18 International and European Research on Immigration (FIERI), Turin, Italy
19
20 Michael Collyer, Sussex Centre for Migration Research (SCMR), University of Sussex,
United Kingdom
21
22 Rosita Fibbi, Swiss Forum for Migration and Population Studies (SFM), University of
23 Neuchâtel, Switzerland / Institute of Social Sciences, University of Lausanne
24 Albert Kraler, International Centre for Migration Policy Development (ICMPD), Vienna,
25 Austria
26
Leo Lucassen, Institute of History, Leiden University, The Netherlands
27
28 Jorge Malheiros, Centre of Geographical Studies (CEG), University of Lisbon, Portugal
29
Marco Martiniello, National Fund for Scientific Research (FNRS), Brussels / Center for
30 Ethnic and Migration Studies (CEDEM), University of Liège, Belgium
31
32 Patrick Simon, National Demographic Institute (INED), Paris, France
33 Miri Song, School of Social Policy and Sociology, University of Kent, United Kingdom
34
35 IMISCOE Policy Briefs and more information can be found at www.imiscoe.org.
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
Selected Studies in International Migration 9
10
and Immigrant Incorporation 11
12
edited by 13
Marco Martiniello and Jan Rath 14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
IMISCOE Textbooks 40
41
Amsterdam University Press 42
43

1
2
3
4 The multidisciplinary IMISCOE-AUP Textbook Series encompasses, at present, four volumes, and
5 aims to present both an international comparison of the development of international migration and
6 immigrant integration in Europe and an assessment of theoretical approaches with regard to this
7 issue. Materialisation of this objective strengthens the development and dissemination of a body
8 of common knowledge in this field and consequently boosts the growth of a European research
9 area. The current volume encompasses 25 theoretical papers that have had an impact on research
10 in Europe or reflect a European perspective on international migration and immigrant integration.
11 Our thanks are due to IMISCOE and to all those who have contributed, in whatever way, to the
12 realisation of this first volume. We especially thank Anna Swagerman and, most of all, Kim Jansen.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33 Cover design Studio Jan de Boer, Amsterdam
34 Layout The DocWorkers, Almere
35 isbn 978 90 8964 160 1
36 e-isbn 978 90 4851 104 4
37 nur 741 / 763
38
39 © Martiniello and Rath / Amsterdam University Press, Amsterdam 2010
40
All rights reserved. Without limiting the rights under copyright reserved above, no part of this book
41 may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or
42 by any means (electronic, mechanical, photocopying, recording or otherwise) without the written
43 permission of both the copyright owners and the authors of the book.

1
2
Contents 3
4
5
Introduction: migration and ethnic studies in Europe 7 7
8
Part 1 - The migration process 19 9
1 Stephen Castles and Godula Kosack 10
The function of labour immigration in Western 11
European capitalism 21 12
2 Tomas Hammar 13
Introduction to European immigration policy: 14
a comparative study 45 15
3 Thomas Faist 16
The crucial meso-level 59 17
4 Steven Vertovec 18
Conceiving and researching transnationalism 91 19
5 Russell King 20
Towards a new map of European migration 111 21
6 Virginie Guiraudon 22
The constitution of a European immigration policy 23
domain: a political sociology approach 141 24
7 Abdelmalek Sayad 25
Immigration and ‘state thought’ 165 26
27
Part II - Modes of incorporation 181 28
8 Hans van Amersfoort 29
‘Minority’ as a sociological concept 183 30
9 Tariq Modood 31
‘Black’, racial equality and Asian identity 201 32
10 William Rogers Brubaker 33
Introduction to immigration and the politics of citizenship 34
in Europe and North America 215 35
11 Marco Martiniello 36
Ethnic leadership, ethnic communities’ political 37
powerlessness and the state in Belgium 237 38
12 Michel Wieviorka 39
Racism in Europe: unity and diversity 259 40
13 Rainer Bauböck 41
Changing the boundaries of citizenship: 42
the inclusion of immigrants in democratic polities 275 43

6 contents
1 14 Robert Kloosterman, Joanne van der Leun and Jan Rath

2 Mixed embeddedness: (in)formal economic
3 activities and immigrant businesses in the Netherlands 315
4 15 Patrick Simon
5 The mosaic pattern: cohabitation between ethnic
6 groups in Belleville, Paris 339
7 16 Hassan Bousetta
8 Political dynamics in the city: three case studies 355
9 17 Adrian Favell
10 Integration and nations: the nation-state and research
11 on immigrants in Western Europe 371
12
13 Part III - Conceptual issues 405
14 18 Fredrik Barth
15 Introduction to ethnic groups and boundaries:
16 the social organization of cultural difference 407
17 19 John Rex
18 The theory of race relations: a Weberian approach 437
19 20 Floya Anthias and Nira Yuval-Davis
20 Contextualizing feminism: gender, ethnic and
21 class divisions 469
22 21 John Solomos
23 Varieties of Marxist conceptions of ‘race’, class and the
24 state: a critical analysis 489
25 22 Frank Bovenkerk, Robert Miles and Gilles Verbunt
26 Racism, migration and the state in Western Europe:
27 a case for comparative analysis 517
28 23 Robert Miles and Victor Satzewich
29 Migration, racism and ‘postmodern’ capitalism 537
30 24 Etienne Balibar
31 Class racism 567
32 25 Ceri Peach
33 The ghetto and the ethnic enclave 581
34
35 About the editors 607
36
37 List of sources 609
38
39 Index 613
40
41
42
43

1
2
Introduction: 3
4
migration and ethnic studies in Europe
5
6
8
9
10
Over the past few decades, practically every country in the advanced 11
world has witnessed a substantial increase in immigration (Castles 12
& Miller 2009). Some countries such as Canada or the United States 13
have hosted immigration for centuries, and their mental map and 14
social fabric are consequently geared to accommodating newcomers. 15
But even for those countries, the magnitude of the current flow of 16
people crossing the border with or without valid documents was un- 17
expected. The US had its version of the guest worker system in the 18
Mexican Bracero Program of the 1940s, but the immigration of Latino 19
workers for the agricultural industry is nothing when compared to 20
what was in store. The previous immigration regime favoured immi- 21
grants from Europe, but the abolition of restrictions for immigrants 22
from Africa, Asia or Latin America in 1965 opened the US to non- 23
Europeans (Cornelius, Martin & Hollifield 1994). Immigration laws 24
were tightened in the 1980s and 1990s in response to growing politi- 25
cal pressure against what some regarded as unbridled immigration 26
as well as mounting unemployment and rising public expenditures 27
for documented and undocumented immigrants alike. Meanwhile, 28
Los Angeles outnumbered America’s all-time city of immigration, 29
New York. That being said – and contrary to the general political 30
mood in the US – authorities still maintain that the city warmly wel- 31
comes immigrants. Even if immigrants are not always treated as 32
welcome guests, still acknowledged are the contributions they have 33
made to the metropolis’ flourishing, now and in the past. 34
35
On the other side of the Atlantic, similar developments have occurred, 36
though under different circumstances. One striking difference is 37
that Europe’s nations have never really considered themselves coun- 38
tries of immigration the way North America has. On the contrary, 39
many, including Greece, Ireland, Italy, Portugal. Spain – sending 40
countries in living memory – and even the Netherlands presented 41
themselves as countries of emigration. International migration and international 42
the social problems it allegedly generates – and with which it usually migration 43

8 marco martiniello and jan rath
1 is amalgamated – have in recent years emerged as inevitable issues

2 in the media and politics, especially after 9/11. Migration has been
3 constructed as an international and domestic security issue linked to
4 urban unsafety, international organised crime, terrorism, illegality,
5 environmental issues and public health. This has aggravated the fear
6 of an invasion of Europe by cohorts of poor people. Meanwhile, the
7 issue of the co-existence between nationals and migrant communi-
8 ties has become increasingly interpreted in terms of social tensions
9 and problems (criminality, drugs, unemployment, school drop-out,
10 insecurity, etc.). In several European countries, political parties play
11 on the fears of the electorate with regard to migration in order to
12 gather electoral support. More precisely, since 9/11 and the Madrid
13 bombings of 3/11, there has been real intellectual and political panic
14 surrounding the issue of Islam in Europe and elsewhere.1 To be fair,
15 there is also a more positive approach to migration and multicultur-
16 alism. Some welcome immigration as an answer to the greying of the
17 population. Others see it as a necessary condition for economic ad-
18 vancement in the framework of the Lisbon Agenda. The same holds
19 for diversity. While many politicians and opinion leaders advance an
20 assimilationist policy and thus aim at abolishing any form of ethnic
21 diversity, urban sociologists, economic geographers and city plan-
22 ners are increasingly identifying diversity as key for economic growth
23 (see for instance Florida 2000). Fractions of the general public also
24 value diversity in their social practices and modes of consumption as
25 illustrated by the success of ethnic food, fashion and world music, for
26 example in most European cities.
27
28 Nevertheless, public and political debates about migration are hardly
29 serene. In fact, since the early 1980s, migration has become the focal
30 point for passionate debates and controversies on a regular basis.2
31 In these circumstances, social scientists find themselves caught in a
32 very difficult position, especially if they take seriously the point that
33 their role is to elaborate knowledge free from passions and fears.
34 Their work is, in effect, running the risk of unwillingly reinforcing
35 the excessive dramatisation surrounding migratory phenomena.
36 Even when they assign themselves the precise opposite goal, they
37 are not always immune from distorted interpretations of their work
38 within the public sphere.
39
40 This ambiguity did not, however, preclude social scientists from be-
41 migration and coming very prolific. Proliferation of migration and ethnic studies
42 ethnic studies in Western Europe is a relatively recent phenomenon. This branch
43 of social scientific research took off in several European countries

introduction 9
in the early 1980s and a little earlier in countries such as the United 1
Kingdom. In the first stages, the study of migration was largely re- 2
served for demographers and political economists. Traditionally, 3
it has been a key area of study for the discipline of demography. 4
Political economy has quite logically developed an interest in this 5
field. Until the oil crisis of 1973, the mere economic dimension of 6
migration was actually assumed to be the most obvious and most 7
natural dimension of the process. It was usually portrayed in terms 8
of the movements of the labour force. 9
10
The aim of the introduction to this textbook is not to present a classic 11
state of the art on migration and ethnic studies. This work has already 12
been done several times and has given rise to many publications in 13
different countries (see for instance Penninx, Berger & Kraal 2006). 14
Instead of repeating what has already been achieved, it seems more 15
fruitful in this context to articulate a number of marked features of 16
the field of study. We will briefly reflect on European migration and 17
ethnic studies and highlight a number of academic publications that 18
were central to this development. In our view, two structural factors 19
shape European migration and ethnic studies. Firstly, there is the 20
structure of European academic research, both in terms of disciplin- 21
ary and thematic profile and funding. Secondly, we turn our attention 22
to the dominance of American perspectives in this field and the ten- 23
dency of European researchers to take these perspectives for granted. 24
25
26
European migration and ethnic studies in a wider scientific 27
structure 28
29
The first feature of European migration and ethnic studies is what 30
may be called the problem of the epistemological break, according to absence of an 31
Gaston Bachelard (1983) and Pierre Bourdieu (1973). More precisely, epistemological 32
we should say that a major challenge in the study of migration and break 33
ethnic relations is the absence of any epistemological break, which 34
is often a result of the aforementioned intellectual emergency and 35
the social conditions of production of the social scientific work. As 36
discussed above, the common sense, led by a biased media sociali- 37
sation, conceives of immigration in terms of economic, social and 38
political problems. These include insecurity and criminality, unem- 39
ployment, poverty, urban decay, violence, religious and ethnic con- 40
flicts and the dilution of the nation. Since 1973, this mosaic of folk 41
representation has been widely diffused in the public. Surprisingly, 42
the social sciences as a whole and sociology, more specifically, did 43

1 not represent any exception. Sociologists have actually started cat-

2 egorising the social experience of migrant populations into distinc-
3 tive domains, which they elaborated as specific social problems to be
4 studied and resolved. In fact, we have observed how construction of
5 the sociological perspectives on migration and ethnic relations in the
6 early hours of the discipline simply mirrored the intuitive theories
7 of migration among the wider public. This led to the development
8 binary perspectives of a literature rife with binary perspectives, such as immigrants and
9 housing, immigrants and school, immigrants and criminality, im-
10 migrants and security, immigrants and health, immigrants and cul-
11 ture, immigrants and the labour market. A great number of studies
12 has been produced – and continues to be – in all these sub-fields of
13 research. In the worst cases, they have been either flatly empiricist or
14 simply unfruitful due to their redundancy. On the whole, one must
15 reckon with this first major difficulty in order to account for the rela-
16 tive theoretical stagnation of the field. (For a more critical point of
17 view, see Rath 2001.)
18
19 It’s as though migration and ethnic studies were meant to contribute
20 to solving the social problems associated with a phenomenon still
21 dominantly perceived as a threat to the social order (Sayad 1984).
22 Insofar as it tends to answer a social demand more or less directly,
23 the sociology of migration has been constrained. It has been forced
24 to internalise the problematised and dramatised perception of the
25 common sense – which is itself largely determined, as stated above,
26 by a concern for social order. In this situation, it is quite difficult to
27 establish a positive assessment in terms of the scientific value of the
28 works produced. As noted by Michel Oriol:
29
30 In their concern for solving concrete problems quickly, they [the re-
31 searchers] can only raise the problems in terms comparable to those
32 of the public opinion. It becomes therefore more difficult to break
33 off with ideology in order to establish a properly scientific approach.
34 (1981: 6)3
35
36 The tight entanglement of social debates and policies helps explain
37 the weaknesses of the sociology of migration processes and ethnic
38 relations, as well as the predominance in the field of the flattest em-
39 piricism (Noiriel 1989).
40
41 Some claim that it is hard to talk of migration and ethnic studies as
42 a firm, coherent theoretical corpus in Europe. In other words, this
43 field of research would not have reached the status of a branch of

introduction 11
the social sciences in its own right. The study of migration and ethnic 1
relations could hardly pretend to compete academically with more es- 2
tablished branches of sociology, anthropology, political science and 3
so forth because of its major theoretical weaknesses and fragmenta- 4
tion. Others believe that mainstream sociology is not theoretically 5
stronger. As such, the problem would be related to the structure of 6
social science research, which is fairly disciplinarily oriented, with 7
disciplinary-based institutes, evaluations and funding. Meanwhile, 8
migration and ethnic studies is thematically oriented and multi- 9
disciplinary. For sociologists, this field is not sociological enough; for 10
anthropologists, geographers and political scientists the same holds 11
true. Consequently, scholars publish in specialised migration and 12
ethnicity journals that attract fewer readers, reach lower citations and 13
have less impact scores. The list goes on. 14
15
It is apparent that migration and ethnic studies was for a long time undervaluation of 16
marginalised in academic circles and universities. As already under- European migration 17
scored by Abdelmalek Sayad (1984) and Philippe Lorenzo (1989), and ethnic studies 18
it was an undervalued field of research. The field consequently re- 19
mained unattractive for academic researchers until not so long ago. 20
This is mainly the case in Continental Europe. In the US and, to a 21
lesser extent, in the UK, things are different. In the New World, the 22
professionalisation of sociology happened in the context of a country 23
conceiving its history as one of immigration. It comes therefore as 24
no surprise that this discipline has grown while maintaining immi- 25
gration as a central concern. For instance, the research produced in 26
this field has allowed the Chicago School to develop and to become 27
a world-famous school of sociology. In many other European coun- 28
tries, the leading figures of social sciences were until rather recently 29
not interested in these phenomena. When they did show an interest, 30
they did it in a way that was once characterised by Lorenzo (1989) as 31
marginal, periodical and brief. 32
33
As far as social sciences and the study of migration are concerned, 34
researchers are all too often constrained by having to chase down 35
funding and research contracts at various ministries and govern- 36
mental agencies. The fact that immigration and integration have, 37
in the course of the last twenty years, remained highly contentious 38
and sensitive from an electoral point of view has had various con- 39
sequences. Most often, elected politicians holding executive offices 40
are particularly careful in selecting the research projects that may be 41
immediately useful in terms of policymaking. Sometimes, an advan- 42
tage is given to research projects that give academic alibis – often of a 43

1 quantitative nature – to policies already agreed upon. In other words,

2 politicians in executive offices have a strong tendency to intrude
3 upon the academic debate by imposing the ‘legitimate’ research
4 problematics and themes without taking into account the research-
5 ers’ properly scientific concerns and agenda. One can observe how,
6 in recent years, themes worth receiving subsidies were the control
7 of asylum seekers and refugee flows, the control of external borders,
8 criminality, migrant insecurity and employment and unemployment
9 and, last but not least, Islamic terrorism issues.
10
11 contractualisation The scarcity of sources of funding and the ‘contractualisation’ of
12 of research research do not easily accommodate the theoretical concerns of the
13 researchers. There is a power struggle between the politicians and
14 policymakers in one camp and the academics in the other. The latter
15 seem to be at the base end of it. However, the relative autonomisation
16 of the academic field is still a precondition for an effective epistemo-
17 logical break in the course of a solid research process. Furthermore,
18 it constitutes an important difference between non-academic exper-
19 tise and scientific research.
20
21 Researching and teaching in this field have, for a long time, remained
22 poorly valued on the whole. Nor have the pursuits been very reward-
23 ing in terms of academic prestige. Investing in these themes has not
24 been the most direct way forward for those willing to join the elite
25 of social science research. As a respondent of Lorenzo put it: ‘You
26 don’t make a career in academia with immigration’ (Lorenzo 1989:
27 9). Sayad once asked the very uneasy question: ‘Is the science of the
28 “poor”, of the “small people”, (socially) a poor science, a small one?’4
29 (1984: 20). There is no doubt about the answer: the sociology of im-
30 migration was a minor sociological subject matter.
31
32 Moreover, it seems that immigration and ethnic relations have almost
33 exclusively been studied by researchers who were in one way or an-
34 other complacent to the subject. A number of researchers in the field
35 were either migrants themselves or of migrant descent. The same
36 narrow relationship between personal experience and research experi-
37 ence was observable among native researchers. They often had a spe-
38 cial relationship with immigrant population, either through marriage
39 or friendship. In other cases, they had close links with the migrants’
40 countries of origin. It should be said that many of these researchers,
41 both natives and migrants, occupied precarious and unstable po-
42 sitions within the academic world and were often badly dependent
43 on external funding. One could contend that, on the social scale

introduction 13
of academic prestige, migration and ethnic studies is still too often in 1

the hands of ‘second-class’ researchers. This latter statement is im- second-class 2
mune from any judgment of their scientific competence. It actually researchers 3
aims to emphasise how their social and national backgrounds, i.e. 4
the weakness of their position in the academic field, do not generally 5
qualify them for the most academically valued positions. Moreover, 6
it is often expected that ethnic minority researchers should work on 7
ethnic and migration issues, just as it is usually considered ‘natural’ 8
that gender studies be foremost a matter for female researchers. This 9
situation has significantly evolved over the course of the 1990s and 10
2000s. Although theoretical divergence within the European field 11
on the relevance of ethnicity as a mobilising social and political force 12
remains important, a form of decompartmentation and demarginali- 13
sation is undoubtedly at work. From either analytical angle, migra- 14
tion and ethnicity have become key issues in the social analysis of 15
contemporary Europe. 16
17
In the course of the 1990s and the 2000s, European migration and 18
ethnic studies has undergone a process of change, of demarginalisa- professionalisation 19
tion and of professionalisation. There are many specialised academic of European 20
journals ranked in the ISI Web of Knowledge (e.g. Ethnic and Racial migration and 21
Studies, International Migration Review, Journal of Ethnic and Migration ethnic studies 22
Studies, Revue européenne des migrations internationales). There are 23
many workshops, conferences and international networks dedicat- 24
ed to the study of migration and integration. There is a number of 25
specialised research institutes at various universities and a growing 26
number of Master’s and PhD programmes in fields related to mi- 27
gration and ethnic studies. Moreover, main funders have launched 28
special programmes for research projects that revolve around mi- 29
gration and integration (e.g. the European Commission’s Seventh 30
Framework Programme, the New Opportunities for Research 31
Funding Co-operation in Europe network known as NORFACE, 32
the Foundation for Population, Migration and Environment (PME/ 33
BMU) and various national research councils). In short, migration 34
and ethnic studies is, more and more, gaining respect as a legitimate 35
academic field worth an investment by students who hope to find a 36
job in the domain. 37
38
39
European social scientists’ fascination for the Americas 40
41
The second feature of European migration and ethnic studies is the 42
adoption – without sufficient care – of conceptual and theoretical ele- 43

1 ments developed in other social and national contexts. As observed

2 by Oriol:
3
4 Sociology has experienced the same enthusiasm as the population
5 in general for the Americas and sought its paradigms there, just as
6 people have sought their fortune. (Oriol 1981: 24)5
7
8 In fact, a wide number of theoretical constructions in the European
9 sociology of migration has been imported from the US. The Chicago
10 theoretical and School and the structural-functionalism among other schools have
11 conceptual imports provided European researchers with a huge stock of concepts. We can
12 mention here a number of examples: assimilation, adaptation, mar-
13 ginality, inclusion, integration. The reason for these abundant theo-
14 retical imports seems to lie in the fascination for the US mentioned
15 by Oriol, as well as the fact that the discipline of sociology in America
16 was far more advanced in the study of migration than the European
17 one when this theme became topical among European researchers.
18 Acknowledging the richness and relevance of the American concep-
19 tual legacy cannot preclude expressing explicit reservations in terms
20 of the very questionable way in which these concepts were used and
21 applied by European researchers.
22
23 A major problem lies in the fact that divergence has been underes-
24 timated, in terms of the historical, social and economic background
25 of Europe and the US. This divergence should have, at the very least,
26 stimulated a careful transferring of concepts from one context to the
27 other. Indeed, different historical and spatial contexts never corre-
28 spond in every respect, and therefore it is somehow illusory to use
29 theories and concepts developed for explaining and accounting for
30 the situation in one context for the other. Before they can be intro-
31 duced in a given context, theories and concepts external to a social
32 formation should first undergo a critical and thorough examination.
33 They must be deconstructed and reconstructed in order to be adapt-
34 ed satisfactorily to a new context. This work has not been sufficiently
35 achieved in this field of study, especially when it comes to importing
36 elements of the American intellectual tradition. Moreover, the intrin-
37 sic problems of these imported concepts and problematics were not
38 definitively solved even in the American context. Therefore, by intro-
39 ducing them uncritically in Europe, theoretical difficulties have also
40 been unwillingly taken on board. This factor may in itself account for
41 the uneasy development of a European sociology of migration and
42 ethnic relations.
43

introduction 15
These two problems of the theoretical and conceptual imports, es- 1

pecially from the US, may be illustrated briefly through the example 2
of the late introduction and the development of concepts linked to 3
‘ethnicity’, ‘multiculturalism’ (Martiniello 1997) and ‘underclass’ in ethnicity 4
Continental Europe. It is unquestionable that these external elements 5
of debate can potentially reinvigorate this field of research. However, 6
these categories must be used carefully. Indeed, can we assert that 7
the concept of ‘ethnicity’ refers to the same intellectual representa- 8
tion in a society that has always conceived of itself as an immigration 9
country? This representation has been shaped for a long time by the 10
powerful ideology of the ‘melting pot’. Countries with old and strong 11
national and nationalist traditions have traditionally considered mi- 12
grant populations as a temporary labour force. European researchers 13
have often neglected this crucial question. Beyond that, sociological 14
debates about ‘ethnicity’ in the US gave rise to the creation of com- 15
peting schools of thought. Today, the advocates of the substantial- 16
ist conception of ‘ethnicity’ seem to be mostly minorised because 17
of the thorough criticism of their position in the early 1960s and, 18
even more sharply, after the publication of the influential works of 19
Nathan Glazer and Daniel Moynihan (1972). Now, among European 20
researchers manipulating the concept of ethnicity in migration and 21
ethnic studies, some still adopt an ambiguous position concerning 22
substantialism, which may bring the theoretical debate a few decades 23
back. 24
25
Another example concerns ‘underclass’. The concept is highly con- underclass 26
tested in American academia, notably for having a strong moralistic 27
content. By reintroducing it in French social sciences in the early 28
1990s, Didier Lapeyronnie imported the American controversy and, 29
to a certain extent, the moralistic approach to the issue of social and 30
economic exclusion in Europe. Importing a concept without refer- 31
ring to the context in which it was created or the controversies it has 32
produced is problematic. We cannot assume a priori that underclass 33
is a useful concept for Europe. 34
35
36
European migration and ethnic studies 37
38
The Europeanisation and the internationalisation of research through 39
several networks and programmes, such as those in the European 40
Union’s scientific research frameworks, can give a fresh new theo- 41
retical orientation to the discipline. It is indisputable that immense 42
conceptual and methodological problems have yet to be solved (Lloyd 43

1 1995), and that there exist only very narrow margins for developing
2 crucial scientific research activities such as data collection and stan-
3 dardisation on an international level. However, at present, there are
4 wider opportunities being offered to European researchers, allowing
5 them to meet on a more or less regular basis and to exchange ideas
6 in collaborative research projects.
7
8 Cooperation needs to be structured. Research must, above all, focus
9 on European issues. Relevant questions must be asked. For instance,
10 how does one regulate supply- and demand-driven migration? What
11 is the best way to integrate for immigrants who stay? How can insti-
12 tutional arrangements be adapted so that social cohesion does not
13 vaporise? In an effort to answer such questions, the research network
14 IMISCOE IMISCOE, which stands for International Migration, Integration
15 and Social Cohesion in Europe, implements a rigorously compara-
16 tive multidisciplinary research programme with Europe as its central
17 focus. This is a promising venture for designing truly transnational
18 and transdisciplinary research projects in Europe, while also foster-
19 ing cooperation with academics interested in the same issues world-
20 wide.
21
22 To conclude, it seems indisputable that we need more profound re-
23 flection on the core features of European migration and ethnic stud-
24 ies. Such a reflection implies that students of migration and ethnic
25 studies familiarise themselves with key texts in this field. For this
26 volume, we collected a number of texts that we believe were crucial
27 for the development of European research in our field. To first iden-
28 tify these texts, we consulted with several dozen key academics in
29 migration and ethnic studies, asking them to ‘nominate’ Europe’s
30 most classic publications. As could be predicted, we ended up with
31 a very long list of titles and authors. Some names, however, were
32 unanimously regarded as crucial in the development of European
33 migration and ethnic studies.
34
35 We take sole responsibility for the next phase of the selection process
36 during which we reduced the list to those comprising the chapters
37 of this volume. We acknowledge that the selection process was, at
38 the end of the day, arbitrary since other works could certainly have
39 been chosen. Our selection, however, provides a compelling repre-
40 sentation of European migration and ethnic studies. The chapters
41 address the main issues dealt with over the years within different
42 academic disciplines, different schools of thought and in a number
43 of European countries. We chose to organise the chapters themati-

introduction 17
cally. Chapters 1 through 7 deal with the migration process and its 1
related policies. Chapters 8 through 17 discuss modes of incorpora- 2
tion. Finally, chapters 18 through 25 bring together works dedicated 3
to transversal conceptual issues. Although some formatting changes 4
have been made, the substance of each chapter is a reproduction of 5
the text as it appeared in its original publication. In each thematic 6
section, the chapters appear in chronological order of their publica- 7
tion. We hope this organisation will help contextualise the works, 8
giving readers a sense of when and how these specific topics and ap- 9
proaches in European migration and ethnic studies emerged. 10
11
12
Notes 13
14
1 See for example Johnson & Crawford (2004) ‘New Breed of Islamic Warrior 15
is Emerging: Evidence in Madrid Attack Points to Takfiris, Who Use
Immigration as a Weapon’, The Wall Street Journal 29 March: A16.
16
2 See 2002’s special issue of the Journal of International Migration and 17
Integration, 3 (3/4). 18
3 Free translation of: Par souci de résoudre vite des problèmes concrets, ils (les 19
chercheurs) ne peuvent guère les poser que dans les termes où l’opinion publique 20
les reconnaît. Il sera alors d’autant plus difficile de s’arracher à l’idéologie, pour
21
essayer de fonder une démarche proprement scientifique... (Oriol 1981: 6).
4 Free translation of: La science du ‘pauvre’, du ‘petit’ (socialement) est-elle une 22
science ‘pauvre’, est-elle une ‘petite science’? (Sayad 1984: 20). 23
5 Free translation of: La Sociologie a connu la même fascination que les peuples 24
pour les Amériques et vint y chercher ses paradigmes tandis qu’ils y quêtaient for- 25
tune. (Oriol 1981: 24). 26
27
28
References 29
30
Bachelard, G. (1973), La formation de l’esprit scientifique. Paris: Vrin. 31
Bourdieu, P., J. C. Chamboredon & J.C. Passeron (1973), Le métier de 32
sociologue. Paris: Mouton. 33
Cornelius, W., P. Martin & J. Hollifield (1994), Controlling 34
Immigration: A Global Perspective. Stanford: Stanford University 35
Press. 36
Castles, S. & M. Miller (1999), The Age of Migration, 4th edition 37
(2009). New York: Guilford Press. 38
Florida, R. (2000), The Rise of the Creative Class: And How It’s 39
Transforming Work, Leisure, Community and Everyday Life. New 40
York: Basic Books. 41
Glazer N. & D. P. Moynihan (1970), Beyond the Melting Pot: The 42
Negroes, Puerto Ricans, Jews, Italians, and Irish of New York City. 43
Cambridge: MIT Press.

1 Lloyd, C. (1995), ‘International comparison in the field of ethnic rela-

2 tions’, in A. Hargeaves & J. Leaman (eds.), Racism, Ethnicity and
3 Politics in Contemporary Europe, 31-44. Aldershot: Edward Elgar.
4 Lorenzo, P. (1989), Approche qualitative des recherches sur l’immigration
5 en France. Paris: Centre de Recherche et d’Études d’Anthropologie
6 et D’Urbanisme.
7 Martiniello, M. (1997), Sortir des ghettos culturels. Paris: Presses de
8 Sciences Po.
9 Noiriel, G. (1989), ‘Enjeux: Une histoire sociale du politique est-elle
10 possible?’, Vingtième Siècle October/December: 81-89.
11 Oriol, M. (1981), Bilan des études sur les aspects culturels et humains
12 des migrations internationales en Europe Occidentale 1918-1979.
13 Strasbourg: Fondation Européenne de la Science.
14 Penninx, R., M. Berger & K. Kraal (eds.) (2006), The Dynamics of
15 Migration and Settlement in Europe: A State of the Art. IMISCOE
16 Joint Studies. Amsterdam: Amsterdam University Press.
17 Rath, J. (2001), ‘Research on Immigrant Ethnic Minorities in the
18 Netherlands’, in P. Ratcliffe (ed.), The Politics of Social Science
19 Research: ‘Race’, Ethnicity and Social Change, 137-159. Houndmills/
20 Basingstoke: Palgrave MacMillan.
21 Sayad, A. (1984), ‘Tendances et courants des publications en Sciences
22 Sociales sur l’immigration en France depuis 1960’, Current
23 Sociology 32 (3): 219-304.
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
Part I 4
5
6
7
The migration process 8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
1. The function of labour immigration in 3
4
Western European capitalism
5
6
Stephen Castles and Godula Kosack 7
8
9
10
11
As of the 1950s, migrant workers flocked to Western Europe to take up the 12
13
manufacturing industries’ low-paid and low-qualified jobs. Some came from
14
former colonial areas. Others were recruited under a guest worker regime.
15
Virtually all occupied an inferior social position in key domains of social life,
16
notably the labour market and housing. A growing number of scholars, many 17
of whom were inspired by Marxist thought, tried to explain this phenomenon. 18
The sociologist and political economist Stephen Castles and the sociologist 19
and ethnologist Godula Kosack formulated this problematic in a compre- 20
hensive way in their seminal book from 1973, Immigrant Workers and Class 21
Structure in Europe. They had previously expounded their views in an article 22
published in 1972 in the journal New Left Review. Here they claimed that mi- 23
grant work fulfilled an economic and socio-political function for capitalism, 24
being a fresh reservoir of labour and a means of dividing the working class. 25
They further located the origin of racism in capitalist expansion. Castles and 26
Kosack garnered much praise for drawing connections between the political, 27
social and ideological demands of capitalism and migrant labour, and for 28
criticising studies that dealt only with the problems of assimilation of indi- 29
vidual migrants. 30
31
32
The domination of the working masses by a small capitalist ruling 33
class has never been based on violence alone. Capitalist rule is based 34
on a range of mechanisms, some objective products of the economic mechanisms of 35
process, others subjective phenomena arising through manipulation domination 36
of attitudes. Two such mechanisms, which received considerable at- 37
tention from the founders of scientific socialism, are the industrial 38
reserve army, which belongs to the first category, and the labour ar- 39
istocracy, which belongs to the second. These two mechanisms are 40
closely related, as are the objective and subjective factors which give 41
rise to them. 42
43

22 stephen castles and godula kosack
1 industrial Engels pointed out that ‘English manufacture must have, at all times
2 reserve army save the brief periods of highest prosperity, an unemployed reserve
3 army of workers, in order to produce the masses of goods required by
4 the market in the liveliest months.’1 Marx showed that the industrial
5 reserve army or surplus working population is not only the necessary
6 product of capital accumulation and the associated increase in labour
7 productivity, but at the same time ‘the lever of capitalist accumula-
8 tion’, ‘a condition of existence of the capitalist mode of production’.2
9 Only by bringing ever more workers into the production process can
10 the capitalist accumulate capital, which is the precondition for ex-
11 tending production and applying new techniques. These new tech-
12 niques throw out of work the very men whose labour allowed their
13 application. They are set free to provide a labour reserve which is
14 available to be thrown into other sectors as the interests of the capi-
15 talist require. ‘The whole form of the movement of modern industry
16 depends, therefore, upon the constant transformation of a part of
17 the labouring population into unemployed or half-employed hands.’3
18 The pressure of the industrial reserve army forces those workers who
19 are employed to accept long hours and poor conditions. Above all:
20 ‘Taking them as a whole, the general movements of wages are exclu-
21 sively regulated by the expansion and contraction of the industrial
22 reserve army.’4 If employment grows and the reserve army contracts,
23 workers are in a better position to demand higher wages. When this
24 happens, profits and capital accumulation diminish, investment falls
25 and men are thrown out of work, leading to a growth of the reserve
26 army and a fall in wages. This is the basis of the capitalist economic
27 cycle. Marx mentions the possibility of the workers seeing through
28 the seemingly natural law of relative over-population, and undermin-
29 ing its effectiveness through trade-union activity directed towards co-
30 operation between the employed and the unemployed.5
31 labour aristocracy The labour aristocracy is also described by Engels and Marx. By
32 conceding privileges to certain well-organized sectors of labour,
33 above all to craftsmen (who by virtue of their training could not be
34 readily replaced by members of the industrial reserve army), the capi-
35 talists were able to undermine class consciousness and secure an
36 opportunist non-revolutionary leadership for these sectors.6 Special
37 advantages, sometimes taking the form of symbols of higher status
38 (different clothing, salary instead of wages, etc.) rather than high-
39 er material rewards, were also conferred upon foremen and non-
40 manual workers, with the aim of distinguishing them from other
41 workers and causing them to identify their interests with those of
42 the capitalists. Engels pointed out that the privileges given to some
43 British workers were possible because of the vast profits made by the

part i – the migration process 23
capitalists through domination of the world market and imperialist 1

exploitation of labour in other countries.7 Lenin emphasized the ef- 2
fects of imperialism on class consciousness: ‘Imperialism... makes it 3
economically possible to bribe the upper strata of the proletariat, and 4
thereby fosters, gives shape to, and strengthens opportunism.’8 ‘... A 5
section of the proletariat allows itself to be led by men bought by, or 6
at least paid by, the bourgeoisie’, and the result is a split among the 7
workers and ‘temporary decay in the working-class movement’.9 8
The industrial reserve army and the labour aristocracy have not 9
lost their importance as mechanisms of domination in the current 10
phase of organized monopoly capitalism. However, the way in which 11
they function has undergone important changes. In particular the 12
maintenance of an industrial reserve army within the developed 13
capitalist countries of West Europe has become increasingly diffi- 14
cult. With the growth of the labour movement after the First World 15
War, economic crises and unemployment began to lead to political economic crises 16
tensions which threatened the existence of the capitalist system. 17
Capitalism responded by setting up fascist régimes in the areas 18
where it was most threatened, in order to suppress social conflict 19
through violence. The failure of this strategy, culminating in the de- 20
feat of fascism in 1945, was accompanied by the reinforcement of the 21
non-capitalist bloc in East Europe and by a further strengthening of 22
the labour movement in West Europe. In order to survive, the capital- 23
ist system had to aim for continuous expansion and full employment 24
at any price. But full employment strikes at a basic principle of the 25
capitalist economy: the use of the industrial reserve army to keep 26
wages down and profits up. A substitute for the traditional form of 27
reserve army had to be found, for without it capitalist accumulation 28
is impossible. Moreover, despite Keynsian economics, it is not pos- 29
sible completely to avoid the cyclical development of the capitalist 30
economy. It was therefore necessary to find a way of cushioning the 31
effects of crises, so as to hinder the development of dangerous social 32
tensions. 33
34
35
Immigrants as the new industrial reserve army 36
37
The solution to these problems adopted by West European capital- 38
ism has been the employment of immigrant workers from under employment of 39
developed areas of Southern Europe or from the Third World.10 immigrant workers 40
Today, the unemployed masses of these areas form a ‘latent surplus- 41
population’11 or reserve army, which can be imported into the de- 42
veloped countries as the interests of the capitalist class dictate. In 43

1 addition to this economic function, the employment of immigrant

2 workers has an important socio-political function for capitalism: by
3 creating a split between immigrant and indigenous workers along
4 national and racial lines and offering better conditions and status to
5 indigenous workers, it is possible to give large sections of the work-
6 ing class the consciousness of a labour aristocracy.
7 The employment of immigrant workers in the capitalist produc-
8 tion process is not a new phenomenon. The Irish played a vital part
9 in British industrialization. Not only did they provide a special form
10 of labour for heavy work of a temporary nature on railways, canals
11 and roads;12 their competition also forced down wages and condi-
12 tions for other workers. Engels described Irish immigration as a
13 ‘cause of abasement to which the English worker is exposed, a cause
14 permanently active in forcing the whole class downwards’.13 Marx
15 described the antagonism between British and Irish workers, artifi-
16 cially created by the mass media of the ruling class, as ‘the secret of
17 the impotence of the English working class, despite their organiza-
18 tion’.14 As industrialization got under way in France, Germany and
19 Switzerland in the latter half of the 19th century, these countries too
20 brought in foreign labour: from Poland, Italy and Spain. There were
21 800,000 foreign workers in the German Reich in 1907. More than a
22 third of the Ruhr miners were Poles. Switzerland had half a million
23 foreigners in 1910 – 15 per cent of her total population. French heavy
24 industry was highly dependent on immigrant labour right up to the
25 Second World War. According to Lenin, one of the special features
26 of imperialism was ‘the decline in emigration from imperialist coun-
27 tries and the increase in immigration into these countries from the
28 more backward countries where lower wages are paid’.15 This was
29 a main cause of the division of the working class. The fascist form
30 of capitalism also developed its own specific form of exploiting im-
31 migrant workers: the use of forced labour. No less than 71/2 million
32 deportees from occupied countries and prisoners of war were work-
33 ing in Germany by 1944, replacing the men recruited for the army.
34 About a quarter of German munitions production was carried out by
35 foreign labour.16
36 Compared with early patterns, immigration of workers to con-
37 temporary West Europe has two new features. The first is its charac-
38 ter as a permanent part of the economic structure. Previously, immi-
39 grant labour was used more or less temporarily when the domestic
40 industrial reserve army was inadequate for some special reason, like
41 war or unusually fast expansion; since 1945, however, large numbers
42 of immigrant workers have taken up key positions in the productive
43 process, so that even in the case of recession their labour cannot be

dispensed with. The second is its importance as the basis of the mod- 1
ern industrial reserve army. Other groups which might conceivably 2
fulfil the same function, non-working women, the disabled and the 3
chronic sick, members of the lumpenproletariat whose conditions 4
prevent them from working,17 have already been integrated into the 5
production process to the extent to which this is profitable for the 6
capitalist system. The use of further reserves of this type would re- 7
quire costly social measures (e.g. adequate kindergartens). The main 8
traditional form of the industrial reserve army – men thrown out of 9
work by rationalization and cyclical crises – is hardly available today, 10
for reasons already mentioned. Thus immigration is of key impor- 11
tance for the capitalist system. 12
13
14
The development of immigration since 1945 15
16
There are around eleven million immigrants18 living in West Europe, 17
making up about 5 per cent of the total population. Relatively few have 18
gone to industrially less developed countries like Norway, Austria 19
and Denmark, while large concentrations are to be found in high- 20
ly industrialized countries like Belgium, Sweden, West Germany, 21
France, Switzerland and Britain. Our analysis concentrates on the 22
four last-named which have about 90 per cent of all immigrants in 23
West Europe between them. 24
25
Immigrants in West Germany, France, Switzerland and Britain19 26
27
Immigrants Immigrants as Date of figures 28
(thousands) percentage of (latest available) 29
total population
30
West Germany 2,977 4.8 September 1970 31
France 3,177 6.4 December 1969 32
33
Switzerland 972 16.0 December 1969
34
Britain 2,603 5.0 1966 35
36
Most immigrants in Germany and Switzerland come from Southern 37
Europe. The main groups in Germany are Italians (574,000 in 38
1970), Yugoslavs (515,000), Turks (469,000), Greeks (343,000) 39
and Spaniards (246,000). In Switzerland, the Italians are by far the 40
largest group (532,000 in 1969) followed by Germans (116,000) 41
and Spaniards (98,000). France and Britain also have considerable 42
numbers of European immigrants, but in addition large contingents 43

1 from former colonies in Africa, Asia and the Caribbean. France has
2 617,000 Spaniards, 612,000 Italians, 480,000 Portuguese, as well as
3 608,000 Algerians, 143,000. Moroccans, 89,000 Tunisians, about
4 55,000 black Africans and an unknown number (probably about
5 200,000) from the remaining colonies (euphemistically referred to
6 as Overseas Departments) in the West Indies and the African island
7 of Réunion. The largest immigrant group in Britain comes from
8 the Irish Republic (739,000 in 1966). Most of the other Europeans
9 were displaced persons and the like who came during and after the
10 war: Germans (142,000), Poles (118,000). Cypriots number 60,000.
11 There are also an increasing number of South Europeans, often al-
12 lowed in on a short-term basis for work in catering and domestic
13 service. Coloured immigrants comprise about one third of the total,
14 the largest groups coming from the West Indies (269,000 in 1966),
15 India (240,000) and Pakistan (75,000).20
16 The migratory movements and the government policies which
17 changing function direct them reflect the growing importance and changing function
18 of immigrant labour of immigrant labour in West Europe. Immediately after the Second
19 World War, Switzerland, Britain and France recruited foreign work-
20 ers. Switzerland needed extra labour for the export boom permit-
21 ted by her intact industry in the middle of war-torn Europe. The
22 ‘European Voluntary Workers’ in Britain (initially displaced persons,
23 later Italians) were assigned to specific jobs connected with indus-
24 trial reconstruction. The reconstruction boom was not expected to
25 last. Both Switzerland and Britain imposed severe restrictions on
26 foreign workers, designed to stop them from settling and bringing
27 in their families, so that they could be dismissed and deported at the
28 least sign of recession. France was something of an exception: her
29 immigration policy was concerned not only with labour needs for
30 reconstruction, but also with permanent immigration to counteract
31 the demographic effects of the low birth-rate.
32 When West German industry got under way again after the 1949
33 Currency Reform there was at first no need for immigrants from
34 Southern Europe. An excellent industrial reserve army was provided
35 by the seven million expellees from the former Eastern provinces
36 of the Reich and by the three million refugees from East Germany,
37 many of whom were skilled workers. Throughout the fifties, the
38 presence of these reserves kept wage-growth slow and hence pro-
39 vided the basis for the ‘economic miracle’. By the mid-fifties, how-
40 ever, special labour shortages were appearing, first in agriculture and
41 building. It was then that recruitment of foreign workers (initially
42 on a seasonal basis21) was started. Here too, an extremely restrictive
43 policy was followed with regard to family entry and long-term settle-

ment. ‘Rotation’ of the foreign labour force was encouraged. In this 1

stage, the use of immigrants in the countries mentioned followed 2
the pre-war pattern: they were brought in to satisfy special and, it was 3
thought, temporary labour needs in certain sectors. They were, as an 4
official of the German employers’ association put it, ‘a mobile labour mobile labour 5
potential’.22 potential 6
By the sixties, the situation was changing. Despite mild cyclical 7
tendencies it was clear that there was not going to be a sudden re- 8
turn to the pre-war boom-slump pattern. The number of immigrant 9
workers grew extremely rapidly in the late fifties and early sixties. 10
Between 1956 and 1965 nearly one million new workers entered 11
France. The number of foreign workers in West Germany increased 12
from 279,000 in 1960 to over 1.3 million in 1966. In Switzerland 13
there were 326,000 immigrant workers (including seasonals) in 14
1956, and 721,000 in 1964. This was also the period of mass im- 15
migration to Britain from the Commonwealth.23 The change was not 16
merely quantitative: immigrants were moving into and becoming in- 17
dispensable in ever more sectors of the economy. They were no lon- 18
ger filling gaps in peripheral branches like agriculture and building 19
but were becoming a vital part of the labour force in key industries 20
like engineering and chemicals. Moreover, there was growing com- 21
petition between the different countries to obtain the ‘most desirable’ 22
immigrants, i.e. those with the best education and the least cultural 23
distance from the receiving countries. The growing need for labour 24
was forcing the recruiters to go further and further afield: Turkey and 25
Yugoslavia were replacing Italy as Germany’s main labour source. 26
Portugal and North Africa were replacing Italy and Spain in the case 27
of France. 28
As a result, new policies intended to attract and integrate im- 29
migrant workers, but also to control them better, were introduced. 30
One such measure was the free labour movement policy of the EEC, 31
designed to increase the availability of the rural proletariat of Sicily 32
and the Mezzogiorno to West European capital.24 Germany and 33
Switzerland liberalized the conditions for family entry and long- 34
term settlement, while at the same time tightening political control 35
through measures such as the German 1965 Foreigners Law. France 36
tried to increase control over entries, in order to prevent the large- 37
scale clandestine immigration which had taken place throughout 38
the fifties and sixties (and still does, despite the new policy). At the 39
same time restrictions were made on the permanent settlement of 40
non-Europeans – officially because of their ‘greater difficulties in 41
integrating’. In Britain, racialist campaigns led to the stopping of 42
unrestricted Commonwealth immigration in 1962. By limiting the 43

1 labour supply, this measure contradicted the economic interests of

2 the ruling class. The new Immigration Act of 1971, which could pro-
3 vide the basis for organized and controlled labour recruitment on the
4 German and French pattern, is a corrective, although its application
5 for this purpose is not at present required, since the ruling class has
6 created an internal industrial reserve army through unemployment.
7 In view of the stagnant domestic labour force potential and the
8 long-term growth trend of the economy, immigrant labour has be-
9 come a structural necessity for West European capitalism.25 It has
10 a dual function today.26 One section is maintained as a mobile fluc-
11 tuating labour force, which can be moved from factory to factory or
12 branch to branch as required by the development of the means of
13 production, and which can be thrown out of work and deported as
14 required without causing social tensions. This function was shown
15 clearly by the West German recession of 1966-7, when the foreign
16 labour force dropped by 400,000, although there were never more
17 than 29,000 receiving unemployment benefit. As a United Nations
18 study pointed out, West Germany was able to export unemployment
19 to the home countries of the migrants.27 The other section is required
20 permanent for permanent employment throughout the economy. They are of-
21 employment fered better conditions and the chance of long-term settlement.28
22 Despite this they still fulfil the function of an industrial reserve army,
23 for they are given inferior jobs, have no political rights and may be
24 used as a constant threat to the wages and conditions of the local
25 labour force.
26
27
28 Occupational position
29
30 The immigrant percentage of the population given in the table above
31 in no way reflects the contribution of immigrants to the economy.
32 They are mainly young men, whose dependents are sent for later if at
33 all. Many of them remain only a few years, and are then replaced by
34 others, so that there are hardly any retired immigrants. Immigrants
35 therefore have higher than average rates of economic activity, and
36 make contributions to health, unemployment and pension insurance
37 far in excess of their demands on such schemes.29 Particularly high
38 rates of activity are to be found among recently arrived groups, or
39 among those who for social and cultural reasons tend not to bring de-
40 pendents with them: Portuguese and North Africans in France, Turks
41 in Germany and Pakistanis in Britain. Immigrant workers are about
42 6.5 per cent of the labour force in Brirain, 7-8 per cent in France, 10
43 per cent in West Germany and 30 per cent in Switzerland. Even these

figures do not show adequately the structural importance of immi- 1

grant labour, which is concentrated in certain areas and types of work. 2
The overwhelming majority of immigrants live in highly indus- 3
trialized and fast-growing urban areas like Paris, the Lyon region, 4
the Ruhr, Baden-Württemberg, London and the West Midlands. For 5
example 31.2 per cent of all immigrants in France live in the Paris 6
region, compared with only 19.2 per cent of the total population. 9.5 7
per cent of the inhabitants of the Paris region are immigrants.30 In 8
Britain more than one third of all immigrants are to be found in 9
Greater London compared with one sixth of the total population. 10
Immigrants make up 12 per cent of London’s population.31 11

More important still is the concentration in certain industries. 12
Switzerland is the extreme case: the whole industrial sector is domi- 13
nated by foreign workers who make up more than 40 per cent of the 14
factory labour force. In many branches – for instance textiles, cloth- 15
ing, building and catering – they outnumber Swiss employees.32 Of 16
the nearly two million foreign workers in Germany in September 17
1970, 38.5 per cent were in the metal-producing and engineering in- 18
dustry, 24.2 in other manufacturing branches and 16.7 per cent in 19
building. Foreign workers accounted for 13.7 per cent of total em- 20
ployment in metal producing and engineering. The proportion was 21
even higher in some industries with particularly bad working condi- 22
tions, like plastic, rubber and asbestos manufacture (18.4 per cent). 23
In building, foreign workers were 17.5 per cent of the labour force. 24
On the other hand they made up only 3-4 per cent of all employees 25
in the services, although their share was much higher in catering 26
(14.8 per cent).33 Similar concentrations were revealed by the 1968 27
Census in France: 35.6 per cent of immigrant men were employed in 28
building and 13.5 per cent in engineering and electrical goods. 28.8 29
per cent of foreign women were domestic servants. In Britain the 30
concentration of immigrants in certain industries is less marked, 31
and different immigrant groups have varying patterns. The Irish 32
are concentrated in construction, while Commonwealth immigrants 33
are over-represented in metal manufacture and transport. Pakistani 34
men are mainly to be found in the textile industry and Cypriots in 35
clothing and footwear and in distribution. European immigrants are 36
frequently in the services sector. Immigrant women of all nationali- 37
ties tend to work in services, although some groups (Cypriots, West 38
Indians) also often work in manufacruring.34 39
In general immigrants are concentrated in certain basic indus- concentration in 40
tries, where they form a high proportion of the labour force. Together basic industries 41
with their geographical concentration this means that immigrant 42
workers are of great importance in the very type of enterprise and 43

1 area which used to be regarded as the strongholds of the class-con-

2 scious proletariat. The real concentration is even greater than the
3 figures show, for within each industry the immigrants tend to have
4 become predominant in certain departments and occupations. There
5 can be hardly a foundry in West Europe in which immigrants do not
6 form a majority, or at least a high proportion, of the labour force.
7 The same applies to monotonous production line work, such as car-
8 assembly. Renault, Citroen, Volkswagen, Ford of Cologne and Opel
9 all have mainly foreign workers on the assembly line (the British mo-
10 tor industry is an exception in this respect).
11 socio-economic Perhaps the best indication of the occupational concentration of
12 distribution the immigrant labour force is given by their socio-economic distribu-
13 tion. For instance a survey carried out in 1968 in Germany showed
14 that virtually no Southern Europeans are in non-manual employ-
15 ment. Only between 7 per cent and 16 per cent of the various nation-
16 alities were skilled workers while between 80 per cent and 90 per
17 cent were either semi-skilled or unskilled.35 By comparison about a
18 third of German workers are non-manual, and among manual work-
19 ers between one third and one half are in the skilled category in the
20 various industries. In France a survey carried out at Lyon in 1967
21 found that where they worked in the same industry, the French were
22 mainly in managerial, non-manual or skilled occupations, while the
23 immigrants were concentrated in manual occupations, particularly
24 semi-skilled and unskilled ones. The relegation to unskilled jobs is
25 particularly marked for North Africans and Portuguese.36 In Britain,
26 only about 26 per cent of the total labour force fall into the unskilled
27 and semi-skilled manual categories, but the figure is 42 per cent for
28 the Irish, 50 per cent for the Jamaicans, 65 per cent for the Pakistanis
29 and 55 per cent for the Italians.37
30
Immigrants form the lowest stratum of the working class carry-
31 ing out unskilled and semi-skilled work in those industrial sectors
32 with the worst working conditions and/or the lowest pay.38 The entry
33 of immigrants at the bottom of the labour market has made possible
34 the release of many indigenous workers from such employment,
35 and their promotion to jobs with better conditions and higher status,
36 i.e. skilled, supervisory or white-collar employment. Apart from the
37 economic effects, this process has a profound impact on the class
38 consciousness of the indigenous workers concerned. This will be dis-
39 cussed in more detail below.
40
41
42
43

Social position 1
2
The division of the working class within the production process is 3
duplicated by a division in other spheres of society. The poor living inferior conditions 4
conditions of immigrants have attracted too much liberal indigna- 5
tion and welfare zeal to need much description here. Immigrants get 6
the worst types of housing: in Britain slums and run-down lodging 7
houses, in France bidonvilles (shanty-towns) and overcrowded hotels, 8
in Germany and Switzerland camps of wooden huts belonging to 9
the employers and attics in the cities. It is rare for immigrants to get 10
council houses. Immigrants are discriminated against by many land- 11
lords, so that those who do specialize in housing them can charge 12
extortionate rents for inadequate facilities. In Germany and France, 13
official programmes have been established to provide hostel accom- 14
modation for single immigrant workers. These hostels do provide 15
somewhat better material conditions. On the other hand they in- 16
crease the segregation of immigrant workers from the rest of the 17
working class, deny them any private life, and above all put them 18
under the control of the employers 24 hours a day.39 In Germany 19
the employers have repeatedly attempted to use control over immi- 20
grants’ accommodation to force them to act as strike-breakers. 21
Language and vocational training courses for immigrant workers 22
are generally provided only when it is absolutely necessary for the 23
production process, as in mines for example. Immigrant children 24
are also at a disadvantage: they tend to live in run-down overcrowded 25
areas where school facilities are poorest. No adequate measures are 26
taken to deal with their special educational problems (e.g. language 27
difficulties), so that their educational performance is usually below- 28
average. As a result of their bad working and living conditions, im- 29
migrants have serious health problems. For instance they have much 30
higher tuberculosis rates than the rest of the population virtually ev- 31
erywhere.40 As there are health controls at the borders, it is clear that 32
such illnesses have been contracted in West Europe rather than be- 33
ing brought in by the immigrants. 34
The inferior work-situation and living conditions of immigrants 35
have caused some bourgeois sociologists to define them as a ‘lumpen- 36
proletariat’ or a ‘marginal group’. This is clearly incorrect. A group 37
which makes up 10, 20 or 30 per cent of the industrial labour force 38
cannot be regarded as marginal to society. Others speak of a ‘new 39
proletariat’ or a ‘sub-proletariat’. Such terms are also wrong. The first 40
implies that the indigenous workers have ceased to be proletarians 41
and have been replaced by the immigrants in this social position. The 42
second postulates that immigrant workers have a different relation- 43

1 ship to the means of production than that traditionally characteristic

2 of the proletariat. In reality both indigenous and immigrant work-
3 ers share the same relationship to the means of production: they are
4 excluded from ownership or control; they are forced to sell their la-
5 bour power in order to survive; they work under the direction and in
6 the interests of others. In the sphere of consumption both categories
7 of workers are subject to the laws of the commodity market, where
8 the supply and price of goods is determined not by their use value
9 but by their profitability for capitalists; both are victims of landlords,
10 retail monopolists and similar bloodsuckers and manipulators of
11 the consumption-terror. These are the characteristics typical of the
12 proletariat ever since the industrial revolution, and on this basis im-
13 division of the migrant and indigenous workers must be regarded as members of
14 proletariat the same class: the proletariat. But it is a divided class: the marginal
15 privileges conceded to indigenous workers and the particularly inten-
16 sive exploitation of immigrants combine to create a barrier between
17 the two groups, which appear as distinct strata within the class. The
18 division is deepened by certain legal, political and psychological fac-
19 tors, which will be discussed below.
20
21
22 Discrimination
23
24 Upon arrival in West Europe, immigrants from under-developed ar-
25 eas have little basic education or vocational training, and are usually
26 ignorant of the language. They know nothing of prevailing market
27 conditions or prices. In capitalist society, these characteristics are
28 sufficient to ensure that immigrants get poor jobs and social condi-
29 tions. After a period of adaptation to industrial work and urban life,
30 the prevailing ideology would lead one to expect many immigrants
31 to obtain better jobs, housing, etc. Special mechanisms ensure that
32 this does not happen in the majority of cases. On the one hand there
33 is institutionalized discrimination in the form of legislation which
34 restricts immigrants’ civic and labour market rights. On the other
35 hand there are informal discriminatory practices based on racialism
36 or xenophobia.
37 institutionalized In nearly all West European countries, labour market legislation
38 discrimination discriminates against foreigners. They are granted labour permits for
39 a specific job in a certain firm for a limited period. They do not have
40 the right to move to better-paid or more highly qualified positions,
41 at least for some years. Workers who change jobs without permis-
42 sion are often deported. Administrative practices in this respect have
43 been liberalized to some extent in Germany and Switzerland in re-

cent years, due to the need for immigrant labour in a wider range 1
of occupations, but the basic restrictiveness of the system remains. 2
In Britain, Commonwealth immigrants (once admitted to the coun- 3
try) and the Irish had equal rights with local workers until the 1971 4
Immigration Act. Now Commonwealth immigrants will have the 5
same labour market situation as aliens. The threat of deportation if 6
an immigrant loses his job is a very powerful weapon for the em- 7
ployer. Immigrants who demand better conditions can be sacked for 8
indiscipline and the police will do the rest.41 Regulations which re- 9
strict family entry and permanent settlement also keep immigrants 10
in inferior positions. If a man may stay only for a few years, it is not 11
worth his while to learn the language and take vocational training 12
courses. 13
Informal discrimination is well known in Britain, where it takes informal 14
the form of the colour bar. The PEP study,42 as well as many other discrimination 15
investigations, has shown that coloured immigrants encounter dis- 16
crimination with regard to employment, housing and the provision 17
of services such as mortgages and insurance. The more qualified 18
a coloured man is, the more likely he is to encounter discrimina- 19
tion. This mechanism keeps immigrants in ‘their place’, i.e. doing 20
the dirty, unpleasant jobs. Immigrants in the other European coun- 21
tries also encounter informal discrimination. Immigrants rarely get 22
promotion to supervisory or non-manual jobs, even when they are 23
well-qualified. Discrimination in housing is widespread. In Britain, 24
adverts specifying ‘no coloured’ are forbidden, but in Germany or 25
Switzerland one still frequently sees ‘no foreigners’. 26
The most serious form of discrimination against immigrant 27
workers is their deprivation of political rights. Foreigners may not 28
vote in local or national elections. Nor may they hold public office, 29
which in France is defined so widely as to include trade-union posts. 30
Foreigners do not generally have the same rights as local workers 31
with regard to eligibility for works councils and similar representa- 32
tive bodies. The main exception to this formal exclusion from politi- 33
cal participation concerns Irish and Commonwealth immigrants in 34
Britain, who do have the right to vote (the same will not apply to those 35
who enter under the 1971 Act). But the Mangrove case shows the type 36
of repression which may be expected by any immigrants who dare to 37
organize themselves. Close police control over the political activities 38
of immigrants is the rule throughout Europe, and deportations of po- 39
litical and trade-union militants are common. After the May Events 40
in France, hundreds of foreign workers were deported.43 Foreign lan- 41
guage newspapers of the CGT labour federation have been repeated- 42
ly forbidden. The German Foreigners Law of 1965 lays down that the 43

1 political activity of foreigners can be forbidden if ‘important interests

2 of the German Federal Republic require this’ – a provision so flexible
3 that the police can prevent any activity they choose. Even this is not
4 regarded as sufficient. When Federal Chancellor Willy Brandt visited
5 Iran in March 1972 to do an oil deal, the Shah complained strongly
6 about Iranian students being allowed to criticize him in Germany.
7 The Greek and Yugoslav ambassadors have also protested about the
8 activities of their citizens. Now the German Government is working
9 on a new law which would go so far as to make police permission
10 necessary even for private meetings of foreigners in closed rooms.44
11
12
13 Prejudice and class consciousness
14
15 Discrimination against immigrants is a reflection of widespread hos-
16 tility towards them. In Britain, this is regarded as ‘colour prejudice’
17 or ‘racialism’, and indeed there can be no doubt that the hostility of
18 large sections of the population is at present directed against black
19 people. Race relations theorists attribute the problems connected
20 with immigration partly to the immigrants’ difficulties in adapting
21 to the prevailing norms of the ‘host society’, and partly to the in-
22 digenous population’s inbred distrust of the newcomers who can be
23 distinguished by their skin colour. The problems are abstracted from
24 the socioeconomic structure and reduced to the level of attitudes.
25 Solutions are to be sought not through political action, but through
26 psychological and educational strategies.45 But a comparison of sur-
27 veys carried out in different countries shows that hostility towards
28 immigrants is everywhere as great as in Britain, even where the im-
29 migrants are white.46 The Italian who moves to the neighbouring
30 country of Switzerland is as unpopular as the Asian in Britain. This
31 indicates that hostility is based on the position of immigrants in so-
32 ciety and not on the colour of their skin.
33 racialism and Racialism and xenophobia are products of the capitalist nation-
34 xenophobia al state and of its imperialist expansion.47 Their principal historical
35 function was to split the working class on the international level, and
36 to motivate one section to help exploit another in the interests of the
37 ruling class. Today such ideologies help to deepen the split within
38 the working class in West Europe. Many indigenous workers do not
39 perceive that they share a common class position and class interests
40 with immigrant workers. The basic fact of having the same relation-
41 ship to the means of production is obscured by the local workers’
42 marginal advantages with regard to material conditions and status.
43 The immigrants are regarded not as class comrades, but as alien in-

truders who pose an economic and social threat. It is feared that they 1
will take away the jobs of local labour, that they will be used by the 2
employers to force down wages and to break strikes.48 Whatever the 3
behaviour of the immigrant workers – and in fact they almost invari- 4
ably show solidarity with their indigenous colleagues – such fears 5
are not without a basis. It is indeed the strategy of the employers to 6
use immigration to put pressure on wages and to weaken the labour 7
movement.49 The very social and legal weakness of the immigrants is 8
a weapon in the hands of the employers. Other points of competition 9
are to be found outside work, particularly on the housing market. 10
The presence of immigrants is often regarded as the cause of ris- 11
ing rents and increased overcrowding in the cities. By making im- 12
migrants the scapegoats for the insecurity and inadequate conditions 13
which the capitalist system inevitably provides for workers, attention 14
is diverted from the real causes. 15
Workers often adopt racialism as a defence mechanism against a racialism 16
real or apparent threat to their conditions. It is an incorrect response defence mechanism 17
to a real problem. By preventing working-class unity, racialism as- 18
sists the capitalists in their strategy of ‘divide and rule’. The function 19
of racialism in the capitalist system is often obscured by the fact that 20
racialist campaigns usually have petty-bourgeois leadership and di- 21
rect their slogans against the big industrialists. The Schwarzenbach 22
Initiative in Switzerland – which called for the deportation of a large 23
proportion of the immigrant population – is an example,50 as are 24
Enoch Powell’s campaigns for repatriation. Such demands are op- 25
posed by the dominant sections of the ruling class. The reason is 26
clear: a complete acceptance of racialism would prevent the use of 27
immigrants as an industrial reserve army. But despite this, racial- 28
ist campaigns serve the interests of the ruling class: they increase 29
tension between indigenous and immigrant workers and weaken 30
the labour movement. The large working-class following gained by 31
Powell in his racialist campaigns demonstrates how dangerous they 32
are. Paradoxically, their value for capitalism lies in their very failure 33
to achieve their declared aims. 34
The presence of immigrant workers is one of the principal fac- 35
tors contributing to the lack of class consciousness among large sec- 36
tions of the working class. The existence of a new lower stratum of 37
immigrants changes the worker’s perception of his own position in 38
society. Instead of a dichotomic view of society, in which the working 39
masses confront a small capitalist ruling class, many workers now 40
see themselves as belonging to an intermediate stratum, superior to 41
the unskilled immigrant workers. Such a consciousness is typified by 42
an hierarchical view of society and by orientation towards advance- 43

1 ment through individual achievement and competition, rather than

2 through solidarity and collective action. This is the mentality of the
3 labour aristocracy and leads to opportunism and the temporary decay
4 of the working-class movement.
5
6
7 Immigration and society
8
9 The impact of immigration on contemporary West European society
10 may now be summarized.
11
12 effects of Economic effects: the new industrial reserve army of immigrant work-
13 immigration ers is a major stabilizing factor of the capitalist economy. By restrain-
14 ing wage increases, immigration is a vital precondition for capital ac-
15 cumulation and hence for growth. In the long run, wages may grow
16 more in a country which has large-scale immigration than in one
17 which does not, because of the dynamic effect of increased capital
18 accumulation on productivity. However, wages are a smaller share,
19 and profits a larger share of national income than would have been
20 the case without immigration.51 The best illustration of this effect is
21 obtained by comparing the German and the British economies since
22 1945. Germany has had large and continuous increases in labour
23 force due to immigration. At first wages were held back. The result-
24 ing capital accumulation allowed fast growth and continuous ratio-
25 nalization. Britain has had virtually no growth in labour force due
26 to migration (immigration has been cancelled out by emigration of
27 British people to Australia, etc). Every phase of expansion has col-
28 lapsed rapidly as wages rose due to labour shortages. The long-term
29 effect has been stagnation. By the sixties, German wages overtook
30 those of Britain, while economic growth and rationalization contin-
31 ued at an almost undiminished rate.
32 Social effects: The inferior position of immigrant workers with re-
33 gard to employment and social conditions has led to a division of
34 the working class into two strata. The split is maintained by various
35 forms of discrimination and is reinforced by racialist and xenophobic
36 ideologies, which the ruling class can disseminate widely through
37 its hegemony over the means of socialization and communication.
38 Large sections of the indigenous workers take the position of a la-
39 bour aristocracy, which objectively participates in the exploitation of
40 another group of workers.
41 Political effects: the decline of class consciousness weakens the
42 working-class movement. In addition, the denial of political rights
43 to immigrants excludes a large section of the working class from po-

litical activity, and hence weakens the class as a whole. The most 1
exploited section of the working class is rendered voiceless and pow- 2
erless. Special forms of repression are designed to keep it that way. 3
4
5
Working-class movement and immigrant labour 6
7
Immigrant labour has an important function for contemporary West 8
European capitalism. This does not mean, however, that socialists 9
should oppose labour migration as such. To do so would be incorrect 10
for two reasons. Firstly, it would contradict the principle of proletar- 11
ian internationalism, which rejects the maintenance of privileges for 12
one section of the working class at the expense of another. Secondly, 13
opposition to immigration would cause immigrants in West Europe 14
to regard the working-class movement as its enemy, and would 15
therefore deepen the split in the working class – which is exactly 16
what the capitalists are hoping for. The aim of a socialist policy on 17
immigration must be to overcome the split in the working class by 18
bringing immigrant workers into the labour movement and fight- 19
ing against the exploitation to which they are subjected. Only by de- 20
manding full economic, social and political equality for immigrants 21
can we prevent the employers from using them as a weapon against 22
working-class interests. 23
The policies of the trade unions with regard to immigration have opposition and 24
varied widely. The Swiss unions oppose immigration, and have since acceptance 25
the mid-fifties campaigned for a reduction in the number of foreign 26
workers. At the same time, they claim to represent all workers, and 27
call upon foreigners to join – not surprisingly, with little success. 28
The British unions opposed the recruitment of European Voluntary 29
Workers after the war, and insisted upon collective agreements lim- 30
iting their rights to promotion, laying down that they should be dis- 31
missed first in case of redundancy and so on.52 The policy towards 32
Commonwealth immigration has been totally different: the TUC has 33
opposed immigration control, and rejected any form of discrimina- 34
tion. This rejection has, however, been purely verbal, and virtually 35
nothing has been done to organize immigrants or to counter the 36
special forms of exploitation to which they are subject. The CGT in 37
France opposed immigration completely during the late forties and 38
the fifties, condemning it as an instrument designed to attack French 39
workers’ conditions. More recently the CGT, as well as the two other 40
big labour federations, the CFDT and the FO, have come to regard 41
immigration as inevitable. All have special secretariats to deal with 42
immigrant workers’ problems and do everything possible to bring 43

1 them into the unions. In Germany, the DGB has accepted immigra-
2 tion and has set up offices to advise and help immigrants. The mem-
3 ber unions also have advisory services, and provide foreign language
4 bulletins and special training for immigrant shop-stewards. In gen-
5 eral, those unions which have recognized the special problems of im-
6 migration have not done so on the basis of a class analysis (here the
7 CGT is to some extent an exception). Rather they have seen the prob-
8 lems on a humanitarian level, they have failed to explain the strategy
9 of the employers to the workers, and the measures taken have been
10 of a welfare type, designed to integrate immigrants socially, rather
11 than to bring them into the class struggle.
12 Therefore, the unions have succeeded neither in countering ra-
13 cialism among indigenous workers, nor in bringing the immigrant
14 workers into the labour movement on a large scale. The participation
15 of immigrant workers in the unions is on the whole relatively low. This
16 is partly attributable to their rural background and lack of industrial
17 experience, but in addition immigrants often find that the unions do
18 not adequately represent their interests. The unions are controlled by
19 indigenous workers, or by functionaries originating from this group.
20 In situations where immigrant and indigenous workers do not have
21 the same immediate interests (this happens not infrequently due to
22 the differing occupational positions of the two groups, for instance
23 in the question of wage-differentials), the unions tend to take the
24 side of the indigenous workers. Where immigrants have taken ac-
25 tion against special forms of discrimination, they have often found
26 themselves deserted by the unions.53 In such circumstances it is not
27 surprising if immigrants do not join the unions, which they regard
28 as organizations for local labour only. This leads to a considerable
29 weakening of the unions. In Switzerland many unions fear for their
30 very existence, and see the only solution in the introduction of com-
31 pulsory ‘solidarity contributions’, to be deducted from wages by the
32 employers. In return the unions claim to be the most effective instru-
33 ment for disciplining the workers. When the employers gave way to
34 a militant strike of Spanish workers in Geneva in 1970, the unions
35 publicly attacked them for making concessions.
36 the formation of Where the unions do not adequately represent immigrant work-
37 immigrant unions ers, it is sometimes suggested that the immigrants should form their
38 own unions. In fact they have not done so anywhere in contemporary
39 West Europe. This shows a correct class position on their part: the
40 formation of immigrant unions would deepen and institutionalize
41 the split in the working class, and would therefore serve the interests
42 of the employers.54 On the other hand, all immigrant groups do have
43 their own organizations, usually set up on the basis of nationality,

and having social, cultural and political functions. These organiza- 1

tions do not compete with the trade unions, but rather encourage 2
their members to join them. The aim of the political groups have 3
so far been concerned mainly with their countries of origin. They 4
have recruited and trained cadres to combat the reactionary regimes 5
upon returning home. At present, as a result of greater length of stay 6
and increasing problems in West Europe, many immigrant political 7
groups are turning their attention to class struggle in the countries 8
where they work. 9
It is the task of the revolutionary movement in West Europe to revolutionary 10
encourage this tendency, by making contact with immigrant groups, movement 11
assisting them in co-ordinating with immigrants of other nationali- 12
ties and with the working-class movement in general, giving help in 13
political education and cadre-training, and carrying out joint actions. 14
Such co-operation means surmounting many problems. Firstly, lan- 15
guage and culture may make communications difficult. Secondly, 16
the risk of repression to which immigrant militants are exposed may 17
make them reluctant to make contacts. Thirdly, the experience of dis- 18
crimination may cause immigrants to distrust all local people. This 19
leads in many cases to cultural nationalism, particularly marked for 20
historical reasons among black people. In order to overcome these 21
difficulties, it is essential for indigenous political groups to study the 22
problems of immigrants and the special forms of discrimination and 23
exploitation to which they are exposed. Concrete attempts to combat 24
these must be made. Indigenous groups must offer co-operation and 25
assistance to immigrants in their struggle, rather than offering them- 26
selves as a leadership. 27
It is not only when revolutionary groups are actively trying to co- 28
operate with immigrant workers’ organizations that they come up 29
against the problems of immigration. The majority of immigrants 30
are not politically organized, whether through apathy or fear of re- 31
pression. Groups agitating in factories or carrying out rent cam- 32
paigns are likely to come up against large numbers of unorganized 33
immigrants in the course of their daily work. It is then essential to 34
take special steps to communicate with the immigrants and to bring 35
them into the general movement. Failure to do so may result in the 36
development of petty-bourgeois chauvinism within factory or hous- 37
ing groups, which would correspond precisely with the political aims 38
of the capitalists with regard to labour migration. In Germany, the 39
large numbers of revolutionary groups at present agitating in facto- 40
ries almost invariably find it necessary to learn about the background 41
and problems of immigrant workers, to develop special contacts with 42
them, and to issue leaflets in the appropriate languages. The same 43

1 applies to housing groups, which frequently find that immigrants

2 form the most under-privileged group in the urban areas where they
3 are working.
4 Immigrant workers can become a class-conscious and militant
5 section of the labour movement. This has been demonstrated repeat-
6 edly; immigrant workers have played a leading part in strike move-
7 ments throughout West Europe. They are at present in the forefront
8 of the movement which is occupying empty houses in German cities.
9 Immigrant workers showed complete solidarity with the rest of the
10 working class in May 1968 in France, they were militant in strikes
11 and demonstrations and developed spontaneous forms of organiza-
12 tion in the struggle.
13 pre-conditions for But such successes should not make us forget the capitalist strat-
14 class-consciousness egy behind labour migration. Powerful structural factors connected
15 and class struggle with the function of immigrants as an industrial reserve army, and
16 with the tendency of part of the indigenous working class to take on
17 the characteristics of a labour aristocracy, lead to a division between
18 immigrant and indigenous workers. Solidarity between these two
19 sections does not come automatically. It requires a correct under-
20 standing of the problems within the revolutionary movement and
21 a strategy for countering ruling-class aims. It is necessary to assist
22 the immigrant workers in fighting exploitation and in defending
23 their special interests. At the same time revolutionary groups must
24 combat racialist and xenophobic ideologies within the working class.
25 These are the pre-conditions for developing class-consciousness and
26 bringing the immigrant workers into the class struggle.
27
28
29 Notes
30
31 1 Engels, ‘The Condition of the Working Class in England’, in Marx and
32 Engels, On Britain, Moscow 1962, p. 119.
2 Marx, Capital, Vol. I, Moscow 1961, p. 632.
33 3 Ibid., p. 633.
34 4 Ibid., p. 637.
35 5 Ibid., p. 640.
36 6 Engels, Preface to the English edition of ‘The Condition of the Working
37 Class in England’, op. cit., p. 28.
7 Engels, ‘The English Elections’, in On Britain, op. cit., p. 505.
38
8 Lenin, Imperialism – the highest Stage of Capitalism, Moscow 1966, pp. 96-7.
39 9 Ibid., pp. 99-100.
40 10 In this article we examine the function of labour migration only for the
41 countries of immigration. Migration also plays an important stabilizing role
42 for the reactionary regimes of the countries of origin – a role which is un-
43 derstood and to some extent planned by the ruling class in West Europe.

Although we are concerned only with West Eucope in this article, it is im- 1
portant to note that the use of certain special categories of workers, who 2
can be discriminated against without arousing general solidarity from other 3
workers, is a general feature of modern capitalism. The blacks and chicanos
are the industrial reserve army of the USA, the Africans of white-dominated
4
Southern Africa. Current attempts by ‘liberal’ capitalists to relax the colour 5
bar to allow blacks into certain skilled and white-collar jobs, both in the USA 6
and South Africa, however estimable in humanitarian terms, are designed 7
mainly to weaken the unions and put pressure on wages in these sectors. 8
11 Marx mentions several forms taken by the industrial reserve army. One is
9
the ‘latent’ surplus-population of agricultural labourers, whose wages and
conditions have been depressed to such an extent that they are merely wait- 10
ing for a favourable opportunity to move into industry and join the urban 11
proletariat. (Capital, Vol. I., op. cit., p. 642.) Although these workers are not 12
yet in industry, the possibility that they may at any time join the industrial 13
labour force increases the capitalist’s ability to resist wage increases. The 14
latent industrial reserve army has the same effect as the urban unemployed.
Unemployed workers in other countries, in so far as they may be brought
15
into the industrial labour force whenever required, clearly form a latent in- 16
dustrial reserve army in the same way as rural unemployed within the coun- 17
try. 18
12 See E. P. Thompson, The Making of the English Working Class, Harmondsworrh 19
1968, pp. 469-85.
20
13 ‘The Condition of the Working Class in England’, op. cit., p. 123.
14 Letter to S. Meyer and A. Vogt, 9 April 1870, in On Britain, op. cit., p. 552. 21
15 Imperialism, op. cit., p. 98. 22
16 Hans Pfahlmann, Fremdarbeiter und Kriegsgefangene in der deutschen 23
Kriegswirtschaft, 1939-1945, Darmstadt 1968, p. 232. 24
17 For the role of the lumpenproletariat in the industrial reserve army, see 25
Capital, Vol. I, op. cit., p. 643.
26
18 We use ‘immigrants’ in a broad sense to include all persons living in a West
European country which is not their country of birth. Much migration is of 27
a temporary nature, for a period of 3-10 years. But such temporary migration 28
has effects similar to permanent migration when the returning migrant is 29
replaced by a countrymen with similar characteristics. Such migrants may 30
be regarded as a permanent social group with rotating membership.
31
19 For sources, as well as a detailed analysis of social conditions of immigrants,
see Stephen Castles and Godula Kosack, Immigrant Workers and Class 32
Structure in Western Europe, London, Oxford University Press for Institute of 33
Race Relations, 1972 (forthcoming). 34
20 The 1966 Census figures are at present the most recent ones available. It 35
should, however, be noted that, for technical reasons, they seriously un- 36
der-enumerate the Commonwealth immigrants in Britain. Moreover, the
37
number has grown considerably since 1966, particularly if we look at the
whole community including children born to Commonwealth immigrants 38
in Britain, who were not counted by the census. We shall have to wait for 39
the results of the 1971 Census to obtain a more accurate picture of the im- 40
migrant population in Britain. 41
21 Many foreign workers are still employed on a seasonal basis in building,
42
agriculture and catering in France and Switzerland. This is a special form
43

1 of exploitation. The worker has no income in the off-season and is therefore

2 forced to work very long hours for the 9-10 months when he does have work.
3 He cannot bring his family with him, he has even more limited civic rights
than other immigrants, and he has absolutely no security, for there is no
4
guarantee that his employment will be continued from year to year.
5 22 Ulrich Freiherr von Gienanth, in Der Arbeitgeber, Vol. 18, 20 March 1966, p.
6 153.
7 23 For Commonwealth immigration see E.J.B. Rose et al., Colour and Citizenship,
8 London 1969.
24 Eurocrats refer to the free movement policy as the beginning of a ‘European
9
labour market’. But although EEC citizens have the right to choose which
10 country to be exploited in, they lack any civic or political rights once there.
11 Moreover, the Southern Italian labour reserves are being absorbed by the
12 monopolies of Turin and Milan, so that intra-EEC migration is steadily de-
13 clining in volume, while migration from outside the EEC increases.
14 25 Where formalized economic planning exists, this necessity has been publicly
formulated. Prognoses on the contribution of immigrants to the labour force
15
were included in the Fourth and Fifth Five-Year Plans in France, and play
16 an even more prominent part in the current Sixth Plan. See Le VIe plan et les
17 travailleurs étrangers, Paris 1971.
18 26 Cf. Ruth Becker, Gerhard Dörr, K.H. Tjaden, ‘Fremdarbeiterbeschäftigung
19 im deutschen Kapitalismus’, Das Argument, December 1971, p. 753.
27 United Nations Economic Commission for Europe, Economic Survey of
20
Europe 1967, Geneva 1968, Chapter I, p. 49.
21 28 The distinction between the two sections of the immigrant labour force
22 is formalized in the new French immigration policy introduced in 1968.
23 There are separate regulations for South Europeans, who are encouraged
24 to bring in their families and settle permanently, and Africans (particu-
25 larly Algerians) who are meant to come for a limited period only, without
dependents.
26
29 It is estimated that foreign workers in Germany are at present paying about
27 17 per cent of all contributions to pension insurance, but that foreign-
28 ers are receiving only 0.5 per cent of the total benefits. Heinz Salowsky,
29 ‘Sozialpolitische Aspekte der Auslanderbeschaftigung’, Berichte des Deutschen
30 Industrie instituts zur Sozialpolitik, Vol. 6 (8), No.2, February 1972, pp. 16-22.
30 Calculated from: ‘Statistiques du Ministère de l’Intérieur’, Hommes et
31
Migrations: Documents, No. 788, 15 May 1970; and Annuaire Statistique de la
32 France 1968.
33 31 1966 Census.
34 32 Statistisches Jahrbuch der Schweiz 1967, pp. 140-1.
35 33 Ausländische Arbeitnehmer 1970, Nürnberg 1971.
36 34 1966 Census. For a detailed analysis of immigrants’ employment see: K.
Jones and A.D. Smith, The Economic Impact of Commonwealth Immigration,
37
Cambridge 1970.
38 Also Immigrant Workers and Class Structure in Western Europe, op. cit., Ch.
39 III.
40 35 Ausländische Arbeitnehmer 1969, Nürnberg 1970, p. 86.
41 36 L’insertion sociale des étrangers dans l’aire métropolitaine Lyon-Saint-
42 Étienne’, Hommes et Migrations, No. 113, 1969, p. 112.
37 1966 Census.
43

38 Some employers – particularly small inefficient ones – specialize in the ex- 1

ploitation of immigrants. For instance they employ illegal immigrants, who 2
can be forced to work for very low wages and cannot complain to the authori- 3
ties for fear of deportation. Such cases often cause much indignation in the
4
liberal and social-democratic press. But, in fact, it is the big efficient firms
exploiting immigrants in a legal and relatively humane way which make the 5
biggest profits out of them. The function of immigration in West European 6
capitalism is created not by the malpractices of backward firms (many of 7
whom incidentally could not survive without immigrant labour), but by the 8
most advanced sectors of big industry which plan and utilize the position of
9
immigrant workers to their own advantage.
39 ‘So far as we are concerned, hostel and works represent parts of a single 10
whole. The hostels belong to the mines, so the foreign workers are in our 11
charge from start to finish’, stated a representative of the German min- 12
ing employers proudly. Magnet Bundesrepublik, Informationstagung der 13
Bundesvereinigung Deutscher Arbeitgeberverbande, Bonn 1966, p. 81. 14
40 A group of French doctors found that the TB rate for black Africans in the
15
Paris suburb of Montreuil was 156 times greater than that of the rest of
the local population. R.D. Nicoladze, C. Rendu, G. Millet, ‘Coupable d’être 16
malades’, Droit et Liberté, No. 280, March 1969, p. 8. For further examples 17
see Immigrant Workers and Class Structure in Western Europe, op. cit., Ch. 18
VIII. 19
41 For a description of how a strike of Spanish workers in a steel-works was
20
broken by the threat of deportation, see P. Gavi, Les Ouvriers, Paris 1970, pp.
225-6. 21
42 W. W. Daniels, Racial Discrimination in England, based on the PEP Report, 22
Harmondsworth 1968. 23
43 See Review of the International Commission of Jurists, No. 3, September 1969, 24
and Migration Today, No. 13, Autumn 1969. 25
44 Cf. Der Spiegel, No. 7, 7 February 1972.
26
45 See Mark Abrams’ study on prejudice in Colour and Citizenship, pp. 551-604.
The results of the study are very interesting, but require careful interpre- 27
tation. The interpretation given by Abrams is extremely misleading. The 28
results of the prejudice study, which was said to indicate a very low level 29
of prejudice in Britain, attracted more public attention than all the other 30
excellent contributions in this book. For a reanalysis of Abrams’ material see
31
Christopher Bagley, Social Structure and Prejudice in five English Boroughs,
London 1970. 32
46 We have attempted such a comparison in Immigrant Workers and Class 33
Structure in Western Europe, op. cit., Chapter IX. Historical comparisons also 34
tend to throw doubt on the importance of race as a cause of prejudice: white 35
immigrants like the Irish were in the past received just as hostilely as the 36
black immigrants today.
37
47 Oliver Cromwell Cox, Caste, Class and Race, New York 1970, p. 317 ff. This
superb work of Marxist scholarship is recommended to anyone interested in 38
racialism. 39
48 Surveys carried out in Germany in 1966 show a growth of hostility towards 40
immigrants. This was directly related to the impending recession and local 41
labour’s fear of unemployment. 42
43

1 49 Historically, the best example of this strategy was the use of successive waves
2 of immigrants to break the nascent labour movement in the USA and to
3 follow extremely rapid capital accumulation. The Jungle by Upton Sinclair
gives an excellent account of this. Similar was the use of internal migrants
4
(the ‘Okies’) in California in the thirties – see John Steinbeck, The Grapes of
5 Wrath.
6 50 Although the Federal Council, the Parliament, the employers, the unions
7 and all the major parties called for rejection of the Schwarzenbach Initiative,
8 it was defeated only by a small majority: 46 per cent of voters supported the
Initiative and 54 per cent voted against it.
9
51 Many bourgeois economists and some soi-disant Marxists think that im-
10 migration hinders growth because cheap labour reduces the incentive for
11 rationalization. Bourgeois economists may be excused for not knowing (or
12 not admitting) that cheap labour must be the source for the capital which
13 makes rationalization possible. Marxists ought to know it. A good study on
14 the economic impact of immigration is: C.P. Kindleberger, Europe’s Postwar
Growth – the Role of Labour Supply, Cambridge (Mass.) 1967.
15
52 See Bob Hepple, Race, Jobs and the Law in Britain, London 1968, p. 50 and
16 Appendix II.
17 53 For details of such cases see Immigrant Workers and Class Structure in Western
18 Europe, op. cit., Chapter IV.
19 54 We do not wish to imply that it is always incorrect for minority groups to
form new unions, if the existing ones are corrupt and racialist. It was obvi-
20
ously necessary for militant blacks in the USA to do this, as the existing
21 union structure was actively assisting in their oppression. But organizations
22 like the Detroit Revolutionary Union Movement (DRUM), though consist-
23 ing initially of blacks only, were not separatist. They had the perspective of
24 organizing class-conscious workers of all ethnic groups. Such organizations
25 appear to be neither necessary nor possible in the present stage of struggle
in West Europe.
26
27
28 For references please consult the bibliography of the book in which
29 this article was originally published. (see List of sources, page 609)
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Introduction to European immigration 3
4
policy: a comparative study
5
6
Tomas Hammar 7
8
9
10
11
12
This article is the introduction to the book European Immigration Policy: A
13
Comparative Study, edited by political scientist Tomas Hammar. At the time
14
of its publication in 1985, the construction of a European immigration policy 15
was not yet a hot policy issue. Moreover, the systematic comparative study 16
of immigration policies across the Continent was underdeveloped. In fact, 17
Hammar’s book was one of the first – and most convincing – collective at- 18
tempts to compare immigration patterns and policies in different European 19
countries. This is one valuable reason to include the introductory chapter of 20
the book in this volume even though it does not deal with all the member 21
states of the European Community at the time. Another reason is the utility 22
of Hammar’s analytical distinction between two related parts of immigration 23
policy – immigration regulation and aliens control versus immigrant policy 24
– which has become a classic distinction. After more than twenty years, it 25
remains an excellent point of entry into the study of immigration and integra- 26
tion policies in Europe. 27
28
29
The six immigration countries studied in this book have experienced 30
a period of large-scale immigration caused mainly by similar factors. 31
None of these countries had planned or even foreseen an internation- 32
al migration of the size that actually occurred. Their reaction to this similar experiences 33
migration has been strikingly similar and at the same time decisively 34
different, but in the long run immigration control has become more 35
strict everywhere and active labor recruitment has been stopped; at 36
the same time, there have been a number of improvements in the 37
social and cultural situation of immigrants. 38
39
40
41
42
43

46 tomas hammar
1 Selecting six countries

2
3 The project countries have been chosen partly because of their size
4 and their large immigrant populations and partly because they of-
5 fer a high degree of variation in the regulation of immigration and
6 representative in immigrant policy. Germany, France and Britain were included
7 countries from the outset because of their sizeable immigrant populations, and
8 Switzerland because of its high proportion of foreigners. In addition
9 these four countries provide examples of very different sorts of inter-
10 national migration as well as different immigration policies. Sweden
11 could not be left out, partly because the initiative and financing of
12 this study was Swedish, but more important it deserves a place as the
13 Scandinavian country which has both admitted the most immigrants
14 and developed first a specific immigration policy. The Netherlands
15 was included as the sixth country because of its mixture of post-co-
16 lonial and Mediterranean labor immigration, and also because of its
17 traditional emphasis on cultural pluralism and its influence on cur-
18 rent “ethnic minorities” policy.
19 The selection of countries was also made with the idea that the
20 two major ways of regulating immigration should be represented:
21 the “guestworker” or rotation system (Germany and Switzerland),
22 and the policy of permanent immigration (Britain and Sweden). The
23 post-colonial immigration that prevails in Britain and has played a
24 major role in France and in the Netherlands is included as well as
25 immigration to countries with no such colonial ties, represented by
26 Sweden, Germany, and Switzerland. We further hoped that our selec-
27 tion would give examples of various types of immigrant policy, based
28 on different welfare ideologies and on the different social and political
29 organizations of the societies represented.
30 Other immigration countries, of course, could have been included
31 as well, had the project resources not required that the number of
32 selected countries be limited. Norway and Denmark have both admit-
33 ted immigrants from, among other countries, Turkey and Pakistan,
34 and they offer interesting cases for policy comparison. Yet immigra-
35 tion to these two countries has been relatively small, and if only one
36 Scandinavian country can be included, Sweden is the logical choice.
37 Belgium had a large immigrant population of some 900,000 in
38 1980. The number of foreign citizens residing in Austria at the same
39 time was estimated to be about 250,000. Although both countries
40 have adopted policies directed towards the temporary employment of
41 foreign workers, they have found that their immigrants tend to stay
42 permanently. They would offer excellent additional studies, but their
43 exclusion does not significantly reduce the breadth of our study.

Since we study changes over several decades in immigration and 1

immigrant policy, we may claim that we cover more than six cases. 2
We are able to present data for each country emanating from differ- 3
ent time periods. The comparison of six national cases will improve 4
our knowledge about the preconditions of immigration policy, about 5
the interrelations between regulation of immigration and immigrant 6
policy, and in general about the dynamics of international migration 7
and national policymaking. 8
9
10
Migratory paths 11
12
Postwar migration to and within Europe has been characterized as 13
a movement from south to north, although such postwar migration movement from 14
would be better characterized as a movement from the periphery to periphery to center 15
the center. Migration from Italy reached considerable proportions in 16
the 1950s and was joined during the 1960s by an even larger migra- 17
tion from Spain and Portugal in the southwest and from Yugoslavia, 18
Greece, and later Turkey in the southeast. African migration has 19
gone mainly to France, while the bulk of transoceanic migration 20
from the West Indies, Pakistan, and India has gone to Britain. The 21
Netherlands has had immigration from Indonesia and Latin America 22
as well as from Morocco and Turkey. 23
On the map (Figure 1.1) two additional arrows from Ireland to 24
Britain and from Finland to Sweden reinforce the impression of a 25
movement from periphery to center. Nevertheless, although both ar- 26
rows show a movement across national boundaries, one is reluctant 27
to say that they represent “international migration” in the same sense 28
as do the other arrows. Irish immigrants have always been allowed to 29
enter Britain and seek employment without restriction. Until at least 30
1948 they were regarded as full British citizens. Finnish immigrants 31
have a similarly privileged position because of the common Nordic 32
labor market and their country’s traditional ties to Sweden. In con- 33
trast to the Irish, however, a large number of Finnish immigrants 34
have considerable language difficulties after arrival, and in this re- 35
spect they resemble the immigrant groups in Sweden that have more 36
distant origins. 37
Eastern Europe is blank on the map, not because it has no migra- 38
tion or exchanges of labor, but because we lack information about 39
it. The sizeable immigration to West Germany from East Germany 40
and from Poland is discussed in the chapter on Germany, but it 41
would also be interesting to have had examples of migration within 42
Eastern Europe. We probably would have found surprising similari- 43

48 tomas hammar
1 ties and enormous differences from the immigration phenomenon

2 in Western Europe. For example, the German Democratic Republic
3 has signed an agreement with Algeria that provides for the trans-
4 fer of workers with relatively stringent provisions which might be
5 compared with similar agreements in the West. In the countries of
6 Eastern Europe, however, state planning and control of the economy,
7 including labor mobility, predominates, which means that the back-
8 ground for immigration and immigration policy is completely differ-
9 ent there. Thus, we leave this part of the map blank, mainly because
10 a thorough study of migration in Eastern Europe requires a separate
11 research project.
12
13 Figure 1.1. Postwar migration to Europe
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

More than three fourths of the foreign citizens in the immigration 1

countries live in France, Germany, and Britain. Each of these coun- 2
tries has approximately four million resident immigrants, although 3
the statistics are difficult to compare and in some cases are rather 4
unreliable. Except for Liechtenstein and Luxemburg, Switzerland 5
has the highest percentage of foreign citizens in its population (14.5 6
percent in 1982). If one compares statistics on the percentage of for- 7
eign workers in the project countries, they are about the same as the 8
percentage of foreign residents (see Table 1.1). 9
These figures do not reveal that immigrants in Western Europe distribution of 10
represent a great number of different nationalities, nor do they nationalities 11
show how immigrants with the same nationality often settle in the 12
same country and even the same region. Spanish and Portuguese 13
immigrants have gone mainly to France, and to a lesser extent to 14
Switzerland. Yugoslavs and Turks have gone mainly to Germany. 15
Italians are an older immigrant group and have settled primar- 16
ily in Switzerland and to a lesser extent in Germany and France. 17
Immigrants from North Africa have gone to France and later to the 18
Netherlands as well, although the bulk of immigration to the latter 19
country has come from its former colonies in Asia and Latin America. 20
21
Table 1.1. Foreign citizens residing in the European project countries in 22
1983 (thousands) 23
24
All residents Labor force 25
Foreign Percent Foreign Percent 26
citizens of total citizens of total 27
28
Sweden 405.5 4.9 2227.7 5.2
29
Netherlands* 543.6 3.7 208.4 3.7 30
France* 4,459.0 7.2 1,436.4 6.3 31
32
Great Britain † 1,705.0 3.1 931.0 3.8
33
West Germany 4,666.9 7.6 2,037.6 9.2 34
Switserland ‡ 925.8 14.5 647.9 21.9 35
36
Source: OECD, Continuous Reporting System on Migration, SOPEMI 1983, for all coun- 37
tries except Great Britain. 38
Notes:
* Data from 1982, and for labor force in France 1981. Based on number of residence
39
and work permits, and therefore an overestimate of the size of the foreign population. 40
† Data from 1981, Labour Force Survey. 41
‡ Yearly average. Seasonal workers (13,400) and frontier workers (108,400) are in- 42
cluded. 43

50 tomas hammar
1 The same is true for Britain, where almost all postwar immigration
2 has come from former colonies in the West Indies and from India
3 and Pakistan. The majority of immigration to Sweden has come
4 from Finland and from the other Nordic countries, although there
5 has also been a significant inflow of immigrants from Yugoslavia,
6 Greece, and Turkey.
7 There are a number of possible explanations for the distribution
8 bilateral agreements of nationalities among the receiving countries. In many cases bilater-
9 al agreements and recruitment practices based on such agreements
10 have led to concentrations of certain nationalities, for example,
11 Turks and Yugoslavs in Germany or Moroccans in the Netherlands.
12 Geographical proximity between sending and receiving countries
13 has often had a similar effect, particularly when accompanied by a
14 history of close relations. Geographical distance has sometimes re-
15 duced the potential for certain kinds of immigration. Since Britain
16 and Sweden are located somewhat on the periphery of continental
17 European migration, they have not received as many immigrants
18 from Southern Europe and Turkey. Ex-colonies and countries with
19 whom they have historically had close contact have provided much of
20 the immigration to France, the Netherlands, and especially Britain.
21 Finally, the distribution of immigrants by nationality can also be ex-
22 plained by “chain migration”, which occurs when an initial group of
23 immigrants settles in a country and then, by encouraging others in
24 their home country or by providing a model for them, attract others
25 of the same nationality to a particular receiving country.
26 The sources of migration to Europe have progressively moved to
27 areas farther and farther away. While immigration from Southern
28 Europe, initially quite extensive, has decreased in recent years, immi-
29 gration from Africa, Asia, and especially the Near East has increased.
30 The change in the sources of immigration has meant that many of
31 the new minority groups are more highly visible, as they differ more
32 in culture and tradition from indigenous European population than
33 did the so-called “traditional” immigrant groups of the past. There
34 are indications that this newer long-distance immigration will con-
35 tinue and increase in the future.
36 An important change in immigration policy occurred during the
37 restricted policies period from 1970 to 1974. For economic and other reasons the im-
38 migration countries of Western Europe heavily restricted or usually
39 stopped recruiting foreign labor, and since then only refugees and
40 the relatives of resident aliens are admitted. Policymakers have now
41 come to realize, to their surprise, that many foreign workers are like-
42 ly to remain as permanent residents.
43

This change in immigration policy, which we will call the “turn- the turning point 1
ing point”, was the first clear break with the relatively open and un- 2
restricted policies of the previous two decades. The change was de- 3
clared in Switzerland (1970), Sweden (1972), Germany (1973), and 4
France (1974). Though it was made with the consent of each national 5
government, it was made without open political debate and without 6
any formal, official decisions. It is important to note that this turning 7
point should be thought of as a policy change towards stricter regula- 8
tion but not necessarily as a “stop” for labor migration. 9
In Britain and the Netherlands, where most immigrants came 10
from colonies or former colonies and usually held the citizenship of 11
the mother country, the turning point in immigration policy did not 12
occur at a specific time but came gradually. In Britain this process 13
has involved the gradual elimination of the immigration rights of 14
colonial citizens. Though this process began there in 1962 and has 15
not yet ended, one can nevertheless say that the passage of the 1971 16
Immigration Act was perhaps the most significant legislation in this 17
area. In the Netherlands there was a major revaluation of immigra- 18
tion policy at the end of the 1970s. The number of new work permits 19
issued fell sharply in 1973, but labor immigration was never formally 20
“stopped”. Not until 1980 did the government impose serious re- 21
strictions on post-colonial immigration and begin to develop a new 22
immigrant policy. 23
Immigration to the six European project countries has changed 24
during the past decade in other ways as well. While the number of 25
single, male immigrants has decreased, mainly because of the policy 26
change that occurred at the turning point, the immigration of refu- 27
gees and the dependants of resident aliens has increased. In other 28
words, the total amount of immigration to the project countries has 29
not decreased substantially as a result of the “stop” in labor recruit- 30
ment, but has remained constant or in some cases has actually in- 31
creased. Thus, there is a relationship between the imposition of the 32
“stop” and the change in the composition of immigrant population. 33
This relationship is discussed in more detail in the comparative anal- 34
ysis presented in Part II. 35
36
Immigration policy 37
There are many definitions of immigration policy. They vary even a working definition 38
within a single country. Yet when we compare a number of coun- 39
tries, we need a working definition that is relevant to all these coun- 40
tries. Thus, under our scheme, “immigration policy” will consist of 41
two parts which are interrelated, yet distinct: (a) regulation of flows of 42
immigration and control of aliens, and (b) immigrant policy. 43

52 tomas hammar
1 Immigration regulation and aliens control

2 Regulation of immigration is the oldest, the most obvious, and ac-
3 cording to some people the only aspect of immigration policy.
4 Immigration regulation refers to the rules and procedures gov-
5 erning the selection and admission of foreign citizens.
6 It also includes such regulations which control foreign citizens
7 (aliens) once they visit or take residence in the receiving country,
8 including control of their employment. Deportation also falls under
9 these regulations. Employers may be allowed to recruit foreign labor
10 regulation on their own, or labor transfer agreements may be entered into by
11 the state and official information and recruitment bureaux be opened
12 abroad. All this, of course, is a part of immigration regulation and
13 must be included along with measures taken to restrict immigration
14 or to stop it completely. The free movements of peoples, such as oc-
15 cur in the common labor markets of the EEC and Nordic areas, are
16 also an aspect of immigration regulation; even though in these two
17 cases policymakers have decided that certain kinds of immigration
18 should not be regulated.
19 In general, all sovereign states reserve the right to determine
20 whether foreign citizens will be permitted to enter their territory and
21 reside there, and in all the project countries this power of the state is
22 found in law or in administrative regulations. Most changes in im-
23 migration policy, for example the changes at what we call the “turn-
24 ing point”, have been made by changing the application of existing
25 aliens laws and not by changing the laws. Such laws were applied in
26 a liberal way as long as immigration was encouraged, but later, when
27 the goal was to limit the volume of immigration, discourage potential
28 immigrants, and reduce the total number of foreigners in the coun-
29 try, the application of the same aliens laws became more strict. At
30 the same time, however, immigration regulation was abandoned for
31 certain groups of foreigners who were admitted without restrictions.
32 Examples of this are, as already mentioned, the free circulation of
33 labor in the EEC and the Nordic area and the acceptance on a perma-
34 nent basis of political refugees.
35 Immigration regulation implies that foreign citizens remain un-
36 der some kind of aliens control until they become naturalized citi-
37 zens. The conditions that foreign citizens are subject to during this
38 period of “controlled” residence vary greatly from country to country.
39 Some countries at an early stage guarantee their foreign residents
40 the right to remain permanently. Other countries keep them in a
41 position of legal insecurity and uncertainty for many years. Some
42 countries admit foreign workers for seasonal employment and re-
43 quire them to leave when the season ends, although they are often

permitted to return again the following season. Some countries or- 1

ganize so-called “rotation” systems under which foreign workers are 2
allowed to stay in the country only a maximum number of months or 3
years, after which (in theory at least) they must depart to make room 4
for new workers. In this way these countries hope to avoid the es- 5
tablishment of any new, permanent population groups whose needs 6
and demands would be considerably greater than those of temporary 7
“guestworkers”. 8
Even in countries that do not apply seasonal employment or rota- 9
tion systems, however, it often takes many years before foreign citi- 10
zens are guaranteed that they will not be forced to leave the coun- 11
try against their will. By delaying “permanent status”, immigration 12
countries retain the legal right to repatriate foreign workers when 13
desired, even those with many years of residence. The conditions at- 14
tached to permanent status can thus function as a means of control- 15
ling the size or composition of immigration and must therefore also 16
be included as a part of immigration regulation. 17
Compulsory repatriation of large groups of immigrants is rare. 18
Nevertheless, it has long been a possibility which hangs over the 19
heads of many of the foreign workers employed in Western Europe. 20
Though seldom utilized, it nonetheless influences their living condi- 21
tions and their attitudes towards residence in the host country. 22
Thus, the very existence of the possibility of compulsory repatria- legal insecurity 23
tion is a factor in a country’s immigrant policy. Immigration regula- 24
tion may be said to foster a considerable degree of legal insecurity 25
because decisions concerning permanent status are made by admin- 26
istrative authorities who have much discretion in interpreting such 27
regulations. Such legal insecurity is made worse when foreign citi- 28
zens have no right to appeal against the decisions of administrative 29
authorities. 30
31
32
Immigrant policy 33
34
Immigrant policy is the other part of immigration policy and refers 35
to the conditions provided to resident immigrants. It comprises all 36
issues that influence the condition of immigrants; for example, work 37
and housing conditions, social benefits and social services, educa- 38
tional opportunities and language instruction, cultural amenities, 39
leisure activities, voluntary associations, and opportunities to par- 40
ticipate in trade union and political affairs. Immigrant policy may be 41
either direct or indirect. 42
Immigrants have a number of special needs to begin with be- 43

54 tomas hammar
1 cause they are different from the host population. They often speak a
2 foreign language and represent a different culture. Immigrants also
3 have special economic interests and ambitions for the future. All of
4 this may sometimes prompt a country of immigration to devise spe-
5 cial measures to improve the situation of its immigrants. Since these
6 ‘direct’ immigrant measures do not usually apply to the non-immigrant population, we
7 policy will call them “direct” immigrant policy.
8 Like the non-immigrant population, immigrants are also affected
9 by a country’s general public policy, which involves economic, social,
10 political, and other measures. These measures are not designed with
11 only immigrants in mind; instead, they are intended to apply to all
12 inhabitants of a country whether citizens or not. Yet they may not
13 be applied to all inhabitants in the same way, i.e. there may be dis-
14 crimination, both positive and negative, in the allocation of resources
15 and opportunities. When general public policy affects immigrants
16 ‘indirect’ immigrant substantially, we will talk about “indirect” immigrant policy.
17 policy Indirect immigrant policy can be termed “inequitable” or “dis-
18 criminatory” when immigrants receive significantly less than others,
19 and when they are denied opportunities to participate in society. Even
20 when the distribution of benefits is perfectly equal, however, immi-
21 grants can still remain in an inferior position, primarily because they
22 have recently made a new start in the host country and experience
23 less favorable circumstances than the rest of the population. This
24 situation can be ameliorated if immigrants are given greater benefits
25 than other people, e.g. special language instruction, special cultural
26 support, and so on. These measures are the tools of direct immigrant
27 policy.
28 To summarize in outline form, immigration policy comprises:
29
30 1. Immigration regulation and aliens control
31 (a) “strict” or “liberal” control of the admission and residence of
32 foreign citizens
33 (b) guarantees of “permanent status”; legal security versus vulner-
34 ability to arbitrary expulsion
35 2 Immigrant policy
36 (a) indirect: immigrants’ inclusion in the general allocation of
37 benefits; “equal” versus “discriminatory” distribution
38 (b) direct: special measures on behalf of immigrants; “affirmative
39 action” and the removal of legal discrimination
40
41 Although we will in our analysis distinguish between these two parts
42 of immigration policy, they are of course in practice at work simul-
43 taneously. What is very often not understood is the profound effect

that they can have on one another. A system of rotation might, for 1
example, leave most immigrants in a very weak legal position as 2
residents. This may in turn impede integration and the full enjoy- 3
ment of social and civil rights – both areas of concern to immigrant 4
policy. Another example of the mutual influence between immigra- 5
tion regulation and immigrant policy would be when a country uses 6
instruments of immigrant policy (e.g. housing applications, school 7
registers, and so on) to identify and expel illegal immigrants, thus 8
accomplishing a task of immigration regulation. 9
10
11
General preconditions 12
13
Immigration policy should be analyzed in the context of a country’s 14
history, economy, geography, population, international relations, 15
etc., for these are factors that affect immigration to a country, both 16
quantitatively and qualitatively. Valid comparisons between the proj- 17
ect countries are possible only when the general preconditions for 18
the countries’ immigration policies are analyzed. 19
Policymakers in each country may have tried to shape immigra- 20
tion policy on the basis of their own experience and their particular 21
national needs, but the policies of all the project countries neverthe- 22
less have numerous features in common. Periods of passport exemp- 23
tion, rigid immigration control, and active recruitment of foreign 24
workers have come at the same or almost the same time in every local conditions 25
country. Thus, it seems that the shaping of immigration policy is matter 26
determined in part by conditions beyond the control of policy makers 27
in the individual countries. For example, two world wars have dis- 28
rupted long-standing patterns of habitation and have forced people 29
to flee their home countries. Economic disruptions, resulting either 30
from the wars or from other causes, have been possibly even more 31
unsettling than the wars themselves. The Great Depression in the 32
1930s affected the entire industrialized world and resulted in the 33
widespread traumatic belief that future economic crises had to be 34
avoided at all costs. During the following decades, Keynesian eco- 35
nomic theory gradually provided new policy options, starting with 36
active budget policies, which were applied to counter depressions. 37
Of course, all countries have not been affected by war and eco- 38
nomic crisis to the same degree, and partly because of this, there 39
are significant differences in the immigration policies of the proj- 40
ect countries. One might say that although they came from different 41
parts, they are all sailing on the same heaving ocean, all exposed to 42
the same fluctuations in weather, winds, and currents. Yet because 43

56 tomas hammar
1 they each set a different course and sail in a different kind of vessel,
2 no two voyages are ever exactly alike. Similarly, no two countries’ im-
3 migration policies are ever exactly alike, even though all countries are
4 affected by and must contend with the same external conditions.
5 “General preconditions”, as the term will be used here, are back-
6 ground conditions which, on the whole, remain stable for a consid-
7 erable period of time and are not easily influenced or altered in the
8 short term. For the general as well as attentive public, and also for
9 conditional policymakers, these conditions act as constraints on the possibilities
10 constraints for state action; in other words, they form a factual, concrete frame-
11 work for immigration policy over a relatively long period of time.
12
13
14 Terminology
15
16 definition of Two of the key concepts in this comparative study are immigrant
17 immigrant and and immigration. The term “immigrant” is sometimes used in the
18 immigration very broad sense of its root-word “migrant”, a person who moves
19 from one country to another. In common usage, however, the term
20 “immigrant” has acquired the narrower meaning of “a person who
21 migrates to a country with the intention of taking up permanent resi-
22 dence”, something akin to the term “settler”. The definition of im-
23 migrant that will be used in this book lies somewhere in between the
24 broad sense of “migrant” and the narrow sense of “settler”:
25
26 “Immigrant” is a person who migrates to a country and then actually
27 resides there longer than a short period of time, i.e. for more than
28 three months.
29
30 “Immigration” refers to the physical entrance of immigrants as here
31 defined, either singly or as a group, into a country.
32
33 This definition thus excludes people that pay only a short visit to a
34 country; for example, those who come on vacation or to visit rela-
35 tives, or those who come on business trips or to do some specific
36 job (a mechanic to install machinery for instance, or artists to give a
37 performance), as long as their stay is for less than three months. On
38 the other hand, “immigrant” does not only refer to those who plan
39 from the beginning to stay permanently in a country. Thus, students,
40 scholars, artists, and others who spend longer than three months as
41 “guests” in a country are considered immigrants although they do
42 not plan to stay permanently.
43 The decisive criterion is the actual length of time that a person

resides in the country of immigration. People that intend to remain 1

permanently, i.e. “settler” immigrants, are not included in the defini- 2
tion if they return home after only a couple of months; on the other 3
hand, people that intend to remain only a couple of months but later 4
change their mind and stay for several years are included. Obviously, 5
the length of residence necessary for a person to be included in our 6
definition of “immigrant” cannot be determined in any but an arbi- 7
trary fashion. Each project country allows most foreign citizens to 8
take up residence for a limited period of time, usually three to six 9
months, without requiring visas or residence permits, and for this 10
reason we have set the residence criterion in our definition at three 11
months. 12
Foreign citizens that remain in a country for longer than three 13
months must usually obtain a residence permit; therefore, any for- 14
eign citizen who has such a permit is likely to become an immigrant, 15
and is therefore considered such under our definition. But the defi- 16
nition also includes people who do not have residence permits, in 17
particular illegal or “undocumented” aliens. In general, it is difficult 18
to say with certainty that people are or are not immigrants when they 19
arrive, although those who have applied for residence permits in ad- 20
vance are of course more likely to stay longer than those who have 21
not. Under our definition, the criterion determining whether or not 22
a foreign citizen should be considered an immigrant is if he or she 23
stays in the country for longer than three months. 24
The terms “immigrant” and “immigration” are applied in a dif- 25
ferent manner in each project country, and their meanings have 26
changed over time. The definition used here will for this reason 27
cause more difficulties in some project countries than in others. 28
As the following chapters will show, there is an obvious relation 29
between a country’s immigration policy and its terminology. In 30
Germany and Switzerland immigrants are “foreign workers” (aus- 31
ländische Arbeitnehmer in Germany and Fremdarbeiter in Switzerland) 32
and they are controlled by “aliens bureaux” (Ausländerbehörde, or in 33
Switzerland Fremdenpolizei). France has always used the terms les 34
immigrés and l’immigration, and Sweden used similar terms (invan- 35
drare and invandring) in the 1960s when its new immigrant policy 36
was launched. In Britain the term “immigrant” has been applied par- 37
ticularly to colored people, while in the Netherlands the new policy 38
envisioned for immigrants is called a “minorities” policy. 39
The technical language used in each country is adjusted so that terminology’s 40
it best describes and explains the country’s policy. Terminology also influence 41
influences the way in which immigration policy is conceived and un- 42
derstood in each country; terms that should be instruments of de- 43

58 tomas hammar
1 scription gradually become fixed concepts that limit flexibility and

2 creativity. For this reason it is important that our comparative discus-
3 sions use terms that are well defined. The above definitions of “im-
4 migrant” and “immigration” will be used in a strict sense in the com-
5 parative chapters and will also serve as the general frame of reference
6 in the country chapters, although each author has naturally chosen to
7 use the terminology of his particular country by way of illustration.
8
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
The crucial meso-level 3
4
Thomas Faist1 5
6
7
8
9
10
11
This article was originally published in 1997 in a volume entitled International 12
13
Migration, Immobility and Development: Multidisciplinary Perspectives, edited
14
by political scientists Thomas Faist and Tomas Hammar and sociologists
15
Grete Brochman and Kristof Tamas. Historically, most theories of migration
16
have focused either on global and structural factors explaining the different 17
patterns of population movement (macro theories) or on individual determi- 18
nants of migration (micro theories). The meso level that exists between indi- 19
viduals and larger structures has been, for a long time, relatively neglected. 20
Faist argues very convincingly in favour of meso level theories. It is crucial to 21
understand the impact on migration of social relations or social ties across 22
individuals in kinship groups, households, local settings, formal organisa- 23
tions and friendship circles. It is also vital to examine the relationships be- 24
tween migrants and those who stayed behind since they may explain the 25
reproduction of migration patterns. With his characteristic theoretical rigor, 26
Faist offers here one of the first systematic arguments in defence of meso 27
theories of migration while, at the same time, recognising the importance of 28
micro and macro theories. By doing so, he has promoted a comprehensive, 29
multilevel approach to migration that has since been further developed in 30
the literature. 31
32
33
Lacunae in sociological theories of international migration 34
35
Sociological approaches have presented an impressive array of plau- 36
sible arguments as to why people move from one place to another, 37
especially across the borders of nation-states. However, these theo- 38
ries have not directly addressed the question of why so few people 39
migrate from so few communities and why so many return. Firstly, 40
the total migrant population in the world is estimated to about 2 per 41
cent of the world’s population. Secondly, return migration consti- 42
tutes an important fact. The social ties between movers and stayers 43

60 thomas faist
1 are not automatically ruptured. For example, between 1960 and 1993
2 out of an estimated total of 12 million labour migrants and depen-
3 dants from the Mediterranean countries of Southern Europe and
4 North Africa, 9.3 million returned to their countries of origin from
5 Germany (own calculations, based upon Statistisches Bundesamt
6 1955-95). Nevertheless, the immigrant population in Germany in-
7 creased as a result of family reunification during the later 1970s and
8 1980s after the ending of guestworker recruitment.
9 In short, any theoretical attempt should therefore not focus on
10 movers and stayers movers only, but on both movers and stayers, and also on how stayers
11 who once make a move shuttle back and forth, or become stayers
12 again, be it in the countries of origin or destination.
13 Most theoretical efforts have mostly focused either on global
14 structural factors inducing migration and refugee movements (mac-
15 ro-theories) or on factors motivating individuals to move (micro-
16 theories). This review and partial reconstruction of theories about
17 international South to North migration emphasises the meso-level
18 between what are usually called the micro- and the macro-levels, the
19 level of analysis between individuals and larger structures such as
20 the nation-state. It does so in focusing on social relations (social ties)
21 between individuals in kinship groups (e.g. families), households,
22 neighbourhoods, friendship circles and formal organisations.
23 meso-level Two strands of literature have paid attention to the meso-level.
24 Firstly, in recent years the processes of immigrant incorporation
25 have been studied in economic sociology (Portes 1995). However,
26 so far little has been said about the costs and benefits involved in
27 transferring human capital abroad or about the mediating role of
28 resources inherent in social relations (social capital) in the decision-
29 making process. Secondly, there is a huge and impressive empirical
30 literature on migrant networks (Massey et al. 1993). There are also
31 plausible arguments as to why these migrant networks embedded in
32 migration systems (Kritz and Zlotnik 1992) are crucial elements in
33 explaining international migration. Yet this literature is more suc-
34 cessful in explaining the direction (e.g. from former colonies to the
35 European and North American core) than the volume of international
36 movement. In particular, it is not clear what exactly happens in net-
37 works and collectives that induces people to stay, move and return.
38 The specific characteristics of social capital are important in ex-
39 plaining the low volume of international movement, chain migra-
40 tion and often high rates of return migration. It is very difficult to
41 transfer social capital abroad; even harder than the transfer of hu-
42 man capital. However, once pioneer migrants have moved abroad,
43 relatives, friends and acquaintances can draw upon social capital and

processes of ‘chain migration’ develop. Nevertheless, social ties of 1

movers and stayers do not simply vanish in the course of interna- 2
tional migration. This is why many movers return to the countries of 3
origin. 4
The following discussion evaluates micro-level rational choice 5
theory and macro-level migration systems theories. Secondly, it in- 6
troduces three levels of analysis – the structural (political-econom- 7
ic-cultural factors in the sending and receiving countries and at the 8
international level), the relational (social ties of movers and stayers) 9
and the individual (degrees of freedom of potential movers). Thirdly, 10
the decision-making processes and the dynamics of migration are 11
partially reconstructed. Two crucial categories are used as a point of 12
departure: social ties and social capital in social networks and collec- 13
tives. 14
15
16
Dominant theories of international migration 17
18
This section appraises micro- and macro-level theories about the vol- micro- and 19
ume and dynamics of South to North movement. The idea is not macro-level theories 20
to evaluate these theories as such but what they say about decision 21
making and the dynamics of international migration. Theoretical 22
and empirical work started with Sir Ernest George Ravenstein (1885 23
and 1889). He perceptively analysed relations between distance and 24
propensity to move, developing seven ‘laws’ of migration. 25
The laws are: (1) The majority migrate only short distances and 26
thus establish ‘currents of migration’ towards larger centres. (2) This 27
causes displacement and development processes in connection with 28
populations in sending and destination regions. (3) The processes of 29
dispersion and absorption correspond to each other. (4) Migration 30
chains develop over time. (5) Migration chains lead to exit move- 31
ments towards centres of commerce and industry. (6) Urban resi- 32
dents are less prone to migrate than rural people. (7) This is also true 33
for the female population. 34
These observations are a useful starting point as empirical rules 35
of thumb that may apply to certain regions of the world at specif- 36
ic time periods. Ravenstein himself found abundant evidence for 37
these ‘laws’ in mid-nineteenth-century internal English migration. 38
However, his generalisations and later those of Everett Lee ( 1964) 39
must be placed into more general sociological frameworks if we 40
want to know whether and why their rules of thumb are true or not. rational choice and 41
Rational choice and systems theories may provide such frameworks. systems theories 42
43

62 thomas faist
1 The rational choice approach: between preferences and

2 opportunities
3 The basic instrumental statement is: In choosing between at least
4 two alternative courses of action, a person is apt to choose the one for
5 which the perceived value of the result is the greater. It is assumed
6 that the actor is able to make rational decisions on the basis of a set
7 of tastes or preference orderings.
8 Some sociological rational choice theories take as the basic com-
9 ponent not only the values (goals, preferences) but also the expectan-
10 cies (subjective probabilities) a potential mover holds (DeJong and
11 Fawcett 1981; see also chapter 3).
12 basic The basic value-expectancy model is straightforward:
13 value-expectancy
14 model MM = S(i) P(i) E(i)
15
16 where ‘MM’ is the strength of the motivation to migrate, ‘P’ is the
17 preferred outcome, ‘E’ is the expectancy that migration will lead to
18 the desired outcome, and ‘i’ refers to the specific preferences (values)
19 potential movers hold.
20 The preferences may be most diverse. They may be related to im-
21 proving and securing: wealth (e.g. income), status (e.g. prestigious
22 job), comfort (e.g. better working and living conditions), stimulation
23 (e.g. experience, adventure and pleasure), autonomy (e.g. high de-
24 gree of personal freedom), affiliation (e.g. joining friends or family),
25 exit from oppression of all kinds (e.g. refugees), meaningful life (e.g.
26 improving society), better life for one’s children, and morality (e.g.
27 leading a virtuous life for religious reasons). In this view the poten-
28 tial migrant might not only be a worker, a member of a household
29 or a kinship group, but also a voter, a member of ethnic, linguistic,
30 religious and political groups, a member of a persecuted minority, or
31 also, among many other things, a devotee of arts or sports.
32 In addition to values (preferences) and expectancies Hartmut
33 Esser explicitly adds a third important element, opportunities
34 and constraints. Therefore, his approach can be called structural
35 individualism.
36 We could restate the above equation to read:
37
38 MM = S(i) V(i) P(i), depending on O/C
39
40 ‘O/C’ is the set of external opportunities and constraints encoun-
41 tered by a potential migrant.
42 Esser’s theoretical approach deals with assimilation and accultur-
43 ation of immigrants in the receiving country. Yet Esser’s premises

could be used to deal with decision making in the sending countries 1

as well. His first fundamental hypothesis (1980: 210-11) could be re- 2
stated as follows: The more intense the motives of a migrant are re- 3
garding a specific goal, the stronger the expectation that she can fulfil 4
her goals by (temporary) territorial exit, the higher the propensity 5
to attribute a high preference (value) to exit and the fewer the con- 6
straints working against exit, the more likely a potential mover will 7
choose the exit option. These constraints and opportunities could in- constraints and 8
clude factors such as societal and cultural norms (e.g. gender roles), opportunities 9
state policies (admission policies of the receiving countries) and eco- 10
nomic differentials related to income or employment. 11
In addition to opportunity structures information plays a decisive 12
role for migration decision making. Depending on the availability of 13
information on transport and opportunities for jobs and housing, po- 14
tential migrants can optimise their benefits. Such information may 15
flow along various communication channels, such as mass media 16
and friends who migrated before but also pioneer migrants outside 17
the inner circle of relatives and friends. 18
An important prerequisite of immobility then is that a potential 19
mover has sufficient information as to what goals can be better ac- 20
complished in the sending or the receiving country. If a potential 21
migrant decides to be mobile, the question arises whether the neces- 22
sary resources can be transferred abroad. The territorial restriction 23
of certain assets has been termed ‘location specific capital’ (DaVanzo location-specific 24
1981: 116). It is a widespread phenomenon that highly educated and capital 25
trained movers, especially refugees, cannot enter at the same occu- 26
pational level in the receiving country. For example, lawyers, physi- 27
cians and engineers may not get accredited to practise law, medicine 28
and mechanics and may have to look for work outside their field. 29
Information about these and other limitations may prohibit interna- 30
tional movements although they would not discourage the internal 31
movement of migrants. In these cases it is more likely that internal 32
and not international migration occurs. 33
Rational choice accounts certainly are a powerful tool with which 34
to model migration decision making and action. Yet, we have to ex- 35
amine what is meant by opportunities and constraints in order to 36
understand more clearly the decision-making process. Sociological 37
and anthropological studies have frequently found that migration de- 38
cisions are taken in social units such as the family, extended families 39
or even whole communities. These social units use available resourc- 40
es in their perceived self-interest. Often, in patriarchal systems. the 41
male head decides at the expense of females and younger members 42
of the family. 43

64 thomas faist
1 This problem of defining a supra-individual decision-making unit

2 is partly remedied by the ‘new economics of migration’ (Stark 1991),
3 whose theorists do not prejudge the sole social unit of decision mak-
4 ing to be the individual actor but try to aggregate the utilities of the
5 individuals involved, especially in the case of rural economies. Yet by
6 considering family utility in aggregate terms, these theorists have ig-
7 nored or simplified the relations between family members, the social
8 ties that bind or separate family or household members. If basic so-
9 cial relations are disregarded in this way, we do not get a good idea of
10 power and authority relations, (mis)trust and solidarity. For example,
11 who decides which member of a social unit such as a household mi-
12 grates and what is the legitimation of the decision maker?
13 Even if we specify the structural opportunities and constraints, we
14 should still explain how they relate to individual rationality. Rational
15 choice approaches to migration do not specify how structural oppor-
16 tunities are translated into individual action and vice versa. In es-
17 sence, we encounter the problem of linking macro- and micro-levels
18 of analysis: ‘For example, a sophisticated individual might be aware
19 of the level and nature of foreign investment in his or her country,
20 but would still be unlikely to perceive it as immediately affecting a
21 residence desire and possible decision to move’ (Gardner 1981: 73).
22 To make this link we need to complement micro-level approaches
23 with more elaborate concepts of social relations and social ties.
24
25 The migration-systems approach: between the world system and
26 networks
27 While rational choice theories of migration have evolved from the
28 micro-level to consider macro-level factors also (structural individ-
29 ualism), systems theorists have come full circle: They were at first
30 exclusively concerned with the macro-level (migration systems), but
31 have gradually come to introduce lower-level concepts such as mi-
32 grant networks.
33 The most elaborate effort at developing a fully-fledged system-the-
34 oretic analysis is Hoffmann-Nowotny’s concept. encompassing four
35 levels, the individual, national subsystems, national societies and the
36 international society (1970 and 1973). Hoffmann-Nowotny applies
37 general social systems theory to the phenomenon of international
38 migration. He starts with the fundamental relation between power
39 prestige and power and prestige in a society. In his conceptual universe ‘prestige’ legiti-
40 mises ‘power’. Hoffmann-Nowotny posits that in any society there
41 exists some sort of consensus about the value attributed to material
42 and immaterial goods (e.g. education). Power and prestige in a social
43 system are determined by the position and by the status attributed to

their positions. ‘Structural tensions’ arise from inequalities and sta- 1

tus inconsistencies in the sending country. These structural tensions 2
may generate ‘anomic tendencies’, i.e. an imbalance between power 3
and status. Action directed to resolve these tensions may take forms 4
such as social mobility, giving up the social position held or emigra- 5
tion to a country where status aspirations can be attained (Hoffmann- 6
Nowotny 1973: 11-14). In essence, for Hoffmann-Nowotny (interna- 7
tional) migration constitutes an ‘interaction between societal systems 8
geared to transfer tensions and thus balancing power and prestige 9
(1973: 19; translation T.F). 10
Later migration-systems approaches have four main characteris- four main 11
tics. Firstly, migration-system theories assume that migration sys- characteristics 12
tems pose the context in which movement occurs and that it influ- 13
ences actions on whether to stay or to move. An analysis of trade and 14
security linkages and colonial ties helps to explain the origin and direc- 15
tion of international movement. Basically, a migration system is here 16
defined as two or more places (most often nation-states) connected 17
to each other by flows and counterflows of people (see Faist 1995). 18
Secondly, using dependency-theory and world-systems approaches, 19
systems theories have stressed the existence of linkages between coun- 20
tries other than people, such as trade and security alliances, colonial 21
ties and flows of goods, services, information and ideas (Portes and 22
Walton 1981). These linkages often have existed before migration 23
flows occurred. For example, in the case of European receiving coun- 24
tries (e.g. France, Netherlands and Great Britain) most movers come 25
from former colonies. 26
Thirdly, migration systems theory focuses on processes within mi- 27
gration systems. Movement is not regarded as a one-time event but 28
rather as a dynamic process consisting of a sequence of events across 29
time (Boyd 1989: 641). Already Mabogunje suggests in his program- 30
matic article on rural-urban migration in Africa that migration needs 31
to be studied as ‘a circular, interdependent, progressively complex 32
and self-modifying system’ (1970: 4). Theorising the dynamics of 33
migration has thus moved from a consideration of movement as a 34
linear, unidirectional, push-and-pull, cause-effect movement to no- 35
tions that emphasise migration as circular, interdependent, progres- 36
sively complex and self-modifying systems in which the effect of 37
changes in one part can be traced through the entire system. For 38
example, once it has started, international migration turns into a self- 39
feeding process. Petersen assumed that pioneer migrants or groups 40
set examples that can develop into a stream of what he called ‘mass 41
migration’ (1958: 263-4). This helps to explain international move- 42
ment as a self-feeding process that gains in momentum as networks 43

66 thomas faist
1 reduce both the direct monetary costs of movement and the opportu-
2 nity costs (that is, the earnings forgone while moving, searching for
3 work and housing, learning new skills), and also decrease the psy-
4 chological costs of adjustment to a new environment in the receiving
5 country. Movers and stayers are regarded as active decision makers
6 (Fawcett 1989).
7 Fourthly, within the context of important factors such as econom-
8 ic inequalities within and between nation-states and the admission
9 policies of the receiving states, individuals, households and families
10 develop strategies to cope with stay-or-go alternatives. Lately, systems
11 theorists have started to apply social network theory vigorously. The
12 main assumption is aptly summarised in Charles Tilly’s provocative
13 phrase that it is ‘not people who migrate but networks’ (1990: 75). In
14 other words, migrants are not ‘atomistic flies’ (Cohen 1987). Social
15 networks consist of more or less homogeneous sets of ties between
16 three or more actors. Network patterns of social ties comprise eco-
17 nomic, political networks of interaction, as well as collectives such
18 as groups (e.g. families, communities) and (public) associations.
19 ‘Network theory builds its explanations from patterns of relations.
20 It captures causal factors in the social structural bedrock of society,
21 bypassing the spuriously significant attributes of people temporarily
22 occupying particular positions in social structure’ (Burt 1986: 106).
23 Migrant networks, then, are sets of interpersonal ties that connect
24 movers, former movers and non-movers in countries of origin and
25 destination through social ties, be they relations of kinship, friend-
26 ship or weak social ties (see Choldin 1973). In international migra-
27 tion, networks may be even more important than in domestic migra-
28 tion because there are more barriers to overcome, e.g. exit and entry
29 permits, and if not available, costs for illegal border crossing.
30 Concerning migration and non-migration, a system-theoretic
31 perspective emphasises that predisposing factors of very different
32 kinds can enhance migration (e.g. wage differentials between coun-
33 tries, population growth, civil wars) when embedded in the context of
34 historically grown political, economic and cultural linkages between
35 senders and receivers, while other macro-factors may lead to non-mi-
36 gration, such as very restrictive exit and entry policies. Precipitating
37 events (e.g. economic crises in sending countries) and intervening fac-
38 tors (e.g. migrant networks) are then thought to enhance migration.
39 An important insight is that migration processes are accompanied by
40 feedback effects affecting decisions to stay or go. For example, earlier
41 internal migrations may lead to international migration or pioneer
42 migrants may serve as role models for other potential migrants.
43 In sum, migration-systems theories constitute a great advance in

the explanation of the dynamics of international movements. Yet, the 1

real significance of social and political units between the micro- and 2
macro-levels remains blurred. Contrary to what is claimed, we get no 3
clear understanding of the mechanisms by which macro-factors shape 4
micro-level decision making. To posit the relevance of intermediate 5
structures such as the family, household and migrant networks is 6
not sufficient to establish a meso-level. It begs the question as to how 7
intermediate structures systematically pattern decision making, and 8
are shaped both by the actions of potential and actual movers and by 9
larger social structures. 10
Both rational choice and migration-systems theories have started 11
to place more emphasis on processes linking the micro- and macro- linking micro- and 12
levels: Rational choice theories have come to consider social units macro-levels 13
such as families and migration-systems theories emphasise net- 14
works. But both show a decisive weakness in conceptualising the 15
social ties of movers and stayers within families or households and 16
networks. Processes within these social units and relations between 17
them and larger aggregates (e.g. state institutions) have to be brought 18
into this analysis. One of the crucial factors is the lack of an appro- 19
priate conceptual framework. The following exposition of a social 20
relational approach is therefore not meant to substitute but rather 21
to enrich the rational choice and migration-systems approaches to 22
international migration by paying more systematic attention to the 23
meso-level. 24
25
26
Three levels of analysis: macro-structural, relational and 27
individual 28
29
In its most general form spatial movement can be understood as a 30
transfer not only from one place to another but also from one social 31
unit or neighbourhood to another. This transfer may strain, rupture, 32
change or reinforce previous social ties. In a sociological analysis 33
of international migration three levels are relevant: (1) political-eco- 34
nomic-cultural structures on the level of the international system, 35
the country of origin and the country of destination (structural level), 36
(2) density, strength and content of social relations between stayers 37
and movers within units in the areas of origin and destination (rela- 38
tional level), and (3) the degree of freedom or autonomy of a potential 39
mover (individual level), i.e. the degree to which he or she has the 40
ability to decide on moving or staying. 41
(1) Political-economic-cultural structures denote an array of factors political-economic- 42
in the sending and receiving countries and in the international politi- cultural structures 43

68 thomas faist
1 cal and economic system of nation-states. The nation-states differ in

2 the political realm as political and administrative units. For example,
3 sending countries may vary with respect to political stability. This has
4 consequences for the emergence of refugee flows. The admission
5 and integration policies of sending countries also vary. Nation-states
6 also differ along characteristics such as living standards, jobs and
7 working conditions, unemployment rates and wages in the sending
8 and receiving countries. Such differentials are important prerequi-
9 sites for migration to occur between nation-states. Finally, in the
10 cultural realm there are differences in normative expectations and
11 collective identity. For example, in some areas of the world, ‘cultures
12 of migration’ have developed (e.g. Caribbean islands and the Indian
13 island of Goa). International norms and organisations also have an
14 impact on the mobility of persons (e.g. international convenants on
15 human and social rights by the International Labour Organisation
16 and the Geneva Convention on refugees and asylum seekers).
17 Research into structural opportunities has been abundant, espe-
18 cially into the history of international labour migration. Hatton and
19 Williamson (1994) summarise their findings on transatlantic migra-
20 tions from Europe to America around the turn of the century, saying
21 that demographic growth in the sending regions and income gaps
22 between home and overseas destinations were both important, while
23 industrialisation (independent of its influence on real wages) made
24 a modest contribution. Frank Thistlethwaite argued in his précis on
25 earlier transatlantic migrations that ‘the inner secrets of emigration
26 are to be sought in the working of those two revolutions which are so
27 interconnected, the demographic and the industrial’ (1991: 236-7).
28 With respect to political refugees, however, large refugee flows
29 have been caused by international wars, especially the Second World
30 War but also the Cold War. Many more recent refugee flows have
31 originated as a by-product of the formation of new states in the
32 South, or as a result of social transformations (e.g. revolutions) and
33 ethnic conflicts in both old and new states. External intervention in
34 less developed countries has also been a common cause of refugee
35 flows, for example in the South (Zolberg et al. 1989). Also, the very
36 formation of territorially bounded states in the South after decolo-
37 nisation resulted in migration and refugee flows. Moreover, in pro-
38 cesses of state formation and the rebuilding of states the persecution
39 and expulsion of minority groups can achieve a high priority.
40 (2) The social ties of the movers and stayers vary with respect to
41 density, strength density, strength and content. These ties may go to the receiving or the
42 and content sending countries or to both at the same time. They can range from
43 a dense network of social ties to the country of origin to a total break,

i.e. no social relations anymore and a reorientation to the country 1

of destination in the process of settlement. Yet even in the case of 2
permanent settlement abroad, social ties can be established or re- 3
inforced both in the country of origin and in the receiving country. 4
Therefore, permanent settlement in the receiving country does not 5
necessarily mean fewer social ties to the area of origin. If these social 6
ties are systematically patterned in networks and collectives, we can 7
link the relational to the structural level. 8
(3) On the individual level international movements can be char- 9
acterised by a continuum along the degrees of freedom for potential degrees of freedom 10
movers. At one end, in some instances – for example, slaves, con- 11
victs, some refugees, contract workers, sometimes children and 12
spouses – the essential decision maker is not the migrant him- or 13
herself. At the other end, there are individuals with a high degree of 14
autonomy, based on resources such as money, information and con- 15
nections. The degree of freedom or autonomy is circumscribed in a 16
context in which the main sets of parties involved in migration deci- 17
sion making and the dynamics of migration are: (1) individuals in the 18
place of origin; (2) collectives and social networks of potential and ac- 19
tual movers and stayers such as families, households, friendship and 20
kinship circles, neighbourhoods, ethnic, religious and professional 21
associations, but also (3) interested collective actors in the countries 22
of origin and destination (e.g. Non-Governmental Organisations, su- 23
pra-national organisations such as UNHCR, sending and receiving 24
country governments, political parties, unions and employer organi- 25
sations). 26
27
Characteristics of a meso-level approach 28
Firstly, emphasis needs to be placed on how decisions on moving 29
and staying are made in and between groups of people (e.g. families 30
and various forms of larger territorial and extra-territorial communi- 31
ties) rather than by isolated individuals or groups where economic- 32
political-cultural structures only come in as external constraints and 33
opportunities. A processual account will help us to specify the mech- 34
anisms causing changes in social relations. In this interpersonal and interpersonal and 35
inter-group perspective decisions over moving and staying may be inter-group decisions 36
taken on different levels – for example, by individuals and differently 37
sized groups – or imposed upon these groups by outside collective 38
actors such as governments of nation-states. The basic assumption 39
is that potential migrants and groups always relate to other social 40
structures along a continuum of degrees of freedom. Particular units 41
such as households or families therefore deserve special attention. 42
Empirical studies muster abundant evidence that these units have 43

70 thomas faist
1 figured most prominently not only in earlier transatlantic migrations

2 from Europe to the white-settler colonies (Bodnar 1985), but also do
3 so in contemporary movements from the South to the North, espe-
4 cially from rural areas in the South (Hugo 1995). It would be naive
5 to conceptualise all social units such as households as single-interest
6 decision-making bodies. There is too much evidence on the impor-
7 tance of diverging interests and of power relations within these units,
8 for example expressed in hierarchical and patriarchal decision making.
9 internal dynamics Secondly, the internal dynamics of migration can indeed be de-
10 of migration scribed as self-feeding processes of cumulative causation, usually in
11 ways that reinforce existing staying/moving patterns. Historically,
12 waves of international moving and staying usually had a clearly dis-
13 cernible beginning, a climax and an end – with dynamics somewhat
14 independent even from economic and political conditions in the
15 receiving and sending countries once migration started (Thomas
16 1973). A relational analysis tries to capture the dynamics of migration
17 by a close analysis of collectives (e.g. families or households) and net-
18 works. This implies that international migration is not simply seen
19 as a straight line, only interrupted by external factors that may or may
20 not capture ‘mass migration’. Instead, movers and stayers take ad-
21 vantage of the opportunities offered by macro-level constraints such
22 as demographic, economic and political developments.
23
24 A pioneering exemplar: the polish peasant in Europe and America
25 One exemplar that implicitly sketches theoretical considerations and
26 empirical evidence along these lines is Thomas’s and Znaniecki’s ac-
27 knowledged masterpiece on The Polish Peasant in Europe and America
28 (1918-20). It deals with transatlantic migration of peasants from
29 Russian Congress Poland to the United States. According to Thomas
30 and Znaniecki, the decisions of movers and stayers can be described
31 by reference to the breakup of traditional society, and particularly of
32 its extended-family system due to the marketisation of economic life
33 in the areas of origin. The breakup of the peasant family was said to
34 create new possibilities, especially through the ‘growing assertion of
35 the personality’ (ibid. 2: 217). This evolutionary determinism may be
36 criticised, but the shift from affectual to purposive and rational forms
37 of action is the most relevant aspect of The Polish Peasant for the
38 study of the causes and dynamics of migration. Importantly, Thomas
39 and Znaniecki argued that this development of more abstract, com-
40 plex and cognitive levels of social reorganisation did not entail the
41 disappearance of primary-group attitudes and values but was largely
42 constituted out of them.
43 Newer research has focused on migration not as an expression

of societal disorganisation but as an active strategy to diversify in- 1

come in rural households dependent on crops, etc. Yet what may 2
be needed most for a comprehensive interpersonal and processual 3
account is a focus on migration that includes processes of both soci- societal organisation 4
etal organisation and disorganisation. Clearly, the focus of these two and disorganisation 5
authors on household, communal and other ties remains valuable 6
because it helps to construct the meso-level, whether we focus on 7
disorganisation (e.g. persecution of political refugees) or organisa- 8
tion (e.g. migration as a household strategy for economic survival or 9
even advancement). 10
Thomas and Znaniecki observed that potential migrants can 11
reorganise both in the country of origin and in the new country of 12
settlement. In the former country examples of co-operative collec- 13
tive action included education of peasants through the press and the 14
emergence of co-operative institutions, such as co-operative shops, 15
lean and savings banks, and agricultural improvement societies (ibid. 16
4: 178-304). We could add forms of political voice such as peasant 17
protests (see Scott 1976). Indeed, there were alternatives to moving 18
in improving the life situation in the country of origin. In the main 19
country of destination, the United States, Polish immigrants came to 20
be members of various forms of communal life, ranging from mu- 21
tual aid societies and parishes to cultural organisations. Typically, 22
immigrants such as Poles used their investment in family, ethnic- 23
ity and religion as resources to redefine their situation, as workers, 24
citizens, and members of household and religious groups. A parallel 25
story could be told about political refugees. Although the root causes 26
may differ and options to stay without endangering their lives may 27
be minimal for refugees at the time of flight, the same principles of 28
social analysis could be applied. 29
30
31
The decision-making process 32
33
Social ties and social capital 34
Social relations in collectives and social networks constitute distinct 35
sets of intermediate structures on the meso-level. It is via these social 36
relations that the resources of individuals are related to opportunity 37
structures (figure 7.1). According to rational choice approaches deci- 38
sions to move or to stay are inevitably made by individual or collective 39
actors who weigh the costs and benefits involved. What migration- 40
systems theories emphasise is that these decisions are always made 41
within specific economic, political and cultural contexts that are de 42
termined by larger opportunity structures – reflected in the family, 43
neighbourhood, workplace and community.

72 thomas faist
1 Figure 7.1. Three levels of migration analysis

2
3 MACRO-LEVEL: MESO-LEVEL: MICRO-LEVEL:
4 STRUCTURAL RELATIONAL INDIVIDUAL
5
6 opportunity structures collectives and social values, expectancies and
7 networks recources
8
(political-economic-
9
cultural structure) (social relations) (degrees of freedom)
10
11 economics: social ties: individual values
12 – income and – strong ties: families goals, preferences and
13 unemployment and households; expectancies)
differentials; access – weak ties: networks – improving and
14
to capital of potential movers, securing survival,
15
politics: brokers and stayers; wealth, status, com-
16 – regulation of spatial – symbolic ties: fort, stimulation,
17 mobility (nation- ethnic and religious autonomy, affiliation
18 states and interna- organisations and morality
19 tional regimes);
20 – political repression,
21 etnic and religious
22 conflicts
– interdependence in
23
international system social capital:
24 of states resources available to
25 cultural setting: potential movers and
26 – dominant norms stayers by participa- individual resources:
27 and discourses tion in networks and – financial capital
28 demography and collectives through – human capital:
29 ecology: weak, strong and educational creden-
30 – population growth; symbolic social ties tials; professional
– availability of arable skills
31
land – cultural capital:
32 – level of technology common worldviews,
33 forecast, memories,
34 symbols
35 – political capital: voice
36
37
38 social ties and The macro- and micro-levels of analysis can be connected by
39 social capital the concepts of social ties and social capital. Movers and stayers are
40 embedded in a social-relational context characterised by social ties, a
41 continuing series of interpersonal transactions to which participants
42 attach shared interests, obligations, understandings, memories and
43 forecasts. Strong ties are characterised by direct, face-to-face trans-

actions between the actors involved. They are durable and involve 1
obligations and substantial emotions. They are most widespread in 2
small, well-defined groups such as families, kinship and communal 3
organisations. By contrast, weak ties are defined by indirect relation- 4
ships. They involve no direct or only fleeting contact. Weak ties refer 5
to a more narrow set of transactions. Transactions among ‘friends of 6
friends’ is an apt shorthand for weak social ties. 7
Social capital are those resources inherent in patterned social 8
ties that allow individuals to co-operate in networks and collectives, 9
and/or that allow individuals to pursue their goals.2 Such resourc- 10
es include information on jobs in a potential destination country, 11
knowledge on means of transport, or loans to finance a journey to 12
the country of destination. Social capital also serves to connect indi- 13
viduals to networks and collectives through affiliations. Social capital 14
thus has a dual thrust: it facilitates co-operation between individual 15
(and group) actors in creating trust and links individuals to social 16
structures. Furthermore, social capital serves to mobilise financial, 17
human, cultural and political capital. (For other and differing defini- 18
tions of social capital, see Bourdieu 1983 and Portes 1995.) 19
Social capital is not simply an attribute of individual actors. The 20
amount of social capital eventually available to individuals depends 21
on the extent of the network of social ties that can be mobilised and 22
the amount of financial, cultural and political capital that members 23
of collectives or network participants can muster. In short, social cap- 24
ital is created and accumulated in social relations, but can be used by 25
individuals as a resource. Social capital is thus primarily a meso-level 26
category. 27
The primary question concerning the meso-level is how social 28
capital is created, accumulated and mobilised by collectives and net- 29
works, given certain macro-conditions. Moreover, how is this capital 30
made available to individuals, members and non-members of these 31
collectives? How does it serve to mobilise other forms of capital such 32
as financial, cultural and political capital? It certainly makes a dif- 33
ference whether we deal with first-time movers, return movers or 34
non-movers. For the sake of simplicity this section deals exclusively 35
with first-time movers while the section on the dynamics of migra- 36
tion takes up the issue of return movers and their influence on deci- 37
sion making. 38
Analytically, we can distinguish three different macro-level di- three macro-level 39
mensions for this relational analysis: functional considerations, dimensions 40
normative expectations and collective identity (distinction based on 41
Peters 1993; see also Habermas 1981). On the level of potential mov- 42
ers and stayers we can then make an ideal-typical distinction between 43

74 thomas faist
1 interest-related, norm-oriented, and expressive behaviour and action.

2 Along this typology we are then able to chart various forms of social
3 capital that facilitate decision making in collectives and networks –
4 exchange, reciprocity and solidarity (figure 7.2).
5
6 Figure 7.2. The meso-level: three forms of social capital in interper-
7 sonal relations
8
9 macro-level functional normative collective
10 dimensions considerations expectations identity
11 orientation of
12 interest-related norm-oriented expressive
movers and stayers
13
14 forms of social capital exchange reciprocity solidarity
15
16
17 exchange, reciprocity The first context in which social capital figures prominently is ex-
18 and solidarity change relationships. This is the classical case analysed by rational-
19 choice approaches. Accordingly, migrants move when they expect
20 that they can reap higher benefits in another location. Persons who
21 are involved in aiding these movers (facilitators) can also expect to
22 benefit through material (e.g. money) and immaterial (e.g. social sta-
23 tus) gains. Favours, information, approval and other valued items are
24 given and received in transactions between movers and facilitators
25 (e.g. pioneer movers who return to the place of origin). In the course
26 of social interaction the movers, stayers and facilitators involved ac-
27 cumulate deposits based on previous favours by others, backed by
28 the norm of reciprocity.
29 Reciprocity does not imply that favours given and received must be
30 of the same value or identical. For example, in many cases the head
31 of the family is responsible for the flow of the household income. Yet
32 this does not mean that the head moves himself or herself in order to
33 supply cash. Reciprocity is a form of social capital when at least two
34 norms are adhered to: Firstly, persons help those who have helped
35 them, and secondly, persons should not harm those who helped
36 them before (Gouldner 1960). Reciprocity may serve to increase the
37 financial capital available in collectives such as families or house-
38 holds. Migrant labour is a means to get much-needed cash to supple-
39 ment income earned through crops. In case of crop failure income
40 through labour migration can even act as a temporary substitute. In
41 this case reciprocity would mean that, on the one hand, the moving
42 family members remain loyal and actually send money back home
43 and, on the other hand, the remaining family members work in the

fields. Trust between members of relevant collectives such as fami- 1

lies or households is a very valuable resource upholding reciprocity. 2
This norm-related aspect of reciprocal transactions also refers to the 3
third type of social capital, solidarity. 4
Solidarity is based on a group identity (‘we’) that refers to a unity 5
of wanting and action. It is an expressive dimension to be distin- 6
guished from interest-based and norm-oriented behaviour. The 7
group’s self- and other-definition makes it meaningful to talk about 8
the importance for potential movers that membership of a collective 9
and participation in a network have. 10
Usually, transactions of the exchange type are characterised by social ties and 11
weak social ties, while reciprocity and solidarity require strong so- symbolic ties 12
cial ties (Sahlins 1965). Yet norms of solidarity go with weak social 13
ties, when individual or collective actors feel closely bound to eth- 14
nic, religious and national identities. Movers and recipients may be 15
connected through symbolic ties, characterised by transactions based 16
on shared worldviews, understandings, forecasts and memories. For 17
example, in many African countries borders of nation-states are the 18
result of drawing-board exercises by the former colonial powers, and 19
arbitrarily cut across ethnic and linguistic groups. Refugees who 20
cross international borders are often more generously received by 21
groups with whom they share strong ethnic and linguistic affinities. 22
The existence of symbolic ties across nation-states and the fact that 23
most refugees in the South are movers with few resources explain 24
why many refugees, especially in Africa, end up in countries adjoin- 25
ing the state of origin, and why only a minority ever moves on to 26
countries in the North. 27
Taking Talcott Parson’s distinction between self-orientation and self-oriented and 28
collectivity-orientation as a point of departure (Parsons 1951: 60), we collectivity-oriented 29
can further distinguish between migration decision making that is decisions 30
oriented towards the self and towards relevant collectives. Tensions 31
can arise between, for example, occupational self-fulfilment and the 32
expectation to contribute to the sustenance of the family in the coun- 33
try of origin, as Thomas and Znaniecki have amply demonstrated. 34
For example, movers at the onset of political persecution could de- 35
cide further to support their family (collectivity orientation) although 36
imminent danger of being singled out as a target of violence strongly 37
suggests that they move immediately, albeit individually. To compli- 38
cate matters even further, potential movers are not only members of 39
families but also citizens of a nation-state, members of religious or 40
ethnic groups, etc. In short, they occupy several roles, i.e. there are 41
cross-cutting ties. 42
43

76 thomas faist
1 The difficult transfer of social capital

2 In order to say what contributes to migration or enhances immobil-
3 ity, we have to start from the fundamental insight that many resourc-
4 es are local assets and transferring them to foreign countries would
5 high transferral involve high transferral costs. This does not apply only to the transfer
6 costs of human capital discussed earlier. Networks of social ties connect
7 migrants to other migrants and natives in the receiving country (who
8 hold various amounts of human, financial and political capital). It
9 takes social capital to build such networks and substantial resources
10 are required. It is quite time- and energy-consuming to construct or
11 join new networks in the receiving country, especially in those cases
12 where it is not the whole family that is moving. It is even more dif-
13 ficult to establish and join new collectives. Also, if a mover leaves
14 behind family, friends and other important persons – relationships
15 that are characterised by strong and affective social ties – it involves
16 high costs to maintain these ties while abroad, for example economic
17 costs (return trips) and psychological costs of adjustment to a new
18 environment. Costs are especially high for pioneer migrants who
19 cannot rely on established networks of movers to guide and facili-
20 tate their migration. Only if expected gains in transferring various
21 forms of capital exceed perceived costs are potential migrants seri-
22 ously encouraged to move. In sum, local assets that are undergirded
23 by financial, human, political and social capital can lead a potential
24 mover to prefer in situ adjustment in the sending country to adjust-
25 ment abroad because transferral costs are high.
26 social capital Secondly, social capital is often a prerequisite for the accumulation
27 and mobilisation of human, financial, cultural and political capital. New
28 social ties in the receiving countries have to be well established, be-
29 fore migrants can make use of their financial and human capital or
30 that available to other migrants who may help them in finding work
31 and housing. If there is no access to social capital, it is extremely
32 difficult to invest resources such as money and skills in a beneficial
33 way. This is especially true when there are no pioneer migrants and
34 brokers who act as intermediaries for scarce resources. Moreover,
35 without social capital there is no basis for a rich cultural life in mi-
36 grant communities; for example, no religious institutions will be es-
37 tablished. Similar things can be said about political participation. If
38 migrants do not engage in collective action to voice their interests,
39 they will probably face more discrimination in the receiving coun-
40 tries. For a political voice, they need to form associations.
41 Therefore, we would expect that potential migrants prefer those
42 forms of movement that allow them to keep their social ties intact
43 (circular migration), to interrupt them only briefly (seasonal migra-

tion) or to transfer the whole set of important social ties abroad (e.g. 1
family migration in the context of chain migration). 2
3
The first-time decision-making process 4
We may now conceptualise decision making and dynamics of move- 5
ment in various networks and collectives. The most relevant units 6
constituting meso-levels are households and families, groups of kin- 7
ship, the reference community, but also friends and acquaintances 8
in the workplace, and groupings such as ethnic, religious and politi- 9
cal associations. Interest-guided survival strategies, normative obli- 10
gations of family members to each other and expressions of collec- 11
tive identity are not mutually exclusive realms, the first relating to 12
hard-core purposive (economic) action, the second to the soft fringe 13
of social and the third to the even softer fringe of cultural action. 14
We must analyse the set of social relations that structures decision 15
making and the dynamics of migration, the social connectivity itself, 16
the direct and indirect connections between actors. Here, we have to 17
measure the density, strength, symmetry, range, and so on, of the 18
ties that bind and the transaction and conversion costs and gains of 19
various forms of capital. Furthermore, we must study the cultural 20
content of functional imperatives and normative expectations. 21
Using the threefold typology developed earlier, we can hypoth- exchange 22
esise that exchange relationships, albeit asymmetrical regarding relationships 23
power and authority, may explain why family or household members 24
engage in a division of labour and migration. Thanks to reciprocity 25
as a form of social capital, household members can count on a fair 26
division of burdens and benefits. As a subsistence and socialising 27
unit, the household allocates economic roles and assigns tasks ac- 28
cording to age, sex and kinship ties. It may give incentives to house- 29
hold members – both at home and abroad – to forgo more immedi- 30
ate satisfactions and carry burdens in the expectation that migratory 31
arrangements serve the household and its members in the long run 32
through factors such as acquisition of land, durable consumer goods 33
and improved human capital. Also, reciprocity could lead movers to 34
continue sending remittances home although they do not intend to 35
return. In cases of refugee flows social ties with actors in the country 36
of origin are likely to be severed quite abruptly. Family members are 37
often separated for long time periods. In these situations solidarity 38
between family members really needs a basis in past practices and 39
family bonds, including both reciprocity and solidarity as forms of 40
social capital. 41
On a cautionary note it should always be remembered that fami- 42
lies or households are defined by different economic, political, cul- 43

78 thomas faist
1 tural, demographic and ecological settings and are not social units
2 with universal behaviour (see chapter 8). For example, it certainly
3 makes a difference whether we analyse movement from Africa to
4 Europe or from Latin America to the United States as well as from
5 various communities, regions or countries within these continents.
6 Factors such as household size and expectations directed towards
7 family members are likely to differ, not to speak of the variations
8 pertaining to historical links between sending and receiving regions,
9 current exit and admission policies, income, wage and unemploy-
10 ment differentials between sending and receiving countries.
11
12
13 The dynamics of international migration
14
15 So far, the main question has been why potential migrants decide
16 either to stay or to go. If we consider the dynamics of moving, ques-
17 tions then arise as to what happens after the migrants have moved
18 and why they return to the country of origin or stay in the receiving
19 subsequent country. After an analysis of first-time decisions on moving or stay-
20 developments ing we shall now specify the causal mechanisms that allow us to fol-
21 of choice processes low subsequent developments in the flow of choice processes over
22 time.
23 All the previous conceptual considerations on migration decision
24 making at the different levels of households, kinship relations (e.g.
25 families), friends and even larger groups suggest that there is a con-
26 tinuum along the definiteness of the break of social ties with the ori-
27 gin. Return migration is one case in which strong social ties between
28 sending and receiving regions matter.
29 Historical evidence of earlier transatlantic migrations also at-
30 tests to this thesis: While estimates vary and although most records
31 of immigration are imprecise, return rates probably ranged from
32 25 to 60 per cent for European immigrants in the United States in
33 the late nineteenth and early twentieth centuries (Piore 1979: 110).
34 Sometimes, even permanent migrants retained strong ties with their
35 rural regions of origin; they maintained their location-specific hu-
36 man and social capital, e.g. bought land, built houses, and contrib-
37 uted to village and city projects.
38 Furthermore, leaving and returning may not be decisions taken
39 only once. Empirical research suggests that they occur repeatedly over
40 the life course of a mover. This suggests that space in international
41 migration is inadequately described by focusing solely on countries
42 of origin and destination (see chapter 2). Rather, as international
43 migration proceeds, transnational spaces unfold that cross-cut na-

tion-states. A flow of people, goods, capital and services emerges. In 1

sum, in addition to the interpersonal and inter-group dimension, all 2
these aspects concern the intertemporal dimension of international 3
migration. 4
Three questions have to be addressed: Firstly, how do networks 5
of movers and stayers come into existence? Secondly, how do migra- 6
tion flows turn into ‘chain migration’ – migration as an established 7
pattern that may depart from its original incentive? Thirdly, are there 8
discernible patterns concerning the feedback effects on the sending 9
side? 10
11
How transnational networks are formed and function 12
Exchange relationships partly account for network formation. Clearly, 13
cost-benefit calculations could lead the actors involved to intensify 14
social contacts. Migrant and refugee networks and organisations 15
facilitate social and individual action in reducing information and 16
transport costs as well as costs of integration in the country of desti- 17
nation. For example, migrants may get information about prospec- 18
tive employment by mail or telephone, and for refugees information 19
about reception centres in potential destination countries may be a 20
valuable resource. Also, exchange relations decrease the risk of not 21
finding a job and income in the country of destination. Very often, 22
movers know who awaits them and many probably already know 23
their prospective employer. 24
For the brokers facilitating international migration, migrant net- facilitating social 25
works can provide a lucrative business. Brokers can be pioneer mi- networks 26
grants or refugees who capitalise on their experience, professionals 27
in organisations concerned with labour recruitment, or respected 28
individuals in the sending or receiving communities who facilitate 29
or enable contacts of potential and actual migrants to employers and 30
legal authorities. These brokers or gatekeepers thus turn into trans- 31
national entrepreneurs. They benefit through money or social debts 32
incurred to them in the process of migration (exchange). Yet they 33
are themselves constrained by social norms to respond to legitimate 34
claims for assistance (reciprocity transactions). Exchange relation- 35
ships can also be applied for sending-community strategies chosen. 36
For example, inhabitants in some Mexican villages can best expect to 37
reap results from international migration if they all agree to sponsor 38
selected individuals for graduate studies in the United States (Pries 39
1996). The individualised strategy would be illegal entry in the coun- 40
try of destination. 41
Reciprocity is another source of network formation. For example, 42
when migrants arrive in the country of destination on prepaid tick- 43

80 thomas faist
1 ets, they are expected to pay back the expenses defrayed beforehand.
2 Often only formal agreements and not legal contracts undergird these
3 kinds of transactions between movers and intermediaries. Solidarity
4 may be a prime resource when the actors living and working abroad
5 send back remittances or arrange for their family members to join
6 them in the country of destination.
7 selective access Access to migrant networks tends to be selective. Usually, it is not
8 open for all members of a sending. Access is governed by available
9 information and financial resources, but also by (in)formal norms
10 of reciprocity and solidarity. For potential movers to get access to
11 migrant networks does not necessarily require everyday social in-
12 teraction and direct acquaintance within a community. Indirect so-
13 cial contacts maintained over large geographical distances may also
14 work. Although there is no empirical evidence yet, we can draw on
15 the ‘strength of weak ties’ (Granovetter 1973). The argument here
16 is that weak ties may break more easily, but also transmit distant
17 information on migration opportunities more efficiently under cer-
18 tain circumstances, for example, potential movers may remember
19 persons in destination and sending countries with whom some kind
20 of contact existed in the past, or who know friends who know mi-
21 grants. These persons then serve as brokers of information or even
22 gatekeepers for entry into the receiving countries, and access to jobs
23 and housing. Those to whom potential movers are weakly tied are
24 more likely to move in circles different from theirs and will thus have
25 access to resources such as information different from that of the
26 community of origin.
27 The value of networks for international movers and stayers dif-
28 fers, among other things, by the amount of human, financial, cul-
29 tural and political capital available to the participants. We may hy-
30 pothesise that if the amount of financial, human and cultural capital
31 held by individuals or collectives forming a network is very low, net-
32 works may act to retard the adjustment of movers into the receiving
33 nation-state (see also Pohjola 1991). The reason is that the capacity
34 to employ social capital crucially depends on the amount of other
35 forms of capital the respective network participants can muster. For
36 example, a comparative study on Colombian and Dominican immi-
37 grants in New York City during the 1980s found that movers with
38 higher amounts of human and financial capital were found to be less
39 likely to rely on kin at the place of destination, while movers who had
40 lower amounts of capital depended more on kinship networks to get
41 established (Gilbertson and Gurak 1992). Among others, the latter
42 group relied more heavily on relatives to assist them with housing
43 upon arrival. They received assistance in seeking employment. The

immigrants who reported heavy assistance from family networks 1

when they arrived were also found to be culturally and socially much 2
less integrated in New York. They had less language ability and held 3
lower-status jobs. 4
Not only individuals can participate in networks but also collec- participation of 5
tives such as households, kinship groups or organisations (figure collectives in 6
7.3). networks 7
8
Figure 7.3. Networks of movers and stayers and organisations in inter- 9
national migration 10
11
12
Networks of Movers and Stayers Organisations
13
– sending networks: aid with travel – elite institutional networks 14
arrangements, financial support, (e.g. transnational corpora- 15
etc. tions) 16
– illegal intermediaries – legal/extra-legal agencies
17
(e.g. smugglers) (e.g. recruitment bureaus)
– state labour recruitment 18
(e.g. national labour offices) 19
– refugee-aid organisations 20
(e.g. UNHCR and privately 21
sponsored associations) 22
– receiving networks: aid with – support associations in the 23
legal systems, housing, jobs, receiving country (e.g. human 24
schooling, capital for enterprices, -rights organisations) 25
language training 26
27
Networks with strong ties may constitute secure environments that 28
not only supply valuable information and provide emotional encour- 29
agement (or the opposite!) but often arrange for the subsequent 30
move of members from various collectives. Once migrants have ar- 31
rived at their destination, these collectives lend valuable assistance in 32
adjusting to the new environment, especially in finding housing and 33
employment. Also, the migrant networks in which collectives par- 34
ticipate need not only consist of migrants themselves. Often, patron- 35
client relationships emerge in the employment field between natives 36
and newcomers. 37
Finally, the strongest form of regularization of social interaction 38
is found in various organisations in the field of international migra- 39
tion, which for their purposes apply institutional rules and resources. 40
These may be transnational companies sending personnel abroad 41
(e.g. management and/or construction workers), labour-recruitment 42
agencies (often supervised or even run by state institutions in Asian 43

82 thomas faist
1 sending countries), or human-rights organisations in the countries

2 of origin and destination which extend shelter. The most regularised
3 forms of migrant selection are labour recruitment directly performed
4 by the receiving country in the sending countries (e.g. German
5 labour-office authorities in Turkey during the 1960s), or the selec-
6 tion of refugees in camps near the region of origin (e.g. Canadian
7 government in Africa since the 1980s).
8
9 Chain migration and relative deprivation
10 At some point in migration processes, networks sustain population
11 flows in ways that are less dependent on objective economic-political
12 conditions in the areas of origin and destination (for example, see
13 Shah 1994a: 34). The hypothesis would be that, once the number
14 of network connections reaches a certain level, international move-
15 ments become self-perpetuating because they create the social struc-
16 ture necessary to sustain them. In other words, it is likely that net-
17 works of circular migration – a regular circuit in which migrants
18 retain claims and contacts and routinely return home – transform
19 themselves into chain migration – the following of related individu-
20 als or households (‘friends and relatives effect’). The processes can
21 a ‘snowball’ effect be described as a ‘snowball’ effect: The more immigrants of a given
22 place and state in the destination region, the more want to come. It
23 takes time to develop the chain and this is the reason why we see it
24 fully-fledged only in later phases of international migration. When
25 the accumulated capital finds better opportunities for investment
26 and exchange in the countries of destination, and brokers and gate-
27 keepers find worthwhile benefits in advising and channelling mov-
28 ers (exchange relationships), when norms of reciprocity can be en-
29 forced (e.g. money remitted to family) and when forms of mutual
30 aid among migrants create broad commitments to other migrants
31 (solidarity), networks of movers and stayers begin to flourish.
32 For this to occur, those not yet migrating need to receive informa-
33 tion from earlier migrants, or even to see the concrete results of the
34 ventures of those who migrated before. Therefore, (pioneer) return
35 migrants play an important role in spreading information on oppor-
36 tunities regarding where to go, work and live.
37 However, this does not answer the question of how the process of
38 chain migration starts, given favourable macro-conditions. To name
39 norms, motives, preferences and various forms of capital that guide
40 the behaviour of potential movers does not suffice to account for a
41 phenomenon such as chain migration. We might compare places of
42 origin that are very similar regarding both people’s preferences to
43 move or stay and the opportunity structures they are faced with. Yet

it has been repeatedly observed that the number of people moving 1

abroad from two most similar villages in this regard is not seldom 2
vastly different. In this virtually unexplored area, threshold models 3
of collective behaviour could be used to give situation-specific expla- 4
nations of moving and staying that do not explain outcomes solely 5
in terms of structures, goals and expectancies of actors before the 6
movement begins (Granovetter 1978). Only when we view decisions 7
on moving and staying as being also dependent upon the number or 8
proportion of other potential movers, who must make the decision 9
before another stayer does so, can we start to understand the process 10
of chain migration. The cost-benefit calculations involved in thresh- 11
old behaviour are easiest to follow in the case of strong, symbolic and 12
affective social ties, for example when all family members move to 13
live together abroad. 14
Migration may bring about more migration by changes in social 15
status and income distribution. Relative deprivation theory posits that relative deprivation 16
individual and household satisfaction arise not only from improve- theory 17
ments in absolute economic status but also through comparison with 18
other actors in the reference community. If a potential migrant’s level 19
of income is low, the level of motivation to exit will also be low as 20
long as incomes are low across the board. However, if some actors 21
in the reference community experience an increase, then a poor ac- 22
tor will feel relatively deprived. This can be a direct effect of migration. 23
When household members migrate abroad for work, they earn higher 24
incomes than those available locally, and when they send money 25
home, they increase the amount of income available at the top of the 26
income distribution in the country of origin. This may lead to more 27
international migration. Relative deprivation need not necessarily 28
relate only to income but also to ways of life. For example, in a way 29
that is poorly understood, cultural norms of potential migrants have 30
evolved in the Caribbean to form a veritable ‘culture-of-migration’ 31
(Marshall 1982). 32
One of the key elements introduced by economists into the analy- inverted U-curve 33
sis has been the so-called ‘inverted U-curve’ thesis: development often thesis and S-shaped 34
first enhances and thereafter reduces the scope and incentives for curve 35
migration (see chapter 4). This inverted U-curve depends upon ex- 36
ternal factors such as the level of income (economic development). 37
In addition, we could also speculate about an ‘s-shaped curve’ concern- 38
ing the social diversity of migration (figure 7.4). An s-shaped curve 39
would depend upon factors that arise from the very process of migra- 40
tion itself, i.e. that are internal to migration processes. 41
42
43

84 thomas faist
1
The Crucial Meso-Level
2
3
4
5
6
7 Number or
8 Percentage of
9 Migrants
(Cumulative)
10
11
12
13
14
15 Time
16 Figure 7.4. A stylised S-shaped migration curve
17
18 Massey et al. (1994) found in research on Mexico-US migration that
19 social diversity was low in the initial stages of migration, increased
20 dramatically during the intermediate stages, and then stayed con-
21 stant or fell slightly as a level of mass migration was reached. In this
22 view migration begins with a narrow range of each community’s socio-
23 economic structure, but over time broadens to incorporate other social
24 groups. How could we explain this s-shaped pattern?
25 In an initial period, migration turns into a self-feeding process
26 that gradually encompasses more and more groups and social classes
27 from a local community because of declining costs. In a second pe-
28 riod, the movement becomes somewhat independent of economic
29 conditions in the host country as immigrants acquire social benefits
30 in the receiving country and as family reunification and marriage mi-
31 gration quicken due to guaranteed civil rights and the establishment
32 of immigrant communities. This contributes both to rising numbers
33 of migrants and to less selectivity as to social class. At this stage even
34 children and older kin migrate in growing numbers. There could be
35 spill-over effects even to relatives and friends in other communities
36 in the same country of origin. In a third phase, migration may be-
37 come more selective again; this time in favour of groups that have
38 been underrepresented in the beginning (e.g. members of lower-
39 class or lower-status groups from remote parts of the sending re-
40 gions). Finally, in a fourth period, as migration has captured virtually
41 all groups and classes, the value of migration declines for potential
42 migrants. Those who could not migrate are not only relatively but
43 also absolutely deprived and even socially and economically margin-

alised in the community. Yet all those who could participate in mi- 1
gration had a chance trying to overcome their sense of relative depri- 2
vation vis-à-vis the early movers. As the migrant potential is gradually 3
exhausted in the sending communities, some migrants settle in the 4
country of destination, some return to the country of origin for good 5
– and others, probably a minority, may continue to move back and 6
forth for extended periods of time. Eventually, the volume declines 7
again. 8
9
Cumulative causation: feedback effects in the sending regions 10
Some of the feedback effects of migration that lead to further mi- 11
gration are part of a process called cumulative causation, dating back 12
to Gunnar Myrdal’s use of the term. As is clearly seen by the ‘new 13
economics of migration’, temporary migration may be a strategy of 14
risk diversification in rural households. Foreign wages sometimes 15
lead farmers to farm their land less intensively than before or even 16
let it lie fallow. If these migrants buy land, the outcome might be that 17
there is less land under intensive cultivation in the community, that 18
local food production is reduced, the price of staple crops raised and 19
the demand for labour decreased. These consequences may give in- 20
centives to the remaining members of the community to move, too. 21
Also, if land is more intensively cultivated, as farmer migrants 22
can now afford more capital, this could lead to more out-movement 23
because less manual labour is needed (Massey (1990). However, re- 24
mittances spent on agriculture could actually increase agricultural 25
profits. In some Mexican villages, for example, the money from ‘El 26
Norte’ has helped to develop productivity and output, and migrant 27
farmers have even been able to keep marginal land under produc- 28
tion (Cornelius 1991: 108). In this latter case we could not expect 29
economic feedback effects to encourage further migration. 30
Even very high and increasing levels of migration do not neces- 31
sarily imply the exodus of virtually all potential movers or the settle- 32
ment of all movers in the receiving country. Assets and capital may 33
be location-specific and the transferral costs of social and other capi- 34
tals may keep the volume lower than expected. 35
As to return rates, movers may maintain social ties with the send- 36
ing region and build new ones in the receiving country. Caces and 37
others have tried to capture the first phenomenon on the household 38
or family level by using the concept of the shadow household. It in- shadow household 39
cludes ‘all individuals whose principal commitments and obligations 40
are to a particular household but who are not presently residing in 41
that household’ (Caces et al. 1985: 8). The intensity of their commit- 42
ments or obligations can be operationalised as indicators of house- 43

86 thomas faist
1 hold affiliation. Of course, they may differ from one culture to an-
2 other, and depend on the closeness of kinship and other social or
3 symbolic ties that keep the family or household together.
4 Therefore, decisions over moving or staying made by families and
5 individuals not only influence later decisions made by other indi-
6 viduals and households but also the long-term social and economic
7 arrangements within the families, households and the sending com-
8 munities. Furthermore, changes in the networks and collectives in
9 the country of origin could be expected during the absence of movers
10 and upon their return. For example, migration may entail the real-
11 location of responsibilities which ultimately impact on the roles and
12 status of household members. In the absence of male adult members
13 of the household, the gendered division of labour may change, as
14 women may take over additional roles, or vice versa. Female con-
15 tract workers from Indonesia in the oil-exporting countries of the
16 Gulf have often spent months away from their families, and special
17 arrangements have been made for the care of their children. In ad-
18 dition, there is empirical evidence that the traditional division of
19 labour along gender lines has broken down as women have taken
20 pride in autonomy and competence in handling family affairs in the
21 absence of their husbands, or as men have taken more responsibility
22 in childrearing during the overseas employment of their wives or as
23 women have increased their involvement in financial affairs upon
24 returning home (Hugo 1995).
25 Women more than men may be willing to settle in the receiving
26 countries. For example, male and female migrants from a Mexican
27 village in the United States in the late 1980s differed strongly in their
28 responses to whether they planned to return to Mexico on a relatively
29 permanent basis. In general, women looked much less favourably
30 than men on the idea of returning to live in Mexico. It could be that
31 women may not get a job on the formal labour market there, and that
32 women’s housework in the Mexican countryside generally involves
33 more drudgery than it does in US cities. For men, however, rural
34 Mexico represents a place where tradition is adhered to and men can
35 be men through either work or leisure activities, while the United
36 States remains the place of work, proletarian and spatial discipline,
37 and diminished male authority (Goldring 1995).
38 virtuous and vicious On the community level in the country of origin the feedback ef-
39 cycles fects can be conceptualised as virtuous and vicious cycles: In some
40 cases a virtuous cycle evolves because migration eases the pressure
41 on land and labour. Remittances enable subsistence. However, one
42 also has to consider that the dependence on harvests or crop price is
43 replaced by one on urban wages. Moreover, not only economically,

but also politically, this may strengthen the voice option. This is es- 1
pecially the case when members of groups opposing the political re- 2
gime in the country of origin move back and forth between the two 3
regions. Even political campaigning may take place in the country 4
of destination, e.g. Dominicans in New York City and Algerians in 5
France. Refugees in the country of destination may stay in contact 6
with political activists in the sending country. Sikh secessionists in 7
the United Kingdom and Kurdish activists in Germany constitute 8
clear-cut examples of this. 9
It is equally plausible that a vicious cycle evolves. When labour mi- 10
gration grows in importance, this works against economic and politi- 11
cal co-operation at the village level. Financially, external links might 12
become the most significant and the nexus of social pressures and 13
economic imperatives that held a subsistence-oriented village together 14
could weaken. Here, new forms of solidarity and reciprocity may 15
arise – as described by Thomas and Znaniecki (1918-20). If efforts to 16
build mutually beneficial arrangements of exchange, reciprocity and 17
solidarity fail, however, social disorganisation may ensue that rules 18
out the mutuality and the shared poverty, replacing it with involu- 19
tion and mutual hostility. What Edward Banfield has termed ‘amoral 20
familism’ in Southern Italy is perhaps the accumulation of migra- 21
tion feedback effects in a village that became economically marginal. 22
According to Banfield this effect has been produced by three factors 23
acting in unison: a high death rate – and important for our context – 24
certain land tenure conditions and the absence of the institution of 25
the extended family (Banfield 1958: 10). 26
27
The importance of change and stability 28
One hypothesis is: the stronger the commitment of migrants to so- 29
cial units in the country of origin (not only in terms of strength of 30
social ties – weak and strong – but also regarding the content – reci- 31
procity and solidarity), the more likely it is that return migration of 32
successful migrants takes place. In turn, the higher the rate of this 33
kind of return migration, the greater the likelihood that positive eco- 34
nomic feedback effects occur. 35
To determine the rates of return, we have to ask to what degree 36
the goals of the actual movers could be fulfilled while living abroad 37
and whether a change in their preferences has taken place in the 38
course of their absence from the sending place. Firstly, high rates of 39
return migration may attest to the fact of the successful achievement 40
of some goals involved (e.g. transfer of remittances and skills). Or, goal attainment and 41
alternatively, it could be an indicator that the goals aspired to could social ties 42
not be achieved, a sign of failure. Secondly, return may also indicate 43

88 thomas faist
1 the existence, maintenance and further development of social ties

2 that bind movers to those left behind, sometimes despite the strains
3 and changes created by international mobility. It would indeed be
4 unwarranted to assume that potential and actual movers can only
5 maintain social ties to either side, the sending or the receiving coun-
6 try. Therefore, it is feasible to assume that migrants use social capital
7 to retransfer various other forms of capital.
8 In an age of increasing international migration we can also ob-
9 serve that migrants not only cultivate social ties to the area of origin
10 but, simultaneously, also in the country of destination. At first sight,
11 this is somewhat counterintuitive. There is a continuum regarding
12 social ties between temporary commuting and circulation, on the
13 one hand, and permanent emigration and immigration, on the other
14 hand. Commuting and circulation are terms that denote a great va-
15 riety of movement, usually short-term, repetitive or cyclical in na-
16 ture, but all having in common the lack of any declared intention
17 of a permanent or long-lasting change in residence. They imply few
18 breaks of links with the place of origin and little distance regarding
19 the political, economic and cultural sphere. At the other end of the
20 continuum, permanent emigration and thus immigration are more
21 likely to change significantly the character of social ties and involve
22 greater economic, political and cultural distances. Regarding short-
23 term movements we would expect a higher degree of a sojourning
24 orientation (e.g. towards seasonal and cyclical movement) than in
25 the case of permanent settlement in the country of destination (Tilly
26 1978).
27 The intentions of migrants to stay are relatively clear-cut, if we dif-
28 ferentiate between those who intend to stay permanently and those
29 who come temporarily. However, there are labour migrants or refu-
30 gees who did not come to stay permanently, but eventually settle and
31 still indicate that they wish to return to their homeland. This phe-
32 ‘illusion’ of return nomenon has often been referred to as the ‘illusion’ of return. In these
33 cases we must look not only at the social ties of migrants to persons
34 in the sending countries, but also at the symbolic ties, namely the
35 set of collective representations (e.g. religious symbols), memories,
36 forecasts and worldviews that migrants perceive to have in common
37 with those in the sending countries. The prevalence of symbolic ties,
38 a basis for cultural capital, is one important element in the explana-
39 tion of actual settlement and declared return.
40 In short, it is the differential strength and the content of social
41 and symbolic ties of movers to the place of origin as well as destina-
42 tion that can be used to classify different types of spatial mobility on
43 the domestic and international level across different administrative

units such as nation-states. However, trans-national social spaces 1

suggest that even more permanent settlement in the receiving coun- 2
try does not necessarily imply a complete rupture of social ties and 3
other forms of linkages. 4
The existence of transnational social spaces attests to the ability of transnational 5
movers creatively to pattern their occupational and personal expe- social spaces 6
rience. In this perspective it would often seem appropriate to talk 7
of transnational migrants instead of emigrants and immigrants. We 8
need to develop concepts that can not only be applied in either the 9
sending or the receiving regions but can also refer to emerging trans- 10
national social linkages, such as those between Algeria and France, 11
India and the United Kingdom, Turkey and Germany, and Mexico 12
and the Caribbean and the United States. Glick-Schiller and her as- 13
sociates give a vivid picture of social ties in transnational spaces: 14
15
Whether the transnational activity is sending the barbecue to Haiti, 16
dried fruits and fabric back home to Trinidad so these goods can be 17
prepared for a wedding in New York, or using the special tax status 18
of Balikbayan boxes to send expensive goods from the United States 19
to families back home in the Philippines, the constant and various 20
flows of such goods and activities have embedded within them re- 21
lationships between people. These social relations take on meaning 22
within the flow and fabric of daily life, as linkages between different 23
societies are maintained, renewed, and reconstituted in the context 24
of families, of institutions, of economic investments, business, and 25
finance and of political organizations and structures including na- 26
tion-states. (Glick-Schiller et al. 1992: 11) 27
28
29
Towards a meso-level in international migration 30
31
This analysis suggests that answers to pressing issues of international 32
migration can be found in supplementing the dominant micro- and 33
macro-sociological theories and including an explicit social relation- 34
al perspective. Conceptual meso-levels introduce a distinct layer of 35
analysis to the already rich empirical literature working on this level. 36
Ironically, the study that comes closest to the social relational con- 37
cepts advanced in this appraisal is the one that stood at the begin- 38
ning of the sociology of international movement, namely William 39
I. Thomas’s and Florian Znaniecki’s theoretical-empirical study on 40
The Polish Peasant in Europe and America. These authors have posed 41
the core questions of staying or moving and the feedback effects in 42
a way that also deserves much more attention than it has received 43

90 thomas faist
1 lately. Looking at moving and staying as both an interpersonal and

2 an intertemporal process, we can analyse first-time moves, repeated
3 migration and return migration with the same conceptual tools.
4 Using these tools we come to realise not only that territorial exit is
5 one of several possible strategies to respond to declining or increas-
6 ing opportunities. In situ adjustment and change have to be consid-
7 ered as well. We also pay more attention to the importance of local
8 assets, high transaction costs for social capital and the difficulties
9 involved in converting various forms of social capital because they
10 do not seem to be traded in a common ‘currency’. Also, the analysis
11 of transnational social spaces developing within migration systems
12 offers a way to study the transfer and retransfer of various forms of
13 capital.
14 Moreover, various forms of migration and economic mobility al-
15 possibility for voice ways have to be complemented by the possibility for voice. Sometimes
16 voice is directly or indirectly one of the immediate causes for moving,
17 as in the case of persecution. And even in the case of labour migrants
18 the feedback effects of migration on opportunities to express voice
19 can be important. For example, political activists move between and
20 within both the sending and receiving countries. The current con-
21 flicts surrounding the political role of Islam in West European and
22 North American countries is a vivid case in point. One of the ques-
23 tions to be addressed is to what degree these conflicts are transferred
24 from the sending to the receiving country, and to what extent these
25 politicisation processes are outcomes of migrant adjustment to new
26 centres of work and life.
27
28
29 Notes
30
31 1 The author would like to thank his collaborators in the ‘Migration and
32 Development’ project for fruitful comments. Thanks also go to the au-
thor’s colleagues at the centre for Social Policy Research and at the Institute
33 for Intercultural and International Studies at the University of Bremen.
34 Moreover, various individuals contributed stimulating criticism that some-
35 times differs vigorously from the positions taken by the author: Hartmut
36 Esser, Jutta Gatter, Jürgen Gerdes, Hans-Joachim Hoffmann-Nowotny,
37 Stefan Leibfried, Bernhard Peters, Stefan Sandbrink, Charles Tilly,
Madeleine Tress and Carsten G. Ullrich.
38
2 ‘Social capital… is created when the relations between persons change in
39 ways that facilitate action’ (Coleman 1990: 304).
40
41

1
2
Conceiving and researching transnationalism 3
4
Steven Vertovec 5
6
7
8
9
10
11
12
Anthropologist Steven Vertovec played a crucial pioneering role in the de-
13
velopment of transnationalism studies in Europe. In this article published
14
in 1999 in a special issue of Ethnic and Racial Studies on transnational com- 15
munities, Vertovec reviews the literature on transnationalism and suggests 16
several themes to disentangle the term. He presents transnationalism suc- 17
cessively: as a social morphology, as a type of consciousness, as a mode of 18
cultural reproduction, as an avenue of capital, as a site of political action and 19
as a reconstruction of locality. This categorisation helped to renew the field of 20
transnationalism studies, which was at the time slowly running out of breath 21
and imagination. Furthermore, Vertovec presents a very clear research agen- 22
da for his programme on transnational communities, sponsored by the UK’s 23
Economic and Social Research Council. Today, this programme remains in- 24
tact as one of the major attempts to systematise research on transnational- 25
ism in Europe. It is also one of the more innovative initiatives in the area of 26
migration studies. marge tekst 27
28
29
To the extent that any single ‘-ism’ might arguably exist, most so- 30
cial scientists working in the field may agree that ‘transnationalism’ 31
broadly refers to multiple ties and interactions linking people or in- 32
stitutions across the borders of nation-states. Of course, there are 33
many historical precedents and parallels to such patterns (see, for in- 34
stance, Bamyeh 1993 as well as the introduction to this special issue). 35
Transnationalism (as long-distance networks) certainly preceded ‘the 36
nation’. Yet today these systems of ties, interactions, exchange and 37
mobility function intensively and in real time while being spread 38
throughout the world. New technologies, especially involving tele- 39
communications, serve to connect such networks with increasing 40
speed and efficiency. Transnationalism describes a condition in 41
which, despite great distances and notwithstanding the presence of 42
international borders (and all the laws, regulations and national nar- 43

92 steven vertovec
1 ratives they represent), certain kinds of relationships have been glob-

2 ally intensified and now take place paradoxically in a planet-spanning
3 yet common – however virtual – arena of activity (see among oth-
4 ers, Glick Schiller, Basch and Szanton-Blanc 1992; Castells 1996;
5 Hannerz 1996).
6 Transnationalism represents a topic of rapidly growing interest
7 witnessed in the proliferation of academic articles, university semi-
8 nars and conferences devoted to exploring its nature and contours.
9 While broadly remaining relevant to the description of ‘transnation-
10 alism’ offered above, however, most of this burgeoning work refers
11 to quite variegated phenomena. We have seen increasing numbers of
12 studies on ‘transnational...’ communities, capital flows, trade, citizen-
13 ship, corporations, inter-governmental agencies, non-governmental
14 organizations, politics, services, social movements, social networks,
15 families, migration circuits, identities, public spaces, public cultures.
16 These are obviously phenomena of very different natures, requiring
17 research and theorization on different scales and levels of abstrac-
18 tion. In the excited rush to address an interesting area of global ac-
19 tivity and theoretical development, there is not surprisingly much
20 conceptual muddling.
21 defining It is a useful exercise therefore to step back at this point in or-
22 transnationalism der to review and sort out the expanding repertoire of ideas and ap-
23 proaches so as perhaps to gain a better view of what we are talking
24 about as transnationalism is variously discussed.
25
26
27 Transnationalism as...
28
29 In the Introduction to this special issue of Ethnic and Racial Studies,
30 Alejandro Portes, Luis E. Guarnizo and Patricia Landolt (1999) rig-
31 orously describe the meaning of transnationalism as it pertains to a
32 significant, and arguably new, category of contemporary migrants.
33 While others have approached migration by way of addressing trans-
34 nationalism, Portes, Guarnizo and Landolt emphasize that it is the
35 scale of intensity and simultaneity of current long-distance, cross-
36 border activities – especially economic transactions – which provide
37 the recently emergent, distinctive and, in some contexts, now nor-
38 mative social structures and activities which should merit the term
39 ‘transnationalism’. This is a compelling contribution to theory.
40 In a number of recent works on transnationalism (many of which
41 do not focus on migration) the characteristics of intensity and simul-
42 taneity are also, in different ways, offered as the term’s hallmarks.
43 However, such works offer an often confusing array of perspec-

tives. Nevertheless, theory and research on transnationalism has 1

been grounded upon rather distinct conceptual premises, of which 2
six merit closer scrutiny. The different ‘takes’ on the subject are, of different 3
course, not exclusive; indeed, some rely on others. Nevertheless, the foundations 4
meaning of transnationalism has been variously grounded upon argu- 5
ably distinct conceptual premises, of which six merit closer scrutiny. 6
7
1. Social morphology 8
The meaning of transnationalism which has perhaps been gaining 9
most attention among sociologists and anthropologists has to do with 10
a kind of social formation spanning borders. Ethnic diasporas – what ethnic diasporas 11
Kachig Tölölyan (1991, p. 5) has called ‘the exemplary communities 12
of the transnational moment’ – have become the paradigm in this 13
understanding of transnationalism. To be sure, diasporas embody 14
a variety of historical and contemporary conditions, characteristics, 15
trajectories and experiences (see Tölölyan 1996, Cohen 1997, van 16
Hear 1998), and the meaning of the term ‘diaspora’ itself has been 17
interpreted widely by contemporary observers (Vertovec 1999). One 18
of the hallmarks of diaspora as a social form is the ‘triadic relation- 19
ship’ (Sheffer 1986; Safran 1991) between (a) globally dispersed yet 20
collectively self-identified ethnic groups, (b) the territorial states and 21
contexts where such groups reside, and (c) the homeland states and 22
contexts whence they or their forebears came. 23
Another feature central to the analysis of transnational social for- 24
mations are structures or systems of relationships best described as 25
networks. This is a handle on the phenomena in line with Manuel 26
Castells’ (1996) analysis of the current Information Age. The net- 27
work’s component parts – connected by nodes and hubs – are both 28
autonomous from, and dependent upon, its complex system of rela- 29
tionships. New technologies are at the heart of today’s transnational 30
networks, according to Castells. The technologies do not altogether 31
create new social patterns but they certainly reinforce pre-existing 32
ones. 33
Dense and highly active networks spanning vast spaces are trans- 34
forming many kinds of social, cultural, economic and political rela- 35
tionships. Akhil Gupta and James Ferguson (1992, p. 9) contend that 36
37
Something like a transnational public sphere has certainly rendered 38
any strictly bounded sense of community or locality obsolete. At the 39
same time, it has enabled the creation of forms of solidarity and iden- 40
tity that do not rest on an appropriation of space where contiguity 41
and face-to-face contact are paramount. 42
43

94 steven vertovec
1 Furthermore, Frederic E. Wakeman (1988, p. 86) suggests that the

2 ‘loosening of the bonds between people, wealth, and territories’
3 which is concomitant with the rise of complex networks ‘has altered
4 the basis of many significant global interactions, while simultane-
5 ously calling into question the traditional definition of the state’.
6 In these ways the dispersed diasporas of old have become today’s
7 ‘transnational communities’ sustained by a range of modes of social
8 organization, mobility and communication (see especially Guarnizo
9 and Smith 1998). The examples and discussions concerning trans-
10 nationalism and migration offered in the Introduction to this special
11 issue (Portes, Guarnizo and Landolt 1999) clearly contribute to this
12 perspective. In addition to the longstanding ethnic diasporas and
13 newer migrant populations which now function as transnational
14 communities, many illegal and violent social networks also operate
15 transnationally as well. For the United States Department of Defense,
16 transnationalism means terrorists, insurgents, opposing factions in
17 civil wars conducting operations outside their country of origin, and
18 members of criminal groups (Secretary of Defense 1996). These
19 kinds of cross-border activities involving such things as trafficking
20 in drugs, pornography, people, weapons, and nuclear material, as
21 well as in the laundering of the proceeds, themselves require trans-
22 national measures and structures to combat them (see, for instance,
23 Stares 1996; Williams and Savona 1996; Castells 1998).
24
25 2. Type of consciousness
26 Particularly in works concerning global diasporas (especially within
27 Cultural Studies) there is considerable discussion surrounding a
28 kind of ‘diaspora consciousness’ marked by dual or multiple iden-
29 tifications. Hence there are depictions of individuals’ awareness of
30 decentred attachments, of being simultaneously ‘home away from
31 home,’ ‘here and there’ or, for instance, British and something else.
32 ‘While some migrants identify more with one society than the other,’
33 write Nina Glick Schiller, Linda Basch and Cristina Szanton-Blanc
34 (1992, p. 11), ‘the majority seem to maintain several identities that
35 link them simultaneously to more than one nation.’ Indeed, James
36 Clifford (1994, p. 322) finds, ‘The empowering paradox of diaspora
37 is that dwelling here assumes a solidarity and connection there. But
38 there is not necessarily a single place or an exclusivist nation... [It is]
39 the connection (elsewhere) that makes a difference (here).’
40 Of course, it is a common consciousness or bundle of experiences
41 which bind many people into the social forms or networks noted in
42 the section above. The awareness of multi-locality stimulates the de-
43 sire to connect oneself with others, both ‘here’ and ‘there’ who share

the same ‘routes’ and ‘roots’ (see Gilroy 1987, 1993). For Stuart Hall 1
(1990), the condition of diaspora or transnationalism is comprised of 2
ever-changing representations that provide an ‘imaginary coherence’ imaginary 3
for a set of malleable identities. Robin Cohen (1996, p. 516) develops coherence 4
Hall’s point with the observation that 5
6
transnational bonds no longer have to be cemented by migration or 7
by exclusive territorial claims. In the age of cyberspace, a diaspora 8
can, to some degree, be held together or re-created through the mind, 9
through cultural artefacts and through a shared imagination. 10
11
A wealth of personal and collective meanings and perspectives may 12
subsequently be transformed, such that, as Donald M. Nonini and 13
Aihwa Ong (1997) describe, transnationalism presents us with ‘new 14
subjectivities in the global arena’. 15
Further aspects of diasporic consciousness are explored by Arjun 16
Appadurai and Carol Breckenridge (1989, p. i), who suggest that 17
whatever their form or trajectory, ‘diasporas always leave a trail of col- 18
lective memory about another place and time and create new maps 19
of desire and of attachment’. Yet these are often collective memories 20
‘whose archaeology is fractured’ (ibid). Compounding the awareness 21
of multi-locality, the ‘fractured memories’ of diaspora consciousness 22
produce a multiplicity of histories, ‘communities’ and selves – a re- 23
fusal of fixity often serving as a valuable resource for resisting repres- 24
sive local or global situations. 25
Finally, in addition to transformations of identity, memory, 26
awareness and other modes of consciousness, a new ‘the transna- 27
tional imaginary’ (Wilson and Dissanayake 1996) can be observed 28
reshaping a multitude of forms of contemporary cultural production. 29
30
3. Mode of cultural reproduction 31
In one sense depicted as a shorthand for several processes of cultural 32
interpenetration and blending, transnationalism is often associated 33
with a fluidity of constructed styles, social institutions and everyday 34
practices. These are often described in terms of syncretism, creoliza- 35
tion, bricolage, cultural translation and hybridity. Fashion, music, 36
film and visual arts are some of the most conspicuous areas in which 37
such processes are observed. The production of hybrid cultural phe- 38
nomena manifesting ‘new ethnicities’ (Hall 1991) is especially to new ethnicities 39
be found among transnational youth whose primary socialization 40
has taken place with the cross-currents of differing cultural fields. 41
Among such young people, facets of culture and identity are often 42
self-consciously selected, syncretized and elaborated from more than 43
one heritage.

96 steven vertovec
1 An increasingly significant channel for the flow of cultural phenom-

2 ena and the transformation of identity is through global media and
3 communications. Appadurai and Breckenridge (1989, p. iii) com-
4 ment that
5
6 Complex transnational flows of media images and messages perhaps
7 create the greatest disjunctures for diasporic populations, since in
8 the electronic media in particular, the politics of desire and imagina-
9 tion are always in contest with the politics of heritage and nostalgia.
10
11 Gayatri Spivak (1989, p. 276) describes ‘the discourse of cultural
12 specificity and difference, packaged for transnational consumption’
13 through global technologies, particularly through the medium of
14 ‘microelectronic transnationalism’ represented by electronic bulletin
15 boards and the Internet.
16 Many other forms of globalized media are having considerable
17 impact on cultural reproduction among transnational communities
18 too, for example, diasporic literature (Chow 1993; King, Connell and
19 White 1995). Concerning television Kevin Robins (1998) describes
20 aspects of de-regulation affecting broadcasting regions that effect
21 the emergence of ‘new cultural spaces’ necessitating a ‘new global
22 media map’. The expansion of satellite and cable networks has seen
23 the spread of channels targeting specific ethnic or religious diaspo-
24 ras, such as Med TV for Kurds, Zee TV for Indians, and Space TV
25 Systems for Chinese, Vietnamese, Japanese and Koreans. Viewing
26 is not solely passive, and there are emerging multiple and complex
27 ways in which these media are consumed (see, for instance, Gillespie
28 1995, Morley and Robins 1995, Shohat and Stam 1996).
29
30 4. Avenue of capital
31 transnational Many economists, sociologists and geographers have seen transna-
32 corporations tional corporations [TNCs] as the major institutional form of trans-
33 national practices and the key to understanding globalization (see,
34 for instance, Sklair 1995). This is due not least to the sheer scale
35 of operations, since much of the world’s economic system is domi-
36 nated by the TNCs (Dicken 1992). TNCs represent globe-spanning
37 structures or networks that are presumed to have largely jettisoned
38 their national origins. Their systems of supply, production, market-
39 ing, investment, information transfer and management often create
40 the paths along which much of the world’s transnational activities
41 flow (cf. Castells 1996).
42 Alongside the TNCs, Leslie Sklair (1998) proposes that there has
43 arisen a transnational capitalist class comprised of TNC executives,

globalizing state bureaucrats, politicians and professionals, and con- 1

sumerist elites in merchandizing and the media. Together, Sklair 2
claims, they constitute a new power elite whose interests are global, new global power 3
rather than exclusively local or national, and who thereby control elite 4
most of the world economy. 5
In addition to the Big Players in the global economy, however, the 6
little players who comprise the bulk of transnational communities 7
are making an ever greater impact. The relatively small amounts of 8
money which migrants transfer as remittances to their places of ori- 9
gin now add up to at least $75 billion world-wide (Martin 1994). The 10
scale of this activity has soared over the past thirty years: in Algeria, 11
the value of remittances climbed from $178 million in 1970 to $993 12
million in 1993; in India from $80 million in 1970 to over $3 billion 13
in 1993; and in Egypt from $29 million in 1970 to nearly $5 billion 14
in 1993 (World Bank 1995). 15
Beyond what they mean to the families receiving them, for national 16
governments remittances represent the quickest and surest source 17
of foreign exchange. Indeed, a great number of national economies 18
today, such as the Philippines, Pakistan and many Latin American 19
states, absolutely depend on monetary transfers of many kinds from 20
‘nationals’ abroad. This fact has prompted many countries to develop 21
policies for the ‘transnational reincorporation’ of ‘nationals’ abroad 22
into the home market and polity (Guarnizo and Smith 1998). One 23
often cited case is India, which provides a range of favourable con- 24
ditions for ‘non-resident Indians’ [NRIs] to use their foreign-honed 25
skills and capital to invest in, found or resuscitate Indian industries 26
(Lessinger 1992; ct. The Economist, 6 June 1998). Such policies have 27
impacts beyond the economic dimension. As Katharyne Mitchell 28
(1997b, p. 106) observes, ‘the interest of the state in attracting the in- 29
vestments of wealthy transmigrants widens the possibilities for new 30
kinds of national narratives and understandings’. 31
Resources do not just flow back to people’s country of origin but 32
to and fro and throughout the network. Robin Cohen (1997, p. 160) 33
describes part of this dynamic; anywhere within the web of a global 34
diaspora, 35
36
Traders place orders with cousins, siblings and kin “back home”; 37
nieces and nephews from “the old country” stay with uncles and 38
aunts while acquiring their education or vocational training; loans 39
are advanced and credit is extended to trusted intimates; and jobs 40
and economically advantageous marriages are found for family 41
members. 42
43

98 steven vertovec
1 The strategy is often one of spreading assets (particularly if one of

2 the geographic contexts of activity – ‘at home’ or ‘away’ – is deemed
3 unstable for reasons of political turmoil, racism, legal bureaucracy,
4 shrinking labour market or simply bad business environment).
5 While many transnational communities have found themselves dis-
6 persed for reasons of forced migration (van Hear 1998), others have
7 largely spread themselves for economic reasons. Thus among the
8 Chinese diaspora, Nonini and Ong (1997, p. 4) state that ‘it is im-
9 possible to understand such transnational phenomena unless strat-
10 egies of accumulation by Chinese under capitalism are examined,
11 for such strategies penetrate these phenomena and are in turn af-
12 fected by them’. Yet while economic objectives may be catalyst to the
13 formation of transnational groupings, such activities give rise to a
14 transnational host of others. Transnational activities are cumulative in character,
15 activities Alejandro Portes (1998, p. 14) notes, and ‘while the original wave of
16 these activities may be economic and their initiators can be properly
17 labeled transnational entrepreneurs, subsequent activities encom-
18 pass political, social, and cultural pursuits as well’.
19
20 5. Site of political engagement
21 ‘[T]here is a new dialectic of global and local questions which do not
22 fit into national politics,’ writes Ulrich Beck (1998, p. 29), and ‘only
23 in a transnational framework can they be properly posed, debated
24 and resolved.’
25 Such a transnational framework – a global public space or forum
26 – has been actualized largely through technology. Publishing and
27 communications technologies make possible rapid and far-reaching
28 forms of information dissemination, publicity and feedback, mobili-
29 zation of support, enhancement of public participation and political
30 organization, and lobbying of intergovernmental organizations (see
31 Alger 1997; Castells 1997). Certainly much needs to be done to real-
32 ize the full civic potential offered by these, yet a considerable amount
33 of political activity is now undertaken transnationally.
34 INGOs The most obvious and conventional forms of such activity are
35 represented by international non-governmental organizations
36 [INGOs], including the International Red Cross and various United
37 Nations agencies. Their number has been rapidly increasing and in
38 1993 INGOs totalled 4,830 (Kriesberg 1997). The transnational di-
39 mensions are reflected in their ability to provide and distribute re-
40 sources (especially from constituent bodies in wealthy countries to
41 ones in poorer countries), facilitate complementary or cross-cutting
42 support in political campaigns, and provide safe havens abroad for
43 activities of resistance which are illegal or dangerous in home con-

texts. However many INGOs, claims Louis Kriesberg (ibid), simply 1

reflect the status quo of hierarchy and power. Transnational Social 2
Movement Organizations [TSMOs], on the other hand, are INGOs 3
that seek to change the status quo on a variety of levels. ‘TSMOs,’ ac- 4
cording to Kriesberg (ibid, p. 12) ‘work for progressive change in the 5
areas of the environment, human rights, and development as well as 6
for conservative goals like opposition to family planning or immigra- 7
tion.’ The issues which concern TSMOs themselves are transbound- 8
ary in character, and they draw upon a ‘planetization’ of people’s 9
understandings (Cohen 1998). Citing information published in the 10
1993 Yearbook of International Organizations, Jackie Smith (1997) ob- 11
serves that among 631 TSMOs 27 percent are explicitly concerned 12
with human rights, 14 per cent with the environment, 10 per cent 13
with women’s rights, 9 per cent with peace, 8 per cent ‘world order/ 14
multi-issue’, 5 per cent with development, and 5 per cent ‘self-deter- 15
mination/ethnic’ . 16
Transnational political activities are also undertaken by ethnic ethnic diasporas 17
diasporas. Robin Cohen (1995, p. 13) reasons that ‘Awareness of their 18
precarious situation may also propel members of diasporas to ad- 19
vance legal and civic causes and to be active in human rights and so- 20
cial justice issues’. Yet the nature of much diasporic politics is quite 21
contested. Katharyne Mitchell (1997a) deeply criticizes the assump- 22
tions of many postmodernist theorists (especially Homi Bhabha 23
1994) who contend that hybrid, diasporic ‘third space’ standpoints 24
are inherently anti-essentialist and subversive of dominant hegemo- 25
nies of race and nation. Mary Kaldor (1996) points to the presence 26
of both cosmopolitan anti-nationalists and reactionary ethno-nation- 27
alists within diasporas. And Arjun Appadurai (1995, p. 220) writes 28
that among transnational communities 29
30
These “new patriotisms” are not just the extensions of nationalist 31
and counter-nationalist debates by other means, though there is cer- 32
tainly a good deal of prosthetic nationalism and politics by nostalgia 33
involved in the dealings of exiles with their erstwhile homelands. 34
They also involve various rather puzzling new forms of linkage be- 35
tween diasporic nationalisms, delocalized political communications 36
and revitalized political commitments at both ends of the diasporic 37
process. 38
39
The ‘politics of homeland’ engage members of diasporas or trans- 40
national communities in a variety of ways. The relations between 41
immigrants, home-country politics and politicians have always been 42
dynamic, as Matthew Frye Jacobson (1995) and Nancy Foner (1997) 43

100 steven vertovec
1 remind us with regard to the Irish, Italians, Poles and Jews in turn-
2 of-the-century America. Yet now expanded activities and intensified
3 links are creating, in many respects, ‘deterritorialized’ nation-states
4 (Basch, Glick Schiller and Szanton-Blanc 1994). Political parties now
5 often establish offices abroad in order to canvass immigrants, while
6 immigrants themselves organize to lobby the home government.
7 Increasingly, emigrants are able to maintain or gain access to health
8 and welfare benefits, property rights, voting rights, or citizenship in
9 more than one country (around half the world’s countries recognize
10 dual citizenship or dual nationality; see ‘Traces’ world news digest
11 No.1 on the Transnational Communities Programme website, URL
12 address below). Other forms of recognition have developed as well.
13 For instance, in Haiti, a country that is politically divided into nine
14 departments or states, during President Aristide’s regime overseas
15 Haitians were recognized as the Tenth Department complete with
16 its own ministry (Basch, Glick Schiller and Szanton-Blanc 1994).
17 And in one of the strangest cases of transnational politics, the gov-
18 ernment of El Salvador has provided free legal assistance to political
19 refugees (fleeing their own regime!) in the United States so that they
20 may obtain asylum and remain there, remitting some $1 billion an-
21 nually (Mahler 1998).
22
23 6. (Re)construction of ‘place’ or locality
24 Practices and meanings derived from specific geographical and his-
25 torical points of origin have always been transferred and regrounded.
26 Today, a high degree of human mobility, telecommunications, films,
27 video and satellite TV, and the Internet have contributed to the cre-
28 ation of translocal understandings. Yet nevertheless, these are an-
29 chored in places, with a variety of legal, political and cultural ramifi-
30 cations, not only for the practices and meanings, but for the places as
31 well (cf. Kearney 1995; Hannerz 1996).
32 Some analysts have proposed that transnationalism has changed
33 social fields people’s relations to space particularly by creating ‘social fields’ that
34 connect and position some actors in more than one country (Glick
35 Schiller, Basch and Szanton-Blanc 1992; Castells 1996; Goldring
36 1998). Appadurai (1995, p. 213) discerns that many people face in-
37 creasing difficulties of relating to, or indeed producing, ‘locality’
38 (‘as a structure of feeling, a property of life and an ideology of situ-
39 ated community’). This, he reckons, is due not least to a condition
40 of transnationalism which is characterized by, among other things,
41 ‘the growing disjuncture between territory, subjectivity and collective
42 social movement’ and by ‘the steady erosion of the relationship, prin-
43 cipally due to the force and form of electronic mediation, between

spatial and virtual neighbourhoods’. There have emerged, instead, 1

new ‘translocalities’ (Appadurai 1995; Goldring 1998; Smith 1998). 2
3
4
Researching transnationalism 5
6
The subject of transnationalism is receiving ever greater attention 7
through a range of approaches and disciplines. Nonini and Ong 8
(1997, p. 13), however, are critical of the creeping dilution of research 9
by a cultural studies approach ‘that treats transnationalism as a set 10
of abstracted, dematerialized cultural flows, giving scant attention 11
either to the concrete, everyday changes in people’s lives or to the 12
structural reconfiguration that accompany global capitalism’ (cf. 13
Mitchell 1997a,b). 14
While there is certainly much to be learned about the construc- 15
tion and management of meaning offered by cultural studies, there 16
is immediate need for more, in-depth and comparative empirical 17
studies of transnational human mobility, communication, social ties, 18
channels and flows of money, commodities, information and images 19
– as well as how these phenomena are made use of. In addition to 20
helping us to understand the rapid forms of change (and their his- 21
torical antecedents) which transnationalism represents, more social 22
scientific studies will help us to recognize how and why, as Nancy 23
Foner (1997, p. 23) puts it, ‘some groups [and places] are likely to be 24
more transnational than others – and we need research that explores 25
and explains the differences. Within immigrant groups, there is also 26
variation in the frequency, depth and range of transnational ties’. 27
Luis Eduardo Guarnizo and Michael Peter Smith (1998) outline 28
some serious shortcomings in contemporary theorization of transna- shortcomings in 29
tionalism. Perhaps foremost among these is the question of the ap- transnationalism 30
propriate level of analysis and the connection between scales. In the theories 31
introduction to this special issue, Alejandro Portes, Luis E. Guarnizo 32
and Patricia Landolt (1999) have addressed these issues and made 33
significant strides in establishing, delimiting, analytically defining 34
and typologizing transnational phenomena. 35
George E. Marcus (1995) has provided a useful methodological 36
outline of ‘multi-sited ethnography’ essential to the study of transna- 37
tionalism. Such research involves ‘tracing a cultural formation across 38
and within multiple sites of activity’ (ibid, p. 96) by way of methods 39
‘designed around chains, paths, threads, conjunctions, or juxtaposi- 40
tions of locations’ (ibid, p. 105). Marcus advocates approaches which 41
either ‘follow the...’ people (especially migrants), the thing (com- 42
modities, gifts, money, works of art, and intellectual property), the 43

102 steven vertovec
1 metaphor (including signs and symbols or images), the plot, story or

2 allegory (narratives of everyday experience or memory), the life or bio
3 graphy (of exemplary individuals), or the conflict (issues contested in
4 public space).
5 While broadly concurring with the advantages of such a meth-
6 odology, Ulf Hannerz (1998) adds that ‘the research may need to
7 be not merely multilocal but also translocal... Serious effort must
8 thus be devoted to an adequate conceptualization and description
9 of the translocal linkages, and the interconnections between these
10 and the localized social traffic.’ Hannerz (ibid) also sees the need
11 for collaborative, multidisciplinary teamwork among colleagues in
12 a variety of locations, themselves supported by the new information
13 and telecommunications technologies. Following and drawing upon
14 all these approaches and insights, a major new multidisciplinary re-
15 search programme has been developed with the aim of advancing
16 both our empirical knowledge and theoretical understanding of con-
17 temporary forms of transnationalism.
18
19
20 ESRC research programme on transnational communities
21
22 In 1997 the Economic and Social Research Council of Great Britain
23 [ESRC] launched a £3.8 million ($7 million) research programme
24 on the subject of Transnational Communities. Following a national
25 call for projects, some 170 proposals were received. Together with
26 a Selection Committee comprised of fourteen academics and non-
27 academics, over 250 peer referees contributed towards the final
28 choice of projects to be funded. Nineteen projects have been com-
29 missioned, some within a single discipline, but most linking several.
30 While the programme’s Directorship is based at Oxford University,
31 the projects themselves are managed from a variety of British universi-
32 ties with multi-site research to be undertaken throughout the world.
33 The programme projects will be linked by common methodologi-
34 cal concerns surrounding the formation and maintenance of ‘com-
35 munity’ based especially on social, economic and political networks,
36 the construction and expression of identity focused on the refashion-
37 ing of cultural forms and symbols, and the reproduction or contesta-
38 tion of social relations including issues of gender and power. The
39 four project themes projects are grouped under four themes (which coincidentally paral-
40 lel themes proposed in the introduction to this special issue):
41
42
43

1. New approaches to migration 1

• Comparative Diasporas – commissioned studies within this theme 2
look at notions of incorporation within the Armenian diaspora, 3
Hungarians of Hungary’s periphery, Soviet Jews and Aussiedler 4
(returned ‘ethnic Germans ‘) in Germany; 5
• Transversal Migration – projects here concern the social and cul- 6
tural communities of seafarers and the expansion of transnational 7
Chinese migration circuits; 8
• Refugees and Asylum-Seekers – comprised of comparative research 9
on the role of exiles in post-conflict reconstruction in Eritrea and 10
Bosnia; 11
12
2. Economics 13
• Global Economic Networks – a theme representing a core area of 14
the programme, including a study of the Russian diaspora and 15
post-Soviet economic restructuring, research on British experts in 16
global financial centres, an examination of Chinese global entre- 17
preneurship with special reference to Southeast Asia, plus a study 18
of production and marketing strategies surrounding commodity 19
flows between India and Britain; 20
• Transnational Corporations [TNCs] – focused on a study of 21
Japanese and Korean corporations and their managers in Britain; 22
• Transnational Household Strategies – work assessing the impact 23
of legal status and children on the strategies of female migrant 24
domestic workers in Britain, plus research on remittance patterns 25
among Pakistanis and Bangladeshis in Britain; 26
27
3. Politics 28
• Global Political Networks – includes research on Turkish political 29
networks in Europe and on the indigenous people’s movement 30
and its localization in Ecuador and Bolivia; 31
• City, Region, National and Supra-National Policies – consisting of 32
a comparative study of dual citizenship strategies, of the state and 33
of immigrants, in Canada, Germany and Britain; 34
• Gender, Communities and Power – addressed by a project examin- 35
ing gendered aspects of British and Singaporian transmigration 36
to China; 37
38
4. Society and culture 39
• Social Forms and Institutions – concentrating on a set of three 40
interlinked projects concerning culture flows in societies of the 41
Arab Gulf; 42
• Cultural Reproduction and Consumption – addressed by two teams, 43

104 steven vertovec
1 one concerned with literature and film within a variety of diaspo-

2 ras, the other with the place of broadcast media among Turks in
3 Europe;
4 • Transnational Religious Communities – devoted to a multi-sited
5 study of a prominent Sufi Muslim movement.
6 While conducted independently, the projects will gain a kind of syn-
7 ergy through their coordination as a programme.
8 The programme does not exist solely for the projects, however.
9 Other facets include: a weekly seminar series; an annual conference,
10 each year devoted to one of the programme’s key themes; workshops
11 organized within Britain and abroad focusing on a variety of issues
12 and bringing together academics and non-academics. A Working
13 Paper series including papers by such distinguished writers as
14 Alejandro Portes (1998), Zygmunt Baumann (1998) and Stephen
15 Castles (1998) has been established in both hardcopy and internet-
16 downloadable formats. The Transnational Communities programme
17 will also be supporting a newsletter, world news digest, and three
18 book series. Information on the projects and all other aspects of
19 the research programme can be found on the ESRC Transnational
20 Communities Programme website (http://www.transcomm.ox.ac.uk).
21 transnationalism as Although invoked with a variety of meanings, ‘transnationalism’
22 an umbrella concept provides an umbrella concept for some of the most globally transfor-
23 mative processes and developments of our time. The term’s multi-
24 vocality may actually prove to be advantageous: as Alejandro Portes
25 (1998, p. 2) points out, ‘the concept may actually perform double
26 duty as part of the theoretical arsenal with which we approach the
27 world system structures, but also as an element in a less developed
28 enterprise, namely the analysis of the everyday networks and pat-
29 terns of social relationships that emerge in and around those struc-
30 tures’. The ESRC Transnational Communities Programme, working
31 in conjunction with parallel projects and programmes in Europe,
32 North America and the Asia-Pacific will add significant new data and
33 analyses to test some of transnationalism’s more speculative concep-
34 tualizations.
35
36
37
38
39
40
41
42
43

References 1
2
ALGER, CHADWICK F.1997 ‘Transnational social movements, 3
world politics and global governance’, in Jackie Smith, Charles 4
Chatfield and Ron Pagnucco (eds), Transnational Social Movements 5
and Global Politics, Syracuse, NY: Syracuse University Press, pp. 6
260-75 APPADURAI, ARJUN 1995 ‘The production of locality’, 7
in Richard Fardon (ed.), Counterworks: Managing the Diversity of 8
Knowledge, London: Routledge, pp. 204-25 9
– and CAROL BRECKENRIDGE 1989 ‘On moving targets’, Public 10
Culture, vol. 2, pp. i-iv 11
BAMYEH, MOHAMMED A. 1993 ‘Transnationalism’, Current 12
Sociology, vol. 41, no. 3, pp.1-95 13
BASCH, LINDA, GLICK SCHILLER, NINA and SZANTON BLANC, 14
CRISTINA 1994 Nations Unbound: Transnational Projects, 15
Postcolonial Predicaments and Deterritorialized Nation-States, 16
Amsterdam: Gordon & Breach 17
BAUMAN, ZYGMUNT 1998 ‘Europe of Strangers,’ ESRC 18
Transnational Communities Programme Working Paper No.3 19
BECK, ULRICH 1998 ‘The cosmopolitan manifesto’, New Statesman, 20
20 March, pp. 28-30 21
BHABHA, HOMI 1994 The Location of Culture, New York: Routledge 22
CASTELLS, MANUEL 1996 The Rise of the Network Society, Oxford: 23
Blackwell 24
– 1997 The Power of Identity, Oxford: Blackwell 25
– 1998 End of Millennium, Oxford: Blackwell 26
CASTLES, STEPHEN 1998 ‘New Migrations, Ethnicity and 27
Nationalism in Southeast and East Asia,’ ESRC Transnational 28
Communities Programme Working Paper No.9 29
CHOW, REY 1993 Writing Diaspora: Tactics of Intervention in 30
Contemporary Cultural Studies, Bloomington, IN: Indiana 31
University Press 32
CLIFFORD, JAMES 1994 ‘Diasporas’, Cultural Anthropology, vol. 9, 33
pp. 302-38 COHEN, ROBIN 1995 ‘Rethinking “Babylon”: icono- 34
clastic conceptions of the diasporic experience’, New Community, 35
vol. 21, pp. 5-18 36
– 1996 ‘Diasporas and the nation-state: from victims to challeng- 37
ers’, International Affairs, vol. 72, pp. 507-20 38
– 1997 Global Diasporas: An Introduction, London: University 39
College London Press 40
– 1998 ‘Transnational Social Movements’, ESRC Transnational 41
Communities Programme Working Paper (in press) 42
DICKEN, PETER 1992 Global Shift: The Internationalization of 43
Economic Activity, London: Paul Chapman, 2nd edn

106 steven vertovec
1 FONER, NANCY 1997 ‘What’s new about transnationalism? New

2 York immigrants today and at the turn of the century’, paper for
3 the conference on ‘Transnational Communities and the Political
4 Economy of New York in the 1990s’, New School of Social
5 Research, New York
6 GILLESPIE, MARIE 1995 Television, Ethnicity and Cultural Change,
7 London: Routledge GILROY, PAUL 1987 There ain’t no Black in
8 the Union Jack, London: Hutchinson
9 – 1993 The Black Atlantic: Modernity and Double Consciousness,
10 London: Verso
11 GLICK SCHILLER, NINA, BASCH, LINDA and SZANTON BLANC,
12 CRISTINA 1992 ‘Transnationalism: a new analytic framework for
13 understanding migration’, in Nina Glick Schiller, Linda Basch &
14 Cristina Szanton-Blanc (eds), Toward a Transnational Perspective
15 on Migration, New York: New York Academy of Sciences, pp. 1-24
16 GOLDRING, LUIN 1998 ‘The power of status in transnational social
17 fields’, in Michael Peter Smith and Luis Eduardo Guarnizo (eds),
18 Transnationalism from Below, New Brunswick, NJ: Transaction
19 Publishers, pp. 165-95
20 GUARNIZO, LUIS EDUARDO and SMITH, MICHAEL PETER
21 1998 ‘The locations of transnationalism’, in Michael Peter Smith
22 and Luis Eduardo Guarnizo (eds), Transnationalism from Below,
23 New Brunswick, NJ: Transaction Publishers, pp. 3-34
24 GUPTA, AKHIL and FERGUSON, JAMES 1992 ‘Beyond “cul-
25 ture”: space, identity, and the politics of difference’, Cultural
26 Anthropology, vol. 7, pp. 6-23
27 HALL, STUART 1990 ‘Cultural identity and diaspora’, in Jonathan
28 Rutherford (ed.), Identity: Community, Culture, Difference, London:
29 Lawrence and Wishart, pp. 222-37
30 – 1991 ‘Old and new identities, old and new ethnicities’, in
31 Anthony D. King (ed.), Culture, Globalization and the World-System,
32 Houndmills: Macmillan, pp. 41-Q8
33 HANNERZ, ULF 1996 Transnational Connections: Culture, People,
34 Places, London: Routledge
35 – 1998 ‘Transnational research’, in H. Russell Bernard (ed.),
36 Handbook of Methods in Anthropology, Walnut Creek, CA: Altamira
37 Press (in press)
38 JACOBSON, MATTHEW FRYE 1995 Special Sorrows: The Diasporic
39 Imagination of Irish, Polish, and Jewish Immigrants in the United
40 States, Cambridge, MA: Harvard University Press
41 KALDOR, MARY 1996 ‘Cosmopolitanism versus nationalism: the
42 new divide?’, in Richard Caplan and John Feffer (eds), Europe’s
43 New Nationalism: States and Minorities in Conflict, Oxford: Oxford
University Press, pp. 42-58

KEARNEY, MICHAEL 1995 ‘The local and the global: the anthro- 1
pology of globalization and transnationalism’, Annual Review of 2
Anthropology, vol. 24, pp. 547-65 3
KING, RUSSELL, CONNELL, JOHN and WHITE, PAUL (eds) 1995 4
Writing across Worlds: Migration and Literature, London: Routledge 5
KRIESBERG, LOUIS 1997 ‘Social movements and global transfor- 6
mation’, in Jackie Smith, Charles Chatfield and Ron Pagnucco 7
(eds), Transnational Social Movements and Global Politics, Syracuse, 8
NY: Syracuse University Press, pp. 3-18 9
LESSINGER, JOHANNA 1992 ‘Nonresident-Indian investment and 10
India’s drive for industrial modernization’, in Frances Abrahamer 11
Rothstein and Michael L. Blim (eds), Anthropology and the Global 12
Factory, New York: Bergin & Garvey, pp. 62-82 13
MAHLER, SARAH J. 1998 ‘Theoretical and empirical contributions 14
toward a research agenda for transnationalism’, in Michael Peter 15
Smith and Luis Eduardo Guarnizo (eds), Transnationalism from 16
Below, New Brunswick, NJ: Transaction Publishers, pp. 64-100 17
MARCUS, GEORGE E. 1995 ‘Ethnography inlof the world sys- 18
tem: the emergence of multi-sited ethnography’, Annual Review of 19
Anthropology, vol. 24, pp. 95-117 20
MARTIN, PHILIP 1994 ‘International migration and trade’, HCO 21
Dissemination Notes No. 29, The World Bank 22
MITCHELL, KATHARYNE 1997a ‘Different diasporas and the hype 23
of hybridity’, Environment and Planning D: Society and Space, vol. 24
15, pp. 533-53 25
– 1997b ‘Transnational discourse: bringing geography back in’, 26
Antipode, vol. 29, pp. 101-14 27
MORLEY, DAVID and ROBINS, KEVIN 1995 Spaces of Identity: 28
Global Media, Electronic Landscapes and Cultural Boundaries, 29
London: Routledge 30
NONINI, DONALD M. and ONG, AIHWA 1997 ‘Chinese transna- 31
tionalism as an alternative modernity’, in Aihwa Ong and Donald 32
M. Nonini (eds), Ungrounded Empires: The Cultural Politics of 33
Modern Chinese Transnationalism, London: Routledge, pp. 3-33 34
PORTES, ALEJANDRO 1998 ‘Globalisation from Below: the Rise 35
of Transnational Communities’, ESRC Transnational Communities 36
Programme Working Paper No. 1 37
– GUARNIZO, LUIS E. and LANDOLT, PATRICIA 1999 ‘The 38
study of transnationalism: pitfalls and promise of an emergent re- 39
search field’, Ethnic and Racial Studies (this issue) 40
ROBINS, KEVIN 1998 ‘Spaces of Global Media’, ESRC Transnational 41
Communities Programme Working Paper No.6 42
SAFRAN, WILLIAM 1991 ‘Diasporas in modern societies: myths of 43
homeland and return’, Diaspora, vo1. 1, pp. 83-99

108 steven vertovec
1 SECRETARY OF DEFENSE, UNITED STATES 1996 Proliferation:

2 Threat and Response, Washington, DC: U.S. Government Printing
3 Office
4 SHEFFER, GABRIEL 1986 ‘A new field of study: modern diasporas
5 in international politics’, in Gabriel Sheffer (ed.), Modern Diasporas
6 in International Politics, London: Croom Helm, pp. 1-15
7 SHOHAT, ELLA and STAM, ROBERT 1996 ‘From the imperial
8 family to the transnational imaginary: media spectatorship in the
9 age of globalization’, in Rob Wilson and Wimal Dissanayake (eds),
10 Global/Local: Cultural Production and the Transnational Imaginary,
11 Durham, NC: Duke University Press, pp. 145-70
12 SKLAIR, LESLIE 1995 Sociology of the Global System, London: Prentice
13 Hall, 2nd edn
14 – 1998 ‘Transnational Practices and the Analysis of the Global
15 System’, ESRC Transnational Communities Programme Working
16 Paper No.4
17 SMITH, JACKIE 1997 ‘Characteristics of the modern transnational
18 social movement sector’, in Jackie Smith, Charles Chatfield and
19 Ron Pagnucco (eds), Transnational Social Movements and Global
20 Politics, Syracuse, NY: Syracuse University Press, pp. 42-58
21 SMITH, ROBERT C. 1998 ‘Transnational localities: communi-
22 ty, technology and the politics of membership within the con-
23 text of Mexico and U.S. migration’, in Michael Peter Smith and
24 Luis Eduardo Guarnizo (eds), Transnationalism from Below, New
25 Brunswick, NJ: Transaction Publishers, pp. 196-238
26 SPIVAK, GAYATRI 1989 ‘Who claims alterity?’, in Barbara Kruger
27 and Phil Mariani (eds), Remaking History, Seattle, WA: Bay, pp.
28 269-92
29 STARES, PAUL 1996 Global Habit: The Drug Problem in a Borderless
30 World, Washington, DC: Brookings Institution
31 TOLOLYAN, KACHIG 1991 ‘The nation-state and its others: in lieu
32 of a preface’, Diaspora, vol. 1, pp. 3-7
33 – 1996 ‘Rethinking diaspora(s) : stateless power in the transna-
34 tional moment’, Diaspora, vol. 5, pp. 3-36
35 VAN HEAR, NICHOLAS 1998 New Diasporas: The Mass Exodus,
36 Dispersal and Regrouping of Migrant Communities, London:
37 University College London Press
38 VERTOVEC, STEVEN 1999 ‘Three meanings of “diaspora”, exem-
39 plified among South Asian religions’, Diaspora, vol. 6, No.3 (in
40 press)
41 WAKEMAN, FREDERIC E. 1988 ‘Transnational and comparative re-
42 search’, Items, vol. 42, no. 4, pp. 85-7
43

WILLIAMS, PHIL and SAVONA, ERNESTO U. (eds), 1996 The 1

United Nations and Transnational Organized Crime, Special 2
Issue Transnational Organized Crime, vol. 1, No.3 3
WILSON, ROB and DISSANAYAKE, WIMAL 1996 ‘Introduction: 4
tracking the global/local’, in Rob Wilson and Wimal Dissanayake 5
(eds), Global/Local: Cultural Production and the Transnational 6
Imaginary, Durham, NC: Duke University Press, pp. 1-18 7
WORLD BANK 1995 World Development Report 1995, Washington, 8
DC: The World Bank 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Towards a new map of European migration 3
4
Russell King 5
6
7
8
9
10
11
Geographer Russell King’s work on migration is best described by two key 12
words: interdisciplinarity and innovation. In this article on European mi- 13
gration, first published in 2002 by the International Journal of Population 14
15
Geography, King seeks to advance the knowledge on migration by system-
16
atically questioning what he calls the old dichotomies of migration studies:
17
international versus internal, forced versus voluntary, temporary versus per-
18
manent, legal versus illegal. In the new European age of migration, these 19
binary distinctions are increasingly blurred. The motivations and modalities 20
of migrations are much more diverse than in the past. It is therefore useful to 21
explore relatively new patterns of migration, such as retirement migration or 22
the hybrid concept of tourism-migration. To do so, King invites us to develop 23
an interdisciplinary and comparative approach to human spatial mobility 24
and therefore to stretch the frontiers of migration research in a groundbreak- 25
ing way. 26
27
28
29
Introduction 30
31
Established forms of international migration which have historically 32
been very important (nineteenth-century settler migrations from 33
Europe to the Americas, post-war guest-worker migration from the 34
Mediterranean to northwest Europe, refugee migrations post-World 35
Wars) have for too long now shaped our thinking about how migra- 36
tion is conceptualised and theorised.1 These migrations, and their 37
conceptual codification by writers ranging in time from Ravenstein 38
(1885, 1889) to Sjaastad (1962), Lee (1966), Harris and Todaro (1970) 39
and White and Woods (1980), have led to the assumption, or at least 40
the inference, that all migrants are poor and uneducated. This as- 41
sumption, when applied to European (and other) migrations today, 42
leads to false characterisations: for instance, to the notion that the es- 43

112 russell king
1 sence of a definition of a migrant is someone who is poor, uprooted,

2 marginal and desperate; or to the automatic assumption that all mi-
3 grants from, say, Bangladesh, the Philippines, Senegal or Albania are
4 uneducated and therefore somehow socially inferior to the members
5 of the host societies with whom they interact.2
6
We therefore need to appreciate that many of the key questions
7 that were frequently asked in order to frame our understanding of
8 the functioning of migration as a historically ubiquitous social pro-
9 cess (Why does migration take place? Who migrates? What are the
10 spatial and temporal patterns of flows? What are the effects of mi-
11 gration on the places of origin and destination and on the migrants
12 themselves?) now have a different array of answers than the mainly
13 economic and political ones which shaped our earlier analyses. Even
14 where economic rationales remain paramount, new mobility strate-
15 gies are deployed to achieve the economic (and other) objectives.
16 new geographies of In this paper I attempt to offer an overview of some new geogra-
17 migration phies and typologies of international migration in Europe. My analy-
18 sis will not be a rigorous mapping of the new flows, but rather a qual-
19 itative, even intuitive, exploration of a range of new, and not-so-new,
20 types of migration and mobility. These relatively new forms of mi-
21 gration derive from new motivations at both macro- and micro-level
22 (the retreat from those Fordist-type migrations which were linked to
23 mass production, and from an individual desire to see migration as
24 a route to a stable industrial job), new spacetime flexibilities, various
25 new globalisation forces and new international divisions of labour,
26 and changing views of consumption and self-realisation. Amongst
27 these changes in migration types, patterns and motivations, there are
28 important implications for defining and studying migration which
29 tend to blur further the never-straightforward boundary between mi-
30 gration and mobility, and to melt away some of the traditional di-
31 chotomies which have shaped the study of migration in the past. I
32 propose to deal with some of these conceptual and methodological
33 questions first, and then turn to the new geographies and typologies
34 of migration in Europe.
35
36
37 Towards a more integrated approach to migration studies
38
39 Despite a long history of scholarly study into the field, today migra-
40 tion still tends to remain a dichotomised and fragmented area of en-
41 quiry.
42 More than 30 years ago the sociologist Clifford Jansen (1969: 60)
43 wrote that migration is a problem of many disciplines: it is essen-

tially geographical, involving human movement across space, influ- 1

encing and changing the environments of both the places of arrival 2
and of departure; it is demographic, since it affects the structures of 3
the populations at both origin and destination; it is economic to the 4
extent that many shifts in population (especially of workers) are due 5
to economic imbalances between areas; it may be a political problem, 6
where states feel the need to control or restrict departure or entry of 7
international migrants and refugees; it involves social psychology inas- 8
much as migrants’ motives for leaving and their problems of adapt- 9
ing to the new host society have to be studied; and it is a sociological 10
phenomenon since the social structure and cultural systems, again 11
both in the places of origin and arrival, are affected by migration, and 12
in turn affect the migrant. 13
Anthropologists might feel offended at being left out of the above 14
list, but to some extent their fields of enquiry have been subsumed by 15
Jansen under his definition of migration as a sociological phenom- 16
enon; nevertheless, the important recent research by anthropologists 17
on a wide range of migration-related issues to do with culture, iden- 18
tity, transnationalism and gender deserves more prominent mention 19
here (even if nearly all of it post-dates Jansen’s overview). And the 20
above list is by no means exhaustive, given the interest shown in mi- 21
gration studies by historians, lawyers and human rights specialists, 22
social policy analysts, philosophers, literary and media scholars and 23
others. As the map of learning constantly evolves, so fresh perspec- 24
tives are opened up; in recent years, for instance, migration has come 25
to be seen as a crucial element in cultural studies. 26
The need, therefore, is for an interdisciplinary (rather than a 27
cross-disciplinary or a multidisciplinary) synthesis which brings to- interdisciplinary 28
gether and integrates a range of perspectives, frameworks, theoretical study of migration 29
stances and methodologies in order to study migration (or the vari- 30
ous forms of migration) in a manner which is holistic (embedding 31
migration in its social context) and which recognises its multifaceted 32
diversity. This sounds like a challenging agenda, but it can be (and is 33
being) achieved. 3 Too often, on the other hand, does one read papers 34
which attempt to ‘model’ or ‘explain’ migrant behaviour by reference 35
to economic or psychological variables which seem to have scant 36
linkage with the reality of the migrant experience in the specific con- 37
text in which they are being studied; too often are the economic data 38
upon which some analyses are built insufficiently scrutinised (if they 39
are questioned at all) for the accuracy and relevance of the sources. 40
Too often, also, does one come across qualitative research which has 41
insufficient claim to rigour or representativeness; the insights might 42
be valid for the group studied, but often the reaction is – so what? 43

114 russell king
1 The results generated by a given micro-scale study may be very differ-

2 ent from those of similar groups studied elsewhere, but comparative
3 analysis necessary for migration studies to reach a mature stage as
4 a ‘unidisciplinary’ or ‘postdisciplinary’ branch of the social sciences
5 and humanities is too often lacking.
6 The interdisciplinary study of migration is only achieved over
7 time: by studying migration assiduously in different contexts, by
8 having benefited from an interdisciplinary formation (something
9 not easy to achieve within the UK university system), and by wide
10 reading and engagement with migration scholars from different
11 barrieres to disciplinary and interdisciplinary backgrounds. The objective is to
12 interdisciplinarity overcome single-discipline narrowness – for instance, by exposing
13 the lack of reality and humanity in many econometric studies or by
14 critiqueing the myopia of folkloric studies carried out in some tiny
15 corner of the world – and also to be open-minded towards the nu-
16 merous ideological paradigms which often underlie discipline-based
17 studies (neo-classical economics, Marxist sociology, systems theory
18 in its various forms, theories of transnational identity or hybridity,
19 etc.). Further barriers to a holistic, synthesising study of migration
20 are posed by the division of the migration process into its many
21 fragmented component stages (departure, arrival, return) and by the
22 hegemonic role of national models and discourses of immigration
23 and ethnicity (assimilation, integration, multiculturalism, ius sangui-
24 nis, etc.). In short, disciplinary and paradigmatic closure are the en-
25 emy of an effective, sympathetic study of human migration (Castles,
26 2000: 15-25).
27
28
29 Deconstructing the binaries of migration
30
31 New forms of mobility and migration, and new integrated ways of
32 studying these mobilities, also imply a reappraisal of the longstand-
33 ing heuristic divides within the field of migration study. As will be-
34 come more specifically apparent later on, we need to deconstruct tra-
35 ditional migration dichotomies – or migration dyads as Cohen (1995:
36 6) calls them. Whilst these binaries perhaps continue to have some
37 use for the beginner to construct a mental map of the field of migra-
38 tion studies, they are less solid devices for understanding migratory
39 phenomena in Europe in the late twentieth and early twenty-first
40 centuries. What are these dichotomies? I would list the following.
41
42
43

Process and product 1

The field of migration studies consists of two rather distinct branches 2
and, hence, two rather separate literatures: the study of the actual act 3
of migration as movement across space (often undertaken by geogra- 4
phers and economists); and the study of the ethnic communities and 5
diasporas that are the product of migration (analyses of integration, 6
ethnic relations, cultural characteristics, etc.). Although these two 7
subfields of migration studies are analytically distinct, the linkages 8
between them have been insufficiently explored. Now this is begin- the migratory 9
ning to be rectified by longitudinal or life-history approaches that process and the 10
trace the migrant from origin to destination (and, where relevant, product of migration 11
back again), linking pre- with post-migration characteristics, some- 12
times across more than one generation, and often employing a social 13
networks approach. According to Castles, this dynamic whole, which 14
encompasses all aspects of the lived reality of migrants: 15
16
‘may be referred to as the migratory process, a term which underlines 17
that migration is not a single event (i.e. the crossing of a border) but a 18
life-long process which affects all aspects of a migrant’s existence, as 19
well as the lives of non-migrants and communities in both sending 20
and receiving countries.’ (Castles, 2000: 15-16) 21
22
One might also add that migration influences the lives of other mi- 23
grants in the destinations. The study of transnational communities, 24
for many scholars the new migration paradigm of the last half-dozen 25
years, affords an integration of patterns of movement within the es- 26
tablishment, maintenance and evolution of migrant communities in 27
two or more countries (Glick Schiller et al., 1995; Portes et al., 1999; 28
Pries, 1999; Faist, 2000). 29
30
Internal versus international migration 31
We have another primary distinction between studies of internal and 32
those of international migration. Again, rather separate literatures 33
have evolved, with somewhat different conceptual frameworks and 34
models.4 Only very recently has research begun to link the two scales: 35
searching for common conceptual models; noting how internal mi- 36
gration is often sequenced or interleaved with international migra- 37
tion; examining how international migrants and ethnic minorities 38
are mobile within the host countries; and realising that, as nation- 39
states become less important, so the distinction between internal and 40
international mobility becomes blurred. This is obviously the case 41
within the European Union, and has particular meaning for third- 42
country nationals for whom different types of European boundaries 43

116 russell king
1 (e.g. within and outside ‘Schengenland’) present different (im)per-

2 meabilities for their movement and access to rights.
3 Studies of migration that focus on the household or family have
4 often noted how, within such a unit, different individuals migrate in
5 different ways to different destinations, both internal and interna-
6 tional. Often such a division of labour in migration may be gendered,
7 with a difference between men and women as to who goes abroad
8 and who migrates internally.
9 Another blurring of the difference between internal and inter-
10 national migration occurs when international borders change. The
11 breakup of Yugoslavia or the former Soviet Union, or the unification
12 of Germany, are examples of significant international frontier shifts
13 which affect migration status, in effect turning internal migrants
14 into international movers, and vice versa. This raises an interesting
15 question: are there internal migrants who are destined to become in-
16 ternational migrants at some stage in the future, not through actual
17 movement but through some hitherto unforeseen political event?
18
19 Voluntary versus forced migration
20 There is a commonly-used distinction between voluntary and forced
21 migration. This is the dichotomy used to structure Aaron Segal’s
22 Atlas of International Migration, for instance, together with a third
23 part on diasporas (Segal, 1993). Whilst it is easy to think of migrations
24 which are unequivocally forced (slave migrations, or migrations of
25 ‘ethnic cleansing’ or of religious persecution), as well as those which
26 are unequivocally voluntary (such as Northern European retirement
27 migrants who settle on the Costa del Sol), in practice many migra-
28 tions are not so easily categorised. Is a young Filipina woman sent
29 by her family to work as a domestic helper in Rome or Madrid a vol-
30 untary or a forced migrant? Segal classifies Asian indentured migra-
31 tion as voluntary – a highly dubious categorisation. Clearly there is a
32 complex continuum of coercion and free-will in migration decisions,
33 as some later examples will testify.
34 Such a continuum might contain the following stages:
35
36 • Migrants of ‘free will’, who choose to migrate to satisfy largely
37 non-economic life-choice ambitions – for a better education, or to
38 retire to a pleasant scenic or climatic environment.
39 • Migrants who are encouraged or ‘pushed’ to migrate by life cir-
40 cumstances, such as ‘economic migrants’ seeking to avoid un-
41 employment and very low incomes by seeking better-paid jobs
42 abroad.
43 • Migrants who are more or less compelled to migrate by circum-

stances which are largely beyond their control – extreme poverty, 1

famine, environmental crisis, political chaos, inter-ethnic ten- 2
sion, etc. 3
• People who are forced to migrate by others and who therefore 4
have no control over their decision to move – slave migrations, 5
refugees fleeing to save their lives, extradition, abduction, forced 6
repatriation, or children taken abroad by their parents. 7
8
However, even between these four types there are blurred boundar- 9
ies between the ‘migration forces’ of free-will, encouragement, virtu- 10
al compulsion and force exerted by violence or threat. Moreover, both 11
‘forced’ and ‘voluntary’ migrants can use similar means of migration 12
(e.g. clandestine border crossing by smugglers) and can have similar 13
impacts on destination areas. 14
15
Temporary versus permanent migration 16
Next, we can make a basic distinction between temporary migra- 17
tion (followed by return) and permanent migration (where there is 18
no return).5 This seems a simple enough distinction, but often the 19
intention (to emigrate for good, or to return sooner or later) is quite 20
different from the outcome. Also, there are different degrees of tem- 21
porariness: one year, five, twenty. Return migrants to Greece are 22
defined by the Greek government as those who have lived abroad 23
for at least one year and been resident back in Greece for at least a 24
year, whereas return studies of the Mexico-US labour migration are 25
based on the notion of return to Mexico after ‘settlement’, this be- 26
ing defined as three years continuous residence in the United States 27
(Massey et al., 1987: 310; King, 2000: 9). But the time-space con- 28
tinuum of migration/ mobility is truly continuous; threshold levels 29
are arbitrary tools for statistical measurement (and perhaps too for 30
policy), but they can obscure more than they reveal. For migrants 31
they can have real significance as they trigger residency, citizenship 32
or other rights. Seasonal and shuttle migration of a to-and-fro kind 33
(weekly, monthly, occasional) must also fit into the continuum, blur- 34
ring the distinction between migration and other forms of spatial 35
mobility which, although they may not be regarded as ‘convention- 36
al’ migration, nevertheless carry similar sorts of motivation (for in- 37
stance, economic) and intentionality. 38
Psychologically, many longer-term migrants are torn between 39
the desire to return and the desire (or need) to stay: the ‘myth of re- the ‘myth of return’ 40
turn’ (Anwar, 1979) is just one way of expressing this ambivalence. 41
Another is the notion of ‘being a migrant’ becoming a permanent 42
state of mind: a true home doesn’t exist any more. Perhaps we can 43

118 russell king
1 call this a state of ‘migrancy’ (Chambers, 1994). The construction

2 of transnational communities can be seen as another expression of
3 this condition of being neither (or both) ‘here’ and ‘there’, with the
4 migrant moving back and forth across and within this transnational
5 social and cultural space. For yet others, the true return can never
6 take place, for home is another time, another place – held in the
7 memory by nostalgia but not recoverable because of changes which
8 have occurred in the meantime.
9
10 Legal and illegal migration
11 Reflecting the renewed globalisation of migration over the last 20
12 years, and the increasing perception – in some quarters – of migra-
13 tion as a ‘crisis’ and ‘unwanted’ (hence the growing discourse of
14 ‘migration control’), we can distinguish between legal and illegal mi-
15 gration. Whilst this may be an easy distinction to defend in strictly
16 legal terms, once again the dichotomy fails to match many aspects of
17 contemporary migratory reality. Many are unhappy with the term ‘il-
18 legal’ and prefer terms such as ‘irregular’ or ‘undocumented’. There
19 are many ways of interpreting the growth of ‘illegal’ movement. For
20 some it represents the growing undesirability of ‘mass’ migrations
21 and the need to control and manage migration in the face of appar-
22 ently increasing pressures for people to migrate. For others it is al-
23 most the reverse – a reflection of the fact that the ‘natural forces’
24 of migration will always overcome regimes of control and contain-
25 ment. Hence, is it something to be repressed or a phenomenon to
26 be creatively managed? Moreover, the boundary between legality
27 and illegality is easily crossed. An amnesty or a regularisation law
28 may transform illegal into legal immigrants virtually overnight. Or
29 an apparently unproblematic legal migrant may suddenly become
30 illegal the moment he or she becomes unemployed, is suspected of
31 a petty crime, or fails to renew the permit to stay. Furthermore, a
32 ‘legal’ migrant may work in the ‘illegal’ (or informal) economy; or an
33 ‘illegal’ migrant may work without hindrance in the legal or formal
34 economy. And who defines a migrant as an ‘illegal’? The country of
35 origin, of destination, of transit, or some international organisation?
36
37 Blurring the distinction
38 To sum up this part of the discussion, the multiplicity and variety of
39 types of migration and movement observable today blur the distinc-
40 tion between the migratory dyads, turning them into continua and
41 mixing them up into new matrices and combinations rather than
42 preserving them as readily identifiable polar types. The old certain-
43 ties – if ever they were certainties – disappear. How voluntary is

voluntary migration? How temporary is temporary? What is meant, 1

exactly, by illegal migration? How is movement within the EU’s 2
‘Schengenland’ to be defined: as internal or international migration? 3
For individuals who are frequently on the move, circulating between 4
two or more countries, according to fixed or irregular rhythms and 5
circuits, are they engaging in ‘true’ migration? Or is this some other 6
kind of spatial mobility? 7
Finally, I address the wider question: is migration the exception or 8
the norm? On a world scale, about 150 million people are reckoned 9
to be international migrants, less than 3% of the world’s population 10
(International Organization for Migration, 2000: 5). On the other 11
hand, in Europe (and other parts of the more developed world), only 12
a minority of people are born, live their lives, and die in the same 13
community or settlement; some kind of migration inevitably takes 14
place. I wonder how many of you, reading this paper, have never 15
engaged in some kind of migration. We should also remember that 16
there are many people and cultures in the world whose very existence 17
is based on migration or on a history of migration: nomads, transhu- 18
man shepherds, Roma, international business executives, and so on. 19
So are migrants therefore still to be regarded as the ‘others’ who are 20
different from ‘us’? Or is it the case that all of us are, in some way or 21
another, migrants or the product of migration? Is it not the case that 22
migrants are the perfect exemplars of the post-modern condition? 23
And if so, does this not bring migration studies from the fringes of 24
the social sciences and the study of humanity in to its very core? The 25
postmodernist emphasis on permeability of borders, connectivities 26
and identities lends itself by nature to the study of migration; and the 27
study of migration, in response, shifts its focus to a new emphasis on 28
culture, subjectivity and identity, reflecting the general cultural turn 29
in the social sciences in the past couple of decades (Cohen, 1995: 8). 30
31
32
New motivations and settings for migration 33
34
At a more concrete level, new connectivities, new space-time flex- 35
ibilities, and the embedding of migration/mobility within the forc- 36
es of globalisation, have served to blur the correlative conception globalisation 37
of migration as a measurable spatio-temporal phenomenon (i.e. a 38
movement across a threshold distance for a specified threshold of 39
time). New mobilities have emerged which confound the conven- 40
tional divide between migration on the one hand and other forms 41
of human spatial mobility on the other – travel, tourism, circulation, 42
commuting. Globalisation and the post-1989 New World Order cre- 43

120 russell king
1 ate new geographies of movement into and around Europe – from

2 new globe-spanning migrations which have no historical precedent,
3 to local-scale cross-border dynamics where none existed for half a
4 century before.
5 The motivations, too, have fundamentally changed. Under the
6 earlier migration epochs of European transatlantic settlement and
7 postwar European labour migration, linked to the relatively fixed
8 parameters of the respective productive regimes of colonialism and
9 Fordism, the migration variables were more or less certain – the
10 destination, the type of job, the level of pay, the means of transport,
11 privatised the likelihood of stay or return. Now migrants’ motives, and the out-
12 international comes of their actions, are far more diverse, as are their geographi-
13 migration regime cal origins, destinations, routes and modes of travel. As ‘Fortress
14 Europe’ imposes its own logic of migration control, new migration
15 processes and patterns open up, driven by new market dynamics.
16 Migration has become a new global business with a constantly shift-
17 ing set of agents, mechanisms, routes, prices and niches. Very differ-
18 ent from the Fordist labour migration system of Europe in the 1960s
19 and early 1970s, the new migration regimes of the 1980s and 1990s
20 were based on fast-evolving European and global conditions: the
21 escalation of push pressures from the global South, the new-found
22 economic prosperity of southern Europe (combined with ease of en-
23 try), and the removal of the Iron Curtain as a barrier to emigration
24 (only for it to be partly replaced by a West European set of barriers
25 to immigration including a new ‘Fortress Europe’ frontier along the
26 border of the former Soviet Union). Episodic migrations of crisis and
27 flight from political turmoil and environmental catastrophe add to
28 the cocktail of new migration factors.
29 Consistent with the post-Fordist ‘privatisation’ of migration and
30 with the creation of a kind of ‘migration plc’ come other market con-
31 cepts: growth in the number of agents, intermediaries, traffickers,
32 and a pricing structure for each route, each origin nationality and
33 each destination country. Within this new privatised, semi-illegal
34 international migration regime, some migrants set off with no par-
35 ticular destination country in mind: they go where the agents and
36 smugglers take them, or abandon them. Others are able quite explic-
37 itly to ‘shop’ for opportunities and destinations, measuring the costs
38 and benefits of risk, insecurity, quality of life, anticipated income,
39 cultural (un)familiarity, and existence of social and kin contacts.6
40 These types of migrant, described above, are still largely to be
41 characterised as ‘economic migrants’ although they do differ from
42 the classic ‘labour migrant’ type where recruitment is managed by
43 the host country. Another difference is the diverse educational, skill

and status levels of recent immigrants to Europe, whether they come 1

from (say) Morocco, Kurdish areas of Iraq or Bangladesh. Many are 2
highly educated and some have considerable professional experi- 3
ence, but the opportunities available to them are severely restricted 4
to the low-status jobs rejected by West European nationals. Rhode 5
(1993) has described this phenomenon as ‘brain waste’; highly edu- 6
cated migrants and refugees are attracted to menial jobs in Europe 7
because the pay they get, even for cleaning houses or selling newspa- 8
pers at street corners, is much higher than pursuing a professional 9
career in their home countries where jobs are often extremely scarce 10
and incomes very low and unreliable. 11
Yet, perhaps reflecting their more educated background and their 12
possession of a kind of anticipatory socialisation into West European 13
culture by their consumption of global media and images of Western 14
lifestyles, their motives are not necessarily purely economic. For 15
many of these migrants, moving to Europe, by whatever means (and 16
often the price is very high), is part of a dream of self-realisation. self-realisation 17
Their migration may be a gesture of escape, an adventure, a rite of motives 18
passage (King, 1996). Shuttleworth and Kockel (1990), in their study 19
of young Irish emigrants, have described this type of emigration as 20
‘emigration as walkabout’. Hence to the traditional economic motiva- 21
tion of labour migration we add other rationales: excitement, experi- 22
ence, leisure, ‘seeing the world’. Migration itself becomes a desirable 23
act rather than an economic means to an end: a consumption good 24
rather than a strategy which satisfies the production needs of another 25
country’s economy or the private survival needs of an individual mi- 26
grant; and the projection of an individual’s identificatory experience 27
beyond what are perceived as the restricting confines of his or her 28
own country. 29
30
31
New European migrations: some examples 32
33
To list fully all ‘new’ forms of migration affecting Europe is beyond 34
the scope of this paper, quite apart from the issue of what is new 35
and what is not. What I have tried to do in the preceding sections of 36
the paper is to set out some of the new contexts for recent migratory 37
phenomena and to link these to the need for changing approaches 38
to how we define and study migration/mobility. Let us now be more 39
specific and examine a selection of new migration types and flows in 40
Europe. The following is by no means an exhaustive list and is sub- 41
ject to the caveats drawn above. The list extends and elaborates some 42
of the types identified by Cohen (1997) and its time-frame is roughly 43

122 russell king
1 the last 15 years, since the European migrations of Fordism, fam-

2 ily reunion and post-Fordist economic restructuring (King, 1993a;
3 Blotevogel and King, 1996; Koser and Lutz, 1998).
4
5 Migrations of crisis: refugee, irregular and ‘illegal’ migrations
6 One of the main features of the global and European map of migra-
7 tion since the mid-1980s has been the strong growth in refugee mi-
8 grations, especially in respect of people who do not satisfy the 1951
9 UN convention definition of a ‘well-founded fear of persecution on
10 account of race, religion, nationality, membership of a particular
11 social group, or political opinion’, and who are thus condemned to
12 remain asylum-seekers or displaced persons. The UN definition of
13 refugees is being rendered out-of-date by political, religious, ethnic
14 and environmental crises. At the same time, there has been a sharp
15 increase in the phenomenon of ‘illegal’ or irregular migration. An
16 crisis-driven estimated 500,000 foreigners entered the EU clandestinely in 2000,
17 migrations five times the number estimated to have entered in 1994 (Ratnesar,
18 2001). Of course, such estimates must be regarded as highly approxi-
19 mate given the obvious problems of measuring clandestine migra-
20 tion, but few would dispute the general trend to a marked increase.
21 This has happened in response to strong push factors operating from
22 the countries of origin, and in the context of increasingly harsh re-
23 gimes of immigration control imposed by West European states, in-
24 cluding stricter criteria and more rigid and mechanistic processing
25 of asylum-seekers’ claims for refugee status. Two main mechanisms
26 of irregular migration can be recognised: deliberate illegal entry
27 (forged documents, landing on remote southern European coasts,
28 crossing poorly guarded borders, etc.); and legal entry (e.g. on a tour-
29 ist visa) followed by overstaying. Increasing evidence exists for the
30 orchestration of illegal entry by semi-criminal organisations – ma-
31 fia groups, traffickers and agents at various points in key smuggling
32 routes. Laczko and Thompson (2000) and Salt (2000) have provided
33 useful overviews of human smuggling and migrant trafficking in
34 Europe, including conceptual issues, bibliographic surveys and sta-
35 tistical estimates.
36 Crisis-driven migrations affecting Europe as a destination can oc-
37 cur in any part of the world. Since 1990 they have emanated from
38 the Gulf War, the persecution of Kurdish populations in Turkey and
39 Iraq, war and famine in various parts of Africa, and the break-up
40 of the former Yugoslavia. In the last of these cases, war and ethnic
41 cleansing led to massive displacements of population, both within
42 the region and, more particularly, the 1 million Bosnians who be-
43 came refugees in Western Europe in the early to mid-1990s, many

of whom have now been pressured to return in the wake of the 1995 1
Dayton Agreement which ended hostilities in Bosnia and provided 2
for the planned repatriation of the displaced and refugee popula- 3
tions. 4
The Albanian emigration of the last ten years is a good example Albanian 5
of how the notion of ‘crisis’ can differentially interact with migration, emigration 6
producing a continually evolving dialogue between the two terms 7
(Pastore, 1998): 8
9
• Firstly, the mass exodus of 1991 can be seen as a direct response 10
to the Albanian political, economic and social crisis accompany- 11
ing the abrupt post-Communist transition. This is an obvious 12
point, but a deep understanding of the Albanian context is nec- 13
essary to comprehend fully the resultant dynamics of migration 14
to Italy and Greece. To view Albanians fleeing their country in 15
the early 1990s as either refugees or economic migrants fleeing 16
political chaos and economic collapse is too simplistic. As Mai 17
(2001) shows in an interesting analysis of the role of Italian tele- 18
vision in the Albanian emigration, the collapse was also a moral 19
and an ethical one. Young Albanians, in particular, were suffering 20
a collective identity crisis which counterposed a forced, ethicised 21
identity of the heroic nature of work in an Albanian society that 22
was projected by Enver Hoxha to his information-starved people 23
as ‘paradise on earth’, with the increasing identificatory appeal of 24
la dolce vita on the other side of the Adriatic. 25
• But the migration of the early 1990s provoked further crisis in 26
Albania, leading to both short- and longer-term instability. The 27
key to this vicious cycle of linkages was the investment of mi- 28
grant remittances in informal pyramid savings schemes which 29
collapsed in early 1997, bankrupting the majority of the Albanian 30
population and provoking a second mass exodus. Longer-term 31
effects of emigration on the re-making of the Albanian crisis, 32
according to Pastore (1998), were the establishment of crimi- 33
nalised emigration rackets and the demographic distortion of the 34
Albanian population by the emigration of so many young people. 35
• Thirdly, the Albanian migrations were also seen as a crisis for 36
the Italian and Greek states: how were they to deal with the tens 37
of thousands of migrants entering the country without legal 38
documents and by whatever means they could find? As Lazaridis 39
(1996) and Zinn (1996) have shown, policy paralysis, dithering 40
and contradictions have been the main reactive outcomes in both 41
countries. This has had the effect of prolonging and almost insti- 42
tutionalising the ‘crisis’ nature of the Albanian immigration into 43

124 russell king
1 a kind of semi-permanent feature of the Italian and Greek politi-

2 cal and press discourse, which tends overwhelmingly to stigma-
3 tise Albanians as ‘criminals’ (Jamieson and Silj, 1998; Lazaridis
4 and Wickens, 1999). On the one hand this might be thought to be
5 a negation of the very meaning of the term ‘crisis’; on the other,
6 it asks important questions about how media representations of
7 migrants come to be constructed, and about the power of such
8 representations to influence public opinion.
9
10 ‘Sisters are doing it for themselves’: growth in independent female
11 migration
12 Until the early 1980s, there was an overwhelming and regrettable
13 trend to consider women migrants as dependants or followers of ‘pri-
14 mary’ male migrants – to use Cohen’s (1997) phrase, as the ‘baggage
15 of male workers’. Numerically and sociologically (Cohen’s words
16 again), we have entered a new phase of female migration, charac-
17 terised by the independent migration of females in response to the
18 needs of the European and global service economies.
19 Campani (1995) and Phizacklea (1998) have been important voic-
20 es in the debate on the contemporary global and European contexts
21 ‘sex, marriage for female migration. ‘Sex, marriage and maids’ describe, somewhat
22 and maids’ over-simplistically, the three sectors of activity which are important
23 for female migrants in Europe (Phizacklea, 1998: 31-4), but few data
24 are available to quantify the relative importance of these three fe-
25 male migratory types – the migration (including trafficking) of sex-
26 workers, the international bride trade, and the migration of domestic
27 and care workers. More broadly, it is important to realise how the
28 demand for women migrants has increased through the centrality
29 of the types of service activities in post-industrial society which have
30 traditionally been associated with female labour or are those which
31 only women are willing to supply (Campani, 1995: 546).
32 There has been quite an impressive amount of literature on fe-
33 male migration experiences in Southern Europe published since the
34 late 1990s; of particular value are the collections edited by Anthias
35 and Lazaridis (2000) and Ribas Mateos (2000). Anthias notes that
36 women migrants provide the flexibility and low cost that appeals
37 both to global capital, and to middle-class households seeking to hire
38 domestic help: ‘they are located in or within a secondary, service-
39 oriented, hidden (economy) ... that reproduces an ethically and gen-
40 dered divided labour market’ (Anthias, 2000: 25). Against this struc-
41 turalist perspective are a number of ethnographic documentaries
42 which tell a variety of stories – of exploitation and empowerment,
43 of patriarchy and liberation, of isolation and solidarity, of sacrifice

and achievement (for some accessible studies in English, see Andall, 1

1998, 1999, 2000; Chell, 1997, 2000; Escrivà, 1997; Lazaridis, 2
2000; Zontini, 2001). More often than not, women are the social 3
glue which holds the ethnic community together, especially in na- 4
tional communities (e.g. Filipinos in Spain and Italy, Cape Verdeans 5
in Italy) where women migrants were the pioneers and where they 6
remain in numerical dominance. Moreover, they are playing increas- 7
ingly active roles in processes of integration with the host society, be- 8
coming important agents of cultural change. Undoubtedly, migrant 9
women in Southern Europe are at the forefront of the interesting 10
work being done in gender issues in migration in recent years. 11
12
Playing the global labour market: skilled and professional migrants 13
This type of migration has been thoroughly researched by Salt (1984, 14
1992) and Findlay (1993) since the mid-1980s. Hence it is question- 15
able how new it is for the European setting, although new flows of 16
skilled migration have emerged from Eastern Europe since 1989 skilled migration 17
(Rhode, 1993). Findlay and Salt write about professionals, business 18
executives, accountants, engineers, consultants and the personnel of 19
international organisations. To these I would add sports stars and 20
entertainers. The flows, by and large, are not one-way but multidi- 21
rectional and temporary, although East-West flows are likely to be 22
more permanent and unidirectional. This is a fluid type of migration 23
which merges with (and is tending to be substituted by) business 24
travel and short-term contract and trouble-shooting visits (Salt and 25
Ford, 1993). Highly-skilled and professional migration also overlaps, 26
at the individual level if not conceptually, with the next two types 27
I am going to consider: cross-border shuttle migration and student 28
migration. 29
The movement of skilled persons lies at the heart of attempts to 30
integrate Europe through the free movement of people, goods, ser- 31
vices and capital within the EU. This increasing ease of movement 32
for elites and highly-skilled labour creates a polarisation of migration 33
types with, at the other end, ‘poor’ immigrants and asylum-seekers 34
from outside the EU. This emerging hierarchical division is one of 35
the clearest contrasts opening up in the new map of European migra- 36
tions (Koser and Lutz, 1998: 2). 37
38
Here and there and back and forth: shuttle migration 39
The bipolar fixity of conventional studies of migration – based on an 40
origin, a destination, and a more-or-less definitive and statistically 41
measurable relocation between the two – has been challenged both 42
by the heightened role of mobility in (European, Western) society 43

126 russell king
1 at large (Urry, 2000), and by new geographies and temporalities of

2 movement (Cwerner, 2001). Now, many movements are multiple
3 and spatially capricious – in Kevin McHugh’s catchy words, ‘in-
4 side, outside, upside down; backward, forward, round and round’
5 (McHugh, 2000).
6 The dual role of borders and frontiers is interesting here: on the
7 one hand the removal of frontiers within the EU facilitates an inten-
8 sification of mobility between and across the states of the Union;
9 on the other, the juxtaposition of countries at different levels of eco-
10 nomic development and with different social and cultural systems,
11 notably inside and outside the EU frontier, creates the conditions for
12 new dynamics of movement. Much of this may be ‘illegal’, but much
13 of it represents an accommodation of new economic mobility types
14 to the visa and access regimes that are imposed by the EU.
15 In particular, since 1989 there has been a sharp rise in cross-bor-
16 der shuttle migration across the eastern frontier of the EU; this has
17 tended to replace the mass East-West migrations originally feared
18 by the West as soon as the Iron Curtain was dismantled. Although
19 some instances of cross-border shuttle migration are of fairly long
20 standing (e.g. that of Slovenians to Trieste), others have risen with
21 dynamic new rhythms during the 1990s, for instance the migration
22 of Poles to Germany (Iglicka, 2000). Iglicka distinguishes shuttle or
23 pendular migrants (who stay for less than three months) from short-
24 term migrants (more than three months, less than one year), long-
25 term migrants (more than one year), and settler migrants (such as
26 the Aussiedler). Cross-border shuttle migration can be for short-term
27 work opportunities, for instance in construction or agriculture, or for
28 trading – buying and selling of goods with different prices and mar-
29 ket situations either side of the border. It is important to emphasise
30 how this type of movement is facilitated by, and represents an adap-
31 tation to, the availability of tourist visas; it is also important to realise
32 that many trips are multipurpose, combining tourism and shopping
33 with trading and short-term work.
34
35 Student migrations: from the year abroad to the ‘Big OE’
36 Student migrations are a long-overlooked but increasingly important
37 form of European mobility. Whilst there are some historical paral-
38 lels (the medieval wandering scholar, or colonial patterns of stu-
39 dent migrations to France, the UK, the Netherlands, etc.), since the
40 1980s student mobility within Europe has been strongly promoted
41 by the European Commission via schemes such as the Erasmus and
42 Socrates exchanges, whose initial aims – which look unlikely to be
43 achieved, at least for the forseeable future – were to have one in ten

students studying at a university in another EU country.7 Since the 1

launch of the Erasmus scheme in 1987, around 750,000 universi- 2
ty/third-level students have spent a period of 3-12 months studying 3
abroad; this figure covers the academic years 1987-88 to 1999-2000. 4
Numbers have grown steadily year-on-year, with a seven-fold in- 5
crease in annual movers between the late 1980s and the late 1990s. 6
On a broader front it is important, once again, to recognise the 7
variety of migratory subtypes under this general category. Student 8
migrations are an important part of the internal mobility of young 9
adults within European countries, particularly the UK, where there 10
has been a continuous tradition of ‘going away’ to university (in 11
many other European countries the dominant tradition has been for 12
university students to live at home). Surprisingly, the migrational 13
significance of students going to university has scarcely been stud- 14
ied. This significance lies in two areas: the initial move to univer- 15
sity, and the implications of this for subsequent national population 16
distribution (do students tend to stay on in their university towns, 17
return home, or move elsewhere?). For international student migra- 18
tions, such as those involving a ‘Year Abroad’ at a foreign university, 19
the same questions arise: do students tend to preserve their affective 20
and institutional links to their Year Abroad destination, or are their 21
future migration propensities unaffected? 22
It is also possible to see student migration as a subset of youth mi- 23
gration motivated by a mixture of broader educational goals and ex- 24
perience/travel/pleasure-seeking, perhaps facilitated or interleaved 25
with casual or temporary work. Amongst European students, espe- 26
cially those from northern countries, the ‘gap year’ between school 27
and university, or between graduation and employment, exemplifies 28
this, as does young Australian and New Zealanders’ predilection for 29
their ‘Big Overseas Experience’. Here, again, we see complex over- 30
lappings of socio-economic and mobility types (students/workers/ 31
tourists, but also travellers, drifters, hobos...) which defy neat migra- 32
tory and motivational categorisations (Bianchi, 2000). 33
34
Love migrations: the transnationalisation of intimacy 35
The explanation of contemporary migrations increasingly with ref- 36
erence to individual and personal factors (which, nevertheless, at a 37
societal scale have considerable significance) opens up other possi- 38
bilities. Students and tourists travel, study abroad, have sex, fall in 39
love.8 Their subsequent locational behaviour and mobility/migration 40
regimes may be more related to this libidinal factor than to any other. 41
Indeed, love migration can probably be found in all types of migra- 42
tion. Maybe, as far as migration factors are concerned, ‘love conquers 43

128 russell king
1 all’. The possibility for the initiation of such ‘transnational intimacy’

2 is greatly increased by mass travel, study abroad, and tourism; whilst
3 the accelerating speeds and technologies of travel and communica-
4 tion in a shrinking Europe increase the chances of such transnational
5 love being maintained.
6 Technology apart, several important global sociological factors lie
7 behind the growth of love migrations within (and outside) Europe.
8 The expansion of linguistic competence is one factor (young Britons
9 are an exception here: hiding within their global language, they are
10 less multilingual than their other European counterparts). Another
11 is the linked expansion of the ‘global experience’ industries (tourism,
12 travel, leisure, education, networking) with the extension of youthful
13 attitudes and lifestyles to later ages. Together these factors produce
14 an expansion of individual transnational interfaces resulting from
15 mobility and migration; the major cities (London, Paris, Brussels,
16 Frankfurt, Barcelona, Rome, Geneva – the list is much more exten-
17 sive), especially those with explicit multinational functions, are the
18 principal nodes for this intensification of cross-national personal
19 contact, relationships, partnerships and marriages. My conclusion is
20 simple: do not underestimate the libidinal factor in migration. You
21 read it here first.
22
23 Heliotropes and rural idyllists: migrations of environmental
24 preference
25 Finally, there has been a steady growth in what we might call environ-
26 mental preference migration (Williams et al., 1997). These are migra-
27 tions that are the very antithesis of being economically motivated,
28 and are undertaken by those who prioritise quality-of-life and aes-
29 thetic considerations over income. Having said that, they are often
30 undertaken by those who can afford to take such choices, such as
31 people of wealth or independent means, including retired persons.
32 lifestyle migrations These are ‘lifestyle migrations’ in which a move to a pleasant rural
33 landscape or a sunnier climate enables certain individuals to enjoy
34 a more relaxing and healthier life in a culture which is somewhat
35 different from and more appealing than their own. There are, how-
36 ever, many variants on this theme, including those who wish to ‘es-
37 cape to the sun’ by settling in a Spanish Mediterranean coastal resort
38 (O’Reilly, 2000), those who are ‘international counterurbanisers’
39 such as the British home-owners in rural France studied by Buller
40 and Hoggart (1994), and Kockel’s (1991) ‘countercultural migrants’
41 – Dutch and Germans who have settled along the remote western
42 seaboard of Ireland in order to pursue ‘alternative’ rural lifestyles.
43 Heliotropic migrants – Northern Europeans who spend varying

amounts of time during the year living and relaxing in the warm 1
south – illustrate very well one of the dimensions along which the 2
divide between migration and more frequent forms of mobility is 3
particularly difficult to draw. The spectrum of movements ranges 4
from tourism through seasonal residence to permanent relocation to 5
a holiday area, such as international retirement migration (Williams 6
and Hall, 2000). In some recently completed work I carried out 7
with Tony Warnes and Allan Williams (King et al., 2000), we found 8
that British retiree migrants to southern Spain and the Portuguese 9
Algarve generally had extensive prior experience of visiting the re- 10
gion on holiday before making the semi-permanent retirement 11
move.9 Repeated holidays in sunny seaside resorts had frequently 12
led to a progressively more committed engagement with a destina- 13
tion which was seen as both enjoyable and desirable, and as increas- 14
ingly familiar. Often the purchase of a flat or holiday villa as a second 15
home became a stepping-stone to a more-or-less permanent transfer 16
of residence upon retirement. 17
These forms of movement and dual place connections are not dis- 18
similar to movement patterns associated with other kinds of trans- 19
national community, although the motivations behind the establish- 20
ment of such transnational communities may be very different. In 21
contrast to diasporic communities spawned by refugee scatterings 22
or transnational communities built out of labour migrations, the 23
British on the Costa del Sol (or the Germans in Majorca, or whatever) 24
are engaging in migration and resettlement as a ‘lifestyle activity’. 25
They have become heliotropes, permanent sun-seekers, and all the 26
evidence suggests their numbers are set to grow (King et al., 2000). 27
28
29
Conclusion 30
31
This paper has attempted to map out both some new migratory forms new migratory 32
and processes in Europe, and the attendant conceptual and method- forms 33
ological challenges of how to approach their study. These new forms 34
of migration derive from new international divisions of labour, the 35
new European geopolitics after the Cold War, new motivations of 36
migrants (above all the retreat from labour migrations linked to 37
Fordist production systems), new space-time flexibilities and tech- 38
nologies, and the relatively new notion of migration as consumption 39
and self-discovery. Thus, and in a variety of ways, migration process- 40
es in Europe (and globally) have certainly become more diverse in 41
the past 20 years or so. Whilst the structural underpinnings of the 42
new migrations have been implicit throughout much of the forego- 43

130 russell king
1 ing account, there remain some reservations about how new these
2 migrations are. Koser and Lutz (1998: 4-5), for example, cautioned
3 against a posteriori descriptions of newness and pointed out that his-
4 torical analysis often exposes the arbitrariness of the application of
5 the term ‘new’ to a social phenomenon such as changing migration.
6 Nevertheless, they seem to have been broadly happy with the appella-
7 tion ‘new’, and theirs is not the only book on European migration to
8 include this word in its title (King, 1993b; Thränhardt, 1996).
9 This diversification and (albeit contested) newness of migratory
10 forms encourages both the reassertion of some basic tenets of migra-
11 tion study, and opens up the potential, indeed the necessity, of new
12 methodological approaches.
13 Firstly, I reiterate my earlier plea for an integrated interdisciplin-
14 ary approach which also recognises paradigmatic plurality and the
15 value of mixed methodologies – combining, for instance, economic
16 analysis, class analysis, studies of ethnicity and culture, and attempts
17 to capture the richness of the human experience of migration.
18 Secondly, the need for comparative analysis remains paramount if
19 studies of migration are to rise above the ideographic. Comparisons
20 can be between migratory groups (in the same country), or across
21 countries (comparing similar or contrasting migratory groups), or
22 across time.
23 Thirdly, we need to recognise what I would call the double em-
24 beddedness of migration; at the individual scale, migration must be
25 embedded in a migrant’s life-course (and in some cases of the life-
26 course of the family, even across generations); and at the macro scale,
27 the study of migration must be embedded in the societies and social
28 processes of both the countries/places of origin and of destination.
29 Fourthly, it has to be acknowledged that many of the new forms
30 of migration/mobility surveyed or mentioned in this paper are inad-
31 equately captured by statistics, if at all. There is a tendency for migra-
32 tion not to be documented if it is not seen as problematic. Hence less
33 and less reliance can be placed on data sources such as Eurostat or
34 the OECD’s SOPEMI database for measuring human spatial mobil-
35 ity in Europe. More reliance will need to be put on primary research
36 surveys carried out on the new migratory forms.
37 As well as new data-frames, new terms and metaphors are re-
38 quired to describe the new mobility types which challenge the bi-
39 nary polarisation of origin and destination and the semi-permanence
40 of the common notion of migration. Regarding new metaphors of
41 migration, I am much attracted by the notion put forward by Ribas
42 Mediterranean Mateos (2001) of the ‘Mediterranean caravanserai’ – a common
43 caravanserai space for migrant groups and flows where they can arrive, stay a

while, and then move on to other destinations, perhaps returning 1

for a later staging stay prior to other moves. To borrow a current 2
EU mobility term, migrants become stagiaires, interposing migra- 3
tions and journeys with periods spent sojourning and working in 4
a variety of destinations. In his book Sociology beyond Societies, John 5
Urry (2000) goes much further: for him, mobility is the metaphor 6
of contemporary global society. He goes the whole globalisation hog 7
by concentrating his entire ‘post-society’ analysis on migration, mo- 8
bility and interfaces, setting aside social structures and processes. 9
Life is a trip engaged on by contemporary, capitalist nomads moving 10
through fluid, deterritorialised spaces; the place-specific metaphors 11
are spaces of movement, pausing and meeting – the hotel lobby, the 12
motel, the airport transit lounge (Urry, 2000: 26-32). Urry would 13
certainly agree with Berger’s (1984) statement that migration is the 14
quintessential experience of our time, even more so at the dawn of 15
the new millennium. 16
But we should be careful not to be carried away by such hyperbole. 17
The shrinking of a borderless Europe is the privilege of a relatively 18
small section of European society – perhaps above all those lucky aca- 19
demics who are amongst the greatest beneficiaries of this travelling, 20
networking, conferencing, migration culture as they move about 21
their spatially extensive but socially restricted ‘small world’ (Lodge, 22
1983). Of course a globalised Europe is far from a borderless utopia, no borderless utopia 23
as any Albanian or Moroccan migrant will affirm (Urry, 2000: 13, 24
22). 25
Throughout this paper we have seen how the traditional binaries 26
of migration study have been bridged and broken up by new flexible 27
and evolving mobility patterns. How to handle, for instance, cross- 28
border shuttle migration (is it really migration?); or how to categorise 29
migrations driven by poverty as voluntary or involuntary; or how to 30
unravel the space-time configurations of long-stay tourism, foreign 31
second-home ownership, residence abroad and expatriacy? We have 32
also seen how legal versus illegal is a particularly blurred dichotomy 33
of migrant reality. Illegality seems to be constructed in an illogical 34
(but perhaps also cynical) way by host societies which seem willing 35
to exploit cheap migrant labour (and even be structurally dependent 36
upon it) yet at the same time to deny the legal and civic existence of 37
migrants. In this way, migration into Europe has become more and 38
more of a global business (cf. Salt and Stein, 1997) which has its own 39
set of private market mechanisms – competition, prices, agents, bro- 40
kers, buyers and sellers of migrants and migration services. 41
Moreover, there are other, less often recognised migration di- 42
chotomies than those discussed and deconstructed in this paper. 43

132 russell king
1
2 Migrations can be spectacular or mundane, or, as noted a little ear-
3 lier, regarded as problematic or non-problematic. By and large, the
4 mundane, unproblematic forms of movement are left unrecorded
5 and often unstudied. The spectacular, problematic ones get all the
6 attention, although here it must be stressed that the nature of the
7 ‘spectacle’ is often exaggerated and distorted by its media portrayal
8 and politicisation. Even the notions of ‘home’ and ‘away’ or ‘abroad’
9 have become blurred. Members of transnational communities
10 may feel ’at home’ in two or more places (or not feel at home any-
11 where). Furthermore, one can be ‘homeless at home’, as evidenced
12 by Jansen’s (1998) narratives of post-Yugoslav identities; or one can
13 be ‘transnational at home’, without ever having migrated, as Golbert
14 (2001) demonstrates in her study of Ukrainian Jews.
15 These new, more diverse and flexible varieties of mobility/migra-
16 tion pose obvious challenges for migration policy, especially within
17 the mind-set of ‘Fortress Europe’, and for attitudes towards regula-
18 tion, governance and citizenship (Pugliese, 1995; Geddes, 2000).
19 The issue is further complicated by the fact that, in contrast to earlier
20 generations of migrants (for example the European ‘guestworkers’
21 of the 1960s who were functionally and sociologically rather homo-
22 geneous and whose migration was highly regulated), many national
23 mixed migration migration flows into Europe nowadays are mixed flows made up of
24 flows refugees, economic migrants, people with high skills and those with
25 no skills. Moreover, many migrants change categories in order to
26 maximise the success of their migration project, or they may move
27 between destinations for the same reason. All these facets of the
28 contemporary map of European migration sit uneasily with regula-
29 tory regimes of migration management and control. National bodies
30 regulate contiguous space, whereas migrations function in network
31 space. States want to ‘sedentarise’ and ‘integrate’ migrants (or certain
32 accepted categories of them), but mobile people with multiple place
33 affiliations and hybrid or cosmopolitan identities have no wish to fit
34 in to the ideology of one national identity. Meanwhile, all around
35 Europe there seems to be a constantly shifting discourse as to the
36 desirability of migration, now very much related to economic, labour
37 force and demographic projections for the next few decades (see, for
38 instance, Visco, 2000).
39 Finally, in stressing the importance of the new migratory circum-
40 stances of a post-industrial, post-modern Europe, I draw attention
41 again to movements motivated above all by non-economic, or only
42 partly economic, considerations – those linked to life-cycle such as
43 student and retirement migrations, both of which have potential

1
for future expansion. Within the same vein, the migration of chil- 2
dren has scarcely been studied, at least from the child’s perspective 3
(Dobson and Stillwell, 2000). Quite rightly, women have become an 4
important new focus for migration research in Europe, recognising 5
their central role in the migration process and as cultural agents in 6
the structuring of ethnic communities and their relation with host 7
societies. On this, as on so many other topics in the unfolding map 8
of new migrations, much still needs to be done. These are exciting 9
times to be a migration researcher in Europe! 10
11
12
Acknowledgements 13
14
This paper is a revised version of a keynote address to the conference 15
on ‘Strangers and Citizens: Challenges for European Governance, 16
Identity, Citizenship’, University of Dundee, 17-19 March 2001. 17
Earlier versions were presented and discussed at the conference 18
on ‘Old Differences and New Similarities: American and European 19
Immigration in Comparative Perspective’ (Italian Academy, 20
Columbia University, New York, 12-13 November 1999) and at ‘New 21
Patterns, New Theories: A Conference on International Migration’ 22
(Nottingham Trent University, 11-13 September 2000). I thank con- 23
tributors to the discussions following the presentation of the pa- 24
per at these three fora, and also the many postgraduate students in 25
Migration Studies at the University of Sussex for their stimulating 26
conversations – Clara Guillo, Nick Mai, Enric Ruiz-Gelices and Chris 27
Whitwell will all recognise their own individual inputs somewhere in 28
the text. 29
30
31
Notes 32
33
1 Curiously, each of these evolved in ways somewhat different to those origi- 34
nally expected and defined by the terminology: for instance ‘settler migra- 35
tions’ involved a lot of unanticipated (and unrecorded) return, and ‘guest-
workers’ generally ended up by staying and transforming themselves into
36
more or less settled ethnic communities (King, 2000). 37
2 I nominate these examples of particular nationalities because recent work 38
on these migrant groups in Europe has demonstrated that they often have 39
high levels of education and professional expertise which, by and large, they 40
are compelled to leave behind when they take up what are (for them) much
41
more remunerative jobs as cleaners, building labourers, streethawkers or
farm workers in destination countries such as France, Italy or Greece: see, 42
43

134 russell king
1 for example, Chell (1997); Knights (1996); Lazaridis and Wickens (1999);
2 Riccio (2001). Whilst experiences and reactions differ between and amongst
3 the various migrant nationalities, some are able to draw strength from their
own cultural values and self-knowledge of their own multilingualism and
4
cosmopolitan experiences – see, for instance, Riccio (2001) and Zinn (1994)
5 on the Senegalese in Italy.
6 3 See, for instance, a number of recent books which attempt an interdisci-
7 plinary analysis of the general field of international migration: Brettell and
8 Hollifield (2000), Faist (2000), Hammar et al. (1997), and Papastergiadis
(2000).
9
4 Although it is also true, as Cohen (1995: 5) points out, that some of the early
10 pioneering studies of migration as a generic process sought to minimise or
11 overlook this distinction (cf. Lee, 1966; Petersen, 1958; Ravenstein, 1885,
12 1889; Zelinsky, 1971).
13 5 Except, perhaps, after death. The burial place of migrants has particular sym-
14 bolic meaning, the implications of which have scarcely been considered by
researchers.
15
6 The term ‘migrant shopping’ comes originally, I believe, from a workshop
16 paper prepared by Robin Cohen (1997). Enlarging Cohen’s notion, the
17 ‘shopping market’ for migrants functions in two directions. Firstly, indi-
18 vidual countries shop for migrants within a global market in order to satisfy
19 certain needs characterised by domestic labour supply shortfall. The UK,
for instance, has recently recruited nurses from Spain and the Philippines.
20
According to Cohen, the two countries which have perfected the system of
21 ‘immigration shopping’ are Australia and Canada. They have structurally
22 linked their economic development, manpower and immigration depart-
23 ments and are intent on finding selected migrants to fill slots in their labour
24 market, including business entrepreneurs who bring investment and cre-
25 ate new wealth and jobs, and skilled labour migrants for the labour-short
IT sector. The second expression of the migrant shopping market is where
26
individual migrants shop around for possibilities and opportunities in dif-
27 ferent countries, often moving on when better economic or social openings
28 become available in another country. Andall (1999) presents a well-worked
29 case of this type of migrant shopping in her study of Cape Verdean women
30 in Europe, whilst Guiraudon (2000) tackles the issue of ‘venue shopping’ on
the part of asylum-seekers, also in the European context.
31
7 To be more precise, the target proposed by the then European Commission
32 in 1987 was that, by 1992, a tenth of EU graduates would have spent at
33 least three months of their higher education in another country. By 1992 the
34 achieved figure was 4% rather than 10% (Adia et al., 1994: 2, 39). Although
35 the 10% objective was reaffirmed in 1997, this was accompanied by a state-
36 ment that its achievement would be unlikely, due above all to financial pres-
sures on students (Jallade et al., 1997). Meanwhile, the total European popu-
37
lation of students has grown considerably.
38 8 At a recent Erasmus conference in Spain, the Italian philosopher and se-
39 miologist Umberto Eco said that the main benefits of the EU’s Erasmus
40 programme were as much sexual as cultural. According to Eco, student
41 exchanges and bi- and multi-lingualism encouraged mixed marriages and
42 relationships across Europe’s national frontiers. See report in Times Higher
Education Supplement, 6 July 2001.
43

9 This experience of holidaying in the region prior to the migration upon re- 1
tirement was less important in the other two southern European destina- 2
tions we surveyed, Malta and Tuscany. Here, career links, family ties and 3
military service were common additional factors (King et al., 2000: 94-5).
4
5
6
References 7
8
Adia E, Stowell M, Higgins T. 1994. Higher Education Sans Frontières: 9
Policy, Practice and the European Student Market. Heist and UCAS: 10
Leeds. 11
Andall J. 1998. Catholic and state constructions of domestic workers: 12
the case of Cape Verdean women in Rome in the 1970s. In The 13
New Migration in Europe: Social Constructions and Social Realities, 14
Koser K, Lutz H (eds). Macmillan: Basingstoke; 124-142. 15
Andall J. 1999. Cape Verdean women on the move: ‘immigration 16
shopping’ in Italy and Europe. Modern Italy 4: 241-258. 17
Andall J. 2000. Organising domestic workers in Italy: the challenge 18
of gender, class and ethnicity. In Gender and Migration in Southern 19
Europe, Anthias F, Lazaridis G. (eds). Berg: Oxford; 145-171. 20
Anthias F. 2000. Metaphors of home: gendering new migrations 21
to southern Europe. In Gender and Migration in Southern Europe, 22
Anthias F, Lazaridis G. (eds). Berg: Oxford; 15-47. 23
Anthias F, Lazaridis G. (eds) 2000. Gender and Migration in Southern 24
Europe. Berg: Oxford. 25
Anwar M. 1979. The Myth of Return. Heinemann: London. 26
Berger J. 1984. And our Faces, my Heart, Brief as Photos. Writers and 27
Readers: London. 28
Bianchi RV. 2000. Migrant tourist workers: exploring the ‘contact 29
zones’ of post-industrial tourism. Current Issues in Tourism 3: 107- 30
137. 31
Blotevogel HH, King R. 1996. European economic restructuring: de- 32
mographic responses and feedbacks. European Urban and Regional 33
Studies 3: 133159. 34
Brettell CB, Hollifield JF (eds). 2000. Migration Theory: Talking 35
Across Disciplines. Routledge: London. 36
Buller H, Hoggart K. 1994. International Counterurbanization: British 37
Migrants in Rural France. Avebury: Aldershot. 38
Campani G. 1995. Women migrants: from marginal subjects to so- 39
cial actors. In The Cambridge Survey of World Migration, Cohen R. 40
(ed.). Cambridge University Press: Cambridge; 546-550. 41
Castles S. 2000. Ethnicity and Globalization: from Migrant Worker to 42
Transnational Citizen. Sage: London. 43

136 russell king
1 Chambers I. 1994. Migrancy, Culture and Identity. Routledge: London.

2 Chell V. 1997. Gender-selective migration: Somalian and Filipina
3 women in Rome. In Southern Europe and the New Immigrations,
4 King R, Black R. (eds). Sussex Academic Press: Brighton; 75-92.
5 Chell V. 2000. Female migrants in Italy: coping in a country of new
6 migration. In Gender and Migration in Southern Europe, Anthias F,
7 Lazaridis G. (eds). Berg: Oxford; 103-123.
8 Cohen R. 1997. Seven Forms of International Migration: a Global
9 Sketch. Background Paper for the Summer School on Key Issues
10 in Migration Research, Ydra, Greece, 4-17 May 1997.
11 Cohen R. (ed.). 1995. The Cambridge Survey of World Migration.
12 Cambridge University Press: Cambridge.
13 Cwerner SB. 2001. The Times of migration. Journal of Ethnic and
14 Migration Studies 27: 3-31.
15 Dobson J, Stillwell J. 2000. Changing home, changing school: to-
16 wards a research agenda on child migration. Area 32: 395-401.
17 Escriva A. 1997. Control, composition and character of new migration
18 to south-west Europe: the case of Peruvian women in Barcelona.
19 New Community 23: 43-57.
20 Faist T. 2000. The Volume and Dynamics of International Migration
21 and Transnational Social Spaces. Clarendon Press: Oxford.
22 Findlay AM. 1993. New technology, high-level labour movements
23 and the concept of the brain drain. In The Changing Course of
24 International Migration. OECD: Paris; 149-159.
25 Geddes A. 2000. Immigration and European Integration: Towards
26 Fortress Europe? Manchester University Press: Manchester.
27 Glick Schiller N, Basch I, Szanton-Blanc C. 1995. From immigrant to
28 transmigrant: theorizing transnational migration. Anthropological
29 Quarterly 68: 48-63.
30 Golbert R. 2001. Transnational orientations from home: construc-
31 tions of Israel and transnational space among Ukrainian Jewish
32 youth. Journal of Ethnic and Migration Studies 27: 713-731.
33 Guiraudon V. 2000. European integration and migration policy: ver-
34 tical policy-making as venue shopping. Journal of Common Market
35 Studies 38: 251-271.
36 Hammer T, Brochmann G, Tamas K, Faist T (eds). 1997. International
37 Migration, Immobility and Development: A Multidisciplinary View.
38 Berg: Oxford.
39 Harris JR, Todaro MP. 1970. Migration, unemployment and devel-
40 opment: a two-sector analysis. Economic Review 60: 126-142.
41 Iglicka K. 2000. Mechanisms of migration from Poland before
42 and during the transition period. Journal of Ethnic and Migration
43 Studies 26: 61-73.

International Organization for Migration 2000. World Migration 1

Report 2000. I0M and the UN: Geneva. 2
Jallade J-P, Gordon J, Lebeau N. 1997. Student Mobility within the 3
European Union: A Statistical Analysis. DG XXII of the European 4
Commission: Brussels. 5
Jamieson A, Silj A. 1998. Migration and Criminality: the Case of 6
Albanians in Italy. Ethnobarometer Working Paper 1: Rome. 7
Jansen C. 1969. Some sociological aspects of migration. In Migration, 8
Jackson JA (ed.). Cambridge University Press: Cambridge; 60-73. 9
Jansen S. 1998. Homeless at home: narratives of post-Yugoslav 10
identity. In Migrants of Identity: Perceptions of Home in a World of 11
Movement. Rapport N, Dawson A (eds). Berg: Oxford; 85-109. 12
King R (ed.). 1993a. Mass Migrations in Europe: the Legacy and the 13
Future. Belhaven Press: London. 14
King R (ed.). 1993b. The New Geography of European Migrations. 15
Belhaven Press: London. 16
King R. 1996. A Celebration of Migration. University of Sussex 17
Research Papers in Geography 25. 18
King R. 2000. Generalizations from the history of return migration. 19
In Return Migration: Journey of Hope or Despair?, Ghosh B. (ed.). 20
United Nations and the International Organization for Migration: 21
Geneva; 7-45. 22
King R, Warnes AM, Williams AM. 2000. Sunset Lives: British 23
Retirement Migration to the Mediterranean. Berg: Oxford. 24
Knights M. 1996. Bangladeshis in Rome: the political, economic 25
and social structure of a recent immigrant group. In Questioni di 26
Popolazione in Europe: una prospettiva geografica, Gentileschi ML, 27
King R (eds). Pàtron: Bologna; 129-142. 28
Kockel U. 1991. Countercultural migrants in the west of Ireland. In 29
Contemporary Irish Migration, King R (ed.). Geographical Society 30
of Ireland Special Publication: Dublin; 6: 70-82. 31
Koser K, Lutz H (eds). 1998. The New Migration in Europe: Social 32
Constructions and Social Realities. Macmillan: London. 33
Laczko F, Thompson D (eds). 2000. Migrant Trafficking and Human 34
Smuggling in Europe. I0M: Geneva. 35
Lazaridis G. 1996. Immigration to Greece: a critical evaluation of 36
Greek policy. New Community. 22: 335-348. 37
Lazaridis G. 2000. Filipino and Albanian women migrant workers 38
in Greece: multiple layers of oppression. In Gender and Migration 39
in Southern Europe, Anthias F, Lazaridis G (eds). Berg: Oxford; 49- 40
79. 41
Lazaridis G, Wickens E. 1999. ‘Us’ and the ‘Others’: ethnic minori- 42
ties in Greece. Annals of Tourism Research 26: 632-655. 43

138 russell king
1 Lee ES. 1966. A theory of migration. Demography 3: 47-57.

2 Lodge D. 1983. Small World. Penguin: Harmondsworth.
3 Mai N. 2001. ‘Italy is beautiful’: the role of Italian television in
4 the Albanian migratory flow to Italy. In Media and Migration:
5 Constructions of Mobility and Difference, King R, Wood N (eds).
6 Routledge: London; 95-109.
7 Massey D, Alarcón R, Durand J, González H. 1987. Return to Aztlán:
8 The Social Process of International Migration from Western Mexico.
9 University of California Press: Berkeley and Los Angeles.
10 McHugh K. 2000. Inside, outside, upside down, backward, forward,
11 round and round: a case for ethnographic studies in migration.
12 Progress in Human Geography 24: 71-89.
13 O’Reilly K. 2000. The British on the Costa del Sol. Routledge: London.
14 Papastergiadis N. 2000. The Turbulence of Migration. Polity:
15 Cambridge.
16 Pastore F. 1998. Conflicts and Migrations: a Case Study on Albania.
17 Centro Studi di Politica Internazionale: Rome.
18 Petersen W. 1958. A general typology of migration. American
19 Sociological Review 23: 256-266.
20 Phizacklea A. 1998. Migration and globalization: a feminist perspec-
21 tive. In The New Migration in Europe: Social Constructions and Social
22 Realities, Koser K, Lutz H (eds). Macmillan: Basingstoke; 21-38.
23 Portes A, Guarnizo LE, Landolt P. 1999. The study of transnational-
24 ism: pitfalls and promise of an emergent research field. Ethnic and
25 Racial Studies 22: 217-237.
26 Pries L. ed. 1999. Migration and Transnational Social Spaces.
27 Aldershot: Ashgate.
28 Pugliese E. 1995. New international migrations and the ‘European
29 fortress’. In Europe at the Margins: New Mosaics of Inequality,
30 Hadjimichalis C, Sadler D (eds). Wiley: Chichester; 51-68.
31 Ratnesar R. 2001. Sea of promise. Time Europe. 12 February.
32 Ravenstein EG. 1885, 1889. The laws of migration (I and II). Journal
33 of the Statistical Society 48: 167-235; 52: 214-301.
34 Rhode B. 1993. Brain drain, brain gain, brain waste: reflections on
35 the emigration of highly educated and scientific personnel from
36 Eastern Europe. In The New Geography of European Migrations,
37 King R (ed.). Belhaven: London; 228-245.
38 Ribas Mateos N (ed.). 2000. Female immigration in Southern Europe.
39 Special issue of Papers: Revista de Sociologia 60.
40 Ribas Mateos N. 2001. Revising migratory contexts: the Mediterranean
41 caravanserai. In The Mediterranean Passage: Migration and New
42 Cultural Encounters in Southern Europe, King R (ed.). Liverpool
43 University Press: Liverpool; 22-40.

Riccio B. 2001. Following the Senegalese migratory path through 1

media representation. In Media and Migration: Constructions of 2
Mobility and Difference, King R, Wood N (eds). Routledge: London; 3
110-126. 4
Salt J. 1984. High-level manpower movements in North-West Europe 5
and the role of careers. International Migration Review 17: 633-651. 6
Salt J. 1992. Migration processes amongst the highly-skilled in 7
Europe. International Migration Review 26: 484-505. 8
Salt J. 2000. Trafficking and human smuggling: a European per- 9
spective. International Migration (Special Issue on Perspectives on 10
Trafficking of Migrants) 38: 31-56. 11
Salt J, Ford R. 1993. Skilled international migration in Europe: the 12
shape of things to come? In Mass Migration in Europe: the Legacy 13
and the Future, King R (ed.). Belhaven: London; 293-309. 14
Salt J, Stein J. 1997. Migration as a business: the case of trafficking. 15
International Migration 35: 467-491. 16
Segal A. 1993. An Atlas of International Migration. Hans Zell: London. 17
Shuttleworth I, Kockel U. 1990. Aspects of Irish Migration in the 1980s. 18
University of Glasgow Applied Population Research Unit APRU 19
Discussion Paper 90/3. 20
Sjaastad LA. 1962. The costs and returns of human migration. 21
Journal of Political Economy 70: 117-130. 22
SOPEMI 1995. Trends in International Migration: Annual Report 1994. 23
OECD: Paris. 24
Thränhardt D. ed. 1996. Europe: A New Immigration Continent. Lit 25
Verlag & Transaction: Münster. 26
Urry J. 2000. Sociology beyond Societies: Mobilities for the Twenty-First 27
Century. Routledge: London. 28
Visco I. 2000. Immigration, Development and the Labour Market. 29
Paper prepared for an International Conference on Migration 30
Scenarios for the 21st Century, Rome. 12-14 July 2000. 31
White PE, Woods RI (eds). 1980. The Geographical Impact of 32
Migration. Longman: London. 33
Williams AM, Hall CM. 2000. Tourism and migration: new relation- 34
ships between production and consumption. Tourism Geographies 35
2: 5-27. 36
Williams AM, King R, Warnes AM. 1997. A place in the sun: inter- 37
national retirement migration from Northern to Southern Europe. 38
European Urban and Regional Studies 4: 115-134. 39
Zelinsky W. 1971. The hypothesis of the mobility transition. 40
Geographical Review 61: 219-249. 41
Zinn DL. 1994. The Senegalese immigrants in Bari: what happens 42
when the Africans peer back? In Migration and Identity, Benmayor 43

140 russell king
1 R, Skotnes A (eds). Oxford University Press: Oxford; 53-68.

2 Zinn DL. 1996. Adriatic brethren or black sheep? Migration to Italy
3 and the Albanian crisis. European Urban and Regional Studies 3:
4 241-249.
5 Zontini E. 2001. Family formation in gendered migrations: Moroccan
6 and Filipino women in Bologna. In The Mediterranean Passage:
7 Migration and New Cultural Encounters in Southern Europe, King R
8 (ed.). Liverpool University Press: Liverpool; 231-257.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
The constitution of a European immigration 3
4
policy domain: a political sociology approach
5
6
Virginie Guiraudon1 *
7
8
9
10
11
12
This article by political scientist Virginie Guiraudon was awarded the prize 13
for 2001’s best European Union Studies Association conference paper. 14
Combining the insights of James March and Johan Olsen’s Garbage Can 15
Model with a sociological approach focused on power competition between 16
actors, it explains the incomplete and complex constitution of the European 17
immigration policy domain. Guiraudon is one of the first scholars to demon- 18
strate so brilliantly that the melding of policy studies and political sociology 19
can be fruitful. She helps make sense of the gradual Europeanisation of im- 20
migration, asylum and anti-discrimination policy. What’s more, Guiraudon’s 21
work seeks to overcome shortcomings of the simple legal approach long 22
dominant in European Union studies. 23
24
25
*EDITOR’S NOTE An earlier version of this article was presented as 26
a paper to the 2001 EUSA conference and was awarded the prize for 27
the best 2001 EUSA conference paper. 28
The prize selection committee (Dorothee Heisenberg, James 29
Hollifield, George Ross) noted that Guiraudon’s paper ‘captures the 30
complexity of contemporary EU policy formation in the immigration 31
area ... [and] is remarkable for its recognition and mastery of differ- 32
ent streams of policy-making over time. It foregrounds real EU poli- 33
tics in an unstable, constantly changing set of institutional arenas 34
without imposing artificial social science parsimony. Reading the 35
paper we enter the EU as it is, not as we would like it to be in our a 36
priori models. Guiraudon’s refreshing theoretical quest instead goes 37
toward the sociology of organizations, borrowing from March and 38
Olson’s “garbage can” approach.’ garbage can model 39
40
41
42
43

142 virginie guiraudon
1 While ‘first generation’ European Community (EC) policies such as

2 the common agricultural policy are under fire, new policy domains
3 are emerging in the European Union (EU) framework. At the 1999
4 Tampere summit, EU leaders declared that the development of
5 a comprehensive immigration and asylum policy was a top prior-
6 ity and the EU’s next large-scale enterprise after the single market
7 and European monetary union (EMU).2 Three years later, the June
8 2002 Seville summit was still largely dedicated to immigration as
9 European leaders invoked populist electoral breakthroughs in vari-
10 ous European elections to step up the ‘fight against illegal migration’.
11 The official narratives behind the development of this common
12 policy have taken two forms. Before 1992, the predominant dis-
13 course within the ‘Schengen laboratory’ and the ‘ad hoc immigration
14 group’ was that free movement within the EC required compensa-
15 tory measures at the external borders lest Europe become a ‘sieve’.
16 This political version of the ‘spillover’ theory of integration was large-
17 ly replaced in the 1990s by a more securitarian perspective (Bigo
18 1996), as the number of asylum-seekers and persons displaced by
19 war rose. ‘Asylum shopping’ and ‘immigration risks’ were now com-
20 mon ‘problems’ that could best be dealt with through co-ordination.
21 The view here resembles liberal intergovernmentalism, which poses
22 that major member states co-operate to upgrade common interests
23 and reduce transaction costs.
24 The reconstruction of the rationale behind the rise of immigra-
25 masked complexity tion on the political agenda masks the complexity and incomplete-
26 and incompleteness ness of current EU-level policies and considers them to be an in-
27 evitable solution to a commonly defined problem. They postulate a
28 rationality long criticized when analysing national policy processes.
29 Among them, March and Olsen (1989) focused on situations of orga-
30 nized anarchy whereby the elements of decision-making are thrown
31 into the process as they appear as in a ‘garbage can’. The elaboration
32 of an EU immigration policy presents similarities with their model.
33 My main claim is that, regarding both immigrant and migration
34 control policy, only one side of the debate ‘venue shopped’ at the
35 international level to pursue their own ends, primarily to escape do-
36 mestic adversaries. In the case of migration control, bureaucrats sit-
37 ting in interior ministries sought to regain the discretion taken away
38 by courts and the leeway lost to inter-ministerial arbitrage. Regarding
39 immigrant policy, the domestic challenge came from electoral poli-
40 tics that forestalled policy change and innovation. The migration pol-
41 icy domain cannot be understood as the bargaining outcome among
42 states with a coherent or aggregated set of preferences on these is-
43 sues. Instead, only one ‘camp’ in the national policy field went trans-

national, and this article provides an account of the ways they did so. 1
It thus examines the dynamics of the constitution of this policy 2
domain to better apprehend its timing, form and content. After set- 3
ting out the analytical framework that focuses on power struggles 4
among groups seeking legitimacy (I), I turn to the main chapters of 5
the story so far: the bureaucratic rivalry that led to Title IV of the EU 6
Treaty and the incorporation of Schengen via protocol at Amsterdam 7
which sets the frame for a common immigration and asylum poli- 8
cy (II); the rivalry of non-governmental organizations (NGOs) that 9
carved out a space for EU policy in the area of migrant incorporation, 10
which resulted in Article 13 on anti-discrimination and a ‘race direc- 11
tive’ in 2000 (III); and, finally, the parallel activities of the European 12
Court of Justice (ECJ) and the Commission Trade directorate in the 13
area of freedom of services that affect migration flows within and 14
into the EU. 15
16
17
I. Theoretical and contextual premises 18
19
The story of the rise of immigration on the EU policy agenda is 20
that of governmental and non-governmental actors arriving on the 21
European scene to escape domestic constraints and open up new 22
spaces for action. In this motley crew, we find law and order offi- 23
cials from Interior, Justice and Foreign Affairs ministries, interna- 24
tional NGOs, activists and Commission fonctionnaires from different 25
directorates. Although each came to believe that there should be a 26
European immigration policy, they exploited different policy venues 27
and frames resulting in a set of policy instruments involving varying 28
degrees of supranationalization and distinct decision-making rules. 29
These groupings are not monoliths. National and EU bureaucrats, 30
NGOs compete among their own kind as much as they fight among 31
themselves in a struggle for legitimacy and autonomy. In this re- 32
spect, the Bourdieusian notion of ‘field’ (Bourdieu 1981) is helpful field 33
since it focuses on the power struggles within each group of actors 34
(see also Favell 2000). 35
This actor-oriented approach generates several research ques- actor-oriented 36
tions: why did certain groups decide to ‘go transnational’? Who were research questions 37
they competing with at the national and transnational level? Why did 38
certain groups gain a monopoly of expertise in the European sphere? 39
What policy venues and frames did they exploit? What opportunities 40
could they seize upon (allies in EU institutions or member states, 41
actors in other policy areas, treaty revisions, changes in the global 42
economic or strategic context)? The empirical study of these mobili- 43

1 zation strategies explains the particular timing, form and content of

2 EU policies that affect both migration flows and the conditions of im-
3 migrant minorities in Europe. In other words, our approach should
4 be able to explain:
5 • when international co-operation started and when competence
6 was shifted to the EU (timing);
7 • why certain rules and procedures for EU decision-making were
8 adopted (form);
9 • why a particular policy toolbox was adopted (content).
10
11 public policy The insights of public policy studies can be fruitfully combined with
12 insights that of political sociology to grasp the development of a European
13 policy domain.3 March and Olsen suggest that, although the choices
14 made by the various selfinterested actors can be said to be rational
15 from their perspective, one should not reconstruct a non-linear pol-
16 icy process as inevitable. Contingencies and reversals closed certain
17 paths and cleared others along the way. The ‘garbage can’ model un-
18 derlines that interests, institutions, ideas, problems and solutions
19 appear in the process in no preordained sequence as ‘exogenous
20 streams flowing through a system’ (Olsen 2001: 191),4 yet, as we will
21 see, the order in which each element appears has a bearing on the
22 eventual outcome.5
23 I build upon the public policy agenda-setting literature, includ-
24 ing John Kingdon’s work (1995) which was directly influenced by
25 the garbage can model and Frank Baumgartner and Bryan Jones’s
26 concept of policy frames and venues (1993). If a solution is defined
27 before a problem is identified, issue framing will be crucial to re-
28 constitute a ‘causal story’ (Stone 1989). Similarly, the success of a
29 particular frame will depend upon windows of opportunities. This
30 implies that, once actors have decided to shift their strategies to a
31 European policy venue, their ability to do so will depend on the avail-
32 ability of relevant frames and their seizing of opportunities.
33 In methodological terms, I have consequently favoured a gene-
34 alogical approach that starts before the rise of immigration on the
35 European agenda. To capture the cross-national and cross-sectoral
36 dynamics of EU policy-making, I chose a comparative approach. I
37 focused on immigration politics in three founding members of
38 Schengen (France, Germany and the Netherlands) since the 1970s to
39 apprehend the relative position and constraints of national migration
40 policy players before and during the start of European co-operation.
41 To understand the choices and the fate of the various groups that
42 mobilized transnationally and that of EU institutional actors, I inter-
43 viewed the national civil servants in charge of immigration issues in

international forums and also conducted research in Brussels among 1

NGOs and EU institutional actors. 2
Before analysing the scope of EU immigration policy, the contours 3
of national policy-making in this area should be drawn. Migration as 4
a policy issue was never confined to a single ministry since it had 5
implications for labour, economics, foreign affairs, social affairs and 6
internal affairs (etc.). In federal systems, the division of labour is 7
even more complex. There is no tradition in Europe of ‘immigra- 8
tion ministries’ as there is for agriculture or defence. Immigration is 9
a transversal issue where cross-sectoral conflicts often arise. Cross- 10
national convergence was significant in the 1990s, yet there remain 11
different models of incorporation, different priorities in migration 12
control based on previous colonial and labour market histories or 13
geopolitical position. Therefore, for immigration scholars, the ques- 14
tion regarding the policy sectors and the national models and priori- 15
ties that prevail in the European sphere is a fascinating one. Not all 16
sectoral and national interests were weighed in the policy process 17
and not all actors were deemed legitimate to set the agenda. 18
The shift of competence to the EU greatly narrowed the scope 19
of migration-related policies. With regard to migration control, the 20
European Economic Community (EEC) was limited by the treaty to 21
the free movement of EC workers, later EU citizens, although the 22
ECJ has extended some aspects of free movement to the families of 23
Community nationals and to citizens of countries that have signed 24
association treaties with the EC such as Turkey. One of the possible 25
routes that an EU policy could still take but has not, in spite of a 1997 26
Commission proposal6 is to extend free movement to resident third- 27
country nationals. 28
Instead, migration management in the EU context is focused 29
on preventing unwanted migration, through visa policy and car- 30
rier sanctions, the establishment of buffer zones on the east of 31
Europe, the constitution of a database of inadmissible aliens (the 32
Schengen Information System) and of asylum-seekers’ fingerprints 33
(EURODAC). European asylum policies aim at preventing migration 34
with accelerated procedures for examining asylum requests, a com- 35
mon definition of a refugee, the notion of ‘safe third country’ and 36
the 1990 Dublin Convention which organized a system to determine 37
which contracting party is responsible for examining an asylum re- 38
quest. 39
Regarding immigrant policy at the EU level, it has taken two 40
forms. First, the Commission funds projects for the integration of 41
workers and anti-racism, or gives grants to cities and regions that 42
target initiatives at ethnic minorities. Second, in the Amsterdam 43

1 Treaty, an article on anti-discrimination has been added and two di-

2 rectives have since been approved: one covers all forms of discrimi-
3 nation in employment, and the other counter-discriminations on the
4 grounds of race and ethnic origin in many spheres.7 To understand
5 why these particular outcomes and not others such as the extension
6 of free movement or EU citizenship to third-country nationals can be
7 observed, I now turn to the history of EU mobilization around migra-
8 tion, asylum and anti-discrimination.
9
10
11 II. Immigration and asylum: bureaucratic rivalry and
12 security frames
13
14 ‘When policemen replace diplomats’: the emergence of intergovern-
15 mental co-operation on migration control
16 ‘Quand les policiers succedent aux diplomates’: the title of this
17 French Senate report (Turk 1998) sums up in a nutshell the increas-
18 law and order ing involvement of law and order personnel at the European level
19 since the early 1980s and, among them, civil servants in charge of
20 migration management.
21 Migration control experts took advantage of new organizational
22 models: the transgovernmental working groups on security-related
23 issues such as the 1970s Trevi group. These groupings with varied
24 membership were flexible, informal and secretive. This built trust
25 among officials who set the agenda of transgovernmental co-oper-
26 ation by emphasizing the kind of technical solutions that required
27 their expertise. They became inevitable interlocutors at the first nego-
28 tiation stage, that of the Schengen Implementation Agreement (SIA).
29 While the 1985 Schengen agreement only contained three articles on
30 immigration, the issue came to dominate the discussion of the four
31 Schengen groups in charge of the SIA. During the 1985-90 period
32 when the SIA was drafted, inter-ministerial quarrels in the founding
33 Schengen countries flourished. Michel Portal at the French Ministry
34 of Interior recalls that ‘the inter-ministerial conflicts were and still
35 are considerable, terrible, especially when the political leaders totally
36 lost interest’.8 Vendelin Hreblay, a negotiator from the French po-
37 lice, admits that Foreign Affairs ministries – and in Germany the
38 Chancellery – were progressively ousted by Justice and Interior min-
39 istries (1998: 28).9
40 Given that an international agreement was being negotiated
41 and that Foreign Affairs ministries deliver visas through consulates
42 abroad (visa policy being a cornerstone of European co-operation
43 on remote border control), there was no a priori reason to expect a

monopoly of Interior and Justice personnel. Notwithstanding, their 1

domination accounts for the security-oriented content of the SIA 2
and subsequent decisions. It also explains the emphasis on techni- 3
cal issues, border control and surveillance technologies such as the 4
Schengen Information System. 5
Migration control bureaucrats went transnational at that partic- 6
ular moment because they had seen their action increasingly con- 7
strained in the early 1980s (Guiraudon 2000a). First, landmark court 8
decisions in the main European receiving countries that date from 9
the late 1970s had circumscribed administrative discretion. They es- 10
tablished in particular the right to normal family life and to secure 11
residence for long-term residents. In effect, governments could no 12
longer prevent family reunification, diminish the ‘stock’ of legal resi- 13
dents except by financial incentives as the new Kohl government did 14
in 1983, and certain categories of foreigners could no longer be ex- 15
pelled. This period also saw the first major clashes between agencies 16
in charge of the integration of settled foreigners and those in charge 17
of migration control. The incentive to seek new policy venues shel- 18
tered from national legal constraints and conflicting policy goals thus 19
dates from the beginning of the 1980s (see Guiraudon 2000c on this 20
case of ‘venue shopping’). This explains the timing of transgovern- 21
mental co-operation and its character: an emphasis on non-binding 22
decisions and secretive arrangements. Rather than creating an ‘in- 23
ternational regime’, i.e. a constraining set of rules with monitoring 24
mechanisms (Ruggie 1982), national civil servants sought to avoid 25
domestic legal constraints and scrutiny. 26
In 1990, only some elements of March and Olsen’s ‘garbage can’ 27
were to be found in the migration policy domain at the European 28
level. ‘Solutions’ had been devised before ‘problems’ had been de- solution invented 29
fined. The solution was police cooperation and reinforced controls. before problem 30
The problem that these means were meant to address soon became defined 31
apparent after the end of the Cold War in the form of an influx of 32
asylum-seekers in Germany and many emotional debates over im- 33
migration in other core member states, largely covered in the media 34
which prophesized ‘tides’ of ‘bogus refugees’. International migra- 35
tion was also added to the list of transnational phenomena consid- 36
ered by a plethora of experts as the ‘new threats’ which replaced Cold 37
War ideology: Islamic fundamentalism, global mafias and terrorism 38
(Huysmans 2000). 39
While the 1980s had seen the emergence of a particular group 40
of policy actors seeking to further their interests in transgovermental 41
forums on migration and asylum, ideas and institutions were still in 42
their infancy. The ideas that framed intergovernmental co-operation 43

1 hinged on linking migration and crime and considering that they

2 constituted the dark side of ‘globalization’ requiring a supranational
3 response. The lack of an alternative policy frame can be attributed
4 not only to the end of the Cold War security paradigm but also to
5 economic slump and high unemployment, which demobilized busi-
6 ness interests, which traditionally lobby for openness. These conjec-
7 tural elements should not be neglected in understanding why migra-
8 tion became a ‘security’ (as opposed to a labour market) issue in the
9 1990s.
10 The institutional framework set up at the EU level with the cre-
11 the Third Pillar on ation of a Third Pillar on Justice and Home Affairs (JHA) confirmed
12 JHA that European cooperation allowed Justice and Interior personnel
13 to regain a certain margin of manoeuvre and can be described as
14 flexible multilateralism. One full group (GDl) of the K4 committee
15 of the Third Pillar was dedicated to asylum, visa and migration yet
16 the framework required unanimous decisions by the Council and
17 remained outside the community legal order, thereby excluding the
18 ECJ and the European Parliament. The Commission did not have a
19 right of initiative and thus could not play its agenda-setting ‘entrepre-
20 neurial’ role.
21 Although a small task force was set up within the General
22 Secretariat of the Commission to liaise with the Council on JHA mi-
23 gration discussions, they did not come from the units that had always
24 defended the rights of thirdcountry nationals (the Employment and
25 Social Affairs and the Internal Market Directorate-Generals (DGs)),
26 which task force personnel considered ‘oldfashioned’ and ‘maximal-
27 ist’. According to Wenceslas de Lobkowicz of the task force, they
28 wanted to leave the field to the discretion of member states and avoid
29 debates over sovereignty (1994). Jean-Louis de Brouwer, now head of
30 the Commission unit ‘External Borders, Immigration and Asylum’,
31 also points out that ‘one need[ed] to talk to the big players, the minis-
32 ters of Interior of the member states who usually are political heavy-
33 weights in their respective governments’. 10
34 From Maastricht to Amsterdam, the JHA Council only agreed on
35 one joint position on the common definition of a refugee and on five
36 legally binding joint actions, for instance, on school travel for third-
37 country national children and airport transit procedures. The lack of
38 formal agreements has been attributed to the complicated decision-
39 making structure of the Third Pillar. Yet, it is the same large mem-
40 ber states (France, Germany) most concerned with immigration
41 that stalled the process by insisting on labyrinthine procedures and
42 unanimous voting, thus undermining Ugur’s intergovernmentalist
43 account of the upgrading of common interests in the face of massive

asylum requests (1995). The only operative agreements, the 1990 1

Dublin and Schengen agreements, were in fact adopted outside the 2
EU framework and their implementation delayed respectively until 3
1997 and 1995. 4
Moreover, a number of parallel forums on migration and asy- parallel forums 5
lum were set up during this period, making the ‘Third Pillar’ one on migration 6
of many other venues: among them and aside from the Schengen 7
executive committee, Intergovernmental Consultations on Asylum, 8
Refugees and Migration Policies, the Vienna Club (Germany, 9
Austria, Switzerland, France and Italy), the Vienna Group and 10
Budapest process, the Central European initiative, the Ad Hoc 11
Committee of Experts for Identity Documents and the Movement of 12
Persons, the Council of Europe Committee of Experts on the Legal 13
Aspects of Territorial Asylum, Refugees, and Stateless People, the 14
UN Commission on Crime Prevention, and the Organization for 15
Security and Co-operation in Europe (OSCE). The institutions of the 16
EC were not considered as the legitimate set of institutions to de- 17
velop common policies, thus contradicting the neo-functionalist ac- 18
count of a spillover of the creation of the EC single market. 11 19
20
The diplomats strike back? Amsterdam and beyond 21
The decisions to shift co-operation on migration into the Community 22
framework and incorporate Schengen via protocol during the 23
last stage of the Amsterdam negotiations came as a surprise. The 24
Commission negotiating team headed by Michel Petite won a battle 25
if not the war in Amsterdam. Interior officials were taken aback since the Amsterdam 26
they themselves were unclear about the content of the Schengen ac- Treaty 27
quis, 3,000 pages of various legal standing. They did not want the ac- 28
quis published and given a legal character. The Treaty actually came 29
into force in May 1999 before member states had agreed on its con- 30
tent and its incorporation. 31
To understand the Amsterdam outcome, one must remember 32
that ministries of Foreign Affairs negotiate treaty revisions in the EU. 33
They were not concerned with the consequences of the Schengen 34
protocol, a task that their colleagues sitting in Interior and Justice 35
ministries would have to undertake.12 Having seen their negotiating 36
role diminished during the Schengen process, Foreign Affairs were 37
keen to rein in transgovernmental processes dominated by law and 38
order civil servants which had multiplied and run amok. By neglect- 39
ing the Third Pillar and preferring the Schengen group, the bureau- 40
crats in the Schengen founding member states had unwittingly con- 41
tributed to that outcome. They could not count on the support of later 42
Schengen members such as Italy or Greece who had not been treated 43

1 as equal partners. Among the three founding Schengen members

2 studied, only the Dutch favoured a ‘communitarization’ of asylum
3 and immigration. Given that the interests of the larger member states
4 (France and Germany) were better preserved in a flexible multilateral
5 setting such as Schengen, the Dutch preference for the inclusion of
6 Community actors and a more constraining framework should not
7 be surprising. Even less so given that the French in particular had
8 bullied the Dutch in the Schengen context over drugs policy.
9 Notwithstanding, the German and French delegations success-
10 fully lobbied for provisions that limited the role of EC institutions in
11 the new Title IV of the Amsterdam Treaty on the progressive estab-
12 lishment of an ‘area of freedom, security and justice’. The Germans
13 obtained unanimous voting in the Council of Ministers and, under
14 French pressure, the role of the ECJ was circumscribed. The applica-
15 tion of preliminary rulings to the ECJ in areas covered by Title IV is
16 restricted since only courts of last instance will be able to use Article
17 177. Furthermore, the Court of Justice cannot rule on national mea-
18 sures adopted in relation to the crossing of borders to safeguard in-
19 ternal security, and its rulings ‘shall not apply to judgments of courts
20 or tribunals of the Member States which have become res judicata’.
21 The defended position reflects the original motivation of intergov-
22 ernmental co-operation, which was to avoid judicial scrutiny that had
23 undermined migration control policy at the domestic level. 13
24 It remains puzzling that the state most concerned with the is-
25 sue (Germany) did not wish to shift competence to the Community
26 or ‘lock in’ commitments, and lobbied for unanimity, given that
27 their priority was refugee ‘burdensharing’.14 We know that, during
28 negotiations, like-minded officials and national governments do
29 not share the same ‘preferences’ (Lord and Winn 2000) and here it
30 seems that the German Interior officials’ reluctance to depart from
31 a Schengen model of secretive inter-bureaucratic co-operation led to
32 a sub-optimal outcome for Germany, the main recipient of asylum-
33 seekers and displaced persons. Yet this only points to the lack of do-
34 mestic co-ordination on the issue in the German case. This is why
35 the Interior-Foreign Affairs Chancellery rivalry that dated from the
36 Schengen negotiations still mattered at Amsterdam. It led to what
37 aggregation failure Andrew Moravcsik has termed an ‘aggregation failure’ whereby the
38 emergence of a coherent national position out of disparate demands
39 is blocked, a situation that, in his view, allows supranational entre-
40 preneurs to play the role of ‘two-level network manager’ (1999: 283):
41 here the rejoicing Michel Petite who could claim victory for the policy
42 shift from the Third to the First Pillar, although with limitations on
43 the role of EC institutions.

Amsterdam has also not solved the question of ‘opt-outs’. 1

Amsterdam consecrates the idea of a Europe à la carte. The UK, 2
Northern Ireland and, consequently, the Republic of Ireland have 3
opted out of the new area of freedom, security and justice. Denmark, 4
albeit a member of Schengen, is not bound by the new title and co- 5
operates only on visa policy – a legal nightmare since it requires the 6
signing of a separate Danish-EU treaty every time a decision is taken. 7
Since Amsterdam, developments suggest that, given the rules of 8
the game, the logic of the policy process has not drastically changed. 9
The Commission 2000 Communication on a community immi- 10
gration policy resubmitted texts that had been discussed under the 11
Third Pillar framework.15 It faces competition from member states 12
that have a right of co-initiative. Typically, the country that holds the 13
Presidency of the Union uses this platform to push its pet projects 14
to satisfy its domestic electoral interests. Law enforcement measures 15
such as those proposed by the French Presidency in the fall of 2000 16
have been more successful than those emphasizing migrants’ rights. 17
The French proposals on carrier sanctions, expulsion of third-coun- 18
try nationals or the fight against smuggling were adopted in May 19
2001 under the Swedish Presidency, while the latter had to accept a 20
much watered down version of its own text on temporary protection 21
status to ensure passage. 22
What has been confirmed is the importance of migration in 23
the foreign relations of the EU. For instance, all concerned parties 24
agree that the JHA acquis has gone up the agenda in the accession 25
negotiations in which Justice and Interior ministers take a large 26
part (Lavenex 2001). Ten per cent of PHARE funding (130.7 million 27
Euros in 2000) goes to JHA issues, half of which concern border is- 28
sues (House of Lords Select Committee on European Union 2000, 29
part 3, p. 4). ‘Preventing migration at the source’ has become an EU 30
mantra.16 It has resulted in an number of initiatives including the 31
Dutchinspired cross-pillar High-Level Working Group (HLWG) set 32
up in December 1998. The latter drew up action plans for the six 33
main countries of origin of unwanted migrants in Europe to assess, 34
inter alia, the possibility of readmission agreements, ‘safe returns’ 35
and ‘transit zones’. The HLWG’s 1999 report stressed the ‘general 36
recognition that a cross-pillar and comprehensive approach [was] 37
needed’ and stated that ‘the expertise of the Member States needs to 38
be made available in various policy fields’.17 The group’s ‘trans-pillar’ 39
approach seemed to herald a new era when the prevailing ‘preven- 40
tion-by-policing’ policies would be accompanied by policies that ad- 41
dressed the root causes of migration. Yet, the country reports simply 42
restated the six ‘action points’ set out by the JHA K4 Committee in 43

1 March 1998 regarding immigration from Iraq in which Turkey was

2 expected to prevent Iraqis from arriving in the EU.
3 Thus, although diplomats at Amsterdam took their revenge
4 on Interior and Justice personnel, the latter still dominate and are
5 becoming more involved in diplomatic forums. For instance, in
6 February 2000, during the negotiations of the revision of the fourth
7 Lome Convention between the EU 15 and seventy one Mexican,
8 Caribbean and Pacific countries, Interior ministers insisted that a
9 clause of readmission of illegal migrants be included in the final text
10 at the risk of blocking the agreement.
11 In sum, transgovernmental co-operation allowed law and order
12 officials to gain autonomy and devise policies without accommodat-
13 ing judges or conflicting sectoral interests. They successfully defined
14 a frame that equated migration with transnational security threats
15 and favoured intergovernmental secretive forums. Over time, they
16 security threats were perhaps too successful and, after Amsterdam, they have to co-
17 operate with EU institutions and publish their decisions. They do
18 remain key players.
19
20
21 III. Pro-migrant forces go transnational too: NGO rivalry
22 and the social exclusion paradigm
23
24 In 1985, when the first Schengen agreement was signed, the
25 Commission issued new guidelines on migration (CEC 1985) and
26 argued that European integration entailed a better access to rights for
27 foreign residents. In July, it adopted a Decision setting up a procedure
28 for prior consultation of new policy in this area. Five member states
29 contested the move and the ECJ annulled the Decision in 1987.18 The
30 Commission’s competence was confined to the free movement of EU
31 citizens. Yet, this did not deter the Commission unit that had been
32 pro-migrant pushing for this change and pro-migrant transnational organizations
33 transnational such as the Churches’ Commission for Migrants from carving out
34 organizations a space for the defence of the rights of ethnic minorities in Europe.
35 The unit (0.4) within the Commission Directorate for Employment
36 and Social Affairs now called ‘Free Movement of Workers, Migrant
37 Integration and Anti-racism’ was created in 1958 to handle issues
38 related to free movement of labour and later handled many budget
39 lines related to the integration of migrants and refugees and, since
40 1986, anti-racism.19 Annette Bosscher, the head of the unit until the
41 late 1990s, and Giuseppe Callovi, who later moved to other director-
42 ates, firmly believed that European integration should go hand in
43 hand with the integration of non-Europeans.

Their unit has faced many challenges, given the thin treaty basis 1
for its actions. Its ‘institutional activists’ (Ruzza 1999) had to find 2
other bases for intervention. As Adrian Favell recalls: 3
4
as a ‘political’ as opposed to ‘economic’ agenda began to differen- 5
tiate itself in the Commission’s corridors, certain DGs less power- 6
fully placed in the central drive towards EMU, seized on alternative 7
European ‘public interest’ agendas, following the path pioneered by 8
the highly active and progressive minded DG XI (Environment). 9
(Favell 2000: 167) 10
11
Indeed the attitude of civil servants in the Employment and Social 12
Affairs DG resembled the ‘purposeful opportunism’ (Cram 1997) 13
found in other directorates whereby larger policy agendas are instru- 14
mentalized to increase their scope for action. 15
A few individuals committed to a progressive agenda in fairly 16
marginal parts of the Commission could become ‘policy entrepre- 17
neurs’ (Geddes 2000a, 2000b) precisely because their activities were 18
sheltered from public scrutiny. Daniel Cohn-Bendit, a Member of the 19
European Parliament (MEP) who once headed the Frankfurt Bureau 20
for Multicultural Affairs, has thus analysed the situation: ‘Europe 21
is full of promises for the future because the Commission and the 22
Parliament are not exposed to immediate electoral pressures.’20 23
Similarly, the successful initiatives in the area of immigrant policy 24
concerned a few Commission insiders and small NGO structures 25
that may have publicly decried the ‘democratic deficit’ yet practised 26
top-down élite politics. 27
The first opportunity before Maastricht was to build upon the no- European 28
tion of ‘European citizenship’ that was meant to herald a ‘people’s citizenship 29
Europe’ and the end of the ‘democratic deficit’. To help mobilization 30
on this agenda, the Commission sought to increase its legitimacy as 31
a spokesperson for ‘civil society’ by engineering an official channel 32
of interest representation. In 1991, the European Commission acting 33
upon an initiative of the European Parliament founded the Migrants’ 34
Forum that spoke for 130 migrant associations that held an annual 35
general assembly. Yet, the Migrants’ Forum failed to find common 36
ground (Kastoryano 1994; Geddes 1998).21 Turks and Moroccans 37
vied for control of the organization, with the Moroccans eventually 38
winning out and giving the organization a Francophone cast that set 39
it apart from the largely Anglophone NGO world of Brussels. The 40
Forum’s activists also had different conceptions of citizenship and 41
cultures of contention depending on the nation states in which they 42
had settled. 43

1 In any case, using the concept of citizenship to further the rights

2 of third–country nationals failed in 1992. The gap between EU and
3 non-EU citizens widened when the Treaty on European Union
4 granted special rights to EU citizens residing in other member states
5 such as local voting rights. Both the Commission and the Brussels-
6 based NGO Migration Policy Group (MPG) refocused their agenda.22
7 war on social They jumped on the bandwagon of the EU war on ‘social exclusion’
8 exclusion (Article 137 of the Treaty of Amsterdam). Commission documents
9 insist that migrants and their descendants are prime victims of social
10 exclusion and that NGOs know best how to fight it.23 Social exclusion
11 encompasses a wide range of programmes and the MPG promptly
12 responded to this signal by linking migrant integration in the 1996
13 Intergovernmental Conference (IGC) to this agenda rather than to
14 the debates on European citizenship, which had focused the ener-
15 gies of the Migrants’ Forum (Geddes 2000b). Indeed, the ‘social
16 exclusion’ frame benefited pre-existing transnational networks that
17 could draw upon their credentials and expertise in the area of anti-
18 discrimination.24 In fact, the Starting Line Group (SLG) founded in
19 1992 by academic and NGO legal experts and co-ordinated by the
20 MPG to draft an anti-discrimination article for the pre-Amsterdam
21 IGC included members from national anti-discrimination boards:
22 the British Commission for Racial Equality and the Dutch National
23 Bureau against Racism.
24 Citizenship or social exclusion, EU citizenship for third-country
25 nationals or anti-discrimination policies? In the NGO battle for legiti-
26 macy, the SLG supported by the MPG clearly had the organizational
27 structure, and the local and legal knowledge to successfully lobby for
28 its anti-discrimination agenda while the Migrants’ Forum with its
29 cumbersome structure remained focused on citizenship. Moreover,
30 the SLG matched EU technocratic standards. The anti-discrimina-
31 tion clause project was reminiscent of Article 119 and the 1976 Equal
32 Treatment Directive on gender equality in a very Euro-correct way.
33 Leading up to the 1996 IGC, initiatives that showed a gentler, kinder
34 Europe were welcome. The timing was ripe for the SLG initiative.
35 With Commission officials, they were able to informally set the agen-
36 da at the 1996 IGC, thereby confirming accounts of Amsterdam ne-
37 gotiations that build upon Kingdon’s model, such as Mark Pollack’s
38 (1999), and those that focus on the importance of ‘policy framing’,
39 such as Mazey and Richardson’s (1997).
40 Policy framing was key because, if the problem is defined as ‘so-
41 cial exclusion’, the range of solutions is wide. As Andrew Geddes has
42 pointed out:
43

it can be advantageous that the terms inclusion and cohesion are 1

vague and their meanings unclear because it implies that the quest 2
for inclusion is likely to be able to sustain itself in the long term and 3
potentially be institutionalized at the European level. 4
(Geddes 2000a: 224) 5
6
Like ‘sustainable development’, ‘social inclusion’ is an objective that 7
one can hardly oppose. ‘Anti-discrimination’ for its part presented 8
the advantage of not being solely targeted at migrants. Article 13 9
protects people with disability, the elderly and other groups – a plus 10
given that measures specifically protecting migrants are a hard sell. 11
Most member states that did not want to shift competence on im- 12
migrant policy to the EU level nevertheless did because they were 13
led to believe that in fact the issue was social exclusion of a number 14
of groups. 15
The ambiguity of the ‘anti-discrimination’ frame also partly ex- ambiguity of the 16
plains the rapid adoption of the so-called ‘race directive’ in June anti-discrimination 17
2000 (directive 2000/ 43/EC) seven months after the Commission’s frame 18
proposal – ‘a record for the adoption of a piece of Community law 19
requiring substantial legislative changes at national level’ (Tyson 20
2001: 112). The directive also required a unanimous decision in the 21
Council and had an inter-sectoral character that implied interminis- 22
terial co-ordination making it a ‘least likely case’. The single factor 23
most often mentioned by the Council Social Affairs working group 24
interviewed in Brussels is Jorg Halder.25 France was most vocal at 25
condemning the Austrian government for integrating the Freedom 26
Party in February 2000. Ironically, their enthusiasm towards a di- 27
rective that resembled Dutch or British tools for integrating ethnic 28
minorities stemmed from an event, the success of a far-right leader 29
who had praised the Waffen SS, that easily fitted the French concep- 30
tion of anti-racist measures as a means of fighting ideas inspired by 31
Nazi Germany. In a classic ‘Baptist-bootlegger coalition’ situation, 32
the German delegation was also extremely co-operative lest it be as- 33
sociated with the Austrians. The initial policy linkage between the 34
anti-discrimination package and the Austrian far right ensured the 35
passage of a directive. 36
Just as law and order officials, NGOs expanded the realm of com- 37
petence of the EU to include immigrant-related issues. Their agenda 38
had more chances of succeeding through lobbying techniques shel- 39
tered from public scrutiny than at the national level where public 40
opinion, media coverage and the mobilization of anti-immigrant par- 41
ties made the advancement of migrant rights unlikely. Even in the 42
Council, negotiations focused on reaching compromises on techni- 43

1 cal issues and legal wording rather than on the normative underpin-
2 nings of immigrant policy and can be contrasted to the emotional
3 partisan debates observed in many European countries. This closed
4 venue of debate allowed policy change in favour of migrants that is
5 arduous in open national venues.
6
7
8 IV Indirect policy effects: migration and freedom of services
9
10 Beside conscious efforts to mobilize around migration at the
11 European level, one must take into account decisions by EU institu-
12 tions and transnational non-state actors that indirectly bear on mi-
13 gration flows to complete the complex and contradictory set of EU
14 rules that affect migration within and into the EU.
15 The ECJ has traditionally been concerned with extending its ju-
16 risdiction. The Court has had to strike a balance between expanding
17 EC competence and remaining within the legitimate bounds of its
18 sphere of duty (economic rights rather than people’s rights and EC
19 citizens rather than non-EC citizens). Therefore, its jurisprudence
20 on third-country nationals has not been based on human rights but
21 Rush Portuguesa on freedom of services or association treaty provisions. In the Rush
22 decision Portuguesa decision of 27 March 1990 (C-113/89, ECR 1-1417), the
23 ECJ reiterated that the provisions for the suppression of restrictions
24 to the freedom to deliver services entailed that a company could move
25 with its own staff. If the company employs third-country nationals,
26 member states cannot refuse them entry to protect their labour mar-
27 ket on the grounds that immigration from non-EU states is a matter
28 of national sovereignty.
29 The Court decision was in line with the drive towards the single
30 market, which resulted in the 1993 liberalization of service provision.
31 It stirred a controversy in Germany given the important number of
32 posted workers in the construction industry denounced by trade
33 unions as a form of ‘social dumping’. Indeed, no comprehensive su-
34 pranational regulation has been passed on the social and wage con-
35 ditions applicable to posted workers. Instead, a 1996 directive has
36 allowed member states to apply a minimum level of national regula-
37 tions to these posted workers and the Commission has proposed two
38 directives to implement this derived right of third-country nationals
39 (OJ 1999 C 67/9).
40 Meanwhile, at the Trade Commission directorate, developments
41 suggest that the mobility of personnel in the services sector will be
42 extended at the global level and thus affect flows into the EU. Co-
43 optation strategies are at work between the Trade Commission staff

and business interests. One particular non-governmental forum sup- 1

ported by the Trade Commission is the European Services Forum European Services 2
(ESF), an official NGO in the Seattle EU delegation whose focus is to Forum 3
support the Commission’s viewpoint during the General Agreement 4
on Trade in Services (GATS) negotiations. At a conference of the 5
ESF under the patronage of the Commission, Trade Commissioner 6
Pascal Lamy expressed this sentiment: ‘I particularly welcome the 7
participation of ... NGOs. The key to the success of the ESF is that it 8
is a forum, open to all stakeholders, including civil society.’26 9
Pascal Lamy has experience in setting up ‘partnerships’ that 10
short-circuit member states since this was a key strategy of Jacques 11
Delors when Lamy was his chef de cabinet (Ross 1995). Lamy’s refer- 12
ence to civil society is misleading. In fact, the ESE based at UNICE, 13
the European employers’ federation, includes thirty-six European 14
trade federations and fifty EU-based international companies in sec- 15
tors such as banking, insurance, telecommunications, postal ser- 16
vices, aviation, shipping, tourism, retail, legal services, accountancy, 17
management consulting, architecture, engineering, IT services, pub- 18
lishing, audiovisual, energy and environmental services. 19
Part of their agenda is lobbying against ‘barriers to the movement 20
of people’ and in particular the ‘complex, cumbersome, and time- 21
consuming procedures to obtain work permits and visas’ (ESF 2000) 22
and they favour a GATS visa or passport.27 The adversaries are clearly 23
identified: the ESF managing director describes them as ‘the under- 24
standably defensive interests of WTO Member Countries’ immigra- 25
tion and labor market developments officials’ (Kerneis 2000). 26
At an MPG meeting on this issue organized in Brussels in March 27
2001,28 immigration officials’ jaws dropped in silent disbelief when 28
they heard multinational corporations proposing ‘their’ passport. 29
The meeting also showed that strange bedfellows emerge at the 30
European level. European pro-migrant NGOs are not used, as are 31
their American counterparts, to engaging in client politics with busi- 32
ness interests (on the US case, see Freeman 2001). Yet there seems 33
to be a fast learning curve, which is fostered by the MPG’s ‘trans- 34
atlantic dialogue’ with US think-tanks. Strategic alliances between 35
NGOs and business interests are signs that some of the actors in our 36
story are trying to co-ordinate their scripts to seize upon the opportu- 37
nity of the new economic climate and the older free movement and 38
neo-liberal agenda of European integration. 39
40
41
42
43

1 Conclusion
2
3 The coexistence of conflicting discourses that do not speak to one
4 conflicting another, competition among like-minded actors, diverse modes of
5 discourses decision-making (depending on their level of supranationalization),
6 in a period of numerous and rapid ED constitutional changes ex-
7 plains the autocratic and contradictory character of law-making in
8 EU immigration-related policies.
9 Both in the case of migration and asylum and that of anti-dis-
10 crimination policies at the EU level, we observe parallel dynamics.
11 First, a group of actors vie to become the legitimate policy interlocu-
12 tors against other similar groups: interior civil servants vs. their for-
13 eign affairs counterpart, MPG and the SLG vs. the Migrants’ Forum.
14 Each group has a pre-formatted set of policy solutions based on their
15 expertise: policing for the former, anti-discrimination for the latter.
16 They succeeded by momentously seizing upon an emergent broader
17 policy frame: immigration officials built upon the post-1989 new se-
18 curity agenda while NGOs joined calls for the fight against ‘social
19 exclusion’ during the 1996 IGC. They were helped by their adversar-
20 ies’ weaknesses or errors, respectively the lack of supervision of other
21 key ministries whose attention was fixed on the fall of the Berlin Wall
22 when Schengen was being negotiated, and the structural and chronic
23 problems of the Migrants’ Forum. It is telling that initially there was
24 little supervision of these experiments that grew on the margins of
25 the core market-driven project of European integration.
26 Our goal has been to account for the particular timing, form and
27 content of the immigration policy domain. Our focus on the ac-
28 tors who prevailed and the interests that they represented explains
29 the content and form of the European immigration policy domain.
30 Immigration officials sought to avoid national judicial constraints
31 and conflicting bureaucratic views that were experienced in the early
32 1980s. They consequently favoured a secretive intergovernmental-
33 ism where they could exclude other ministries and escape judicial
34 monitoring. Similarly, they have privileged informal co-operation
35 and ‘soft law’. Their own professional identity explains the bias to-
36 wards control and policing. Pro-migrant groups knew that, as in a
37 national context, the institutions most receptive to defending mi-
38 grant interests are restricted venues of debate sheltered from elec-
39 toral fallout such as social administrations and courts. They found
40 European ‘functional equivalents’ in the Commission and the ECJ
41 and have focused on legal solutions such as the inclusion of Article
42 13 in the Treaty. At the European level, small lobby-like structures are
43 the most efficacious which explains the success of the Dutch-British

activists and therefore the emphasis on anti-discrimination. Success 1

only came once they could co-ordinate with Commission officials, 2
who had first sought to build upon free movement to expand their 3
competence and later had tried to foster a more representative as- 4
sembly. 5
The timing of the constitution of the immigration policy do- 6
main itself depends on the windows of opportunity constituted by 7
the emergence of new frames, changes in the strategic or economic 8
context, or constitutional openings such as IGCs or Schengen nego- 9
tiation working groups. In this respect, this is not a straight ‘path- 10
dependent’ account. Today’s winners may yet face challeng- 11
es if the context changes or at the next constitutional moment. 12
Notwithstanding, they have accumulated a legitimacy capital and the 13
policy domain has been institutionalized in a way that cannot be easi- 14
ly undone. For instance, we have seen that the diplomats’ ‘revenge’ at 15
Amsterdam has not altered the predominance of Interior and Justice 16
interests in the management of EU migration policy. 17
It cannot be denied that following 9/11 and the concert of European 18
leaders’ calls for a European border police prior to the Seville sum- 19
mit, a security/restrictive take may prevail. For politicians, this is a 20
convenient way of shifting blame and responsibility. Yet, few con- 21
crete decisions were taken at Seville and harmonization is slow, and 22
EU measures have not resulted in a decrease in illegal immigration. 23
In the end, this strategy may be as dangerous as ‘activating xenopho- 24
bia’ at the national level. Populist parties, which are generally both 25
anti-immigrant and anti-EU, will be further strengthened by the fail- 26
ure of European leaders to support more proactive immigration and 27
integration policies. 28
29
30
Notes 31
32
1 The author thanks Martin Schain and participants of the 2001 EUSA meet- 33
ing who commented on an earlier version of this paper, Andrew Moravcsik 34
for his incisive reading, as well as the two anonymous referees for their in-
sightful suggestions.
35
2 Before 2004, the Council should unanimously adopt measures on asylum, 36
refugees and displaced persons, on the absence of any controls on persons 37
crossing internal borders and on external border control (including rules on 38
visas for stays of less than three months), and on the free travel of third- 39
country nationals within the EU for short-term stays. After 2004, measures
40
should be adopted with respect to refugee ‘burden-sharing’, and the harmo-
nization of the conditions of entry and residence, standards for the issue of 41
long-term visas and residence permits, or the right of residence for third- 42
country nationals wishing to stay in EU states other than their country of 43
residence.

1 3 For a fuller treatment of the application of political sociology to EU studies,

2 see Guiraudon (2000b).
4 ‘The central idea of the garbage can models is the substitution of a temporal
3
order for a consequential order’ (March and Olsen 1986: 17) and thus our
4 research paid particular attention to temporal ordering.
5 5 Given the recent debate in the American Political Science Review on the ‘gar-
6 bage can’ (see Bendor et al. 2001 and the reply by Olsen 2001), I clarify that
7 my reference to Cohen et al.’s famed 1972 article respects the spirit of their
8 work: the metaphor was not meant as ‘the’ theory but rather as ‘a’ model
to ‘comprehend some features of decision-making ... to extend, rather than
9
replace, understandings gained from other perspectives’ (March and Olsen
10 1986: 12).
11 6 Proposal for a Council Act establishing the convention on rules for the ad-
12 mission of third-country nationals to the member states. COM/97/0387 fi-
13 nal - CNS 97/0227 [Doc 597PC0387].
7 Respectively, Council Directive 2000/43/EC of 29 June 2000 [Official
14
Journal L 180, 19/07/2000, pp. 22-6] and Council Directive 2000/78/EC
15 0(27 November 2000 [O. J. L 303, 02/1212000, pp. 16-22].
16 8 Interview with Michel Portal, chef de bureau, Sous-Direction de la Circulation
17 Transfrontiere et des Visas, Ministry of Interior, Paris, December 1994.
18 Also interviews with M. Malwald, German Federal Ministry of Interior,
19 Bonn, April 1995, with Jürgen Haberlandt, German Federal Ministry of
Interior, Berlin, June 1995, and with Nicolas Franzen, Immigration and
20
Naturalization Department, Ministry of Justice, The Hague, February 1995.
21 The lack of political leadership was heightened by glasnost and Germany’s
22 unwillingness to build a wall to its east.
23 9 Transport ministries had also signed the original 1985 agreement and later
24 disappeared.
10 Interview, General Secretariat of the European Commission, Brussels,
25
March 1999.
26 11 See Guiraudon (2000c) for a fuller analysis of alternative explanations.
27 12 Interview with Michel Petite, chief negotiator for the 1996 IGC, European
28 Commission, Cambridge, MA, April 1999.
29 13 Stetter (2000) refers to these decisions as ‘principals’ seeking to prevent
30 ‘agency loss’ when delegating authority. I would add that, once delegation
had occurred against their views, migration bureaucrats did indeed seek to
31
limit agency loss but that the rules and procedures adopted to do so seem
32 to have been counterproductive given what he views as the motivation for
33 shifting competence, which, again, was not the ex ante preferred option for
34 French and German officials.
35 14 For a thorough test of alternative theories of EU burden-sharing in this area,
see Thielemann (2002).
36
15 COM(2000) 757 final, 22/1112000.
37 16 COM(2000) 757 final, 22/11/2000, section 2.1 ‘partnership with countries of
38 origin’.
39 17 Press release, ‘Final Report of the High-Level Working Group on Asylum
40 and Migration’, 18/9/1999.
41 18 See 9 July 1987 decision in joint cases 281,283-5,287/85, Rec. 1987, 3023.
19 The Unit administers about 10 million ECUs for refugee integration, 6 mil-
42
lion for migrants, and 7 million for anti-racism every year. A 1995 report
43

assessing 200 of the 560 projects on migrant integration that DG V funded 1

between 1991 and 1993 demonstrates that only 32 (16 per cent) were mi- 2
grant-led (CEC 1995: 10). NGOs, churches, trade unions, etc., made up the
3
rest of the beneficiaries. After 1995, Brussels-based NGOs that had submit-
ted 2.6% of the proposals received 6.8% of the total funding – a clear success 4
(CEC 1998). 5
20 Interview, Brussels, May 1995. 6
21 After several mismanagement crises, the Forum has been suspended. 7
22 The MPG staff acknowledged that supranational competencies that would 8
affect diverse national concepts of citizenship or change nationality law were
9
anathema ro member states (Hix and Niessen 1996).
23 See Guidelines on Preparatory Measures to Combat Social Exclusion (CEC 10
1998). 11
24 French or German national activists were interested in citizenship issues yet 12
were rarely present among the personnel of pro-migrant Brussels NGOs. 13
25 Interviews in Brussels with Claire Aubin, Social Affairs attaché, French per-
14
manent delegation to the EU, 5 December 2001, Porfirio Silva, Social Affairs
attache (in charge of presiding Social Affairs and Employment Council work- 15
ing group), Portuguese permanent delegation to the EU, 6 December 2001, 16
John Kittmer, Social Affairs attaché, British permanent delegation to the EU, 17
6 December 2001. 18
26 Speech given at the conference ‘The GATS 2000 Negotiations: new opportu- 19
nities of trade liberalization for all services sectors’, Hotel Sheraton Brussels
20
Airport, Brussels (Zaventem), 27 November 2000.
27 The idea of a ‘GATS visa’ emerged in 1993 at the end of the Uruguay Round 21
and is understood as a passport for different categories of natural persons 22
permitted entry under the schedule of commitments at the horizontal and 23
sectoral levels like Information and Communication Technologies (lCTs), 24
business visitors, contract personnel.
25
28 Transatlantic Workshop on High Skilled Migration (Brussels, 5-6 March
2001). 26
27
28
29
References 30
31
Baumgartner, Frank and Jones, Bryan (1993) Agendas and Imtability 32
in American Politics, Chicago: Chicago University Press. 33
Bendor, Jonathan, Moe, Terry and Shotts, Kenneth (2001) ‘Recycling 34
the garbage can: an assessment of the research program’, American 35
Political Science Review 95(1): 169-90. 36
Bigo, Didier (1996) Polices en réseaux. L’expérience européenne, Paris: 37
Presses de la Fondation Nationale de Sciences Politiques. 38
Bourdieu, Pierre (1981) ‘La répresentation politique: éléments pour 39
une théorie du champ politique’, Actes de la recherche en sciences 40
sociales 36/37: 3-24. 41
CEC (Commission of the European Communities) (1985) Orientation 42
pour une politique communautaire des migration, COM(85) 48 def. 43
Brussels: CEC.

1 CEC (Commission of the European Communities) (1995) Assistance

2 Given to Migrant Association, Brussels: DG V.
3 CEC (Commission of the European Communities) (1998) European
4 Year Against Racism Directory of Projects, Brussels: DG V.
5 Cohen, Michael, March, James and Olsen, Johan (1972) ‘A gar-
6 bage can model of organizational choice’, Administrative Science
7 Quarterly 17 (March): 1-25.
8 Cram, Laura (1997) Policy-Making in the European Union: Conceptual
9 Lemes and the Integration Process, London and New York: Routledge.
10 European Services Forum (ESF) (2000) Second Position Paper on the
11 Temporary Movement or Key Business Personnel, Brussels: ESF, 24
12 October. Available at www.esf.be
13 Favell, Adrian (2000) ‘L’Européanisation ou l’émergence d’un nou-
14 veau “champ politique”: le cas de la politique d’immigration’, in
15 Virginie Guiraudon (ed.), Sociologie de l’Europe: élites, mobilisation
16 et configuration institutionnelles, special issue of Cultures et conflits
17 38(9): 153-85. Available at www.conflits.org
18 Freeman, Gary (2001) ‘Client politics or populism? The politics of
19 immigration reform in the United States’, in Virginie Guiraudon
20 and Christian Joppke (eds), Controlling a New Migration World,
21 London: Routledge, pp. 65-95.
22 Geddes, Andrew (1998) ‘The representation of “migrants’ interests”
23 in the European Union’, Journal of Ethnic and Migration Studies
24 24(4): 695-713.
25 Geddes, Andrew (2000a) ‘Thin Europeanisation: the social rights
26 of migrants in an integrating Europe’, in Michael Bommes and
27 Andrew Geddes (eds), Immigration and Welfare. Challenging the
28 Borders of the Welfare State, London: Routledge, pp. 209-26.
29 Geddes, Andrew (2000b) ‘Lobbying for migrant inclusion in the
30 European Union: new opportunities for transnational advocacy?’,
31 Journal of European Public Policy 7(4): 632-49.
32 Guiraudon, Virginie (2000a) Les politiques d’immigration en Europe.
33 Allemagne, France, Pays-Bas, Paris: L’Harmattan.
34 Guiraudon, Virginie (2000b) L’espace sociopolitique européen, un
35 champ encore en friche?’, in Virginie Guiraudon (ed.), Sociologie
36 de l’Europe: élites, mobilisations et configurations institutionnelles,
37 special issue of Cultures et conflits 38(9): 7-37. Available at www.
38 conflits.org
39 Guiraudon, Virginie (2000c) ‘European integration and migra-
40 tion policy: vertical policymaking as venue shopping’, Journal of
41 Common Market Studies 38(2): 249-69.
42 Hix, Simon and Niessen, Jan (1996) Reconsidering European Migration
43 Policies, Brussels: Churches’ Commission for Migrants in Europe.

House of Lords Select Committee on European Union (2000) 1

Seventeenth Report on Enlargement and EU External Frontier 2
Controls, London: House of Lords, published 24 October. 3
Hreblay, Vendelin (1998) Les accords de Schengen. Origine, fonc- 4
tionnement, avenir, Brussels: Bruylant. 5
Huysmans, Jef (2000) ‘The European Union and the securitization 6
of migration’, Journal of Common Market Studies 38(5): 751-77. 7
Kastoryano, Riva (1994) ‘Mobilisations des migrants en Europe: du 8
national au transnational’, Revue européenne des migrations interna- 9
tionales 10(1): 169-81. 10
Kerneis, Pascal (2000) ‘Letter to Pascal Lamy dated 8 November 11
2000’, Brussels: ESE Available at www.esf.be 12
Kingdon, John (1995) Agendas, Alternatives and Public Policies, 2nd 13
edn, New York: HarperCollins. 14
Lavenex, Sandra (2001) ‘Migration and the EU’s new eastern border: 15
between realism and liberalism’, Journal of European Public Policy 16
8(1): 24-42. 17
Lobkowicz, Wenceslas de (1994) ‘Intergovernmental coopera- 18
tion in the field of migration – from the Single European Act to 19
Maastricht’ in Joerg Monar and Roger Morgan (eds) , The Third 20
Pillar of the European Union: Cooperation in the Fields of Justice and 21
Home Affairs, Brussels: European University Press, pp. 99-122. 22
Lord, Christopher and Winn, Neil (2000) ‘Garbage cans or rational 23
decision? Member governments, supranational actors and the 24
shaping of the agenda for the IGC’, Current Politics and Economics 25
of Europe 9(3): 237-56. 26
March, James and Olsen, Johan (1986) ‘Garbage can models of 27
decision making in organizations’, in James March and Roger 28
Weissinger-Baylon (eds), Ambiguity and Command. Organizational 29
Perspectives on Military Decision Making, Marshfield, MA: Pitman, 30
pp. 11-35. 31
March, James and Olsen, Johan (1989) Rediscovering Institutions. The 32
Organizational Basis of Politics, New York: Free Press. 33
Mazey, Sonia and Richardson, Jeremy (1997) ‘Policy framing: in- 34
terest groups and the lead up to the 1996 Intergovernmental 35
Conference’, West European Politics 20(3): 111-33. 36
Moravcsik, Andrew (1999) ‘A new statecraft? Supranational entrepre- 37
neurs and international cooperation’, International Organization 38
53(2): 267-306. 39
Olsen, Johan (2001) ‘Garbage cans, new institutionalism, and the 40
study of politics’, American Political Science Review 95(1): 191-8. 41
Pollack, Mark (1999) ‘Delegation, agency and agenda setting in the 42
Treaty of Amsterdam’, European Integration online Papers (EIoP) 43
3/6 at http://eiop.or.at/eiop/ texte/1999-006a.htm

1 Ross, George (1995) Jacques Delors and European Integration, Oxford:

2 Oxford University Press.
3 Ruggie, John (1982) ‘International regimes, transactions, and
4 change: embedded liberalism in the postwar economic order’,
5 International Organization 36(2): 379-415.
6 Ruzza, Carlo (1999) ‘Normal protest: social movements and institu-
7 tional activism’. Unpublished ms.
8 Stetter, Stefan (2000) ‘Regulating migration: authority delegation in
9 justice and home affairs’, Journal of European Public Policy 7(1):
10 80-103.
11 Stone, Deborah (1989) ‘Causal stories and the formation of policy
12 agendas’, Political Science Quarterly 104: 281-300.
13 Thielemann, Eiko (2002) ‘Between interests and norms: explaining
14 burden-sharing in the European Union’. Paper for the UACES
15 Workshop on European Burden Sharing and Forced Migration,
16 London: LSE, 12 January.
17 Turk, Alex (rapporteur) (1998) Quand les policiers succèdent aux diplo-
18 mates. Rapport d’information 523 (97-8) de la Commission des lois du
19 Sénat, Paris: Sénat.
20 Tyson, Adam (2001) ‘The negotiation of the European Community
21 Directive on Racial Discrimination’, European Journal of Migration
22 Law 3: 111-229.
23 Ugur, Ehmet (1995) ‘Freedom of movement versus exclusion: a re-
24 interpretation of the “insider-outsider” divide in the European
25 Union’, International Migration Review 29(4): 964-99.
26 Final version accepted for publication 9/10/02
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Immigration and ‘state thought’ 3
4
Abdelmalek Sayad 5
6
7
8
9
10
11
12
In France and in French-speaking academia, the sociologist Abdelmalek 13
Sayad is unanimously considered one of the very best thinkers on migra- 14
tion. His oeuvre, though quantitatively not huge, is qualitatively outstanding. 15
Unfortunately it is not well known beyond the francophone world. This article 16
was first published in French in 1996. It develops one of the major themes 17
in Sayad’s work, the notion of ‘state thought’. As explained here, it can be 18
summarised in the following way: to think about migration is always to think 19
about the state, and more precisely it is to think about the state that thinks 20
about migration. 21
22
23
Although it is a universal phenomenon, migration is always dis- 24
cussed within the framework of the local unit and, insofar as we are 25
concerned, within the framework of the nation-state.1 Despite the ex- 26
treme diversity of situations in which it occurs and despite the varia- 27
tions it displays in time and space, the phenomenon of emigration- 28
immigration does exhibit constants, in other words characteristics constants 29
(social, economic, juridical and political) that reappear throughout 30
its history. These constants constitute, as it were, a sort of common 31
and irreducible basis, which is both a product and an objectification 32
of ‘state thought’. State thought is a form of thought that reflects, state thought 33
through its own structures (mental structures), the structures of the 34
state, which thus acquires a body (see Bourdieu 1993). The categories 35
through which we think about immigration (and, more generally, 36
our whole social and political world), or our social, economic, cultural 37
and ethical categories – and we can never place too much emphasis 38
on the role morality plays in the way we perceive the phenomenon of 39
immigration – and, in a word, our political categories, are definitely 40
and objectively (that is, without our being aware of it and, therefore, 41
independently of our will), national or even nationalist categories. 42
The structures of our most ordinary political understanding, or of the 43

166 abdelmalek sayad
1 understanding that is spontaneously translated into our world-view,

2 shape our perception of immigration, but they are at the same time
3 shaped by it. They are basically national structures and they therefore
4 act as such. They are structured structures in the sense that they are
5 socially and historically determined products, but they are also struc-
6 turing structures in the sense that they predetermine and organize
7 our whole representation of the world, and therefore the world itself.
8 It is, without any doubt, because of all this that the migratory phe-
9 nomenon as a whole – emigration and immigration – can only be
10 described and interpreted through the categories of state thought.
11 That mode of thought is completely inscribed within the line of de-
12 marcation that radically divides ‘nationals’ from ‘non-nationals’. The
13 line itself is invisible or scarcely perceptible but it has major implica-
14 tions. On the one hand, we have those who have quite naturally, or,
15 as the lawyers put it, have ‘by right’, the nationality of the country
16 (their country) from which they come – in other words of the state
17 whose nationals they are and of the territory over which that state
18 has sovereignty – and, on the other, we have those who do have the
19 nationality of the country in which they are resident.
20
21
22 The spirit of the state
23
24 It is also for all these reasons that we can say that thinking about im-
25 migration means thinking about the state, and that it is ‘the state that
26 is thinking about itself when it thinks about immigration’. And this
27 is perhaps one of the last things we discover when we reflect upon
28 the problem of immigration and work on immigration, whereas we
29 should of course have begun with this, or at least should have known
30 this before we started. What we discover in this way is the secret
31 virtue of immigration: it provides an introduction, and perhaps the
32 best introduction of all, to the sociology of the state. Why? Because
33 immigration immigration constitutes the limit of what constitutes the national
34 state. Immigration is the limit that reveals what it is intrinsically, or
35 its basic truth. It is as though it were in the very nature of the state to
36 discriminate and, in order to do so, to acquire in advance all the nec-
37 essary criteria of pertinence that are required to make the distinction,
38 without which there can be no national state, between the ‘nationals’
39 it recognizes as such and in which it therefore recognizes itself, just
40 as they recognize themselves in it (this double mutual recognition-
41 effect is indispensable to the existence and function of the state), and
42 ‘others’ with whom it deals only in ‘material’ or instrumental terms.
43 It deals with them only because they are present within the field of

its national sovereignty and in the national territory covered by that 1

sovereignty. It has been said that this diacritical function of the state, 2
which, strictly speaking, is one of definition, i.e. delineation,2 is in 3
the very nature of the state, and that it constitutes the state in all 4
its forms and throughout its history. The need to discriminate is, it 5
would seem, more imperative and by that very fact more prescriptive 6
in the case of a republican nation-state. Such a state aspires to total 7
national homogeneity – in other words homogeneity at every level: 8
political, social, economic, cultural (and especially linguistic and re- 9
ligious). 10
Quite aside from the fact that it disturbs the national order, blurs 11
the divide or the border line between what is and what is not na- 12
tional, and therefore perturbs or disturbs the order based upon that 13
separation, immigration, or in other words the presence within the 14
nation of ‘non-nationals’ (rather than those who are simply foreign 15
to the nation), infringes upon the integrity of that order. It disturbs 16
the mythical purity or perfection of that order, and it therefore pre- 17
vents the full realization of that order’s implicit logic. We can thus 18
understand why, without taking to extremes the logic implicit in this 19
state of affairs – that is, without perverting it – there is always a great 20
temptation to lapse into a form of fundamentalism that is known all fundamentalism/ 21
over the world, and that is cultivated and celebrated all over the world purism 22
(today’s religious fundamentalism is no more than a variant, and not 23
even a new variant, as it exists prior to national fundamentalism, hav- 24
ing existed before the reality of the nation itself, and because it has 25
always coexisted alongside that fundamentalism). For those who take 26
a ‘purist’ (or fundamentalist) view of the national order, immigration 27
is supposedly the agent of the perversion of the national social order 28
in its integrity and integrality because it concerns people who should 29
not be there (if the national order were perfect, it would not have this 30
flaw, this inadequacy) but who are there (rather as though they were 31
the objectification or materialization of that flaw, that inadequacy 32
and that inability to complete the nation). Immigration is undeniably 33
a subversive factor to the extent that it reveals in broad daylight the 34
hidden truth and the deepest foundations of the social and political 35
order we describe as national. Thinking about immigration basically 36
means interrogating the state, interrogating its foundation and inter- 37
rogating the internal mechanisms of its structuration and workings. 38
Using immigration to interrogate the state in this way means, in the 39
final analysis, ‘denaturalizing’, so to speak, what we take to be natu- 40
ral, and ‘rehistoricizing’ the state or that element within the state that 41
seems to have been afflicted by historical amnesia. It means, in other 42
words, recalling the social and historical conditions of its genesis. 43

1 Time helps us forget all these things, but time is not the only factor
2 involved: time can succeed in this repressive operation only because
3 it is both in our interests and in the interests of the state itself to
4 forget its history. The ‘naturalization’ of the state, or of the state that
5 exists inside our heads, makes it seem as though the state were an
6 immediate given, as though it were an object that existed by itself or
7 that was created by nature. It makes it seem that the state has been
8 in existence from all eternity, that it has been freed of all determina-
9 tions external to itself. It appears to exist independently of all histori-
10 cal considerations, independently of history and of its own history,
11 from which we prefer to divorce it for ever, even though we never
12 stop elaborating and telling that history. Immigration – and this is of
13 course why it is so disturbing – forces us to unveil the state, to unveil
14 the way we think about the state and the way it thinks about itself.
15 And it is the way it thinks about immigration that gives this away.
16 Being children of the nation-state and of the national categories we
17 bear within us and which the state has implanted in us, we all think
18 about immigration (in other words about those who are ‘other’ than
19 ourselves, what they are, and through them, what we ourselves are)
20 in the way that the state requires us to think and, ultimately, in the
21 way that the state itself thinks.
22 ‘State thought’ or ‘spirit of the state’ as analysed by Pierre Bourdieu
23 is a mode of thought and a distinct way of thinking. The two appear
24 to be inseparable. It is state thought that creates the state’s mode of
25 thinking about everything it is and about all the domains to which
26 it is applied. In the same way, state thought may, as a result of its
27 constancy, its repetitions, its own strength, and its ability to impose
28 its way of thinking on others, have generated durable modes of think-
29 critical reflection ing that are typical of state thought. We must therefore subject the
30 of state thought postulates of state thought to critical reflection, to a process of ‘dele-
31 gitimizing’ what is legitimate, of what goes without saying. We must
32 delegitimize it in the sense of objectifying what is most deeply rooted
33 within us, what is most deeply hidden in our social unconscious.
34 Such an operation makes a desanctifying break with doxa. We have
35 here an undertaking that everything within us resists: our entire so-
36 cial being (individual and collective) and everything that we commit
37 to it with such passion – in other words our whole national being.
38 For we exist only in this form and only within this framework: the
39 framework and form of the nation. To take jurists as an example,
40 it took all the audacity of a Hans Kelsen to free himself from state
41 thought and even to rebel against that thought, and ultimately to con-
42 test the opposition that is de rigueur amongst jurists and (elsewhere)
43 between ‘national’ and ‘non-national’ by demonstrating the arbitrary

(or conventional) character of that distinction: the ‘national’ exists 1

de jure, and belongs by nature or by virtue of state (the possession of 2
the state of nationality) to the population that constitutes the state. 3
Anyone who is ‘foreign’ (non-national) is subject to the competence 4
and authority of a state in which he plays no part, and on whose terri- 5
tory he resides, lives and works only as a result of his presence there 6
and for the duration of that presence. His presence does not have the 7
same status as the presence within that territory of a national. Kelsen 8
regards this difference as purely conventional or non-essential, and 9
that leads him to reject the idea that the state is necessarily the juridi- 10
cal expression of a community. 11
12
13
The crimes of immigration; immigration on trial 14
15
Why this preamble about state thought? First, because immigration 16
constitutes the privileged terrain on to which this form of thought 17
is projected, as though on to a mirror. Second, because of all the 18
domains of existence and of all the sectors of social life, delinquency 19
is the one that owes, so to speak, most to this way of thinking. In the 20
case of immigration, delinquency implies not only the offences that 21
the police have to deal with or those recorded by the crime statistics 22
but also, as one delinquency can hide another, a delinquency that 23
might be described as situational or statutory (and almost ‘ontologi- 24
cal’) because, at the deepest level of our mode of thought (i.e. state 25
thought), it is synonymous with the very existence of the immigrant 26
and with the very fact of immigration. 27
Unconsciously, or even when we are not fully conscious of it, the 28
fact of being an immigrant is far from being a neutral element with- 29
in the whole gamut of evaluations and judgements that are passed, 30
should an offence be committed, on the delinquent. Even though 31
those who pass these judgements (both the ones handed down by 32
the juridical apparatus and those of the social apparatus – i.e. social 33
judgements) are unaware of the fact, and even though they almost 34
always do so against their will, the fact of being an immigrant de- 35
linquent (or a delinquent immigrant) constitutes, as a general rule, 36
something of an aggravating circumstance. Because we spontane- 37
ously endorse expressions of public opinion, which exists inside our 38
heads just as it exists inside the heads of everyone around us (this is 39
doxa), we even see such circumstances as a supplementary offence 40
in addition to the offence that has been committed and that has to be 41
judged. Immigration is a latent, camouflaged offence (that of being 42
an immigrant – an offence for which the subject in question bears 43

1 no responsibility), which is brought to light by the actual offence that

2 has been committed, by the objectified offence that has to be brought
3 before the courts. Any trial involving a delinquent immigrant puts
4 the very process of immigration on trial, first as a form of delinquen-
5 cy in itself and second as a source of delinquency. Before we can even
6 double punishment speak of racism or xenophobia, the notion of ‘double punishment’
7 is therefore present within any judgement passed on the immigrant
8 (and not only in the judgements handed down by judges sitting in
9 court). It is rooted in state thought, and is the anthropological basis
10 on which all our social judgements rest. ‘Double punishment’ exists
11 objectively in our way of thinking, even before we make it exist in
12 the objectified form of either the sanction of a legal tribunal or an
13 administrative decision.
14 ‘Double punishment’ exists inside our ‘national’ heads, because
15 the very fact of immigration is tainted with the idea of being at fault,
16 with the idea of anomaly and anomie. The immigrant presence is al-
17 ways marked by its incompleteness: it is an at-fault presence that is in
18 itself guilty. It is a displaced presence in every sense of the term. It is
19 physically and geographically displaced: in other words, it is spatially
20 displaced because migration is primarily a spatial displacement. It
21 is displaced in the moral sense too, in the sense in which we speak,
22 for instance, of ‘speaking out of turn’ or of ‘misplaced’ discourse.
23 It is as though our categories of thought, which are in this respect
24 and as can never be said too often national categories, saw immigra-
25 tion itself as a form of delinquency, as an intrinsic delinquency. It
26 is as though, because the immigrant is already in the wrong simply
27 because he is present in a land of immigration, all his other sins are
28 reduplicated and aggravated by the original sin of immigration. That
29 is his first sin in the chronological sense because it necessarily pre-
30 cedes all the other sins that might be committed during the lifetime
31 of an immigrant. It is a generative sin in the sense that it is the cause
32 not of his actual sins themselves, but of the place, time and context
33 (in other words of the social, economic and economic conditions) in
34 which those sins are committed. Because it is an objective sin, immi-
35 gration can never be totally bracketed out or neutralized, even when
36 we try to do so in all objectivity. Immigration, with all the disparage-
37 ment, disqualification and stigmatization it implies, affects all the
38 most ordinary acts committed by immigrants and, a fortiori, their
39 criminal acts. Conversely, all immigrant behaviour, and especially
40 deviant behaviour, has repercussions on the phenomenon of immi-
41 gration itself, and leads to greater disapproval, greater disqualifica-
42 historically tion and greater stigmatization.
43 situated sins We therefore have two kinds of sin or guilt: a historically situated

sin (that of immigration) and what might be called behavioural sins behavioural sins 1
or crimes, or actual sins that figure in the taxonomy or the usual 2
table of sins that are reprehensible, sanctionable and sanctioned as 3
such (with varying degrees of severity) by the provisions of the Penal 4
Code which, in law (in theory, which means in accordance with a law 5
that has lost all sense of reality), apply to all offenders, whoever they 6
may be. What relationship is there between the two orders of crime? 7
On the one hand, we have a crime that has not been committed in- 8
tentionally. To that extent, none of those involved, or who become 9
involved despite themselves – immigration and the country of im- 10
migration – can admit to it. Even when it is officially authorized, the 11
‘presence’ of the immigrant is still, as we have said, basically at fault 12
(it is a presence that cannot be an end in itself and which, no matter 13
whether it is accepted or denounced, requires constant justification). 14
Those who are most concerned, namely the emigrant-immigrants 15
themselves, appear, finally, to be the real victims of the gigantic farce 16
that is being acted out at their expense. On the other hand, we have 17
the crime that has been committed, reported and recorded in canoni- 18
cal fashion. It is viewed and seen in itself for what it is in its material- 19
ity and, whenever possible, in the same light as all the crimes of the 20
first kind. 21
What is the relationship between the two? In law, there is none. 22
Historically situated sins or crimes cannot be used as an argument 23
for either the defence or the prosecution of second-order crimes, 24
even when those crimes make the criminal liable to the ever-present 25
sanction of deportation, irrespective of whether or not it is actual- 26
ly implemented. Second-order crimes cannot serve as a pretext for 27
making a more serious and unjust case against immigration. But, in 28
practice, there is a relationship that is always present in everyone’s 29
mind. Some strongly deny being influenced in one way or the other 30
by that relationship. Some claim to be totally neutral and to know 31
nothing about the guilty party’s previous record or, in this case, his 32
status and quality as an immigrant. Others, in contrast, do not con- 33
ceal or hide their satisfaction at seeing two different modalities of 34
crime and the two punishments that sanction them overlapping and 35
aggravating one another – in their view, this is only fair and, basi- 36
cally, something that is quite normal and that should be the rule. 37
The case against immigration is always inseparable from the case 38
made against the immigrant because of some offence, even a minor 39
one, that he has committed. The case against immigration in fact 40
involves the whole system of representation through which we con- 41
stitute immigration, and the deviancy or delinquency of immigra- 42
tion, through which we define the immigrant and the acts, criminal 43

1 mental or otherwise, he is permitted to commit. These representations are

2 representations of two kinds. First, we have mental representations that are translated
3 into acts of perception and evaluation, cognition and recognition.
4 They are translated into a whole series of acts in which agents in-
5 vest their material and symbolic interests (and the symbolic are per-
6 haps invested with more force and passion than the material), their
7 social prejudices, their presuppositions and, in a word, their whole
8 object social being. Second, we have what we might call object representa-
9 representations tions. These consist in all the external signs, all the indices, all the
10 features and all the characteristics that can become the object of the
11 manipulative symbolic strategies we use to determine the (mental)
12 representations that others have of those properties – which are all
13 perceptible from the outside – and their bearers. (In the practical
14 mode, an individual exists mainly in the sense that he is seen and
15 that he allows some part of himself to be seen; and the identity we
16 talk about so much is basically this being-perceived that we all share
17 in a social sense, and which basically exists only because it is rec-
18 ognized by others.) That is the way it is in social life, which is an
19 incessant struggle between the perceptions and classifications these
20 representations impose. Everyone would like to impose the defini-
21 tion or (mental) representation that flatters him most and is in his
22 best social interests by using the properties at his disposal and his
23 self-authorized (object) representation. Courts of all kinds are full
24 of these classification struggles, and the greatest condemnation con-
25 sists, of course, in the a priori denegation and dispossession of all
26 the social attributes – even the most elementary, which are also the
27 most essential – that make it possible to take part, even at the lowest
28 and most dominated level, in the play of these struggles between rep-
29 resentations, in the sense of both mental images and manifestations
30 designed to act upon those mental images.
31 The situation of criminality in immigration – a situation which
32 implies, rather than its objective probability, a guaranteed rise in rac-
33 ism, as it always exists in the presence of and under the gaze of the
34 other – raises the issue of the relationship between politics and polite-
35 ness. When an immigrant is involved, breaking the law also means
36 breaking the unwritten law imposing the reserve and neutrality (real
37 and feigned) that befits a foreigner. In such cases, breaking the law
38 means more than the infraction in question: it is an error of a differ-
39 ent order, a lack of politeness. This demand for simple politeness, for
40 good manners and nothing more, in reality implies the renunciation
41 of many things. The apparently minor or purely normal concessions
42 known as ‘politeness’ are valuable only because they are, in reality,
43 or deep inside us, political concessions: enforcing respect for forms

comes down to demanding every form of the respect that is owed to 1

order. The political neutrality that the political demands of foreign 2
residents who are confined to the non-political is certainly more ac- 3
ceptable and more easily obtained if we locate it in the register of 4
politeness rather than in the sphere of the political, even though that 5
is its true territory. At an unconscious level, it is politeness that pre- 6
vents the foreigner from playing a political part in the political affairs 7
(internal and external) of the host country. 8
9
10
Allaying suspicion 11
12
A sort of social hyper-correction is required of the immigrant, espe- 13
cially one of lowly social condition. Being socially or even morally 14
suspect, he must above all reassure everyone as to his morality. There 15
has never before been so much talk of ‘republican values’ in France. 16
That is because it is a way of denouncing what the social and political 17
morality of French society regards as the deviant behaviour of Muslim 18
immigrants: wearing veils to school, statutory discrimination against 19
women, the political use of religion, which is referred to as funda- 20
mentalism, and so on. Being conscious of the suspicion that weighs conscious of 21
upon him and which he cannot escape because he is confronted with suspicion 22
it throughout his immigrant life and in every domain of his exis- 23
tence, it is up to the immigrant to allay it constantly, to foresee it and 24
to ward it off by repeatedly demonstrating his good faith and his good 25
will. He finds himself caught up in social struggles despite himself, 26
because they are of necessity struggles over identity. Because he is 27
involved in them as an isolated individual and almost without wish- 28
ing to be involved – especially in the interindividual interactions of 29
everyday life – he has no choice but to exaggerate in one way or an- 30
other. Making a virtue of necessity, and to a large extent because of 31
the dominated position he occupies in the structure of symbolic pow- 32
er relations, the immigrant tends, no doubt rightly, to exaggerate 33
each of the contradictory options he thinks he has chosen, whereas 34
they have actually been forced upon him. He is condemned to exag- 35
gerate everything; everything he does, everything he experiences and 36
everything he is. At times, he must, as an immigrant (when he is at 37
the bottom of the social hierarchy within the world of immigrants), 38
assume the stigmas which, in the eyes of public opinion, create the 39
immigrant. He must therefore accept (resignedly or under protest, 40
submissively or defiantly, or even provocatively) the dominant defini- 41
tion of his identity. We need only recall, in this connection, the fact 42
that the stigma itself generates a revolt against the stigma, and that 43

1 one of the first forms of that revolt consists in reappropriating or lay-

2 ing claim to the stigma, which is converted into an emblem in accor-
3 dance with the classic paradigm of black is beautiful. This can even
4 lead to the institutionalization of the group, which thus turns the
5 stigma – in other words and roughly speaking the social, economic,
6 political and cultural effects of the stigmatization of which it is the
7 object and in part the product – into its foundation. At other times,
8 in contrast, the immigrant devotes himself to the quest for so-called
9 assimilation assimilation. This presupposes putting a great deal of effort into his
10 self-presentation and representation (the representation others have
11 of him, and the representation he wishes to give of himself). The ef-
12 fort is therefore focused essentially on his body, his physical appear-
13 ance, and those forms of external behaviour that are most loaded
14 with symbolic attributes or meanings. It is intended to remove all the
15 signs that might recall the stigma (physical signs such as complex-
16 ion, skin colour, hair colour, etc; cultural signs such as accent, man-
17 ner of speech, clothes, the wearing of a moustache, a whole lifestyle,
18 etc.). The other strategy involves conspicuous mimicry and the adop-
19 tion of features which, in contrast, seem to be emblematically charac-
20 teristic of those to whom he wishes to assimilate. Whilst they are not
21 mutually exclusive, the two strategies, or at least parts of them, can
22 be simultaneously juxtaposed, though there is a danger that this will
23 exacerbate the contradictions. In all these examples, no matter how
24 contrasted, the issue appears to centre on the use of strategies of
25 simulation and dissimulation, pretence and bluff, and the acquisi-
26 tion and projection of a self-image that pleases [qui plaît] others and
27 in which the immigrant delights [se complaît], the image he would
28 like to be in keeping with his material and symbolic interests, or the
29 image that is least removed from the identity he is laying claim to.
30 On the one hand, his original identity is credited with having a great-
31 er authenticity – the identity of the ‘old man’ which he refuses to kill
32 off. He must preserve, or believe he is preserving, his original iden-
33 tity because he thinks he is doing so in order not to have to experi-
34 ence it in shame, timidity and scorn, and to avoid the risk of exoti-
35 cism, all of which can encourage the racism of which they are a
36 component element. On the other hand is the new identity he wishes
37 to create in order to appropriate, if not all the advantages bound up
38 with the possession of the dominant identity, at least the legitimate
39 identity (i.e. the identity of the dominant) that he will never have and
40 at least the negative advantages he can expect to derive from no lon-
41 ger having to be judged, or having to judge himself, by criteria that
42 he knows will always, and of necessity, work to his disadvantage.
43 There is another point on which the two strategies are basically in

agreement: both contain within them, each in its own way, a forced 1
recognition of legitimate identity. The former recognizes it by refus- 2
ing it, by keeping as great a distance as possible, and by avoiding any 3
superfluous contact or any contact that is not indispensable. The lat- 4
ter, in contrast, recognizes it by taking its inspiration from it, by tak- 5
ing it as a model, by simulating it and by trying to reproduce it as 6
faithfully as possible, but also as slavishly as possible. In both cases 7
– and this is another reason why they converge – what is really at 8
stake in these strategies for social struggles, which are found in any 9
struggle between the dominated and the dominant, or in the face of 10
domination, is not, as is commonly said, the conquest or reconquest 11
of identity. It is the ability to reappropriate for oneself the possibility constructing and 12
of constructing one’s own identity and of evaluating that identity in evaluating one’s 13
complete autonomy. This is the ability that the dominated are obliged identity 14
to surrender to the dominant, so much so that anyone who finds 15
himself in the dominated position within the field of symbolic power 16
relations has only two possible ways of gaining recognition or, more 17
prosaically, continuing to exist. Either he must be negated, and must 18
therefore consent to his own negation and disqualification, or he 19
must accept the risks involved in any attempt to assimilate. If he 20
adopts the first strategy, he must do what he is being asked to do even 21
though he cannot resign or withdraw completely in the strict sense of 22
the term from a game he knows to be basically stacked against him. 23
He must, that is, simply withdraw from the struggle, as he is being 24
asked to do – in other words, abandon it without necessarily leaving 25
the arena (i.e. immigration) in which such struggles take place. He 26
must agree to do no more than watch the struggle being played out, 27
through him and in front of him, without intervening. He must 28
agree to play the role of the victim designate. This is the fate to which 29
one is almost always condemned when one is involved in a game one 30
is not equipped to play and which one can never master (a game one 31
has not chosen to play, which is always played on the home ground 32
of the dominant, in their way, in accordance with their rules and with 33
their weapons of choice). The alternative is to accept the risks in- 34
volved in any attempt at assimilation, in other words in any form of 35
behaviour that is explicitly calculated, designed and organized with a 36
view to bringing about a change of identity, or what he believes to be 37
the transition from a dominated identity to a dominant identity. This 38
implies the danger of denying himself and, correlatively, all of his 39
fellows who reject that choice, who cannot or do not want to act in 40
that way, and thus deny themselves. Abandoning an identity, be it 41
social, political (or more specifically national, as in the case of natu- 42
ralization), cultural, religious or whatever is not without its ambigu- 43

1 ity, especially when it is an identity that is dominated from every

2 point of view, an identity that is stigmatized and despised. In the eyes
3 of those who are being abandoned and left behind, this borders upon
4 treachery; in the eyes of the others, or those one dreams of joining,
5 that one aspires to being, it undeniably implies allegiance, but there
6 is still a suspicion of pretension and selfish calculation.
7 reassuring others Reassuring others and giving them a sense of security, as well
8 as reassuring oneself, and giving oneself a sense of security, con-
9 stitute an imperative incumbent upon any foreign presence. This is
10 the constant preoccupation of any foreigner or anyone who has the
11 feeling of being a foreigner where he is living, of any foreigner to the
12 country and the society in which he lives, often continuously, but
13 who does not experience them as his country and society. He is a
14 foreigner to the economy and culture of that country, and a foreigner
15 amongst the population of that country. As a general rule, this is the
16 case with all traditional immigrants, who never stop emigrating from
17 their homeland. Their children may feel the same even though they
18 are not always or not necessarily, foreigners in the national sense.
19 Anyone who is not in a position of strength, when the balance of
20 power, and especially symbolic power, is not in one’s favour (which
21 is collectively the case with immigrants, or, let me repeat, all those
22 who have a feeling of not really being at home in the place where
23 they are), is anxious not to frighten others. He is anxious not to do
24 so even when there is, objectively, no reason for them to be afraid of
25 him (the immigrant himself has no control over the phantasmatic
26 fears he inspires). He is, to be more accurate, always anxious not to
27 disturb them because a foreign presence is (rightly or wrongly, not
28 that it matters) always a cause for concern (foreigners are those of
29 whom we like to say we don’t know who they are. We don’t know what
30 they are like; we don’t know what makes them tick; we don’t know what
31 they are thinking or how they think; we don’t know what is going on
32 inside their heads; we don’t know how they might react; we cannot
33 understand them; you never know with them).
34 Reassuring the other is often a precondition for one’s own secu-
35 rity. There are only two ways of providing reassurance and self-reas-
36 surance, only two ways of succeeding in reassuring both oneself and
37 others. They complement one another because they are both ways
38 of dispelling the mutual fears. They dispel both one’s own fear (the
39 foreigner’s fear of being in a foreign country) and the fear of others
40 (their fear of a foreigner who is in their country). Both fears (which
41 are different in terms of their form and especially their content) are
42 shared – unequally and differently, of course – by both parties, or
43 by both the dominated and the dominant. The two different fears

feed on one another; and despite all the differences that may exist 1
between them, they are part of the same attempt to reassure. On 2
the one hand, there is the fear of the dominant – in other words 3
and in this case, the masters of the house – who are all nationals, 4
no matter which social class they belong to. It can be allayed by the 5
strength of those who know they are dominant (because they know 6
that they are naturally at home, and know that they are the country’s 7
natural inhabitants), and who know they are in a position of strength 8
because they possess a legitimacy that merges into domination (a 9
legitimacy which, as such, does not realize that it is dominant). On 10
the other hand, there is the fear of the dominated (i.e. immigrants), 11
of the weak who have, in these circumstances, been deprived of all 12
power and all legitimacy. For the dominant, being reassured means 13
no longer having to reassure themselves in the face of some danger 14
(even though there is nothing for them to be afraid of, and even when 15
the danger is completely imaginary) and, at the same time, reassur- 16
ing others whose fear is, so to speak, constitutive of their immigrant 17
condition. For the dominated who, despite their structural weakness, 18
or perhaps because of that weakness, are perceived as dangerous (or 19
at least as constituting a collective danger) or, which is worse, are 20
regarded as ‘enemies’ (and not only as the ‘class enemies’ of old, 21
with whom we were used to coming into conflict), reassuring the 22
dominant is without doubt the price that has to be paid to ensure 23
their own security (which is purely relative). 24
As this self-assurance depends upon a security that has to be won to take refuge 25
from the other or in the face of the other, certain immigrants prefer 26
to withdraw, to take refuge in their hidden fear, and choose (or chose, 27
in an earlier state of immigration) to opt for the greatest possible dis- 28
cretion or, in other words, to become as invisible as they can. They 29
are helped here by the social and spatial relegation of which they are 30
the victims (relegation in space and by space). They also simultane- 31
ously turn it into self-relegation: relegation and self-relegation into 32
the same spaces, the space of social relations, the space of housing 33
and, primarily, the space of work. These are all spaces where they 34
find themselves to be in the majority and amongst other immigrants 35
of the same background (originally from the same country, the same 36
region, the same village, the same kinship group). These are the im- 37
migrants of whom it is said that they ‘hug the walls’, which can only 38
please those who tend to see their reserve as a sign of politeness, or 39
even the eminently reassuring subservience they expect and demand 40
from foreigners. For other immigrants who are sufficiently self- 41
confident, or convinced that they can allay suspicion, providing reas- 42
surance appears to consist in simulating the greatest resemblance 43

1 to or similarity with those they are trying to reassure by disguising

2 their own features, or at least by attenuating the distinctive signs that
3 make them stand out and which are normally described as stigmas.
4 In a word, they do all they can to deny and abolish the radical alterity
5 (or the radicality of the alterity) of which they are the bearers. This
6 attitude, which corresponds to a quest for the greatest proximity and
7 which in fact contains within it all the marks of the allegiance shown
8 to the dominant, is inevitably – despite the objective intentions be-
9 hind it and its self-proclaimed finality – and paradoxically retrans-
10 lated into potential conflicts. It is always liable to be interpreted in
11 terms of rivalry – of unseemly rivalry, illegitimate rivalry and unfair
12 competition. This is an indication of the relatively narrow limits that
13 are ascribed to assimilation, of the limits within which the dominant
14 inscribe the assimilation they wish to impose upon those they domi-
15 nate, and which they are also happy to see them succeed in assimi-
16 lating,3 by conceding them the form without always recognizing its
17 content.
18 children of But the height of both civil and political impoliteness, and the
19 immigrants height of rudeness and violence towards national understanding,
20 seems to be attained by those ‘immigrants’ who are not immigrants:
21 the children of immigrants, those ‘hybrids’ who do not fully share
22 the properties that ideally define the integral immigrant, or the ac-
23 complished immigrant who conforms to the representation we have
24 of him. And nor do they really share the objective, and especially
25 not the subjective, characteristics of nationals. They are ‘immigrants’
26 who have not emigrated from anywhere. They are immigrants who
27 are not, despite that designation, immigrants like any others, in oth-
28 er words foreigners in the full sense of the term. They are not for-
29 eigners in cultural terms, as they are integral products of this society
30 and its mechanisms of reproduction and integration, of a language (a
31 language into which they were born and which, in this country, is not
32 their mother tongue in the literal sense), of education and of all the
33 other social processes. Nor are they foreigners in national terms, as
34 they usually have the nationality of the country in which they are liv-
35 ing. In the eyes of some, they are no doubt ‘bad’ products of French
36 society, but they are still products of that society. Rather like disturb-
37 ingly ambiguous agents, they blur the borders of the national order,
38 and therefore the symbolic value and pertinence of the criteria that
39 found the hierarchy of groups and their classification. And what it is
40 no doubt most difficult to forgive this category of immigrants for is
41 of course the fact that they disrupt the diacritical function and mean-
42 ing of the divorce that state thought establishes between nationals
43 and non-nationals. We therefore do not know how to regard or treat

these new-style immigrants, and nor do we know what to expect of 1

them. And at this point, ordinary fear, if we can put it that way, or the 2
personal or individual fear inspired by the foreign immigrant, turns 3
into a collective anxiety as the traditional separations are abolished 4
and as we lose the simultaneously physical, moral and mental or in- 5
tellectual security and comfort afforded by those eminently reassur- 6
ing separations to the extent that they constitute a protective barrier 7
behind which we can take shelter by asserting that we are ‘at home’, 8
safe from outside interference. 9
This form of anxiety, or this new fear of the immigrant, against disseminated fear 10
which the demand for politeness is powerless, is even more difficult 11
to dispel. It can be disseminated more widely and projected on to 12
a whole series of related objects: young people, difficult neighbour- 13
hoods, bad estates, the suburbs, the unemployed, delinquents and so 14
on. It can be projected on to the same individuals and the same plac- 15
es (the children of immigration or ‘second-generation immigrants’). 16
From that point of view, a radical transformation has taken place 17
within immigration, and the suspicion that continues to weigh upon 18
these new-style immigrants is proportional to the changes brought 19
about by the immigration of families and by their reproduction on 20
the spot. And given these new conditions, we have to go back to the 21
genetic crime that is consubstantial with this immigration, and all the 22
other crimes that have been committed in practice. Basically, we have 23
to go back to the reactions provoked by these crimes, to the way they 24
are judged, and to the ways in which they are assessed. Crimes and 25
infractions are not just forbidden. When they are committed, they are 26
punished accordingly, in other words for what they undoubtedly are, 27
but they are also, surreptitiously and secretly, punished because of 28
the nature of the offender. Even though the immigrant has changed 29
with regard to the outside world, this type of offender is regarded as 30
being illegitimate, as not being allowed to commit infractions, as be- 31
ing forbidden to offend and as not having the right to offend. 32
The suspicion always weighs on the same people. It weighs upon 33
people whose every characteristic – their history and their birth (and 34
in this case, their immigration and their having been born in im- 35
migration) and, correlatively, their social position, their status, the 36
social and especially the symbolic capital they have acquired – desig- 37
nates them as perpetual suspects. The stigmatization revealed by this 38
form of generalized suspicion derives from a schema of thought and 39
social perception with which we are already familiar. In more gen- 40
eral terms, it derives from the suspicious and accusatory relationship 41
we have with the popular classes, which are viewed as dangerous 42
classes. This schema, which is always the same, is as true today as it 43

1 was yesterday, as every age has its own dangerous classes. If the situa-
2 tion specific to the delinquent foreigner (and even more so the ‘immi-
3 grant’, even if he does have the nationality of the country), who is guilty
4 in two ways, or guilty of being guilty, is not necessarily to work to his
5 disadvantage and is not to act as an aggravating circumstance, judges
6 must display great restraint and a lot of self-control, and make an at-
7 tempt at self-correction. Even when it is not openly talked about, this
8 implicit combination of crimes and therefore punishments does give
9 rise to another sanction that is often imposed in addition to the other
10 deportation two. It is intrinsically bound up with the foreigner’s condition, as a for-
11 eigner is by definition liable to be deported, even if, as does happen, it
12 has been agreed not to deport him. Whether the deportation actually
13 takes place or not, the foreigner’s liability to deportation is the sign par
14 excellence of one of the essential prerogatives of national sovereignty.
15 This too is a characteristic of state thought, which is not to say that it is
16 state thought. It is in fact in the very nature of the sovereignty of the na-
17 tion to be able to deport those foreign residents (foreign in the national-
18 ity sense) it sees fit to deport, and it is in the very nature of the foreigner
19 (speaking nationally) to be liable to deportation, regardless of whether
20 or not he is actually deported. Whilst it is not a juridical sanction in the
21 strict sense, as it is not normally pronounced by a court of law, deporta-
22 tion from the national territory, which is an administrative or politico-
23 administrative measure – taken as a result of the judicial condemnation
24 it extends beyond its effects – clearly demonstrates the risks run by any
25 foreigner who infringes the rules of good conduct. Having supplied
26 proof of his lack of discretion, he is subject to administrative sanctions.
27 The same logic governs, a fortiori, the operation of naturalization: the
28 nation and nationality do not naturalize and nationalize just anyone.
29 Being an act that basically results from a decision, naturalization may
30 be incompatible with certain social and cultural characteristics or with
31 certain customs (in the sense of habits and customs). In the French
32 case, it is incompatible with polygamy, which is regarded as an offence
33 against public order in the particular sense in which international pri-
34 vate law understands that term. Naturalization may be incompatible
35 with certain criminal penalties. The nature and hierarchy of some pen-
36 alties disqualify anyone from claiming the quality of being French, but
37 they also vary according to the context and the moment. Not surprising-
38 ly, these crimes reproduce their punishments and bring them into line,
39 roughly speaking, with those that lead to deportation, rather as though
40 the conditions for entering a nationality obeyed, no doubt even more
41 strictly, the same principles as the conditions for entering and residing
42 in the nation, because they precede and prefigure them.
43
For references please consult the bibliography of the book in which
this article was originally published. (see List of sources, page 609)

1
2
Part II 3
4
5
6
Modes of incorporation 7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
marge tekst 36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
‘Minority’ as a sociological concept 3
4
Hans van Amersfoort 5
6
7
8
9
10
11
12
At a time when most scholars were writing about migratory workers and the 13
European guest work system, the geographer Hans van Amersfoort published 14
a study addressing some important questions. Which factors determine the 15
social position of different categories of migrants in the host society? To 16
what extent and under what conditions do ethnic minorities form via the 17
migration process? Van Amersfoort’s work relied heavily on American socio- 18
logical scholarship, resulting in a typology of majority-minority relations and 19
a definition of immigrant ethnic minorities. The study strongly influenced 20
the theoretical basis of both integration studies and integration policy in the 21
Netherlands. 22
23
24
Introduction 25
26
The terms ‘minority’ or ‘minority group’ are widely used in the so- 27
ciological literature. ‘Minority’ appears to be a word with a broad, dif- diffuse meaning 28
fuse meaning and an emotional appeal, exactly the qualities to make of minority 29
it a candidate for political debate. Unfortunately, almost the opposite 30
properties are required if the term is to be used in scholarly analysis. 31
In fact, there are such a variety of meanings and contradictory prop- 32
erties attributed to the term in the scholarly literature that we can 33
hardly speak of a concept that can serve as an analytical tool. 34
The origins of this lack of precision can be traced back to that 35
essay by Louis Wirth which most social scientists take as the start- 36
ing-point of their analyses. Wirth initially describes a minority as: 37
‘A group of people who, because of their physical or cultural charac- 38
teristics, are singled out from the others in the society in which they 39
live for differential and unequal treatment and who therefore regard 40
themselves as objects of collective discrimination’ (Wirth 1945: 347). 41
This is not a very satisfactory definition because it makes the exis- 42
43

184 hans van amersfoort
1 tence of minorities completely dependent on the feelings of minority

2 group members. It is not surprising that Wirth finds it necessary
3 to add new elements to his definition throughout the course of his
4 argument, adding that ‘minorities objectively occupy a disadvanta-
5 geous position in society’ (Wirth 1945: 348). In subsequent passages
6 Wirth mentions a great number of additional properties that are ‘of-
7 ten’ or ‘not necessarily’ present, but which all have some connection
8 with the broad field that he is trying to encompass. He concentrates
9 increasingly on the disadvantageous social position of the minority
10 and tends to neglect its numerical relationship to the wider society,
11 so that ‘the people whom we regard as minority may actually, from
12 a numerical standpoint, be a majority’ (Wirth 1945: 349). If we ac-
13 cept this point of view, in addition to the importance attached to the
14 subjective definition of the situation by the minority found in Wirth’s
15 earlier statements, then every instance of group conflict in society
16 becomes a ‘minority problem’.
17 Other authors have taken even less care about the question of def-
18 inition. Simpson and Yinger, for example, do not arrive at anything
19 like a definition in their textbook Racial and cultural minorities and
20 conclude their introduction with the remark: ‘... we have tried to de-
21 velop a meaning that will be useful in the study of the relationships
22 with which we are concerned’ (Simpson and Yinger 1953: 32). These
23 are simply all kinds of relationship between groups that differ ac-
24 cording to ‘racial’ or cultural criteria. As their starting point for deal-
25 ing with minority group situations they take an earlier classification
26 by Cox in which we find yet another category called a ‘ruling class
27 minority’, such as the Dutch in the colonial East Indies (Simpson
28 conceptual and Yinger 1953: 23). This illustrates the conceptual confusion sur-
29 confusion rounding the term ‘minority’, for Wirth would never have spoken
30 of a ruling class minority, given the weight he attached to the disad-
31 vantageous position and his tendency to disregard the question of
32 numbers. However, Rex has declared that it could never have been
33 Wirth’s intention to go so far as to call, for instance, the Indians dur-
34 ing the British raj a minority (Rex 1970: 25). But Rex seems to under-
35 estimate the consistency with which Wirth’s confusing statements
36 are followed. Thus, in a more recent work, Bloom states explicitly:
37 ‘Minority-majority situations do not depend on mere numbers. In
38 South Africa the twenty-five per cent of the population that is white
39 is the effective majority’ (Bloom 1971: 30). To my mind this is a con-
40 fusing and inaccurate play on words. Moreover, it shows insufficient
41 awareness of the nature of the problem to be analysed, which is the
42 manner and extent to which social position and numerical strength
43 are related.

part ii – modes of incorporation 185
One valuable attempt to make the concept of ‘minority group’ 1

more useful can be found in the work of Wagley and Harris. In their 2
introduction and conclusion to Minorities in the New World they link 3
the concept of minority to the process of state formation and espe- 4
cially to the rise of the modern state (Wagley and Harris 1967: 5, 242 5
ff). It is only in the course of this process that people with different 6
cultural backgrounds become incorporated into one organization that 7
influences an increasing number of aspects of social life. Therefore 8
minorities have the following major characteristics: (1) minorities are 9
subordinate segments of complex state societies; (2) minorities have 10
special physical or cultural traits which are held in low esteem by 11
the dominant segments of society; (3) minorities are self-conscious 12
units bound together by the special traits which their members share 13
and by the special disabilities which they bring; (4) membership in a 14
minority is transmitted by a rule of descent which is capable of affili- 15
ating succeeding generations even in the absence of readily apparent 16
special cultural or physical traits; (5) minority peoples, by choice or 17
necessity, tend to marry within the group (Wagley and Harris 1967: 18
10). On this last point, Wagley and Harris indicate more clearly than 19
Wirth that a minority situation is not only objectively disadvanta- 20
geous; by drawing attention to the continuity of the membership of 21
the minority group over several generations, they stress a subjective 22
side to the definition, a feeling of belonging. Thus minorities are not 23
only categories but collectivities whose sense of solidarity is based 24
on shared values. However, it is especially the first point that adds a 25
valuable element to the discussion and makes it possible to come to 26
a more accurate definition of the concept of a minority. 27
28
29
Minority and state 30
31
The state monopolizes the use of violence within its territory and re- 32
stricts, as far as it is successful, the exercise of power by other units. 33
In this way the state increasingly influences the institutionalized life 34
of society. This says nothing about the question of the relative nu- 35
merical strength of different groups in society. As we emphasized 36
earlier, the concept of minority implies that the numerical strength 37
of a group is connected with its social position. It is not only cumber- 38
some to call the Africans in South Africa a minority, it is also based 39
on the hidden supposition that every state is based on the universalis- 40
tic idea of equality. Petersen has rightly pointed out that the use of the 41
term ‘minority’ implies a ‘democratic moral judgement’ (Petersen democratic moral 42
1965: 235). We therefore have to look for situations in which the state judgement 43

1 is sufficiently developed to exercise a profound influence on social

2 life, and where the numerical size of groups is a decisive factor in the
3 process of policy-making.
4 Wagley and Harris have made this connection between the con-
5 cept of minority, the formation of states and the increasing influence
6 of national, state-controlled institutions on social life. In the Western
7 world, this process runs parallel to the development of the idea of
8 nation-state ‘the sovereignty of the people’ and the rise of the nation state. The na-
9 tion state has, on the one hand, to define who belongs to the ‘nation’
10 and, on the other hand, to decide what are the rights each member
11 should possess. Since the French Revolution the traditional response
12 has been to favour the ‘equality’ of citizens, but this makes it neces-
13 sary to define more carefully what is meant by the terms equality and
14 citizen.
15 Marshall has made a classic analysis of this problem and defined
16 three basic rights three kinds of basic rights and corresponding social institutions: (a)
17 Civil Rights, such as ‘liberty of person, freedom of speech, thought
18 and faith, the right to own property and to conclude valid contracts,
19 and the right to justice’; (b) Political Rights, such as the franchise and
20 the right of access to public office; (c) Social Rights, ranging from ‘the
21 right to a modicum of economic welfare and security to the right to
22 share to the full in the social heritage and to live the life of a civilized
23 being according to the standards in the society’ (Marshall 1964: 71-3).
24 These rights find their expression respectively in the courts, the vari-
25 ous representative bodies, the social services and the schools. Bendix
26 and Rokkan have analysed how these formal rights have been extend-
27 ed to increasingly larger sections of the population. Exceptions that
28 were made for categories such as illegitimate children, Jews, women
29 and illiterates have vanished in the course of time (Bendix 1964: 74-
30 104). In the first instance this applied only to the recognition of for-
31 mal rights, leaving open the question of how far these rights could be
32 exercised in practice. Such a discrepancy between formal rights and
33 the actual possibility of exercising these rights produced an impor-
34 tant focus of social tension during the course of this process.
35 A complete realization of these rights, developed on the basis of
36 individual equality can, in certain cases, result in a burden rather
37 than a privilege. Rights in these circumstances acquire the character
38 of duties imposed by the state. Examples of such duties might be
39 the introduction of compulsory education or compulsory vaccination
40 against certain contagious diseases.
41 In the course of time, state-controlled institutions founded on
42 a conception of ‘equal rights’ replace the institutions of the pre-in-
43 dustrial society based on a kinship and locality. Those sections of

the population that lag behind such developments – for whatever 1

reasons – will also stay behind in their degree of participation in so- 2
ciety. When this ‘backwardness’ is defined as a problem it raises the 3
dilemma of how far the state should transform rights into duties. 4
This is a conflict that often arises with minority groups and deserves 5
attention, because the nature of the conflict is rarely given sufficient 6
emphasis in minority studies. There are many examples of such con- 7
flict, especially in the sphere of education where the absolute author- 8
ity of the father, an integral part of the culture of many minorities 9
from a peasant background, clashes with the state’s desire to uphold 10
the individual rights of women and children. 11
Finally, we must draw attention to the process of decision-making decision-making 12
in modern states. As citizens are rarely unanimous, decisions are 13
taken by a majority vote. It is in this situation that the term ‘minor- 14
ity’ becomes salient. For in the feudal, colonial or totalitarian state 15
the question of majority or minority participation in the decision- 16
making process simply does not arise. The relative numbers of the 17
different segments of the population and the various strengths of in- 18
terest groups do not count in the political process, political life being 19
by definition a matter for the elite alone. Inequality for these states is 20
a natural condition, for they are based on it. 21
In such states the emancipation struggle of underprivileged 22
groups aims primarily at establishing formal political rights. As the 23
ruling elites are small they would soon lose much of their power if 24
formal political rights were used effectively. The characteristic prob- 25
lem for a minority group is not so much that it is difficult to ensure 26
formal rights, but that the numerical situation restricts the possibil- 27
ity of translating such rights into social influence. 28
The concept of citizenship, as it has been developed by Marshall, is 29
used by Parsons in his analysis of the social position of the American 30
Negro (Parsons and Clark 1965: 709-54). Rose also falls back on the 31
concept of citizenship in the introduction to his work on the coloured 32
immigrants in Britain, because it is particularly useful for describ- 33
ing a situation of social deprivation and oppression in general terms 34
(Rose 1969: 13-17, 27-33). But it is even more valuable in that it makes 35
it possible to isolate a few strategic fields in which we can compare 36
the situations of minorities in different states, or of different minori- 37
ties in one state. 38
In the first place we can turn our attention to the legal position. legal position 39
Is the minority situation characterized by a special legal position as 40
compared to the majority? These legal rights can apply to all three 41
spheres of citizenship – civil, political and social. The difference be- 42
tween formal rights and the actual possibility of using them is much 43

1 more difficult to establish. But it is important to examine, when the

2 formal position of the minority is one of equality, whether the minor-
3 ity can in fact use the formal rights in the same ways as the majority.
4 Secondly, the position of the minority with respect to the education
5 system is of great significance. The school is a strategic institution
6 in modern society and regulates further participation in most other
7 social institutions and activities. Organizing and controlling the edu-
8 cation system is therefore one of the major tasks of modern govern-
9 ments.
10 labour and housing Besides these two areas that are controlled directly by the state,
11 markets there are other fields in which, at least in Western countries, the
12 control is not direct, but in which, nevertheless, the state exercises a
13 substantial influence. These are the crucial areas of the labour mar-
14 ket and the housing market. One’s position in the labour market in
15 a modern society is the key to the distribution and personal alloca-
16 tion of goods and services, and it also regulates, to a large extent,
17 an individual’s chances in the housing market. Although this is not
18 completely true, as governments also have a direct influence on the
19 housing market under several types of legislative-provision which
20 may vary between different countries, it does demonstrate that some
21 form of government regulation is generally thought necessary.
22 For the sake of argument, so far I have described the state as a
23 uniform, monolithic body, which is, of course, a gross oversimplifi-
24 cation. But I do want to stress that these four fields – law, education,
25 employment and housing – can be looked upon as being in the ‘pub-
26 lic’ domain. They are, therefore, open to regulation or ‘interference’
27 by the state, although this clearly varies from one society to the next.
28 A number of interest groups, operating both outside and within po-
29 litical parties, try to steer government policy in their direction, and
30 it is certainly possible that some groups are more successful than
31 others in getting their wishes fulfilled. It is not the author’s intention
32 to suggest that there is a necessary or ‘inevitable’ development in the
33 direction of increasing equality.
34 It is exactly in the case of minorities that such a development is
35 not in the least clear. For they are a relatively permanent collectivity
36 opposed in many respects to the majority. This is the crucial distinc-
37 tion between minorities and interest groups, such as farmers, pen-
38 sioners, higher civil servants or divorced women. All these categories
39 are also minorities from a numerical point of view, but their position
40 is not continuously opposed to that of the majority. They can trade
41 their interests in the process of bargaining, propaganda and, most
42 importantly, by forming coalitions. It is precisely in this last respect
43 that a minority does not form part of the political arena. The forma-

tion of a coalition presupposes that there is not a permanently domi- 1

nant majority; if there is such a dominant group then the rationale 2
for coalitions collapses. This is decisive in the case where the formal 3
rights of minorities are recognized. For in mass societies there are 4
always more needs and wants than can be fulfilled at any given mo- 5
ment. The political goal of each group is to get its own needs and 6
wants placed somewhat higher on the priority list than those of other 7
groups, and the formation of coalitions is an essential part of this 8
process. 9
By concentrating on these four basic fields we can substantiate 10
more precisely the claim that a minority must be in a disadvanta- 11
geous position. It is in these fields that the norms of public life should 12
prevail, and these norms are derived from the concept of the ‘equality 13
of the citizen’ and are not based on ascriptive criteria. This implies 14
that there must be a degree of consensus about the prevailing norms 15
in public life. Alternative norms in these fields would not only be an 16
impediment to required social interaction, but would undermine the 17
whole idea of equality on which it is based. It is on this point – the 18
necessity for uniform norms regulating public life – that many stud- 19
ies about the relation between majority and minority concentrate. 20
21
22
Minority and pluralism 23
24
It is possible to think of many different relationships between mi- 25
nority and majority groups. Several writers have categorized these 26
relations and have tried to construct systematic classifications. This 27
raises a fundamental question: is it possible that a minority position 28
can cease to exist without the disappearance of the collectivity? 29
Wirth sees such a possibility as the most desirable solution to ma- 30
jority-minority relations and calls it ‘pluralism’, a term used by many pluralism 31
other writers after him. This term, however, appears to have a great 32
variety of meanings and it is sometimes used with a favourable con- 33
notation, and on other occasions in a negative sense. This diversity 34
of meaning has also led to subtle variations in the term; sometimes a 35
distinction is made between plural, pluralistic and pluriform. I think 36
it is unwise to elaborate on these distinctions: rather, the content of 37
the term could be analysed more carefully. 38
Wirth describes pluralism as: ‘the conception that variant cul- 39
tures can flourish peacefully side by side in the same society. Indeed 40
cultural pluralism has been held out as one of the necessary pre- 41
conditions of a rich and dynamic civilization under conditions of 42
freedom’ (Wirth 1945: 354). He adds the important qualification that 43

1 the majority must not feel threatened, thus implying that there is a
2 limit to the degree of pluralism that a society can tolerate. Simpson
3 and Yinger also seem to recognize this limitation when they define
4 pluralism as ‘cultural variability within the range still consonant with
5 national unity and security’ (Simpson and Yinger 1953: 27). However,
6 neither Wirth nor Simpson and Yinger are explicit about the precise
7 limits of pluralism. It is difficult to see what Wirth has in mind, since
8 pluralism, which seems to me to be a property of society, for Wirth
9 appears to be primarily a property of the minority group. Moreover,
10 there are situations of pluralism which hardly can be looked upon as
11 favourable; for Wirth also states: ‘If there is a great gulf between their
12 own status and that of the minority groups... the toleration of minori-
13 ties may go as far as virtually to perpetuate several sub-societies with
14 the larger society’ (Wirth 1945: 355). This is not a case of toleration in
15 the accepted meaning of the term but a strategy for oppression, and
16 pluralism, in this sense, is in total contradiction with the previous
17 definition of the concept.
18 To make things even more complex, there is another scholarly
19 tradition which uses the term ‘pluralism’ in the analysis of societ-
20 ies that have a culturally diversified population. As Wirth is writing
21 within the American political tradition he tends to use the term in
22 a positive sense. The other tradition stems from the description of
23 colonial and post-colonial societies and there it generally has a nega-
24 tive connotation. These traditions form separate circuits and most
25 writers do not appear to link the two, thus giving us a second reason
26 for analysing the use of the term in both contexts. But our main aim
27 remains to find an answer to the question of what degree of norma-
28 need for tive consensus is necessary to allow a minority full participation in
29 consensus society.
30 Furnivall introduced the term ‘plural society’ to describe the co-
31 lonial societies of South East Asia. He regarded such societies as the
32 product of the colonial state that brought a number of peoples and
33 cultures together as far, and only as far, as this was necessary for eco-
34 nomic purposes (Furnivall 1948: 304). In Holland, the Furnivall tra-
35 dition has been continued by van Lier in his social analysis of the his-
36 tory of Surinam. He argues that ‘every reasonable complex society is
37 made up of elements held together by the state’. In a non-segmented
38 or ‘pluralistic society’ the component parts are the result of a strict
39 division of labour and an unequal distribution of the material and
40 cultural property of the population. This results in the appearance of
41 social strata with different styles of life and diverse customs and tra-
42 ditions. But these differences are mere gradations within one and the
43 same culture, the major portion of which is the common property of

all. Moreover, the members of a pluralistic society are usually of one 1

race and share a common language and religion, and the economic 2
behaviour of the different groups is generally governed by the same 3
motivation. A plural society is marked by an absence of unity of race 4
and religion and, furthermore, the different groups live in separate 5
economic spheres. The differences that arise in this type of society 6
are not gradations within a single culture, but are the result of groups 7
stemming from different ethnic origins and diverse cultural back- 8
grounds. Social strata usually coincide with groups that differ ‘on 9
the basis of racial, cultural and economic criteria as well’ (van Lier 10
1971: 10). 11
It is especially this last sentence that points to the central issue 12
of the whole discussion: are these plural or segmented societies re- 13
ally different from societies that are ethnically stratified? M.G. Smith, 14
who has used the concept of the ‘plural society’ particularly for de- 15
scribing situations in the British Caribbean, is strongly opposed to 16
this view. He states, ‘what characterizes a plural society is that its 17
different segments have different institutional systems. Such insti- 18
tutional systems include kinship, religion, property and economy, 19
recreation etc....’ The only common institutional system is the state 20
or, as Smith phrases it, ‘government’. There is no common value sys- no common value 21
tem and, in the Parsonian tradition where stratification is seen as the system 22
result of the value system, Smith finds ‘no inherent reason why all 23
cultural sections in a plural society should be ranked hierarchically’ 24
(Smith 1965: 82-3). 25
I think it is necessary to stress the word ‘all’ in the preceding sen- 26
tence, because it is certainly not true that access to the only joint 27
institution, ‘government’, is shared equally by all segments. This is 28
not only historically the case in the Caribbean, but it is also neces- 29
sary, from a theoretical point of view, that there should be a centre of 30
integration that keeps the different segments together in one society. 31
The vagueness about the extent to which there is a single hierarchy 32
is a problem that keeps emerging with writers who work with a more 33
or less modified model taken from Smith. 34
These modifications sometimes go to extreme lengths. Rex, for 35
instance, has used Smith’s plural concept in an article on South 36
Africa where he modifies it to the point where it is virtually unrecog- 37
nizable (Rex 1971). He emphasizes the hierarchical relations in the 38
field in which they are concluded: the labour market. It is a major 39
conceptual weakness of Smith’s analysis (as it is in van Lier’s work) 40
that he evades the problem of the values that operate when there is 41
interaction between the members of different groups. In fields in 42
which there is no interaction, such as leisure activities or religious 43

1 ceremonies, the distinct groups can adhere to their own value sys-
2 tems, but, in a domain where they interact, the pattern of relation-
3 ships will be dominated by a single value system.
4 However, it is far from clear how significant the cultural differ-
5 ences must be before we can speak of a plural society. Van Lier, for
6 instance, stresses differences in religion and it is generally agreed
7 that, from an historical or theological standpoint, the gulf between
8 the religions in Surinam is greater than the differences between, for
9 example, Protestants and Catholics in Holland. But does this mean
10 that the saliency of religion as a factor in social life is therefore auto-
11 matically greater? The Northern Ireland situation might cause us to
12 hesitate before jumping to such a conclusion.
13 All these reservations about the way various authors use the con-
14 cept of ‘plural society’ does not alter the fact that there are societies
15 in which we not only have stratification but also a significant degree
16 social distance of social distance of a non-hierarchical character. To a certain extent
17 we find examples of such social distance in all societies. There is
18 everywhere a social distance between rural and urban peoples, be-
19 tween groups with different religions and philosophies of life, and
20 these need not be hierarchical. When such differences become in-
21 stitutionalized then we see the emergence of a plural society. Van
22 den Berghe describes this as follows: ‘social structure is compart-
23 mentalized into analogous, parallel, non-complementary sets of in-
24 stitutions’. Moreover, there is: ‘primacy of segmental utilitarian non-
25 affective and functionally specific relationships between corporate
26 groups and of total, nonutilitarian affective, diffuse ties within such
27 groups’ (van den Berghe 1967: 34-5). Van den Berghe calls this ‘so-
28 cial pluralism’, which he distinguishes from ‘cultural pluralism’. The
29 emergence of ethnic groups in a society he calls ‘cultural pluralism’,
30 whereas a society that is racially structured, but culturally homoge-
31 neous, such as the South of the United States, he terms a socially
32 plural society (van den Berghe 1967: 35-6, 132-3). It seems difficult to
33 make this distinction operational because, in order to be stable, cul-
34 tural pluralism must result in a certain degree of institutionalization.
35 ‘Any form of cultural pluralism’, writes van den Berghe, ‘has a struc-
36 tural facet which can be treated as social pluralism.’ Furthermore,
37 a culturally homogeneous society in which the social strata have no
38 interaction except in strictly specified roles, will become divided into
39 clearly distinguished subcultures (van den Berghe 1967: 135).
40 This is of little help if, instead of trying to analyse an historical
41 situation with the benefit of hindsight, we are confronted by a con-
42 crete society. The more so because the significance of these parallel,
43 non-complementary institutions for the functioning of the society as

a whole is not made more explicit. As I argued before with regard to 1

Smith, so with van den Berghe it is clear that access to the political 2
arena is not in the least organized through these parallel, non-com- 3
plementary institutions. On the contrary, political power is vested in 4
one of the segments of the population which looks upon this ‘cul- 5
tural pluralism’ as a strategy to continue its monopoly of power. 6
It follows from what has been said about the concept of minority 7
so far, that it is the spheres of public life, those that are important for 8
all the inhabitants of the state, that deserve most attention. Cultural 9
diversity in itself is not a problem but is a reality in all states, and 10
particularly the modern industrial state. Structural pluralism, in the 11
sense of parallel institutions, is the accepted rule in areas that are 12
now considered to be part of the private domain, as religion has be- 13
come with the rise of the secular state in the West. How far these 14
parallel institutions may promote or hinder minority participation in 15
the general ‘public’ fields is a separate question, which Gordon has 16
put forward sharply with regard to the United States (Gordon 1964: 17
233-65). 18
However, the more important question is the extent of these ‘pub- 19
lic’ or joint fields. If we want to identify a position as ‘disadvanta- 20
geous’ there has to be a standard of comparison, there must be a 21
field of interaction where the roles of the parties can be described as 22
asymmetrical. ‘Plural’, according to this definition, is a term that can 23
only be applied to a society when: (a) the ‘plural’ organization regu- 24
lates ‘communal’ fields, that is, the public domain; (b) the ‘plural’ 25
organization consists of institutions that are indeed parallel. Where 26
the organization brings groups together in an hierarchical order, we 27
should rather speak of ethnic stratification. 28
One of the few democratic societies meeting these conditions 29
seems to be the Dutch society, as described by Lijphart. Lijphart uses Lijphart’s three 30
the term plural in the American political science tradition and not conditions 31
in the Furnivall-Smith sense. He calls every society plural that ex- 32
hibits: ‘clearly discernible, racial, linguistic and religious differences’ 33
(Lijphart 1968: 3). This description applies to almost every modern 34
society, as long as we do not specify how significant these differences 35
must be. However, he adds two further elements. The first is the 36
obvious point that differences must be institutionalized or, in his 37
words, ‘organized’ (Lijphart 1968: 5). Although this comes close to 38
the ideas of Smith and van den Berghe, Lijphart argues that such a 39
society could hardly continue to exist unless these organized contacts 40
are cross-cutting rather than reinforcing. His third condition is that 41
there must be a diffusion of solidarity because participation, or po- 42
tential participation, is ‘overlapping’. We can easily imagine a society 43

1 where linguistic, religious and economic lines do not run parallel

2 to each other, and where there is such a diffusion of solidarity. To a
3 certain extent Belgium fits this model.
4 This last condition represents a development of the concept that
5 brings it into total contradiction with the formulations of Smith and
6 van den Berghe. These writers would never call a society ‘plural’ if it
7 had cross-cutting loyalties. The interest of the Dutch case, however,
8 is that it does not meet Lijphart’s third condition. Dutch society is, or
9 was at least until the 1960s, organized in a pattern of parallel, mutu-
10 ally reinforcing institutions. That Lijphart continues to label it plural
11 seems inconsistent, but, because we have now reached the situation
12 where writers from the other tradition would use the term, it is inter-
13 esting to follow his analysis.
14 Dutch society: The Dutch ‘pillarized’ (verzuilde) society corresponds to the pic-
15 pillarization ture of a society where there is a lack of consensus between the con-
16 stituent parts of the population, so that it is on the verge of conflict
17 and instability unless power can be monopolized by one of the seg-
18 ments. This raises the key question of Lijphart’s book: how can de-
19 mocracy function under such circumstances? In order to answer this
20 question, Lijphart looks to the specific rules of the political game as
21 they are accepted in this type of society. The fundamental rule is that
22 leaders are obliged to find, in some form or another, a practical com-
23 promise. Furthermore, these rules stipulate that every type of coali-
24 tion must be possible, thus giving every pillar a realistic chance that
25 a good deal of its objectives could be achieved at some time or other.
26 The Dutch system seems to correspond quite closely to the cen-
27 tral feature of plural societies as expounded by Smith and van den
28 Berghe. It is a society in which the population segments are integrat-
29 ed into ‘blocks’ that are not hierarchically ranked one over the other.
30 Such pluralism seems to be only possible under three conditions. In
31 the first place, none of the pillars must be able, through numerical
32 strength or any other factor, to monopolize political power. Should
33 there be a dominant party then there will inevitably develop a system
34 of stratification. Secondly, no party or block must be permanently
35 excluded from participating in government. In Holland there have
36 always been some smaller splinter parties, but these have tradition-
37 ally associated themselves with one of the main pillars. For instance,
38 several of the more strictly orthodox Protestant churches have given
39 birth to such mini-parties, but they have always considered them-
40 selves to be the ‘conscience’ of the major parties. In the third place,
41 these pillars or blocks must succeed in integrating the total popula-
42 tion by these indirect means.
43 It is very doubtful whether this pillar system can be still regarded

as an accurate description of Dutch society. Increasingly individuals 1

and groups have broken away from the traditional pillars so that this 2
third condition is no longer satisfied. Further, it can also no longer 3
be taken for granted that the second condition will be met in the fu- 4
ture. Immigration since the Second World War has brought several 5
categories of immigrants into the country that cannot be regarded 6
as becoming integrated into any of the existing pillars, neither can 7
they hope to develop into a pillar of their own. This is the case with 8
the Islamic peasant migrants from Morocco and Turkey, a category, 9
even when it possesses the vote, that is too small to create its own 10
pillar. Consequently, for groups that are numerically small relative to 11
the society in which they live, a plural organization in itself does not 12
guarantee participation in social life. 13
If we use the term ‘plural’ in the very vague sense of ‘heteroge- obscured profits 14
neous’, it has hardly any meaning. We can say, for example, that 15
Dutch society has become more plural because of the immigration of 16
Muslims, but this obscures more than it illuminates. There is a good 17
example of the confusion to which the unspecific use of the term 18
may lead in Bagley’s study of race relations in Holland (Bagley 1973). 19
By using the term plural he suggests a consistency and continuity in 20
Dutch policy towards immigrants that simply is not supported by the 21
facts. This policy could only be labelled plural if the aim was either 22
to integrate newcomers into one or all of the existing pillars, or to let 23
them develop into a pillar of their own. This last possibility has never 24
been attempted and is, in any case, political nonsense. 25
Furthermore, there can also be cited examples of a great variety 26
of reactions of Dutch society towards immigrants. The early Roman 27
Catholic immigrants from Poland and Italy were absorbed in the 28
Roman Catholic pillar; the Indo-Dutch were absorbed in all three 29
pillars. In the case of the Ambonese soldiers, now generally known 30
as ‘South Moluccans’, absorption was explicitly excluded as an aim 31
during the first decade of their stay. Prior to 1975, there was simply 32
no policy at all with regard to the West Indians. To label all these 33
reactions as ‘plural’ does not clarify the situation in the slightest. 34
If we use plural in the strict sense, as outlined above, it offers 35
a model for social organization that can be profitable to countries 36
which have a population divided along cultural lines, provided there 37
is a certain balance of power between the blocks. However, if we have 38
segments that are numerically small, relative to the other blocks in 39
the total population, such an organization would not give them a 40
chance to promote their interests or to participate in public life. 41
42
43

1 Towards a typology of majority-minority relations

2
3 In the preceding discussion it has repeatedly been stressed that the
4 disadvantageous position of a minority must be distinguished from
5 other positions of disadvantage in which categories and collectivities
6 might find themselves. Schermerhorn has characterized the position
7 of minorities by constructing a typology which is presented here in a
8 somewhat modified form (Schermerhorn 1970: 13).
9
10 Schermerhorn’s typology
11
12 Numerical strength
13 Social power Strongest party Weakest party
14
Dominant Majority Elite
15
16 Subordinate Subordinated masses Minority
17
18
19 In his comment on this typology, Schermerhorn stipulates that there
20 are two characteristic configurations: elite – masses and majority –
21 minority. This is helpful in that we can at least distinguish between
22 these two fundamentally different situations.
23 There are still many other factors that can cause differentiation
24 between the concrete situations of minorities. Wirth has made a clas-
25 sification based on the objectives of the minority, and distinguished
26 between minorities aiming at pluralism, assimilation, secession and
27 domination (Wirth 1945: 354 ff). A practical difficulty with these cat-
28 egories is that it is not always easy to establish what the aims of the
29 minority are. A far more serious criticism for the present discussion
30 is that such a classification does not take into account enough dimen-
31 sions to construct a global typology. If the orientation of the minority
32 is important, we should at least expect that the aims of the majority
33 three dimensions of should also be taken into account. In my opinion, there are at least
34 minority- three dimensions along which minority-majority relations are basi-
35 majority relations cally differentiated.
36
37 Concentration-dispersion
38 In the first place a distinction should be made between concentrated
39 and dispersed minorities. If we take the state as the unit of analysis
40 it is possible that a minority may be numerically stronger in a par-
41 ticular region. This makes it difficult to apply the central idea of the
42 concept of minority group, and it will be necessary in such situations
43 to consider the four public fields with due regard to regional varia-

tions. For ‘concentrated minorities’ other factors play an interven- 1

ing role in minority situations, and these include the absolute size 2
of the group, its size relative to that of the majority at the national 3
and regional level and the practical support given to the group from 4
other countries. These factors will also influence the situation of a 5
dispersed minority but, because such minorities have no core area, 6
their whole orientation will be different and their aspirations will ac- 7
quire a different political expression. 8
9
Particularism-universalism 10
We have already mentioned Wirth’s idea that minorities differ in 11
their aims and that this can be seen as an important aspect of their 12
situation. It is also possible to introduce the orientation of the mi- 13
nority as a simple, dichotomous variable in the construction of a 14
typology. Either the minority can aim at participation in society, or 15
it can be focused exclusively on its internal affairs. In the first case 16
the objective will be to remove any barriers preventing participation. 17
The minority bases its demands in such a situation on the principle 18
of ‘equality’, and, in general, will also demand the preservation of 19
alternative roles. The objective is precisely to get these kinds of roles 20
recognized as ‘alternative’. One wishes to be free to participate as a 21
Jew, Sikh or Black in the society because the public domain ought to 22
be ‘neutral’ with regard to these properties. I will use the standard 23
sociological term ‘universalism’ to label such situations. In the case 24
of concentrated minorities, this orientation can acquire the specific 25
form of regionalism. 26
Particularistic minorities also aim at improving their position 27
but their perception of rights and duties is fundamentally different. 28
They do not demand ‘equal’ rights with the majority, but derive their 29
rights from their own particularistic value system. The extreme case 30
of this is when the minority aims at dominating the majority. 31
32
Emancipation, continuation and elimination 33
A third possibility for classifying minority situations is to use the 34
aims of the majority as a criterion. Simpson and Yinger have made 35
such a classification, but because they do not combine this variable 36
systematically with others, their approach has little practical value. 37
For instance, they give as an example of their variant ‘pluralism’ only 38
cases of what I have called concentrated minorities. (Simpson and 39
Yinger 1953: 24ff). However, the policy of the majority is an impor- 40
tant aspect of the relationship between majority and minority, and 41
it is possible to distinguish between three major types of objective. 42
43

1 (a) Emancipation: The aim is that there should be full participation

2 by the minority in the society. Special legal and social measures may
3 be undertaken to achieve this end. It presupposes that there is a suf-
4 ficiently clear definition of what is thought to be ‘participation’ in
5 society. The vital distinction between other reactions that I will call
6 ‘elimination’ is that such participation need not result in the minor-
7 ity group becoming invisible or ceasing to exist as a collectivity.
8
9 (b) Continuation: Another objective of the majority can also be to
10 continue the present situation. This is an obvious possibility when
11 the minority fulfils certain functions for the majority, as when the
12 minority is exploited. However, this is not necessarily the case, for it
13 depends on the fact that the absolute and relative ‘size of the minor-
14 ity must be substantial. In some situations the goal of minority group
15 continuation may be the result of passive rather than active policies.
16 It is not so much that the objective is to exploit the minority, for the
17 simple reason that there is little to exploit, but rather there is a refusal
18 to pursue a policy of active emancipation for a minority that has be-
19 come part of the society in the course of historical development.
20
21 (c) Elimination: The majority can also aim at the elimination of the
22 minority as a recognizable collectivity. We can distinguish between
23 two variants of this category. In the first case, the majority can aim
24 at forced assimilation by suppressing the constituent elements of a
25 minority such as language or religion. In the second case, the major-
26 ity can attempt the physical extermination of the minority by deporta-
27 tion, population transfer or even genocide.
28
29 From a logical point of view these three variants could be subsumed
30 under the same dichotomy that we used when describing the mi-
31 nority’s orientation. In that case the categories continuation and
32 particularism elimination would merge and come under the label ‘particularism’.
33 However, I think that the distinction between the two categories is
34 sufficiently illuminating to justify retaining this subdivision.
35
36
37 The construction of a typology
38
39 By using these three dimensions we can construct a typology of twelve
40 cells which represent different types of minority-majority relations.
41 advantages While such a typology has its limitations, it also has a number of ad-
42 vantages. It demonstrates how rare are the possibilities for a positive
43 development of relationships between majority and minority. Only

the cells marked ‘emancipation process’ and ‘federalism’ suggest 1

the prospect of a stable form of minority participation in society. All 2
other forms are unsatisfactory for one or both of the parties and are 3
therefore inherently unstable, though the majority may be successful 4
for a period of time in consolidating its position by repression. There 5
may also be a temporary acceptance of the continuation of the status 6
quo by a strongly particularistic minority, which is in effect accept- 7
ing a reservation situation. The development of communications in 8
modern societies, however, is so strong that such groups tend, in the 9
long run, to adopt the wider society as their frame of reference. Thus 10
over several generations they will develop a desire for participation, 11
be it initially only in such fields as consumption and education. 12
In the dimension of ‘elimination’ it is possible to reach a stable 13
situation, although this will also result in the end of the majority- 14
minority relationship. Particularly in the case of concentrated mino 15
rities the outcome of the conflict may be strongly influenced by the 16
international political scene, as has clearly been the case in the popu- 17
lation transfer between Greece and Turkey after the First World War 18
and in the secession of Bangladesh. 19
It is not my intention to suggest that the development in the first 20
two cells I mentioned will necessarily run smoothly. On the contrary, 21
these processes generally provoke a number of disputes and conflicts 22
concerning the extent of emancipation and the degree of autonomy, 23
and the relationship between rights and duties. However, such con- 24
flicts can be resolved within these basic processes. 25
There are a number of objections that can be raised against the ty- disadvantages 26
pology. First, it may give the impression that once they have acquired 27
a certain character these relationships are unchanging. In reality, 28
majority-minority relations can change their character in the course 29
of time, but the typology gives us no information about the direction 30
in which these processes develop. A second objection is the unspe- 31
cific nature of the classification criteria. How are we to determine the 32
‘orientation’ or the ‘objectives’ of a majority or a minority? There are 33
situations in which these orientations are relatively homogeneous 34
and it is not difficult to state in what direction they are pointing. But 35
in many cases the majority or the minority are far from homoge- 36
neous and may aim simultaneously at several different goals, which 37
may even be contradictory. 38
Nevertheless, this typology represents an improvement as com- 39
pared to the lack of precision which characterizes the scholarly, and 40
the more popular literature, on the subject of minorities. It should be 41
remembered that the typology is an elaboration of the one developed 42
by Schermerhorn, and it only deals with majority-minority relation- 43

1 ships. It combines two dimensions that appear in the literature, but

2 only as separate criteria for classification. Schermerhorn has stipulat-
3 ed that the similarity in orientation between majority and minority is
4 in itself an important variable in minority situations (Schermerhorn
5 1970: 83). The combination of orientations can be clearly seen in
6 our typology. Moreover, I have clearly differentiated between concen-
7 trated and dispersed minorities because I have not found this distinc-
8 tion systematically treated elsewhere. It seems, however, that this is a
9 basic variable in the situations in which minorities find themselves.
10
11
12 Conclusion: a revised definition
13
14 To conclude this analysis of the concept of minority or minority
15 group, it is necessary to present a revised definition. It is essential
16 that the concept contains no contradictory properties and that these
17 properties are sufficiently explicit to distinguish the phenomenon
18 from any other phenomena. As a result of the present analysis, it
19 could be argued that a minority has three constituent properties that
20 can be summarized as follows:
21 revised definition 1 A minority is a continuous collectivity within the population of a
22 state. This continuity has two important aspects:
23 (a) the minority consists of several generations,
24 (b) membership of the minority has priority above other forms of
25 social categorization.
26 2 The numerical position of a minority excludes it from taking ef-
27 fective part in the political process.
28 3 A minority has an objectively disadvantageous position in the
29 sense that its members do not participate to the same degree as
30 the majority population in the four following ‘public’ fields:
31 (a) the legal system
32 (b) the educational system
33 (c) the labour market
34 (d) the housing market.
35
36 The translation of this chapter first appeared, in slightly modified
37 form, in Ethnic and Racial Studies, vol. 1, no. 2, April 1978, 218-33. I
38 am grateful to the editor, John Stone, for his help with the translation
39 and for his permission to reproduce it here.
40
43

1
2
‘Black’, racial equality and Asian identity 3
4
Tariq Modood 5
6
7
8
9
10
11
12
In the 1980s, within the sociology of race, a particular concept of blackness 13
was de rigueur. This political concept, inspired by the struggle of racialised 14
minorities in American society, was appealing – at least for those who em- 15
braced it – because it lumped all non-white groups together, thereby sug- 16
gesting they had a lot in common. In an influential, provocative article pub- 17
lished by the journal New Community in 1988, the sociologist Tariq Modood 18
criticised the hegemony of this concept because it would not do justice to 19
the history and experiences of Asian immigrants. The concept of blackness, 20
even when used in a political way, overstates the significance of colour and 21
colour discrimination, while simultaneously understating the ethnic identity 22
of Asians and the ethnicisms they suffer from. 23
24
25
The interpretation of our reality through patterns not our own serves only 26
to make us ever more unknown, ever less free, ever more solitary. 27
Gabriel Garcia Marquez 28
29
Over the last few years a consensus had developed amongst race 30
equality professionals and activists that the term ‘black’ should be 31
used to describe all those who because of their race are unfavour- 32
ably treated within British society. While this idea originated in the 33
contracts with the US Black Power groups and is associated with the the use of ‘black’ 34
Left, particularly with the Black Section movement in the Labour 35
Party, and indeed is most zealously pursued by some Labour con- 36
trolled local authorities, it has become a commonplace so that, for in- 37
stance, it is now current practice of the media not least the BBC. The 38
argument behind this usage of ‘black’ is that it provides the means 39
of affecting a unity between otherwise very diverse, powerless mi- 40
norities that is necessary for an effective anti-racist movement. This 41
argument is thought to be so decisive that it is rare, at least in print, 42
to see it critically considered.1 I believe however, that it required too 43

202 tariq modood
1 high a price in terms of loss of principle from anti-racists and sells

2 short the majority of the people it identifies as ‘black’. In particular it
3 has the effect of imposing a professional-political consensus on the
4 Asian community that was formed by those largely outside it and
5 at a time when Asians as a community were barely participants in
6 debates on race.
7
8 In the use of the term ‘Asian’ I assume that ethnic identities can co-
9 exist at different levels of generality (e.g. Mirpuri, Pakistani, Asian).
10 What I mean by an ‘Asian’ identity is some share in the heritage of
11 the civilisations of old Hindustan prior to British conquest. Roughly,
12 it is those people who believe that the Taj Mahal is an object of their
13 history. As neither they nor the British public in general have yet dis-
14 covered the academic term ‘South Asian’ I shall refer to such people
15 in Britain as ‘Asian’.
16
17
18 Racial simplicities
19
20 The idea that race equality involves the recognition of cultural and
21 ethnic diversity is one that is widely paid lip-service. It is true that
22 talk about cultural variety is sometimes ill-informed or patronising
23 and all too often an evasion from a serious commitment to fight-
24 ing against racial discrimination. Indeed, the factual recognition of
25 cultural plurality is not logically incompatible with forms of hierar-
26 chy. Apartheid is a classic case in point. Nevertheless an organisation
27 which is itself pledged to racial equality cannot but be opposed to
28 the crude categorisations which divide societies and humanity into
29 white and black. While the reduction of an over-lapping and inter-
30 simplistic dualism related plurality into a simplistic dualism is the stock-in-trade of rac-
31 ist thinking it is not a tool available for anti-racists. For the latter are
32 committed to challenging the gross ignorance about peoples and the
33 indifference to their variety that racists utilise. If anti-racists borrow
34 the racists’ classifications in order to defeat racism (‘racists have no
35 trouble in saying who is black, so why should we?’ it is often said)2
36 then however successful or not they may be as an interest group
37 they will have lost their opposition to racism as a way of thinking. In
38 particular, they will have lost the ideal of a multi-racial society for a
39 model of society as composed of two and only two ‘races’ which for
40 the forseeable future must live in conflict.
41
42 If this seems somewhat abstract it is worth noting, in contrast to
43 say the USA or Canada, which are often the models for British race

egalitarians in respect of government action especially in employ- 1

ment policy, the decline of the vocabulary of multi-racialism in the 2
UK. Similarly, one has to note the divergence of the new British race 3
vocabulary from that of America, Africa, India or just about any other 4
part of the globe where ‘black’ continues to mean ‘of sub-Saharan 5
African origin’ (cf. ‘black is beautiful’, ‘black music’, ‘black Africa’). 6
An anomaly which, for example, leads to even greater confusion in 7
the international than the British press as to how many ‘black’ MPs 8
were elected in June 1987.3 9
10
11
Description and identity 12
13
One justification for the new use of the term ‘black’ is that regardless 14
of these inadequacies in other respects it is said to have a descriptive 15
clarity in objectively and factually picking out all those who suffer 16
in common ways from a single form of racism. The drawback here, 17
however, is that most ordinary people wish to be defined in terms of 18
a historically received identity, a distinctive set of beliefs and prac- 19
tices or in terms of their aspirations for themselves and their fami- 20
lies. They may seek more government attention for their problems 21
but just as they wish not to be trapped in a problematic condition 22
– in inner-city decay, in conflict with authority, in alienation from 23
the mainstream, without hope of winning acceptance and graduated 24
progress for their children – so most people do not wish to be defined 25
in terms of a problem or as victims. The situation is exactly analo- 26
gous to the one where social theorists identify persons as proletariat 27
who may have nothing else in common other than this condition and 28
are then surprised to find that the people in question do not make 29
that identity their own; or that the term ‘working class’ fails to offer 30
inspirational identity to a large number of people who on all socio- 31
economic criteria are evidently working class. Most people wish to 32
put on show their best features, those qualities in which their indi- 33
vidual and collective pride resides in and by which they want the rest 34
of the world to know them. They wish to be known for what they are, 35
not for what others find problematic about them. 36
37
38
A black identity 39
40
Now, of course, for many who suffer from white domination ‘black’ black pride 41
has become a focus of collective pride. This is certainly true of 42
many ‘black’ activists and is perhaps quite widely true now of Afro- 43

204 tariq modood
1 Caribbeans here4 and particularly so of Afro-Americans. They may

2 not all share identical notions of black let alone a common political
3 perspective but they do all believe that the term ‘black’ should be
4 used to promote a positive identity. The important point to note is
5 that this use of the term ‘black’, unlike the one I started off with, is
6 no longer descriptive. It is evaluative or aspirational for it denotes
7 not just the negative treatment of others to oneself (of white people
8 to those of another colour) but what one wishes to be or ought to
9 wish to be. One important implication is that, to be slightly technical
10 for a moment, the generic term ‘black’ covers cases which are not
11 equally examples of the genus. Let me illustrate what I mean by an
12 example. ‘Democracy’ is an evaluative generic term which may be
13 used to cover a range of organisations some of which may be more
14 democratic than others, and so the term applies more to some of the
15 organisations than to others. And there is a further assumption that
16 the lesser cases ought to be more like the major cases for the genus
17 is something worthy: the less democratic ought to be more like the
18 more democratic. So similarly the aspirational use of the term ‘black’
19 implies that while some persons or groups are more black than oth-
20 ers, insofar as being black is something to be encouraged, the lesser
21 or more ambiguous blacks ought to aspire to be more like the ‘true’
22 blacks.
23
24 This use of the term ‘black’ may not in itself, any more than ‘de-
25 mocracy’, present any special difficulties. However, there are several
26 factors about the British situation that conspire to make this posi-
27 tive notion of black harmful to British Asians.5 Firstly, because as a
28 matter of historical and contemporary fact this positive black iden-
29 tity has been espoused by peoples of sub-Saharan African roots, they
30 naturally are thought to be the quintessential or exemplary cases of
31 British Asians black consciousness and understand black consciousness to be at its
32 fullest, something only achieved by people of African ethnicity. The
33 Handsworth Harambee organisation thus defined itself on a BBC
34 ‘Open Door’ television programme as rooted in a belief system ‘in-
35 fluenced by Pan-Africanism, African socialism, and parts of the black
36 power philosophy. It is believed that though conditions of black peo-
37 ple are influenced by what happens in Africa, in this way black people
38 can carve out for themselves a decent existence in Handsworth’.6 So
39 if Asians in Britain, by virtue of the discrimination practised against
40 them, come to believe that they too are black in a positive sense it is
41 obvious that only some of the concepts forged by creators of black
42 consciousness will be applicable to Asians so that they will necessar-
43 ily not be capable of being black in the full sense but be only second-
ary or ambiguous blacks.

Secondly, some may claim that when Asians are encouraged to think 1
of themselves as black what is on offer is not the old black conscious- 2
ness, the one forged exclusively by people of African ethnicity and for 3
people of African ethnicity, but a new Afro-Asian identity. Leaving new Afro-Asian 4
aside the question of what is supposed to be the link here between identity 5
the old and the new, the problem here is to know what content this 6
new identity has. For the attempt to reduce several groups who have 7
nothing more in common with each other (except the negative con- 8
dition of discrimination) than any of those groups have with white 9
people to a single identity makes, I must confess, such little sense to 10
me that this concept to me is nothing but a meaningless chimera. 11
At best it marks not so much a positive identity but a positive de- 12
termination to oppose white racism, and the adoption of the term 13
‘black’ here usually means by implication and certainly as a matter 14
of fact, the acceptance by Asians of an Afro political leadership. The 15
latter is evidenced by the relative numbers and especially positions 16
of power (e.g. Chairs of Committees) of Afro and Asian members in 17
inner London Councils’ ruling groups,7 black workers groups such 18
as that in the National Association of Local Government Officers 19
(NALGO), black caucuses and other similar organisations prefixed 20
by the term ‘black’.8 An Afro leadership has of course had some ben- 21
efits for Asians, as for example in the West Indian lobbying which led 22
to the inclusion of the concept of indirect discrimination in the 1976 23
Race Relations Act, but presumably it need not be, as it has pres- 24
ently become, at the price of subordinating their identity to political 25
concerns. Indeed, if the primary mode by which Asians are made to 26
publicly relate to the rest of British society is through a black political 27
identity then no one should be surprised if Asians remain politically 28
under-represented and misrepresented and increasing numbers of 29
successful Asians try to make themselves inconspicuous and opt for 30
a path of apolitical assimilation. 31
32
That Asians cannot be served by a black identity equal to its use for, 33
say, Afro-Caribbeans is perhaps most pointedly illustrated by the fact 34
that when even explicit users of the new concept, in the moments 35
they wish to refer only to Asians do so by the term ‘Asian’, while a 36
book sub-titled ‘West Indians in British Politics’ after a few introduc- 37
tory remarks about the wider ‘black’ community thereon confidently 38
speaks of West Indians as the ‘black community’.9 What is particu- 39
larly significant here is not that the author, Trevor Carter, in writing 40
exclusively of West Indian experience should use the term ‘black’ as 41
an ethnically specific term. What is significant is that Carter, despite 42
his introductory remarks, is able to use the term in this narrower 43

206 tariq modood
1 way without any loss of intelligibility or plausibility. More damaging

2 still to the desire of any Asians to be included in a ‘black’ identity is
3 Peter Fryer’s history of ‘black’ people in Britain which, again, de-
4 spite the usual prefatory remarks about Asians as an integral part
5 of black Britain devotes less than twenty of its six hundred pages to
6 them.10 This process of paying lip-service to the idea of British Asians
7 as blacks while actually being interested in developing a black ethnic
8 identity reaches its apogee in Paul Gilroy’s recent book in which, de-
9 spite some occasional and incidental uses of ‘black’ as a descriptive
10 term (e.g. pp. 45-46), the interest in Asians, spanning two or three
11 sentences only, is confined to the extent that they approximate to black
12 youth culture11. A far more honest approach is that of Fraser’s and
13 Douglas’s Black Heroes in the Hall of Fame in which there is no place
14 for Asians – not out of any hostility but simply inappropriateness.12
15
16
17 ‘Doublespeak’ and racial inequality
18
19 The drawback with ‘black’ used as a descriptive term, then, is that it
20 unconscious defines people not in terms of their own identity but by the treatment
21 doublespeak of others; the aspirational use, on the other hand, overcomes this
22 deficiency but at the price of making British Asians have to define
23 themselves in a framework historically and internationally developed
24 by people in search of African roots. These two situations, of course,
25 describe ideal or abstract cases. Real life is never so simple and the
26 present British situation certainly is not. For that consists of a largely
27 unrecognised ambivalence or confusion arising from the following:
28
29 i) wishing that a single term could be used for all non-white peoples;
30 ii) feeling that ‘non-white’ is a term of negative contrast and noting
31 that at least for some of the referent groups ‘black’ is a positive
32 term and hence to be preferred;
33 iii) noting that the term ‘black’ is not adequately comprehensive nor
34 neutral between different ethnic minority groups for it seems
35 much more apt for some of those groups rather than others.
36
37 Hence we have a kind of doublespeak which in charity one has to
38 suppose is unconscious for otherwise one would have to question the
39 intelligence and/or motives of its users. A sentence like the following
40 is what I have in mind:
41
42
43

Too often when the party discusses the membership of black and 1
Asian people it centres on the level of black public representative- 2
ness, magistrates and MPs, rather on ways in which black people can 3
play a role in the party without necessarily aspiring to hold office; this 4
is not to diminish the important point that many more black people 5
should hold such offices.13 6
7
A sentence which boldly begins with one meaning of ‘black’ imme- 8
diately gives way to an entirely different meaning without any sug- 9
gestion of having done so. Another example of the same general 10
phenomenon is when local authority job advertisements proclaim 11
a desire to attract applications from ‘black and other ethnic minori- 12
ties’ or ‘black and Asian people’. That in each case the second half of 13
these conjunctions is very definitely secondary, an irritating addition, 14
is clear from the fact that regardless of how often these conjunctions 15
are used their order always follows strict precedence. Rare indeed 16
in these contexts would a statement be made in terms of ‘all ethnic 17
minorities including black people’. And to expect a phrase such as 18
‘Asian and Black’ might not seem unreasonable given the size of the 19
respective populations14 or even the convention of alphabetical pre- 20
cedence, let alone the variety normal in the use of language; but it is 21
an expectation which will invariably be disappointed for it misses the 22
hierarchical politics of such formulae. When added to this an institu- 23
tion as central to public opinion formation as the BBC decides that 24
the term ‘Black or Asian’ is too cumbersome and that for the sake of 25
editorial simplicity programme makers have the right to abbreviate 26
that term to ‘Black’, what are Asians in Britain supposed to conclude 27
about their significance as a community in Britain? What is the mes- 28
sage that is being sent out to them? As anyone involved in race equal- 29
ity issues knows, constantly being described as an appendix or as an 30
afterthought erodes one’s sense of one’s worth so that one comes to 31
believe that one perhaps is as secondary or inferior as the benevolent 32
authorities and the media imply. 33
34
35
Ethnic self-definition 36
37
This brings us to the central point at issue, namely, the principle of 38
ethnic self-definition as a basic element of racial equality and multi- 39
racialism. When some time ago American blacks insisted on calling 40
themselves ‘black’ and on being so-called by others this was right- Afro-American 41
ly thought to be an assertion of collective self-respect and respect collective self-respect 42
for which by other races was a basic step towards racial equality. 43

208 tariq modood
1 Similarly, a while ago when many people here of West Indian origins
2 took on the term Afro-Caribbean it was part of a search for an iden-
3 tity in which one could have a sense of worth and resist denigrators.
4 And yet Asians in Britain who do possess a sense of common his-
5 tory and ethnic identity are finding it difficult to hold on to, let alone
6 develop, this identity by the activities of the very people who publicly
7 profess racial equality and in many cases are publicly invested with
8 the task of promoting it. Let me confine myself to one example, the
9 Commission for Racial Equality (CRE). The CRE in so much of its
10 publicity literature, videos, recommended ethnic monitoring catego-
11 ries, through the work of its professionals and so on, increasingly
12 refers to the people about whom it is concerned as ‘black’.15 Yet it de-
13 nies treating Asians in any way less than their due and rejects that it
14 is smothering any distinctive group identity. The CRE’s view seems
15 to be that its proposed categorisation of Asians as black for, say, pur-
16 poses of ethnic monitoring as a tool in equal opportunity strategies,
17 is not an attempt to define Asians as such. Rather it is to pick out an
18 important but limited feature about Asians in Britain while leaving
19 them free to develop their distinctive identity along lines congenial
20 to themselves.
21
22 If I am right in thinking that this is the CRE’s view (in the absence
23 of any official statement it is gleaned from private correspondence
24 and conversations) then it is morally fraudulent. For when local au-
25 thorities,16 academics, politicians, the media and public in general
26 in unison use the categories by which Asians are blacks, and this
27 categorisation becomes second nature so that anyone who questions
28 it is thought to be out of touch, there can be no doubt that the funda-
29 mental identity of Asians in Britain has been defined for them by the
30 mode of reference of the race relations establishment. When I raised
31 this matter with the Community Relations Council of one London
32 late political self- Borough I was told that this issue was out of date, that it had already
33 consciousness been settled by various conferences of professionals and that the
34 among Asians fight against racial discrimination would be best served if the Asian
35 community – coming late to political self-consciousness – accepted
36 it as a fait accompli.
37
38
39 Who knows what Asians think? Who cares?
40
41 Of course some Asians, including prominent figures, do accept the
42 term ‘black’ of themselves. However, this fact has to be balanced by
43 three others. Namely, that there are three other groups of Asians,

each of which is larger than the group just referred to. The largest 1
perhaps is the group that knows that society now refers to them as 2
black, tolerates this while studiously avoiding referring to themselves 3
as black. Then there is the group that feels politically obligated to talk 4
of themselves as black for they see that their political champions, 5
sponsors and other sympathisers talk of them in these ways and ex- 6
pect them to do so too. Finally, there is the group of Asians to whom 7
it simply has not occurred that when local authorities, politicians, 8
media, etc. speak of ‘blacks’, for example as in job advertisements 9
which say ‘applications from black people are welcome’, that they are 10
being referred to.17 It might be thought that this last group must con- 11
sist of those who are least educated, least connected to British society 12
and live in areas of the country where race equality is not a major is- 13
sue. My experience is that this is not so at all and this group can still 14
be found in large numbers in areas such as Brent in London. They 15
persist in a cocoon of ignorance because their own understanding of 16
themselves and of other groups is so different from the assumptions 17
of the local public vocabulary that those assumptions do not even 18
register as possibilities within their framework of understanding.18 19
20
I have made assertions here about what I believe to be true about 21
the large majority of Asians in Britain. It may be asked of me how I 22
can prove these assertions. Perhaps the strict answer is that I cannot 23
and that no one can prove the opposite either. For – and this speaks 24
more loudly than any words – there are very few figures available 25
on this matter. Virtually no one, certainly not the CRE nor the local 26
authorities who confidently assume that Asians think of themselves 27
as ‘black’, nor again those who despite what they know feel no inhi- 28
bition in imposing this identity upon Asians, has thought the Asian 29
community important enough to merit this research and consulta- 30
tion.19 The one research project that has specifically examined grass- 31
roots thinking on this matter has been recently published by the 32
Office of Population and Censuses Surveys (OPCS).20 Their research OPCS research 33
consisted of three separate field tests using three different question on Asians 34
formats and on each occasion in several parts of the country. They 35
found that when in the few cases that Asians ticked themselves as 36
‘Black’ it was mainly done in error due to the design of the form.21 37
While it did not specifically test for this it found no wish amongst 38
Asians to be subsumed under a black identity. It will be interesting to 39
see whether the issue is thought important enough for others to un- 40
dertake further research and for race relation professionals to finally 41
come to respect the principle of ethnic self-definition. 42
43

210 tariq modood
1 It will, incidentally, also be interesting to see if the government it-

2 self listens to the message of the OPCS research. For while it has
3 been the case that the Civil Service Commission has for a number of
4 years been using ethnic classifications which are in tune with OPCS
5 findings, the Home Office in its recent ethnic survey of probations
6 service staff and the Department of Education and Science’s survey
7 of schoolchildren and teachers to begin in autumn 1988 have used
8 the categories fashioned by the CRE and in the former case despite
9 the protest and noncooperation of the National Association of Asian
10 Probation Staff.22 It seems that some government departments have
11 been persuaded by race professionals on this issue just at the mo-
12 ment when Asian opinion is beginning to stir on this point.
13
14
15 Political realism and Asian identity
16
17 I said at the start that the argument behind the new usage of ‘black’
18 consists in the unity it provides for anti-racism. Against this I have
19 tried to show how the argument is not worthy of race egalitarians
20 and necessarily devalues Asians, the numerically larger party in this
21 prospective unity. What now in conclusion must be stressed is that
22 one does not necessarily have to choose between these rival two posi-
23 tions, between an anti-racist common front on the one hand and a
24 respect for Asian identity on the other. What does follow, however,
25 a ‘rainbow is that it is foolish to expect a ‘rainbow coalition’ (as Jesse Jackson
26 coalition’ calls the non-white political alliance in America) to be successful if
27 it involves asking a partner to this coalition to adopt an identity false
28 to their own being. Such a coalition will be only skin-deep and will
29 be betrayed at the first opportunity.23 It is already quite clear how
30 unattractive current race equality campaigning is to the majority of
31 Asians who consistently cross the street to avoid it unless some grant-
32 aid is in prospect. The current uses on the term ‘black’, particularly
33 those which associate it with what is coming to be called a ‘culture of
34 resistance’,24 may create unity amongst a band of militants but will
35 lead more Asians to seek a life of quiet assimilation than otherwise
36 would.
37
38 If we follow further the reference to Jesse Jackson and look more
39 clearly at the American experience we will learn, I believe, that racial
40 inequality and exclusion is overcome not simply by political institu-
41 tional change but with an accompanying restoration of ethnic pride.
42 restoration of This latter was achieved by the ‘black is beautiful’ campaign and is to
43 ethnic pride some extent being emulated by Afro-Britons. Similarly, it is my con-

viction that what we Asians need at present is to develop and project 1

a public identity which will be readily communicable and be true to 2
our own being and our own sense of worth. Only this will give us the 3
confidence to play a more active role in public affairs and assert our- 4
selves as we are and not as others – friend or foe – would wish us to 5
be. An identity which, on the one hand, is capable of fostering pride 6
in our historical heritage and ethnicity and, on the other hand, which 7
can earn us the respect due to us in British society by virtue of the 8
hard work and disciplined commitment that we or our parents have 9
made in establishing ourselves in this country, and by virtue of our 10
growing contribution in the many areas of commerce, law, medicine, 11
education, science, technology and so on. A public identity which is 12
true to our thinking and being is of value in itself; it is also of benefit 13
against those who would distort us into schemes and theories for 14
their own political purposes; and finally, it is of benefit in inspiring 15
and achieving those aspects of racial equality and social success that 16
political initiatives by themselves cannot deliver. 17
18
The development of an authentic public identity is, of course, not 19
an alternative to a politics along the lines of a ‘rainbow coalition’ or 20
a common front against racism. But – and this is important – nor 21
is it in opposition to it. The real question is whether current modes 22
of anti-racism will be sufficiently adaptive on this point. For Asians 23
cannot be expected to embrace a political race equality which denies 24
them the distinctive public recognition that they seek. 25
26
The choice, then, is not between a separatist Asian ethnicity and uni- 27
ty of the racially oppressed; the choice is between a political realism 28
which accords dignity to ethnic groups on their own terms and a 29
coercive ideological fantasy. 30
31
32
Notes 33
34
1 See, however, Sandip Hazareesingh’s excellent ‘Racism Cultural Identity: 35
An Indian Perspective’, Dragon’s Teeth 1986, 24:4-10. See also the brief edi- 36
torial in the ‘Race and Society’ supplement of New Society, 6.11.1987. For an
example of the anti-Asian prejudice that the latter hints at, see the editorial
37
and ‘Voice of Alex’ in Platform, October 19 and 25, 1987 respectively. 38
2 This argument is often generalised from a reference to racists to white so- 39
ciety in general: ‘the dominant popular culture continues to insist on us- 40
ing the word ‘black’ to identify people of both Afro-Caribbean and South 41
Asian descent’ (Richard Jenkins ‘Countering Prejudice – Anthropological or
42
Otherwise’, Anthropology Today 1987, 3:2). Popular culture, however, has yet
43

212 tariq modood
1 to oblige: it is a common experience of race/racism awareness trainers that

2 most white people use the term ‘coloured’ when they should use ‘black’ and
3 that they need to be taught the new vocabulary.
3 See, for example, Newsweek, 4 January 1988: 32-33. Newsweek, however, can
4
be forgiven for overlooking that in Jonathan Sayeed, the Conservative M.P.
5 for Bristol East, Westminster has had a half-Indian M.P. since 1983 for none
6 of the British media has noted the fact either.
7 4 This is not, of course, universally so even amongst the younger generation.
8 For a voice of dissent see Ferdi Dennis ‘History – Fact or Fiction?’, The Voice,
week ending 25 March 1987. Indo-Caribbeans of course are defined out of
9
existence by the current idea that the term ‘Afro-Caribbean’ is simply an
10 update on the term ‘West Indian’, see Lynette Lithgow ‘East Indians and the
11 West Indies’, Asian Times, 4.3.1988, pp. 4-5.
12 5 Probably also harmful to (amongst others) some black people: it can’t be
13 much help to develop a term of ethnic pride and then see it applied indis-
14 criminately to non-white peoples. I confine my concern here to Asians.
6 Quoted in John Rex and Sally Tomlinson Colonial Immigrants in a British
15
City, London, 1987, Routledge and Kegan Paul, p. 260.
16 7 See Peter Hamid ‘Asian Involvement in the Political Life of Great Britain with
17 Parallels Drawn from the Afro-Caribbean Experience’, Shakti, November
18 1986, pp. 16-17.
19 8 Research suggests that not only do the majority of Asians not join such or-
ganisations but they do not even approve of their existence. The Harris poll
20
for Caribbean/Asian/African Times found only 31 per cent of Asians in favour
21 of the setting up of a Black Section in the Labour Party (African Times, 5 June
22 1987: 22).
23 9 Trevor Carter Shattering Illusions, West Indians in British Politics, London,
24 1986, Lawrence and Wishart.
25 10 Peter Fryer Staying Power: The History of Black People in Britain, London,
1984, Pluto Press Ltd.
26
11 Paul Gilroy There Ain’t No Black In The Union Jack, London, 1987,
27 Hutchinson.
28 12 Flip Fraser and J.D. Douglas Black Heroes in the Hall of Fame, a stage musical
29 currently (December 1987) showing at the Astoria Theatre, London to wide
30 critical acclaim, not least because of its contribution to black pride.
13 Chosen at random from Positive Discrimination: Black People and the Labour
31
Party (The Labour Party, 1985: 20). ‘Doublespeak’ sentences like these can
32 be found daily in virtually any book, article or newspaper item on race. Even
33 an Asian paper like New Life which is normally very clear on these matters
34 can occasionally find itself in this sort of incoherence (see the editorial 19
35 June 1987: 6).
36 14 Most lay (and some professional) white, black and Asian people seem to be
actually unaware of the numbers - for example, that there are more than
37
twice as many British Asians as British Afro-Caribbeans. If as some guess
38 (Gujarat Samachar, Special Issue, August 1987, p. 19) that over 35 per cent of
39 West Indian immigrants were people of Asian origin then the statistics are
40 considerably further complicated.
41 15 This is particularly true of the Employment Division. See, for example, its
42 Positive Action and Equal Opportunity in Employment (CRE, 1985) which on
p. 3 states that ‘the term “black” is used as a general description for ethnic
43 minority groups, including those of Afro-Caribbean and Asian origin and

other groups who experience discrimination on grounds of race, colour or 1

national origins.’ This comprehensive use sits cheek by jowl with a quote on 2
the very next page from the White Paper preceding the 1976 Race Relations 3
Act which speaks of ‘black and brown workers’.
4
16 Taking the country as a whole many local authorities do not yet think of
Asians as ‘blacks’ though it certainly is a growing trend. When they do switch 5
to the new usage the usual justification is that they are merely following CRE 6
guidelines. Nevertheless a noticeable counter current too has emerged. In 7
the face of local Asian protest some councils, usually only in areas where 8
Asians greatly outnumber Afro-Caribbeans, have recently formally decided
9
to not classify Asians under the term ‘black’. Three cases known to me are
Leicestershire, L.B. of Hounslow and Peterborough City Council. The issue 10
is at the centre of considerable controversy in L.B. of Brent where Asians 11
outnumber Afro-Caribbeans by only two to one, and is a live issue in L.B.s of 12
Tower Hamlets, Waltham Forest and Hillingdon. 13
17 In which case such job advertisements seem to be prima facie unlawful indi- 14
rect discrimination though as far as I know no cases have appeared before an
15
industrial tribunal.
18 Richard Jenkins in ‘Countering Prejudice – Anthropological and Otherwise’ 16
in Anthropology Today 1987, 3:2 offers ethnic chauvinism as the one and only 17
reason why Asians cannot identify themselves as ‘blacks’. This is, of course, 18
as absurd as saying the only reason the Welsh have for objecting to the popu- 19
lar international conflation between English and British is Welsh chauvin-
20
ism! Anti-racist intellectuals, even when friends, would do well to extend the
sources of their ethnic understanding. 21
19 I understand that the CRE is coming to the view that this is an issue which 22
will not go away and some consultations ought to be undertaken. If this is 23
indeed so I hope that these consultations will not be confined to or centre on 24
race professionals and CRCs (their majority view is not a secret and indeed 25
is the problem) but will be directed to Asian organisations and could very
26
simply be supplemented by commissioning an opinion poll.
20 Office of Population, Census and Surveys Developing Questions on Ethnicity 27
and Related Topics for the Census, Occasional Papers, 36, 1987. 28
21 Ibid: 64. Indeed, two details which emerge from this research show that 29
Asian and black modes of self-identity continued to diverge rather than con- 30
verge. While in the course of their field tests OPCS removed ‘West Indian’
31
and ‘African’ as sub-sections of ‘Black’ because a distinct number of black
people objected to being defined in terms of overseas origins and any sub-di- 32
visions, Asians while accepting ‘Asian’ as a generic term wished to be further 33
classified by reference to national origins (e.g., ‘Indian’). Similarly, while a 34
number of young blacks advocated the category ‘Black British’ for blacks 35
born in the U.K., Asians deprecated the category ‘British Asian’ for it implied 36
that British Asians were not British unless born here. Asians, it seems, are
37
searching for a British identity which is not incompatible with overseas ori-
gins and, no doubt, continuing overseas links. 38
22 The National Association of Asian Probation Staff described the use of a 39
‘black/white’ classification as ‘divisive and itself racist’ (quoted in New Life, 40
25 December 1987: 2). 41
23. Consider the Harris Poll, op. cit., which suggests that the Conservative vote 42
amongst Asians living in areas of low ethnic concentration is now not much
below the national average. 43

214 tariq modood
1 24 See, e.g., C. Gutzmore ‘The Notting Hill Carnival’, Marxism Today, August
2 1982. Though I cannot speak with any authority here I do not believe that
many black people welcome such descriptions of themselves.
3
4
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Introduction to immigration and the politics of 3
4
citizenship in Europe and North America
5
6
William Rogers Brubaker 7
8
9
10
11
12
This article is the introductory chapter of a 1989 publication edited by soci- 13
ologist Rogers Brubaker, entitled Immigration and the Politics of Citizenship in 14
Europe and North America. The book was one of the first transatlantic com- 15
parisons of the links between immigration and citizenship. Though the vol- 16
ume only deals with six countries, Brubaker’s introductory chapter well places 17
the debates on citizenship, membership and immigration in a historical per- 18
spective. It also elegantly presents the major issues to be discussed when 19
dealing with issues of nationhood and citizenship in a migratory context. 20
21
22
23
Massive postwar migrations have posed a fundamental challenge to 24
the nation-states of Europe and North America. They have compelled 25
these countries to reinterpret their traditions, to reshape their institu- 26
tions, to rethink the meaning of citizenship – to reinvent themselves, 27
in short, as nation-states. 28
This book addresses one important aspect of this challenge. It 29
is concerned with the implications of immigration for the theory 30
and practice of citizenship and membership in the United States, citizenship 31
Canada, the United Kingdom, France, West Germany, and Sweden. 32
Much has been written about immigration to these countries, but 33
little has been written about citizenship.1 Through a broad compara- 34
tive discussion of citizenship and social membership – the first of 35
its kind – the book aims to bring fresh perspectives to bear on the 36
intensifying policy debates about immigration and citizenship. 37
The authors make arguments about how citizenship and mem- 38
bership ought to be organized. And they make clear how citizenship 39
and membership are in fact organized. The essays in the first part of 40
the book incline toward political argument, the essays in the second 41
part toward policy analysis. But the distinction is not a rigid one. 42
Most of the essays involve both argument and analysis. 43

216 william rogers brubaker
1 The essays in the first part of the book articulate a wide variety
2 of viewpoints. This reflects the preference of the German Marshall
3 Fund (and of this writer) for a lively clash of perspectives over a cho-
4 rus of carefully orchestrated bromides. The authors challenge tradi-
5 tional views of citizenship and membership. Joseph Carens disputes
6 the traditional, state-centered view that moral considerations are out
7 of place in decisions about admission to citizenship. Peter Schuck
8 argues that American citizenship has lost much of its value and
9 meaning. Kay Hailbronner challenges the widespread notion that
10 the Federal Republic of Germany has an unreasonably restrictive citi-
11 zenship policy. And Tomas Hammar takes issue with the traditional
12 negative attitude towards dual citizenship.
13 The essays in the second part of the book look through a com-
14 parative lens at citizenship and membership policies and practices.
15 I discuss citizenship law and naturalization practices. Mark Miller,
16 questioning the traditional view of noncitizens as politically passive,
17 analyzes the many ways in which noncitizen immigrants participate
18 in politics. And in the concluding essay, I discuss the economic and
19 social rights of noncitizens.
20 comparing The six countries examined in the book have very different tradi-
21 six countries tions of immigration and citizenship. Canada and the United States
22 are classical countries of immigration whose citizenship policies
23 have long been geared to mass immigration. Britain and France are
24 former colonial powers whose immigration and citizenship poli-
25 cies reflect in complex ways the legacy of colonialism. Sweden and
26 Germany are traditional countries of emigration whose postwar
27 prosperity led to the recruitment, initially on a temporary basis, of
28 migrant workers.2
29 Despite these differing traditions, each of these countries today
30 confronts similar problems. During the last quarter-century, each has
31 experienced a “new immigration” – to borrow the expression used
32 to describe the surge in immigration from Southern and Eastern
33 Europe to the United States in the late 19th century. And the United
34 States has experienced a “new ‘new immigration.’” Thus Asia is now
35 the leading source of immigration to both Canada and the United
36 States; the Indian subcontinent and the Caribbean displaced Ireland
37 in the 1960s as the leading source of immigration to Britain; half of
38 the foreign population in France is now from Africa or Asia (mainly
39 from North Africa); Turks surpassed Italians during the 1970s as the
40 largest group of foreign workers in Germany; and Asia has recently
41 displaced Nordic countries as the leading source of immigration to
42 Sweden.
43 Contemporary debates about citizenship are simultaneously de-

bates about nationhood. They are debates about what it means, and 1
what it ought to mean, to be a member of a nation-state in today’s 2
increasingly international world. To place these debates in perspec- 3
tive, this introductory essay begins by evoking in general terms the 4
challenge posed by immigration to the nation-state and sketching 5
the historical background to current debates about immigration and 6
citizenship in each of the six countries. Next, it outlines the major 7
questions facing policy makers and sketches the options they have in 8
addressing these questions. The introduction concludes with some 9
remarks about the individual essays. 10
11
12
The challenge to the nation-state 13
14
Citizenship today means membership of a nation-state. To note this 15
is to point to a basic fact of political and social organization. We live 16
in a world of nation-states. Each claims a certain fraction of the hu- 17
man population as its own, and each aspires to mould this popula- 18
tion – its citizenry – into something more than a mere aggregate 19
of individuals or a mere congeries of groups. Each aims to create a 20
cohesive and in some respects homogeneous nation. The persistent 21
ethnic strife that afflicts many polities is a brutal reminder that this 22
aspiration often goes unrealized. The aspiration, though, is shared 23
even by such fundamentally multicultural polities as India and the 24
Soviet Union. 25
But the nation-state is not only a fact. It is also an idea or ideal – a ideal of nation-state 26
way of thinking about political and social membership.3 It is a deeply 27
influential model of membership that informs much current debate 28
on immigration and citizenship. Membership, according to this 29
model, should be egalitarian, sacred, national, democratic, unique, 30
and socially consequential. The membership status of postwar im- 31
migrants to Europe and North America, however, deviates from this 32
model in every respect. This has strained deeply rooted shared un- 33
derstandings about the way social and political membership ought 34
to be organized, and it has occasioned talk of a “crisis of the nation- 35
state.” 36
Because it remains so influential, I want to look more closely at 37
this model of membership and say something about each of its com- 38
ponents. In sketching this model, I am not endorsing it. I want sim- 39
ply to summarize certain inherited ideas and ideals that continue to 40
inform political debates and discussions about immigration, about 41
nationality and citizenship, about patriotism and national identity, 42
about military service and the welfare state. I want to sketch the back- 43

1 drop of taken for-granted ideas and ideals against which the politics
2 of membership unfolds today.
3 six ideals of state- What are these ideas and ideals? First, state-membership should
4 membership be egalitarian. There should be a status of full membership, and no
5 other (except in the transitional cases of children and persons await-
6 ing naturalization). Gradations of membership status are inadmis-
7 sible; nobody should be a second-class citizen.
8 Second, membership should be sacred. Citizens should be pre-
9 pared to make sacrifices – etymologically, to perform “sacred acts”
10 for the state. They should be willing to die for it if need be. “Profane”
11 attitudes toward membership, involving calculations of personal ad-
12 vantage, are profoundly inappropriate.
13 Third, state-membership should be based on nation-membership.
14 The political community should be simultaneously a cultural com-
15 munity, a community of language, mores, or belief. Only thus can a
16 nation-state be a nation’s state, the legitimate representative and au-
17 thentic expression of a nation. Those aspiring to membership of the
18 state must be or become members of the nation. If not (presumptive-
19 ly) acquired through birth and upbringing, such nation-membership
20 must be earned through assimilation.
21 Fourth, membership should be democratic. Full membership
22 should carry with it significant participation in the business of rule.
23 And membership itself should be open: since a population of long-
24 term resident nonmembers violates the democratic understanding
25 of membership, the state must provide some means for resident
26 nonmembers to become members. Over the long run, residence and
27 membership must coincide.
28 Fifth, state-membership should be unique. Every person should
29 belong to one and only one state. Statelessness can be catastrophic
30 in a world in which even so-called human rights are enforceable for
31 the most part only by particular states. And dual (or multiple) citizen-
32 ship has long been considered undesirable for states and individuals
33 alike. There are legal techniques for regulating and mitigating the
34 conflicts, inconveniences, and ambiguities it causes. But these tech-
35 niques cannot solve the central political problem of dual citizenship
36 – the problem of divided allegiance.
37 Lastly, membership should be socially consequential; it should be
38 expressed in a community of well-being. Membership should entail
39 important privileges. Together with the duties mentioned above,
40 these should define a status clearly and significantly distinguished
41 from that of nonmembers. Membership should be objectively valu-
42 able and subjectively valued – it should be prizeworthy and actually
43 prized.

This model of membership is largely vestigial. It is riddled, vestigial model 1

moreover, with unresolved internal tensions. The idea of an egalitar- 2
ian and democratic membership points in one direction; the idea 3
that membership should be sacred and based on cultural belonging 4
points in a very different direction, with different policy implications. 5
That the model survives is due mainly to the lack of a coherent and 6
persuasive alternative. We lack a developed political theory of partial 7
or limited state-membership. We lack a political theory of “desacral- 8
ized” membership, based solely on calculations of personal advan- 9
tage, or of political membership dissociated from cultural belonging, 10
or of dual or multiple membership. 11
Because it is vestigial, the model is significantly out of phase with 12
contemporary realities of state-membership. There are conspicuous 13
deviations from the model that have nothing to do with immigration. 14
The desacralization of state-membership, for example, has more to 15
do with the emotional remoteness of the bureaucratic welfare state 16
and the obsolescence of the citizen army in the nuclear age than 17
it does with immigration and occasional naturalizations of conve- 18
nience. And if modern-day citizenship is not very robustly demo- 19
cratic, this has more to do with the attenuated participation of most 20
citizens in the exercise of sovereignty than it does with the exclusion 21
of noncitizens from the franchise. 22
Still, the postwar immigration has accentuated existing deviations 23
from the nation-state model and generated new ones. These include 24
the proliferation of statuses of partial membership; the declining 25
value of citizenship; the desacralization of membership through the 26
calculating exploitation of the material advantages it confers; the in- 27
creasing demands for, and instances of, full membership of the state 28
without membership of the cultural nation; the soaring numbers of 29
persons with dual citizenship; and the long-term exclusion of large 30
numbers of apparently permanent residents from electoral participa- 31
tion. These membership trends deviate from every component of the 32
nation-state model. And each one arises from the unexpected devel- 33
opment of postwar immigration. 34
Unexpected especially on the Continent: for what has become a 35
settlement immigration began in France and Germany and Sweden 36
as a temporary labor migration. Neither a strictly temporary guest- 37
worker system nor unambiguous and accepted settlement immigra- 38
tion poses insuperable problems of membership. But an impercep- 39
tible slide from labor migration to settlement immigration, a slide 40
only partially and belatedly acknowledged by the immigrants them- 41
selves and by the country of immigration, could not help generating 42
delicate problems of membership. And equally delicate problems of 43

1 membership are posed by the gradual settlement of undocumented

2 alien workers and their families in the United States.
3 The membership status of these migrants-turned-immigrants
4 has developed on the Continent in an ad hoc fashion with the piece-
5 meal administrative, legislative, and judicial acknowledgment of
6 their membership status.4 This piecemeal process of inclusion con-
7 trasts with the “total” transformation effected by naturalization.
8 Paradoxically, the further this process has gone, the weaker the in-
9 centive to naturalize. Ad hoc enlargements of migrants’ rights may
10 obstruct rather than clear the path to full membership, trapping large
11 numbers of migrants-turned-immigrants in an intermediate status,
12 carrying with it many of the privileges and obligations of full mem-
13 bership but excluding two of the most important, symbolically and
14 practically: the right to vote and the duty of military service.
15 The immigration was unexpected, too, in its volume and in its
16 steadily increasing ethnic diversity. This holds for the United States
17 and Britain (and, as far as ethnic diversity is concerned, for Canada)
18 as well as for the Continent. Against the backdrop of the model of
19 membership sketched above, this threefold unexpectedness helps to
20 explain the profound political uncertainty of North America and es-
21 pecially Europe in the face of today’s increasingly settled and increas-
22 ingly assertive immigrant population.
23 fundamentalism Not everyone shares this uncertainty, of course. Fundamentalists
24 defend the traditional model of the nation-state, stressing in particu-
25 lar the idea that state-membership presupposes nation-membership.
26 Multicultural pluralists, on the other hand, deny any validity to this
27 model, arguing for new forms of political membership that would
28 mirror an emerging postnational society. Fundamentalists demand
29 of immigrants either naturalization, stringently conditioned upon
30 assimilation, or departure; multicultural pluralists demand for im-
31 migrants a full citizenship stripped of its sacred character and di-
32 vorced from nationality. Neither position is particularly nuanced.
33 Fundamentalists treat the nation-state as something frozen in social
34 and political time; theirs is a profoundly anachronistic interpretation.
35 Multicultural pluralists, in their haste to condemn the ‘nation-state
36 to the dustbin of history, underestimate the richness and complexity
37 of the nation-state model. If suitably reinterpreted to take account of
38 the changing economic, military, and demographic contexts of mem-
39 bership, the nation-state model may have life in it yet.
40
41 Traditions of nationhood and the politics of citizenship
42 The ideas and ideals sketched above inform the politics of citizenship
43 on both sides of the Atlantic. Yet, for historical reasons, the contours
of debate vary from country to country.

Above all, there is a basic difference between nations constituted 1

by immigration and countries in which occasional immigration has immigration and 2
been incidental to nation-building. Canada and the United States nation-building 3
have a continuous tradition of immigration. They were formed and 4
reformed as nations through immigration, and immigration figures 5
prominently in their national myths. 6
No European country is a classical country of immigration in this 7
sense. This is not to say that Europe has no historical experience with 8
immigration. Industrialization in Europe as elsewhere was accom- 9
panied by massive labor migrations, often across state boundaries, 10
and often leading to settlement. Poles in the coal mines of the Ruhr 11
and on the Junker estates of Prussia, Irish in the northern industrial 12
cities of England, Belgians and Italians in the frontier and industrial 13
regions of France – these and other labor migrants of the second half 14
of the 19th century became permanent settlers. 15
Yet immigration has not been central to European nation-build- 16
ing, not even in France. Concerned about the low birth rate and about 17
the devastating losses in the world wars of this century, the French 18
state has long promoted immigration for demographic reasons. In 19
sheer numbers, immigration has been much more important in 20
France during the last hundred years than in any other European 21
country. Yet not even in France does immigration form part of the 22
national myth. 23
The massive immigration of the last quarter-century has not 24
transformed European countries into countries of immigration in 25
the classical North American sense. Even Sweden, which has gone 26
furthest in acknowledging and accepting its postwar labor migrants 27
as permanent settlers, makes it clear that it is not and cannot become 28
a country of immigration in the classical sense. 29
Debate about immigration and citizenship in each of our six coun- 30
tries is informed by distinctive traditions of nationhood – by deeply 31
rooted understandings about what constitutes a nation. A few observa- 32
tions about these traditions may help to explain some of the striking 33
national differences in the contemporary politics of citizenship. 34
France was the first nation-state, and it has remained the nation- 35
state par excellence. French conceptions of nationhood and citizen- 36
ship bear the stamp of their revolutionary origin. The nation, in this 37
tradition, has been conceived mainly in relation to the institutional 38
and territorial frame of the state: political unity, not shared culture, 39
has been understood to be its basis. “What is a nation?” asked Abbé 40
Sieyès in his famous pamphlet of 1789, and answered: “a body of 41
associates living under one and the same law and represented by 42
one and the same legislature.” But if political unity has been funda- 43

1 mental, the striving for cultural unity has been crucially expressive of
2 French nationhood. Political inclusion has entailed cultural assimila-
3 tion, for regional cultural minorities and immigrants alike. The uni-
4 versalist, inclusive theory and practice of citizenship have depended
5 on confidence in the assimilatory workings of schools, the army, the
6 church, unions, and political parties – confidence that has waned
7 markedly in recent years.5
8 If the French conception of nationhood has been universalist, as-
9 similationist, and state-centered, the German conception has been
10 particularist, organic, and Volk-centered. Because national feeling
11 developed before the nation-state, the German idea of the nation
12 was not originally a political one, nor was it linked with the abstract
13 idea of citizenship. This pre-political German nation, this nation in
14 search of a state, was conceived not as the bearer of universal politi-
15 cal values, but as an organic cultural, linguistic, or racial community
16 – as a Volksgemeinschaft. On this understanding, ethnic or cultural
17 unity is primary and constitutive of nationhood, while political unity
18 is derivative. While this way of thinking about nationhood has never
19 had the field to itself, it took root in early 19th century Germany and
20 has remained available for political exploitation ever since; it finds
21 expression even in the Basic Law of the Federal Republic.6
22 One would expect citizenship defined (as in France) in political
23 terms to be more accessible to culturally distinct immigrants than
24 membership defined (as in Germany) in ethnic or cultural terms.
25 This is in fact the case. The policies and politics of citizenship in
26 France and Germany have been strikingly different since the late
27 19th century, and they remain so despite converging immigration
28 policies and comparable immigrant populations. As a result, a sub-
29 stantial fraction of the French immigrant population has French citi-
30 zenship, while only a negligible fraction of the corresponding West
31 German population has German citizenship.
32 assimilation The postwar migrations, to be sure, have placed considerable
33 strain on French and German traditions alike. The French tradition
34 of assimilation finds few defenders today: the multiculturalist left
35 and immigrant organizations argue that immigrants should not be
36 assimilated, the exclusionary right that they (the North Africans in
37 particular) cannot be assimilated. The far right, led by Jean-Marie
38 Le Pen, has embarked on a major campaign to “revalorise” French
39 citizenship by restricting immigrants’ access to it. Le Pen’s slogan –
40 ”Etre français, cela se mérite” – means roughly: “to be French, you
41 have to deserve it.”
42 Nor is it only the French tradition of inclusion via assimilation
43 that is under strain. The current conservative government of West

Germany has had to acknowledge that large numbers of Turkish mi- 1

grants have in fact become permanent immigrants. It has even pro- 2
claimed a public interest in the naturalization of second-generation 3
immigrants. 4
It is too early to predict the outcome of the contemporary poli- 5
tics of citizenship in France and West Germany. But the bearing of 6
traditional shared understandings of nationhood on this politics is 7
clear. French moves toward a more restrictive and German moves 8
toward a more liberal politics of citizenship have encountered strong 9
resistance. The French government withdrew its proposed, mildly 10
restrictive reform of nationality law in December 1986 after meeting 11
unexpectedly strong opposition. (Dissenters included the venerable 12
Council of State, which criticized the reform as “contrary to republi- 13
can tradition and principles.”) It subsequently appointed a nonpar- 14
tisan commission to study the issue; the changes proposed in the 15
commission’s report, if enacted, would actually liberalize access to 16
French citizenship for second-generation immigrants. On the other 17
hand, every recent proposal to liberalize German nationality law has 18
foundered in the upper house. A central argument has been that the 19
current restrictive nationality law is appropriate for a country that, by 20
inescapable tradition, is not and cannot become a country of immi- 21
gration.7 22

In Sweden, as in France, national feeling and state institutions Sweden: absence of 23
developed in tandem long before the age of nationalism. The sense ethnic nationalism 24
of nationhood emerged in the course of political and military strug- 25
gles against Denmark in the late 15th and 16th centuries, before a 26
distinctively Swedish culture existed. Literature, art, and language 27
were then permeated by Danish and German influence. Nor were 28
there sharp ethnic distinctions between Swedes and Danes. In these 29
circumstances, national feeling was expressed in an attachment to 30
political and institutional traditions, not in the sense of ethnic or cul- 31
tural distinctiveness. Later, to be sure, national feeling did find ex- 32
pression in a distinctive culture. And contemporary Sweden certainly 33
has a relatively homogeneous national culture. But this national cul- 34
ture has never carried a strong political charge in the Swedish tradi- 35
tion. It was not harnessed to a project of domestic assimilation and 36
overseas imperialism, as in France, nor to a movement for national 37
unification, as in Germany, nor to a campaign for national autonomy 38
or independence, as occurred in 19th century Finland and Norway, 39
neither at that time a sovereign state. Sweden’s long, continuous his- 40
tory as an independent state with a more or less homogeneous popu- 41
lation, and its position as the dominant Scandinavian power from 42
the 17th century on, provided no occasion for the politicization of 43
cultural identity.

1 The absence of a tradition of ethnic or cultural nationalism may

2 help explain why Sweden has been able to make citizens of its post-
3 war immigrants with so little fuss or friction. A further reason is to be
4 found in the composition of the immigrant population, which, until
5 recently, was two-thirds Nordic and overwhelmingly European. The
6 ethnic diversity of the immigrant population has increased markedly
7 in the last decade, as large numbers of refugees from Chile, Turkey,
8 Vietnam, Iran, and Iraq have been granted immigrant status. And
9 a small fundamentalist opposition has recently made some gains.
10 But while this may encourage centrist politicians to adopt a more
11 restrictive policy on refugee admissions, it seems unlikely to affect
12 Sweden’s liberal policy on admission to citizenship.
13 England: fuzzy Early political unification led to the early development of national
14 conception of feeling in England. Yet neither England nor Britain ever became a
15 citizenship nation-state on the French model – a tightly integrated political and
16 cultural community. English rule over Scotland, Wales, and espe-
17 cially Ireland gave the state a composite character, and nationhood
18 an ambiguous character. British national feeling developed, but it did
19 not supersede English, Scottish, Welsh, or Irish national feeling.
20 Just as there has been no clear conception of British nationhood,
21 so too there has been no clear conception of citizenship. The concept
22 of citizenship as membership of a legal and political community was
23 foreign to British thinking. Legal and political status were conceived
24 instead in terms of allegiance – in terms of the vertical ties between
25 individual subjects and the king. These ties of allegiance knit togeth-
26 er the British empire, not the British nation. Until 1948, all persons
27 born within the dominions of the king were British subjects. There
28 was no specific citizenship status for the colonies, for Britain itself,
29 or even for the independent Commonwealth countries.
30 With the dismantling of its empire, Britain has had to redefine
31 itself as a nation-state, and to create for the first time a national citi-
32 zenship. The transition has been an awkward one. France too had
33 to negotiate the dismantling of a huge colonial empire. And, unlike
34 Britain, it became involved in a bloody, bitter, protracted war. But
35 at least France already had a strong identity as a nation-state and a
36 well established national citizenship. Britain had neither, and this
37 contributed to the confused and bitter politics of immigration and
38 citizenship during the last quarter-century.
39 Lacking a national citizenship until 1981, Britain lacked a clear
40 criterion for deciding whom to admit to its territory. In the early post-
41 war years, inspired by a heady vision of itself as the center of a vast
42 multiracial Commonwealth of Nations, it continued the traditional
43 practice of admitting all British subjects – a category now includ-

ing citizens of the independent Commonwealth countries. But con- 1

trols were imposed on this latter group in 1962 after a significant 2
immigration developed from Jamaica, India, and Pakistan. This was 3
inevitable, in view of the huge population disparity between the inde- 4
pendent Commonwealth countries and Britain itself. More troubling 5
was the fact that the government later drew distinctions in immigra- 6
tion law between persons possessing the same formal citizenship 7
status – citizenship of the United Kingdom and Colonies. While 8
other countries were debating the citizenship status of immigrants, 9
Britain was debating the immigration status of citizens.8 10
Britain now has a national, postcolonial citizenship, and with it 11
a clear criterion of admission to the territory. But it achieved this, 12
in the eyes of some critics, only by drawing the lines of the national 13
community of citizens too narrowly, and by creating a special sec- 14
ond-class citizenship status, without the right of immigration, for 15
residents of Hong Kong and others. 16
In the domain of economic, social, and political rights, immi- 17
grants in Britain generally have more rights than elsewhere. This 18
too results from the fact that Britain has not traditionally defined 19
itself as a nation-state. British law imposes relatively few disabilities 20
on aliens; more important, relatively few of Britain’s postwar immi- 21
grants have been aliens. Neither Irish citizens nor citizens of inde- 22
pendent Commonwealth countries are considered aliens. Outside the 23
domain of immigration law itself, immigrants from the Caribbean, 24
from India, from Pakistan, and elsewhere have virtually the same 25
rights as British citizens, including the right to vote and to run for 26
office. 27
American and Canadian conceptions of citizenship and nation- Europe vs. America 28
hood reflect the historical and contemporary importance of immigra- 29
tion. This distinguishes them sharply from their European counter- 30
parts. Even before American independence, the pressing need for 31
settlers had established naturalization as central to the theory and 32
practice of citizenship. Characteristics of naturalization – a process 33
through which an individual expresses his or her voluntary adhesion 34
to a state – came to be ascribed to American citizenship as such. The 35
War of Independence reinforced this understanding of citizenship, 36
for it led to sharp criticism of the British conception of unchosen and 37
perpetual subjectship.9 And since the new nation lacked a distinctive 38
ethnic or cultural identity, American nationhood and nationalism 39
had to be defined in terms of a universalistic political formula that 40
would set it apart from the mother country.10 41

The Civil Rights Act of 1866 and the 14th Amendment of 1868 42
definitively established birth in the territory (jus soli) as the criterion 43

1 for the attribution of citizenship and affirmed, in principle, the pri-

2 macy of national over state citizenship. In the aftermath of the Civil
3 War, the affirmation of jus soli and of national citizenship had an
4 explicitly egalitarian, inclusive meaning.
5 The traditional inclusive and universalistic self-understanding
6 of the United States has always stood in tension with a much less
7 pretty practice. Free blacks, as well as slaves, were excluded from
8 U.S. citizenship before the Civil War, even when they possessed state
9 citizenship. Blacks continued to be excluded from full citizenship
10 after the Civil War through a restrictive judicial reading of the 14th
11 Amendment. American Indians were not granted automatic citizen-
12 ship at birth until 1924. And the category of “aliens ineligible for
13 citizenship,” first introduced to exclude Chinese in 1882, was not
14 finally abolished until 1952. Ethnic exclusion based on national-or-
15 igin immigration quotas, moreover, persisted until 1965. Still, the
16 voluntaristic and universalistic understanding of citizenship helped
17 eventually to undermine the legitimacy of these exclusionary prac-
18 tices. High rates of immigration, liberal naturalization provisions,
19 and the jus soli rule have made the United States, for most of its
20 history, exceptionally open to the political incorporation of ethnically
21 and culturally distinct immigrants.
22 phases of This tradition of inclusion has been interrupted by periodic phas-
23 exclusiveness es of exclusiveness. One such phase, marked by the surge of the
24 KnowNothings in the 1850s, occurred in response to the dramatic
25 increase in Catholic immigration after 1830; another, culminating in
26 the severely restrictive legislation of 1917-1924, occurred in response
27 to the “new immigration” from southern and eastern Europe after
28 1890. Today, after twenty years of the “new ‘new immigration’” ush-
29 ered in by the liberal Hart-Celler Act of 1965 and twenty years of
30 high levels of illegal immigration, we may be entering another such
31 phase. Even in the present political climate, however, debates about
32 immigration and citizenship continue to be informed by the distinct-
33 ly inclusive American understanding of nationhood. Thus the legal-
34 ization program of the Immigration Reform and Control Act of 1986
35 acknowledged the legitimate membership claims of long-settled un-
36 documented immigrants and of seasonal agricultural workers. And
37 it has been taken for granted that legalized immigrants would be-
38 come citizens. Newspaper reports on the legalization program some-
39 times described undocumented aliens as applying for citizenship, al-
40 though in fact they were applying for temporary resident status and,
41 if successful, would qualify for permanent resident status only after
42 18 months, and for citizenship only after another five years.11
43
Canada, in some respects, has been even more strongly marked

by immigration than the United States. Immigration has amounted 1

to as much as five percent of the total population in a single year 2
(1913), more than three times the highest percentage ever recorded 3
in the United States. And the foreign-born are currently twice as nu- 4
merous, in relation to population, in Canada. 5
Immigration policy has followed similar rhythms in the two coun- 6
tries. Canada, too, excluded the Chinese in the late 19th century, re- 7
stricted entry after World War I, abolished discrimination by national 8
origin in the 1960s, and has since admitted immigrants of steadily 9
increasing ethnic diversity. Rapid naturalization has long been pro- 10
moted in Canada, perhaps somewhat more vigorously than in the 11
United States. 12
Yet the centuries-old French-English dualism has complicated 13
the relation between immigration, citizenship, and nationhood in 14
Canada. The tensions that peaked in the late 1970s have abated, but 15
Canadian nationhood remains ambiguous and problematic. The 16
most basic question – is Canada one nation, or two ? – remains con- 17
troversial.12 18
Immigration has been related in complex ways to this dualism. 19
Historically, dualism has not meant pluralism. Immigrants have 20
been expected to assimilate to the French- or the English-speaking 21
community. The large majority, even those settling in Quebec, have 22
done the latter – a fact that sparked French resentment of immigra- 23
tion as an instrument of English domination. On the other hand, 24
dualism may have engendered in recent years a greater sensitivity to 25
the cultural identity of immigrants. A few years after becoming bilin- 26
gual on the federal level, Canada adopted an official policy in support 27
of multiculturalism. It is not clear what this means in practice. But 28
it may encourage Canada’s increasingly diverse immigrants to natu- 29
ralize quickly, without feeling that they must thereby abandon their 30
cultural identity. 31
32
33
Questions of membership13 34
35
The nation-state is doubly bounded. It has a bounded territory and 36
a bounded membership. States make decisions about whom to ad- 37
mit to their territories, and about whom to admit as members. This 38
book is not concerned with admission to the territory. Not that this is 39
unimportant. Quite the contrary: the intensifying demand for entry 40
raises urgent and troubling questions about territorial boundaries. territorial 41
Most fundamentally: what right do states have forcibly to deny entry boundaries 42
into their territories – particularly to persons in urgent need of food, 43
shelter, or protection?14

1 Questions of membership, though, differ from questions of en-

2 try. Questions of membership concern persons already present in
3 the territory (although not all such persons: the vast majority of those
4 admitted to the territory of another state are short-term visitors for
5 business or pleasure; their membership status is not in question).
6 Problems of membership arise, rather, for persons whose residence
7 and participation in the economic and social life of a country have
8 engendered significant ties to that country.15
9
It is of course impossible to delimit this group with any preci-
10 sion. Ties develop gradually, and there is no sharp divide between
11 shortterm visitors whose attachments remain firmly anchored in
12 their country of origin and persons whose developing attachments
13 to a new country begin to raise questions of membership – personal
14 questions in the mind of the migrant, and policy questions for the
15 country in which he or she resides. It is just for this reason that the
16 personal questions and the policy questions are such difficult ones.
17 The policy questions are of two sorts. First, under what condi-
18 tions and on what terms should such persons be admitted to full
19 citizenship? Second, what is the appropriate status for persons who
20 are not, or not yet, full citizens? What civil, political, economic, and
21 social rights should they enjoy? To what obligations should they be
22 subjected?
23
24 Access to citizenship
25 Citizenship is at the vital center of the political life of the modern
26 nation-state. Whom should the state admit to the privileges of citi-
27 zenship, and on what terms and conditions?
28 The individual essays have much to say about this question.
29 Without rehearsing their arguments here, let me simply note that
30 the essays of Part One make arguments about admission to citizen-
31 ship on two levels, linking political philosophy and public policy.
32 They raise broad questions of political philosophy, but these ques-
33 tions have definite – and sometimes quite far-reaching – policy im-
34 plications.
35 Central to the essays of Joseph Carens and Kay Hailbronner, for
36 example, is a perennial conundrum of political philosophy: how
37 weighing interests should one weigh the claims and interests of individuals against the
38 claims and interests of the state? Professor Carens articulates and as-
39 serts the claims of individuals, Professor Hailbronner the claims of
40 the state. These arguments have diametrically opposed implications.
41 Carens would compel the state to grant citizenship to all persons re-
42 questing it, providing they meet minimum residence requirements.
43 Hailbronner defends the state’s discretionary power to grant or deny
naturalization in accordance with its own interests.

Tomas Hammar, too, considers the interests of individuals 1

and the interests of the state in his discussion of dual citizenship. 2
Traditional antipathy to dual citizenship, he suggests, results from 3
the tendency to look at the matter primarily from the point of view of 4
the state. From the point of view of the individual, which Hammar 5
thinks ought to be given much more weight, the inconveniences of 6
dual citizenship are minimal, the advantages considerable. 7
These essays make arguments about how the state should regulate 8
access to citizenship. My own essay on citizenship law and natural- 9
ization practice looks at the way states do regulate access to citizen- 10
ship. I consider in detail the choices open to policymakers. And I 11
discuss the reasons that have led some countries to base citizenship 12
on birthplace, others on parentage, some to adopt liberal, others re- 13
strictive naturalization policies. There is thus no need for further dis- 14
cussion here of the problem of admission to citizenship. 15
16
The membership status of noncitizens 17
Citizenship is a neat category. It is simple and straightforward from 18
the point of view of the individual and from the point of view of the 19
state. One either is or is not a citizen of a particular state. There is no 20
middle way, no more or less, no ambiguity – except, of course, when 21
one is a citizen of two or more states. 22
Membership, in contrast, is a messy category. It is complex and ambiguity of 23
ambiguous from the point of view of the individual and from the membership 24
point of view of the state. Unlike citizenship, membership is not an 25
all-or nothing, yes-or-no variable. The world cannot be neatly divided 26
into those who are and those who are not members of a particular 27
state. One can be more or less a member; one can be a member in 28
one respect but not in another. 29
One of the major themes of this volume – developed in different 30
ways by Carens, Schuck, and Hammar, and in my own concluding 31
essay – is that membership is a broader and more inclusive category 32
than formal citizenship. In each of our six countries, there is a large 33
and growing group of noncitizen members. What sort of member- 34
ship status should these resident noncitizens enjoy? 35
There are two ways of approaching this tangled and complex ques- 36
tion. One can focus on different types of membership. This approach 37
asks what distinctions should be drawn between citizens and nonciti- 38
zens, and between different categories of noncitizens. Alternatively, 39
one can focus on different types of membership goods. One would 40
then ask what sorts of goods should be reserved for citizens, and 41
what sorts of goods should be made available to noncitizens as well. 42
Consider each approach in turn. 43

1 Types of membership. The ideal of equality – more precisely,

2 formal equality of status – is deeply rooted in the Western political
3 tradition. With philosophical sources in Stoicism and Christianity,
4 this ideal was elaborated by liberal political philosophers, propagated
5 by the French Revolution, and gradually realized in practice over the
6 19th and 20th centuries. I noted above the central place of this ideal
7 in our inherited understanding of nation-state membership; Peter
8 Schuck discusses its importance in the American political tradition.
9 partial membership Given the strength of this egalitarian ideal, partial membership is
10 always in need of special justification. It is always vulnerable to con-
11 demnation as second-class citizenship. To our modern egalitarian
12 sensibility, partial membership is legitimate only if it is temporary.
13 Partial membership may be a way station on the road to full member-
14 ship; or it may accommodate temporary participants in our society
15 who remain full members of another. Even ardent egalitarians would
16 be willing to accept some kind of transitional status for permanent
17 immigrants and some kind of temporary status for resident sojourn-
18 ers – persons whose attachments remain anchored elsewhere but
19 whose residence and participation in the society distinguish them
20 from short-term visitors such as tourists and business travellers.
21 If the principle of transitional or temporary partial membership
22 is acceptable, why is the practice so problematic? The reason, I think,
23 is that the social realities of partial membership do not correspond
24 to the models just sketched. Millions of people in Europe and North
25 America have been partial members for a decade or more. They are
26 not – or not any more – the sort of temporary participants for whom
27 partial membership is appropriate. And if they are on the road to full
28 membership, the road is a long one indeed. By their own accounts,
29 though, many do not seem to be on the road to citizenship at all.
30 They seem likely to remain partial members for the indefinite future.
31 There are strong arguments, informed by the principle of equal-
32 ity, for extending to these long-term residents the rights enjoyed by
33 full citizens. Yet as Peter Schuck points out, to carry this process of
34 inclusion to its logical – or illogical – conclusion would erase the dis-
35 tinction between citizens and resident aliens and deprive the status
36 of citizenship of any distinctive value or meaning. Given the impor-
37 tance of citizenship in the theory and practice of democratic nation-
38 states, this would be deeply problematic. Indeed, fundamentalists
39 argue that the process of inclusion has already been carried too far;
40 they propose to restore value and meaning to citizenship by reserv-
41 ing a wider range of rights for citizens.
42 The ideal of equality and the ideal of citizenship are both deeply
43 ingrained in the political culture of Western nation-states. The two

ideals need not clash. Indeed citizenship is an inherently egalitarian 1

ideal. It implies full legal and political equality among citizens. Yet 2
the equality inherent in the idea of citizenship is a bounded equality. 3
It is necessarily restricted to citizens. Full equality between citizens 4
and noncitizens would render citizenship meaningless. For this rea- 5
son, the ideal of citizenship may clash with the principle of equality. 6
This makes the question of the extent to which long-term resident 7
noncitizens ought to share in the rights of citizenship a difficult and 8
deeply contested one. The tendency seems to be to extend many, 9
even most of the rights of full membership to long-term resident 10
aliens, while reserving certain core political rights and functions to 11
citizens. 12
Another response to long-term partial membership is to encour- 13
age naturalization. This sounds innocuous enough, and it would naturalization 14
seem to be less controversial than extending citizenship rights to 15
noncitizens. But in practice it too is controversial. For one can pro- 16
mote the passage to full citizenship with a carrot or with a stick. One 17
can liberalize access to citizenship, or one can make partial member- 18
ship less attractive. The latter can be done by limiting the rights of 19
partial members or by imposing new obligations on them (e.g., mili- 20
tary service). At the limit, it can be done by requiring partial mem- 21
bers to apply for naturalization or leave the country.16 22

Partial membership for immigrants, then, too often becomes a 23
final station rather than a way station on the road to full citizenship. 24
Partial membership for short-term sojourners poses a different set of 25
problems. Should sojourners have the chance to become settlers? If 26
so, which sojourners, and under what conditions? What provisions 27
should be made for the passage from temporary to permanent mem- 28
bership? 29
These questions are difficult partly because the category of short- 30
term sojourners is so heterogeneous. It includes all short- to medi- 31
umterm residents whose attachments and interests remain centered 32
in their country of origin, but who are in the process of creating a 33
new set of attachments and interests. 34
One large group includes those who are resident in order to re- 35
ceive some kind of education or training. Even this category is quite 36
heterogeneous, with the education varying from the general to the 37
highly technical and the length of residence from a couple of months 38
to several years. Persons resident for work or business represent an 39
equally heterogeneous category, ranging from unskilled laborers to 40
the international professional and corporate elite; for this group, too, 41
stays may be measured in months or in years. 42
Can the state insist on a sharp distinction between immigrants 43

1 immigrants and and sojourners, keeping the latter in a strictly temporary status? Or
2 sojourners must it grant them the opportunity to become permanent members?
3 The question is by no means academic. Each of our six countries, wary
4 of increasing “backdoor” immigration on the part of persons admitted
5 for temporary stays, has taken steps in recent years to restrict passage
6 from temporary to permanent status. When directed against tourists
7 or persons on short-term business visits, such measures seem un-
8 objectionable. But when directed against students or workers whose
9 stays may span several years, they raise difficult questions.
10 These questions arise even when persons are admitted on the ex-
11 plicit understanding that they will eventually have to leave. When the
12 U.S. Immigration and Naturalization Service recently announced
13 that tens of thousands of nurses admitted on non-immigrant H-1
14 visas would not be able to extend or renew their visas after six years,
15 it was only confirming the explicit terms on which the visas had been
16 issued. 17 Yet the decision does seem troubling. The state was under
17 no obligation, legal or moral, to admit the nurses in the first place.
18 But having permitted them to work and live and form ties for six
19 years, it may have acquired a moral obligation to let them remain.
20 The debate about seasonal worker programs pivots on similar
21 questions. Seasonal workers permit states to meet certain manpower
22 needs cheaply while externalizing various costs, including the cost of
23 unemployment. Although the limitation of work and residence to
24 a certain number of months per year is intended to hinder the for-
25 mation of social ties and thus to prevent settlement, many seasonal
26 workers – particularly those hired year after year – develop signifi-
27 cant attachments to the country in which they work. It seems only
28 fair that they be given the chance to graduate to permanent status.
29 What about students? Most countries discourage the settlement
30 and naturalization of foreign students. One important rationale –
31 since many of the students are from developing countries – is that
32 this policy will hinder the brain drain from the third world to the
33 first. This is surely a legitimate consideration, but what exactly justi-
34 fies the differential treatment of workers and students? One could
35 argue that, for equal periods of residence, work in a country creates a
36 stronger claim to membership than study. Work – so the argument
37 might run – makes a direct contribution to the wealth and welfare
38 of a country, while study primarily prepares an individual for his or
39 her own projects. But would this apply to all types of work? Does it
40 apply equally to the executive of a multinational corporation and to
41 the unskilled laborer? Or is there a sense in which the latter has spe-
42 cial membership claims, perhaps because his or her presence in the
43 territory is the result of what some analysts characterize as an “un-

equal exchange”? Certainly persons actively recruited by employers 1

or the state to perform work shunned by citizens – as is the case for 2
the nurses mentioned above – would seem to have especially strong 3
membership claims. 4
Perhaps the only point on which wide agreement might be se- 5
cured is one developed in this volume by Joseph Carens. Professor 6
Carens argues that the claim to citizenship varies directly with the 7
strength of social ties and thus, normally, with length of residence. 8
One implication of this view is that whatever right the state might 9
have to limit noncitizens’ stays must be exercised sooner rather than 10
later. It is not a right that can be reserved for eventual use whenever 11
this might seem opportune. Failure to exercise it within a reasonable 12
period leads to its expiration. State acquiescence in continued resi- 13
dence eventually creates an individual right to remain. This, by the 14
way, is no mere philosopher’s argument; the principle has been ac- 15
knowledged by courts, among them the highest administrative court 16
in West Germany. 17
Special problems of partial membership are raised by persons 18
residing and working in the territory without the permission of the 19
state. This question has dominated the politics of immigration and 20
citizenship in the United States, and it has been important in France 21
as well. To what extent should such persons be included in the bene- 22
fits of membership? In the United States, this is in part a constitution- 23
al question, resting on the interpretation of the 14th Amendment’s 24
equal protection clause. It is on the basis of this clause, for example, 25
that the Supreme Court ruled that undocumented immigrant chil- 26
dren could not be excluded from the public schools. But it is more 27
profoundly a political question. To what extent do their economic 28
contribution, their de facto integration, and what Senator Simpson 29
has called the “statutory encouragement to migrate illegally” (i.e., the 30
absence of penalties on employers) give undocumented immigrants 31
a claim to some form of membership? Most would probably agree 32
that the prolonged government acquiescence in massive employ- 33
ment of undocumented immigrants gives these immigrants a stron- 34
ger membership claim than those who entered the country after the 35
imposition of employer sanctions (assuming that these are actually 36
enforced). 37
The goods of membership. An important aspect of citizenship 38
(and other forms of state-membership) is the access it provides, di- 39
rectly or indirectly, to a wide range of goods. These include such basic access to goods 40
goods as public order, physical safety, and access to a labor market; 41
the complex array of civil, political, social, and economic rights; and 42
even intangibles such as a feeling of belonging or collective identity. 43

1 The enjoyment of some of these goods depends directly on mem-

2 bership status – on citizenship, permanent residence, or some other
3 status. Other goods, though, do not depend directly on membership
4 status, being available to all persons who happen to be present in the
5 territory. The public peace, for example, may be enjoyed by those il-
6 legally or temporarily in the territory as well as by members. Yet even
7 this good depends indirectly on membership, for only some form
8 of membership can secure long-term residence and thus long-term
9 enjoyment of the good.
10 From a global perspective, the most important basic goods today
11 are public peace and access to a relatively promising labor market
12 (one affording a reasonable chance of realizing personal or familial
13 aspirations). Both goods depend at least indirectly on membership,
14 and both goods are distributed among states in a highly unequal
15 manner. It is this that accounts for the unprecedented migratory
16 pressure and for the increasing salience and urgency of the politics
17 of immigration and citizenship today.
18 What is it about the various membership-dependent goods that
19 makes it reasonable to set different conditions of eligibility for them?
20 What goods ought to be reserved for full citizens, and why? At the
21 other end of the spectrum, what goods should be extended to all per-
22 sons in the territory, regardless of membership?
23 agreement on the Michael Walzer has suggested that shared understandings about
24 meaning of goods the meaning of goods should guide policy deliberations about their
25 distribution.18 The principle can be applied to the goods of member-
26 ship. It is the different moral and political meanings of these goods,
27 I think, that may explain why some are reserved for citizens, others
28 extended to permanent residents, and others available to all without
29 regard for membership.
30 To agree on this principle is simply to agree on a mode of argu-
31 ment. It does not, of course, settle any substantive questions of eligi-
32 bility. Disagreement about the meaning of particular goods or about
33 the implications of this meaning for eligibility is not only possible,
34 it is inevitable. The following remarks are merely illustrative; I make
35 no attempt to establish the meanings of different sorts of member-
36 ship goods.
37 Consider voting. Even those who wish to extend to noncitizens
38 most rights of citizenship often concede that there is something spe-
39 cial about voting in national elections. The fact that national elections
40 influence policy in the domains of defense and foreign affairs may
41 justify reserving the right to vote in such elections to citizens, bound
42 to the state by ties of allegiance and obligations of service. Voting
43 in local elections, however, has a different meaning. It involves lo-

cal self-administration, not high politics on the international scene. 1

Questions of ultimate allegiance, it maybe argued, are simply irrel- 2
evant to local voting. Thus voting rights in local elections have been 3
granted to resident noncitizens in a number of European countries. 4
Or consider social benefits. Some derive their meaning and jus- 5
tification in reference to work: they are intended to replace lost in- 6
come when a person is unable to work because of injury, involuntary 7
unemployment, or old age. Such benefits are financed through em- 8
ployer and employee contributions. Worker’s compensation, unem- 9
ployment insurance, and social security are examples. Other social 10
benefits have a different meaning. They are justified with reference 11
to membership and financed out of general revenues. Family allow- 12
ances, housing assistance, and income-supplement programs in 13
general are examples. A third type of benefit is justified with respect 14
to urgent need: this includes emergency medical care and emergency 15
assistance generally. 16
The meanings of these goods have implications for eligibility. eligibility 17
Most people would probably agree that anyone granted access to the 18
labor market, whatever his or her membership status, should qualify 19
for such directly work-dependent benefits as worker’s compensa- 20
tion, unemployment insurance, and social security. Membership- 21
independent eligibility for family allowances or housing assistance, 22
however, is more controversial. This is because these latter could be 23
understood as a form of mutual aid provided by members of a pol- 24
ity for one another. (“Members” might be interpreted restrictively to 25
mean citizens only, or it might include permanent resident aliens as 26
well.) The meaning of emergency assistance, finally, requires that it 27
be extended to all persons in need, whatever their membership sta- 28
tus. This includes illegal immigrants. 29
30
For references please consult the bibliography of the book in which 31
this article was originally published. (see List of sources, page 609) 32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Ethnic leadership, ethnic communities’ political 3
4
powerlessness and the state in Belgium
5
6
Marco Martiniello 7
8
9
10
11
12
Based on political sociologist Marco Martiniello’s doctoral study, which won
13
the award for best thesis at the European University Institute in Florence in
14
1993, this article was first published by the journal Ethnic and Racial Studies in 15
the same year. Presented here are the results of one of Continental Europe’s 16
first studies on the links between ethnic leadership formation, the role of 17
the state and the reproduction of political powerlessness among immigrant 18
ethnic communities. Martiniello articulates a theoretical approach inspired 19
by Marxian insights combined with Stephen Lukes’ theory of power and 20
American theories of ethnic leadership, on the one hand, and a qualitative 21
empirical research, on the other. The article sparked renewed interest in the 22
issues of political inclusion, inclusion of immigrants and their descendents 23
in Western Europe. It also showed how American theoretical categories 24
need to be reconstructed to fit the European migration and post-migration 25
context. 26
27
28
Introduction 29
30
Although to state that no human society is homogeneous appears 31
to be a banal remark, this simple observation constitutes the very division principles 32
basis of diverging sociological approaches. In our post-industrial 33
Western societies there are various principles of division, the rela- 34
tive importance of which social scientists have long been discussing 35
and arguing about. For many Marxian scholars class constitutes the 36
foremost criterion for the breakdown of our societies and, in their 37
view, should therefore be the basic unit of social and political analy- 38
sis. For other social scientists gender division seems to be a more sig- 39
nificant dimension of differentiation in human societies and gender 40
for them, should therefore be the main unit of research. Yet a third 41
group stresses the predominance of divisions along racial and ethnic 42
lines which they consider to be the chief organizational principles in 43

238 marco martiniello
1 our societies. Race and ethnicity then become central to their work.
2 Social scientists who definitely and exclusively choose one rather
3 than another of these division principles grow fewer and fewer in
4 number. There is currently some kind of recognition that a better
5 understanding of our societies stems from a masterly combination
6 of all those dimensions in sociological analysis. It is more and more
7 accepted that class, gender, race and ethnicity, seen as research units,
8 are not necessarily mutually exclusive. In this view, either none of
9 those principles is crucial or they all are. In other words, indepen-
10 dently of how they are conceptualized, class, gender, race and, eth-
11 nicity appear to be interrelated and in some cases even to overlap.
12 Furthermore, the nature of such interrelations and overlaps is nei-
13 ther definite nor fixed in time. One of the issues facing social re-
14 search is precisely to try to understand and explain those historical
15 changes.
16 The interrelations and overlaps concern class, gender, race and
17 ethnicity, whether they be considered as analytically distinct research
18 units or as a basis for individual identity formation, or, as mobiliz-
19 ing principles for collective action. On the one hand, more and more
20 scholars seek to discover the connections between race and class
21 (Anthias 1990), between class and gender, and between class, gender
22 and race. On the other hand, individuals seldom define themselves
23 simply as ‘black’, or ‘female’ or ‘Moroccan’. Usually a person’s iden-
24 tity is a combination of several of those dimensions, which prefer to
25 as many identification processes. At the level of collective action, the
26 same phenomenon may be empirically observed. Frequently, several
27 of the four dimensions presented are used simultaneously as orga-
28 nizing principles.
29 The aim of the present article is not to tackle the issue of the inter-
30 connection between class, gender, race and ethnicity in a straightfor-
31 ward manner. However, in looking at our post-industrial post-World
32 War II societies – especially Belgium – as massive international labour
33 and political immigration countries, it will be dealt with indirectly. The
34 arrival and settlement of immigrants have had significant and com-
35 plex effects on the class, gender, racial and ethnic composition of
36 Belgium, as well as on the emergence of new forms of identity and
37 collective action.
38 labour and political In order to avoid a sterile and endless theoretical discussion about
39 immigration the interconnection between class, gender, race and ethnicity, it is
40 useful to introduce the concepts of ‘labour’ and ‘political immigra-
41 tion’ as an alternative division principle in our societies. From this
42 standpoint, Belgium can be characterized by the presence of two
43 types of human groups: the ‘native’ population, and the population

resulting from post-World War II mass immigration. Both are cul- 1

turally, socially and politically heterogeneous and socio-economically 2
stratified, even though the socio-economic: stratification follows dif- 3
ferent patterns in each case. 4
Situated in that broad context, this article deals mainly with the re- 5
lations between the Belgian state and political system on the one hand, 6
and the ethnic communities from immigrant origin on the other. In 7
Brass’s (1985) terms, this is a study of the relations between ‘ethnic 8
groups’ and the state in which only specific ethnic groups are con- the state 9
cerned. However, the use of the concept ‘ethnic community from im- 10
migrant origin’ does not imply that central importance is given to ‘sub- 11
stantial’ (i.e., which had a substance, a content) ethnicity. As is shown 12
in the following section, the proposed definition of ethnic groups is 13
largely non-ethnic in the primordialist sense of the expression. 14
In order to clarify the concepts and to avoid confusion about 15
terms, the next section specifies the main units of analysis used in 16
this article. The central hypothesis will be dealt with in the second 17
section. The third and final section presents the main results of an 18
empirical case-study in Belgium and an evaluation of the previously 19
stated hypothesis as well as two sets of conclusive remarks. 20
21
22
The units of analysis 23
24
As Brass (1985) correctly observed, most studies of the relations be- 25
tween ethnic groups and the state present two limitations. Firstly, 26
they show a tendency to ‘reify’ ethnic groups – to attribute to mere 27
categories a reality that they may not necessarily have. Secondly, one 28
can observe a certain ‘objectification’ of the units of analysis, where- 29
by one or other dimension is considered to be of greater importance 30
than any of the others, the latter being seen as secondary. 31
One way to avoid such problems is to draw a clear-cut distinction 32
between objective and subjective social entities. In this article, two 33
types of ethnic collectivities from immigrant origin are distinguished, 34
namely ethnic categories1 and ethnic communities. This conceptualiza- 35
tion is analogous to the Marxist distinction between class-in-itself 36
and class-for-itself (Marx 1956). An ethnic category is a collection of 37
individuals who share a set of common objective features and who ethnic categories 38
live generally in an analogous situation characterized by a set of col- 39
lective disadvantages that define a status of minority. The first objec- 40
tive feature is national origin (Schermerhorn 1974). It is somehow 41
linked to cultural features such as language, dietary habits and reli- 42
gion but these elements are not constitutive of the definition. Here, 43

1 ‘ethnicity’ refers only to national origin (Alba 1976; 1895) that stems
2 mainly from a juridical classification2 of human beings.
3 Paradoxically, the definition of ethnic collectivities is largely non-
4 ethnic because it is not based at all on ‘cultural’ or ‘ethnic’ (in the
5 primordialist sense) elements beyond national origin.3 The second
6 objective feature is the migratory origin that people have in common.
7 The reason why people are classified in an ethnic category can be
8 traced back to one and the same phenomenon, namely post-World
9 War II international labour and political migration. At the origin, an
10 ethnic category is made up or can be constituted by migrant work-
11 ers or political refugees, that is, by people who came from abroad.
12 However, their family and children, often born in their parents’ ar-
13 rival country, are also part of the ethnic category, even though they
14 have no personal experience of migration and cannot therefore be
15 considered as immigrants. Consequently, and this is the third fea-
16 disadvantages ture, ethnic categories are reproduced over several biological genera-
17 tions (Keyes 1976). Their lifetime is thus of long duration.
18 The collective disadvantages that people classified into ethnic cat-
19 egories face can be observed in many spheres of human life. As far
20 as the socio-economic sphere is concerned, they are usually concen-
21 trated at the level of manual unskilled or semi-skilled labour, often
22 in declining industries but also in other sectors, such as the services
23 sector. This relative homogeneity in the weakest positions on the
24 labour market is to a certain extent reproduced over the biological
25 generations.4 It is rooted in the history of post-World War II labour
26 migration which concerned mainly unqualified – or at least used as
27 such – manpower. In the legal-political sphere their position is also
28 weaker than that of the ‘natives’. As foreigners they are often de-
29 prived of basic political rights such as the right to vote and to be
30 elected. Even where they have obtained the relevant nationality, and
31 consequently those basic rights, they are often the targets of unequal
32 treatment, for example by the police. Their position in education and
33 housing is also disadvantaged in many ways. Furthermore, ethnic
34 categories are numerically small compared to the population of the
35 society at large. Finally, there can be as many ethnic categories as there
36 are successive labour or political migratory waves in one country.
37 The notion of ethnic category as used here, is an abstract one. As
38 a research construction, it is based exclusively on objective criteria.
39 To be part of an ethnic category, it is not necessary to have a self-
40 consciousness or an identity. There is no membership, no belonging
41 to an ethnic category. People are assigned to an ethnic category by a
42 researcher on the basis of some objective features that they share and
43 some common disadvantages that they face.

Besides this scientific classification, ethnic categories are also 1

the object of symbolic social categorizations. This means that the 2
objective features and disadvantages mentioned above are socially 3
perceived as being valid classificatory items on which a set of more 4
or less negative images of the ethnic categories is built and repro- 5
duced until they become a set of prejudices. These prejudices then 6
become yet another disadvantage that the ethnic category faces. For 7
example, because many Turks live in an old and deteriorating part 8
of Brussels (objective disadvantage), they will be spoken of as ‘dirty 9
Turks’ (symbolic categorization) and the prejudice that all Turks are 10
dirty will take the upper hand. In other words, ethnic categories are 11
the products of both a sociological construction and a social construc- 12
tion. However, not all the groups constructed as ethnic categories in 13
the first way are constructed as such in the second way or vice versa. 14
Compared to the ethnic category, the ethnic community has two the ethnic 15
additional characteristics: identity or self-consciousness and organi- community 16
zation. Along with Barth (1969) and Weber (1971), it is held that the 17
basic constitutive element of an ethnic community is not a shared 18
culture but rather a feeling of ‘being a member’, a self-consciousness 19
of belonging. The emergence of this identity may be interpreted as 20
a response to the symbolic social categorization and the prejudice 21
mentioned above, though it is not the only possible interpretation of 22
the process of identity formation. Membership is crucial because it 23
creates the basis for the appearance and development of the organi- 24
zational dimension, which is the second basic characteristic of the 25
ethnic community. In other words, it is only when certain people, 26
who are ethnically categorized, develop a common subjective self- 27
consciousness about some of the objective features which they share 28
or which they are convinced they share, and about some interests 29
which they believe they have in common, that an organizational, in- 30
stitutional and relational web will emerge progressively in order to 31
promote and defend those interests. 32
As a result of the definitions presented above, the ethnic com- 33
munity will be much smaller quantitatively than the ethnic category. 34
All the members of the ethnic community are also part of the ethnic 35
category but the contrary is not only not necessarily true in theory – it 36
is never true empirically. 37
It is outside the scope of this article to enter into a detailed discus- state and polity 38
sion about the definitions of state and polity. It suffices here to state, 39
along with Brass (1985), that the state is looked at neither as a mere 40
arena for group conflicts nor as an instrument of domination in the 41
hands of one social class against the other. The state is conceived 42
as a relatively autonomous entity that tends to act independently in 43

1 what it presents as the nation’s interest by classifying and making

2 distinctions among the population and by distributing collective re-
3 sources in a differentiated way. In this view, the state acts primarily
4 to preserve its monopolistic position and the definition, imposition
5 and representation of the nation’s values, goals and interest, and also
6 to preserve its distributive function. In this sense, the state warrants
7 the conservation and the perpetuation of the existing social order.
8 Formally, it is a set of persisting institutions about the control of
9 which there is a constant conflict between individual and collective
10 actors. Nevertheless, there is a minimal consensus between the vari-
11 ous conflicting parties about the fundamental role of the state as de-
12 fined above.
13 This study goes beyond the concept of state by using the notion
14 of polity. By doing so, reference is made to the set of political institu-
15 tions, or, more precisely, to the set of collective political actors pres-
16 ent in society. This means that, besides the executive, legislative and
17 judiciary powers, the notion of polity includes all the other political
18 actors and institutions who, in one way or another, at least theoreti-
19 cally in a modern democracy, take part in the definition and the man-
20 agement of society’s collective affairs. For instance, political parties,
21 unions and lobbies of every kind are all part of the polity.
22 It is important to underline that in this study about the relation-
23 ship between the state and the polity on the one hand, and the ethnic
24 communities on the other, no postulate is made beforehand about
25 the position of the ethnic communities with regard to the state and
26 the polity. The object of investigation does not imply a prejudgement
27 as to whether or not ethnic communities are included in the state
28 and the polity. Rather, their inclusion or exclusion is precisely a cru-
29 cial issue to be examined thoroughly: how to characterize the ethnic
30 communities’ position with regard to the state and the polity and
31 how to explain it?
32 ethnic leaders The central role given to the concepts of ethnic leaders and elites
33 and elites in this theoretical approach appears to be obvious for two reasons.
34 Firstly, it seems very difficult to analyse the relations between state
35 and polity on the one hand, and ethnic communities on the other,
36 without using the units of ethnic leadership and elites. This is be-
37 cause the state and the polity’s authorities cannot deal with abstrac-
38 tions, but have recourse to privileged actors or individuals, namely,
39 to ethnic leaders and ethnic elites. The theoretical importance of
40 those units is that they can be used to take into account internal con-
41 flicts within the ethnic communities, their external relations, as well
42 as several points of intersection between the two. Secondly, defining
43 ethnic categories in terms of collective features and disadvantages

does not imply that they are fully homogeneous and ‘undifferenti- 1
ated’. Furthermore, the concepts of ethnic elites and ethnic leaders are 2
very useful when it comes to grasping the social and economic dif- 3
ferentiation within ethnic collectivities. 4
The concept of ethnic elite refers to people from the ethnic catego- 5
ry who have reached a significant degree of success – as compared 6
to the average success level of their fellow ‘ethnics’ – in the larger 7
society involving one or more of the various fields of human activity 8
(work and profession, arts and culture, politics, business, etc.). Along 9
with Pareto (1986), one can talk of a plurality of elites, in this case 10
ethnic elites, each one of them corresponding to one specific field of 11
human activity. In any case, ethnic elites are a small but variable sub- 12
category of the ethnic category. Moreover, the dividing line between 13
elite and non-elite is defined in relative terms. It depends on each 14
category’s economic, social and political characteristics and history 15
in the immigration country. 16
The concept of ethnic leader refers to those members of the ethnic 17
community who have the ability to exert intentionally some variable 18
degree of influence on the preferences and/or behaviour of the other 19
members of that community, the aim being to obtain satisfaction of 20
the group’s objective interests as perceived by the leaders. When the 21
influence is exerted effectively, it is done through the leaders-follow- 22
ers’ interactions in the ethnic community’s institutions. Ethnic lead- 23
ers necessarily enjoy some degree of recognition by their followers in 24
the ethnic community on which the leadership’s legitimacy is based. 25
Finally, the approach taken in this article centres on the concepts 26
of power and powerlessness as a valid alternative to the dominant per- 27
spective, at least in continental Europe, that focuses on cultural or 28
ethnic relations using notably the concept of integration (Martiniello 29
1992). The definitions adopted here are largely inspired by the work 30
of Lukes (1974; 1986). Power is conceived as the ability of an ethnic 31
collectivity as a group to control results related to issues affecting its 32
interests. Consequently, an ethnic collectivity is politically powerless, 33
if it is unable to promote and defend its collective interests in the web 34
of political relations in a given society. 35
On the basis of the units of analysis defined above, the next sec- 36
tion develops the main theoretical hypothesis of this article before 37
turning to the case-study. 38
39
40
41
42
43

1 State and ethnic collectivities’ powerlessness in Belgium

2
3 As clarified above, the ethnic collectivities dealt with here are mainly
4 characterized by the objective disadvantages they collectively face in
5 many fields. Nevertheless, their disadvantaged position is not per-
6 fectly stable in time and space. Furthermore, it varies according to
7 the type of migratory experience of each receiving country and the
8 ethnic, collectivity in question. Consequently, not all ethnic collectivi-
9 ties face the same degree of objective disadvantage. For example, the
10 ethnic collectivities from member states of the European Community
11 [EC] are in many ways privileged compared to non-EC ethnic collec-
12 tivities. As workers, they are protected by European law and as ethnic
13 categories they are much less stigmatized than, for example, North
14 African collectivities.
15 However, a certain degree of disadvantage always persists and is
16 reproduced over the biological generations. This can be reduced or.
17 increased, but fundamentally, the ethnic collectivities studied live in
18 a chronically disadvantaged position in the receiving society. To the
19 extent that the continuation of this situation is contrary to the rank-
20 and-file ethnic’s objective interests, which are not efficiently promot-
21 ed and defended through an ethnic community collective action, eth-
22 nic collectivities are politically powerless and this powerlessness is
23 thus fundamentally persisting as well. This article does not concern
24 either the so-called ‘middleman minorities’ (Bonacich 1973) or the
25 powerful ethnic lobbies acting in particular political systems.
26 collective reactions From an analytical point of view, various forms of ethnic collec-
27 tivities’ reactions to their condition can be conceived. Firstly, they
28 can simply accept their position passively, either individually or
29 collectively, in which case no form of active response is elaborated.
30 Secondly, a fraction of the ethnic category can seek individual suc-
31 cess in areas that are left relatively open by the native society through
32 mobilizing personal resources in individual strategies. This process
33 of escape or exit gives birth to the ethnic elites. Thirdly, there is in
34 theory the possibility of elaborating active collective responses to po-
35 litical powerlessness.
36 two phases This process of active collective response could be seen analyti-
37 cally as a two-phased one. In the first phase, the ethnic collectivity
38 gives itself some kind of structure and constitutes itself in a single
39 collective actor or set of juxtaposed collective actors. A more or less
40 dense web of ethnic organizations and institutions takes shape in
41 which emerging ethnic leaders play a central initiating role, notably
42 through moulding and constructing some kind of mobilizable collec-
43 tive identity. This move towards ethnic-community building can be

interpreted as a first reduction in the degree of political powerless- 1

ness of the ethnic category. In the second phase, ethnic leaders will 2
try to promote and defend the ethnic community’s interests, notably 3
through their relations with the state and the polity. They will also 4
strive to keep the support of their fellow ethnics and to reaffirm con- 5
stantly the existence of the ethnic community. 6
The constitution and the structuration of a collective actor is a 7
crucial and much contested issue but one which is not dealt with 8
here. Rather, the problem is considered to have been solved by taking 9
the existence of ethnic communities as defined above as a premiss. 10
Seen in this context, the relations between the ethnic community 11
and the state and polity are the main focus of analysis. In this respect, 12
the study of the role of ethnic leaders in the reduction of the degree 13
of the ethnic community’s powerlessness through those relations is 14
thought to be of utmost importance. 15
The central hypothesis is this. In Belgium, ethnic leaders gen- maintaining 16
erally fail to reduce significantly their ethnic community’s political the status quo 17
powerlessness. They tend either to increase it or to maintain the 18
status quo. The fundamental reasons for that inability are not to be 19
found in the intrinsic characteristics of the ethnic leaders, such as 20
their political inexperience, say, or their incompetence. Rather, it has 21
to be explained by reasons that relate to the general political climate 22
in Belgium and its repercussions on the way in which the state and 23
polity tackle the relevant issues of the ethnic collectivities. Diffused 24
racism and xenophobia characterize the political climate in Belgium 25
and this explains the development of the state and polity exclusion 26
strategies directed towards ethnic collectivities. In other words, the 27
current political climate does not seem to be favourable to the em- 28
powerment of ethnic collectivities through their leaders’ actions, no 29
matter how competent the latter may be. 30
The general will to keep ethnic communities outside or at the 31
margin of the political system is translated into two related strategies 32
as far as ethnic leaders and elites are concerned: the neutralization 33
of ethnic leaders, and the ‘depoliticization’ of their action. The neu- 34
tralization of ethnic leaders is done in comparable ways by the state 35
and other political actors, mainly the political parties and the unions, 36
notably through the establishment of ad hoc peripheral institutions 37
or sub-institutions to deal with the problems of ethnic categories. 38
Furthermore, individual social and economic upward mobility often 39
accompanies the processes of neutralization and depoliticization of 40
ethnic leaders. In this second process, politicized ethnic leaders are 41
transformed into apolitical ethnic elites. 42
The neutralization and depolitization processes can be clarified 43

1 by adopting a historical perspective. In the beginning there is a group

2 of ethnic leaders whose inclusion in the polity follows a threefold
3 pattern. Firstly, they are often co-opted; that is, they are chosen ‘from
4 above’ by the ‘native’ political institutions. Secondly, they can be sup-
5 ported by a more or less massive mobilization of the ethnic com-
6 munity. Thirdly, they can be elected through some kind of formal
7 representation process. However, after intervention by the state and
8 other political actors, the previous group of ethnic leaders is divided
9 into two new groups. On the one hand, some of these leaders contin-
10 ue to act as defenders and promoters of the community’s interests,
11 but since they have been included in powerless ‘buffer’ institutions,
12 and despite the fact that they are still politicized, they are unable to
13 produce any significant effect on the community’s powerlessness.
14 In this sense, they have been neutralized. On the other hand, some
15 ex-leaders are no longer involved in ethnic-community politics. They
16 have been depoliticized and have simultaneously achieved individual
17 success, thereby becoming ethnic elites or confirming their precious
18 position ‘above’ the collectivity’s average.
19 ‘the role of ethnic Thus, the role of ethnic leaders is to be analysed in relation to the
20 leaders’ originally state and polity’s more or less imposed management of ethnic col-
21 published lectivities’ related issues. It could be claimed that the state and polity
22 usually tolerate only those leaders who do not oppose the dominant
23 view on ethnic collectivities’ issues, one important aspect of this be-
24 ing the generally shared willingness to keep immigrant communi-
25 ties outside or at the margin of the political system. Therefore, a cor-
26 ollary hypothesis would be that ethnic leaders could help to reduce
27 the ethnic-collectivity powerlessness only if the state and polity were
28 open enough to accept a real dialogue with them.
29 These hypotheses have been worked out in fieldwork research
30 concerning the collectivity of Italian origin in French-speaking
31 Belgium. The next section presents a summary of the results of that
32 empirical research, as well as a few conclusive remarks.
33
34
35 Italians in French-speaking Belgium: a powerless ‘model
36 of integration’
37
38 The history of the Italian presence in Belgium dates back to a remote
39 past. It is outside the scope of this article, however, to trace the his-
40 torical origins of the phenomenon. Rather, the focus is put on the
41 post-World War II period in which Italian labour immigration can be
42 subdivided into three consecutive phases, according to the type of re-
43 cruitment of the Italian workforce and its position in the productive

system. Between 1946 and 1957 Italian immigrant workers were di- 1
rectly recruited in Italy by the Belgian coal industry, with the help of Italian immigrants 2
the government and the acceptance of the reluctant unions, to work in Belgium 3
in the coal mines. An agreement between the Belgian and Italian 4
governments, signed in Rome in 1946 (Morelli 1988), provided for 5
50,000 workers to be ‘exported’ annually to Belgium. Every week, 6
tightly controlled rail convoys were organized in Italy to bring immi- 7
grants to Belgium. In the early years, these contingents of workers 8
were housed in former German prisoners camps. Immigrants had 9
all signed temporary contracts to work in the mines, and any occu- 10
pational mobility outside the extractive sector was legally prohibited. 11
The period of ‘contingented’ Italian immigration ended with the ac- 12
cident at the Marcinelle mine when 136 Italian miners lost their lives. 13
Between 1958 and 1968 immigration continued at a slower pace and 14
was mainly spontaneous. Italians were coming to Belgium as tour- 15
ists and usually found a non-qualified manual job in the building in- 16
dustry, in the iron industry, metallurgy or the extractive sector quite 17
easily. Since 1968, Italian immigration in Belgium has slowed down 18
considerably and is governed by the principle of free movement of 19
workers in the EC.5 Italian workers have entered every sector of the 20
Belgian economy and their settlement has become more and more 21
visible through the continuation of the family reunification process. 22
As a result of these three migratory phases, the Italian population 23
in Belgium nowadays amounts to about 240,000 people (Martiniello 24
1990). Including Belgians of Italian origin, the Italian collectivity 25
reaches almost 300,000, which is roughly 3 per cent of the coun- 26
try’s total population and 25-30 per cent of the total immigrant origin 27
population in Belgium. Italians and Belgians of Italian origin are the 28
largest ethnic collectivity living in Belgium. Seventy per cent of them 29
are settled in the French-speaking part of the country (Martiniello 30
1990). 31
The Italian population is increasingly presented as a ‘model of 32
perfect integration’, to be followed and imitated by all other immi- 33
grant origin populations – mainly Moroccans and Turks – present 34
in Belgium. In the discourse of politicians and many social scien- 35
tists, Italians are no longer included in the ethnic categories issue. 36
However, this is popular science which is not supported by fact. On 37
the contrary, fieldwork results show that Italians in Belgium are still 38
an ethnic category as defined above. Taking into account their socio- 39
professional position, their juridico-political status, their positions 40
in education and housing as well as the prejudices they still face, 41
Italians are nevertheless disadvantaged compared to native Belgians. 42
At the same time, when their position is compared to that of the 43
Maghrebins and to that of the Turks, it is a privileged one.

1 a disadvantaged The relatively disadvantaged position that Italians occupy in

2 position Belgium can be briefly illustrated by the following points. Firstly,
3 Italians are to a large extent still concentrated in unskilled and semi-
4 skilled manual jobs. According to different sources, the rate of Italian
5 male workers occupying such jobs varies between 69 and 76 per cent
6 (Martiniello 1992). For Belgians the rate is 47 per cent (Martinello
7 1992), while 88 per cent of male Moroccans and 92 per cent of male
8 Turks are part of the unskilled or semi-skilled labour force. As stated
9 above, Italians clearly appear to be in a sort of intermediary posi-
10 tion between Belgians and other immigrant origin workers. As far as
11 male unemployment is concerned, a recent study shows that the av-
12 erage rate is around 6.4 per cent for Belgians, 15 per cent for Italians,
13 20 per cent for Turks and 25 per cent for Moroccans (Bastenier and
14 Dassetto 1988). For women, the rates are as follows: approximately
15 17 per cent for Belgians and between 30 and 40 per cent for Italians,
16 Moroccans and Turks (Bastenier and Dassetto, 1988).
17 Secondly, being foreigners Italians are still deprived of the main
18 political rights, such as the right to vote and to be elected at all levels
19 (local, provincial, regional, national). In theory, they are free to join a
20 Belgian political party, a union, or a voluntary association. Before the
21 end of 1994 EC citizens may be granted the right to vote at local level,
22 though not necessarily to be eligible, if agreement is reached with-
23 in the framework of the European Political Union. In practice, this
24 would mean that Italians could vote in Belgium for the first time in
25 the year 2000, since the next local elections will take place in October
26 1994 and it is improbable that a positive decision on the matter will
27 be taken before then. As far as Belgians of Italian origin are con-
28 cerned, they enjoy full citizenship the same as any other Belgian.
29 Thirdly, the position of the Italian category in the educational
30 system is as weak as that of the Belgian working class. Italian youth
31 is largely concentrated in technical schools, which are at the bottom
32 level of the secondary school system. Not surprisingly, therefore,
33 they are also underrepresented at university and post-university lev-
34 el. Fourthly, Italians are still disadvantaged compared to Belgians in
35 terms of access to housing and the quality of housing. For example,
36 in Brussels 35 per cent of Belgians own their own house as against
37 only 27 per cent of Italians – taking into account that many of them
38 are probably European civil servants – 12.5 per cent of Turks and 9.5
39 per cent of Moroccans (Kesteloot 1987).
40 Finally, there still seems to be a disguised hostility towards
41 Italians among the native Belgian population, even though the main
42 targets of racism and xenophobia are the Moroccans and the Turks.
43 A distinction should be introduced at this stage between Flanders

and Wallonia. There are few Italians in Flanders, except in the min- 1
ing area of Limburg. Nevertheless, the common prejudices against 2
Italians, notably concerning the mafia, seem to find some echo in 3
parts of the Flemish population. In Wallonia, one is used to the pres- 4
ence of Italians. Walloons and Italians mix socially and at work, espe- 5
cially in working-class areas, None the less, a recent poll6 shows that 6
an anti-Italian tendency subsists among the Walloon population. To 7
paraphrase Romeo and Juliet, nobody knows why there is a war but 8
there is a war. In the present case, the word ‘war’ cannot be used, but 9
the process is the same, since no one can remember the origin of the 10
hostility. 11
However, the situation, globally speaking, is better than it was 12
forty years ago. Italians no longer live segregated in prisoner-of-war 13
camps like the ‘pioneers’ who arrived in 1945 to work in the coal 14
mines. Nowadays Italians are present in almost all sectors of the 15
economy: furthermore, there are businessmen, doctors, lawyers and 16
university students of Italian origin. There are even some Belgians of 17
Italian origin occupying positions of power in politics; For example, 18
the Minister of Education in the French Community government7 19
is Elio Di Rupo, the son of an Italian mineworker who arrived in 20
Belgium after 1945; The period of gang warfare between Belgian and 21
Italian youth in the discotheques during the late sixties and seven- 22
ties is over, and open racism against Italians is often socially con- 23
demned. Yet this incontestable improvement of the Italians’ position 24
in Belgian society is much more the result of general improvements 25
that have affected the whole of Belgian society since World War II, 26
and of a collection of individual and familial efforts, than it is of the 27
collective action of an Italian community organized around its lead- 28
ers in the Belgian state and polity. In that sense, the Italian commu- 29
nity as such is still as politically powerless as it was in the past. politically powerless 30
Contrary to a largely diffused view, ethnic communities are rarely 31
strongly structured groups of people obeying a single leadership. 32
This observation applies perfectly to the present case. The Italian 33
community is no exception. It is a split, heterogeneous and complex 34
set of local micro-communities each consisting of people with family 35
or local association links. These micro-communities are guided by as 36
many local leaders in competition with one another. 37
In 1985 there were more than 300 Italian voluntary associations 38
in Belgium.8 In the Liege area alone there were already around nine- 39
ty associations in 1989.9 Nevertheless, it is important to note that the 40
web of ethnic institutions effectively gathers together a maximum of 41
10 per cent of the ethnic category (Martiniello 1989), that is less than 42
30,000 people. Furthermore, there is no ‘institutional completeness’ 43

1 (Breton 1964) in the Italian community, although its organizational

2 complexity, in spite of the fact that the potential public is decreasing
3 with time, is very high and stable.
4 In order to describe the Italian community in Belgium better, a
5 double distinction can be made. Firstly, some institutions are mainly
6 orientated towards the ‘well-being’ of the collectivity, while others are
7 more orientated towards economic profit (Gans 1962; Nelli 1983).
8 Secondly, the first group of institutions is divided into two sub-
9 groups: the institutions transplanted from Italy, and those created in
10 Belgium. There are four main types of transplanted institutions: (1)
11 the institutions emerging from the Roman Catholic Church, whose
12 main aim is the spiritual welfare of the Italian immigrants; (2) the
13 national associations that correspond to the social and cultural asso-
14 ciations organized at the national level in Italy; (3) the patronati, that
15 is, the social services of the Italian trade unions; and (4) the Italian
16 political parties’ sections that usually follow the line defined in Rome
17 in their respective decision-making centres. Among the institutions
18 created in Belgium, there are all kinds of cultural, folkloristic and
19 sports associations – especially football. There are also regional as-
20 sociations that are in touch with Italian regional governments as well
21 as two newspapers printed in the Italian language for the Italian col-
22 lectivity.
23 fragmented There is no central coordination of all these institutions at the
24 coordination national level in Belgium. At local level some coordination exists by
25 way of various committees linked to the Italian diplomatic institu-
26 tions. More significantly, networks of ethnic institutions are formed
27 on the basis of political allegiance. As in Italy, all the institutions can
28 be classified as three ‘families’: the Catholic family, the Communist
29 or, more precisely, the ex-Communist family, and the Socialist fam-
30 ily. It is thus easily understandable that any collective action at the
31 community level faces serious internal obstacles.
32 As far as leadership is concerned, the situation is equally com-
33 plex. There is no unique leadership recognized by the Italo-Belgians.
34 Italian leadership is as fragmented as the community itself. However,
35 three empirical profiles of Italian political leaders have been distin-
36 guished. The ‘traditional’ leaders came to Belgium as migrant work-
37 ers. They are now in their sixties and have a low level of education,
38 usually not beyond the end of primary school. Their activity in the
39 community is voluntary and generally directed towards Italy. They
40 are recognized as leaders within the community at the local level and
41 they mainly use cultural references from Italy. Most ‘modern’ lead-
42 ers were born in Belgium or arrived there at an early age. They are
43 between thirty and thirty-five years old, and have been educated in

Belgium up to medium or high level. There are professional com- 1

munity workers among them, but mostly they remain involved in 2
community affairs on a voluntary basis. They enjoy internal and ex- 3
ternal recognition at local level and sometimes also at regional level. 4
They direct their action towards life in Belgium where they are cul- 5
turally at ease. The ‘imported’ leaders are neither immigrant workers 6
nor were they born in Belgium. They came from Italy to take care of 7
immigrants on a professional basis. Usually they tend to be elderly 8
people and mostly well educated. Before being recognized within the 9
community, they are recognized by Belgian and Italian authorities at 10
local and regional level. Their cultural references and their activity 11
concern Belgium as much as Italy. 12
In their relations with members of the community most leaders autocratic 13
tend to adopt an autocratic, sometimes even dictatorial, style. This style of leadership 14
is only possible because of the relative apathy of the community. In 15
their relations with the external society, most Italian leaders tend to 16
be accommodation leaders in Myrdal’s (1962) use of the term. 17
Even though there is no single coordinating body in which all 18
leaders come together, Italian leadership is to some extent structured 19
in the following way. An important characteristic of most leaders is 20
their ‘multipositionality’. They are simultaneously members of sev- 21
eral community institutions of the same political ‘family’, in which 22
they are to a greater or lesser degree always active and influential. 23
In addition, they represent one or more of those institutions in the 24
ad hoc bodies established for the relations between the community 25
and the states (Belgian as well as Italian). Leaders get to know each 26
other, therefore, through the various meetings that their ‘multipo- 27
sitionality’ implies and a certain form of privileged relationship de- 28
velops between them. A relatively small circle of competing leaders 29
is thus constituted inside the same political family. Yet even across 30
the borders of these political ‘families’, the leaders mutually recog- 31
nize each other as being the only legitimate and valid political op- 32
ponents. Consequently, there is a kind of common consciousness of 33
‘being leaders’ in leadership circles that must surely be considered as 34
a structuring factor. 35
As far as the Italian leaders’ relations with the Belgian state are 36
concerned, they mainly develop in two specifically created institu- 37
tions: the immigrants’ Communal Consultative Councils [ICCC] 38
that depend on the local level of the state; and the Foreign Origin 39
Populations’ Consultative Councils [FOPCC] that depend on the 40
communitarian level of the state.10 As far as their relations with the 41
rest of the polity are concerned, some Italian leaders also belong to 42
Belgian trade unions and political parties. 43

1 The fieldwork based on semi-participant observation, semi-direc-

2 tive depth interviews and documentary data did not lead to a rejec-
3 tion of the hypothesis, mentioned above, about the inability of ethnic
4 leaders to reduce the ethnic community’s powerlessness in spite of
5 their personal political skilfulness.11 In that sense, the reproduction
6 of the group’s powerlessness is an indicator of the Italian leaders’
7 powerlessness – as ethnic leaders – in the Belgian polity. The Italian
8 leaders’ incapacity is to be explained by the state and polity mode of
9 action towards them.
10 neutralization and The state and polity have neutralized and depoliticized Italian
11 depolitization leaders in two ways. Firstly, they have been confined in consultative
12 structures completely subordinated to the state both at the legal level
13 and at the material level. The weight of structures like the ICCC and,
14 FOPCC has always been virtually nil in Belgian political life. Trade
15 unions have followed the same logic by creating specific sub-sections
16 for ‘immigrants’, far removed from their decision-making centre.
17 Secondly, some Italian leaders and other Italian-Belgians have in-
18 dividually reached positions of power within the polity – some have
19 achieved significant success in other fields of human activity too –
20 but have renounced their leadership role in the Italian community.
21 As such, there are a few important trade unionists of Italian origin, a
22 minister of Italian origin, and a slowly growing presence of Italians
23 in the political parties. These people, who were once actual or poten-
24 tial community leaders, have thus changed into collectivity elites. In
25 other words, the state and the other main political actors have always
26 either to keep Italian leaders outside the centres of power or to allow
27 some of them in on the more or less implicit understanding that
28 they renounce their leadership role. Moreover, the divisions that ex-
29 ist within the Italian community have also been stressed by the state
30 in order to complicate further the task of the ethnic leaders.
31 The emergence of an Italian elite is just the other face of the ex-
32 clusion strategies adopted by the state and the polity. Italian lead-
33 ers have a choice between two options: they can either stick to their
34 leadership role in peripheral and uninfluential institutions or they
35 can seize the opportunity to achieve individual success by escaping
36 from the community. That choice is the core of the Belgian model
37 of insertion of ethnic categories, which is constituted by a certain
38 level of social and economic achievement and, simultaneously, by
39 complete political powerlessness. By offering this choice, the state
40 has kept its autonomy towards ethnic communities and replaced the
41 never-made, let alone implemented, global and coherent ethnic col-
42 lectivities policy.
43 How can one explain these exclusion strategies? Part of the an-

swer is to be found in Belgian political history and part in the inher- 1

ent characteristics of Belgian political life. Belgium has known po- 2
litical unity but never national unity. In the view of many observers, 3
Belgium is a mere accident of history which could be countered at 4
any time. Belgians themselves often question the reality, the exis- 5
tence and the survival of their society (Fox 1978). Since its creation, 6
Belgium has always had to face tensions, divisions, centrifugal forces 7
towards decentralization and centripetal forces towards centraliza- 8
tion. In these conditions, a set of processes and mechanisms aimed 9
at constantly assuring and reassuring the unity and global viability of 10
the society has emerged and become institutionalized. The famous 11
pacte à la belge is one of them. When critical issues are discussed, 12
conflicting groups never oppose each other beyond a certain point 13
which is considered to be dangerous for the survival of the state. They 14
then engage in extraordinary negotiations in an ad hoc commission 15
aimed at re-establishing harmony and peace between the groups in 16
a climate of moderation. This willingness to prevent divisions and 17
conflicts which might lead to the dissolution of the state has been ob- 18
served since its very creation. Belgium has thus developed the ‘art of 19
temporizing’ through setting up multiple commissions and councils, 20
usually consultative bodies, and habitually finding harmonious solu- 21
tions to serious problems on the quiet. As Fox (1978) lucidly states, 22
Belgium is sufficiently concerned ‘with its potentiality for internal 23
conflicts and with its intrinsic risk of self-demolition to establish and 24
maintain permanent pacts between the various actors about social 25
issues considered to be critical’. 26
The hypothesis can be advanced that immigration and the pres- 27
ence of immigrant origin populations are precisely seen as one such 28
critical issue. To the extent that ethnic categories represent about 29
8 per cent of the total population and that they come from various 30
countries whose cultural differences are commonly underlined, their 31
presence is considered to be a potential danger because it compli- potential danger 32
cates even more the already intricate ethno-national Belgian context. 33
This hypothesis is supported by the recent political discourse, admit- 34
tedly during a period of relatively bad relations between the Flemish 35
and the Walloons, in the context of the new discussions about the 36
further federalization of the state after the legislative elections of 37
November 1991. A large consensus has developed between the vari- 38
ous Belgian political actors to keep the threat, that is, immigrants 39
and their descent as communities, outside, or at the margin of, the 40
polity. The inclusion of ethnic communities in the polity is thought 41
to introduce a new and dangerous dimension of the ‘ethnicization’ 42
of Belgian political life. The generally ‘accepted refusal’ of this new 43

1 risk of ‘ethnicization’ of political life is then to be interpreted, in the

2 dominant approach, as a condition for the survival of the present pre-
3 carious equilibrium in the state. In this sense, the exclusion of ethnic
4 communities from the polity can be explained as a survival strategy
5 by a state that feels itself subjectively threatened as well as constitut-
6 ing a symbolic sign of a unity that is unfounded.
7 The issues of the creation and the development of various insti-
8 tutions and councils to deal with the ethnic categories issues under
9 the state’s supervision have to be analysed in that general context.
10 They represent different forms of exclusion and ‘peripheralization’
11 of ethnic categories through the neutralization and ‘depoliticization’
12 of potential and existing ethnic leaders. The Belgian processes of cat-
13 egorization, exclusion and reproduction of powerlessness could be
14 ‘immigrization’ termed ‘immigrization’, because the official vocabulary used refers
15 predominantly to the notions of immigration and immigrants. They
16 constitute the Belgian model of ‘integration by exclusion’: the Belgian
17 state and polity offer some opportunities of social and economic pro-
18 motion to ethnic communities whilst simultaneously keeping them
19 out of the state and polity by ‘using’ ethnic leaders.
20 The Belgian model of integration, as described above, seems to
21 have worked rather well with the ethnic communities and leaders
22 whose presence is the result of the first waves of immigration after
23 World War II. The Italian leaders have been depoliticized and neu-
24 tralized quite easily and, consequently, the Italian community has
25 been kept out or at the margin of the state and polity. Will this model
26 be equally effective when applied to ethnic categories whose pres-
27 ence is more recent – mainly the Moroccans and Turks – not to men-
28 tion current immigration and the movements of political refugees
29 that will certainly lead to the settlement of new ethnic categories in
30 Belgium? At present this crucial question remains unanswered.
31 What is certain, however, is that the social, economic and political
32 conditions are very different now compared to what they were in the
33 sixties, seventies and even the early eighties, so that the viability of
34 the Belgian model of integration can seriously be questioned. Firstly,
35 the working-class organizations, especially the unions, which played
36 such an important role in the processes of creating an Italian socio-
37 political elite and of neutralizing Italian leaders, are less willing and
38 able to exert the same role as far as Moroccan and Turkish leaders
39 and elites are concerned. Secondly, the economic success that some
40 Italians enjoyed in the past is much less evident today, because of the
41 continuing economic crisis that began in the early seventies, because
42 of the high rate of unemployment, and because of a growing dualiza-
43 tion of society. Thirdly, the electoral success of the Vlaams Blok and

extreme-right-wing parties in the general elections of November 1991 1

could provoke a further radicalization of the general political climate 2
and of the discourse of traditional parties on ethnic communities and 3
immigration issues. Fourthly, the domestic ethnic tension between 4
the Flemish and the Walloons has never before been as acute and 5
dangerous for the existence of the Belgian state. The combination of 6
these four elements can, at least in the short and medium term, lead 7
to a radicalization of the exclusion processes towards non-European 8
ethnic categories. 9
At least three types of reactions can then be expected in terms of three types of 10
ethnic leadership and elites. Possibly, there will be an emergence of reactions 11
more radical ethnic leadership, especially among the youth, in self- 12
made political and cultural organizations. Attempts to create ethnic 13
lobbies in party politics will become increasingly probable, since 14
more and more young people of ethnic categories acquire Belgian 15
nationality. Finally, a further development of individual exit strate- 16
gies, for example through small businesses and education, can be 17
envisaged. It should be noted that, in the latter, women could play a 18
vanguard role in the sense that their results seem to be much better 19
than those of male counterparts. Will this lead to the emergence of 20
a female ethnic leadership? This remains another open question. In 21
any case, it would be a great novelty in ethnic leadership in Belgium, 22
since until now it has been almost exclusively a male phenomenon. 23
24
25
Acknowledgments 26
27
I am grateful to Rainer Bauböck, John Bade, Alec Hargreaves, Ronald 28
Kaye, Zig Layton-Henry, Jan Rath and Giovanna Zincone for their 29
helpful comments on a draft version of this article. 30
31
32
Notes 33
34
1 In the remainder of this article the phrase ‘from immigrant origin’ will no 35
longer be used, since it is now clear that the study deals with populations 36
whose presence is a consequence of international labour and political im-
migration in the post-World War II period.
37
2 The expression ‘national-origin category and community from immigrant 38
origin’ might have been used instead of ‘ethnic category and community of 39
immigrant origin’ to avoid any possible confusion in the meaning of ethnic- 40
ity. However, for elegance sake, the ethnic vocabulary has been kept. 41
3 The question until which biological generation does an ethnic category con-
42
tinue to be named as such will not be addressed here, because it is mainly an
empirical one. 43

1 4 Of course, the reproduction is not perfect. There is an individual upward

2 mobility process among ethnic categories, the importance of which will vary
3 from case to case.
5 Regulation no. 1612/68 of the EC Council, Official Journal of the European
4 Communities, no. L257, 19 October 1960.
5 The poll was published by the weekly, Pourquoi Pas?, 17 March 1988.
6 7 The Belgian quasi federal state consists of two kinds of federal institutions,
7 namely the Regions (Flanders, Wallonia and Brussels) and the Communities
8 (French, Flemish and German). Each Community has a government that is
responsible for culture, education, sport, tourism, etc.
9
8 Ministero degli Affari Esteri, Direzione Generale dell’Emigrazione e degli
10 Affari Sociali, Associazioni ltaliane nel mondo 1984, Roma, 1985.
11 9 Official data from the General Consulate of Italy in Liège, October 1989.
12 10 The very complex structure of the Belgian state is very well synthetized in
13 Mean (1989).
14 11 For an extensive analysis of the fieldwork results, see Martiniello (1992). For
details about the methodology, see Martiniello (1990).
15
16
17
18 References
19
20 ALBA, RICHARD D. 1976 ‘Social assimilation among American
21 Catholic NationalOrigin groups’, American Sociological Review, vol.
22 41, no. 6, pp. 1030-1046
23 – 1985 ‘The twilight of ethnicity among Americans of European an-
24 cestry: the case of Italians’, Ethnic and Racial Studies, vol. 8, no. 1,
25 pp. 134-58
26 ANTHIAS, FLOYA 1990 ‘Race and class revisited – conceptualizing
27 race and racisms’, Sociological Review, vol. 38, no. 1, pp. 19-42
28 BARTH, FREDRIK (ed.) 1969 Ethnic Groups and Boundaries: The
29 Social Organization of Culture Differences, London and Oslo: Allen
30 & Unwin/Forgalet
31 BASTENIER, ALBERT and DASSETTO, FELICE 1988 Cycles migra-
32 toires, espaces, insertion, Brussels: CCPOE
33 BONACICH, EDNA 1973 ‘A theory of middleman minorities’,
34 American Sociological Review, vol. 38, no. 5, pp. 583-94
35 BRASS, PAUL (ed.) 1985 Ethnic Groups and the State, London and
36 Sidney: Croom Helm
37 BRETON, RAYMOND 1964 ‘Institutional completeness of ethnic
38 communities and the personal relations of immigrants’, American
39 Journal of Sociology, vol. 70, no. 2, pp. 193-205
40 FOX, RENÉE 1978 ‘Why Belgium?’, Archives Européennes de Sociologie,
41 vol. 19, no. 2, pp. 205-28
42 GANS, HERBERT S: 1962 The Urban Villagers: Group and Class in
43 the Life of Italian-Americans, New York: The Free Press

KESTELOOT, CHRISTIAN 1987 ‘Le marché du logement et les im- 1

migrés à Bruxelles’, Tribune Immigree, nos 26-27, pp. 5-15 2
KEYES, CHARLES 1976 ‘Towards a new formulation of the concept 3
of ethnic group’ Ethnicity, vol. 3, pp. 202-13 4
LUKES, STEVEN 1974 Power, A Radical View, London: Macmillan 5
– (ed.) 1986 Power, Oxford: Blackwell 6
MARTINIELLO, MARCO 1989 ‘L’associationisme regional italien 7
en Belgique: point final de l’immigritude?’, paper presented at the 8
Conference ‘Hommes, cultures et capitaux dans les relations italo- 9
belges aux 19ème et 20ème siècles’, Rome: Academia Belgica 10
– 1990 ‘Élites, leadership et pouvoir dans les communautes eth- 11
niques d’origine immigrée: le cas des Italiens en Belgique fran- 12
cophone’, Florence: European University Institute, unpublished 13
PhD thesis 14
– 1992 Leadership et Pouvoir dans les Communautes d’Origine Immigrée, 15
Paris: CIEMI/L’Harmattan 16
MARX, KARL 1956 The Poverty of Philosophy, Moscow: Progress 17
Publishers [First published in French as Misère de la Philosophie 18
in 1847] 19
MEAN, ANDRE 1989 Comprendre la Belgique fédérale. Les nouvelles 20
institutions, Brussels: La Libre Belgique 21
MORELLI, ANNE 1988 ‘L’appel de la main-d’oeuvre italienne pour les 22
charbonnages et sa prise en charge à son arrivée en Belgique dans 23
l’immédiat apres-guerre’, Revue Belge d’Histoire Contemporaine, 24
vol. 19, nos 1-2, pp. 83-130 25
MYRDAL, GUNNAR 1962 An American Dilemma. The Negro Problem 26
and Modern Democracy, New York: Harper and Row 27
NELLI, HUMBERT S. 1983 From Immigrants to Ethnics: the Italian- 28
Americans, Oxford: Oxford University Press 29
PARETO, WILFREDO 1986 The Rise and Fall of Elites, Salem: Ayer 30
Company 31
SCHERMERHORN, RICHARD A. 1974 ‘Ethnicity in the perspective 32
of the sociology of knowledge’, Ethnicity, vol. I, no. I, 1974, pp. 1-14 33
WEBER, MAX 1971 Économie et société, Paris: Pion. [First published 34
in German in 1922] 35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Racism in Europe: unity and diversity 3
4
Michel Wieviorka 5
6
7
8
9
10
11
12
In the early 1990s, the sociologist Michel Wieviorka was one of the leading 13
specialists on racism in Europe. He published several books in French on the 14
issue. This article was published in 1994 in a book entitled Racism, Modernity 15
and Identity. Here Wieviorka convincingly defends the idea that any analysis 16
of racism in Europe has to recognise the links between racism and moder- 17
nity. Furthermore, Wieviorka distinguishes four forms of racism: universalis- 18
tic, the ‘poor white’ response, anti-modernist and a form of racism linked to 19
intergroup conflict in the modern era. This distinction has become a classic 20
one in the European study of racism. 21
22
23
Observing growing racist tendencies that affect most European 24
countries, an increasing number of scholars feel an urgent need for 25
a comparative reflexion that may bring answers to a central question: 26
over and beyond the empirical evidence of differences, is there not unity in racism 27
a certain unity in contemporary racism in Europe? Is it not possible 28
to elaborate a reasoned set of hypotheses that could account for most 29
national racist experiences in Europe, while shedding some light on 30
their specificities? 31
European unification, in so far as it exists, and the growth of rac- 32
ism are obviously distinct phenomena, and it would be artificial to try 33
and connect them too directly. The most usual frame of reference for 34
any research about racism and race relations remains national. And 35
even the vocabulary or, more deeply, the analytical and cultural cat- 36
egories that we use when dealing with this issue vary so widely from 37
one country to another that we meet considerable difficulties when 38
trying to translate precise terms. There may be large differences in 39
language, and words with negative connotations in one country will 40
have positive ones in another. Nobody in France, for instance, would 41
use the expression relations de race, which would be regarded as rac- 42
ist, although it is commonly employed in the United Kingdom. 43

260 michel wieviorka
1 The key preliminary task, therefore, is not to contribute direct em-

2 pirical knowledge about the various expressions of racism in Europe,
3 as can be found, for instance, in the important survey of ‘Racism and
4 xenophobia’ published in 1989 by the European Community (CCE,
5 1989). Nor is the initial task to compare elementary forms of rac-
6 ism, such as harassment, stereotypes, discrimination or political rac-
7 ism in a certain number of countries, in order to prove that they are
8 more or less similar, or that they follow a similar evolution. Rather
9 the problem is primarily conceptual. If we want to test the idea of a
10 certain unity of contemporary racism in Europe, we must elaborate
11 sociological and historical hypotheses, and then apply them to the
12 facts that we are able to collect. Thus the most difficult aspect of a
13 comparative approach is not to find data, but to organize it with well-
14 thought-out hypotheses.
15 My own hypotheses can be formulated in two different ways, one
16 of which is relatively abstract and the other more concrete.
17
18
19 Racism and modernity
20
21 An initial formulation of the problematic, in effect, consists in the
22 construction of a global argument enabling us to demonstrate that
23 racism is inseparable from modernity, as the latter developed from
24 European origins, and from its present crisis (Wieviorka, 1992a).
25 Racism, both as a set of ideologies and specious scientific doctrines,
26 and as a set of concrete manifestations of violence, humiliation and
27 discrimination, really gathered momentum in the context of the im-
28 mense changes of which Europe was the centre after the Renaissance.
29 It developed further in modern times, with the huge migrations, the
30 extension of trading relationships, the industrialization of Western
31 society and colonization. But racism, in its links with modernity, can-
32 not be reduced to a single logic, and even seems to correspond to
33 processes which are sometimes so distinct that numerous demands
34 are made for the discussion of racisms in the plural. This in fact gives
35 rise to a debate the terms of which are badly posed. It is effectively
36 possible to set up an integrated, global argument in which the vari-
37 sociological ous forms of racism, including anti-semitism, find their theoretical
38 unity of racism place, and which goes in the direction of a sociological, even anthro-
39 pological, unity of racism. One can also consider each of these forms
40 in its historical specificity, which goes in the opposite direction. Both
41 approaches are legitimate and complementary, but since we are
42 thinking here about the unity of contemporary forms of racism in
43 Europe, it is clear that we should privilege the former. This leads us

to distinguish four main lines of argument which cross the space of 1

racism in its relation to modernity. 2
In the first instance, as the companion of modernity triumphant, 3
racism is universalist, denouncing, crushing and despising different 4
identities. Whence the apparition of inferior ‘races’ as an obstacle to 5
the process of expansion, in particular colonial expansion, or des- 6
tined to be exploited in the name of their supposed inferiority. 7
Next, linked to processes of downward social mobility, or exclu- 8
sion, racism is the expression, as well as the refusal, of a situation 9
in which the actor positively values modernity, but lives, or is afraid 10
he/she will be exposed to a form of expulsion which will marginal- 11
ize him/her. The actor then assumes a reflex or an attitude of ‘poor 12
white’, particularly common in contexts of economic crises or of re- 13
traction from the labour market. Racism here is a perversion of a 14
demand to participate in modernity and an opposition to the effective 15
modalities of its functioning. 16
A third line of argument corresponds not to a positive valorization 17
of modernity, the rise of which must be ensured, or from which one 18
refuses to be excluded, but to appeals to identity or to tradition which 19
are opposed to modernity. The nation, religion and the community 20
then act as markers of identity, thus giving rise to a racism which 21
attacks those who are assumed to be the vectors of a detested mo- 22
dernity. The Jews are often the incarnation of these vectors, as are, 23
in some circumstances, those Asian minorities who are perceived 24
as being particularly economically active. Finally, racism can cor- 25
respond to anti- or non-modern positions, which are displayed not 26
against groups incarnating modernity, but against groups defined 27
themselves by an identity without any reference to modernity. It ex- 28
presses, or is an extension of, intercultural, intercommunity, inter- 29
ethnic or similar tensions. 30
It is therefore possible to represent the space of racism around the space of racism 31
four cardinal points: 32
33
Modernity against identities 34
Identities against identities 35
Identities against modernity 36
Modernity against modernity 37
38
In a space of this type, the racist actors do not necessarily occupy one 39
single position, and their speech and their behaviour are frequently 40
syncretic and vary over time. There are even sometimes paradoxi- 41
cal mixtures of these various positions, when people, for instance, 42
reproach a racialized group with symbolizing at the same time mo- 43

1 dernity and traditional values which they consider deny modernity:

2 in the past, but also today, Jews, in many cases, fulfil this double
3 function (Wieviorka, 1992b). They are hated in the name of their
4 supposed identification with political power, money, the mass-media
5 and a cosmopolitan internationalism, but also because of their dif-
6 ference, their visibility, their nationalism and support or belonging
7 to the state of Israel, or because they flaunt their cultural traditions
8 or their religion.
9 This theoretical construction of the space of racism may help us
10 to answer our question. In effect, it enables us to read the European
11 experience, and above all its recent evolution. The latter has long
12 been dominated, on the one hand, by a racism of the universalist,
13 colonial type and, on the other hand, by oppositions to modernity
14 which have assumed the form of anti-Semitism; today, much more
15 than previously, it is directed by the fear or reality of exclusion and
16 downward social mobility, and on the other by tensions around iden-
17 tity and vague fears of which the most decisive concern the question
18 of belonging to the nation.
19
20
21 Formation and restructuration of the European model of
22 national societies
23
24 The argument outlined above can be completed by a much more con-
25 crete historical analysis of the recent evolution of most of the major
26 western European countries. The latter, throughout this century, and
27 up to the 1960s or 1970s, can be defined on the basis of a model
28 which integrates three elements which are then weakened and de-
29 structured, reinvigorating the question of racism.
30
31 The era of integration
32 In most western European countries, racism, before the Second
33 World War, was a spectacular and massive phenomenon, much
34 more widespread than today. Colonial racism postulated the inferior-
35 colonial racism ity of colonized people of ‘races’, and modern anti-Semitism gave a
36 new and active dimension to former anti-Judaïsm. This is why we
37 must introduce a sense of relativity into our perceptions of contem-
38 porary racism. This is why we must also think in terms of periods,
39 with the idea of a certain unity in time for the phenomenon that we
40 are discussing. This idea means not that there is no continuity in
41 racist doctrines, ideologies, prejudice or more concrete expressions,
42 but that a new era in the history of racism began with the retreat, as
43 Elazar Barkan (1992) says, of scientific racism, the end of decoloniza-

tion, and, above all, the ‘economic crisis’ that has in fact meant the 1
beginning of the decline of industrial societies. 2
Until that time, i.e. the 1960s and 1970s, most European coun- 3
tries had succeeded, to a greater or a lesser extent, depending on the 4
country, in integrating three basic components of their collective life: 5
an industrial society, an egalitarian state and a national identity. 6
Most European countries have been industrial societies: that is, 7
they have had a set of social relations rooted in industrial labour and 8
organization. From this point of view, they have been characterized 9
by a structural conflict, which opposed the working-class movement 10
and the masters of industry, but which extended far beyond work- 11
shops and factories. This conflict gave the middle classes a possibil- 12
ity to define themselves by either a positive or negative relationship 13
towards the working-class movement. It brought to unemployed 14
people the hope and sometimes the reality of being helped by this 15
movement. It was also the source of important political debates deal- 16
ing with the ‘social question’. Furthermore, it influenced intellectual 17
and cultural life profoundly, and acted as a point of reference for 18
many actors, in the city, in universities, in religious movements and 19
elsewhere. 20
European countries, and this is the second basic component of 21
our model of analysis, have also been able to create and develop insti- 22
tutions which aimed at ensuring that egalitarian treatment was im- 23
parted to all citizens as individuals. The state has generally taken over 24
various aspects of social welfare and security. It has become a wel- 25
fare state. The state also introduced or defended a distance between 26
religion and politics. Although countries such as Spain, Portugal 27
and Greece have recently experienced dictatorial regimes, states in 28
Europe have generally behaved, since the Second World War, as war- 29
rants for democracy. 30
Lastly, most European countries have given a central importance 31
to their national identity. This identity has usually included two dif- 32
ferent aspects, sometimes contradictory, sometimes complementary. 33
On one hand, the idea of a nation has corresponded to the assertion 34
of a culture, a language, a historical past and traditions, with some 35
tendencies to emphasize primordial ties and call for a biological defi- 36
nition loaded with racism, xenophobia and anti-Semitism. On the 37
other hand, the nation has also been defined in a more positive way, 38
as bound to the general progress of mankind and to universal values 39
that could be defined in economic, political or ethical terms. In this 40
last perspective, a nation is related to reason, progress, democracy of 41
human rights. industrial society, 42
Industrial society, state and nation: these three basic elements have state and nation 43

1 never been consonant with their highest theoretical image. One can
2 easily show the weakness of the working-class movement in some
3 countries, or its constant subordination to political forces, the limits
4 of the welfare state everywhere in the past, and the domination of the
5 reactionary and xenophobic aspects of nationalism in many circum-
6 stances. Moreover, some European countries have defined them-
7 selves as bi- or plurinational. But since we recognize these limits,
8 and since we recognize many differences between countries, we can
9 admit, without the danger of creating a myth, that our three basic ele-
10 ments are typical of European countries until the 1960s and 1970s.
11 Not only have they characterized three countries, but they have also
12 been relatively strongly articulated, so much so that various terms
13 are used to express this articulation: for instance, integration, nation-
14 state and national society. We must be very cautious and avoid de-
15 veloping the artificial or mythical image of countries perfectly suited
16 to the triple and integrated figure of an industrial society, a two-di-
17 mensional nation and a modern and egalitarian state. But our repre-
18 sentation of the past is useful in considering the evolution of the last
19 twenty or thirty years, an evolution which is no doubt dominated by
20 the growing weakness and dissociation of our three basic elements.
21
22 The era of destructuration
23 All European countries are experiencing today a huge transforma-
24 transformation tion which affects the three components of our reflection, and de-
25 fines what I have called, in the case of France, ‘une grande mutation’
26 (Wieviorka, 1992c).
27 Industrial societies are living their historical decline, and this phe-
28 nomenon should not be reduced to the spectacular closing of work-
29 shops and factories. More important in our perspective is the decay
30 of the working-class movement as a social movement. In the past,
31 the working-class movement was, to various degrees, capable of in-
32 corporating in a single action collective behaviour corresponding to
33 three major levels. There could be limited demands, struggles based
34 on the professional defence of political demands, dealt with by the in-
35 stitutional system, and, at the highest level of its project, orientations
36 challenging the control and the direction of progress and of indus-
37 try. These orientations are quite out of place today: the working-class
38 movement is breaking up, and this decomposition produces various
39 effects (Touraine et al., 1987). Among workers, there is a strengthen-
40 ing of tendencies towards corporatism and selfishness – those work-
41 ers who still have a certain capacity of action, because of their skill or
42 their strategic position in their firm, develop struggles in the name of
43 their own interests, and not in the name of more general or universal
ones.

Sometimes workers’ demands can no longer be taken up by the 1

trade unions, which have been considerably weakened. This can re- 2
sult in violent forms of behaviour, or in spontaneous forms of or- 3
ganization, such as the recent ‘co-ordinations’ in France, which are 4
easily infiltrated by extremist ideologies. 5
In such a context, the middle classes no longer have to define 6
themselves by reference to class conflicts, and they tend to oscillate 7
between, on the one hand, unrestrained individualism and, on the 8
other, populism or national-populism, the latter being particularly 9
strong among those who experience downward mobility or social 10
exclusion. These two distinct phenomena are closely related to so- social and economic 11
cial and economic dualization. In the past, most people could have a dualization 12
strong feeling of belonging to a society, ‘down’ as workers, or ‘up’ as 13
éites, or middle classes. Today, a good number of people are ‘in’, and 14
constitute a large middle class, including those workers who have 15
access to jobs, consumption, health or education for their children, 16
while a growing proportion of people are ‘out’, excluded and margin- 17
alized. 18
Such an evolution may lead to renewed expressions of racism. 19
Those who are ‘out’, or fear to be, have a feeling of injustice and 20
loss of previous social identity. They think the government and the 21
politicians are responsible for their situation, and may develop pop- 22
ulist discourses and attitudes in which anti-migrant or ethnic mi- 23
norities racism can take place. They then impute their misfortune 24
to migrants, even if these migrants share the same experience. And 25
those who are ‘in’ may develop more subtle forms of racism, trying to 26
secure themselves with a colour bar or by individual or collective be- 27
haviours that create social and racial segregation and build symbolic 28
but also real barriers. Furthermore, the logic of segregation, particu- 29
larly at the political level, is always likely to become indistinguishable 30
from a national and populist form of discourse which amalgamates 31
the fears, anger and frustrations of the excluded and the social self- 32
centredness of those who wish to defend their status and their way of 33
life. This merging therefore gives a result which is only paradoxical 34
in appearance, since it results in an identical form of racism in those 35
people who have experienced living with, or close to, immigrants or 36
similar categories of people, and in those who have not actually done 37
so, but who have heard about it through the mass-media or from 38
rumours. 39
A second element of destructuration deals with the state and pub- crisis of state and 40
lic institutions, which encounter increasing difficulties in trying to republic institutions 41
spect egalitarian principles, or in acting as welfare states. Everywhere 42
in Europe, the number of unemployed people has grown, creating 43

1 not only a great many personal dramas, but also a fiscal crisis of the
2 state. The problems of financing old-age pensions, the health care
3 system, state education and unemployment benefits are becoming in-
4 creasingly acute, while at the same time there is a rising feeling of in-
5 security which is attributed, once again, to immigrants. The latter are
6 then perceived in racist terms, accused not only of taking advantage
7 of social institutions and using them to their own ends, but also of
8 benefiting from too much attention from the state. At the same time,
9 the ruling classes have been tempted since the 1970s by liberal policies
10 which in fact ratify and reinforce exclusion and marginalization.
11 The crisis of the state and the institutions is a phenomena which
12 must be analytically distinguished from the decline of industrial so-
13 ciety and the dualization which results from its decline. But the two
14 phenomena are linked. Just as the welfare state owes a great deal,
15 in its formation, to the social and political discussions which are in-
16 separable from the history of the working class, which is particularly
17 clear in the countries endowed with strong social democracy, so too
18 the crisis of the welfare state and the institutions owes a great deal to
19 the destructuration not only of these discussions and conflicts, but
20 also of the principal actor which informed them, the working-class
21 movement.
22 nationalism A third aspect of the recent evolution concerns the national is-
23 sue, which becomes nodal – all the more so as social issues are not
24 politically treated as such. In most European countries, political de-
25 bates about nation, nationality and citizenship are activated. In such
26 a context, nationalism loses its open and progressive dimensions,
27 and its relationship with universal values, and is less and less linked
28 with ideas such as progress, reason or democracy. National identity
29 is increasingly loaded with xenophobia and racism. This tendency
30 gains impetus with the emergence or growth of other identities
31 among groups that are defined, or that define themselves, as com-
32 munities, whether religious, ethnic, national or regional. There is a
33 kind of spiral, a dialectic of identities, in which each affirmation of a
34 specific identity involves other communitarian affirmations among
35 other groups. Nationalism and, more generally speaking, communal
36 identities do not necessarily mean racism. But as Etienne Balibar
37 explains, racism is always a virtuality (Balibar and Wallerstein, 1988).
38 This virtuality is not nurtured uniquely by the presence, at times
39 exaggerated and fantasized, of a more or less visible immigration.
40 It also owes a considerable amount to phenomena which may even
41 have nothing to do with it. Thus national identity is reinforced in its
42 most alarming aspects when national culture appears to be threat-
43 ened by the superficial and hypermodern character of an internation-

al culture which originates primarily in America, by the political con- 1

struction of Europe or, again, by the globalization of the economy. 2
At the same time, it becomes more and more difficult to assert 3
that society, state and nation form an integrated whole. Those who 4
call for universal values, human rights and equality, who believe that 5
each individual should have equal opportunities to work, make mon- 6
ey and then participate fully in cultural and political life – in other 7
words, those who identity themselves with modernity – are less and 8
less able to meet and even to understand those who have the feeling 9
of being excluded from modern life, who fear for their participation 10
in economic, cultural and political life, and who retire within their 11
national identity. In extreme cases, social and economic participa- 12
tion are no longer linked with the feeling of belonging to a nation, 13
the latter being what remains when the former becomes impossible. 14
Reason, progress and development become divorced from nation, 15
identity and subjectivity, and in this split, racism may easily develop. 16
In the past, industrial society often offered workers disastrous 17
conditions of work and existence. But the working-class movement, 18
as well as the rulers of industry, believed in progress and reason, and 19
while they were opposed in a structural conflict, this was precisely 20
because they both valorized the idea of progress through industrial 21
production, and both claimed that they should direct it. The nation, 22
and its state, as Ernest Gellner explains (1983) were supposed to be 23
the best frame for modernization, and sometimes the state not only 24
brought favourable conditions, but also claimed to be the main agent 25
of development. Nationalism could be the ideology linked to that 26
perspective, and not only a reactionary or traditionalist force. Today, 27
waters divide. Nationalism is mainly expressed by social and political 28
groups frightened by the internationalization of the economy and 29
culture. It is increasingly differentialist, and racism develops as so- 30
cial problems such as exclusion and downward mobility grow, and as 31
anxiety develops in regard to national identity. 32
33
34
The categories of the sociological analysis of racism 35
36
The argument outlined above is historical and sociological in nature, 37
but a closer examination of the contemporary phenomena of racism 38
requires explicitness in the instruments and, therefore, the catego- 39
ries of analysis of racism properly speaking (Wieviorka, 1991). 40
41
The two logics of racism 42
Contemporary sociological literature increasingly insists on the idea 43

1 of changing forms of racism. Some scholars, relying on American

2 studies, oppose the old ‘flagrant’ racism to the ‘subtle’ new versions
3 (Pettigrew, 1993). Others emphasize a crucial distinction, which
4 could, in an extreme interpretation, lead to the idea of two distinct
5 kinds of racism. Following authors such as Martin Barker or Pierre-
6 André Taguieff, we should distinguish between a classical, inegali-
7 classical racism tarian racism and a new, differentialist one (Barker, 1981; Taguieff,
8 1988). The first kind considers the Other as an inferior being, who
9 may find a place in society, but the lowest one. There is room for
10 differentialistic inferior people in this perspective, as long as they can be exploited
11 racism and relegated to unpleasant and badly paid tasks. The second kind
12 considers the Other as fundamentally different, which means that
13 he/she has no place in society, that he/she is a danger, an invader,
14 who should be kept at some distance, expelled or possibly destroyed.
15 The point is that for many scholars the new racism, sometimes also
16 referred to as cultural racism, is the main one in the contemporary
17 world, while the inegalitarian one becomes secondary.
18 As long as this remark is intended as a statement of historical
19 fact, based on the observation of empirical realities of present-day
20 racism, it is acceptable. But it must not take the place of a gener-
21 al theory of racism. First, cultural or differentialist perspectives in
22 racism are not new. It is difficult to speak of Nazism, for instance,
23 without introducing the idea that anti-Semitism in the Third Reich
24 was deeply informed by these perspectives. Jews were said to corrupt
25 Aryan culture and race, and the ‘final solution’ planned not to assign
26 them to the lowest place in society, but to destroy them. Second, the
27 opposition between the two main logics of racism should not conceal
28 the main fact, which is that a purely cultural definition of the Other,
29 as well as a purely social one, dissolves the idea of race. On one hand,
30 Claude Levi-Strauss is not a racist when he emphasizes cultural dif-
31 ferentiation. One is a racist only when there is any reference to race
32 in a cultural opposition, when beneath culture we can, explicitly or
33 implicitly, find nature: that is, in an organicist or genetic represen-
34 tation of the Other as well as oneself. On the other hand, when the
35 Other is defined only as socially inferior, exploited or marginalized,
36 the reference to race may disappear or become, as William J. Wilson
37 suggests (1978), less significant.
38 In fact, in most experiences of racism, the two logics coexist, and
39 racism appears as a combination of them both. There are not two
40 racisms, but one, with various versions of the association of cultural
41 differentialism and social inegalitarianism. The general analysis that
42 has been presented for contemporary Europe helps us to refuse the
43 idea of a pure, cultural racism, corresponding to a new paradigm that

would have taken the place of an old one. The sources of European 1
contemporary racism, as I have suggested, are in the crisis of na- 2
tional identities and in the dualization of societies, which favour a 3
differentialist logic. But they are also connected with phenomena of 4
downward social mobility and economic crisis, which lead to popu- 5
lism and exasperation and have an important dimension in appeals 6
for an unequal treatment of migrants. 7
8
Two main levels 9
As I have indicated in a recent book (Wieviorka, 1991), we may dis- four levels of racism 10
tinguish four levels in racism. The way that experiences of racism 11
are articulated at the different levels where they act may change with 12
their historical evolution. Our distinction is analytical, and should 13
help us as a sociological tool. 14
A first level refers to weak and inarticulated forms of racism, 15
whatever they may consist of: opinions and prejudice, which are 16
more xenophobic and populist than, strictly speaking, racist; and dif- 17
fuse violence, limited expression of institutional discrimination or 18
diffusion of racial doctrines, etc. At this first level, racism is not a 19
central issue and it is so limited, quantitatively and qualitatively, that 20
I have chosen to use the term infraracism to characterize it. 21
We may speak of split racism at a second level, in reference to 22
forms of racism which are still weak and inarticulate, but stronger 23
and more obvious. At this stage, racism becomes a central issue, but 24
does not give the image of a unified and integrated phenomenon, 25
mainly because of the lack of a strong political expression. 26
We may speak of political racism, precisely, when political and in- 27
tellectual debates and real political forces bring a dual principle of 28
unity to the phenomenon. On the one hand, they give it an ideologi- 29
cal structure, so that all its expressions seem to converge and define 30
a unique set of problems; on the other hand, they offer it practical 31
forms of organization. 32
At the fourth level, we may call total racism those situations in 33
which the state itself is based on racist principles. There is nowa- 34
days no real threat of total racism in our countries; and we may now 35
simplify the distinction into four levels of racism by reducing them 36
to two main ones, the infrapolitical level, including infra and split rac- 37
isms, and the political one. 38
We can now come back to our general analysis of European con- 39
temporary racism and be more precise. This rise of the phenome- 40
non, following what was previously said, is due to the evolution of 41
three basic elements, and to their destructuration. We may add that 42
it appears first at an infrapolitical level, and that it then ascends to the infrapolitical to 43
political level, with variations from one country to another. political level

1 In certain cases, a rather important political party appears and

2 develops quickly, as in France with the Front National. In other cases,
3 such a party appears but quickly declines, which means not that rac-
4 ism necessarily stays at the infrapolitical level, but that it informs
5 political debates without being the flag of one precise strong orga-
6 nization – this could define the English experience. But above all,
7 the analytical distinction into levels enables us to introduce a central
8 question: is there not throughout Europe the same danger of seeing
9 political actors capable of taking over and of directing infrapolitical
10 racism?
11 On the one hand, we observe in several countries the growing in-
12 fluence of racist ideologies, but also of political organizations which
13 are no longer small groups of activists and which may occupy an
14 important space in political life. The French Front National appears
15 as a leader in Europe, and sometimes as a model, but other parties
16 or movements should be quoted too: the Deutsche Volksunion and the
17 Demokratische Partei Deutschlands in Germany; the FPO in Austria,
18 which gained 22.6 per cent of the votes in the November 1991 elec-
19 tions in Vienna; the Vlaams Blok in Flanders, with twelve members
20 of Parliament since November 1991; and the Italian Leagues.
21 One must be careful, however, not to exaggerate. The more ex-
22 treme-right parties occupy an important place, the more they appear
23 as populist rather than purely racist. Racism, strictly speaking, is
24 only one element, and sometimes a minor one, along with strong na-
25 tionalism or regionalism. Moreover, political and electoral successes
26 force these parties to look respectable, and avoid overtly flagrant ex-
27 pressions of racism.
28 On the other hand, racism appears in non-political contexts, when
29 prejudice and hostile attitudes to migrants develop, when social and
30 racial segregation is increasingly visible (which is the case in France,
31 where the issue of racism is constantly related to the so-called urban
32 crisis and ‘the suburban problem’), when violent actions develop,
33 sometimes with a terrorist aspect, when various institutions includ-
34 ing the police have a responsibility for its growth, when discrimina-
35 tion is obvious (for instance, in relation to housing or employment),
36 and when the media contributes to the extension of prejudice. In
37 such a perspective, all the European democracies have to face the
38 same problem. There is a growing opportunity for extreme-right
39 forces to capitalize on fears, frustrations, unsatisfied social demands
40 and feelings of threat to national identity. Even worse, there is a dan-
41 ger that these forces will introduce new elements into infrapolitical
42 racism. This is the case in France, for instance, where popular racism
43 is strongly hostile to migrants, to black people and to gypsies, rather

than to Jews, and where the Front National tries constantly to instill 1
anti-Semitism. 2
More generally, there is still a real distance between infrapoliti- 3
cal and political racism, and this means that racism is not so much 4
a widely extended ideology offering people a general framework in 5
which to interpret their own lives and personal experiences, but rath- 6
er a set of prejudices and practices that are rooted in these concrete 7
lives and experiences, and which could possibly evolve. 8
In the present state of things, the development is dominated by 9
a process of populist fusion in which popular affects and political 10
discourse converge, but which, paradoxically, protects our societies 11
from extreme and large-scale racist episodes. However, populism is 12
never a stable phenomenon and is always potentially open to more 13
frightening processes. 14
15
16
The diversity of European racist experiences 17
18
In contemporary Europe, our general analysis does not apply every- 19
where in the same way. Many factors intervene, which do not invali- diversifying factors 20
date our global hypothesis, but which oblige us to introduce much 21
more diversified images. 22
Some are related to the social history of each country, to its indus- 23
trialization, or to the making of its working-class movement. Some 24
are related to its political history, to the making of its state, institu- 25
tions and political system, and, possibly, to its dictatorial or totalitar- 26
ian recent past. Some also deal with the specificity of its culture and 27
national identity, and with its international past. Countries that have 28
experienced colonization and decolonization, or that have to face do- 29
mestic tensions due to what many nationalist actors and intellectu- 30
als have called ‘internal colonialism’ differ between themselves, and 31
from countries that are not concerned with these issues. For many 32
years, some European countries have experienced the presence of 33
migrants who have been attracted by agriculture and industry, or 34
who came for other reasons, including political ones. Others, like 35
Italy, are only now discovering this phenomenon. 36
The list of factors of this kind could certainly be extended, but the 37
most important thing is to see that they each affect at least one of the 38
three basic elements of our global analysis. The latter insists on the 39
twofold idea of a process in which, in the first place, industrial society 40
breaks down, the egalitarian state enters into crisis and the nation be- 41
comes paralysed in differentialist and defensive terms; and in which, 42
secondly, these three elements are increasingly dissociated. The pat- 43

1 tern of this process of destructuration and dissociation depends on

2 the various capacities of resistance to decline or crisis of each basic
3 element, and consequently on the various factors listed above.
4 Germany In Germany, for instance, industrial society adapted to the change
5 more efficiently than elsewhere. Trade unions, and mainly the DGB,
6 maintain a much higher capacity for action and bargaining than most
7 of their counterparts in the world. Before the fall of the Berlin Wall,
8 West Germany had a state and a political system which seemed less
9 affected by the crisis than other countries, and it is only recently, with
10 the huge price of the reunification with East Germany, that fiscal and
11 political problems developed and took on acute forms. At the same
12 time, the third element of our general analysis, the national issue,
13 appears as a crucial topic. Racist and neo-Nazi violence, and the ex-
14 tension of skinhead groups, express primarily symbolic and concrete
15 difficulties in implementing national unification, and are particu-
16 larly important in the former East Germany, where immense social
17 tensions and fears for the future are interpreted within the category
18 of nation. The centrality of this issue is also important in Austria. In
19 these two countries, the experience of the 1930s and 1940s informs
20 present political debates, and references to a national culture and
21 identity are so significant that theoretical priority should be given to
22 the national issue. The strength of popular and political anti-Semi-
23 tism in these countries reinforces this point; it strongly supports the
24 hypothesis of anti-modern attitudes linked to a traditional national-
25 ism, or to its revival due to the economic crises that transform social
26 demands into nationalist and racist attitudes.
27 Italy In Italy, to introduce a different case, the decline of industrial so-
28 ciety and the crises of trade unions are obvious, but they do not con-
29 stitute the main problem. In this country, national unification came
30 late, and localism or regionalism are strong, but they do not consti-
31 tute the heart of the problem. Analysing the emergence of racism
32 in Italy, interest must focus on the crisis of the state, of institutions
33 and of the political system, which is expressed by the recent electoral
34 successes of the Leagues in the northern part of the country, by the
35 incapacity of the state to deal with the mafia, and by the renewal of
36 debates concerning the mezzogiorno. Italy has long been a country of
37 emigration, and is just discovering that it has now become a country
38 of immigration. The first expressions of racism should not be overes-
39 timated. The Leagues are much more populist than racist, and con-
40 crete discrimination and acts of violence are not so frequent. When
41 they appear, they express a will for the economic inferiorization of
42 black or Arab migrants; they are not strongly linked to a cultural
43 and differentialist affirmation. The possible extension of the racist

phenomenon, at least at the political and ideological level, should be 1

analysed in terms of the crisis of the state and the political system. 2
This implies paying special attention to intellectual and political ac- 3
tors, who in Italy sometimes have a paradoxical role: by importing, 4
mainly from France, the issue of anti-racism in a context of weak 5
forms or racism, and by developing the image of a differentialist rac- 6
ism. While the main popular expressions are inegalitarian, they are 7
perhaps creating a self-fulfilling prophecy. 8
In other countries such as France, Belgium or the United 9
Kingdom, there is a temptation to use as a starting point for analysis 10
the decay of industrial society and the decline of the working-class 11
movement, one consequence of this being that migrants are defined 12
less as workers and more as members of religious or ethnic commu- 13
nities, even if the very existence of these communities may be over- 14
estimated. But French, Belgian and British experiences deserve in 15
fact an analysis that is directly three-dimensional and that gives equal 16
importance to the decomposition of industrial society, to the crises of 17
the state and institutions, and to the national issue. Let us add that, at 18
least in the Belgian and British cases, the unit for analysis of racism 19
should not be the whole country, but smaller entities, so that differ- 20
ences between, for instance, Scotland and England, or Flanders and 21
Wallony could be seriously taken into account: English nationalism, 22
for example, is much closer to xenophobia and racism than Scottish 23
nationalism. 24
There are therefore considerable differences between countries, 25
but these do not fundamentally challenge our global analysis. Each 26
national experience must be approached in its three-dimensionality, 27
even if, depending on the country, it is better at the outset to focus 28
thinking on only one or other of the three basic elements in our argu- 29
ment. In any event, it is effectively the image of the dissociation of dissociation of 30
these three elements – society, the state and the nation – which is the society, state and 31
origin of the spread of racism. nation 32
33
34
References 35
36
Balibar Etienne, and Wallerstein, Immanuel (1988) Race, classe, na- 37
tion, Paris: La Découverte. 38
Barkan, Elazar (1992) The Retreat of Scientific Racism, Cambridge: 39
Cambridge University Press. 40
Barker, Martin (1981) The New Racism, London: Junction Books. 41
CCE (1989) Eurobaromètre: L’opinion publique dans la Communauté 42
Européenne, Brussels: Commission des Communautés Européen 43
nes.

1 Gellner, Ernest (1983) Nations and Nationalism, Oxford: Blackwell.

2 Pettigrew, Thomas, and Meertens R.F. (1993) ‘Le racisme voilé: com-
3 posants et mesure’, in Racisme et Modernité (under the direction of
4 M. Wieviorka), Paris: La Déceouverte.
5 Taguieff, Pierre-André (1988) La force du préjugé, Paris: La Découverte.
6 Touraine, Alain, Wieviorka, Michel, and Dubet François (1987) The
7 Working Class Movement, Cambridge: Cambridge University Press.
8 Wieviorka, Michel (1991) L’espace du racisme, Paris: Seuil.
9 Wieviorka, Michel (1992a) ‘Racism and modernity’, paper pre-
10 sented at the Congress of the American Sociological Association,
11 Pittsburgh.
12 Wieviorka, Michel (1992b) ‘Analyse sociologique et historique de
13 l’antésimitisme en Pologne’, Cahiers Internationaux de Sociologie,
14 vol. 93, pp. 237-49.
15 Wieviorka, Michel (ed.) (1992c) La France raciste, Paris: Seuil.
16 Wilson, William J. (1978) The Declining Significance of Race, Chicago:
17 University of Chicago Press.
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Changing the boundaries of citizenship: 3
4
the inclusion of immigrants in democratic polities*
5
6
Rainer Bauböck
7
8
9
10
11
12
Normative political philosopher Rainer Bauböck has been a leading world
13
scholar on citizenship issues for over twenty years. This article is the last 14
chapter of his edited book from 1994, From Aliens to Citizens: Redefining the 15
Status of Immigrants in Europe. He suggests a threefold typology of mem- 16
bership in contemporary liberal states: territorial sovereignty, nominal citi- 17
zenship and social membership. Bauböck is one of the very first European 18
thinkers to argue forcefully that a substantial improvement of the legal inte- 19
gration of immigrants can be achieved by combining ‘residential citizenship’ 20
for foreigners with optional naturalisation as well as the toleration of dual 21
citizenship. To this day, these arguments are hotly debated, though not as 22
much as they were during the early 1990s. In any case, this article remains 23
an important contribution to European citizenship and migration studies. 24
25
26
And the resident alien feels himself equal to the citizen and the citi- 27
zen to him (Plato on the “anarchic temper” of democracy, Republic, 28
VIII, 562e) 29
30
31
Who is included in democratic legitimation? 32
33
Every scientific discipline has its own core question. For the philo- 34
sophically oriented branches of social science, their core questions 35
seem to be unanswerable in the sense of finding a definite solution 36
that will be accepted by all rational participants in scientific dis- 37
course. At the same time, these questions appear to be unavoidable 38
and capable of stimulating never-ending debates that reassure social 39
theorists that there is, after all, a raison d’être for their disciplines. 40
I think that the core question for normative political theory has 41
been: What are the conditions for making political rule legitimate? legitimating 42
A general answer that has strongly prevailed, at least since Thomas political rule 43

276 rainer bauböck
1 rational consent Hobbes, is the following one: Political rule must be of a kind that
2 those who are subject to it could rationally consent to being ruled in
3 this way. Theories that have tied political legitimacy to democratic
4 rule have specified further conditions, such as the following ones:
5 The collective of all subjects must be regarded as the ultimate sov-
6 ereign in a political system. Subjects are entitled to elect their politi-
7 cal representatives, counting every vote as one and one only. They
8 can appeal to an independent judiciary against the unlawful exercise
9 of political power and enjoy a right of resistance against illegitimate
10 rule. It is conditions like these that mark the transformation of sub-
11 jects into citizens. When elaborating such answers we will find that
12 the core question can be split into two separate ones: How can po-
13 litical rule be made legitimate? and Who are those towards whom it
14 must be legitimated? It appears that most contemporary democratic
15 theories regard the “how” as much more important than the “who”.
16 There are two reasons for this unequal emphasis. First, contempo-
17 rary liberal democracies differ strongly in their constitutional struc-
18 tures such as in their legal traditions, electoral systems or separation
19 of powers; this variety stimulates the comparison of the virtues and
20 disadvantages of different solutions to the problem of democratic
21 legitimacy. In contrast, the ranges of inclusion appear to be rather
22 similar in all these political systems and minimum standards are
23 much more firmly established in this regard. The exclusion of blacks
24 or women from the franchise, or a decision to deprive an ethnic mi-
25 nority of its citizenship, would be clearly regarded as unjustifiable
26 today.1 The second reason is that most people would probably agree
27 that there is a straightforward answer to the who-question: All those
28 who are affected by political decisions, and who are able to participate
29 in the legitimating activities, should be included in the democratic
30 polity.
31 Of course there are some significant exceptions where contempo-
32 rary democracies seem to fail by this principle. It is by examining the
33 reasons for these exceptions that we can best distinguish the inher-
34 ent limitations of democratic inclusion from unjustified exclusion. I
35 will group these exceptions into three; (1) external exclusion, (2) in-
36 ternal exclusion and (3) internal exclusion with reference to external
37 affiliation.
38
39 external exclusion (1) Citizens of state A may be strongly affected by political decisions
40 taken by state B and legitimated only towards B’s citizens. The wag-
41 ing of an offensive war, occupation and colonization of another coun-
42 try are the most blatant cases where, by definition, the victimized
43 population is excluded from legitimation of the action (although

many of these actions have been labelled by the aggressor’s pro- 1

paganda as liberating or civilizing missions). In other examples, a 2
damage inflicted upon the population of A may be the by-product 3
of some action on the part of B, which is less intended to harm A 4
than to serve B’s interests. Take, as an illustration, the depletion of 5
natural resources at the detriment of some neighbouring country 6
(e.g. when the water of a river is diverted or used at the expense of 7
those living downstream at the other side of the border). Regarding 8
environmental pollution across international borders it is not only 9
the neighbouring areas which suffer, but generally the damage is 10
even greater among the population of the country from which the 11
emissions originate. Politically speaking, the former are nonetheless 12
in the worse position because they and their representatives are for- 13
mally excluded from controlling what affects them. In these cases 14
interstate treaties, rules of international law or pressure may help to 15
restrain the ruthless pursuit of a national policy which does not con- 16
sider the effects on populations beyond the border. However, a fun- 17
damental difference remains between such remedies and the kind of 18
popular involvement which is regarded as essential for democratic 19
legitimacy. 20
Yet another problem of external exclusion results from the opera- 21
tions of a global economy. In the 1980s monetarist policy of Western 22
states pushed up interest rates with the effect of reducing the ability 23
of highly indebted countries in both Eastern Europe and the so-called 24
Third World to pay back credits, forcing many of them to adopt se- 25
vere austerity policies. One could argue that in this example, gov- 26
ernments of debtor nations had agreed to terms of contract which 27
included such a risk. Yet this objection does not fairly represent the 28
unequal balance of power by virtue of which creditor nations can 29
unilaterally influence the capacity of debtors to comply with their 30
obligations. 31
Many more examples could be given of policies that strongly af- 32
fect populations which are excluded from democratic legitimation 33
simply because they live outside the territory of the state which de- 34
termines and controls this policy. The general problem is that of the 35
disjunctures of globalization (Held, 1991; Held and McGrew, 1993). 36
The territorial ranges of ecological systems do not coincide with the 37
boundaries of states and modernization makes economic systems 38
increasingly transnational or even global. The modern bureaucratic 39
state, however, is solidly tied to a territory within which it claims a 40
monopoly of legitimate violence. Democratic legitimation therefore 41
also refers to a territorially bounded population. Involving the popu- 42
lations of other countries in the legitimation of national political de- 43

278 rainer bauböck
1 cisions can normally only be achieved indirectly within a framework

2 of peaceful international cooperation.
3
4 internal exclusion (2) However, democratic legitimation may in certain ways also ex-
5 clude parts of the population living in the territory of a state. In
6 pre-modern democratic constitutions, free citizens were generally a
7 minority of the population. Slaves and women were not considered
8 to be members of the polity. Nineteenth-century democracies still
9 maintained gender and property requirements for active citizenship2
10 and racist exclusion was widespread. In contemporary liberal democ-
11 racies three groups remain internally excluded: minors, the severely
12 mentally handicapped and convicts. There are two significant shifts
13 in the patterns of justification from pre-liberal to contemporary ex-
14 clusions.
15 Firstly, pre-liberal requirements for citizenship referred to gener-
16 alized social conditions for individual autonomy which were seen as
17 preconditions for the formation of an independent judgement about
18 the common good and the interests of the state. Paupers, workers
19 and women had to be excluded from full citizenship because their
20 economic dependency and lack of education presumably prevented
21 them from developing that kind of judgement. This was clearly also
22 a self-defeating ideological argument. How could the privileged class
23 of male property owners be trusted to develop an unbiased view of
24 the common good? Isn’t it more reasonable to assume that they
25 would rather defend their own interests against those of excluded
26 groups? Since then it is not only the argument but also social condi-
27 tions which have changed so that the argument has lost whatever
28 force it once might have carried. On the one hand, almost everybody
29 receives nowadays that kind of elementary education which may be
30 said to be necessary for an active citizen and, on the other hand, a
31 broad middle class now has to rely on wages and salaries for their
32 income and on state bureaucracies for their social security. The capi-
33 talist welfare state has thus created a new social basis for including
34 broader populations into citizenship by generalizing education as
35 well as economic dependency. Any remaining citizenship disabili-
36 ties are seen to result from a lack of relevant mental capacities and
37 moral qualities of individuals rather than being attributed to them as
38 permanent members of ascriptive social groups. Minors are automati-
39 cally included on reaching their age of majority and convicts may
40 regain the status of full citizenship when being released from prison.
41 Mentally handicapped persons may remain permanently disenfran-
42 chised but this is justified with regard to a minimum of dialogic ca-
43 pacities that are essential for participating in political deliberation
(see Ackerman, 1980: 78-80).

Secondly, and I think more importantly, those who remain dis- 1

enfranchised are no longer excluded from citizenship. Minors, men- 2
tally handicapped persons and even criminal convicts are citizens in 3
the latter sense even though they may be excluded from the vote. 4
Basic mental capacities and moral qualities are not required for 5
membership in the polity but serve as criteria for the distribution of 6
the core rights of political participation within the polity. Citizenship 7
is acquired at birth rather than at the age of majority and generally 8
it cannot be taken away by the state or abandoned by citizens them- 9
selves as long as they live in the territory. From a liberal democratic 10
point of view, the status of citizenship, by which a state recognizes 11
an individual as its member, is not a formal legal concept lacking any 12
particular content;3 it implies substantial rights to protection, as well 13
as those against interference, by the state. Democratic legitimation 14
is not confined to the activity of political participation but rests on 15
this more comprehensive bundle of rights. For a liberal conception, 16
in contrast with the republican tradition of Aristotle, Rousseau or 17
Hannah Arendt, the inclusion of the inactive and even the incom- 18
petent as equal members in the polity is a basic achievement of con- 19
temporary democracy. This is a guarantee against the degeneration 20
of democracy into the rule of a self-proclaimed enlightened elite. 21
Modern liberal citizenship therefore emerges from a dual movement 22
of (a) turning the narrow privileges of active political participation 23
into general rights and (b) enriching the generalized condition of 24
protected subjecthood with the enjoyment of basic rights. 25
26
(3) There is, however, one kind of persistent internal exclusion which 27
can only be justified by arguments for external exclusion. This is the 28
peculiar status of resident aliens. Their position in contemporary de- 29
mocracies is a paradoxical one. They are clearly affected by political 30
decisions in much the same way as citizens. Provided that they speak, 31
or have learned to speak, the language of their country of residence, 32
they are not different in their general capacities that quality them for 33
citizenship. They do, in most cases, enjoy fundamental rights, such 34
as equal rights in court, civil liberties, social rights to elementary 35
education and equal employment-related benefits of social security. 36
Their rights thus go considerably beyond universal human rights, 37
however, they are granted to them as residents rather than as citi- 38
zens. On the one hand, this convergence between the rights of resi- 39
dents and of citizens demonstrates that the basic democratic norm of 40
legitimation applies to a resident population rather than only to those 41
individuals who are formally recognized as members of a polity. On 42
the other hand, why are there still so many significant distinctions 43

280 rainer bauböck
1 between the status of aliens and of citizens, especially concerning

2 the right to permanent residence and voting rights? Do not states of
3 immigration with a large and growing disenfranchised alien popula-
4 tion fail to meet the norm of inclusion which characterizes liberal
5 democracy? Yet the charge that these distinctions of rights and status
6 between citizens and foreigners lack democratic justification raises
7 some additional questions which might be more difficult to answer.
8 Should one draw from this the consequence of automatically natu-
9 ralizing all alien residents? Or should one go even further and aban-
10 don the formal concept of citizenship altogether? What do we need
11 a formal status of citizenship for if all residents already enjoy equal
12 rights? The dynamics of modernization implies a long-term trend
13 towards increasing international migration. Assuming that liberal
14 democracies will be preferred targets and should be relatively open
15 to new admissions, they will have to rethink their principles for the
16 allocation of rights and membership among citizens and foreigners.
17 This will inevitably break up national frameworks which have been
18 used to define the boundaries of membership. But what should re-
19 place them? These are some of the questions I aim to address in this
20 paper.
21
22
23 Limits for inclusion: individual choice, political allegiance
24 and societal membership
25
26 reasons for Two kinds of reasons might be given for the substantial curtailment
27 exclusion of rights and formal exclusion of foreign residents. The first one
28 is that this alien status is essentially a chosen one. Immigrants are
29 supposed to have come of their own free will and to know that they
30 will not be regarded as equal citizens.4 Their discriminated status
31 as aliens is the result of a social contract by which they gained the
32 desired admission. Furthermore, many who could have naturalized
33 have not chosen this option and thus seem to voluntarily accept their
34 exclusion and discrimination. This line of argument does not apply
35 to those who have come as refugees rather than as voluntary immi-
36 grants. They have not chosen their fate and have been deprived of
37 their rights as citizens of their home countries. If there is a reason-
38 able presumption that the situation causing their flight will not per-
39 sist for long, they will need temporary protection and assistance in
40 order to return to their homes. But if they need more permanent
41 protection, an appropriate answer to their plight is to offer them the
42 citizenship of the country which has granted them asylum.5
43 choice
However, the argument referring to choice is difficult to accept,

even for voluntary migrants. Most foreign citizens are not given the 1
option of naturalization. Admission procedures in ordinary natural- 2
izations are normally discretionary - the final decision is taken by the 3
naturalizing state, not by the applicant.6 Now suppose, for the sake 4
of argument, that naturalization became fully optional, i.e. available 5
upon request after a relatively short time of residence and without 6
further conditions attached.7 Even then, the question remains why 7
full rights for resident immigrants should depend upon their opting 8
for legal membership. Native citizens who enjoy these rights have 9
not chosen to be members, but have acquired their status at birth 10
and they are generally denied an option to renounce it while stay- 11
ing in the country. This indicates that, from the perspective of a lib- 12
eral democratic polity, inclusion seems to be more important than 13
choice. If a substantial number of the population is excluded from 14
the polity because of their foreign citizenship, this creates a problem 15
for the legitimacy of political decisions – even if this exclusion were 16
a voluntary one. Nevertheless, migrants may have special reasons 17
not to choose naturalization which ought to be taken into account. 18
Intuitively, it seems obvious that forcing a migrant to adopt a citizen- 19
ship she or he does not want cannot be compared to the automatic 20
attribution of citizenship at birth. So the balance between inclusion 21
and choice should be a different one for native citizens and migrants. 22
However, this argument does not provide a justification for any kind 23
of discrimination. The question which I shall take up again in the 24
concluding section is rather: How different should the status of citi- 25
zens and resident aliens be in terms of rights in order to make opting 26
for naturalization a meaningful choice? 27
The second type of reasoning for maintaining a clear line between political allegiance 28
foreigners and citizens emerges from the perspective of the receiv- 29
ing state. The argument is that this line is constitutive for the pol- 30
ity itself and thus cannot be blurred by some democratic principle. 31
Democracy would become self-destroying if the imperatives of legiti- 32
mation made it impossible to maintain the boundaries of the polity. 33
In the framework of Carl Schmitt’s politics of friend and foe, and 34
Thomas Hobbes’s view of international relations as a latent state of 35
war, it is quite plausible to deny foreigners essential rights of citi- 36
zenship as well as the optional access to naturalization. The reason 37
for this is that their allegiance and obligations tie them to another 38
sovereign. It may be in the interest of a state to encourage immigra- 39
tion (if there is a strong demand for labour), it may even be expedi- 40
ent to naturalize immigrants in great numbers (if there is a lack of 41
soldiers). However, admission to the polity must remain under the 42
control of the receiving state in the same way as immigration8 and 43

282 rainer bauböck
1 the essential qualifying criterion for naturalization is not the period

2 of residence but a credible change of loyalty.
3 In this view, the boundaries of a polity do not relate to a territory
4 or to the population living there, but define mutually exclusive sets
5 of persons who are citizens or subjects of sovereign states. These
6 boundaries emerge in interaction and confrontation with other poli-
7 ties by identifying those who cannot be included because they belong
8 somewhere else. Such a definition of external boundaries is not ar-
9 bitrary and can be well combined with a broad internal inclusion.
10 It need not fall back on Joseph Schumpeter’s dictum that a general
11 theory of democracy must “leave it to every populus to define himself”
12 (Schumpeter, 1950: 245).9 At the same time, it postulates that the
13 democratic norm of inclusion ceases to apply where another sover-
14 eign state has a prior claim to regard some individual as its member.
15 A foreigner may live permanently in the territory of state A, but, as
16 a citizen of B, all claims of democratic legitimation which she or he
17 might raise are addressed to that state. Such membership is not a
18 social relation which might become weaker as time passes but a le-
19 gal one that retains its binding force over time and might even be
20 transferred to the immigrant’s children. As above with the argument
21 referring to voluntary choice, this argument about the mutually ex-
22 clusive nature of sovereignty does not apply to refugees and stateless
23 immigrants. But it is still the conventional wisdom which supposedly
24 justifies the legal discrimination of foreigners and the discretionary
25 procedures of naturalization.
26 I believe that this view is at odds with modern liberal conceptions
27 of democracy. It is also incapable of accounting for the dynamics of
28 the extension of legal rights for long-term resident foreigners, for the
29 tendency to recognize that immigrants may acquire a moral entitle-
30 ment to be naturalized and, finally, for a trend in Western Europe to
31 tolerate dual citizenship. Just as I have acknowledged that a certain
32 differentiation of status between foreigners and citizens may be jus-
33 tified within a framework of choice, I am also inclined to support the
34 idea that in an international system with a multiplicity of states, poli-
35 ties have to be externally bounded. However, it is far from obvious
36 that these boundaries have to be mutually exclusive in the way that
37 territorial ones are. If it is not membership in a different polity which
38 sets the external limits for the range of inclusion in democracy, what
39 could then determine these limits?
40 societal membership I want to defend the proposition that the basic standard for inclu-
41 sion in a liberal democratic polity is based on a specific notion of so-
42 ciety – the outlines of which can be determined by applying the norm
43 of democratic legitimacy to the social instead of the political sphere.

From the perspective of individuals, a society in this sense comprises 1

all whose social position durably relates them to a certain state so that 2
they depend on this state for their protection and rights. Seen from 3
the perspective of a state, a society is the basic ensemble of popula- 4
tions permanently affected by its collectively binding decisions. 5
We might characterize this as the political concept of society. It 6
contrasts with the narrower notion of the polity, on the one hand, and 7
with the wider sociological concept of society as an open system of 8
interaction and communication, on the other. A polity only includes 9
citizens, i.e. those whose state membership is of a political rather 10
than a social nature. The boundaries of polities can be controlled by 11
the political decisions made on membership so that individuals who 12
are not admitted, or who are excluded, will clearly not be members 13
regardless of their social relation to the state. The boundaries of so- 14
ciety are not subject to political decision in this way but they result 15
from the exercise of political power. Liberal democratic legitimation 16
requires inclusion of the whole society in the sense that the distribu- 17
tion of rights must correspond to the impact of political power and in 18
the sense that the polity must be genuinely open for the admission 19
of everybody who can claim membership in society. As we shall see 20
later on, this does not bring with it a total equality of political status 21
and rights of citizenship throughout society. 22
In contrast with the world economy of modern capitalism and 23
global flows of information, the global political system remains seg- 24
mented into a multitude of states. This is why there is also a mul- 25
tiplicity of societies which relate to these states. However, while 26
the political image of societies (in contrast with a sociological or 27
economic one) is always one of bounded populations, the shape of 28
these boundaries remains to be determined. I will defend the idea 29
that from the perspective of a system of liberal democratic states they 30
are permeable and overlapping, and they include foreign residents 31
in the territory as well as citizens, and even some foreigners living 32
abroad. Nevertheless, political societies are not unbounded and soci- 33
etal membership will set the limits within which the norm of inclu- 34
sion applies. 35
36
37
Orders of membership: territorial sovereignty, nominal 38
citizenship and societal membership 39
40
Before discussing the norms that can be applied to determine the sta- 41
tus and rights of immigrants in receiving societies in more detail, let 42
me first take a bird’s-eye view of the kinds of orders of membership 43

284 rainer bauböck
1 that states produce among populations. As I use the term here, an

2 order does not refer to an internal structure (such as the hierarchical
3 or egalitarian features of a political system) but simply to the sorting
4 of individuals into different sets which are characterized by their re-
5 lation to a state. The following diagram graphically symbolizes three
6 different types of orders for three states A, B and C.
7
8
9
10 A B A B A B
11
12
13
14
15 C C C
16
17
18 The simplest order is that of territorial sovereignty. Each state rules
19 a particular stretch of land and everybody who happens to be in that
20 state’s monopoly land is, in an elementary way, subject to that state’s monopoly of
21 of violence violence. States generally also claim the right to make laws that are
22 binding for anybody who is in the territory even for a short time.
23 Exceptions do exist but they are few and well-defined. Apart from sit-
24 uations of military conflict, these exceptions result from legal norms
25 or coordinated actions of the international community of states rather
26 than from uncoordinated policies of individual states. Foreign diplo-
27 mats enjoy a special immunity that partially exempts them from the
28 rules of territorial sovereignty as they apply to persons. Exceptions
29 with regard to unique sovereignty over a territory may occur after
30 a war when one or several victorious powers occupy the aggressor
31 state (as was the case with Germany and Austria after the Second
32 World War), or when an embattled territory is temporarily put under
33 the authority of the United Nations. Another unique exception is the
34 international status of Antarctica which, however, is due to the fact
35 that there are no permanent resident populations in this territory.
36 Apart from this continent, the whole land mass of the globe is now
37 divided into mutually exclusive state territories and all human beings
38 relate to the state of their present abode as their territorial sovereign.
39 A substantial body of international law has attempted to resolve any
40 remaining ambiguities such as that concerning the status of persons
41 on board of ships in international waters.
42 The order that territorial sovereignty produces can thus be called
43 complete and discrete. I define these two features in the following

way: Completeness means that everybody is at any point in time sub- 1

ject to the territorial sovereignty of a state; discreteness implies that 2
nobody is subject to more than one state simultaneously. Such an 3
order can be represented, as in the above diagram, by a political map 4
of states without stretches of no-man’s land or water between them. 5
However, in contrast with such a geographical representation, this 6
order can be highly volatile. We can define the stability of an order 7
as the average probability that an individual who is classified as a 8
member of set A in t1 will be classified as a non-member of this set 9
in t2. Individuals who cross international borders will be subject to 10
different territorial sovereigns before and after this move. 11
The second kind of order is that of citizenship in the sense which 12
is also called nationality. I will use the term “nominal citizenship” nominal citizenship 13
when I want to distinguish it from the substantial aspect of citizen- 14
ship as a bundle of rights and obligations that individuals hold in 15
their relation to a state. Citizenship in the former sense identifies 16
persons in the international arena by using the name of a country 17
in a manner similar to the use of family names in social interaction 18
outside the family. Both indicate that an individual belongs to a state 19
or family, but the name also belongs to the individual; it is a personal 20
attribute which the individual has the right to carry.10 If we put in- 21
dividuals into sets, first, with regard to their subjection to territorial 22
sovereignty and then once more with regard to their citizenship, we 23
shall find that the sets broadly overlap but are usually not identical. 24
Foreign residents will be included in the former but excluded from 25
the latter, while the reverse is the case with emigrant citizens. 26
Apart from this incongruency, the above-mentioned characteris- 27
tics clearly distinguish the two kinds of orders from one another. 28
Firstly, the order of nominal citizenship is more stable than that of 29
territorial sovereignty, secondly, it is neither discrete nor complete. 30
Citizenship is acquired at birth and most people never change it dur- 31
ing their lives. Citizenship is not an ascriptive feature like gender 32
or “race” where the immutability of societal membership is empha- 33
sized by relating it to innate differences of human bodies, but it is 34
still intended to last for life. All states’ rules for naturalization em- 35
phasize this temporal stability by inhibiting frequent change. This 36
can be achieved by residence requirements, by extended waiting pe- 37
riods prior to naturalization, by an oath of allegiance which is meant 38
to express commitment for an indefinite future and by denying or 39
delaying expatriation even after emigration. There are important 40
political reasons for enhancing stability. The exercise of state power 41
that turns people into subjects is spatially constrained by the range 42
of territorial sovereignty, but it does not require all who are liable to 43

286 rainer bauböck
1 obey the law to be bound to the state by any lasting ties of member-
2 ship. However, any system of government also calls for a durable
3 relation between the state and those who can be identified as subjects
4 in a narrower sense of the term. Obligations that states impose on
5 their subjects can only be enforced when the relation is relatively
6 stable. Conscription, collecting taxes or enforcing criminal punish-
7 ment require that people can be identified and that they cannot evade
8 their obligations by simply moving somewhere else. There is also a
9 strong democratic argument in favour of stability. Citizens who par-
10 ticipate in political deliberation or who elect representatives who are
11 to take collectively binding decisions need a common temporal per-
12 spective that reaches back into the past and forward into the future.
13 They cannot form reasonable judgements on political matters unless
14 they share some experience with past decisions and given institu-
15 tions of their state. Furthermore, they must also share a perspective
16 of knowing that they themselves, or their children and others close to
17 them, will be affected by the decisions they support. In contrast with
18 republican thinking, liberal democracy allows for a wide diversity of
19 interests that can be legitimately expressed in political choices. While
20 in such a polity common interests may be reduced to quite a small
21 number, there must nevertheless be a common time-horizon for all
22 interests that are put forward in the process of political deliberation.
23 stateless While the nominal order of citizenship is more stable over time
24 than that produced by territorial sovereignty, it is at the same time
25 less perfect with regard to the criteria of discreteness and complete-
26 ness. Individuals may be multiple citizens or stateless. These phe-
27 nomena are widely perceived as irregular. Yet, in contrast with a
28 breach of the principles of territorial sovereignty, such irregularities
29 generally do not cause conflicts between states and they emerge from
30 the very rules that guide the allocation of nominal citizenship in the
31 international system of states. State sovereignty ends where the ter-
32 ritory of a neighbouring state begins, but it does not necessarily end
33 where another state claims an individual as a member. Each state
34 reserves the right to set up its own rules for the acquisition and loss
35 off citizenship as a core expression of its sovereignty. Statelessness
36 and multiple citizenship can thus emerge from a conscious policy of
37 ignoring the rules of another state or as an unintended side-effect of
38 rules applied separately by each state involved.
39 Let me give a few examples. Political refugees who want to natu-
40 ralize in their state of asylum are sometimes denied voluntary ex-
41 patriation by their state of origin or they are unwilling to submit to
42 the procedures for obtaining it from the authorities of the persecut-
43 ing state. Western democracies normally accept that the person will

become a dual citizen in this case. Dual citizenship may also result 1
from rules of optional or automatic admission that are applied by 2
some Western European states to foreigners born in the country 3
when attaining their majority or to those who have been married to a 4
citizen for a certain time. Most cases of dual citizenship emerge from 5
birth in mixed marriages if both countries involved apply ius sangui- 6
nis from both parents11 or result from a simultaneous application of 7
ius soli by the state of birth and ius sanguinis by the parents’ state. In 8
contrast to dual citizenship, statelessness may be the intended effect 9
of a policy of disenfranchising an ethnic minority or depriving it of 10
any kind of state protection. Another origin of statelessness is the de- 11
naturalization of emigrants regardless of whether they have already 12
acquired their host country’s citizenship or not. Finally, statelessness 13
may also result from voluntary expatriation. The right to a national- 14
ity has been established in Article 15 of the Universal Declaration of 15
Human Rights of 1948 and many states have signed the 1961 United 16
Nations Convention on the Reduction of Statelessness. I think that 17
normative arguments for avoiding statelessness are strong enough 18
to warrant making the emigrants’ right to expatriation conditional 19
upon another state’s willingness to naturalize them. However, the 20
rules of international law still do not provide sufficient guarantees 21
for preventing the re-emergence of these areas of no man’s land in 22
the international order of citizenship. 23
What I have called the political concept of society points to a third societal membership 24
kind of order. In contrast to the two preceding ones, this order of 25
societal membership is not formalized in the legal relations between 26
individuals and states. Its contours emerge, on the one hand, from 27
sociological observations about the role the state plays in regulating 28
the conditions for the individual’s life prospects and opportunities. 29
On the other hand, the order is constructed from a normative point 30
of view in order to answer the question posed in this paper: Who 31
can claim a right to inclusion in a liberal democratic polity? As illus- 32
trated in the diagram above, such an order resembles that of nominal 33
citizenship because there are overlapping areas, only that here these 34
are much more extensive. Individuals can be members of more than 35
one society simultaneously without this fact being reflected in mul- 36
tiple citizenship. At the same time, the order of societal membership 37
shares the feature of completeness with that of territorial sovereign- 38
ty. There are hardly any individuals for whom we cannot identify at 39
least one state to which they are socially tied. Statelessness is not a 40
condition of cosmopolitan detachment but just on the contrary; it is a 41
status of extreme dependency upon the protection offered by specific 42
states without the formal entitlement to claim that protection. 43

288 rainer bauböck
1 Within the state territory international migration makes the num-

2 ber of societal members larger than the population of nominal citi-
3 zens, but smaller than the aggregate of everybody physically present
4 at a certain point in time. Resident foreigners have to be regarded as
5 members of society but individuals who are passing through on their
6 way to another destination or who have come for a short temporary
7 stay need not be counted.12 Outside the state’s territory the number
8 of societal members may be either larger or smaller than that of emi-
9 grant citizens. If a state adopts a policy of indefinite transmission of
10 citizenship by ius sanguinis, a third or later generation may still be
11 registered as citizens of the state where their ancestors have come
12 from without having any significant social ties to that country them-
13 selves. Conversely, an individual may have strong social ties to a state
14 or depend upon its protection without living there or being one of its
15 citizens. Two relevant examples may be mentioned as an illustration.
16 The first is ethnic diaspora minorities who regard a foreign state as
17 their national homeland from which they expect protection of their
18 rights. Frequently, these rights will include that of being admitted
19 to the territory of that state. Some states recognize these claims and
20 treat such minorities as “ethnic citizens” abroad without nominal
21 membership. Germany and Israel are extreme cases who grant their
22 co-ethnics not only a right of immigration but also immediate access
23 to nominal citizenship thereafter.13 A second example is that of fam-
24 ily members of immigrants who have stayed in the country of origin,
25 or of migrants who had to return there after a long residence abroad.
26 Maybe the most obvious case of societal membership of foreigners
27 who are neither citizens nor residents is that of second-generation
28 young people who were born in the country of immigration but were
29 turned into aliens by ius sanguinis and later had to return to their
30 parents’ country of origin, either because their parents demanded it
31 or because they had lost their residence permit.14
32 general indicator
In its temporal aspects the order of societal membership is cer-
33 tainly more stable than that of territorial sovereignty but need not be
34 as rigid as that of nominal citizenship. People can change their social
35 affiliations that tie them to a state several times during their lives
36 and, coming to a country where one takes up a permanent residence
37 does not imply a promise or commitment to stay there for good.
38 Societal membership does not strictly require a perspective which
39 reaches back into the past and forward into the future (as democratic
40 citizenship does). The time of residence is no more than a general
41 indicator for the consolidation of social ties.
42 Along the time axis, the transition from one societal membership
43 to another may follow different paths for different categories of mi-

grants. For some, emigration means dissolving their households in 1

their country of origin and dissociating themselves from that state. 2
This will be often true for refugees who have little hope of returning 3
but it may also be the case for some long-distance migrants who con- 4
sciously choose another state as their home for the rest of their lives. 5
For these people, migration does not generate an overlapping area 6
of societal membership. They simply cross a societal border and a 7
territorial one simultaneously. The number of migrants of this kind 8
is rapidly diminishing and has probably always been overestimated 9
for most migration flows. Even for the classical overseas labour mi- 10
gration from Europe to North America around the last turn of the 11
century the idea that most immigrants had simply burned the bridg- 12
es is a misconception. Immigration history is always written from 13
the perspective of the receiving country and if that country, more- 14
over, sees itself as a nation of immigrants, cyclical and return migra- 15
tion simply drop out of sight. For a second group of migrants the 16
time of dual societal membership may be a transitional period. They 17
leave family members behind and frequently contribute remittances 18
to their household budget; they visit their country of origin during 19
vacations or at least on the occasion of important family events such 20
as births, marriages, or deaths; they often also plan themselves to 21
return after having achieved a certain target in savings, when retiring 22
or when the economic and political situation has improved there. For 23
some, these plans may work out and their dual membership was a 24
temporary extension of their societal affiliation during a certain pe- 25
riod of their lives. If their stay abroad has been a prolonged one, they 26
will, nevertheless, normally also retain significant social ties to that 27
country after returning to their country of origin. Others may finally 28
bring all their close family into the country of immigration and cut 29
their ties to the society of origin after a long residence abroad. In this 30
case, the overlapping area forms a passage in a slow but unidirec- 31
tional shift of membership. An ever-growing number of migrants, 32
however, acquires a social status as dual members for the rest of their 33
lives, regardless of whether they stay or return. 34
35
36
Contractarian, libertarian, republican and nationalist 37
inclusion 38
39
I have said above that the political concept of societal membership 40
emerges not only from sociological observation but also from a nor- 41
mative perspective of inclusion in a liberal democratic polity. This liberal democratic 42
seems to lead into a circular argument where the norm of inclusion polity 43

290 rainer bauböck
1 first refers to the reality of spatially bounded societies whose bound-

2 aries are, then, defined by specifying to whom the proposed norm
3 should apply. I admit that some circularity of this kind appears to me
4 unavoidable. It mirrors the fact that transnationally mobile societies
5 do not only overlap but are also blurred at their margins. There is
6 thus always some latitude for the contestation of societal member-
7 ship which can only be decided by specifying normative criteria with-
8 in the context of a particular society. By contrasting the liberal demo-
9 cratic perspective with alternative ones, I hope to be able to show that
10 it is not quite so indeterminate as it might seem. No comprehensive
11 political ideology and system of political rule can do without a po-
12 litical concept of society that sets a standard for inclusion. However,
13 rival strands of political thought differ in how they construct their
14 respective concepts of societal membership.
15
16 (1) For a Hobbesian Leviathan the basic relation between individuals
17 and states is that of subjection to a territorial sovereign. However, as
18 I have already pointed out above, the dense web of obligations that
19 binds the subject to the sovereign does not necessarily include every-
20 body in the territory nor exempt all those living abroad. The question
21 is how those who are permanently obliged in this specific way can be
22 distinguished from those who are only temporarily subject to territo-
23 rial sovereignty. The most plausible answer to this is that anybody
24 subject by birth born within the territory has to be regarded as a subject by birth. Ius
25 soli has its roots in feudal and absolutist systems where the rule over
26 people is derived from ownership of the land. The basic idea about
27 the status of foreigners under the latter kind of rule is concisely ex-
28 pressed by Hobbes:
29
30 But he that is sent on a message, or hath leave to travel, is still
31 Subject; but it is, by Contract between Soveraigns, not be vertue of
32 the covenant of Subjection. For whosoever entreth into anothers do-
33 minion, is Subject to all the Laws thereof; unlesse he have a privilege
34 by the amity of the Soveraigns, or by speciall license (Hobbes, 1973,
35 XXI: 117).
36
37 John Locke’s reformulation of the social contract allows for a differ-
38 ent and somewhat more liberal interpretation that concedes a claim
39 to protection to foreigners and opens the door to voluntary natural-
40 ization but still emphasizes their exclusion, as foreigners, from the
41 commonwealth:
42
43

[F]oreigners by living all their lives under another government, and 1

enjoying the privileges and protection of it, though they are bound 2
even in conscience to submit to its administration as far forth as any 3
denizen, yet do not thereby come to be subjects or members of that 4
commonwealth. Nothing can make any man so, but his actually en- 5
tering into it by positive engagement, and express promise and com- 6
pact (Locke, 1956, VTII, §122: 62-63). 7
8
The status of foreign residents is defined as one of non-membership 9
in both cases. At the times of Hobbes and Locke, the very idea of 10
society as a conceptual unit for the study of human relations and in- 11
teraction, independent of a country’s political constitution, probably 12
made no sense. The notions of commonwealth or civil society refer, 13
in Locke’s words, to a “political society” constituted by the original 14
contract of which resident foreigners were clearly not seen to be party. 15
16
(2) A political theory which would tie societal membership even 17
more strongly to territorial sovereignty is the libertarian utopia of libertarian utopia 18
Robert Nozick (1974). Nozick’s world is one of minimal states whose 19
functions are reduced to exercising a territorial monopoly of vio- 20
lence. Nozick dismisses the idea of social compact (p. 131-132) and 21
replaces it with an “invisible-hand” explanation (p. 118-119) of how 22
such a state might come about from the rights of individuals to own- 23
ership, self-defence and free association for purposes of protection. 24
In contrast with an “ultraminimal state” whose monopolistic protec- 25
tive agency only protects clients who have purchased its services, a 26
minimal state protects everybody living permanently in a territory. 27
So resident foreigners cannot be excluded just because they have 28
never formally joined and this kind of protective association neither 29
has the right not to admit them as formal members if they wanted to 30
join. As far as Nozick’s extreme individualism allows for any concep- 31
tion of society at the level of states,15 the range of this society should 32
relate to all residents of a state territory. I refrain from speculating 33
how Nozick would define the status of transient migrants and tem- 34
porary residents. Generally speaking, his kind of theory would maxi- 35
mize inclusion with regard to territorial sovereignty while leaving 36
little scope for also taking the social affiliations that go beyond this 37
into account. More importantly, the deficiency of the theory is that it 38
achieves inclusion only at the expense of reducing the substance of 39
rights, which citizens expect to enjoy in a democratic state, to a bare 40
minimum. 41
42
43

292 rainer bauböck
1 republican (3) The republican tradition of political thought has emphasized ac-
2 conception tive participation by citizens in politics more than passive recipience
3 of protection by a state. Citizenship is seen as a set of obligations
4 more than of rights, as an office more than a status.16 Inclusion in cit-
5 izenship is not so much connected to territorial residence but to mu-
6 tual recognition within a community of equal members of the polity
7 who experience themselves to be the sovereign political authority. In
8 this approach, the order of citizenship seems to be the only reference
9 point to which the norm of inclusion can be applied. A republican
10 conception thus appears to come close to Schumpeter’s self-defining
11 populus. However, republican norms of inclusion would still not be
12 completely redundant. Firstly, they can specify certain features of a
13 desirable order of citizenship. Republican thought has always strong-
14 ly objected to multiple membership in different polities, whereas
15 multiple subjecthood in a Hobbesian world could be perfectly ac-
16 ceptable as long as it is supported by “the amity of the soveraigns”. A
17 person can be the loyal servant of two masters but nobody can simul-
18 taneously be a full member of two collectives that regard themselves
19 as sovereign. Secondly, in contrast with the ancient conception of the
20 polis, modern republicanism has to answer the question: What status
21 should be given to those who do not qualify as active citizens? Even
22 if active citizens are seen as an egalitarian political elite among a
23 broader population, they must refer to a broader concept of society in
24 their pursuit of the common good. Passive citizenship thus comple-
25 ments the activist conception as a second and wider frame of inclu-
26 sion. In this respect, the problem with contemporary neo-republican
27 thought is not the range of inclusion but the dichotomy of active and
28 passive citizenship that is overemphasized within this range. Seeing
29 active political participation and voluntary compliance with civic du-
30 ties as the core expression of citizenship leads to a devaluation of
31 the enjoyment of rights and liberties as a merely passive experience.
32 In contrast, a liberal democratic perspective would emphasize the
33 enabling and activating qualities of civil and social rights which are
34 the essential precondition for making democracy representative of a
35 broad population with widely diverse interests, rather than of a small
36 and socially homogeneous political elite.
37 The active/passive dichotomy that tends to split the polity into
38 two classes of citizens is complemented by a second one that divides
39 a state’s population into those included in, or excluded from, the pol-
40 ity. Republicanism conceives the bond of citizenship as the essential
41 factor of social cohesion. From classic contractarian doctrines it in-
42 herits the idea that the mere social fact of residence in a territory can-
43 not qualify individuals for full membership. This does not rule out a

policy of encouraging naturalization. Citizenship results from an act 1

of will and mutual consent, and each naturalization is a particular 2
instance which highlights and celebrates this general idea that citi- 3
zens freely consent to their membership. However, the republican 4
view is incompatible with a flattening of the threshold of citizenship 5
by granting foreigners rights that ought to remain a prerogative of 6
active citizens only. Voting rights of any kind (even at the local level) 7
must be strictly denied to those who have not been recognized as citi- 8
zens. In contrast with Locke’s proposal that each individual should 9
individually decide on her or his membership on attaining the age of 10
majority (see section “Tensions between...” below), Rousseau’s for- 11
mula for the social contract envisages a ritualistic mutual confirma- 12
tion of membership: 13
14
Each of us puts his person and all his power in common under 15
the supreme direction of the general will, and, in our corporate ca- 16
pacity, we receive each member as an indivisible part of the whole 17
(Rousseau, 1973, VI: 192). 18
19
If a common will, rather than a shared experience of dependency and 20
need for protection, unites the political community, then the act of 21
will that marks the boundary between foreigners and citizens must 22
be regarded as truly constitutive for the polity. 23
24
(4) Ethnic nationalism is the strongest rival for liberal democracy, ethnic nationalism 25
not in the field of political theory where it has hardly a significant 26
following, but in the discursive struggles for political legitimation 27
that unfold in the public arenas of Western democratic states. The 28
two competitors have one feature in common: both support a strong 29
norm of inclusion that applies to a conception of society which does 30
not coincide with the polity. However, they are fundamentally op- 31
posed to each other in the way they determine the boundaries of so- 32
ciety. A nation shares a comprehensive and peculiar culture and his- 33
torical experiences which reach back many generations into the past. 34
Ethnic nationalism conceives of the nation also as a self-reproducing 35
biological group of common descent. In nations like the French or 36
U.S. American ones the ethnic interpretation that searches for its 37
origins in some mythical ancestry (the Gauls or the Pilgrim Fathers) 38
is counterbalanced by others which refer to a historical event of state 39
foundation (the French and American Revolutions). This “political 40
nationalism” comes much closer to the truth, i.e. that it was the mod- 41
ern nation-state which created the nation as a cultural community 42
rather than the other way round (Gellner, 1983). However, even this 43

294 rainer bauböck
1 version is still fundamentally opposed to a liberal conception of soci-

2 ety. For nationalism of any kind, the character and boundaries of the
3 community have been shaped by some irreversible historical process
4 or event. They are thus given independently of the present shape of
5 a state territory or the network of social interactions that connect a
6 population to a state. The nationalist programme is to emancipate
7 the nation in a sovereign state by uniting its dispersed communities,
8 by conquering or liberating the territories they inhabit or where their
9 origins lie, and by assimilating, expelling, or keeping out minori-
10 ties that do not fit into the national community. Nationalism thus
11 attempts to make the boundaries of territory and of cultural groups
12 coincide (Gellner, 1983). Nationalism’s success is rooted in the drive
13 for cultural homogenization of populations within state territories
14 that comes with the development of industrial economies and of the
15 modern state bureaucracy. Nationalism’s failure lies in the prolifera-
16 tion of rival claims to nationhood that have led to an uneasy truce be-
17 tween national factions in pluri-national states, to the survival or new
18 formation of ethnic and linguistic minorities resisting assimilation,
19 and to chain-reactions of separation into ever smaller states that can
20 hardly claim to be independent in their economic or foreign policies
21 (Hobsbawm, 1990).
22 Nationalisms operate with an imaginary map of spheres of hege-
23 mony that nations claim over territories and populations. Seen from
24 the point of view of each single nation, this map resembles that of
25 territorial sovereignty. It is discrete in terms of populations – nobody
26 can be simultaneously a member of two nations – and complete in
27 terms of territories. It need not, however, be complete for all human
28 groups: some have been denied the capacity of belonging to any na-
29 tion or of forming one themselves. This is a characteristic of racism
30 in both its anti-Semitic and anti-Black varieties. Moreover, the terri-
31 torial map is no longer discrete when combining the perspectives of
32 nations that raise rival claims to the same stretch of land.
33
34
35 Inclusion in mobile societies
36
37 A liberal democratic norm of inclusion with reference to a political
38 paradox concept of society faces a paradox. On the one hand, if people did
39 not move across state borders the whole range of inclusion would be
40 perfectly identical with that of territorial sovereignty and the very no-
41 tion of social ties as different from political subjection would become
42 redundant. On the other hand, once societies become transnationally
43 mobile, there is no hard criterion for determining individual mem-
bership.

We might reassure ourselves that only borderline cases will be 1

indeterminate. There is no natural threshold in the time of residence 2
after which a foreigner must be regarded as a member of society. 3
However, agreement about when a person has in fact acquired a kind 4
of residential membership should be rather easy to find if one de- 5
taches the question in a first step from its political consequences in 6
terms of the implied entitlements. It should neither be difficult to es- 7
tablish a list of indicators that obviously turn a person into a member 8
of society in a liberal view, although they might not qualify from the 9
nationalist or republican perspectives: being born in a country and 10
spending one’s early childhood there;l7 being a member of a house- 11
hold where one lives for at least several months each year; going to 12
school or being regularly employed. Other criteria are of a more dubi- 13
ous nature. Consider for example the frequently heard argument that 14
immigrants ought to be given the vote because they pay taxes just 15
as citizens do. However, tax requirements have been generally abol- 16
ished for the franchise. Why should the political rights of foreign- 17
ers depend upon their contributions if modern liberal citizenship 18
has generally dissolved the former nexus between such rights and 19
obligations of this kind? A more difficult criterion is that residence 20
must have been legal in order to qualify for membership. Certainly, 21
a liberal welfare state must be interested in maintaining the rule of 22
law and, more specifically, in preventing the spread of illegal employ- 23
ment. However, the facts of societal membership depend on the time 24
of residence more than on legal status. If a state has been unable 25
or unwilling to control illegal entries, residence and employment, 26
it ought to consider the claims of those who have been residing in 27
the country for a long period of time as relevant. This line of reason- 28
ing could support a general amnesty or an individual procedure for 29
regularization of long-term irregular immigrants. The implications 30
of liberal norms of inclusion are more obvious with regard to depor- 31
tations of legally resident foreigners who have committed a crime. 32
There can be little objection against expulsion when the crime has 33
been committed shortly after a temporary admission into the coun- 34
try. But the current practice of some European states (among them 35
my own country, Austria) of deporting even young foreigners, who 36
have committed minor offences, from their state of birth is certainly 37
indefensible. 38
This idea of inclusion with reference to membership in politically the new nomads 39
bounded societies would fail to provide a satisfactory solution for no- 40
mads. However, the post-modern metaphor of “the new nomads” for 41
modern migrants is completely besides the point when applied to 42
modern migrants. Nomads do not move as individuals but it is rather 43

296 rainer bauböck
1 their societies which move. Social structures, and the individual’s

2 positions in them, are generally rigidly fixed. This creates a kind of
3 societal membership which is dissociated from territorial location.
4 Contemporary migrations show hardly any resemblance with those
5 societies whose movement in space does not expose their internal
6 structure to change. Modern states are strictly tied to a territory not
7 only with regard to their boundaries but even in their microstructure
8 of local administrations. Societies, however, become mobile in the
9 transition from the agrarian to the industrial age and this territorial
10 mobility increasingly affects large majorities. The process is like the
11 transition of a liquid matter into a gaseous state with the effect that
12 the rapid movement of molecules can no longer be contained within
13 the old vessel. International migration expands mobile societies be-
14 yond the borders of territorial states. This does not dissolve the bor-
15 ders; it even leads to their fortification in attempts to enforce political
16 control over the movement of people, but it changes the composition
17 of society as well as internal social and cultural structures. In demo-
18 cratic political systems this change must be reflected in the member-
19 ship composition of the polity as well as in the public recognition of
20 specific interests of migrants.
21 From the perspective of individual migrants, the difference be-
22 tween their situation with that of nomads is that their homes do
23 not move along with them but they have to leave them in search of
24 new ones.18 Two different types of migration result from this, none
25 of which resembles nomads: those who have lost their membership
26 without gaining a new one and those who have retained it until, or
27 three policy even after, they have found a new one. For the latter group, three
28 propositions policy propositions can be derived from the liberal norm of inclu-
29 sion: optional naturalization, toleration and recognition of dual citi-
30 zenship and residential citizenship, i.e. equal basic rights for all resi-
31 dents independent of their nominal citizenship.
32 The former category is that of refugees. In refugee policies norms
33 of admission must precede and supplement those of inclusion.
34 Refugees can raise a claim to be admitted because they have been de-
35 prived of membership whereas family members of immigrants can
36 raise similar claims that are based on their existing ties of member-
37 ship. I think that both claims are strong and there is no need to give
38 general priority to one or the other.19
39 The question as to which norms of admission could be defended
40 from a liberal democratic point of view goes far beyond the scope of
41 this paper.20 Here I only want to point out that the perspective of in-
42 clusion may be widened in order to address one of the most difficult
43 normative problems of refugee policies. The problem can be stated

generally as that of unallocated state obligations (see O’Neill, 1991). 1

One may easily agree that those who have been deprived of their 2
state’s protection and have left that state’s territory can raise indi- 3
vidual claims towards liberal democracies to receive protection there. 4
However, which is the one state, among all possible states of asylum, 5
that is obliged to honour this claim? Virtually all states of Western 6
Europe have now adopted the principle that it is only the first state on 7
the asylum seeker’s route where he or she could file an application 8
for admission. Yet this principle of first country of asylum obviously 9
leads to an inequitable distribution of obligations and burdens and 10
means that large numbers have no chance to be admitted into those 11
countries where they could be most easily accommodated. Even the 12
fact that refugees may already have close family members in some 13
European state is deliberately ignored in order to curb the inflows. 14
After serious consideration one can come to the conclusion that quite 15
often the same circumstances that drive refugees to seek specific des- 16
tinations should also be given some weight in deciding whether a 17
particular state rather than another one ought to admit them. Among 18
such factors may be: economic prosperity which enhances immigra- 19
tion capacities, a common history (often that between a former colo- 20
ny and its colonizing state), cultural ties such as a common language 21
or religion, economic or political involvement of the target state in 22
the state of origin and, finally, geographical proximity. The logic of 23
inclusion might in all these circumstances be extended beyond the 24
boundaries of societal membership in order to determine the special 25
obligations of receiving states. There will be many remaining catas- 26
trophes such as the present one in Rwanda where only a joint effort 27
by the international community and a commitment to cooperate in 28
schemes of burden-sharing will be an adequate answer to a refugee 29
crisis. However, in a world of sovereign states, special obligations 30
always carry more weight than those that fall upon the community of 31
states. 32
These rather sweeping generalizations do not exclude the pos- 33
sibility that there are, or will be in the future, new nomads whose 34
inclusion in territorial states raises a different set of problems. It is 35
quite possible that we may see the emergence of tightly-knit ethnic 36
groups who develop a nomadic way of life because they adapt in this 37
way to special niches in a global economy. However, in most cases, 38
those who are perceived as nomads in Western societies are sim- 39
ply forced to move because they are not allowed to stay anywhere. forced movement 40
Central Europe’s Romanies have been often quoted as an example 41
for the former category, i.e. as an ethnic group whose nomadic be- 42
haviour results from a mutual reinforcement of cultural traditions 43

298 rainer bauböck
1 with a specialization in certain trades. Today, however, they clearly

2 find themselves, once again, in the latter situation of people who
3 are forced to cross borders in order to escape racist discrimination
4 or miserable economic conditions. What often distinguishes them
5 from refugees with similar motives is that they have learned not to
6 expect protection from state authorities and consequently frequently
7 resort to irregular routes of entry. Without societal membership ac-
8 quired through continuous residence and without seeking protection
9 by a state, how can they be included in a liberal democratic polity?
10 The answer probably lies in the difficult task of combining a general
11 improvement in their social conditions with respect for cultural dif-
12 ferences and with the recognition of special minority rights, includ-
13 ing an extended right to travel across borders.
14 liberal citizenship The overall direction of policies suggested in this section can be
15 with transnational characterized as an enrichment of liberal democratic citizenship
16 elements with transnational elements. How does this compare with the supra-
17 national citizenship of the European Union that has been strength-
18 ened by a number of provisions in the 1992 Maastricht Treaty? In
19 addition to existing protection against the discrimination of citizens
20 of other member states with regard to civil and social rights, the trea-
21 ty creates three new rights, two of which penetrate into traditional
22 spheres of national sovereignty. These are: equal active and passive
23 voting rights for resident citizens of other member states in local
24 elections and in elections for the European Parliament (Article 8b),
25 diplomatic protection of European Union citizens in third countries
26 by the representatives of other member states if there is no represen-
27 tation of their own state (Article 8c) and a right to petition European
28 Parliament (Article 8d). The most essential rights conferred by EU
29 citizenship are, however, those of free movement across internal
30 borders and of settlement and access to employment for EU resi-
31 dents in other member states. Such rights to admission go beyond
32 imperatives of inclusion derived from societal membership. In other
33 aspects, European Union citizenship has remained deficient with re-
34 gard to this same norm. There is as yet no generalized option of natu-
35 ralization for EU citizens in other states of the Union. Cases of multi-
36 ple citizenship have strongly increased in number, some states have
37 recently changed their laws and have abandoned the requirement to
38 renounce a previous citizenship on naturalization and there are at-
39 tempts at the level of the European Council to eliminate obstacles to
40 the toleration of multiple citizenship which have been enshrined in a
41 1963 convention. However, so far there is no policy of harmonizing
42 and liberalizing citizenship laws at the level of the Union.
43

The most glaring discrepancy between the supranational model of 1

European Union citizenship and the transnational approach which I 2
try to defend concerns the exclusion of “third country aliens”. Only 3
those foreigners who are nationals of a member-state are also recog- 4
nized as members of the wider community and can enjoy the new 5
rights conferred by citizenship of the Union. Constitutional amend- 6
ments have become necessary in several member states in order to 7
extend local voting rights and access to employment in the civil ser- 8
vice to all Union citizens. By simultaneously confirming the exclu- 9
sion of extracommunitari from these rights, the states of the European 10
Union have actually moved away from a liberal democratic model of 11
inclusion.21 Instead of widening the range of inclusion beyond the 12
combination of national citizenships, EU membership has been con- 13
structed by using them as the elementary building blocks. It links the 14
separate columns of national citizenship by developing the common 15
structure of a roof above them. The foundations of residential rights 16
for all remain as low, and spaces between the columns as empty, as 17
before, only that now these gaps have become much more visible as 18
an element in the design of the building. It looks like the lofty struc- 19
ture of a Greek temple rather than like the much-quoted European 20
house that would accommodate all who live on this continent in its 21
many rooms. 22
A general objection might be raised against my model. Inclusion 23
of migrants will increase the internal heterogeneity of political com- increase of internal 24
munities and could thereby diminish social resources for solidarity heterogeneity 25
among its members. If those who have not been born and raised in 26
the society, and who have not pledged their commitment by naturaliz- 27
ing, are given equal rights of membership, will this not further strain 28
the attenuating sense of mutual obligations in modern societies? I 29
readily concede that it is generally easier to foster such motivations 30
within homogeneous and immobile communities. However, I do 31
not think that this argument provides support for maintaining pres- 32
ent forms of exclusion. Firstly, it could also have been used against 33
the dismantling of gender, race and class barriers which had main- 34
tained a high level of homogeneity among the citizenry for a long 35
time. Achieving the minimum levels of social welfare throughout 36
society which are essential for participatory citizenship was relative- 37
ly easy in earlier forms of democracy, when citizens were a socially 38
rather homogeneous group. This has become a much more difficult 39
task in modern welfare states where it requires extensive redistribu- 40
tion. Secondly, the exclusion of immigrants from rights of citizen- 41
ship reinforces their social segregation within receiving societies. It 42
makes these societies more unequal and their democratic systems 43

300 rainer bauböck
1 less representative over time. Thirdly, cutting back immigration can-

2 not be accepted as a long-term strategy to increase homogeneity for
3 both pragmatic and normative reasons. The acceleration of territorial
4 mobility is inherent in the process of modernization. Contemporary
5 democracies have to accept this as a fundamental condition of mo-
6 dernity which will undermine their bases of legitimacy unless they
7 adjust to it. This does not mean denying the necessity or legitimacy
8 of immigration control, but it does imply that such control must it-
9 self be constantly scrutinized for how well it complies with universal
10 human rights and with specific rights that can be derived from exist-
11 ing ties of membership.
12
13
14 Tensions between inclusion, equality and consent
15
16 Inclusion is not the only relevant norm for the allocation of member-
17 ship status and rights in liberal democratic polities. After discussing
18 the range of claims to societal membership let me now turn to the
19 question as to how such wider inclusion could affect the two other
20 norms for the allocation of membership in liberal democratic poli-
21 ties: the norms of equality and choice.
22 The liberal idea of equality is much stronger with regard to the
23 equal polity polity than with regard to society. As citizens, individuals must be
24 treated with equal respect and concern (Dworkin, 1977), no matter
25 how unequal they may be in their social status. This equality of citi-
26 zenship rests on a foundation of basic individual rights, such as those
27 of equal status in court, equal entitlement to school education or vot-
28 ing rights. However, not all rights of citizenship are perfectly equal
29 and individual. The wider the social range of inclusion becomes, the
30 stronger becomes the urgency to take social inequalities and differ-
31 ences into account by differentiating rights according to social posi-
32 tions and groups. Citizenship in a multi-ethnic welfare democracy
33 is a complex bundle of equal individual rights, as well as of highly
34 differentiated collective ones. Nevertheless, the norm of equality pro-
35 vides the yardstick. A justification of collective rights must show that
36 they contribute towards equalizing the standing of individuals in the
37 polity. In contrast with Hannah Arendt’s strict division between the
38 polity as the sphere of equality, and society as that of discrimination
39 (Arendt, 1958), a liberal democratic approach uses political instru-
40 ments for combating social discrimination. This fight is a precondi-
41 tion for including discriminated groups as equal citizens in the pol-
42 ity. The norm of equality is thus in a relation of productive tension
43 rather than of contradiction or identity with that of inclusion. Wider

inclusion transforms simple into complex equality (Walzer, 1983: 1

330), while liberal equality drives formal inclusion in terms of legal 2
status towards substantial inclusion in terms of positive rights. 3
Equality thus implies, first and most obviously, that the status of 4
nominal citizenship is a homogeneous one. Distinctions between le- 5
gal categories of citizens that exclude some from the enjoyment of 6
basic rights are not permissible (with the three exceptions of minors, 7
convicts and mentally handicapped mentioned above). This is a rath- 8
er recent achievement. Until not so long ago, a number of Western 9
democracies distinguished citizens according to their country of 10
birth or their mode of acquiring citizenship. In France naturalized 11
citizens were excluded from public offices until as late as 1983. Only 12
last year Belgium abolished the distinction between naturalisation 13
ordinaire and grande naturalisation which reserved voting rights to 14
those who had passed the second admission procedure. An example 15
of largely symbolic significance is that in the USA only a native-born 16
citizen can become President. 17
The more important question is: In which ways does the norm 18
of equality extend beyond nominal citizenship and territorial resi- nominal citizenship 19
dence? The bundle of rights enjoyed by emigrant citizens can never and territorial 20
be the same as that of citizens living in the territory. Too many ele- residence 21
ments of citizenship are conditional upon residence. Emigrants will 22
also require specific rights that only concern them, such as diplo- 23
matic protection by their country of citizenship or the right to return 24
to this country. However, monetary entitlements of social citizenship 25
and voting rights can also be exercised while staying abroad. Social 26
security benefits such as retirement pensions are now mostly trans- 27
ferable between states and bilateral agreements often also allow con- 28
tributions and employment periods accumulated in other countries 29
to be added to the claims attained in the country of present residence. 30
Others, like unemployment benefits generally cannot be transferred 31
in this way because they are granted under the condition of searching 32
for a job in a national labour market. Traditional states of emigra- 33
tion often not only discourage expatriation but also try to integrate 34
emigrant citizens actively into the polity by granting them the right 35
to participate in national elections. From the point of view of liberal 36
democracy this may be seen as an ambiguous achievement. On the 37
one hand, retaining one’s citizenship voluntarily when living abroad 38
can be taken as an indicator of subjective affiliation and possible in- 39
tentions of returning which give some credibility to the claims of 40
emigrants to be included in elections on equal terms. On the other 41
hand, most among them will not be affected by the decisions taken 42
by their representatives in parliament and they cannot participate 43

302 rainer bauböck
1 fully in public political discourse and deliberation before elections

2 and referenda. There is certainly some scope for reasonable differ-
3 ence of opinions on this. States with a long tradition and constant
4 stream of emigration will have stronger reasons for enfranchising
5 emigrant citizens than nations of immigrants. In this respect, the
6 limit to acceptable diversity in this regard should be that subsequent
7 generations born abroad might have a claim to their parents’ citizen-
8 ship but certainly not to representation in political decision-making
9 if they have never lived in the country.
10 The norm of equality applies much more extensively to resident
11 foreigners than to emigrant citizens.22 This means that foreigners do
12 not just enjoy protection and rights in their host states (as is already
13 implied by John Locke in the above citation), but that the rights of
14 resident citizens provide the benchmark for the normative evalua-
15 tion of their claims. This idea could be operationalized as a constitu-
16 tional principle: There ought to be a general presumption in favour
17 of the equal treatment and rights of foreign residents and citizens
18 unless expressly decided otherwise by legislation. Furthermore, no
19 such discriminatory exception should be derived from criteria such
20 as national, ethnic, or “racial” origins. This seems to conform to
21 the current interpretation of the equal protection clause in the U.S.
22 American constitution. However, most continental European consti-
23 tutions still do not support such a principle but assume the contrary,
24 i.e. that foreigners will be unequal with regard to public law unless
25 their rights have been explicitly legislated. Only in the realm of pri-
26 vate law is there a general presumption of equality independent of a
27 person’s citizenship.
28 Listing all the different areas of legal discrimination of foreign
29 residents in European countries of immigration would take up too
30 much time and space. I will thus concentrate on the issue which
31 is probably the most controversial one from the point of view of a
32 voting rights normative theory of democracy. This is the question of voting rights.
33 Voting rights for foreigners in general and state elections have a ven-
34 erable tradition which starts with the French revolution. In the USA
35 in the 1880s, 18 states granted alien suffrage to foreigners who had
36 simply filed a declaration of intent to naturalize. These voting rights
37 were only finally abolished after the First World War (Ueda, 1982:
38 128f.). Today, New Zealand is the only country I know of that grants
39 foreign citizens a general franchise in national elections. Others like
40 Great Britain, give, however, such rights of political participation to
41 a large number of non-citizens from Ireland and Commonwealth
42 states. In Sweden, plans to introduce national suffrage for foreigners
43 were seriously considered in the 1980s but were later abandoned.

Sweden, Norway, Finland, Denmark, the Netherlands, the Irish 1

Republic and the Swiss cantons of Neuchâtel and Jura do, however, 2
grant active and passive local voting rights to foreigners. In France 3
and Germany, an extensive debate during the 1980s ended with a 4
defeat of the proponents of local franchise for foreign residents. 5
Local voting rights have been proposed or defended with the ar- 6
gument that municipalities, in contrast with provincial or national 7
parliaments, do not exercise legislative functions. This would imply 8
that foreigners could only be admitted for second-rate political par- second-rate political 9
ticipation. I think there is a much stronger argument for a specific participation 10
priority of access to the local vote. Firstly, just as nominal citizens, 11
foreigners do not formally choose to be members of a certain local 12
community; membership results automatically from residence. It 13
therefore makes sense to also derive the rights of active participation 14
directly from that fact of residence rather than from nominal citizen- 15
ship. Secondly, in contrast with nation-states, local communities are 16
without proper borders. They are open for access to anybody who has 17
a right to live in the national territory, citizens and foreigners alike.23 18
Local democracy does not have to operate under the same constraints 19
of bounded territorial sovereignty as democracy at the national level. 20
This shows that, ultimately, the control over the movement of people 21
is also no necessary condition for democratic legitimacy within ter- 22
ritorial states. A further implication of this view is that municipalities 23
are no longer seen as merely the local sub-units of a single sovereign 24
political power, but on the contrary. They are political communities 25
of a particular character whose rights to local self-determination of 26
their own affairs under democratic control of their own citizenry can 27
be seen as an important contribution towards making representative 28
democracy less indirect.24 Thus, the justification for equal political 29
rights of foreigners at the local level could enhance rather than de- 30
valuate this form of citizenship. 31
Still, this argument does not fully satisfy the criterion of demo- 32
cratic legitimation. There is no reason to assume that local decisions 33
affect foreigners more than national ones. Just on the contrary, their 34
specific discrimination as aliens is rooted in national legislation and 35
this seems to provide a strong argument for also including them in 36
parliamentary elections. I think that this demand is irrefutable from 37
a liberal democratic point of view, as long as we only apply the norms 38
of inclusion and equality. 39
It is at this point that we have to consider whether contractar- 40
ian and republican arguments still carry some weight when deciding 41
about the rights of resident foreigners. In my view, they fail to pro- 42
vide any reason for denying long-term immigrants the quintessential 43

304 rainer bauböck
1 right which enables them to claim substantial equality with citizens:

2 the right of permanent residence (which in a wider sense includes
3 the right to family reunification and to return after a temporary stay
4 abroad). Once this is granted, the scope for redefining the rights of
5 citizenship so that they become rights of residents instead has been
6 considerable extended. Why would this not also apply to national vot-
7 ing rights? The answer is that nominal citizenship would then no
8 longer make any difference. But why should it? What is the value of
9 a status that merely serves to discriminate against some residents?
10 consent The basic idea captured by contractarian and republican doctrines
11 is that the status of citizenship is not only inclusive and egalitarian,
12 but also expresses consent. As I have stated in the first section, demo-
13 cratic legitimacy is based on rational consent, but not necessarily on
14 active, direct, and explicit consent of each individual with each collec-
15 tive decision by which she or he is affected. In some theories, political
16 legitimacy is achieved by hypothetical consent only. Representative
17 democracy requires more than this, although it generally gives only
18 mediated and diluted expression to popular consent in legislation
19 and government. Individuals must be empowered to actually express
20 consent or dissent in a way that has an impact on collectively binding
21 decisions. Public discourse and deliberation among citizens precede
22 decision-making by their representatives.
23 A similar pattern of mediated consent can also be found in the
24 allocation of nominal citizenship. For contractarian theorists it was
25 essential to demonstrate that legitimate rule depends upon individ-
26 ual consent in membership of the polity. Locke was the most radical
27 thinker in this respect when he stated that
28
29 a child is born a subject of no country or government. He is under his
30 father’s tuition and authority till he comes to age of discretion, and
31 then he is a freeman, at liberty what government he will put himself
32 under, what body politic he will unite himself to (Locke, 1956, VIII,
33 §118: 61).
34
35 Yet this requirement is not met by any contemporary liberal democ-
36 racy. Consent to membership is generally not expressed on its acqui-
37 sition but only on its loss in voluntary expatriation and even then the
38 choice of exit from citizenship is made conditional upon a previous
39 exit from the territory and society. In naturalization, on the other
40 hand, the requirement is heavily biased against the individual who is
41 not at all “at liberty to decide what body politic he will unite himself
42 to”. The choice is restricted to that between the countries of origin
43 and of present residence and it is normally the state authorities who

grant admission rather than the applicant who simply chooses a new 1
membership. 2
As I have explained above, I think that a strong case can be argued making 3
in favour of enhancing the element of choice by making expatriation expatriation and 4
and naturalization symmetrical. Both should be individually chosen naturalization 5
rather than either imposed or discretionarily denied and the option to symmetrical 6
change one’s membership should be only conditional upon the cri- 7
teria of residence and societal membership in both cases. Liberalism 8
increases the scope of individual rights and choice by normatively 9
constraining the requirement of collective or majoritarian consent 10
where it threatens to interfere with equal individual liberties and op- 11
portunities. This should hold for naturalization in the same way as it 12
holds for expatriation. 13
However, why should nominal citizenship not be imposed on 14
foreigners if the norms of inclusion and equality are of overriding 15
importance for liberal democracy? The answer is obvious, as long 16
as we assume that multiple citizenship is not generally tolerated. In 17
contrast with new-born natives, foreigners have a citizenship to lose 18
which might be of essential value for their life-projects. Their mul- 19
tiple societal membership gives migrants a strong claim that a natu- 20
ralizing state must respect their existing affiliation and should not 21
require its renunciation as the price for the ticket of entry. However, 22
I do not think that choice loses all importance once dual citizenship 23
has been granted. A receiving state should not naturalize foreigners 24
without their consent even if their previous citizenship remains un- 25
affected. 26
One potential consequence of citizenship which makes the im- 27
portance of choice obvious is that of military conscription. Not every 28
state imposes this obligation on its citizens and most states which 29
do, impose it only on their male citizens of a certain age group. 30
Moreover, liberal democracies permit conscientious objectors to 31
refuse military service without forcing them out of the country or 32
denaturalizing them. Even under these preconditions, I think that 33
resident foreigners have a stronger reason not to be drafted than 34
either native or naturalized citizens. U.S. law is rather unique in 35
making foreign residents liable to be drafted (in case that general 36
conscription were introduced). This seems to result from a biased 37
view on immigration that sees the choice of a country of residence 38
as already implying a decision for the rest of one’s life and regards 39
naturalization as the natural outcome of the process of settlement. 40
If there is any obligation of citizenship which can be said to require 41
a conscious expression of consent, it must certainly be that to kill or 42
die in the defence of one’s country. While under certain conditions of 43

306 rainer bauböck
1 emergency, conscription of native citizens may be justified by invok-

2 ing their hypothetical rational consent without giving them an actual
3 choice, I believe that no such implication can be inferred from the
4 fact of residence and societal membership of foreigners.
5 Apart from this example, the main reason for insisting on the
6 importance of choice of membership in the polity is the following
7 one: When applied to migration, liberalism’s emphasis on individual
8 rights and choice means that the long-term political target is to turn
9 all migration into voluntary movement, rather than to eliminate all
10 root causes of emigration. Enhancing the scope of choice between
11 different migration targets is one principle that can be derived from
12 this guideline. As I have argued above, it applies even to involuntary
13 forms of migration such as refugee migration. It would be inconsis-
14 tent with this line of argument to deny immigrants the choice as to
15 whether they want to become nominal citizens of their country of
16 residence. The imperatives of inclusion and equality are not strong
17 enough to override the manifold individual reasons migrants may
18 have to refuse applying for naturalization. This is only so because
19 substantial and nominal citizenship need not be strictly tied to each
20 other. Resident foreigners can be included and enjoy equal rights
21 without and before naturalization as “residential citizens” or, in the
22 denizens terminology revived by Tomas Hammar, as “denizens” (Hammar,
23 1990).
24 In spite of its emphasis on choice, this is not an altogether vol-
25 untaristic conception of citizenship. Inclusion primarily relates to
26 an objective criterion of societal membership and makes optional
27 naturalization only available to persons who have entered this range
28 – just as voluntary expatriation is only offered to those who have
29 moved out of it. The same criterion also prevents that the toleration
30 of dual citizenship could lead to an accumulation of memberships
31 which no longer correspond to a social involvement of individuals in
32 the affairs of the polity whose members they are. Neither is mutual
33 consent replaced by unilateral individual choice. It is still the politi-
34 cal community which grants naturalization and thereby expresses its
35 consent. What changes with the move from discretionary to optional
36 naturalization is the sequence of interaction. In the usual procedure,
37 the last word is said by the authorities of the receiving state after the
38 applicant has already documented her or his will and qualification.
39 In optional naturalization, the state first lays down the rules for eligi-
40 bility and the final decision is then the applicant’s.
41 This still does not fully answer the question where the impor-
42 tance of the choice lies if it does not imply any consequences for the
43 legal status and rights of those who have to choose. I think that there

are two different answers to this question and each of them seems 1
to me defensible from a liberal democratic point of view. The first 2
answer could be called a liberal-communitarian one. It would affirm 3
that in spite of reasons for differentiating certain obligations such 4
as military service between citizens and foreigners, there is indeed 5
no reason for differentiating citizenship rights. This means that in 6
addition to all the rights foreigners have already been granted in dif- 7
ferent countries, they ought to enjoy the full franchise as soon as they 8
satisfy the general conditions of residence. This total equalization of 9
rights need not deprive the status of nominal citizenship of any at- 10
traction and meaning. It would retain its symbolic value as a formal 11
expression of membership in the polity, whereas the others would be 12
only informal members. Immigrants could choose this status as an 13
expression of their commitment to their society of residence. Indeed, 14
we can assume that an equalization of rights before naturalization 15
will strengthen such feelings of commitment and compensate for 16
the decline in instrumental rationality of naturalization.25. As long 17
as a sufficient number of immigrants can be motivated to make a 18
voluntary choice in favour of naturalization, there is little reason to 19
abandon the nominal distinction between foreigners and citizens, 20
even though it might have turned into a largely symbolic one. One 21
might object that commitment of a purely symbolic nature is always 22
likely to assume a nationalist tinge. However, opting for naturaliza- 23
tion under conditions where full rights can also be enjoyed without 24
taking this step would express a rather harmless kind of patriotic 25
pride in the achievements of a liberal democratic polity. 26
There may be reasonable disagreement about such a total dis- 27
sociation of legal status and rights of citizenship. If the essence of 28
democratic legitimacy lies in the kinds of rights that it establishes for 29
citizens, should not admission to the polity be more than a merely 30
symbolic inclusion into the community whose process of demo- 31
cratic decision-making establishes and confirms the validity of these 32
rights? After all, individuals are actively involved in democratic legiti- 33
mation as members of the polity rather than of society. A collective 34
constitutes itself as a polity distinct from society by institutionalizing 35
democratic deliberation at the highest level of sovereignty. Should 36
this not be reflected in making the suffrage at this level conditional 37
upon a decision to become a member of the polity for all those who 38
had previously been a member of a different polity? Again, I think 39
there are some drawbacks in this argument. The most important one 40
is that if the incentives for naturalization are not strong enough, a 41
large percentage of the population in societies of immigration might 42
remain permanently excluded from the most important mechanism 43

308 rainer bauböck
1 of democratic legitimation. However, under ideal conditions of op-

2 tional naturalization, there is little reason to fear such an outcome.
3 This solution, which we could call liberal-republican, would there-
4 fore be equally permissible as the liberal-communitarian one. In any
5 case, both solutions go far beyond present policies of inclusion in all
6 Western democracies. There is considerable scope for a simultane-
7 ous improvement of records along all three normative dimensions of
8 inclusion, equality and consent before one reaches the point where
9 tensions between them might manifest themselves as dilemmas.
10
11
12 Conclusions
13
14 I have argued in this paper that the norm of inclusion is central to
15 a liberal understanding of democracy and that it refers to a concept
16 of society that is wide enough to include foreign residents and their
17 family members abroad as well as emigrant citizens. Nonetheless, in-
18 clusion is not the only relevant norm for liberal democracy. Equality
19 of membership and of rights in the polity, and consent expressed
20 in political deliberation and in agreement to membership are of the
21 same significance.
22 conflicts Inclusion and equality may come into conflict when individuals
23 enjoy a common status of membership, but unequal rights or, in-
24 versely, when they enjoy equal rights, but unequal nominal status.
25 While the former tension develops with the accumulation of collec-
26 tive rights in addition to individual ones, the latter one results from
27 the extension of citizenship rights beyond nominal membership in
28 the polity. Both these outcomes can be well justified in a liberal ap-
29 proach.
30 Inclusion comes into tension with consent already with the auto-
31 matic attribution of citizenship at birth. The conditions of consent in
32 membership can, however, be restored by making both expatriation
33 and naturalization optional. This may diminish formal inclusion of
34 resident foreigners compared with a solution that would attribute a
35 status of citizenship automatically after some time of residence. But
36 that latter policy would ignore the specific interests and autonomous
37 choices of immigrants. So the balance seems to be well-drawn in the
38 way that I have suggested.
39 Finally, equality and consent seem to conflict with each other if
40 equal rights can be had without any conscious decision for member-
41 ship. However, as I have argued, even this radical solution would not
42 make the choice of membership meaningless. A different position
43 that insists on tying national voting rights to nominal citizenship

seems to be equally defensible within a framework where resident 1

foreigners can freely choose to be naturalized. 2
None of these solutions takes fully into account the situation of 3
those who have not, or not yet, achieved full societal membership. 4
Transient and temporary migrants as well as those who have just 5
arrived but intend to stay will not be fully included, will not enjoy 6
completely equal rights, and will not be offered all the options of 7
membership. However, in Michael Walzer’s words, the basic norm 8
of inclusion requires that they “must be set on the road to citizen- 9
ship” (Walzer, 1983: 60). 10
At first sight, the overall distribution of rights and legal status in 11
liberal democracy, which emerges from our normative discussion, 12
seems to violate all three principles of inclusiveness, equality and 13
consent. Instead of being a homogeneous status, citizenship is ac- 14
quired in a different way by natives and naturalized immigrants; it 15
is different in its meaning of affiliation to a polity for single and dual 16
citizens; it is different in its substance of rights for emigrant citizens, 17
for temporary immigrants and for long-term residents. Nevertheless, 18
this multi-layered structure of citizenship can be regarded as a conse- 19
quence of combining the three norms and applying them to a world 20
where societies have become mobile across state borders. However, 21
the very same principles which can justify such distinctions also guiding principles 22
point to many obstacles which ought to be removed from the path to 23
citizenship before the terms of admission can be regarded as fair. 24
25
26
Notes 27
28
* This contribution draws on arguments developed at more length in a 29
forthcoming book: Transnational Citizenship. Membership and Rights in 30
International Migration, Edward Elgar, Avebury, UK, 1994. It was first pre-
sented at a panel organized by Joseph Carens at the 90th Annual Meeting
31
of the American Political Science Association, 14 September in New York. 32
Amy Gutman’s critical comments at the conference stimulated some clari- 33
fications in revising the paper. Credits are also due to Ulrike Davy for chal- 34
lenging me to elaborate the apparent contradictions between a strategy of 35
equalizing rights for foreign residents and citizens and one of making natu-
36
ralization optional.
1 The last relics of gender discrimination in Western European systems of 37
franchise have been abolished with the recent introduction of full voting 38
rights for women in the Swiss canton Appenzell-Innerrhoden. The problem 39
of denaturalization of ethnic minorities is still acute in some newly democ- 40
ratized states of Central and Eastern Europe. In June, the Latvian parliament 41
adopted a citizenship law that would make 500,000 ethnic Russians who
have immigrated after 1940 stateless until the year 2000. (The Latvian presi-
42
43

310 rainer bauböck
1 dent objects to the law which has not yet come into force.) In July, the Czech
2 republic turned 70,000 Roma into a stateless minority because they had not
3 applied in time for citizenship of the new state.
2 John Stuart Mill, for example, denounced the exclusion of women but de-
4 fended a franchise limited to taxpayers and a system of plural votes for
5 citizens with a higher education (Mill, 1972, On Representative Government,
6 chapter 8).
7 3 As regarded by legal positivists (see de Groot, 1989: 10-17).
8 4 Joseph Carens has objected that “[a]fter a while, the terms of admission be-
come irrelevant” (Carens, 1989: 44).
9
5 Article 34 of the Geneva Refugee Convention obliges states of asylum to
10 facilitate the integration and naturalization of refugees and to reduce the
11 costs of the procedure as far as possible. A number of signatory states take
12 this into account by reducing the required period of residence prior to the
13 naturalization of refugees.
14 6 In some Western democracies an option exists for those foreigners who are
not immigrants but have been born in the country, or for immigrants who
15 have married a citizen.
16 7 Canada and Australia are probably the two countries of immigration that
17 today come closest to this model of optional naturalization. The Canadian
18 Citizenship Act includes ordinary naturalization in a section under the title
19 “The Right to Citizenship”. Article 5 of the Act specifies that the Minister
shall grant citizenship to any person who meets the requirements whereas
20
the Minister may, in his discretion, waive on compassionate grounds some of
21 these requirements in favour of the applicant.
22 8 David Hendrickson points out that in a realist perspective “[t]he acquisition
23 of nationality is a more momentuous step, and it would not be inconsistent
24 with this formulation to hold that the state’s discretion is much wider in
25 deciding upon membership and nationality than in rejecting admission to
visitors” (p. 219).
26 9 A view which has been strongly criticized by Robert Dahl, who insists that
27 “[t]he demos must include all adult members of the association except tran-
28 sients and persons proved to be mentally defective” (Dahl, 1989: 129).
29 10 See de Groot, 1989: 12-13.
30 11 This has been an unintended effect of eliminating gender discrimination
in citizenship laws. Until well after the Second World War, citizenship was
31
transmitted only by the father in most Western democracies. The mother’s
32 membership became then only relevant if the child was born out of wedlock.
33 12 Apart from being subjected to territorial sovereignty these “transients”
34 (Robert Dahl) may certainly have rights towards their temporary host coun-
35 try but such rights are not based on their societal membership. They result
36 rather from a commitment to respect human rights when no significant
ties of membership are involved. This same kind of commitment opens the
37
boundaries of liberal polities to claims of refugees that their admission is a
38 matter of right rather than merely of generosity, clemency or expediency.
39 13 Claims to external ethnic membership can be based on a purely nationalist
40 line of argument that replaces the political concept of society with that of
41 a national community of descent and culture. In a liberal democratic view
membership requires ongoing social ties of interaction and communication
42
and/or dependence from a state for protection. Where neither is the case the
43 claims to national solidarity beyond borders become spurious.

14 The German Aliens Act of 1990 has, for the first time recognized that young 1
foreigners who had to return to their parents’ country of origin enjoy a right 2
to re-immigration into Germany, i.e. a prerogative that has traditionally been 3
reserved for citizens only. However, beneficiaries are defined very narrowly
as those who have spent at least eight years in Germany and have visited
4
school there for six years, who have sufficient means of subsistence and who 5
apply for return between their 15th and 21st birthdays and within five years 6
after leaving Germany (Gesetz zur Neuregelung des Ausländerrechts, § 16). 7
15 Nozick’s theory defends an atomistic individualism only at the level of states 8
but envisages the flourishing of a multiplicity of associational communities
9
within that framework (Nozick, 1974: chapter 10).
16 See, for example, Oldfield (1990) or van Gunsteren (1992). 10
17 The extreme interpretation of ius soli in the USA, which merely focuses on 11
territorial birth and attributes citizenship automatically even if a child is born 12
on board an aircraft flying over the territory, need not necessarily be seen as 13
a model for other countries of immigration. (For an interesting controversy 14
about the attribution of citizenship to native-born children of illegal immi-
grants see Schuck and Smith, 1985 and Carens, 1987.) For European states
15
that consider reforming their ius sanguinis laws, it would probably make 16
more sense to apply ius soli to native-born children under the condition that 17
one parent has been resident in the country for at least a short period of time. 18
The solution that seems most attractive to me would be to give alien parents 19
a choice, whether they want their children to acquire citizenship at birth and
20
to give the children themselves a second option at an age well before they
attain the age of majority. However, a third generation, i.e. children born in 21
the country of parents themselves born in the country, ought to be attributed 22
automatic citizenship. This is the rule of double ius soli which is, among oth- 23
ers, established in French and Belgian law. 24
18 This criterion distinguishes migrants not only from nomads but also 25
from tourists who visit other countries without searching for a new home.
26
International tourism is a major consequence of the modern revolution in
transportation technology. It strongly affects the economy, ecology and cul- 27
ture of states but it raises no challenge for their definitions of membership. 28
In nomadic migration, societies move while individuals stay put within their 29
structure; in tourism, societies stay put while individuals move. In modern 30
migration the movement of individuals causes an expansion of the social
31
basis of membership.
19 As Mark Gibney does when he defends a liberal admission policy for refu- 32
gees by attacking the U.S. immigration priority for relatives of citizens and 33
immigrants (Gibney, 1986). 34
20 I have tried to address this question in two other papers (Bauböck, 1994a, 35
1994b). 36
21 See Marco Martiniello’s contribution in this volume.
37
22 The third category of persons to whom the norm of inclusion may apply are
those who are neither citizens nor residents. For them there is little substan- 38
tial equality. They may claim admission to the territory but not many other 39
rights which they could exercise beforehand. 40
23 See Article 13 of the Universal Declaration of Human Rights and Article 12 41
of the International Covenant on Civil and Political Rights.
42
24 The classical statement on this point is Tocqueville’s analysis of New England
township democracy (Tocqueville, 1954: chapter 5). 43

rainer bauböck
1 25 Relatively high rates of naturalization in countries such as Sweden, Australia,

2 Canada, which grant both easy naturalization and substantial rights for for-
3 eign residents, seem to provide empirical illustration for this point.
4
5
6 References
7
8 Ackerman, Bruce A. (1980) Social Justice in the Liberal State. New
9 Haven: Yale University Press.
10 Arendt, Hannah (1958) The Human Condition. Chicago: University
11 of Chicago Press.
12 Bauböck, Rainer (1994a) ‘Citizenship and Ethical Problems of
13 Immigration Control’, in Robin Cohen (ed.) (1994) The Cambridge
14 Survey of World Migration. Cambridge: Cambridge University
15 Press.
16 Bauböck, Rainer (1994b) ‘Legitimate Immigration control’, in
17 Adelman, Howard (ed,) Legitimate and Illegitimate Discrimination:
18 New Directions in Migration. Toronto: York Lane’s Press and
19 UNESCO.
20 Carens, Joseph H. (1987b) ‘Who Belongs? Theoretical and Legal
21 Questions about Birthright Citizenship in the United States’, The
22 University of Toronto Law Journals 37: 413-443.
23 Carens, Joseph H. (1989) ‘Membership and Morality, in Brubaker,
24 Rogers W. (ed.), op. cit.
25 de Groot, Gerard-René (1989) Staatsangehörigkeitsrecht im Wandel.
26 Eine rechtsvergleichende Studie über Erwerbs- und Verlustgründe der
27 Staatsangehörigkeit. Köln: Carl Heymans Verlag.
28 Dworkin, Ronald (1977) Taking Rights Seriously. Cambridge, Mass.:
29 Harvard University Press.
30 Gellner, Ernest (1983) Nations and Nationalism. Oxford: Blackwell.
31 Gibney, Mark (ed.) (1988) Open Borders? Closed Societies? New York:
32 Greenwood Press.
33 Hammar, Tomas (1990) Democracy and the Nation State. Aliens,
34 Denizens and Citizens in a World of International Migration.
35 Aldershot: Avebury.
36 Held, David (1991b) ‘Democracy, the Nation-State and the Global
37 System’, in Held, David (ed.), op. cit.
38 Held, David/McGrew, Anthony (1993) ‘Globalization and the Liberal
39 Democratic State’, Government and Opposition 28, no. 2.
40 Hendrickson, David C. (1992) ‘Migration in Law and Ethics: A Realist
41 Perspective’, in Barry, Brian/Goodin, Robert E. (eds), op. cit.
42 Hobbes, Thomas (1973) Leviathan. London: Everyman’s Library.
43 Hobsbawm, Eric (1990) Nations and Nationalism Since 1780.
Cambridge: Cambridge University Press.

Locke, John (1956) The Second Treatise of Government and A Letter 1

Concerning Toleration. Edited with an Introduction by J.W. Gough. 2
New York: Macmillan. 3
Mill, John Stuart (1972) Utilitarianism, On Liberty and Considerations 4
on Representative Government. London: Everyman’s Library. 5
Nozick, Robert (1974) Anarchy, State, and Utopia. Oxford: Basil 6
Blackwell. 7
Oldfield, Adrian (1990) Citizenship and Community. Civic 8
Republicanism and the Modern World. London: Routledge. 9
O’Neill, Onora (1991) ‘Transnational Justice’; in Held, David (ed.), 10
op. cit. 11
Rousseau, Jean-Jacques (1973) The Social Contract and Discourses. 12
London: Dent, Everyman’s Library. 13
Schuck, Peter H./Smith, Rogers M. (1985) Citizenship without 14
Consent. Illegal Aliens in the American Polity. New Haven and 15
London: Yale University Press. 16
Schumpeter, Joseph A. (1950) Capitalism, Socialism and Democracy, 17
third edition. New York: Harper Torchbooks. 18
Tocqueville, Alexis de (1954) Democracy in America, vol. 1. New York: 19
Random House. 20
Ueda, Reed (1982), ‘Naturalization and Citizenship’, in Thernstrom, 21
Stephan (ed.) Immigration, Dimensions of Ethnicity, A series of 22
Selections from the Harvard Encyclopedia of American Ethnic 23
Groups. Cambridge, Mass.: Harvard University Press. 24
van Gunsteren, Herman R. (1992), Eigentijds Burgerschap. Den Haag: 25
Wetenschappelijke Raad voor het Regeringsbeleid. 26
Walzer, Michael (1983), Spheres of Justice: A Defense of Pluralism and 27
Equality. New York: Basic Books. 28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Mixed embeddedness: (in)formal economic 3
4
activities and immigrant businesses in
5
the Netherlands* 6
7
Robert Kloosterman, Joanne van der Leun and Jan Rath 8
9
10
11
12
13
In the field of immigrant or ethnic entrepreneurship, several theoretical ap- 14
15
proaches have emerged. Some emphasise the cultural endowments of im-
16
migrants (such as a cultural inclination in certain groups towards risk-taking
17
behaviour), while others highlight racist or ethnic exclusion and blocked mo-
18
bility in the mainstream labour market. Other approaches revolve around 19
issues of social embeddedness, arguing that individual entrepreneurs take 20
part in ethnically specific economic networks that facilitate their business op- 21
erations. The economic geographer Robert Kloosterman, the criminologist 22
Joanne van der Leun and the sociologist Jan Rath have explored these com- 23
plex interactions along with the array of regulatory structures that promote 24
certain economic activities while inhibiting others. This innovative approach 25
– dubbed ‘mixed embeddedness’ – emphasises the importance of regulation 26
and market dynamics. It is more encompassing in that it links social rela- 27
tions and transactions to wider political and economic structures. Moreover, 28
it acknowledges the significance of immigrants’ concrete embeddedness in 29
social networks while understanding that their relations and transactions 30
are more abstractly embedded in wider economic and political-institutional 31
structures. 32
33
34
Immigrant entrepreneurs and advanced urban economies 35
36
The impact of immigrants has very noticeably changed the outlook 37
of larger Dutch cities in the last quarter of this century. Beginning 38
with the crowds in the streets, by now this demographic shift has also 39
manifested itself in the rising number of immigrant entrepreneurs. 40
Because of this, the four largest Dutch cities (Amsterdam, Rotterdam, 41
The Hague and Utrecht) have not only acquired a distinctly more cos- cosmopolitan 42
mopolitan outlook (Rath and Kloosterman, 1998b), but have also be- outlook 43

316 robert kloosterman, joanne van der leun and jan rath
1 come more like other advanced urban economies, such as New York,
2 Los Angeles, London, Paris and Marseilles, where immigrants and
3 immigrant entrepreneurs are a prominent presence as well (Body-
4 Gendrot and Ma Mung, 1992; Barrett et al., 1996; Häußermann and
5 Oswald, 1997). These immigrant entrepreneurs are affecting cities in
6 numerous – and sometimes quite unexpected – ways, as, for exam-
7 ple, by revitalizing formerly derelict shopping streets, by introducing
8 new products and new marketing strategies (Rath and Kloosterman,
9 1998a), by fostering the emergence of new spatial forms of social
10 cohesion (see, for example, Tarrius and Péraldi, 1995; Simon, 1997),
11 by opening up trade links between faraway areas that were hitherto
12 unconnected through so-called transnational communities (Tarrius,
13 1992; Portes and Stepick, 1993; Portes, 1995b; Guarnizo, 1996;
14 Faist, 1997; Wallace, 1997; The Economist, 1998) and by posing chal-
15 lenges to the existing regulatory framework through being engaged
16 in informal economic activities (Kloosterman et al., 1998). As for the
17 latter, contemporary urban economic sociological studies suggest
18 that immigrants and especially immigrant entrepreneurs play a piv-
19 otal role in these informal economic activities. According to Portes
20 and Sassen-Koob (1987: 48), ‘immigrant communities have provided
21 much of the requisite labor for these activities, have frequently sup-
22 plied sites for their development, and have furnished the entrepre-
23 neurial drive to initiate them’.
24 Below, we explore the role of immigrant entrepreneurs in in-
25 formal activities. We will show that the socio-economic position of
26 immigrant entrepreneurs – and, consequently, also their prospects
27 with respect to upward social mobility – can only properly be under-
28 stood by taking into account not only their embeddedness in social
29 networks of immigrants but also their embeddedness in the socio-
30 economic and politico-institutional environment of the country of
31 mixed settlement. We therefore propose the use of a concept mixed embed-
32 embeddedness dedness, which encompasses both sides of embeddedness to analyse
33 processes of insertion of immigrant entrepreneurs. Complex con-
34 figurations of mixed embeddedness enable immigrant businesses to
35 survive partly by facilitating informal economic activities – in seg-
36 ments where indigenous firms, as a rule, cannot.
37 With the rising number of immigrants and, more particularly, of
38 immigrant entrepreneurs in Dutch cities, the issue arises of whether
39 the small shop run by an immigrant is a step up on the avenue of so-
40 cial mobility or whether it is located on a dead-end street. Exploring
41 these forms of mixed embeddedness among immigrant entrepre-
42 neurs in concrete Dutch metropolitan milieus will eventually allow
43 us to assess to what extent immigrant entrepreneurship in conjunc-

tion with informal economic activities constitutes a distinct trajectory 1

of incorporation. 2
We will start with a short overview of recent developments in 3
immigrant entrepreneurship in the Netherlands. After that, we will 4
present a more thorough account of mixed embeddedness and its 5
relationship with informal economic activities. We will then use this 6
concept to explore a case study of a specific group of immigrant en- 7
trepreneurs in more depth, namely, that of Turkish and Moroccan 8
Islamic butchers. Finally, we will offer some conclusions on the re- 9
lationship between immigrant entrepreneurs and the context of the 10
receiving country. 11
12
13
The rise of immigrant entrepreneurs in the Netherlands 14
15
At first glance, the rise in immigrant entrepreneurship in advanced 16
cities seems to be the obvious outcome of, on the one hand, the inflow 17
of immigrants and, on the other, the resurgence of self-employment 18
in general (OECD, 1992; Martinelli, 1994; Light and Rosenstein, 19
1995). The share of a particular group or category among the ranks of 20
the self-employed is, however, anything but a straightforward reflec- 21
tion of its share of the population as a whole. The rate of participation 22
in entrepreneurship of a particular group of immigrants depends on 23
the intricate interplay between socio-economic and ethno-social char- 24
acteristics of the group in question and the opportunity structure. 25
This opportunity structure – which in itself is primarily a function opportunity 26
of the state of technology, the costs of production factors, the nature structure 27
of the demand for products and the institutional framework – deter- 28
mines when, where and to what extent openings for such businesses 29
will occur. 30
Immigrants in the Netherlands, have found themselves – from 31
a socio-economic point of view – in a rather marginalized position. 32
Despite the fact that the Dutch job machine has been churning out 33
jobs at a very high rate almost continually in the 1990s, unemploy- 34
ment among immigrants has remained relatively high. In 1997, when 35
the Dutch economy was booming, the average rate of unemployment 36
among immigrants still stood at 18%; whereas only 6.3% of the indig- 37
enous workforce was out of work. Turks (31%) and Moroccans (24%) 38
are especially hard hit by unemployment (CBS, 1998). Excluded to a 39
considerable extent from the mainstream labour market, an increas- 40
ing number of immigrants have opted to set up shops themselves. In 41
1986, 11,500 firms in the Netherlands were run by immigrant entre- 42
preneurs. This number had doubled in 1992 and trebled to 34,561 in 43

1 1997, which amounts to about 5.5% of all non-agricultural firms in

2 the Netherlands. The share of self-employed in the total population
3 of immigrants from non-industrialized countries rose from 3.3% in
4 1986 to 7.4% in 1997 (Tillaart and Poutsma, 1998: 40-6).
5
6 Figure 1 The number of immigrant entrepreneurs in the four largest
7 Dutch cities, 1989-97
8
9 7000
10 6000
11 5000 Amsterdam
12
4000 Rotterdam
13
14 3000 Den Haag
15 2000 Utrecht
16 1000
17
0
18
1989 1991 1993 1995 1997
19
20 Source: Based on van den Tillaart & Poutsma 1998: 186
21
22 uneven distribution This spectacular rise of immigrant entrepreneurship has not been
23 evenly distributed in a spatial sense nor with respect to economic
24 activities (cf. Kloosterman, 1996; Kloosterman and Van der Leun,
25 1999). Firstly, immigrant entrepreneurs are heavily concentrated in
26 the four largest cities and especially in Amsterdam. In 1997 about
27 40% of all immigrant entrepreneurs could be found in these cities
28 and about 20% in Amsterdam alone. The rise of the number of im-
29 migrant entrepreneurs in the four largest Dutch cities is shown in
30 Figure 1.
31 This specific spatial pattern clearly reflects the demographic
32 distribution of immigrants in the Netherlands: about 44% of the
33 population of immigrants from non-industrialized countries live
34 in Amsterdam, Rotterdam, The Hague or Utrecht (CBS, 1997).
35 Kloosterman and Van der Leun (1999) have shown that the rela-
36 tionship between demographic trends and the development of im-
37 migrant entrepreneurship also holds within cities. Neighbourhoods
38 with high shares of immigrants in their population turn out to be
39 those with a relatively high number of business start-ups by immi-
40 grants compared to business start-ups by indigenous entrepreneurs.1
41 This relationship suggests the importance of social networks that are
42 mainly based on the proximity of co-ethnics for fledgling firms run
43 by immigrants.

Immigrant entrepreneurship is not just selective in a spatial specific economic 1

sense but also heavily skewed towards specific economic activities. activities 2
Lacking in most cases access to substantial funds of (financial) capi- 3
tal and also deemed lacking in appropriate human capital (educa- 4
tional qualifications), most fledgling immigrant entrepreneurs from 5
non-industrialized countries can, generally, only set up shop at the 6
lower end of this opportunity structure, i.e. in markets with low bar- 7
riers of entry in terms of capital outlays and required educational 8
qualifications. Notwithstanding these ostensibly atavistic characteris- 9
tics, these markets are part and parcel of advanced urban economies 10
(cf. Sassen, 1991; Barrett et al., 1996). 11
As Figure 2 shows, about three in five of the immigrant entrepre- 12
neurs in the largest four cities in the Netherlands have set up shop in 13
either wholesale, retail or restaurants. 14
15
Figure 2 Immigrant entrepreneurs in wholesale, retail and restaurants 16
as a share of the total number of immigrant entrepreneurs 17
in the four largest Dutch cities, 1997 18
19
30 20
25 Wholesale
21
22
20 Retail
23
15 Restaurants 24
10 25
5 26
27
0
28
Amster-
Utrecht
Rotter-
Hague
dam
dam
The
29
30
Source: Based on van den Tillaart & Poutsma 1998: 186 31
32
These are not only economic activities that may cater for an ethnic 33
demand (ethnic foodstuffs, specific clothing), but also sectors where 34
businesses may be started with, in principle, relatively small outlays 35
of capital and limited educational qualifications. Our research find- 36
ings show that immigrants gravitate to businesses at the lower end of 37
the market (Kloosterman et al., 1997; Rath, 1998b; 1999a). 38
Low barriers of entry is one side of the coin, fierce competition 39
the obvious flip side in these highly accessible economic activities. 40
Survival, therefore, is generally difficult and profits can be very low 41
and in many cases even non-existent. The survival of immigrant 42
businesses in these cut-throat markets depends to some extent on 43

1 the fact that many immigrant entrepreneurs (and their families)

2 have different sets of preferences still partly rooted in the sending
3 countries that allow for long hours and low pay (Waldinger, 1996).
4 Their survival is, however, also possible because of the fact that en-
5 trepreneurs are embedded in specific social networks that enable
6 them to reduce their transaction costs in formal but also in informal
7 ways (Zhou, 1992; Portes and Sensenbrenner, 1993; Roberts, 1994;
8 Portes, 1995a). Informal practices can thus be seen as an intrinsic
9 part of contemporary capitalism, especially as the rise of the service
10 sector contributes to a more favourable environment for small firms
11 that partly rely on informal production (Castells and Portes, 1989;
12 Fainstein et al., 1992; Tarrius, 1992; Engbersen, 1997; Kloosterman
13 et al., 1997; 1998; Rath, 1998b).
14
15
16 Informal economic activities
17
18 income generation Informal production encompasses those ‘activities aimed at produc-
19 ing a positive effect on income (for the person executing the activities
20 and/or for the person receiving the results), for which the terms of
21 legislation and regulations (planning requirements, social security
22 legislation, collective labour agreements, and the like) applicable
23 to the activities are not being met’ (Renooy, 1990: 24). Portes and
24 Sassen-Koob (1987: 31) explain that ‘although this definition encom-
25 passes criminal activities, the term is customarily reserved for such
26 activities as those in the food, clothing, and housing industries that
27 are not intrinsically illegal but in which the production and exchange
28 escape legal regulation’. The informal economy is thus conceived as
29 a process of income generation rather than a characteristic of an in-
30 dividual (if only because a moonlighter may have an entirely legal job
31 at another time of the day). The decisive characteristic of the infor-
32 mal economy which distinguishes it from the formal economy is the
33 lack of governmental control (Renooy, 1990: 25).
34 five observations Although a useful definition, five observations have to be made.
35 Firstly, this definition is wholly contingent on the regulatory con-
36 text and this may differ from time to time and from place to place.
37 What is informal in one place may be completely legal in anoth-
38 er. Prostitution, for instance, is completely illegal in a number of
39 American states but (partly or entirely) legal in other states such as
40 Nevada or in countries like the Netherlands. Moreover, specific regu-
41 latory contexts may even create distinct informal economic activities.
42 By establishing monopolies for the sale of cigarettes, the Italian gov-
43 ernment – inadvertently, one presumes – also created the somewhat

peculiar sight of African immigrants selling cigarettes along major 1

exit routes in cities. This points to the dynamic character of the in- 2
formal economy. Activities that are entirely legal at one time, may 3
turn out to be illicit at another. Only three decades ago, the Dutch 4
government welcomed undocumented immigrants who were rep- 5
resented as ‘spontaneous guestworkers’ (Wentholt, 1967: 189; Rath 6
and Schuster, 1995: 103). They were duly regularized as soon as they 7
found a job. Today, the Dutch government perceives undocumented 8
immigrants as ‘profiteers’ who should be expelled instantly. 9
Secondly, being contingent on the regulatory framework also pre- 10
cludes an unequivocal description of the nature and extent of the in- 11
formal economy. Depending on the rules and regulations one takes 12
as a point of departure, one may, in one case, describe and measure 13
the informal economy in terms of money (e.g. the amount of evaded 14
tax money). In another case, however, the unit of measure could be 15
in terms of persons (e.g. the number of undocumented workers). 16
Thirdly, the difference between informal economic activities and 17
criminal activities is not always easy to make. If certain activities are 18
illegal under certain circumstances, does breaching of the rules im- 19
ply an informal or a plain criminal activity? In the Netherlands, legal 20
actions are taken against people who trade in hard drugs; their eco- 21
nomic activities are considered as violations of the Criminal Law. 22
The tax authorities, however, usually consider these same activities 23
as just another form of income generation and, accordingly, levy 24
taxes. By the same token, the criminal is entitled to the same kind of 25
tax deductions with respect to the costs accrued in the process of in- 26
come generation. Having said this, it can be argued that the risks of 27
criminal activities are higher and this may affect the strategies of the 28
(criminal) entrepreneur. However, this makes the criminal economy 29
at best a special variant of the informal economy. 30
Fourthly, within one given regulatory context and regardless of 31
its potential criminal content, it may still be hard to delineate for- 32
mal and informal economic activities. If an employee of a firm parks 33
their van without paying the parking fee to deliver a package, should 34
this be considered as a form of informal production? Furthermore, 35
no ordinary person of flesh and blood is capable of knowing all the 36
rules and regulations by heart, let alone live by them. Especially in 37
advanced welfare states such as the Netherlands with its enormous 38
corporatist legacy of rules and regulations, it seems nigh impossible 39
to avoid some form of informal production. What is considered to 40
be an informal economic activity then becomes, to a certain extent, a 41
matter of arbitrariness. 42
Fifthly, to examine informal economic activities one also has to 43

1 take into account the ways in which government agencies actually

2 deal with these activities. Law enforcement is not self-evident nor
3 does it follow a fixed course. This has, amongst other things, to do
4 with the way in which it is organized. The government may assign
5 many or only a few powers to enforcement agencies, and/or may
6 give them the power to establish their own priorities. These priorities
7 are affected by political decisions as well as the caseload, the inter-
8 pretation of the job, the motivation and the efforts of the law en-
9 forcers themselves. The system of law enforcement may undermine
10 itself when the financial or other costs of policing the rules become a
11 greater burden to society or to the industry concerned than taking a
12 more lenient position. For this reason, the new Labour government
13 in Britain intended to relax the procedures regarding the employ-
14 ment of illegal immigrants (The Independent, 29 May 1997). Labour,
15 moreover, believes that strict law enforcement has deterred many
16 employers from giving jobs to immigrants and that it has created a
17 new market for false ID papers. By relaxing the rules the government
18 tries to prevent such perverse effects. This actual law enforcement
19 is to some extent dependent on the culture of public administration
20 in general and the enforcing agencies in particular. Perhaps part-
21 ly as a result of their plethora of rules, the Dutch have developed a
22 rather awkward way of dealing with certain infringements: they are
23 officially tolerated as part of the typical Dutch policy of gedogen, ‘a
24 nigh-untranslatable term that means looking the other way when you
25 must’ (The Economist, 12 October 1996; see, for a more sophisticated
26 view, Blankenburg and Bruinsma, 1994). Most famous (or notori-
27 ous) in this respect is the Dutch policy towards soft drugs; although
28 illegal, they are tolerated within certain limits. This same approach
29 of gedogen is also used towards certain activities by immigrants. The
30 sale of foodstuffs in mosques in Rotterdam, for instance, provides
31 the cemaat with important financial resources which it otherwise
32 cannot obtain. This sale is illicit but nevertheless tolerated (Rath et
33 al., 1996). Likewise, the government has designed a quite flexible
34 set of transitional arrangements for Islamic butchers working with
35 no permits because of the prevailing importance of the sale of hâlal
36 meat. We will turn to this issue in more detail below.
37 implications The implications of these observations are far-reaching. Against
38 the grain of many popular views, there is no sharp demarcation be-
39 tween ‘the’ formal and ‘the’ informal economy. On the contrary,
40 there is an extensive and ever changing transitional area in which
41 the formal economy gradually transforms into a more informal one.
42 How can the emergence of immigrant businesses and their involve-
43 ment in the informal economy be understood?

Mixed embeddedness 1
2
Embeddedness has become a crucial concept in explaining the suc- 3
cess of entrepreneurs in general and that of immigrants in particular 4
(Granovetter, 1985; Granovetter and Swedberg, 1992; Portes, 1995a; 5
Waldinger, 1995; 1996; Rath, 1999b), in the latter case especially with 6
respect to informal economic activities as they take place outside the 7
regular framework (Epstein, 1994; Roberts, 1994). Embeddedness, 8
however, tends to be mainly used in a rather one-sided way, referring 9
almost exclusively to the social and cultural characteristics of groups 10
that are conceived a priori to consist almost solely of co-ethnics. 11
Using embeddedness in this circumscribed way, neglects the wider 12
economic and institutional context in which immigrants are inevi- 13
tably also inserted or embedded (cf. Cassarino, 1997; Rath, 1997; 14
1999b; Rath and Kloosterman, 1998a). We therefore propose to use 15
the more comprehensive concept of mixed embeddedness – a con- 16
cept that is much closer to the original meaning of embeddedness as 17
intended by Polanyi (1957) – encompassing the crucial interplay be- interplay 18
tween the social, economic and institutional contexts (Kloosterman 19
et al., 1998). In this view, the rise of immigrant entrepreneurship is, 20
theoretically, primarily located at the intersection of changes in socio- 21
cultural frameworks on the one side and transformation processes in 22
(urban) economies on the other. The interplay between these two 23
different sets of changes takes place within a larger, dynamic frame- 24
work of institutions on neighbourhood, city, national or economic 25
sector level. As such, relevant research into immigrant entrepreneur- 26
ship (and its relationship to informal economic activities) has to be 27
located at the crossroads of several disciplines (cf. Granovetter, 1994: 28
453; Martinelli, 1994: 487; Rath and Kloosterman, 1999). 29
The exact shape of the opportunity structure with respect to open- 30
ings for businesses that require only small outlays of capital and rela- 31
tively few educational qualifications constitutes a crucial component 32
in this mixed embeddedness. Market conditions determine to a very 33
large extent in which segments these kinds of openings occur. These 34
conditions have to be taken into account to explain (immigrant) en- 35
trepreneurship. Markets and economic trends themselves, however, 36
are embedded and enmeshed in institutions (cf. Esping-Andersen, 37
1990; 1996). Institutions such as the welfare system, the organiza- 38
tion of markets, the framework of rules and regulations together 39
with their enforcement, housing policies (impacting on the residen- 40
tial distribution of immigrants) and also business associations and 41
specific business practices which regulate particular markets signifi- 42
cantly affect opportunity structures at national, sector and local lev- 43

1 els (Freeman and Ögelman, 1999; Kloosterman, 1999; Kloosterman

2 and van der Leun, 1999; Rath, 1999b).
3 In the case of the Dutch corporatist welfare state, as in other con-
4 tinental European welfare states, the opportunity structure at the
5 lower end is – in marked contrast to the United States – curtailed
6 by relatively high minimum wages which choke the growth of low-
7 value added activities. These low-value added activities include loca-
8 tion-bound manufacturing such as sweatshops, but also potentially
9 booming post-industrial personal services such as child care and
10 housecleaning.
11 low-end opportunity Openings at the lower end of the opportunity structure do oc-
12 structure openings cur even in these highly regulated welfare states as invasion and
13 succession processes in neighbourhoods affect local businesses
14 (Kloosterman, 1999). Two types of processes by which openings
15 are created can theoretically be distinguished, although they tend to
16 blend in the real world. Firstly, openings are created by the emer-
17 gence of a demand for ‘ethnic’ products, such as specific clothing
18 and foodstuffs. Secondly, long-established, native shop owners leave
19 neighbourhoods where the number of immigrants rises and they
20 are replaced by immigrant entrepreneurs. Partly driven by the lack
21 of prospects as employees, and by the near absence of openings in
22 personal services, immigrant entrepreneurs in continental European
23 states flock towards these kinds of opportunities and set up shop in
24 especially wholesale, retail and restaurants (cf. Body-Gendrot, 1992;
25 Body-Gendrot and Ma Mung, 1992; Barrett et al., 1996). As we have
26 seen, this pattern is also found in the Netherlands (Figure 2).
27 Many of the markets where these kind of openings emerge will
28 be near saturation given the easy entry and the push to become self-
29 employed due to exclusion from the labour market. Inevitably, cut-
30 throat competition will evolve in these already shrinking markets.
31 Firms that operate in these markets at the lower end of the opportu-
32 nity structure, compete primarily on flexibility of supply and on price
33 rather than on quality. Hence, the most evident route to survival is
34 cutting (labour) costs. This strategy, however, is only partly feasible
35 within the prevailing regulatory framework. If one goes beyond
36 this framework, by, for instance, evading payment of taxes or social
37 contributions or by ducking the minimum wage and working-hour
38 regulations, the room to manoeuvre increases considerably. These
39 strategies – illegitimate as they may be – can be very profitable, as
40 production at the lower end of the opportunity structure is typical-
41 ly very labour intensive. The entrepreneurs may, moreover, tap re-
42 sources such as social capital. Through their networks of relatives,
43 co-nationals or co-ethnics they have privileged and flexible access

to information, capital and labour at relatively little (monetary) cost. 1

The use of social capital within the current opportunity structure 2
gives these businesspeople a competitive advantage, both within the 3
formal and informal economies. 4
Mixed embeddedness does not only refer to market conditions on 5
a more structural level. Immigrant entrepreneurs are also concretely 6
embedded in Dutch society in other ways as they operate in cities 7
with their own morphology, socio-economic, cultural and political 8
dynamics as well as in sectors with more or less established traditions 9
of doing business. Immigrant entrepreneurs in the Netherlands are 10
predominantly to be found in neighbourhoods with high shares of 11
immigrants. Due to the particular history of migration in conjunc- 12
tion with Dutch housing policies, levels of ethnic concentration in low levels of 13
most of these neighbourhoods remain rather low (Musterd, 1997). ethnic concentration 14
This implies that immigrant neighbourhoods in the Netherlands 15
have a very diverse immigrant population and cannot be equated 16
with American ‘ethnic neighbourhoods’. This diversity may reduce 17
the possibilities for immigrant entrepreneurs catering for a group- 18
specific demand by lowering the number of potential consumers in 19
the vicinity. 20
Being embedded in Dutch society may also refer to (voluntary or 21
obligatory) membership of organizations such as shop-owner asso- 22
ciations (based either on operating in a common line of business or 23
on being located in the same street). These organizations may pro- 24
vide mutual assistance and may also furnish a common set of largely 25
unwritten rules with respect to business practices. In Dutch society, 26
displaying its outspoken corporatist legacy, these kinds of organiza- 27
tion are quite important. They tend to protect the insiders – the al- 28
ready established entrepreneurs – at the expense of the outsiders – 29
the would-be entrepreneurs – by throwing up barriers of entry such 30
as minimum requirements with regard to the shop interior. 31
Configurations of mixed embeddedness may be very complex and 32
manifold. We now turn to a specific case of immigrant entrepreneur- 33
ship to investigate this mixed embeddedness in more depth. 34
35
36
The case of Islamic butchers 37
38
In their recent report to the Dutch Ministry of Economic Affairs, 39
Tillaart and Poutsma (1998: 50) counted 360 butcher shops run by 40
immigrant entrepreneurs in the Netherlands in 1997. These shops 41
are almost exclusively run by Moroccans (51%) and Turks (38%). 42
Islamic butchers are located at the lower end of the market where 43

1 openings are created by vacancy-chain processes of markets in neigh-

2 bourhoods where indigenous butchers quit the business. In the case
3 of Islamic butchers, the vacancies are, however, only part of the story.
4 Islamic butchers cater for a group-specific demand by selling hâlal
5 meat. Islamic dietary laws prescribe Muslims to refrain from eat-
6 ing pork and animals that have not been slaughtered according to
7 the Islamic rite. These products are considered unclean (haram) and
8 therefore strictly taboo.
9 Consequently, the arrival of hundreds of thousands of Islamic
10 immigrants, particularly immigrants from Turkey and Morocco,
11 from the 1960s onwards has had a substantial impact on the mar-
12 ket for meat in the Netherlands. In the early 1960s, when the first
13 Turkish and Moroccan guestworkers arrived, hâlal meat was obtain-
14 able virtually nowhere. This was, of course, due to the then extremely
15 small size of the market, but also to the fact that there were legal
16 impediments to Islamic slaughtering. Jews, long-term citizens in the
17 Netherlands, had already obtained special statutory arrangements for
18 slaughtering but these did not apply to Muslims. In order to meet
19 the demand for hâlal meat, immigrants from Turkey and Morocco
20 illegal butchers slaughtered animals illegally. These illegal butchers got caught every
21 now and then and were fined under the Law on Economic Criminal
22 Offences.
23 According to Bakker and Tap (1985: 37), the first Islamic butch-
24 ers set up shop in the late 1960s. They started without the proper
25 permits and in accommodation that hardly resembled that of regular
26 butchers. The demand for hâlal meat rose steadily, however, and in
27 the 1970s a few dozen Islamic butchers were already running their
28 businesses. It was not until 1975, that a small number of butchers re-
29 ceived temporary official permission to slaughter animals according
30 to the Islamic rite. This interim ruling was replaced two years later
31 by a more definitive regulation when the Ministry of Public Health
32 and Environmental Protection altered the Ministerial Order on Meat
33 Inspection (Vleeskeuringsbesluit) (Bakker and Tap, 1985: 36; Rath et
34 al., 1996).
35 In 1985, the total number of Islamic butchers amounted to
36 224, 138 of which were in Amsterdam, Rotterdam, The Hague and
37 Utrecht. Today, the number of Islamic butchers officially stands at
38 340. However, according to the Commodity Board for Cattle, Meat
39 and Eggs (Produktschap voor Vee, Vlees en Eieren) there are actually
40 more than 500 butchers, as some Islamic butchers work ‘on the sly’
41 (de Volkskrant, 8 March 1996). Some slaughterhouses supply private
42 customers too, while meat is also obtainable (informally) from coffee
43 shops and mosques.

Islamic butchers initially catered almost solely for a specific 1

ethno-religious clientele, mainly made up of Turkish, Moroccan, 2
Surinamese and other Muslims from other countries. This clientele 3
can be seen as a captive, but relatively stable market in the sense 4
that Islamic customers will rarely if ever buy their meat from native 5
Dutch butchers. Whether this will continue remains to be seen as 6
some Dutch supermarkets have started selling pre-packed hâlal meat 7
and have thus become potential competitors (de Volkskrant, 27 March 8
1996). 9
On the other hand, there seems to be a growing native Dutch cli- 10
entele. These customers, mostly living nearby in these mixed neigh- 11
bourhoods, want to prepare ‘exotic’ meals, are attracted by the low 12
prices or do not want to queue up in supermarkets. Some of the in- 13
digenous Dutch customers also appreciate the cutting of the meat in 14
their presence (Baetsen and Voskamp, 1991: 55-6). Islamic butchers, 15
however, are still trapped in severely limited markets as many native 16
Dutch customers do not feel encouraged to enter their shops, partly 17
because of doubts concerning hygiene. 18
These limited opportunities for market expansion are reflected in limited 19
the fact that only a few Islamic butchers are doing well. In the 1970s, opportunities for 20
on Fridays and Saturdays, numerous customers queued at their market expansion 21
shops; but these ‘golden years’ are over as the market has become 22
saturated with Islamic butchers. Too many butchers set up shop in a 23
neighbourhood, which subsequently leads to cut-throat competition. 24
A Turkish entrepreneur wonders: 25
26
Who starts a butcher or greengrocers where four others are perish- 27
ing? I do not understand how they can make a living, really I don’t. 28
But I do know three businessmen who are nearly bankrupt (quoted 29
in de Volkskrant, 4 June 1995). 30
31
This saturation of the market has led to high turnovers and a rela- 32
tively short average life span of Islamic butcher shops (Bakker and 33
Tap, 1985: 82). Most do not exist for more than three or four years. 34
Operating in saturated markets, Islamic butchers are clearly con- 35
strained on the demand side. They do possess, however, certain 36
competitive advantages with respect to their production costs in 37
comparison to indigenous Dutch butchers. Firstly, Islamic butch- 38
ers can make use of many more parts of a body of an animal than 39
their Dutch counterparts who mostly only sell legs and haunches. 40
Secondly, they can keep a smaller range of meat and related products 41
and they also invest much less money in their presentation and their 42
shop interior. The Trading Association of Butchers (Bedrijfschap 43

1 Slagersbedrijf) – a key institution in the field promoting the inter-

2 ests of butchers – voices the opinion that the shop fittings, the range
3 of products and the marketing strategy are rather ‘unusual’, ‘messy’
4 and ‘old-fashioned’. Thirdly, they accept smaller profit margins than
5 Dutch butchers do.
6 Faced with stiff competition, Islamic butchers in many cases re-
7 vert to cutting costs by paying workers off the books, by saving on
8 investments, or by selling products that are allegedly not in their line
9 of business like bread and other foodstuffs. They also cut corners by
10 insufficiently observing the Code of Hygiene for Islamic Butchers
11 (Groeneveld-Yayci, 1996). In particular, butchers reduce labour
12 costs, firstly, by only employing assistants during peak hours, espe-
13 cially on Friday afternoon. These predominantly male assistants –
14 females are rarely to be found behind the counter, they do mostly
15 back-office tasks – are generally recruited from their own group of
16 relatives and friends of co-ethnics and are in many cases employed
17 on an informal basis. Sometimes families enter into an agreement to
18 assist each other (Bakker and Tap, 1985: 110-12). The rewards can be
19 in financial terms, but also payment in kind or in terms of strength-
20 ening social relationships. The shop assistant, for instance, may be
21 allowed to bring foodstuffs home, learn the tricks of the trade as an
22 apprentice or invest in his social relations with his dad or uncle. The
23 butcher, for his part, gets the opportunity to invest in social relations
24 with people that might be beneficial to his enterprise either as poten-
25 relational tial employees, clients or suppliers. This constitutes a clear case of
26 embeddedness Portes’ relational embeddedness (Portes, 1995a).
27 These informal economic activities are not only enabled by so-
28 cial networks based on trust, but also by management practices that
29 contribute to obscuring what is going on in these butcher shops.
30 There are reports that financial management is in many cases to-
31 tally unsound. Financial reserves are practically non-existent, part-
32 ly due to the cut-throat competition, but also due to the inefficient
33 way of price-fixing. A (Turkish) counsellor working for the Trading
34 Association of Butchers commented that ‘those people’ lack the nec-
35 essary know-how in the field and ‘just mess around with the meat’.
36 They:
37
38 haven’t a clue about bookkeeping.... A kilo of minced meat for which
39 they themselves paid six guilders is sold in the shop for eight guil-
40 ders. They seem to forget completely that they have to pay taxes, lev-
41 ies and rent (quoted in de Volkskrant, 8 March 1996).
42
43

In the corporatist and highly regulated Dutch economy, every butch- 1

er is legally obliged to register at the Chamber of Commerce and the 2
Trading Association of Butchers and is also required to have proper 3
professional qualifications. Exemption from the legal requirements 4
regarding professional skills can be granted if the applicant supplies 5
a need that otherwise cannot be filled. Such exemptions have often 6
been granted. Initially, it was assumed that butchers were eligible 7
to exemption if catering to less than 1000 Islamic inhabitants in a 8
neighbourhood. In the meantime, this threshold has been raised to 9
2000 inhabitants for the first butcher in an area, 3000 for the second 10
one, 5000 for the third one and so forth.2 11
Butchers-to-be mostly do sign up at the Chamber of Commerce. 12
However, many of them do not have the proper professional quali- 13
fications, according to the Trading Association of Butchers. Due to 14
communication difficulties, inadequate counselling and the immi- 15
grants’ poor insight into the highly opaque Dutch bureaucracy, many 16
‘forget’ to apply for an exemption, making their enterprise infor- 17
mal outright. This also excludes them from support by the Trading 18
Association of Butchers. The Chambers of Commerce are not au- 19
thorized to close a shop down, while the Economic Control Service different forms 20
has given its priorities to other matters and does not take firm action of regulation 21
(anymore). De facto, Dutch authorities tend to turn a blind eye to this 22
kind of informal economic production by immigrant entrepreneurs. 23
Butchers can qualify for proper professional qualifications after 24
following courses at centres of the Butchers Vocational Training 25
(Slagers Vakopleiding SVO). Not all of the Islamic butchers who are 26
aware of this are willing or able to leave their (informal) shop for 27
these courses. In addition, for many candidates who did follow these 28
courses, the exam proved to be too difficult. The Dutch language 29
appeared to be an insurmountable obstacle and the same could be 30
said for the questions on pork (Tunderman, 1987: 23). In 1975, the 31
exam was replaced by a ‘professional test on the Mohammedan rite’.3 32
In the late 1980s, the Trading Association of Butchers announced 33
that it would provide training specific to immigrants (Tunderman, 34
1987: 24), but it was not until 1993 that the training really proceeded. 35
Since then the special policy regarding Islamic butchers – including 36
the right to obtain an exemption from the legal requirements – has 37
entered a new stage. The changes add up to a policy that is suppos- 38
edly more strict on these matters, although it is still possible to apply 39
for an exemption if the entrepreneur supplies a need that otherwise 40
cannot be filled. An Islamic butcher is now eligible to exemption if 41
the owner enrols in the Butchers Vocational Training and passes 42
its exam and if the manager receives the Training on Commercial 43

1 Practice (Cursus Handelskennis – see Handboek Minderheden, 1994:

2 8).
3 Whereas informal production by Islamic butchers is, at least
4 partly, tolerated by Dutch state agencies, the Trading Association of
5 Butchers has actively set out to reduce informal production by Islamic
6 butchers (and thus stamp out, in a typically corporatist fashion, what
7 is considered as unfair competition). They have appointed a special
8 councillor and set up a special Committee for Islamic Butchers, the
9 objective being to bring the Islamic butchers to ‘a higher standard of
10 quality’. By improving communication with Islamic butchers, pro-
11 moting the Butchers Vocational Training, and by making an invento-
12 ry of specific problems, the Trading Association of Butchers attempts
13 to combat informal economic practices. The association has entered
14 into discussion with the Chambers of Commerce about reducing the
15 ‘mushrooming’ of Islamic butchers. Furthermore, they have also
16 put pressure on the Economic Control Service as well as the Social-
17 Economic Council to enforce the laws more strictly and put a halt to
18 tolerating informal economic activities.
19 These endeavours are hampered by the fact that the Trading
20 Association of Butchers has not managed to organize Islamic butch-
21 ers yet. A key-informant of the Association told us that ‘these people
22 are not capable of organizing. It is not in their culture’. Although
23 this culturalistic statement is clearly at odds with evidence from
24 other fields (cf. Tillie and Fennema, 1997), it does show that there
25 is still a gulf between an established and formal institution like the
26 Association on the one hand and Turkish and Moroccan Islamic
27 butchers on the other.
28
29
30 Conclusions
31
32 An increasing number of immigrants from non-industrialized coun-
33 tries are starting businesses in advanced urban economies. Lacking
34 both in financial and human capital, many of these fledgling entre-
35 preneurs can only set up shop in specific segments of these urban
36 economies that allow for small-scale, labour-intensive, mainly low-
37 skill production. In the Netherlands, with its extensive welfare sys-
38 tem and its relatively high minimum wage, these kinds of openings
39 primarily occur in wholesale, retail and restaurants.
40 informal economic To survive in these mostly saturated markets, many (immigrant)
41 activities entrepreneurs cut corners by engaging in informal economic activi-
42 ties. This informal economic production can only take place on a
43 more permanent basis if a framework of trust exists. This trust can

be generated by social networks that are based on either a shared 1

migration experience or a shared non-indigenous identity. Because 2
of its link with social capital and its subsequent enabling of infor- 3
mal economic activities, the embeddedness of immigrants has quite 4
rightly come to occupy central stage in research into the socio-eco- 5
nomic aspects of immigration. 6
Focusing on embeddedness this way, however, tends to cloud 7
other aspects of embeddedness. Economic activities by immigrants 8
are situated in a wider institutional context. To a large extent this 9
institutional context determines on a macro-level the opportunity 10
structure for businesses in general. Firms are not only embedded in 11
these macro-economic structures but also in sets of rules and regu- 12
lations, neighbourhoods, associations and business traditions. To 13
address the socio-economic position of immigrant entrepreneurs in 14
general and in particular with respect to informal economic activi- 15
ties, one has to incorporate this side of embeddedness as well. We 16
have, therefore, proposed the use of mixed embeddedness to grasp im- mixed 17
migrant entrepreneurial activities. embeddedness 18
We have illustrated this mixed embeddedness by exploring the 19
case of Islamic butchers in the Netherlands. A complex configuration 20
of different types of embeddedness emerged. Islamic butchers in the 21
Netherlands are clearly located at the lower end of the opportunity 22
structure in openings that are partly vacancy-chain and partly group- 23
specific (hâlal meat) driven. They set up butcher shops in neighbour- 24
hoods with high shares of immigrants. Low barriers of entry and a 25
lack of opportunities in other segments funnels many would-be im- 26
migrant entrepreneurs towards this specific line of business. 27
The ensuing cut-throat competition in these highly saturated 28
markets puts pressure on the entrepreneurs to cut costs. This is part- 29
ly done in informal ways, for instance by selling meat off-the-books 30
and employing relatives who are (partly) paid in kind or not at all and, 31
moreover, by setting up a butcher shop without the necessary quali- 32
fications. This informal production is clearly linked to the fact that 33
Islamic butchers in the Netherlands benefit from being embedded in 34
social networks that mainly consist of co-ethnics and co-religionists. 35
These networks generate clients, employees, capital and trust, en- 36
abling them to start a business and engage in informal economic 37
practices. 38
These informal economic activities by Islamic butchers clearly 39
show the dynamic interaction between different domains of embed- 40
dedness. Although by definition unlawful, informal production is 41
to some extent tolerated by government agencies as they consider 42
Islamic butchers to be meeting a demand that otherwise would not 43

1 be met. Moreover, some rules and regulations are changed in such

2 a way that specific informal economic activities become formalized.
3 The case of Islamic butchers also shows that in the specific corporat-
4 ist context of the Netherlands, the role of business associations is very
5 important in embedding entrepreneurs. The Trading Association of
6 Butchers is much more active in combating informal production by
7 Islamic butchers than state agencies. Both immigrant entrepreneurs
8 and indigenous institutions are thus interactively negotiating new
9 territories and, hence, creating new forms of mixed embeddedness.
10 These dynamic processes of constructing new forms of mixed
11 ernbeddedness will be crucial in determining to what extent forms
12 of self-employment will constitute an avenue of social mobility in
13 post-industrial Netherlands. Islamic butchers are located in a specific
14 corner of a stagnant market selling hâlal meat in neighbourhoods
15 with high shares of immigrants. A successful trajectory of incorpora-
16 tion of immigrant entrepreneurs will – after having started on the
17 basis of being embedded in immigrant networks – largely depend on
18 the way they manage to become embedded in the overall Dutch con-
19 text. The case of the Islamic butchers shows that this is not a wholly
20 one-way process nor solely a government affair. Changing the mix of
21 a social process embeddedness is an open, contingent social process in which many
22 social actors may take part and on which the insertion of immigrant
23 entrepreneurs depends.
24
25
26 Notes
27
28 * This research project is part of Working on the Fringes: Immigrant
29 Businesses, Economic Integration and Informal Practices, a thematic
European network for exchange of knowledge and experiences. This interna-
30 tional network, coordinated by Jan Rath and Robert Kloosterman and funded
31 by the European Commission under the Fourth Framework, involves both
32 international comparison and collaboration with regard to research on im-
33 migrant entrepreneurs in Austria, Britain, France, Germany, Israel, Italy
34 and the Netherlands.
1 This relationship between the share of the immigrant population and the
35
ratio of immigrant business start-ups and indigenous business start-ups was
36 found to be statistically significant across all neighbourhoods in Amsterdam
37 and Rotterdam (excluding the centres). By using this ratio, other neighbour-
38 hood characteristics that may influence the number of business start-ups
39 in general (such as the availability of cheap business accommodation) were
40 eliminated and the focus was solely on the number of firms set up by im-
migrant entrepreneurs relative to those started by indigenous entrepreneurs
41 (see Kloosterman and Van der Leun, 1999).
42
43

2 See Handboek Minderheden (1994: 8). Recently it was decided that the Social- 1
Economic Council had to grant exemptions, but the Trading Association of 2
Butchers was still unfamiliar with its policy (see Groeneveld-Yayci, 1996). 3
3 There too, a separate set of regulations was drawn up, which was approved in
4
1977 by the Ministry of Economic Affairs (see Bakker and Tap, 1985: bijlage
III). 5
6
7
8
References 9
10
Baetsen, P. and J. Voskamp (1991) Kopen en verkopen op Zuid. Een 11
onderzoek naar de omvang, betekenis en ontwikkeling van het etnisch 12
ondernemen in Rotterdam Oud-Zuid. Stichting Werkgroep ‘2000, 13
Amersfoort. 14
Bakker, E.S.J. and L.J. Tap (Onderzoekers Kollektief Utrecht) (1985) 15
Islamitische slagerijen in Nederland; Verslag van een onderzoek in op- 16
dracht van Bedrijfschap Slagersbedrijf. Mededelingenreeks nr. 40, 17
Hoofdbedrijfschap Ambachten, Den Haag. 18
Barrett, G., T. Jones and D. McEvoy (1996) Ethnic minority business. 19
Theoretical discourse in Britain and North America. Urban Studies 20
33.4/5, 783-809. 21
Blankenburg, E. and F. Bruinsma (1994) Dutch legal culture. 2nd 22
edn., Kluwer Law Taxation Publishers, Deventer/Boston. 23
Body-Gendrot, S. (1992) Essai de définitions en matière de com- 24
paraisons internationales. Revue Européenne des Migrations 25
Internationales 8.1, 9-16. 26
– and E. Ma Mung (1992) Entrepreneurs entre deux mondes. 27
Revue Européenne des Migrations Internationales Special issue 8.1. 28
Cassarino, J.P. (1997) The theories of ethnic entrepreneurship, and 29
the alternative arguments of social action and network analysis. EUI 30
Working Papers 97/1, European University Institute, Florence. 31
Castells, M. and A. Portes (1989) World underneath. The origins, 32
dynamics, and effects of the informal economy. In A. Portes, M. 33
Castells and L.A. Benton (eds.), The informal economy. Studies in ad- 34
vanced and less developed countries, The Johns Hopkins University 35
Press, Baltimore. 36
CBS (1997) Allochtonen in Nederland 1997. Centraal Bureau voor de 37
Statistiek, Voorburg/Heerlen. 38
– (1998) Enquête Beroepsbevolking. Centraal Bureau voor de 39
Statistiek, Voorburg/Heerlen. 40
De Volkskrant (1995) De slag om de Javastraat. 3 June. 41
– (1996) Lammerramsj. 8 March. 42
– (1996) Super verkoopt ritueel geslacht vlees. 27 March. 43

1 Engbersen, G. (1997) In de Schaduw van Morgen. Stedelijke marginal-

2 iteit in Nederland. Inaugural address, Boom, Amsterdam.
3 Epstein, R.A. (1994) The moral and practical dilemmas of an under-
4 ground economy. Yale Law Journal 103.8, 2157-78.
5 Esping-Andersen, G. (1990) Three worlds of welfare capitalism. Polity
6 Press, Cambridge. Fainstein, S., I. Gordon and M. Harloe (eds.)
7 (1992) Divided cities. New York and London in the contemporary
8 world. Blackwell, Cambridge, MA.
9 Faist, T. (1997) International migration and transnational social spac-
10 es. A new theoretical approach and a case study. Paper presented at
11 the Second International MigCities Conference on Migrants and
12 Minorities in European Cities: The Dynamics of Social Integration
13 at the Neighbourhood Level, Université de Liège, 6-8 November.
14 Freeman, G.P. and N. Ögelman (1999) State regulatory regimes and
15 immigrants’ informal economic activity. In J. Rath (ed.), Immigrant
16 businesses. The economic, politico-institutional and social environ-
17 ment. Macmillan Press, Houndmills, Basingstoke, Hampshire
18 (forthcoming).
19 Granovetter, M. (1985) Economic action and social structure: the
20 problem of embeddedness.
21 American Journal of Sociology 91, 481-510.
22 – (1994) Business groups. In N. Smelser and R. Swedberg (eds.),
23 Handbook of economic sociology, Princeton University Press/Russell
24 Sage Foundation, Princeton/NewYork.
25 – and R. Swedberg (1992) The sociology of economic life. Westview
26 Press, Boulder, CO.
27 Groeneveld-Yayci, A. (1996) Knelpunten islamitische slagerijen.
28 Bedrijfschap Slagersbedrijf, Rijswijk.
29 Guarnizo, L.E. (1996) The Mexican ethnic economy in Los Angeles.
30 Capitalist accumulation, class restructuring, and the transnationaliza-
31 tion of migration. Unpublished paper. University of California at
32 Davis, Davis, CA.
33 Handboek Minderheden (1994) Zelfstandig ondernemerschap,
34 3/1100, 8.
35 Häußermann, H. and I. Oswald (eds.) (1997) Zuwanderung und
36 Stadtentwicklung. Westdeutscher Verlag, Sonderheft 17 Leviathan.
37 Kloosterman, R. (1996) Mixed experiences. Post industrial transi-
38 tion and ethnic minorities on the Amsterdam labour market. New
39 Community 22.4, 637-54.
40 – (1999) Immigrant entrepreneurship and the institutional con-
41 text. A theoretical exploration. In J. Rath (ed.), Immigrant busi-
42 nesses. The economic, politico-institutional and social environment,
43 Macmillan Press, Houndmills, Basingstoke, Hampshire (forth-
coming).

– and J. van der Leun (1999) Just for starters: commercial gen- 1
trification and immigrant business start-ups in Amsterdam and 2
Rotterdam. Housing Studies (forthcoming). 3
–, – and J. Rath (1997) Over Grenzen. Immigranten en de Informele 4
Economie. Het Spinhuis, Amsterdam. 5
–, – and – (1998) Across the border. Immigrants’ economic op- 6
portunities, social capital and informal business activities. Journal 7
of Ethnic and Migration Studies 24.2, 249-68. 8
Light, I. and C. Rosenstein (1995) Race, ethnicity, and entrepreneurship 9
in urban America. Aldine de Gruyter, New York. 10
Martinelli, A. (1994) Entrepreneurship and management. In N. 11
Smelser and R. Swedberg (eds.), Handbook of economic sociology, 12
Princeton University Press/Russell Sage Foundation, Princeton/ 13
New York. 14
Murie, A and S. Musterd (1996) Social segregation, housing tenure 15
and social change in Dutch cities in the late 1980s. Urban Studies 16
33.3, 495-516. 17
OECD (1992) Employment outlook 1992. OECD, Paris. 18
Polanyi, K. (1957) The great transformation. Beacon Press, Boston. 19
Portes, A. (1995a) The sociology of immigration. Essays on networks, eth- 20
nicity, and entrepreneurship. Russell Sage Foundation, New York. 21
– (1995b) Transnational communities. Their emergence and signifi- 22
cance in the contemporary world system. Working papers series no. 23
16, Johns Hopkins University, Baltimore, MD. 24
– and S. Sassen-Koob (1987) Making it underground. Comparative 25
material on the informal sector in western economies. American 26
Journal of Sociology 93.1, 30-61. 27
– and J. Sensenbrenner (1993) Embeddedness and immigration. 28
Notes on the social determinants of economic action. American 29
Journal of Sociology 98.6, 1320-50. 30
– and A. Stepick (1993) City on the edge. The transformation of 31
Miami. University of California Press, Berkeley/Los Angeles/ 32
London. 33
Rath, J. (1997) Ein ethnisches Bäumchen-wechsel-dich-Spiel in 34
Mokum? Immigranten und ihre nachkomen in der Amsterdamer 35
Wirtschaft. In J. Brech and L. Vanhué (eds.), Migration. Stadt im 36
Wandel, Verlag für Wissenschaftliche Publikations/Wohnbund 37
Publikationen, Darmstadt. 38
– (1998a) Een etnische stoelendans in Mokum. Over de econo- 39
mische incorporatie van immigranten en hun nakomelingen in 40
Amsterdam. In A. Gevers (ed.), Uit de Zevende. 50 Jaar Sociaal- 41
Culturele Wetenschappen aan de Universiteit van Amsterdam, Het 42
Spinhuis, Amsterdam. 43

1 – (1998b ) The informal economy as bastard sphere of social in-

2 tegration. In E. Eichenhofer and P. Marschalck (eds.), Migration
3 und Illegalität, Universitätsverlag Rasch, IMIS-Schriften Bd. 7,
4 Osnabrück.
5 – (l999a) A game of ethnic musical chairs? Immigrant busi-
6 nesses and the alleged formation and succession of niches in the
7 Amsterdam economy. In S. Body-Gendrot and M. Martiniello
8 (eds.), Minorities in European cities. The dynamics of social integra-
9 tion and social exclusion at the neighbourhood level, Macmillan Press,
10 Houndmills, Basingstoke, Hampshire (forthcoming).
11 – (ed.) (l999b) Immigrant businesses. The economic, politico-insti-
12 tutional and social environment. Macmillan Press, Houndmills,
13 Basingstoke, Hampshire (forthcoming).
14 – and J. Schuster (1995) Illegalen op de politieke en wetenschap-
15 pelijke agenda.
16 Migrantenstudies 11.2, 102-6.
17 –, R. Penninx, K. Groenendijk and A. Meyer (1996) Nederland en
18 zijn islam. Een ontzuilende samenleving reageert op het ontstaan van
19 een geloofsgemeenschap. Het Spinhuis, Amsterdam.
20 – and R. Kloosterman (l998a) Bazen van buiten. Naar een nieu-
21 we benadering voor het zelfstandig ondernemerschap van immi-
22 grant. In J. Rath and R. Kloosterman (eds.), Rijp en Groen. Het
23 Zelfstandig Ondernemerschap van Immigranten in Nederland, Het
24 Spinhuis, Amsterdam.
25 – and – (eds.) (1998b ) Rijp en Groen. Het Zelfstandig
26 Ondernemerschap van Immigranten in Nederland. Het Spinhuis,
27 Amsterdam.
28 – and – (1999) Outsiders’ business. Research on immigrant entre-
29 preneurs in the Netherlands. International Migration Review (forth-
30 coming).
31 Renooy, P.H. (1990) The informal economy: meaning, measurement
32 and social significance.
33 Netherlands Geographical Studies 115, Koninklijk Nederlands
34 Aardrijkskundig Genootschap/ Regioplan, Amsterdam.
35 Roberts, B. (1994) Informal economy and family strategies.
36 International Journal of Urban and Regional Research special issue
37 18.1, 6-23.
38 Sassen, A. (1991) The global city. New York, London, Tokyo. Princeton
39 University Press, Princeton, NJ.
40 Simon, P. (1997) The mosaïc pattern. The social system of cohabitation
41 between ethnic groups and social classes in Belleville, Paris. Paper pre-
42 sented at the International MigCities Conference on Migrants and
43 Minorities in European Cities: The Dynamics of Social Integration

and Social Exclusion at the Neighbourhood Level, Liège, Belgium, 1

6-8 November. 2
Tarrius, A. (1992) Les fournis d’Europe, migrants riches, migrants pau- 3
vres et nouvelles villes internationales. L’Harmattan, Paris. 4
– and M. Péraldi (eds.) (1995) Marseilles and its foreigners. Revue 5
Européenne des Migrations 6
Internationales special issue 11.1, 1-132. 7
The Economist (1996) The Netherlands: too good to be true? 12 8
October. 9
– (1998) A survey of world trade. 3 October. 10
The Independent (1997) Work can be done? 29 May. 11
Tillaart, H. van den and E. Poutsma (1998) Een Factor van Betekenis. 12
Zelfstandig Ondernemerschap van Allochtonen in Nederland. ITS, 13
Nijmegen. 14
Tillie, J. and M. Fennema (1997) Turkse organisaties in Amsterdam. 15
Een netwerkanalyse. Het Spinhuis, Amsterdam. 16
Tunderman, B. (1987) Concurrentie onder islamistische slagers. 17
Nieuwe opleidingen. Buitenlanders Bulletin 12.4, 22-4. 18
Waldinger, R. (1995) The other side of embeddedness. A case study 19
of the interplay of economics and ethnicity. Ethnic and Racial 20
Studies 18, 555-79. 21
– (1996) Still the promised city? African-Americans and new im- 22
migrants in post-industrial New York. Harvard University Press, 23
Cambridge, MA. 24
Wallace, C. (1997) Crossing borders. Mobility of goods, capital and people 25
in the Central European Region. Sociological Series No. 17, Institute 26
for Advanced Studies, Vienna. 27
Wentholt, R. (ed.) (1967) Buitenlandse arbeiders in Nederland. Een veel- 28
zijdige benadering van een complex vraagstuk. Spruyt, Van Mantgem 29
& De Does, Leiden. 30
Zhou, M. (1992) Chinatown. The socioeconomic potential of an urban 31
enclave. Temple University Press, Philadelphia. 32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
The mosaic pattern: cohabitation between 3
4
ethnic groups in Belleville, Paris
5
6
Patrick Simon 7
8
9
10
11
12
In the early 1990s, the demographer Patrick Simon studied the multiethnic
13
neighbourhood of Belleville, Paris. He analysed how different social groups
14
live together to constitute a complex network of relations organised around 15
cultural associations, community services or economic niches. Simon 16
showed how the various concrete forms of ethnic ‘cohabitation’ at the local 17
level cannot be adequately captured by integration models – e.g. assimila- 18
tion, multiculturalism – that always refer to the nation. Neither can this real- 19
ity be understood in bipolar terms such as whites versus blacks. In a local 20
context, the various social and ethnic groups need to negotiate their position 21
vis-à-vis the others and, in doing so, they follow a rationale that is not neces- 22
sarily compatible with national integration models. The division of urban, 23
political and symbolic space may, paradoxically, promote a certain degree 24
of social cohesion, provided that specific conditions be met for a sharing of 25
these divided spaces. 26
27
28
The image of Paris as a ‘cosmopolitan’ city is as old as Paris itself 29
is. However, only in the 1920s and 1930s did Paris earn its reputa- 30
tion of being a writers’ city, an ‘international republic of artists’, to 31
quote Alejo Carpentier. It became a centre of attraction for the intel- 32
ligentsia worldwide.1 After a period of decline, Paris has once again 33
become a centre of convergence for the world’s elite, a ‘global city’ a ‘global city’ 34
where international executives and financiers run the global econo- 35
my and redistribute the world’s resources. The City of Light owes its 36
cosmopolitan nature not only to its cultural and artistic aura, or to 37
its role in economic exchanges and technological innovation. It also, 38
and maybe even especially, owes it to the fact that from the end of the 39
nineteenth century onwards, immigrants from foreign countries and 40
from the provinces began to flow in massively, fostering an unprec- 41
edented economic and demographic boom. 42
As in all the other international metropolises, immigrants arriv- 43

340 patrick simon
1 ing in Paris were sorted and oriented towards different parts of the
2 segmented city. Throughout the twentieth century, Paris has consis-
3 tently been a centre of attraction and integration. The 1901 census
4 shows that, at the time, a little over 9 per cent of the population was
5 of foreign origin and 56 per cent had been born in the provinces.
6 In 1990, over 25 per cent of the population of Paris were of foreign
7 origin. Since 1982, the proportion of immigrants, which had risen
8 sharply between 1954 and 1975 (from 6 to 14 per cent), has not much
9 changed. As the presence of immigrants in the city increased, two
10 main transformations occurred. First, the origins of the migrants
11 changed as new waves of immigration followed in the wake of those
12 of the 1920s and 1930s. And second, the city’s functional reorganiza-
13 tion modified their distribution in space (Guillon, 1996). One can
14 identify a succession process according to the classical model estab-
15 lished by the urban ecologists of the Chicago school. The Italians,
16 Belgians and Poles who came in the 1920s were followed in the
17 1950s and 1960s by Algerian, Portuguese and Spanish immigrants.
18 Thus emerged the ‘ethnic neighbourhoods’, as they are now called,
19 and as a result, the immigrants became highly visible in the city.
20 multiethnic aspects The aim of this chapter, however, is not to provide a detailed list
21 of Paris’s immigrant neighbourhoods, but to study their multiethnic
22 aspects.2 One of the most striking characteristics of Paris’s ‘ethnic
23 neighbourhoods’ is that they bring together people of many different
24 origins. In addition to ethnic diversity, there is also a certain amount
25 of social diversity. Indeed, the social and symbolic value of neigh-
26 bourhoods with high immigrant concentration has changed since the
27 1970s. ‘Gentrification’ has led many middle- and upper-class house-
28 holds to move to immigrant and working-class neighbourhoods. For
29 these reasons, the bipolar model, such as the ‘whites versus blacks’
30 model, does not really apply to the patterns of segregation and co-
31 habitation observed in Paris. We must imagine a complex network
32 of relations involving many different groups that are more or less
33 organized around cultural associations, community services or eco-
34 nomic ‘niches’, and often circumscribed within a specific area. The
35 various ‘integration models’ – assimilation, multiculturalism, plural-
36 ism, melting-pot – whose context of reference is always the nation,
37 can thus be re-examined and contrasted with actual local situations
38 of ethnic cohabitation. Indeed, by analysing situations from a local
39 point of view, one can avoid the political implications of an analysis
40 of social interactions carried out at the national level. By looking at a
41 neighbourhood, we need not be concerned by questions of national-
42 ity and citizenship, which are of crucial importance in France. Their
43 importance there results from the historical significance of the na-
tion as a political concept in the organization of French society.

In a local context, social and ethnic groups negotiate their posi- 1

tion with their own rationale, which often differs from the national 2
integration model. To quote de Certeau (1986), theirs is a ‘pedagogy 3
of diversity’. We study ‘inter-ethnic’ situations to ‘analyse the condi- 4
tions for the creation of new sociopolitical contracts and in particu- 5
lar shed light on the internal transformations which occur within 6
a dominant group as a result of the presence of other groups’ (de 7
Certeau, 1986: 790-1). To illustrate this point, we have chosen to look 8
at ethnic and class relationships in a formerly working-class neigh- 9
bourhood of Paris, the Belleville quarter, which has become today an 10
emblematic place for several ethnic groups. Our approach is to study 11
the system of regulation of ethnic differences by examining how ur- 12
ban, political, symbolic space is shared between different groups that 13
play an active role on the local scene. We will show how this system, 14
by ensuring a certain degree of social cohesion, promotes the inte- 15
gration of the inhabitants into the city, if not into the nation. 16
17
18
The Belleville context 19
20
The Belleville quarter, one of Paris’s former working-class neigh- 21
bourhoods, is located in the eastern part of the city. It was urban- 22
ized at the end of the nineteenth century and its architecture is typi- 23
cal of working-class areas, with artisan workshops and low-quality 24
apartment buildings. By the early 1960s, the state of upkeep of these 25
buildings was so poor that Belleville had become one of the most in- 26
salubrious neighbourhoods in Paris. The first massive demolitions, 27
carried out in 1956, forced much of the native population out of the 28
neighbourhood. As a consequence of urban renewal, the area ceased 29
to act as a ‘shelter’ for needy people, as it had since the end of the 30
nineteenth century. We have chosen the term ‘shelter’ in reference shelter 31
to Belleville’s role in the wider context of the Paris area housing mar- 32
ket. Indeed, since housing in this neighbourhood was cheap, poor 33
households still wishing to remain in Paris could, as a last resort, 34
find affordable housing in Belleville. The low level of rents was also 35
due to the neighbourhood’s poor reputation. Belleville, home of the 36
lower classes, was considered a dangerous hideout for criminals and 37
political troublemakers, anarchists or communists; for this reason, 38
the area came to represent the epitome of all social ills. 39
Those who first came to live in Belleville were the households 40
evicted from the centre of Paris during Haussmann’s renovations in 41
the 1860s. This population was socially homogeneous: for the most 42
part skilled workers working in small artisan industries. In 1871, 43

342 patrick simon
1 during the Paris Commune, this working-class identity was empha-

2 sized: the actions of revolutionaries from Belleville gave the neigh-
3 bourhood the reputation of a hotbed of rebellion, a reputation it has
4 practically never lost since (Merriman, 1994; Jacquemet, 1984). As a
5 result, Belleville became a socially isolated area with a strong sense
6 of its own identity. In the 1920s, Armenians, Greeks and Polish Jews
7 began to move in. At this time, Belleville became the social and politi-
8 cal centre of the Yiddish and Armenian communities. Stores, work-
9 shops, cafés, places of religious worship or assembly, political news-
10 papers, Zionist, Bundist or communist discussion groups, common
11 interest groups, Jewish or Armenian trade unions formed a dense
12 and dynamic network of community organizations (Roland, 1962).
13 Yiddish and North During the 1950s, the neighbourhood’s ‘Yiddish period’ slowly
14 African period became history, while a new era of immigration dawned with the
15 arrival of massive contingents from Algeria; also came the Tunisian
16 Jews fleeing North Africa in the throes of decolonization. This new
17 wave marked the beginning of Belleville’s ‘North African’ period. At
18 the same time, the neighbourhood’s social composition was chang-
19 ing as French workers, who previously lived in insalubrious housing,
20 moved to the new public housing buildings at the city’s outskirts, or
21 to the suburbs. Immigrants from the transit housing projects or oth-
22 er forms of temporary housing then replaced them and, as a result,
23 the insalubrious housing stock remained permanently occupied.
24 Contrary to what is commonly thought, the departure of French resi-
25 dents was not caused by the arrival of immigrants; instead, the lat-
26 ter’s arrival was made possible by the departure of French residents
27 and the resulting vacancies. Between 1954 and 1982, in a context
28 where the overall density of the neighbourhood dropped consider-
29 ably, the population of French citizens fell to one-half of what it had
30 been (from 45,263 to 24,654), whereas the number of foreigners
31 doubled (from 4696 to 9470). The diversity of origins is quite im-
32 pressive. In 1990, the major groups were Algerians (15 per cent of
33 immigrants), Tunisians (15 per cent), sub-Saharan Africans (9 per
34 cent), Moroccans (8 per cent) and former Yugoslavs (7 per cent).
35 Asians, Turks and Sri Lankans complete the picture of Belleville as a
36 global village (Simon, 1993).
37 Added to the diversity of ethnic origins, there has been a recent in-
38 crease in the variety of socio-professional statuses. Whereas in 1954
39 the neighbourhood was essentially working class, the professional
40 profile of the working population is now changing. The gentrifica-
41 tion process began in 1980, after the partial renovation of several old
42 buildings and the launching of urban renewal programmes. Middle-
43 and upper-class households moved into new apartment buildings

and into existing buildings that were still in good condition; the nu- 1
merous new public housing programmes in renovated areas also 2
attracted new residents. The working-class population fell from 59 3
per cent in 1954 to 31 per cent in 1990, whereas the proportion of 4
liberal and upper-level professionals increased from 4 per cent to 13 5
per cent. 6
Over a period of 30 years, from 1955 to approximately 1985, the 7
neighbourhood underwent several population changes. The pace swift population 8
of these transformations was relatively swift, a fact that partially exchanges 9
plains why the recently arrived populations were able to take over 10
the area’s public space with such ease. Indeed, according to the par- 11
adigm of Elias and Scotson (1965), ‘established’ residents strongly 12
resist the efforts of new residents, or ‘outsiders’, to penetrate the vari- 13
ous spheres of local power. In most cases, the transfer of power from 14
one group to the other occurs over a long period of time. However, 15
in the case of Belleville, the massive departure of part of the popula- 16
tion led to the disappearance of traditional forms of neighbourhood 17
organization; the loss of original structures made it easier for the 18
newcomers to take over. This situation occurs quite frequently in 19
run-down neighbourhoods, before they are renovated (Coing, 1966). 20
Due to the departure of a portion of the ‘established’ population 21
and the ageing of another portion, many small businesses and arti- 22
san workshops closed and a large share of the quarter’s economic in- 23
frastructure was left vacant. Since, due to the neighbourhood’s bad 24
reputation, real estate prices were extremely low, commercial leases 25
became available to people who in normal circumstances would not 26
have been able to afford them. At the same time, immigrants began 27
to purchase property in ‘rundown’ apartment buildings. The fact that 28
the ‘native population’ of Belleville lost interest in the neighbour- 29
hood’s public social life is apparent today in the surprising visibility of 30
several ethnic groups. North African Muslims and Jews, Asians and 31
to a lesser extent Africans can be observed mainly in the local busi- 32
nesses and in the public space. Linked to ‘territorialization strategies’, 33
each group has created highly structured enclaves to serve its own 34
needs; they represent the organizational basis of ethnic cohabitation. 35
36
37
A fragmented area 38
39
Though Belleville as a whole ranks quite low in the hierarchy of 40
Parisian neighbourhoods, it is far from being socially and ethnically 41
homogeneous. At the local level, one can observe the same inequali- 42
ties in the distribution of social or ethnic groups as in the city overall 43

344 patrick simon
1 social and ethnic and as in its different parts. Thus, the middle and upper classes live
2 fragmentation in the high-quality apartment buildings of Belleville heights, where-
3 as the working classes and lower level staff live in the nether part
4 of the neighbourhood, in rundown buildings awaiting demolition.
5 Between 1954 and 1982, the area’s social geography changed as the
6 demolition programmes progressed. As a result of the demolitions,
7 the affordable housing space available to immigrants became scarc-
8 er, while the latter’s numbers increased. This led to the ‘crowding’
9 of many people into a small area, almost reminiscent of a ghetto,
10 unmarked by material boundaries but in fact strictly circumscribed,
11 owing to the pressure of the housing market.
12 Immigrants ended up all living in the same buildings because
13 they used family and community networks whose market was lim-
14 ited. Usually, upon their arrival, Algerian immigrants temporarily
15 settled in cheap hotels whose managers came from the same district
16 as they did (Sayad, 1977). Later on, when their families joined them,
17 they moved to neighbouring flats. A few years later, African immi-
18 grants followed the same itinerary, though the starting points were
19 hostels for migrant workers instead of cheap hotels. Community
20 networks also played an important role in helping immigrants from
21 former Yugoslavia or Portugal settle into vacant apartments with
22 their families. The Tunisian Jews were helped not only by family
23 and friends but also by community associations. The Unified Jewish
24 Social Fund3 helped ‘refugees’ who had had to flee Tunisia during the
25 political crises the country was going through after independence. A
26 strategy consisting of channelling the poorest fringe of immigrants
27 towards Belleville apparently led to the emergence of a ‘Tunisian
28 Jewish ghetto’ (Simon and Tapia, 1998). Finally, the Asians moved
29 into the renovated stock. The latter’s strategy involved property in-
30 vestments thanks to collective funding. Furthermore, special aid pro-
31 grammes also entitled Asian refugees to public housing space.
32 Despite these ‘channelized migration flows’, as B. Thompson
33 (1983) calls them, buildings are never wholly occupied by a single eth-
34 nic group. The distribution of apartments among immigrant groups
35 reflects their diversity, except in the case of hostels and cheap hotels.
36 Thus, at this level, the only really active type of segregation is social
37 segregation. Housing status is determined by income: there are no
38 upper-level professionals living in rundown buildings. Conversely,
39 very few members of the working class can afford to live in renovated
40 buildings with amenities, even if these buildings belong to the public
41 housing stock. From one building to the next, the difference in rent
42 can range from one to ten! Insalubrious buildings thus house immi-
43 grants of all origins, and their only ‘native’ neighbours are working

class. The mixing of different immigrant groups is thus reinforced 1

by social segregation. 2
Although the different ethnic groups tend to mix inside the resi- 3
dential area, the more dynamic groups have divided up public space 4
through a strategy of occupation and control. This strategy is based 5
on the presence of numerous businesses managed by members 6
of the community and aimed mainly at meeting the community’s 7
needs. The shops are used as identity markers (Raulin, 1986): the identity markers 8
shop windows convey specific signals (through signs, displays, and 9
linguistic and colour codes) and sell specific products. When several 10
shops belonging to the same ethnic group are located side by side, 11
they constitute a continuous ‘area’ through the repetition of these 12
‘community markers’. ‘Centres of activity’ thus develop around the 13
business areas with community services, leisure clubs, cafés, cultur- 14
al centres, doctors and places of religious worship. In Belleville, not 15
only are there many shops and businesses, but these are playing an 16
important role in establishing a community’s territory. For example, 17
out of 86 shops in the lower Belleville area, 46 can be considered 18
‘ethnic’ in the sense that they carry mainly imported products that 19
are sold in a specific decor or display according to specific, culturally 20
determined selling practices. (For a description of ethnic shops, see 21
de Rudder, 1987.) Of these 15 belong to the ‘exotic’ type, meaning 22
that though their clientele is not restricted to a single ethnic group, 23
they still refer to a specific culture, visible in the shop windows and 24
on the signs. The three largest communities of the neighbourhood – 25
Sephardic Jews (mainly from Tunisia), Southeast Asians and North 26
African Muslims – manage two-thirds of the local stores. 27
The Sephardic Jewish neighbourhood is located in a small 28
area between the Ramponneau and Dénoyez streets and along the 29
Boulevard de Belleville. Originally, it was much larger, but renova- 30
tions and the departure of part of the Tunisian Jewish community 31
have reduced the ‘little Goulette’ of Paris to its tiny dimensions. 32
Jewish commercial activity in this neighbourhood is linked mainly 33
to the food industry, with kosher butchers, oriental bakeries and gro- 34
cery shops. There are ‘bazaars’ that sell kitchen utensils and vari- 35
ous plastic items, a religious bookshop and several services. Most 36
members of this community participate in its overall economy and 37
its social aid programmes take care of many of them. Several Jewish 38
community organizations are located in the neighbourhood, such as 39
the Paris Jewish social action centre and a Lubavich centre, which 40
has opened two schools in the area. Thanks to the community as- 41
sociations, Belleville is both a commercial and a cultural centre and 42
this enhances its attractiveness for the Tunisian Jewish community 43

346 patrick simon
1 in Paris and entrenches their presence in the area even if most have
2 moved. Thus, Jewish clients who live elsewhere come to shop in local
3 stores. Before religious feasts, as many as 55 per cent of the stores’
4 clients come from other neighbourhoods or from the suburbs.4 The
5 commercial infrastructure is an extremely important factor in a com-
6 munity’s visibility: not only do ‘ethnic’ stores mark the neighbour-
7 hood with their presence, but they also make the community seem
8 larger than it actually is.
9 the Arab city On the other half of the Boulevard de Belleville is the Arab city; its
10 restaurants and grocery stores look very much like those of the Jewish
11 sector, except that the butcher shops are no longer kosher but halal.
12 Mosques have replaced the synagogues and Muslim skullcaps the
13 Jewish kippas. The cheap Kabyle hotels of the 1950s have gone; they
14 have been replaced by a profusion of stores mainly centred on food
15 distribution. This shopping area, which spreads from Ménilmontant
16 to the Père Lachaise cemetery, includes bazaars, cafés, restaurants,
17 travel agencies, secondhand clothing stores, import-export offices,
18 grocery stores, butcher shops and fruit and vegetable stores. In addi-
19 tion to these ordinary commercial activities, a centre of Muslim activ-
20 ity has developed near the Couronnes metro station. Two mosques
21 have been opened there, along with several religious bookshops. In
22 this area, meat sold as strictly halal is under very strict control. Kepel
23 (1984: 190ff.) calls this neighbourhood ‘Paris’s Islamic quarter’. It
24 is controlled by the Tabligh, who are members of the international
25 movement jama’at al tabligh (faith and religious practice).
26 In the Muslim sector, except on market days, far fewer women
27 than men are seen on the streets. The men gather in small tight-knit
28 groups in the central square where Belleville’s market stands are set
29 up twice a week. These groups are often extremely dense, with very
30 little space left unoccupied. The presence of North African Muslims
31 is most noticeable during Ramadan, in which the whole neighbour-
32 hood becomes involved. Social control reaches its highest point dur-
33 ing this period when a Muslim, or a person considered as such, can-
34 not be seen drinking or smoking during the day; if he does, more or
35 less aggressively voiced reprobation will force him to stop. However,
36 Muslims are not the only people concerned with Ramadan: the entire
37 Belleville neighbourhood cannot help but participate in preparations
38 for the feast. Vendors set up shop along the boulevard pavement and
39 sell flat bread, herbs, fruit and sour milk. Shops held by Muslims
40 add special Ramadan products to their usual display. Even Jewish
41 shopkeepers stock up on fruit and drink for the occasion.
42 North African Muslims and Jews have a lot in common, and this
43 is particularly evident when one looks at their economic activities.

Many kosher restaurants employ Muslim waiters, who for a long 1

time made it possible for them to open on the Sabbath.5 The majority 2
of the Jewish bakeries employ Muslims. After emigration, the rules 3
governing the cohabitation of the two communities in North Africa 4
(Lewis, 1986; Memmi, 1974) were reactivated. Tunisian Jews very 5
often speak Arabic, and the memories of Jews and Muslims comple- 6
ment each other within a single North African identity, recognized 7
as such by both groups. They agree to identify the neighbourhood as 8
‘Maghrebian’, meaning neither Jewish nor Muslim. 9
The Asian area was at first limited to the renovated sector of the rue Asian area 10
de Belleville; it subsequently rapidly spread to neighbouring streets.6 11
Asian businesses are extremely varied in nature and meet most of 12
the Asian community’s needs: they include food shops, jewellers, 13
supermarkets, record and video shops, restaurants, bakeries, estate 14
agents, wholesale dealers in fabrics for clothing and leather goods 15
manufacturers. In addition to these businesses, there is a dense net- 16
work of community services, including doctors, letter writers, leisure 17
clubs, cultural associations, and formal and informal information 18
networks. Although the Asians first settled in Belleville at the end of 19
the 1970s, their presence became significant only in the mid-1980s. 20
One reason for their choice of this area was that the Asian quarter 21
of the thirteenth arrondissement was reaching saturation point. The 22
strategy of implantation in Belleville just about matched that applied 23
in the ‘Choisy triangle’: their arrival coincided with urban modern- 24
ization programmes (Raulin, 1988). This ‘penetration’ phase, when 25
Asians began to move into the neighbourhood, mainly into recently 26
built housing, was followed by a ‘consolidation’ phase with the de- 27
velopment of community-oriented businesses. These businesses at- 28
tracted other Asians to the neighbourhood, and many in turn ended 29
up moving there. Between 1982 and 1990, the Asian population in- 30
creased by 63 per cent, the highest increase after that of the Turkish 31
population (76 per cent). 32
The non-Asian shopkeepers feel threatened by the Asian commu- 33
nity’s vitality and expansionist drive, but so far no collective solution 34
enabling them to ensure their own survival has been devised. There 35
has been little group reaction to the massive implantation of Asian 36
businesses, which is so extensive that Belleville is now considered to 37
be Paris’s second ‘Chinatown’. Despite their commercial expansion- 38
ism, there are few Asians in Belleville’s other areas, and it is only at 39
the points of contact between areas that they mix with other groups. 40
This strategy of isolation, though not specifically Asian, tends to sup- 41
port the stereotype of a secretive community that keeps to itself and 42
is unwilling to conform to the neighbourhood’s social order (Live, 43

348 patrick simon
1 1993). The other groups, which consider themselves much poorer

2 than the Asians, are exasperated by their real or imagined financial
3 power. They are envious of their sense of solidarity, thanks to which
4 Asians are much more successful in the interethnic competition
5 than their partners or rivals. However, the population increase in
6 Belleville has led Asian households to disperse, after an initial period
7 of concentration. Thanks to personal contacts made in their residen-
8 tial context, Asians are perceived as individuals instead of simply
9 members of an ethnic group. As grossly simplified ethnic divisions
10 break down and are replaced by daily exchanges - which involve ne-
11 gotiation - the Asian population is gradually adapting to the common
12 social order.
13 Within this overall commercial structure, various other ethnic
14 groups have opened businesses: there are several Spanish grocery
15 stores and restaurants, one or two African restaurants and an increas-
16 ing number of Turkish small businesses, mainly ‘fast food’ outlets
17 (for example, pizza and doner kebab restaurants). Relics of the previ-
18 ous era, the few remaining French-owned shops, are located mainly
19 at the corner of the rue de Belleville and the Boulevard de Belleville.
20 They remain isolated amid the stars of David, Chinese ideograms
21 and Arabic characters and have no influence at all on the atmosphere
22 generated by the dominant groups. The municipality’s renovation
23 plans have included attempts to establish new commercial activities
24 aimed at modifying the neighbourhood’s image, which the authori-
25 ties perceive as too ‘immigrant’. All new apartment buildings include
26 commercial space, but so far Asians or North Africans lease them all.
27 This demonstrates that both communities are trying hard to main-
28 tain their presence in Belleville and that this strategy has won over
29 the municipality’s attempt to ‘requalify’ the neighbourhood.
30 stratified structure Public space is thus appropriated by means of easily identifiable
31 markers: buildings, facilities and other public places are marked off
32 as belonging to a specific, almost private, territory. Those who share
33 its ‘identity’ frequent this territory. These identity signs or mark-
34 ers can be read in shop windows, in the way housing space is oc-
35 cupied, in the playing out of social relationships, or even in people’s
36 personal attributes (such as their clothing and personal demean-
37 our). Schematically speaking, the spatial and social morphology of
38 Belleville is a juxtaposition of ethnic strata, alternately dominated
39 by one or another of the ethnic groups. The strata themselves are
40 first the buildings, then the streets, then the shops, cafes and parks,
41 and finally the whole picture is crossed by a transversal stratum rep-
42 resented by community associations and political groups. The way
43 the various groups adjust to this stratified structure determines the
Belleville cohabitation model.

Cohabitation models 1
2
Now that the framework for our analysis has been established, we can 3
revert to our initial question: how does integration work in Belleville? integration 4
The restriction of certain ethnic groups to a circumscribed territo- 5
ry, the public display and even the exacerbation of one’s specificity, 6
whether religious (Islamic fundamentalism or Jewish orthodoxy) or 7
cultural, are in contradiction with the ‘French model of integration’. 8
According to this model, integration is an individual process enabling 9
immigrants to participate in the activities of mainstream society on 10
condition they accept its rules and that the society in turn is prepared 11
to integrate the immigrants.7 This process is based on a strict dis- 12
tinction between private and public spheres. In the private sphere, 13
cultural specificities can be maintained if they do not contradict the 14
fundamental ‘values’ of the Republic. In the public sphere, however, 15
one must remain ‘neutral’ or, in other words, one’s behaviour must 16
be in conformity with the norms of mainstream society. 17
What is the situation in Belleville? Here, cultural differences, in- 18
stead of being downplayed, are emphasized and play an important 19
role in the definition of relations between the various ethnic groups. 20
Far from being neutral, public space is the object of competition for 21
control over it; but instead of being a cause for social disorder, this 22
competition ensures social stability. Ever since the French working 23
class ceased to be the dominant group in the area, no other group has 24
been able to impose its norms of values on the others. The concept 25
of normative behaviour is no longer relevant, and has been replaced 26
by a much more general attitude based on tolerance and respect of 27
proprieties. Social order in Belleville8 is based first and foremost on 28
a charter of practices devoid of ethnic or cultural references. To use 29
a popular cliché in studies on integration, Belleville’s social order is 30
universalist in both spirit and practice. 31
The coexistence of these groups within a circumscribed area has 32
led to a division of the neighbourhood into small plots. To describe 33
the spatial organization of the groups living in Belleville, the most ac- 34
curate image is that of a mosaic, ‘separate and closed-in worlds which 35
exist side by side but do not mix’, to quote R.E. Park (1925). Each ur- 36
ban segment has its own ‘local colour’ and the atmosphere can differ 37
completely from one street to the next. Each area has its users who 38
feel at home in its atmosphere and contribute, by their presence, to 39
spreading it. These ‘microenvironments’, in which urban functions, microenvironments 40
users and specific practices are combined, are undoubtedly ‘quasi- 41
communities’ (Gans, 1962). The division of space must not be inter- 42
preted as a sign of hostility between the different groups. Indeed, it 43

350 patrick simon
1 is the only way these groups can use the city while maintaining their
2 own specificity. Without such borders, ethnic groups could not keep
3 the distance necessary for them to be able to live together. At the
4 same time, thanks to these borders, which are constantly shifting, a
5 group can define itself in opposition to the ‘others’, as Fredrik Barth
6 (1969), whose book has become a work of reference, has pointed out.
7 As competition for space is high, conflicts can only be regulated
8 if compensation is provided to those groups that are not present on
9 the public scene. If one considers the city according to three impor-
10 tant aspects – urban, political and symbolic – the sharing of space
11 requires that a considerable number of elements be taken into ac-
12 count. Thus, added to the issue of concrete urban space, there is the
13 neighbourhood’s history and collective memory, and in parallel, the
14 political forces and the associations that control the terms of this divi-
15 sion: three distinct yet interlinked spheres of action, whose collective
16 actors may differ. If an actor ceases to participate at one level, his
17 participation may increase at another.
18
19
20 The myth and the multiculturals
21
22 To create this system, history had to be rewritten and the collective
23 memory condensed into a ‘Belleville myth’. The myth has made it
24 possible to create a common area, open to all, and to transcend deep-
25 ‘imagined ly ingrained cultural specificities. The myth has created the ‘imag-
26 community’ ined community’ B. Anderson (1983) described when speaking of
27 nations. Here it is, in a few words. The Belleville myth is based on
28 two assertions: ‘Belleville is an old working-class neighbourhood’
29 and ‘a neighbourhood where immigrants first settled long ago’.
30 These two assertions are of course based on historical fact, but the
31 latter has been modified, in the spirit of what Roland Barthes (1957)
32 called ‘the naturalization of history’. The elements that constitute the
33 Belleville myth are no doubt historically true. But, and it is this sense
34 that a myth has been created, they had neither the impact nor the
35 importance they are believed today to have had. Thus, Belleville is
36 not an old immigrant neighbourhood. Quite the contrary, censuses
37 from the first half of the twentieth century show that Belleville then
38 had the highest proportion of Parisian natives in the city. The im-
39 migrant presence in Belleville has never been as strong as it is today.
40 Similarly, although Belleville was a working-class neighbourhood
41 until the 1970s, this was no longer the case at the time the myth crys-
42 tallized. What is the function of this myth and who perpetrates it?
43 A myth is defined first and foremost by its aim, which is usu-

ally the desire to overcome contradictions. The aim of the Belleville 1

myth is to defuse ethnic conflicts by making them seem outdated. As 2
Claude Levi-Strauss (1958: 231) said: 3
4
A myth is always based on events which occurred in the past: ‘before 5
the creation of the universe’, or ‘at the beginning of time’, in any 6
case, ‘a long time ago’. But the myth’s intrinsic value comes from 7
the fact that these events, which took place at a given time, create a 8
permanent structure; this structure determines the past, the present 9
and the future. 10
11
By associating the immigrants with the neighbourhood’s collective 12
memory, the myth acts as a ‘nativity factory’; thus, ethnic conflicts ‘nativity factory’ 13
cannot be based on the refusal of one group to accept the other’s 14
presence, since they both equally belong to the neighbourhood. In 15
other words, using Elias and Scotson’s (1965) paradigm, thanks to 16
this myth, immigrants cease to be outsiders and can aspire to the 17
more legitimate status of the established. Thanks to the contraction 18
operated by the myth, attitudes of intolerance and rejection, which 19
are often observed in situations where local residents emphasize 20
their cultural specificity, become totally irrelevant. 21
The myth also concerns relations between social classes. By lay- 22
ing emphasis on the neighbourhood’s identity as working class, it 23
aims to make up for the social inequalities reflected in the housing 24
conditions. Acceptance of this myth represents, for members of the 25
middle and upper classes, a guarantee of their own integration into 26
the neighbourhood. Even more so, they play a significant role in cre- 27
ating and spreading the myth, in particular through the action of 28
La Bellevilleuse, a local residents’ association devoted to fighting the 29
neighbourhood renovation programme. 30
Local residents wishing to weigh upon decisions about the lower 31
Belleville area’s renovation programme created the association in 32
1988. Today, it has 500 members, mainly from the recently settled 33
middle and upper classes. Participation in this neighbourhood asso- 34
ciation enables them to express, through militant action, their faith 35
in a certain vision of society. Furthermore, they take an active part in 36
local politics and play a crucial role as intermediaries between society 37
as a whole (represented here by the public authorities and the techni- 38
cal services of the City of Paris) and the minority groups. Because of 39
their strong attachment to ethnic, cultural or social ‘mixing’ or diver- 40
sity, these new residents may be called ‘multicultural’. Their commit- 41
ment to collective action, aimed at defending the right of immigrants 42
and the working class to remain in Belleville, can be interpreted on 43
two levels.

352 patrick simon
1 By insisting on people being rehoused in the same neighbourhood,

2 the ‘multiculturals’ anticipate changes in Belleville’s population: they
3 wish to prevent the too rapid gentrification of the neighbourhood
4 and preserve the atmosphere they came for in the first place. Thus,
5 they have become the advocates of a working-class identity, which
6 is not theirs but for which they feel sympathy. They are themselves
7 often of working-class background, and participation in community
8 action is a manner of ‘reparation’. The aim is to promote, at the local
9 level, a social model that has not taken shape at the national level.
10 The commitment of the ‘multiculturals’ has provided the working
11 class with a new edge in power relations. Indeed, when dealing with
12 the authorities, immigrants and French workers are usually deprived
13 of means of pressure; the ‘multiculturals’ are thus able to serve as
14 mediators, which is what they did in relation to the neighbourhood
15 renovation programme. On a wider scale, their role as mediators has
16 enabled them to create a more positive image of a social world that
17 so far had been perceived as impoverished and pernicious. Through
18 their joint reaction of protest against the renewal programme and the
19 bureaucratic monster that supports it, the neighbourhood’s different
20 groups were able to get together symbolically and, to a certain extent,
21 to come closer operationally. W. de Jong (1989) described a similar
22 process in an old neighbourhood of Rotterdam, ‘Het Oude Westen’,
23 which resembles Belleville in many respects. There, ethnic conflicts
24 were overcome thanks to associations of local residents committed to
25 preventing the deterioration of their neighbourhood.
26 regulation of The Belleville model can thus be seen as a successful system of
27 differences regulation of differences; these differences are asserted within sepa-
28 rate and structured ‘communities’ and expressed in ‘community
29 areas’, which are interlinked without competing one against another.
30 Urban space is identified as belonging to North African Jewish or
31 Muslim immigrants, to Asians and, to a lesser extent, to Africans.
32 Even though they do not have their specific ‘turf’, the ‘native’ resi-
33 dents, that is to say the French workers, who represent the neighbour-
34 hood’s living memory, are a significant component of the Belleville
35 identity. Last, the recently-arrived middle and upper classes, which
36 have the financial means and the extremely valuable ability to cir-
37 culate with ease in the world of social relationships and contacts,
38 have a specific role to play in the sphere of political and community
39 action. In Belleville, each person has a place, has his or her own place
40 within a dynamic system that is constantly changing. Only on this
41 condition can people overcome their objective differences and share
42 a strong local identity. To describe this model, we chose to compare
43 it with a mosaic, a composite image that refers to a surface made up

of assembled pieces as well as to the political system of the ancient 1

Ottoman Empire. Belleville can be compared with both: on the one 2
hand, it is made up of juxtaposed, heterogeneous parts and, on the 3
other, the Ottoman Empire is part of the historical and political back- 4
ground of two of the neighbourhood’s main groups. In this respect, 5
the Ottoman Empire, as an attempt to reconcile different cultures 6
within a unified political system, represents a historical precedent, 7
which has yet to be studied in all its implications (Courbage and 8
Fargues, 1992; Valensi, 1986). 9
The mosaic model owes its existence to historical circumstances 10
in which different population groups going different ways found 11
themselves at the same time in the same place. Many immigrants 12
who managed to improve their social status moved out of the in- 13
salubrious buildings, whereas rehabilitation programmes gradually 14
evicted others. The gentrification process has increased in scope and 15
is now reaching out for the last fragments of territory still accessible 16
to immigrants. Belleville is undergoing a gradual transformation, 17
from ‘ethnic neighbourhood’ to ‘urban immigrant centre’ (espace de 18
centralité immigrée) (Toubon and Messamah, 1991). Even when the 19
members of a community move to another area, they maintain their 20
ties with Belleville, which continues to develop its community-orient- 21
ed economic, cultural and social activities: the area is thus becoming 22
a centre of attraction for both symbolic and practical reasons. This 23
phenomenon of ‘territorial dissociation’, which is characteristic of a 24
‘networked’ society, has been observed in several ethnic neighbour- 25
hoods in Paris, such as the Goutte d’Or (Toubon and Messamah, 26
1991) and the Choisy triangle (Raulin, 1988). This new function 27
seems to be a new stage of ‘transitional area’, or rather, to use the 28
term Ernest Burgess (1928) coined, of ‘first entry ports’, which en- first entry ports 29
able immigrants gradually to adapt to their new society without expe- 30
riencing a total break with their past way of life. The future of these 31
neighbourhoods remains uncertain; the opinion most commonly 32
held is that they will disappear through acculturation. In our opinion, 33
this is not happening in Belleville. Thanks to new forms of ‘distance 34
shopping’ practised by both the older and more recent diasporas - 35
immigrants from Southeast Asia, Ashkenazi and Sephardic Jews, 36
Armenians, North Africans, Africans, Turks and others - ethnic ter- 37
ritories can remain a permanent aspect of the urban environment. 38
They can perhaps even serve as a basis for the elaboration of a com- 39
munity structure of national scope. 40
41
42
43

354 patrick simon
1 Notes
2
3 1 Cf. Hemingway’s celebration of Paris in Paris est une fête, quoted in Ory
4 (1994).
2 This approach owes a great deal to the pioneer (in France) research work
5 carried out by V. de Rudder, M. Guillon and I. Taboada-Leonetti. They fo-
6 cused on multiethnic cohabitation in several neighbourhoods of Paris (the
7 Choisy neighbourhood in the thirteenth arrondissement, the Aligre and Lot
8 Chalon neighbourhoods, and the wealthy neighbourhoods of the sixteenth
9 arrondissement). Summing up the team’s approach, Taboada-Leonetti (1989)
writes: ‘Our aim was to carry out empirical studies to show how people man-
10
age their differences in an ad hoc manner, depending on the issues at stake
11 and the circumstances, and how they produce collective identities which can
12 vary from one situation to the next without necessarily generating social cri-
13 ses, social dysfunction or ethnic identity crises.’
14 3 Unified Jewish Social Fund: this is the main source of funding supporting
15 the various Jewish cultural, social and community institutions in France.
4 Survey conducted in front of shops in Belleville for a study on economic
16 activity in the lower Belleville area (see Fayman and Simon, 1991).
17 5 The religious revival, which has affected the Jewish community in France,
18 was also felt in Belleville. Today, most kosher stores close on the Sabbath.
19 6 A detailed map of Asian businesses in Belleville can be found in Ma Mung
20 and Simon (1990: 99). However, this map dates back to 1985 and the neigh-
bourhood’s business infrastructure has changed considerably since then.
21
More recent information is available in Live (1993).
22 7 This formulation is a condensed synthesis of the definitions of integration as
23 given by two official sources; the Commission de la Nationalité (1988) and
24 the Haut conseil à l’intégration (1991).
25 8 The notion of ‘local social order’ refers to the one G. Suttles formulated
26 about a slum in Chicago. Even though those who live there have been re-
jected by mainstream society as ‘people with disreputable characteristics’,
27
slums are not ‘disorganized’ (Suttles, 1968). Social order is interpreted here
28 as a system of rules, norms and values making it possible for different social
29 groups, which are interdependent yet reject each other, to live together. In
30 Belleville, where residents belong to very different ethnic or social groups,
31 the neighbourhood stands for a reference. Since all these groups live in the
same area, to get along, they must develop a common code of behaviour for
32
the neighbourhood.
33
34
37
38
39
40
41
42
43

1
2
Political dynamics in the city: 3
4
three case studies
5
6
Hassan Bousetta 7
8
9
10
11
12
The article by political scientist Hassan Bousetta first appeared in 2000 in 13
an edited book entitled Minorities in European Cities. Bousetta’s doctoral the- 14
sis, from which this article is drawn, is one of the first systematic qualitative 15
comparisons of the collective dynamics, the socio-political participation and 16
the ethnic mobilisation of immigrant minorities in three mid-size European 17
cities. It was followed in the 2000s by several other studies using a similar 18
theoretical framework and an analogous research methodology. Bousetta’s 19
work is considered pioneering in the field of comparative studies of immi- 20
grant associations in Europe. 21
22
23
This chapter on the collective dynamics, sociopolitical participation 24
and ethnic mobilization of immigrant minorities is based on com- 25
parative case studies of Moroccan communities in three small and ethnic mobilization 26
medium-sized cities in Belgium (Liège), the Netherlands (Utrecht) 27
and France (Liège). Three main ideas inform the design and ratio- 28
nale of this research. 29
The first is that immigrant incorporation is increasingly being 30
shaped by socioeconomic and political dynamics at work locally. In 31
this age of postindustrial transition, inter-ethnic relations are increas- 32
ingly entangled with broader social and economic phenomena affect- 33
ing cities. In countries like France, the Netherlands and Belgium, 34
this is reflected in patterns of policy management of ethnic diversity. 35
The policy interventions of these countries’ public authorities have 36
gradually begun to address the socio-spatial dislocations confronting 37
urban areas. A significant feature of European governments’ poli- 38
cy response to urban decline and immigrant integration has been 39
to decentralize power to local authorities. Whereas migratory flow 40
regulation remains a matter for governmental and European inter- 41
governmental approaches, the integration part of migration policies 42
is often tailored to fit immigrant policy issues emerging in the big 43
cities.

356 hassan bousetta
1 Second, it is important to emphasize that migrant communities

2 are not necessarily at the forefront of the new relationship between
3 economy and society, for which the city has set the stage. From a po-
4 litical sociology point of view, the city has surfaced as a relevant and
5 privileged unit for empirical investigation. For political and social
6 scientists, issues such as the political incorporation of migrants, the
7 enfranchisement of foreigners and immigrant ethnic mobilization
8 provide the basis for a new appraisal of relations between civil and
9 political society. They raise the question of how best immigrant mi-
10 nority groups can organize and participate in local decision-making
11 to defend and preserve their collective interests.
12 The third idea at the heart of this research is its focus on the col-
13 lective response of one immigrant minority group in three settings
14 and to study the focus and patterns of its collective sociopolitical in-
15 sertion.
16
17
18 Ethnic mobilization and sociopolitical participation
19
20 Immigrant Sociopolitical Participation
21 Earlier research on postwar immigration showed that immigrants re-
22 cruited as a labour force of guestworkers quickly confronted the need
23 to organize their collective interests. Initially, they did it within the
24 framework of industrial relations, but their claims quickly moved be-
25 yond that arena. Mark Miller (1981) and Catherine Withol de Wenden
26 (1977, 1978, 1988) were among the first to reflect on these realities
27 immigrant and to challenge the then dominant Marxist assumptions about the
28 labour force political quiescence of the immigrant labour force (Miller, 1981: 22-
29 9). Both authors suggested that migrants were becoming more than
30 a temporary labour force and were developing new kinds of political
31 mobilizations that did not rely on electoral politics. In the framework
32 of this theoretical and empirical reconsideration, immigrants came
33 to be regarded as political subjects, rather than the political objects
34 they had been seen as until then to sustain class divisions and the
35 conservative needs of the capitalist economy.
36 Earlier work on the political sociology of immigration reintro-
37 duced some basic reflections on the boundaries of the nation-state’s
38 political community and on the sustained challenge migration posed
39 to classical conceptions of citizenship and nationality. In most cases,
40 first-generation migrant workers in continental Europe acquired dif-
41 ferentiated and inferior citizenship statuses, to which Hammar later
42 attached the label denizenship (Hammar, 1990). As non-nationals,
43 immigrant workers in countries like France, the Netherlands and

Belgium were granted access to various social and civil rights, but 1
their political rights were restricted.1 They were, in effect, excluded 2
from electoral participation. An important exception to this rule oc- 3
curred when the Dutch, Irish and Scandinavian governments gave 4
foreigners the franchise at the local level. Unlike their counterparts 5
in France, Germany and Belgium, immigrants in these countries 6
were allowed active electoral participation (the right to vote and be 7
elected) at the local level. In terms of political analysis, this was 8
and still is a significant factor because immigrant communities in 9
Belgium, Germany and France have never represented a significant 10
electoral force.2 11
For a number of reasons, the sociopolitical participation of im- 12
migrant ethnic minorities is an important and worthwhile subject 13
of study for the political sociology of liberal democratic societies. In 14
recent years, it has become a bit more multicultural, multiethnic and 15
multi-religious. Withol de Wenden and Hargreaves (1993: 2-3) iden- significance of 16
tify three reasons for the continuing significance of immigrant so- immigrant 17
ciopolitical activism. First, are the memories of alternative means of socio-political 18
political participation open to disenfranchized immigrant communi- activism 19
ties, such as strikes, hunger strikes and marches? Second, consulta- 20
tive institutions have been established in many countries where, as 21
foreigners, immigrants are not entitled to full political rights. Third, 22
immigrants have, to varying degrees, been granted access to national- 23
ity in their receiving countries. This option, which opens the door to 24
full citizenship, has had particular relevance for the second and third 25
generation, particularly in countries that have traditionally based 26
their naturalization procedures on jus soli.3 A fourth reason for study- 27
ing the sociopolitical involvement of immigrants is because the bind- 28
ing relationship between nationality and citizenship, at least in its 29
political dimension, has over the last 20 years been seriously thrown 30
into question. Citizenship of the European Union and foreigners’ 31
experiences of enfranchisement at the local level are instances of a 32
decoupling of citizenship and nationality, the main consequence of 33
which is to open the door towards granting some political rights to 34
non-nationals. 35
These elements indicate that, over the past 20 years, the situa- 36
tion in northwestern immigrant receiving European countries, such 37
as the Netherlands, Belgium and France, has changed qualitatively. 38
Immigrants and their supporters have gained some important victo- 39
ries. Whereas migrant workers and their families were left with prac- 40
tically no access to mainstream political institutions in the 1970s, 41
most immigrant receiving European countries have now established 42
a number of procedures and institutions to increase their political 43

358 hassan bousetta
1 participation and representation. Though some convergence is ob-

2 servable, the nature and scope of these channels of participation dif-
3 fer from one country to another (Layton Henry, 1990). Nevertheless,
4 there are now a number of formal channels through which immi-
5 grants can articulate their political demands.
6 These institutional developments have influenced methods of
7 institutions and theorizing immigrants’ political inclusion. Breaking away from cul-
8 policies turalist interpretations of immigrants’ sociopolitical behaviour, re-
9 cent literature has paid increasing attention to the role and influence
10 of institutions and policies. It has been argued, for instance, that both
11 the nature and impact of immigrant political participation predomi-
12 nantly depend on the political context they confront (Ireland, 1994).
13 This approach leads to a crucial point for European comparative re-
14 search, for it holds that most of the variations that can be identified
15 across national boundaries are more dependent on the specificities of
16 the domestic political context than on the deliberate strategic choices
17 of minority groups.
18 Without going deeper into the complexities of the theoretical de-
19 bate, a cautious interpretation of the actual role of institutions and
20 policies is called for to avoid turning the proper role of immigrants
21 into that of a passive agent determined by structural political and in-
22 stitutional factors. Any attempt to influence politics and to gain more
23 access to the political process necessarily implies the mobilization
24 of collective actors. The organizational basis of immigrant political
25 action should therefore be taken as a focal point in studying immi-
26 grant participatory patterns. Before discussing this in relation to the
27 Moroccan experiences in three cities, a clarification of two related
28 concepts of particular relevance to the problématique is proposed in
29 the next section, namely the concepts of ethnic mobilization and of
30 ethnic minority associationism.
31
32 Ethnic mobilization and ethnic minority associationism
33 As suggested earlier, several channels to political participation are
34 open to ethnic minorities. In the three countries central to this analy-
35 sis, social scientists have pointed out the importance of the liberaliza-
36 tion of foreigners’ rights of association to the political participatory
37 opportunities available to immigrant communities (Layton Henry,
38 1990). The setting up of independent associations has been a major
39 development for immigrant communities denied all the attributes of
40 citizenship of the majority. It has opened a door for them to organize
41 their own sociopolitical interests in institutions independent both of
42 the country of origin and of the host country’s various solidarity or-
43 ganizations. Ethnic minority associational life has in many instances

provided the organizational basis for new types of identity-driven 1

mobilizations, such as ethnic mobilization. However, and this is the 2
point to emphasize here, ethnic minority associations have a twofold 3
orientation, which allows them to distinguish between their role as 4
conveyors of ethnic solidarity and their role as ethnic political actors. 5
The point is that the study of immigrant minority associational life 6
does not provide the basis for a single conceptual approach in terms 7
of ethnic mobilization. Ethnic minority associations can provide an 8
organizational vessel to some forms of ethnic solidarity without nec- 9
essarily being the vector of ethnic political mobilizations. 10
By introducing this distinction, I wish to reinstate a point ex- 11
pressed earlier by ethnic competition scholars who established a ethnic mobilization 12
theoretical and empirical distinction between the concepts of ethnic 13
solidarity and ethnic mobilization (Olzak, 1983; Olzak and Nagel, 14
1986). There has been a tendency in the English-speaking litera- 15
ture to subsume all forms of immigrant collective action under the 16
category of ethnic mobilization. Positing an immigrant ethnic mo- 17
bilization needs a priori definition of what is ethnically defined in 18
their mobilization, as well as a conceptual framework that allows one 19
to account for forms of immigrant mobilization that are not orga- 20
nized solely along ethnic lines. As John Rex’s Barthian perspective 21
on ethnic mobilization suggests, this should depend above all on a 22
situational definition of the projects in which ethnic groups engage 23
(Barth, 1969; Rex, 1991, 1994). In other words, the meaning of eth- 24
nic political mobilization does not rest on the cultural values and 25
norms of the group’s membership, but on a process, which includes 26
boundary drawing, in which ethnicity serves as an instrumental re- 27
source for collective action. This conception of ethnic mobilization is 28
of interest because it provides one with a pivotal concept on which to 29
build a broader conception of multicultural society. For Rex, ethnic 30
mobilization in a multicultural society is a valuable strategy of col- 31
lective action, which immigrant ethnic minorities should pursue to 32
defend and preserve their collective interests (Rex, 1985, 1991, 1994). 33
He does not see ethnic mobilization as being at odds with the defini- 34
tion of the idea of equal citizenship of all individuals of the liberal 35
democratic tradition. As he put it (Rex, 1994: 15), ‘In fact, one of 36
the goals of ethnic mobilization is precisely the achievement of this 37
kind of equal citizenship and it may well be that ethnically mobilized 38
groups will act together to achieve such an end both with other ethnic 39
groups in a similar position and with indigenous peers.’ 40
With this clarification, we can now turn to the role of immigrant 41
ethnic associations in relation to their communities and to the politi- 42
cal process. Ethnic associations have received unequal interest from 43

360 hassan bousetta
1 academics. In France, they have formed the subject of numerous

2 works; in other countries, such as the Netherlands, they have been
3 almost ignored in social science research.4 A brief international over-
4 view of studies of immigrant ethnic associational life shows a great
5 variety of interests and approaches, which cannot be encompassed
6 within a single problématique. Though social science researchers tend
7 to view their roles and functions quite positively, ethnic associations
8 have been analysed in different countries at different times for dif-
9 ferent analytical purposes. In an international comparative study, S.
10 Jenkins and her co-authors looked at ethnic associations from the
11 point of view of the satisfaction they provide to fellow co-ethnics.
12 They suggested that their role be reconsidered for inclusion as policy
13 actors in the delivery of social services (Jenkins et al., 1988). The role
14 and functions of ethnic associations have also received consideration
15 in Rex’s classic community study of Sparkbrook (Rex, 1973). Another
16 study by Rex, Joly and Wilpert (1987) looked at the functions of eth-
17 nic associations from an international comparative perspective and
18 viewed them as a non-transitional phenomenon offering a range of
19 identity options to immigrant populations. Schoeneberg (1983) pro-
20 vided an interesting and comprehensive assessment of the role and
21 functions of ethnic associations in Germany. He sought to establish
22 the relationship between organizational participation in ethnic asso-
23 ciations, direct contact with majority group members and cultural as-
24 similation. From his research, he concluded that these relationships
25 are complex and depend largely on the nature of the organizations,
26 though they can be assumed to have a general positive effect.
27
28
29 Three local case studies5
30
31 Liège
32 In 1996, the Moroccan community of Liège numbered 5270 indi-
33 viduals, most of who had come as immigrant workers or student
34 migrants. This community included numerous organizations dis-
35 playing diverse profiles. Moroccan ethnic associations in Liège are
36 structured along a number of well-established cleavages, including
37 gender, age, ideological orientation towards the country of origin,
38 ideological orientation towards the country of residence, religion
39 or secularism and regional identities (Berbers versus non-Berbers).
40 Though the Moroccan community’s formal organizational structure
41 weak mobilization in Liège does not reveal much variation in comparison with the two
42 other cities, one can contend that this community is weakly mo-
43 bilized in the formal political field. It has also failed to establish a

coherent political movement in the face of deteriorating socioeco- 1

nomic conditions. A good illustration of this is the absence of any 2
significant involvement in electoral politics by Moroccans of Belgian 3
nationality.6 The relationship between the Moroccan community in 4
Liège and local political parties is a chapter that still has to be written. 5
Another indication is that Moroccan ethnic associations are clearly 6
under-represented in local inter-organizational networks mobilized 7
around immigration/integration issues. A range of multiethnic and 8
Belgian solidarity organizations, such as human rights associations 9
and antiracist groups, dominate the mobilized actors. The ideo- 10
logical fragmentation of these organizations may partly explain the 11
Moroccans’ under-representation. Many solidarity organizations are 12
either affiliated to a specific segment of Belgium’s rather pillar-like 13
society, such as the Christian or socialist movement, or are close to 14
alternative political parties such as Ecolo, the green party in French- 15
speaking Belgium. 16
To explain this situation, it is necessary to go beyond normative 17
judgements about the capacity of leaders to articulate the demands 18
of their community. More interestingly, the point is to analyse the 19
interaction between the internal and institutional factors that shaped 20
the sociopolitical trajectory of the Moroccan community in Liège. 21
The most important obstacles that Moroccans, like other smaller eth- 22
nic and religious minorities, have repeatedly confronted in Liège is 23
a shared consensus among the political elite of the majority about 24
the normative meaning of integration. So far, the dominant assimi- 25
lationist ideological framework has impeded the emergence of alter- 26
native ways of representing ethnic minorities either in the formal 27
political process or in the implementation of public policies. To some 28
extent, one could contend that this has resulted in the reproduction 29
of immigrant’s powerlessness through a systematic non-politiciza- 30
tion and non-specific decision-making. In comparison with the three 31
other case studies, the absence of a specifically local policy theorizing 32
on integration issues is evident. 33
In 1973, Liège had, however, experienced a pioneering initiative 34
with the establishment of a consultative institution. This consulta- 35
tive council, the CCILg (Conseil consultatif des Immigrés de la Ville CCILg 36
de Liège), was for a long time the only formal institution where im- 37
migrant minority communities could articulate their political de- 38
mands. Like many peer consultative bodies, the CCILg has steadily 39
confronted a number of difficulties in its communication with the 40
local council and has never managed to increase its power within 41
local politics (see Martiniello, 1992). The CCILg stopped its work in 42
1991 and the new municipal authorities, elected in 1994, have ten- 43

362 hassan bousetta
1 tatively begun to develop a policy of interculturalism. This new policy

2 framework has for the first time sought to stimulate a few associative
3 projects promoting ‘intercultural encounters’. However, the relation-
4 ship between local authorities on the one hand, and multiculturalist
5 and ethnic activists on the other, have suffered from the enduring
6 lack of communication between the local council and voluntary as-
7 sociations. An illustration of this was given recently by a confronta-
8 tional mobilization against the local authorities and Department of
9 Intercultural Relations on the issue of the voluntary sector’s repre-
10 sentation in the newly established regional centres of integration, a
11 new institution promoted by the Walloon government.
12 The lack of consistent and coherent avenues of political participa-
13 tion did not, however, lead to political quiescence. The public politi-
14 cal sphere’s lack of investment is counterbalanced by vigorous activ-
15 institution ity within the community’s institutions and associations. In fact, the
16 building context in which Moroccan sociopolitical action takes place in Liège
17 emerges from a historical outlook towards its institution building. In
18 the earlier phases of Moroccan settlement in Liège, collective struc-
19 turation took on two main orientations, in opposition to one another.
20 The two dominant organizational forms were initially developed by
21 Islamic groups under Moroccan government control7 and by secular
22 leftist groups. The former’s objective was to establish Islamic asso-
23 ciations committed to setting up and managing mosques. Political
24 issues in the homeland, though, largely informed the political ac-
25 tivities of the secularists of the left. However, these types of orga-
26 nizations, which included the Liège section of the National Union
27 of Moroccan Students (UNEM) and Solidarité Arabe, have gradually
28 focused their activities on local issues. Members of the Moroccan
29 secularist left wing have for instance been involved in consultative
30 politics at the city level in Liège within the CCILg and at the level of
31 the French-speaking community within the CCPOE (Conseil consul-
32 tatif pour les Populations d’origine étrangère).
33 A number of Moroccan Islamic organizations have in the past
34 struggled for autonomy against Moroccan consular representatives
35 and have fed a number of conflicts that have resulted in the creation
36 of new mosques.8 These conflicts involved mixed issues of identity,
37 ideology and theology. It is apparent from these internal debates,
38 however, that the sociopolitical interests and attitudes of Moroccan
39 Muslims are fragmented and not amenable to a single strategy of
40 ethnic mobilization. Empirical studies of Islamic institution build-
41 ing reveal considerable dissent among the membership of Islamic
42 associations over the issue of publicizing Islam. Whereas some
43 streams have pleaded for a more visible positioning of Islamic identi-

ties in the public sphere, others have opposed and mobilized to keep 1
their religious space immune from public concern. The El Itissam 2
mosque has undoubtedly gone furthest in the first strategy, while 3
the El Mouahidin mosque has traditionally opted for the second one. 4
The El Iman mosque, a stronghold of Moroccan consular agents and 5
of the friendship societies of Moroccan merchants and workers (ami- 6
cales), has on the other hand relied on forms of ethnic lobbying based 7
on individual networks among the local political elite. These amicales 8
have also had two representatives elected after the CCILg’s elections 9
of 1984. 10
Islamic associations in Liège enter the public political arena not 11
only over local matters, such as a request for Islamic cemeteries9 12
and the organization of educational activities, but over national is- national issues 13
sues such as the representation of Islam according to the Belgian 14
law of 1974 (see Panafit, 1997). The Islamic association El Itissam is 15
at the forefront of this claim and has developed a strategy of vertical 16
integration (at both national and regional levels) with Brussels-based 17
Islamic groups. Unlike the secular left wing, Islamic groups have not 18
participated in regular political relays within the local political arena 19
and have only managed to find occasional access to the policy process 20
on issues of direct concern to them. 21
22
Lille 23
The 6260 Moroccans in Lille represent the most important group 24
of non-nationals. Apart from a small minority who acquired French 25
citizenship, first-generation Moroccan immigrants have had no ac- 26
cess whatsoever to the electoral process. Their status as non-nation- 27
als has denied them access to the most formal political arena. The 28
first significant developments in terms of electoral political partici- 29
pation appeared with the political emergence of the second gener- 30
ation. In Lille, the most recent municipal elections confirmed the 31
slow and uneasy emergence of second-generation individuals in the 32
political arena. In 1989, three candidates from North African youth 33
organizations were put forward by the socialist party. One of them, a 34
co-founder of Les Craignos, was elected and appointed the mayor’s 35
delegate for ‘citizenship and human rights’. In 1995, several North 36
African candidates ran again for a seat in the local council. Among 37
them, two well-known figures in second-generation North African 38
associational life and a social worker of Moroccan origin have been 39
successful.10 40
Before the second generation started to organize politically and to 41
set up its associations in Lille, first-generation Moroccans had been 42
less quiescent than Beur historiography has sometimes tended to 43

364 hassan bousetta
1 suggest. In Lille, as in other European cities, the Moroccan govern-

2 ment became involved early on in setting up collective infrastruc-
3 tures for Moroccan migrants. Setting up a federation of amicales in
4 the north was here again the Moroccan regime’s pivotal instrument
5 for strategy of control. The role of Moroccan diplomats in this pro-
6 cess of community organization and control was never clearer than
7 in the 1986 conflict when Moroccan miners of the northern French
8 coalfield opposed the Charbonnages de France. After a long strike
9 led by a group of Moroccan miners from the French trade union
10 CGT (Conféderation générale du Travail), 3600 Moroccan miners
11 were unfairly dismissed after an agreement was reached between
12 the Moroccan embassy and their employer, the Charbonnages de
13 France (for more details, see Sanguinetti 1991: 75-8). Although many
14 Moroccan miners were forced to return to Morocco, the struggle for
15 their social and economic benefits is still going on today. In 1987, the
16 former Moroccan leaders of the CGT who remained in France found-
17 ed an independent association (Association des Mineurs Marocains
18 du Nord) and joined the national federation of the Association des
19 Travailleurs Marocains en France (ATMF).
20 community Parallel with the first-generation community organizations the
21 organizations second generation, most often headed by young Algerians, has
22 emerged in the sociopolitical field at both local and national levels.
23 As Bouamama recalls, the mobilization of the second generation and
24 the setting up of associations started to become a central issue in Lille
25 with the first nationwide ‘Marches des Beurs’ of 1983 (Bouamama,
26 1989). Texture and Les Craignos are two important associations that
27 were founded in this period. The setting up of a large number of
28 smaller associations, most often youth associations involved at a
29 neighbourhood level, has recently followed their pioneering work in
30 the city of Lille. While Les Craignos has set up a federation of neigh-
31 bourhoods youth associations, the Fédération des Associations des
32 Jeunes de Quartier (FAJQ), Texture has supported the foundation of
33 a multiethnic immigrant women’s association called Femmes d’ici et
34 d’ailleurs.
35 In Lille, as in Liège and Utrecht, in recent years there has been a
36 strong development of Islamic associations. The Lille Sud mosque is
37 at the forefront of the mobilization of North African Muslims in the
38 north. Its activities are strikingly similar to those of the El Itissam
39 association in Liège. Vertical integration with regional Islamic as-
40 sociations and Paris-based federations, mobilization on educational
41 matters, and the provision of services and activities to the second
42 generation are some of the issues with which the Lille Sud mosque
43 is engaged.

There are two interesting points about the nature of North African 1
political incorporation. 2
First, there seems to be a strong generational divide between first- 3
and second-generation collective action. Whereas the first generation 4
relied mostly on ethnic mobilization within trade unions, indepen- 5
dent associations and mosques, the second generation tends more 6
towards universalistic political inclusion. This has given rise to some universalistic 7
interesting debates among members of North African associations political inclusion 8
in Lille. Texture has promoted the idea of intergenerational solidar- 9
ity within the migrant population and has sought to distance itself 10
from narrow forms of ethnic mobilization. In 1989, for instance, it 11
sponsored an electoral list purportedly composed of an aggregate of 12
candidates from migrant communities and socially excluded popula- 13
tions. The mobilizations of France Plus and Espace Intégration are 14
further examples of ethnic mobilizations not necessarily fitting the 15
nature and profile of the organizations in question. In Lille and in 16
the north of France more generally, these two organizations have 17
developed a discursive strategy of republican integration (namely 18
assimilation) into French society, while at the same time activating 19
ethnic boundaries as a basis for political bargaining. This apparent 20
contradiction has been widely discussed in the French literature; it is 21
what Vincent Geisser (1997) tentatively identified as the emergence 22
of a ‘republican ethnicity’. Unlike Texture, which has deliberately 23
avoided grounding sociopolitical activism in ethnic identifications, 24
the latter are interesting examples of ethnic mobilization being em- 25
bedded in discursive strategic use of an assimilationist vocabulary. 26
Second, the so-called town policy (la politique de la ville), which 27
has been implemented as a partnership between national govern- 28
ment, regions and municipalities, has provided a number of profes- 29
sional opportunities to individuals formerly involved in immigrant 30
associational life. This policy has created and sustained a demand for 31
leadership within impoverished immigrant neighbourhoods. One 32
can speak here of the institutional production of an immigrant as- 33
sociational life of proximity. The seamy side of the story, however, 34
is that it has increased control over the practices and ideologies of 35
second-generation activists, while weakening the autonomous politi- 36
cal action of civil society (Bouamama, 1989). 37
38
Utrecht 39
The Moroccan population in Utrecht consists of 13,595 individuals. 40
Unlike their counterparts in Lille and Liège, Moroccans in Utrecht 41
have been enfranchized for local elections since 1986. The Moroccan 42
community has also been identified as a specific target group for the 43

366 hassan bousetta
1 national minority policy implemented since 1983. At the Utrecht city

2 level, integration has been under constant consideration for at least
3 two decades. In 1973, a consultative council was created in Utrecht
4 to advise local authorities on community relations issues (Feirabend
5 amicales and Rath, 1996). The amicales responded very early on to the open-
6 ing up of this avenue of participation. In Utrecht, as in several other
7 Dutch cities, the amicales, with the support of Moroccan diplomats
8 and through their networks of personal contacts within the Moroccan
9 communities, have been acknowledged as legitimate representatives
10 of the political interests of this population,11 though for a very short
11 period. After 1976, the amicales were vigorously challenged by the
12 creation of a nationwide independent organization of Moroccan
13 workers, the KMAN (van der Valk, 1996).
14 Most activists involved in establishing left-wing Moroccan associ-
15 ations in Utrecht have had some initial involvement with the KMAN.
16 This was so for the founders of two very influential associations in
17 Utrecht – AMMU and the KMANU, breakaways from the KMAN.
18 Once the amicales had lost their influence in Utrecht (and in the
19 Netherlands in general), AMMU played an important role as policy
20 adviser to the local council and has come to be the most central actor
21 in Utrecht’s Moroccan community. AMMU has also stimulated the
22 creation of separate ethnic associations for Moroccan women and for
23 Moroccan youth (PMJU).
24 The activities of left-wing Moroccan activists in Utrecht raise im-
25 portant questions about the co-optation of elites. The minority policy
26 in Utrecht (and more generally in the Netherlands) has created and
27 sustained an impressive number of social work, multicultural and
28 antiracist institutions and agencies. This has created numerous op-
29 portunities for elites, both as professionals and as leaders of ethnic
30 communities.
31 Minority representation of these institutions by an elite clearly
32 creates a number of non-political opportunities to voice immigrant
33 claims within the mainstream. However, Moroccans have also pur-
34 sued strategies that challenge the integrationist approach of Utrecht’s
35 Moroccan leaders of the secularist left. Among these are forms of
36 ethnic mobilization around regional identities in the cultural field.
37 Rifan Berbers are currently the most active in this area. Their strat-
38 egy of institution building has steadily confronted the opposition of
39 Moroccan left-wing associations. Ethnoreligious mobilization within
40 Islamic associations is another strategy pursued by Moroccans in
41 Utrecht.12 As Feirabend and Rath (1996) point out, Utrecht is more
42 reluctant than other Dutch cities to create a space for Islamic institu-
43 tions within local sociopolitical life. This development is reflected in

the decision to stop funding the educational activities provided by the 1

El Dawa mosque,13 the biggest mosque in Utrecht. 2
Over the last year or so, the city of Utrecht has completely recon- 3
sidered its policy options in relation to immigrant minority commu- 4
nities. Publication of research the local council commissioned from 5
the University of Utrecht was at the source of a new assessment of 6
the problématique. The Burgers Report called for a shift from a mi- the Burgers Report 7
nority policy towards corrective measures focused on socioeconomic 8
differences (Burgers et al., 1996). The ensuing debate between the 9
municipality and representatives of ethnic minorities led to the defi- 10
nition of a new policy hinged on the operationalization of the concept 11
of ‘interculturalization’ – a far cry, however, from the intercultural 12
approach of the city of Liège. 13
One element of this policy, besides its attempt to combat a 14
dualization of urban life along ethnic lines, is a new partnership be- 15
tween ethnic minority self-organization and the municipality. The 16
framework for this relationship had already been defined in a policy 17
report of 1989. In the programme the municipality recently issued, 18
the role of self-organization is identified as a bridge between societal 19
and internal community dynamics. The concept of ‘interculturaliza- 20
tion’ is a central idea in this policy framework seeking to develop 21
a proactive approach to the forming of a social coalition within so- 22
ciety (maatschappelijke coalitievorming). This reflects an attempt to 23
avoid the separate development of ethnic communities, which was 24
allegedly produced by the earlier minority policy. Indeed, the city of 25
Utrecht’s new policy implicitly gives a positive answer to the follow- 26
ing questions: Has the minority policy led to the isolation of immi- 27
grant minority communities from the mainstream? And was the old 28
policy framework disruptive in terms of social cohesion? 29
30
31
Conclusion 32
33
This comparative overview of three case studies has taught us some 34
important lessons about patterns and forms of immigrant political 35
incorporation. We have observed sociopolitical participation in main- 36
stream political institutions, ethnic mobilization and less politically 37
significant internal community dynamics. The minority response 38
the Moroccan communities exemplified revealed the importance of 39
ethnic mobilization within independent ethnic and religious asso- 40
ciations, the deployment of civic, youth, gender and neighbourhood 41
mobilization, as well as the involvement of minority candidates in 42
mainstream party politics. 43

368 hassan bousetta
1 capacity for The Islamic groups and associations have shown us that their
2 mobilization form of ethnic mobilization may not be temporary. In all three cities,
3 Islamic organizations proved their capacity to attract massive audi-
4 ences within Moroccan communities and one could contend that the
5 impact of Islamic ethnic mobilization is, in political terms, still in its
6 infancy. Although some Islamic associations of the older generation
7 are resisting Islam being brought into the public sphere, the oppo-
8 site phenomenon has been growing in significance within Moroccan
9 communities since the mid-1980s.
10 Though one can, of course, identify more secularized attitudes
11 among the second and third generations, the ethnic mobilization of
12 Islamic associations should not be seen as dependent on cultural and
13 religious values and norms. Islam provides an identity option, the
14 significance of which will depend in the long run on the projects
15 pursued by this youth and by the place open to them within their
16 societies. On the other hand, the secularist left-wing movement of
17 Moroccan workers and students that dominated the stage during the
18 1970s and 1980s has in the three cities lost its capacity to engage in
19 mass contentious collective action. We have also seen appearing the
20 mobilization of youth, gender, generational and locational identities,
21 which proves that minority communities are internally segmented
22 along a number of consequential divides. These factors of internal
23 division should be seen as being a problem intrinsically, even though
24 they preclude the possibility of uniting resources and energies. Of
25 course, a common immigrant political agenda cross-cutting internal
26 and external ethnic boundaries is, under such circumstances, close
27 to utopia.
28 In the three case studies, we have seen external institutional forc-
29 es constrain integrationist forms of political incorporation. We have
30 also seen that local authorities have a number of policy options at
31 hand to deal with the sociopolitical demands of immigrant minority
32 communities. The local authorities of the three cities under review
33 adopted policies of sustained communication with ethnic and mul-
34 tiethnic minority associations (Utrecht, Lille), funding to ethnic and
35 multiethnic associations (Lille, Liège, Utrecht), consultative politics
36 (Liège, Utrecht), and enfranchisement for local elections (Utrecht).14
37 The efficiency of these policies partly depends on their cumulation
38 and coordination. However, as the Dutch case study reveals, a con-
39 sistent, coordinated, multicultural approach still manifests serious
40 difficulties.
41 This latter indication points out that both the institutional politi-
42 cal strategy of incorporation and the minority response have not had
43 far-reaching effects on the collective position of minority communi-

ties in the three societies. In other words, while the nature of im- 1
migrant’s inclusion has diversified, the impact of immigrants’ mo- 2
bilization on a wide number of issues of collective importance has 3
remained extremely weak. The collective position of Moroccans in weak position 4
areas such as education, employment or housing in the three coun- 5
tries, remains an issue of serious concern and the same holds true 6
for the legal position of Moroccan women. Although Miller (1981) 7
was partly right in saying immigrants and their offspring are neither 8
voiceless nor powerless, the reality seems to fall short of his optimis- 9
tic view of foreign workers as an ‘emerging political force’. One must 10
conclude that the social, political and economic emancipation of eth- 11
nic minority groups is still heavily dependent on the implementation 12
of liberal political agendas from the majorities. The experience that 13
Moroccans share with other ethnic minorities in northwest Europe 14
leads to another more general conclusion. Although their demo- 15
graphic share is massively increasing within European urban popu- 16
lations, this has not yet been reflected in the most formal political 17
institutions in which, collectively, they remain under-represented. 18
19
20
Notes 21
22
1 One should, however, call for cautious use of the classical Marshallian 23
distinction of citizenship rights in three spheres: civic, social and political 24
(Marshall, 1950). In many circumstances, political activities are not depen-
dent on the possession of formal political rights. The civil and social rights
25
open to immigrants play in many cases as a legal juridical protection to their 26
extra-parliamentary political activities (see also Miller, 1981: 15-20). 27
2 On this particular point, the situation for foreign communities in continen- 28
tal Europe is substantially different from that in Britain, where foreign resi- 29
dents who are citizens of Commonwealth countries are fully enfranchized.
30
3 Withol de Wenden and Hargreaves (1993: 2) rightly note that this option has
always been more than a theoretical possibility for foreign residents even in 31
countries implementing jus sanguinis-types of naturalization regulations. 32
4 There are some notable exceptions to the rule, including among others de 33
Graaf (1986); de Graaf, Penninx, Stoové (1988) and Van der Valk (1996). 34
5 Use is made in this research of a qualitative methodology based on the se- 35
lection of three urban sites of empirical work in three different countries.
The three urban contexts were chosen in the three countries with the larg-
36
est Moroccan emigrant communities. Among the 1.1 million Moroccan em- 37
igrants settled in Europe, almost half are permanent residents in France, 38
Belgium and the Netherlands. I have selected three cities that attracted sig- 39
nificant numbers of immigrant workers in the period of massive immigra- 40
tion from the Mediteranean (1959-74). It should also be mentioned that they
41
are university cities, which is a relevant consideration given that the migra-
tion of Moroccan students towards European universities has played an im- 42
portant role in the sociopolitical organization of these communities. 43

370 hassan bousetta
1 6 In Begium, the most formal aspects of political participation (the right to

2 vote and to stand as a candidate) are dependent on the possession of Belgian
3 nationality.
7 Historically, the first attempts to create collective infrastructures for
4 Moroccan workers came from the government of the country of origin.
5 These resulted in the establishment of a European-wide network of amicales
6 (friendship societies of Moroccan merchants and workers). Their role con-
7 sisted of organizing political control over the Moroccan communities. The
8 very undemocratic activities of the amicales supported by Moroccan embas-
sies and consulates have, in many middle sized European cities, triggered
9
the same sort of fierce conflicts that were being activated in the same period
10 in bigger cities like Amsterdam, Brussels and Paris (van der Valk, 1996).
11 8 The mosque of El Mouahidin early on refused to make any reference to the
12 ‘the Commander of the Faithfuls’, King Hassan II of Moroccó, during the
13 traditional Friday speech (Saïdi and Aghion, 1987).
14 9 Liège is one of the few Belgian cities with an Islamic cemetery within a
Belgian one. The high demand for burial in this cemetery can no longer be
15 handled, thus the request for a new Islamic cemetery in the region of Liege.
16 10 Farid Sellani, a young Algerian, running on the list of former and re-elected
17 Mayor Pierre Mauroy, has been appointed the delegate to support the ‘asso-
18 ciation’s projects’.
19 11 One of Utrecht’s first amicale activists, and later co-founder of the controver-
sial Union of Moroccan Mosques in the Netherlands (UMMON), recently
20
reflected on this period in a chapter of a book in which the leader of the
21 Dutch right-wing party VVD held conversations with minority leaders (see,
22 Bolkestein, 1997: 45-65).
23 12 There are six mosques in Utrecht, which can be classified in three groups: (1)
24 the mosques controlled by the coalition of Moroccan consular agents, the am-
25 icales and the Union of Moroccan Mosques of the Netherlands (UMMON),
(2) the El Dawa mosque of the Worldwide Islamic League and (3) a group of
26 smaller independent and neighbourhood mosques.
27 13 In the Municipal Department for Welfare’s 1997 programme, this decision is
28 justified as follows: ‘The project has been funded for two years (...) Although
29 it answers a need, we are not ready to extend the subsidies. There is no more
30 funding for 1997. It is important that we do not provide structural fund-
ing to educational activities organized by people who are not independent of
31
religious organizations’ (rough translation of Ontwerp Welzijnsprogramma,
32 1997, City of Utrecht, Department of Welfare).
33 14 Although the enfranchisement of foreigners is a prerogative of national au-
34 thorities, local decision-makers can influence political participation through,
35 for instance, policies of information in the languages of minorities.
36
39
40
41
42
43

1
2
3
Integration and nations: the nation-state 4
5
and research on immigrants in Western Europe1 6
7
Adrian Favell 8
9
10
11
12
The proliferation of integration studies in Europe is, according to the so- 13
ciologist and philosopher Adrian Favell, part and parcel of a wider ‘nation- 14
state-society’ paradigm. Those who work within this paradigm see the nation 15
state as the principal organising unit of society. Moreover, they see society 16
as a bounded, functional whole. The state achieves this by creating policies 17
and institutions. Favell has doubts about whether this ‘nation-state-society’ 18
paradigm is still sufficiently appropriate for understanding the evolving rela- 19
tionship between immigrants and their host context. This article is a strong 20
plea for research that goes beyond such crude and fairly static entities such 21
as nation-states. 22
23
24
Despite its somewhat old-fashioned, functionalist air, ‘integration’ integration 25
is still the most popular way of conceptualizing the developing re- 26
lationship between old European nation-states and their growing 27
non-European, ‘ethnic’ immigrant populations. It is also widely used 28
to frame the advocacy of political means for dealing with the conse- 29
quences of immigration in the post-World War II period. Many simi- 30
lar, difficult-to-define concepts can be used to describe the process of 31
social change that occurs when immigrants are ‘integrated’ into their 32
new host society. But none occurs with the frequency or all-encom- 33
passing scope of the idea of integration across such a broad range of 34
West European countries. This fact continues to decisively structure 35
policy research and policy debate on these subjects in Europe. 36
The wide and varied ordinary language usages of the term are 37
linked to a deeper association of the concept with a longstanding 38
intellectual paradigm at the root of modern western society’s con- 39
ception of itself. This paradigm roots applied social policy thinking 40
in the idea of the ‘nation-state’ as the principal organizing unit of 41
society, with all the epistemological assumptions and political con- 42
straints that this term implies. By using the term, writers continue 43

372 adrian favell
1 to conceive of ‘society’ as a bounded, functional whole, structured by

2 a state which is able to create policies and institutions to achieve this
3 goal. This ‘nation-state-society’ paradigm may now no longer be the
4 appropriate one for charting the evolving relationship of new immi-
5 grants and their host contexts in Europe. In this paper, then, I seek
6 to explore the strengths and weaknesses of ‘integration’ as the seem-
7 ingly inevitable framework for discussing issues in policy-directed
8 research on immigration and ethnic relations.2 After discussing why
9 integration is still such a prevalent term in European thinking – de-
10 spite emerging theoretical challenges associated with globalization
11 and transnationalism – I explore some of the distinct national and
12 supra-national contributions to research in this field. Our compara-
13 tive understanding is often distorted by the predominant focus in
14 much research on big and established country cases such as Britain,
15 Germany or France. I also make reference therefore to newer de-
16 bates surfacing in less central European nations such as Italy, the
17 Netherlands and Denmark, as well as the insights afforded by un-
18 usual cases such as Austria and Belgium.
19
20
21 ‘Integration’ in ordinary language usages
22
23 What is typically spoken of when academics or policy makers use the
24 term ‘integration’ to speak of a collective goal regarding the destiny
25 of new immigrants or ethnic minorities? We can, of course, think
26 long-term of a long list of measures designed to deal with the longer term con-
27 consequences sequences of migration and settlement. These can be distinguished
28 from immigration policies per se, such as policies on border control,
29 rights of entry and abode, or of asylum. ‘Integration’ conceptualizes
30 what happens after, conceiving practical steps in a longer process
31 which invariably includes the projection of both deep social change
32 for the country concerned, and of fundamental continuity between
33 the past and some idealized social endpoint. Measures concerned
34 with integration include (the list is by no means exhaustive, but in-
35 dicative): basic legal and social protection; formal naturalization and
36 citizenship (or residency-based) rights; anti-discrimination laws;
37 equal opportunities positive action; the creation of corporatist and
38 associational structures for immigrant or ethnic organizations; the
39 redistribution of targeted socioeconomic funds for minorities in de-
40 prived areas; policy on public housing; policy on law and order; mul-
41 ticultural education policy; policies and laws on tolerating cultural
42 practices; cultural funding for ethnic associations or religious orga-
43 nizations; language and cultural courses in the host society’s culture,

and so on (for similar checklists of policies, see Kymlicka, 1995, pp. 1

37-8; Soysal, 1994, pp. 79-82; Vertovec, 1997, pp. 61-2). 2
What is interesting is when and why such measures are packaged 3
together and interlinked within the broader concept of ‘integration’. 4
The very difficult-to-define process of social change with historical 5
continuity pictured here, is for sure spoken of using a plethora of 6
other terms: assimilation, absorption, acculturation, accommoda- 7
tion, incorporation, inclusion, participation, cohesion-building, en- 8
franchisement, toleration, anti-discrimination, and so on. Yet other 9
terms on this list are either vaguer (absorption, accommodation, tol- 10
eration); too technically precise, and hence absorbed within integra- 11
tion (such as incorporation, which specifies a legal process, or anti- 12
discrimination, which only describes one type of practical measure); 13
or are concepts which can be used descriptively without necessarily 14
invoking the active intervention of some political agency (assimila- 15
tion, or acculturation). In recent years, less loaded terms such as 16
inclusion and participation have had some popularity, but neither 17
can match the technical ‘social engineering’ quality of the term inte- 18
gration; nor do they invoke a broader vision of an ideal endgoal for 19
society as a whole. Visionary academics and pragmatic policy makers 20
all need a descriptive and normative umbrella term, that can give 21
coherence and polish to a patchy list of policy measures aiming at 22
something which, on paper, looks extremely difficult and improb- 23
able: the (counterfactual) construction of a successful, well-function- 24
ing multicultural or multi-racial society. The identification of this 25
conceptual space in progressive-minded practical thinking about the conceptual space 26
consequences of immigration has – however euphemistic – always 27
been a key part of the term’s success. 28
The other key thing about the list of measures seen to be part of 29
‘integration policy’, is that they are all things that a state can ‘do’. 30
Although for the time being it is rare to come across a specifically 31
designated ‘Ministry of Integration’, the policy field has emerged as a 32
differentiated area of government, often crossing the competences of 33
different departments. Integration is thus not only an ideal goal for 34
society; it is also something a government sets out to achieve. This 35
assumption is crucial to the nation-state centred conceptualization 36
of social processes that will be found at the core of practical ordinary 37
language usages of the term. Such a use precludes the idea that a 38
society might achieve an integrated state of affairs without the state’s 39
intervention. 40
Sociologically speaking, we can, of course, conceive of integra- 41
tion taking place without the structure-imposing involvement of the 42
state. Immigrants can be ‘integrated’ into the local labour market as 43

374 adrian favell
1 employees or service providers, or they can be ‘integrated’ into com-

2 plex inter-community relations at, say, city or district level. Looked
3 at from a bottom-up perspective – where the integration of society
4 as a whole is not assumed as the end goal of interaction between
5 ethnically diverse groups – multicultural relations can be seen to take
6 all kinds of organized and semi-organized forms. These may not at
7 all be encompassed by the top-down, organized structures typical
8 of state thinking on the subject, such as policy frameworks, official
9 channels of participation, or legally circumscribed rights, restrictions
10 and entitlements. Multiculturalism as a descriptive state-of-affairs,
11 in this sense, could be the product of something that never had any-
12 thing to do with the ‘multicultural’ policies or institutions of the
13 state. However – as historical theorists of the state would remind us
14 with their vivid terminology – the state has always constituted itself
15 in the way it imposes formal structures and institutionalizes social
16 relations via a systematic ‘embracing’, ‘caging’ and/or ‘penetrating’
17 of society (Torpey, 2000). This logic of incorporation has invariably
18 in recent history taken a dominant form of collective social power (to
19 borrow the terms of Mann, 1993) that seeks to encompass, contain
20 and bind together the state’s domination of society, and all the varied
21 market or community relations inside it. This form is the modern
22 nation state. And, as soon as we begin to think of integration as a col-
23 lective societal goal which can be achieved through the systematic in-
24 tervention of collective political agency, we inevitably begin to invoke
25 the nation-state in the production of a different, caged and bounded
26 version of multicultural social relations.
27 nation-building It is very difficult, then, to make much sense of the term integra-
28 tion in practical, applied terms, without bringing back in the nation-
29 state, at least in the European political context. This is not only be-
30 cause the term gets monopolized by nationally rooted policy makers
31 who, I will suggest, typically link their ideas about integration and
32 their measures for achieving it – even when they are ‘multicultural’
33 in inspiration – to historical concerns with nation-building. As I will
34 also go on to explain, it is equally because of a range of epistemo-
35 logical constraints imposed by the practical operationalization of in-
36 tegration as a framework for applied research, whether targeted at
37 questions of policy or at generating knowledge through survey-based
38 studies of immigrants and ethnic minorities.
39 Looking across Western Europe in the broadest possible way, it is
40 clear that ‘integration’ has emerged as the most widely used general
41 concept for describing the target of post-immigration policies. This
42 is not to say that every political figure or intellectual in every coun-
43 try likes or uses the term. The synthetic, cross-national pronounce-

ments of international and intergovernmental organizations might 1

be taken as one good indicator of its pervasive acceptance by the end 2
of the 1990s. It is noticeable how, for example, the conclusions of 3
the presidency of the European Council of Ministers at Tampere in 4
October 1999, gestured specifically towards integration as the key 5
term for encompassing the post-immigration processes EU insti- 6
tutions would like to get involved with in this area of rising politi- 7
cal significance. Although rarely defined, it is also noticeably fore- 8
grounded in the formulations of some of the broadest cross-national 9
programmes instigated by organizations as varied as the Council of 10
Europe, the ILO or the OSCE. The formulations of NGOs in Brussels 11
likewise constantly use the term, as do influential transatlantic policy 12
for a such as the Carnegie Endowment for International Peace or 13
Metropolis.3 14
This success echoes the past and recent history of policy debate in 15
individual nation-states. The case of France here is typical. The emer- 16
gence of ‘intégration’ as the central term of the ‘new republican syn- 17
thesis’ of the 1980s, followed a period in which older assimilationist 18
ideas vied with the post-60s inheritance of ideas about cultural dif- 19
ference and the anti-racist struggle (Costa-Lascoux, 1989; Weil, 1991; 20
Haut Conseil à l’Intégration, 1993). Integration became the sensible 21
position for the centre trying to distinguish itself from xenophobic 22
nationalism on the one hand, and radical anti-system discourses on 23
the other. A similar centrist convergence occurred earlier in Britain 24
in the late 1960s, notably in a well remembered quotation from then 25
Home Office minister Roy Jenkins, one of the principal architects of 26
race relations legislation (Rose et al., 1969; Rex, 1991). Although the 27
anti-racist left has always rejected it, the concept has retained a high 28
degree of practical significance for the liberal, cross-party centre. practical 29
Indeed, with the emergence of new migration questions surround- significance 30
ing the reception of asylum seekers, integration has re-emerged as 31
the most comprehensive term for conceiving resettlement policies, 32
and has been central to recent Home Office consultations on im- 33
migration policy (Castles et al., 2002). France and Britain are the 34
paradigmatic early ‘integration nations’ in Europe: turning post-war, 35
post-colonial policies into a mildly nationalist reaffirmation of the 36
tolerant, cosmopolitan, inclusive nature of their conceptions of na- 37
tionhood (on this, see Favell, 1998). 38
Across other European countries, we can find numerous exam- 39
ples of countries converging similarly on integration as the widest 40
frame for discussing postimmigration policies (see Mahnig, 1998). 41
It is used frequently in research in Germany or Belgium, albeit with 42
ambiguity about what the immigrant is integrating into, given the 43

376 adrian favell
1 federal, city-centred and multi-levelled nature of the process here

2 (Esser, 1999; Blommaert and Verschueren, 1998). It has returned
3 to the fore in the Netherlands and Sweden, after periods of flirta-
4 tion with more cultural differentialist thinking, as they seek to recon-
5 nect the provision of welfare benefits and multicultural policy with
6 conditions about the learning of the national language and culture
7 (Fermin, 1999; Soininen, 1999). It has also been the most obvious
8 frame for ‘new’ (or self-discovering) countries of immigration – such
9 as Italy, Spain, Denmark or Austria – finally formulating a cen-
10 trist, more progressive response to their current immigration ‘cri-
11 sis’.4 Perhaps even more importantly, immigrant and ethnic groups
12 themselves speak of desiring integration, or phrase their criticisms
13 of racism and exclusion as barriers to full or fair integration (see, for
14 example, the frequent use of word in Alibhai-Brown, 2000, a well-
15 known ethnic minority spokesperson in Britain).
16 Some of these ordinary language usages shadow the well-estab-
17 assimilation lished American preference for ‘assimilation’ as the core sociologi-
18 cal concept (Alba and Nee, 1997). In terms of recent immigrants,
19 ‘integration’ is here often used interchangeably with assimilation in
20 the US, when it is gesturing to the functional involvement of new mi-
21 grant ‘ethnic’ groups in the society’s housing, educational, welfare or
22 employment systems (Edmonston and Passel, 1994). Here, indeed,
23 the term has been moved away from its discredited links with deseg-
24 regation issues over black/white public relations in the 1960s, to a
25 more European-looking concern with the cultural and social absorp-
26 tion of diverse new populations that have grown dramatically in the
27 US since the opening up of immigration laws in 1965.
28 Europeans, however, usually shy away from the term assimila-
29 tion, which in a European context would smack of biological over-
30 tones and the nasty cultural intolerance of the past. But the European
31 preference for ‘integration’ ahead of ‘assimilation’ is not really the
32 choice of a less loaded or more politically sensitive term over one
33 which implies greater conformist and exclusionary pressures, quite
34 the contrary. It signals, rather, a deeper concern with the fact that the
35 changes brought on by post-war immigration in Europe have raised
36 anew questions over historical continuity – about the substance of
37 nation-building – which echo once again the longer histories of na-
38 tion-building: the more-or-less coercive absorption of minority popu-
39 lations and regions through centralizing processes of modernization
40 (the classic formulation of this is Gellner, 1983). Integration, then,
41 is about imagining the national institutional forms and structures
42 that can unify a diverse population; hence imagining what the state
43 can actively do to ‘nationalize’ newcomers and re-constitute the na-

tion-state under conditions of growing cultural diversity. The nation 1

building institutions of European nations are – unlike the US and 2
other continents of immigration – not historically built on immigra- 3
tion and geographical distance from Europe, but on bounded notions 4
of specific territory and the constant self-distinction of ‘indigenous’, 5
culturally ‘unique’ populations constrained to live alongside very 6
close, and troublesomely similar neighbours. The essential problem- 7
atic worrying European policy makers is, then, the difficult and of- 8
ten only partial accommodation of culturally distinct outsiders and 9
foreigners into longstanding social and cultural institutions which 10
were essentially defined historically within Europe, and for highly lo- 11
cal reasons, in quite exclusive and belligerent terms. The fear which 12
thus defines the problematic of immigrant integration is that full 13
assimilation on these conditions is probably never likely to occur. 14
The everyday popularity of integration as a term may appear pe- 15
culiar at a time when so-called globalization and, in particular, new 16
forms of migration and mobility are said to have generated all kinds 17
of nation-state-transcending ‘transnational’ actors and forms of orga- 18
nization (see Faist, 2000; Papastergiadis, 1999). Our unit of society 19
is now routinely said to be something we must look for beyond the 20
nation-state (Cohen and Kennedy, 2000). In the more speculative 21
fancies of social theorists we are invited to think of the trajectory of 22
(post-) modernity as going beyond society itself (for example, Giddens, 23
1990; Urry, 2000). Under these conditions, migrant groups might 24
be thought of as not following the same westernizing, modernizing 25
integration path into full citizenship, membership and belonging of 26
their new host societies. Pan-national and regional cooperation, as 27
well as the re-emergence of the city as the locus for integration, is 28
also said to have reduced the significance of the nation-state as an 29
exclusive, bounded ‘population container’ in Europe (Torpey, 2000). 30
Yet the endurance of ‘integration’ as the goal of most practical 31
policy thought on this question in Europe – including amongst the 32
leading independent academic authorities – gives us a clue to the 33
vested interests and applied imperatives of the older, nation-state 34
building paradigm. As soon as their minds turn to applied policy for- 35
mulations, these people recognize no ‘beyond-the-nation-state’ to im- 36
migration policy. Europeans continue to speak of the integration of 37
immigrants into bounded, nationally-distinct societal units – focus- nationally distinct 38
ing attention on typical nation-building questions such as naturaliza- societal units 39
tion, access to citizenship, access to the welfare state, participation in 40
political and social institutions, and so on – precisely because any- 41
thing else threatens the basic political ordering of European cultural 42
and social diversity into state-centred, state-organized social forms. 43

378 adrian favell
1 To put it another way, the incentive structures of policy thinking and

2 comparative research on the integration of immigrants in Europe,
3 are still very much set by the imperatives of the singular nation-state-
4 society, which recognizes this and only this as the fundamental prob-
5 lematic at stake here.
6
7
8 Integration as a paradigm for policy research
9
10 Unlike in America, academic research on immigrants and integra-
11 tion in Europe is still dominantly structured by its explicit or implicit
12 links to the knowledge demands of specific policy agendas and po-
13 litical discussions in different national contexts (on these, see Favell,
14 2001). In Europe, the overlap and interpenetration of research and
15 policy making is pervasive at national and, increasingly, internation-
16 al level. Academics are co-opted into politicized roles either through
17 the direct shaping of the research agenda by public and institutional
18 funding opportunities to do ‘applied’ work; by the invitation to take
19 on the role of public intellectual in media or government work; or
20 by their activist involvement as campaigners, in which their work
21 is used to articulate political positions. This involvement clearly is
22 linked to society’s functional need for someone to express political
23 agency, with academics contributing through their research to the
24 construction of both social problems (as they are perceived) and their
25 solution. Insofar as their work also often serves to ‘think’ for the
26 state, it also helps underwrite dominant nation-building ideologies.
27 Such a role has its costs. The involvement of researchers in activism
28 or the policy process can also diminish the intellectual autonomy and
29 viability of independent academic research outside of more instru-
30 mentalized uses.
31 European nations are obviously at different stages of development
32 in their internal debates, but in most cases academic thinking is now
33 moving beyond purely denunciatory work on the negative conse-
34 quences of immigration (such as studies of racism) into the concep-
35 practical tualization of practical integration solutions and trajectories of multi-
36 integration cultural social change. For example, in Britain, the popular sub-field
37 solutions of more critical anti-racist, Marxist and post-Marxist writers (such as
38 the cultural studies writers inspired by Stuart Hall) – whose work
39 tended to focus on condemning the racism of state institutions and
40 celebrating the ‘resistance’ of immigrant cultures – have themselves
41 found there is a limit to what can be done with such arguments.
42 More recently, they have begun to more consciously contribute to
43 debates about multicultural citizenship, in relation to mainstream

policy formulations (i.e., Gilroy, 2000; Alibhai-Brown, 2000).5 The 1

desire to make a respectable intervention into the public debate, or 2
to get hired for research by the government or political think-tanks, 3
can thus be a disciplining experience. Such contributions can, as the 4
evolution of anti-racist and multicultural thinking in Britain shows, 5
play a major role in legitimizing in the mainstream a national sense 6
of ease with difference and diversity. In many other countries, a sim- 7
ilar evolution can be observed, with discussion about ‘integration’ 8
playing the central mainstream role as a focus for constructive, prag- 9
matic, policy-related interventions. 10
National self-sufficiency in policy debates has, however, been the 11
rule. The terms and categories that dominate discussion in different 12
places – for example, ‘multiculturalism’ and ‘race relations’ in Britain, 13
or republicanism and citoyenneté in France – are the product of of- 14
ten exclusively internal national political dynamics. Notably, they are 15
discourses which reflect and reproduce longer standing narratives of 16
nationhood and national destiny popular in these countries. When 17
references to other countries appear, comparison usually enters as a 18
further self-justificatory strategy for the national ideology. In France, 19
for example, a key move among many public intellectuals involved in 20
producing the ‘new republican synthesis’ and idea of intégration of 21
the 1980s was the contrasting of the ‘universalist’ French tradition 22
with the ‘differentialism’ of its European and North American ri- 23
vals (most dramatically in Schnapper, 1991; Todd, 1994): Over time, 24
however, the prejudices of comparison have softened, especially as 25
policy actors and academics have themselves been increasingly ex- 26
posed to debates and consultation with other national counterparts. 27
Under these conditions, their national reflection may begin to incor- 28
porate more explicit elements of structured comparative knowledge, 29
recognizing the specificities of the other national starting points and 30
the opportunities of cross-national policy learning. The emergence of 31
pan-European structures (both EU and Council of Europe) has added 32
to this imperative, tendering research which, in order to get funded, 33
must be explicitly cross-national in scope and personnel, and policy 34
oriented in its objectives. 35
The first result of academic cross-national policy comparison was 36
the identification of ideal-type national ‘models’ of citizenship and ideal-type 37
integration (Hammar, 1985; Castles, 1995). This Weberian com- national models 38
parative impulse was strongly influenced by North American writers 39
bringing a more autonomous set of interests to the study of immi- 40
gration in Europe (especially Brubaker 1989, 1992). The models ap- 41
proach was popular because it proved to be such an effective heuris- 42
tic strategy: reducing the problem of the vague and indefinable object 43

380 adrian favell
1 of enquiry – a national ‘society’ in all its complexity – to a ‘model’

2 which captures the key explanatory variables of social change. These
3 were invariably identified as ‘path dependent’ historical sources of
4 national cultural difference. The most well-known argument linked
5 to the models approach has been the classic distinction between the
6 ‘ethnic’ and ‘civic’ nation in citizenship studies, distilled from a re-
7 ductive (and largely inaccurate) stylization of French and German
8 nationality law as ideal types of ius soli and ius sanguinis citizenship.
9 It was surely questionable to ‘explain’ the differences between these
10 two similar cases by reference to national ideologies, themselves pro-
11 duced in the past by nationalist intellectuals and state actors to distin-
12 guish one nation from the other (on this see Weil, 1996). Yet even if
13 historically dubious, the power of the contrast here worked to gener-
14 ate effective normative arguments about a de facto national conver-
15 gence across Europe foreseeing mixed sources of nationality and a
16 limited recognition of ius soli for second and third generations – thus
17 helping German policy makers to move towards reforms (Hansen
18 and Weil, 2000).
19 The deeper explanatory challenge here would be to produce a
20 ideological modes more reflexive understanding of the ideological modes by which sim-
21 ilar European nation-states have justified and reproduced their own
22 models, as culturally distinct projections of collective identity (see
23 Favell, 1998; Alund and Schierup, 1991; Joppke, 1999). More even
24 handed comparison has gone on to recognize that while national pol-
25 icy legacies matter, they cannot be reduced to positive and negative
26 national examples. One response was the move to introduce typolo-
27 gies of incorporation, factoring in modes of state-society relations
28 and multi-levelled constitutional structures, as a more sophisticated
29 reflection of the different factors determining integration. Soysal’s
30 work in particular had the virtue of turning the ethnic/civic distinc-
31 tion on its head: highlighting in its arguments about the postnational
32 status of migrant groups such as Turks in Germany, the normative
33 dogma involved in always equating full national citizenship with
34 full integration (Soysal, 1994): Structured case-by-case comparisons
35 along these institutionalist lines have enabled a more fruitful type of
36 cross-national work, particularly those located at sub-national levels
37 such as the city (i.e., Ireland, 1994; Bousetta, 2000).
38 However, away from these predominantly North American led
39 comparative efforts, more explicitly policy-oriented studies with
40 a comparative range have tended to follow the least sophisticated
41 academic approaches. This has certainly been the case with work
42 produced through the sponsorship of European institutions. For
43 example, the big winner from an intense bidding struggle among

academics in this field for money from the Targeted Social and 1
Economic Research (TSER) programme on ‘exclusion’ was a nation- 2
al models-based study – led by well-known national figures Friedrich 3
Heckmann and Dominique Schnapper – that explicitly structured its 4
investigations around the idea that immigration and ethnic relations 5
in each country are determined by classic policy ‘models’ rooted in 6
political cultural differences between France, Germany, Britain and 7
so on (Heckmann and Schnapper, 2003). A models-based approach 8
of this kind will often itself reproduce the ideological fictions each 9
nation has of its own and others’ immigration politics. Schnapper 10
and associates duly found that minorities and majorities do indeed 11
talk about the issues in each country in ways that follow the distinct 12
national ideologies. But little or no self-reflexive effort was made to ask 13
how these nation-sustaining ideas about distinct national ‘models’ 14
have themselves been created and sustained by politicians, the media 15
and the policy academics themselves in each country, precisely in 16
order to foreclose the possibility that external international or trans- 17
national influences might begin to affect domestic minority issues 18
and policy considerations. 19
Practical institutional imperatives also dictate that the policy study 20
packages and presents its findings in a narrowly targeted way, which 21
naturally curtails many of the more interesting lines of enquiry. This 22
has been well-understood by one of the more influential NGOs in this 23
field in Brussels – the Migration Policy Group – who have been in- Migration Policy 24
volved in two of the most wide ranging funded surveys on integration Group 25
policies across European society (Vermeulen, 1997; MPG, 1996). In 26
the latter, the ‘societal integration project’, they set up roundtables 27
in around twenty countries, and listened to the expert opinions of 28
policy makers and policy intellectuals, generating a mass of mate- 29
rial about how policy makers talk about the same issues in different 30
places. However, in the end the slim report of highlights and recom- 31
mendations boiled all this down to a reaffirmation that convergence 32
was the source of future norms on citizenship and integration across 33
Europe. Being limited to the typical state-centric talk and self-justifi- 34
cation of policy makers, it was unable to offer any genuine compara- 35
tive evaluation. Moreover, the freedom of reflection of such a project 36
is naturally cut down by the expectations of the sponsors who lay 37
down the lines of research. By definition, such comparative policy 38
studies produce findings which reinforce the state-centred, top-down 39
formulations familiar at national level. The one difference here – as 40
a product of a supra-national European initiative = is that the conclu- 41
sions about the inevitability of convergence underline a familiar EU 42
strategy to focus, not on national exceptionality or uniqueness (as do 43

382 adrian favell
1 national level studies) but rather on the narrowing of national dif-

2 ferences. In other words, as we might expect given the sponsors in-
3 volved, these arguments work to narrow down the freedom of agency
4 of individual states, hence their sovereignty. Convergent citizenship
5 criteria become like convergent criteria for monetary union.
6 To really be able to answer the evaluatory question of which
7 nation-states are doing better on integration than others, we would
8 integration index need some kind of ‘integration index’: a convertible scale which en-
9 abled us to read off across European societies degrees of social segre-
10 gation in housing, success in schooling or employment, differences
11 in resistance of cultural behaviour, persistence of racist attitudes,
12 relative social mobility, or whatever is argued to be the best set of
13 objective measures. These indicators would then have to be linked to
14 the existence, or the success and failure, of specific national policies
15 or institutions. The inevitable impulse to cross-national evaluation
16 of state policy is not only exceedingly difficult to do, given the cross-
17 national data constraints I will go on to discuss. It also imposes as an
18 assumption an untenable automatic correlation between success on
19 the index and the effectiveness of state policies having achieved their
20 goals by shaping or influencing the behaviour of groups and indi-
21 viduals. This assumption itself is a state-reinforcing one, penalizing
22 any society which is less structured by state intervention, regardless
23 of how well ‘integrated’ groups or individuals may in fact be.
24 The one way this kind of approach works is as a comparative
25 shaming strategy directed towards states with less extensive formal
26 rights and entitlements for migrants than others. The most exten-
27 sive survey of this kind was a six nation Austrian study which did
28 just this, in order to shame the Austrian government into better mi-
29 gration policy and anti-discrimination measures (Çinar et al., 1995;
30 Waldrauch and Hofinger, 1997; Waldrauch, 2001). The extensively
31 documented study broke down all formal rights and entitlements of
32 non-nationals across various European states, rating each one be-
33 tween 0 and 1 as an index to barriers to integration. By definition,
34 the approach foresees a state-centred, state-organized solution to in-
35 tegration, and cannot capture any forms of multiculturalism which
36 are the outcome of more laissez-faire style approaches. We end up
37 with the very common conclusion that highly state-organized societ-
38 ies, such as Sweden or the Netherlands, do it best. Yet these are also
39 highly unified national societies, who put high demands of linguistic
40 and cultural assimilation on their inhabitants (something to which
41 the index is blind). They are also societies racked with dilemmas of
42 informal economy, and high degrees of social segregation among
43 their immigrant population. Current discussions on immigration in

Denmark provide a good example of the paradoxes here in some of 1

Europe’s most enlightened social democracies. Laws and policies en- 2
sure excellent access to rights and high rates of formal participation 3
among the so-called ‘new Danes’. Yet the many socioeconomic prob- 4
lems linked to disadvantaged immigrants are routinely interpreted 5
in political discussion as dysfunctional to the smooth running of the 6
Danish national welfare state, and stigmatized as ‘ikke dansk’; i.e., 7
rule-breaking immigrants not behaving in a ‘true’ Danish manner 8
(on Denmark, see Schierup, 1993). 9
Rights-based evaluations of integration contrast dramatically with 10
those which focus on different formal indicators. Britain, with its Britain 11
weak constitutional structures and idiosyncratic race relations insti- 12
tutions, does rather badly in the Austrian study, yet this contrasts 13
sharply with how comparative British evaluations of European expe- 14
riences view the matter. Contrasting its longstanding and successful 15
multicultural practices with the troubled politics and social situations 16
of many continental European societies, the most extensive studies 17
made by British researchers have always found Britain to be far bet- 18
ter endowed with antidiscrimination legislation and multicultural 19
policies (Forbes and Mead, 1992; Wrench, 1996). The British state in 20
fact pursues a minimalist style of intervention into the many and di- 21
verse forms of multiculturalism that have developed in the country. 22
Yet homegrown studies routinely link these successes to the agency 23
of the British state and its policy legacy: what is perceived by them 24
as the existence of a strong state-centred multicultural race relations 25
framework. Multiculturalism is thus claimed as an achievement of 26
the British state, rather than a consequence of the weak penetration 27
of the state in everyday life in Britain. From this point of view which 28
is more plausible in a comparative perspective – it could be argued 29
that it is laissez faire that has enabled London and a small handful 30
of other cities to develop as multicultural cities, in sharp distinction 31
from the white and intolerant provincial hinterlands. 32
As more positive visions of multicultural integration become 33
prevalent across Europe, other less ‘advanced’ integration nations 34
than France or Britain are likely to follow their lead and see their 35
ruling national elites claim the multicultural success in the name 36
of their own tradition of nationhood. For sure, France and Britain 37
look like successful multicultural societies on this score. Yet, it is 38
precisely a country like France which imposes the biggest cultural 39
burdens on newcomers in terms of their adhesion to the particular 40
ways of the nation; or a country like Britain, which buys enlightened 41
race relations as a trade-off for some of the toughest border controls 42
in Europe. These paradoxical results follow from the fact that both 43

384 adrian favell
1 countries practice ‘multiculturalism-in-one-nation’: a multicultural

2 nationalism, that sees no other source of multiculturalism than the
3 miraculously tolerant cosmopolitanism of the home culture. Such
4 countries may then be ‘universalist’, and yet apparently highly intol-
5 erant of specific cultural differences; or they might be highly multi-
6 cultural and multi-racial, and yet be at the same time extraordinarily
7 xenophobic. There are clear costs involved in the stubborn mainte-
8 nance of the fiction of exclusive nation-state agency over the multi-
9 cultural aspects of these locations.
10 national The strong sense of national self-preservation displayed here per-
11 self-preservation haps explains why the European Union has only been able to gain
12 the weakest influence over immigrant integration policies, jealously
13 guarded at the national level. The EU can get involved to identify
14 good practices, or the best convergent norms across societies; but
15 it cannot begin to constitute itself as a political agency here without
16 taking agency (i.e., sovereignty) away from nation-states, which have
17 used issues of immigrant integration precisely to actually underline
18 and reproduce their own existence as coherent, bounded, nation-
19 building societies. European integration is of course itself the search
20 for political agency at a supra-national level; but the fact that it seems
21 to fail to constitute itself as a state, suggests that this is largely be-
22 cause the actual boundaries of European society remain very much
23 fixed at the national level.
24
25
26 Survey and census based work on integration
27
28 It is no surprise that policy-centred studies should inevitably repro-
29 duce the state-centred, nation-building optic in their framing and
30 prescription of ways to achieve integration. As the preceding discus-
31 sion has indicated, such studies by definition can say very little about
32 the kind of less structured social processes that are characteristic of
33 much multiculturalism to be found in Europe’s cities and metropoli-
34 tan regions. Rather, where they recognize multiculturalism, policy
35 and institutional-based studies tend to bolster nationalizing ideolo-
36 gies which affirm the nation-state as the sole relevant locus of politi-
37 cal agency able to shape a ‘society’. They are also, needless to say, the
38 contributions which best chime with the interests of agents of the
39 state, concerned with maximizing their realm of political influence
40 by emphasizing the growing importance of top-down immigration
41 and integration policy.
42 But what of bottom-up studies: empirical work which focuses
43 on the experiences, attitudes or social mobility of the immigrants or

ethnic minority members themselves? Policy and institutional-based 1

studies often have very little to say about actual migrant experiences 2
of integration. Here, more ambitious uses of survey and census-based 3
work, based on studying their values, discourses and behaviour, of- 4
fer a more advanced integration index for measuring and evaluating 5
what is going on. Clearly, this would be material close to the actual 6
process of social change going on inside ‘multicultural’ nation-states; 7
and, it might be thought, material more likely to reveal evidence of 8
tendencies that are decomposing the conventional nation-state inte- 9
gration paradigm. For example, it might be expected to find strong 10
evidence – in those European cities that are significant ‘nodes’ in 11
the global economy – of the growing transnationalism characteristic 12
of the social and cultural forms of migrant groups whose activities 13
are embedded in global economic networks (see Faist, 2000; Rath, 14
2000). 15
Ambitious studies along these lines are now beginning to emerge. ambitious studies 16
The possibility of doing such work has grown out of an increasing so- 17
cietal thirst for more systematic knowledge about immigration phe- 18
nomena as the political salience of the subject has risen. Governments, 19
policy think tanks, international institutions and the media, are all 20
beginning to show interest in funding much more large-scale sur- 21
vey data driven studies of integration issues. The positivistic style 22
of large-scale survey work offers an interesting counterpoint to the 23
normative leanings of policy studies and institutional-based works, 24
which have tended to frame their more journalistic-style methods 25
with the value-laden rhetoric of citizenship and rights. Survey-based 26
researchers, meanwhile, preserve their credibility, not by shadow- 27
ing the language and conceptualizations of policy actors, but by the 28
distinct ‘scientific’ autonomy of their methodology and results. By 29
definition, the kind of work they are doing cannot be mounted by the 30
personnel of governments and newspapers, lacking in the specialist 31
quantitative and qualitative techniques required; such work has to be 32
commissioned, with freedom of research negotiated in advance. This 33
fact creates distinctive material conditions for the kind of work pro- 34
duced. One advantage is that the process of deriving ‘policy’ directed 35
normative conclusions is (or should be) left to post-hoc interpreta- 36
tion, and not in-built in the normative state-centred conceptualiza- 37
tions which typically measure integration: such as those which rate 38
already institutionalized state policy structures linked to citizenship 39
rights or legal and political channels. 40
Numerous examples of impressive large-scale survey work do 41
now exist in various countries at the single-case national level (see 42
the discussion in Phalet and Swyngedouw, 1999; examples are 43

386 adrian favell
1 Modood et al., 1997; Tribalat et al., 1996; Swyngedouw, Phalet, and

2 Deschouwer, 1999; Phalet et al., 2000; Diehl et al., 1999; Veenman,
3 1998; Lesthaege, 2000). The new frontier for survey-based research
4 is the possibility of cross-national comparative survey work on the
5 integration of immigrants. However, as was clear from exploratory
6 discussions at a conference in September 1999 on the subject or-
7 ganized by Hartmut Esser – which brought together the European
8 Consortium for Sociological Research, a grouping of the leading
9 quantitative social scientists in Europe – very few of the epistemo-
10 logical problems of doing such work have yet been considered by
11 researchers more familiar with doing cross-national studies on
12 employment, educational mobility or inequality (e.g., Erikson and
13 Goldthorpe, 1992). Cross-national efforts have to be synthesized
14 from the best of the national level data provided on a nation-by-na-
15 tion basis by governments. The very best of current cross-national
16 efforts in the area of immigration mounted by an international or-
17 ganization, which monitors migration stocks and flows around the
18 SOPEMI developed world – the annual OECD-SOPEMI report – is notoriously
19 hampered by the fact that the expert respondents each report figures
20 for its own country based on different national means of data-gather-
21 ing (SOPEMI, 1998). Moreover, there is nothing like the systematic
22 quantitative effort on integration questions as there is in the report
23 for basic issues of entry, legality, residence and so on. The report
24 does have a growing section on integration, but it is by far the weak-
25 est part of it, reflecting perhaps a lack of sociological expertise among
26 the geographers and economists who make up the immigration spe-
27 cialist panel. The report in fact falls back into a more policy-centred
28 style of analysis: reproducing the same old frameworks about nation-
29 al models and comparative rights indices.
30 We can imagine perhaps a more concerted attempt to concep-
31 tualize the integration questions in a way which escapes this nomi-
32 nalist nation-state centred approach. But the real problem here is
33 that all available data on immigrant or minority numbers basic to
34 the SOPEMI effort, follow the significantly different conventions
35 in each country about collecting population data. There is, in other
36 words, an in-built dependency on nationally-specific research tech-
37 nologies; usually the state apparatus that has been built up around
38 census gathering. The specific methods used to identify populations
39 of immigrant origin in the post-war period vary from country to
40 country, as does the political sensitivity with which this information
41 is released or extrapolated. The technical methods – and the politics
42 surrounding such sensitive state knowledge production – inevitably
43 reflect the national ideology each nation has fashioned for itself as a

narrative of nation-building. No matter how insulated the methodol- 1

ogy, the broader national policy definition of integration as a social 2
process impacts upon the production of categories and numbers elic- 3
ited from survey results. 4
Counting only non-nationals as the immigrant population is still 5
the base-line norm across nearly all European countries except Britain, 6
which has a famously idiosyncratic form of ethnic self-identification 7
in its census. Most comparative tables offer figures for non-nationals 8
by nationality, which works up to point in countries where original 9
nationality remains a distinguishing factor (as, say, in Germany, Italy 10
or Spain; although it runs into problems in Germany, for example, 11
in counting the three million Aussiedler from Eastern Europe). This 12
method is clearly a criterion of declining usefulness, however, as in- 13
creasing numbers of second and third generation immigrant chil- 14
dren in fact accede to full national citizenship; it can indeed be sim- 15
ply a crude measure of administrative exclusion. Naturalization rates 16
over time are a second set of figures, which trace the absorption of 17
immigrants over shorter, given periods of time. Other countries may 18
also offer figures which count those people who identify older family 19
members born outside of the country. From this, a great deal can be 20
extrapolated into second and third generation, but a country such as 21
France still maintains barriers for ideological reasons to researchers 22
using this information, which means that some naturalized second 23
or third generation are lost to studies once they leave the immigrant 24
household. 25
A strong moral prohibition, meanwhile, exists on the classifica- racial classification 26
tion of people by race or religion across Europe. There is little more 27
distasteful to continental Europeans than anything with a whiff of 28
former Nazi racial classifications, or indeed the common practice 29
in multinational empires such as the Soviet Union or Yugoslavia to 30
brand people permanently on their interior passports with an ‘ethnic’ 31
nationality (see Brubaker, 1995). However, a more racially heteroge- 32
neous population such as the Portuguese avoids these racial classi- 33
fications for rather different reasons, to do with the cosmopolitan 34
colonial conception of the nation. In Belgium, you are classified by 35
language according to political records after you vote, religion after 36
you choose university. Here, however, the census is banned by law 37
to answer such questions up front. In the Netherlands, meanwhile, 38
there is no national census at all, after a libertarian public revolt in 39
the 1970s. Ethnic statistics here have to be reconstructed from lo- 40
cal city and police records or special ministry surveys, something 41
that has contributed significantly to the sense of unease about the 42
numbers of ‘undocumented’ residents in the country. Other coun- 43

388 adrian favell
1 tries, however, such as Denmark and Britain – which in other re-

2 spects have very different census methods – are prepared under cer-
3 tain circumstances to make available census data to track specified
4 (anonymous) individuals over time between censuses, in order, for
5 example, to analyze spatial mobility or rates of political participation
6 (see Togeby, 1999; Fielding, 1995). Such a babel of census informa-
7 tion is a difficult starting point. In talking about integration, who are
8 we talking about: ‘legally resident foreigners’, ‘immigrants’, ‘illegal/
9 undocumented residents’, ‘third-country nationals’, ‘ethnic groups’,
10 ‘racial minorities’, new or naturalized ‘citizens’, or simply formally
11 undistinguishable ‘nationals’ with a different de facto cultural history
12 or skin colour?
13 The narrow definition of immigrants as resident non-nationals
14 has the virtue of avoiding the integration issue entirely. It offers the
15 normative panacea of equating citizenship with full integration, an
16 idea which has long reassured French republicans on the virtues
17 normative dogma of a cosmopolitan type of nationhood. A normative dogma such as
18 this makes no sociological sense, of course, once anyone is willing
19 to admit that host populations and migrants alike will continue to
20 informally discriminate themselves and each other regardless of
21 which passport they are holding. Once some outsiders become insid-
22 ers, however, their formal categorization (or ‘recognition’, in more
23 affirmative terms) itself becomes a part of the integration process.
24 Whether or not they are separated off for official monitoring pur-
25 poses, and how and where they can be placed on some path towards
26 full integration, becomes a crucial part of the integrative process it-
27 self, not least because the separation from one’s original nationality
28 may also be a coercive state enforced act (see Simon, 1997). There is
29 a profound moral truth in the French refusal to actually recognize
30 any French citizen of non-national ‘ethnic’ origin as such in official
31 statistics, because the recognition itself can indeed be a form of in-
32 equality or discrimination. The power of naming does indeed count
33 for something. The French refusal is also a dramatic statement of the
34 nation-state’s continued prerogative to nationalize a new citizen as
35 indivisibly French. Yet, on the other hand, no policy can be devised
36 for systematic integration of foreign-origin groups until the nation-
37 state begins to collectively recognize and classify minorities of ethnic
38 origin, with special claims – targeted policies, resources, legal allow-
39 ances, etc – that follow from this (this is the central problematic of
40 the influential work of Kymlicka, 1995).
41 There is another side of the classificatory separation, however.
42 Integration cannot be conceived, identified, let alone measured as
43 degrees of inequality and so on, until a control group representa-

tive of the national population has been specified. But this raises the 1
question: we are talking about integration into what? Here, the logic 2
of classification becomes even more slippery. Are they the indige- 3
nous population (‘de souche’ in French), but if so, what length of ethnic 4
time constitutes ‘roots’; are they defined culturally, by their family self-identification 5
origins, by their length of residence; are they, rather, simply to be question 6
identified as the majority ‘white’ or ‘European’ population; or, are we 7
in fact speaking of some representative sample or statistical mean of 8
the citizenry as a whole, including all those new and culturally exotic 9
recent additions? Moreover, as Michael Banton points out (2001), 10
it makes little sense to measure the integration of an immigrant or 11
ethnic minority population, until we have some precise measure- 12
ment of how well the majority population is integrated as a nation. 13
Whatever method is chosen – however the state chooses to classify, 14
count and control its population or define those who are in and those 15
who are out – will again amount to a pre-determined national ‘sam- 16
pling frame’, that is very closely linked to the ideological concept of 17
nationhood present. Behind this, of course, lies the normative com- 18
mitment to integration as societal end-goal, the underlying assump- 19
tion that holds the nation-state-society unit together. Researchers 20
who thus set out to objectively measure integration, without taking 21
into account how much the nation-state unit has already determined 22
the very quantitative tools they use, will fail to see how much the 23
bounds of what they can discover have already been pre-set for them. 24
If so, they are working no less to underwrite the predominance of the 25
nation-state optic, than policy studies researchers who accept without 26
challenge nation-state centred definitions of ‘universal’ citizenship 27
or ‘cosmopolitan’ multiculturalism. 28
On the whole, however, progressive minded commentators 29
across Europe do not challenge this conceptual recuperation of their 30
very tools of research by a nation-state centred vision of integration. 31
The majority, rather, has been content to push a different, concilia- 32
tory line, that squares the circle between the reality of ongoing na- 33
tion-building efforts and the contrasting idealism of cosmopolitan 34
multiculturalism. They argue that European nations have become, 35
or are becoming, ‘countries of immigration’. Such arguments have 36
been very much present in those countries whose right wing refuses 37
to recognize the reality of continued immigration and settlement at 38
all. Among those promoting this happier version of Europe’s im- 39
migrant future, the coercive weight of ever-present nation-building 40
processes is thus lightened by the claim that the integration of immi- 41
grants in Europe can be equated with what happens to immigrants 42
in Australia, Canada or the US. The normative inspiration is clear 43

390 adrian favell
1 – constitutional universalism, cosmopolitan idealism, the melting

2 pot, open immigration regimes, and so on – but the idea of the old
3 nation-states of Europe metamorphosizing into brand new ‘coun-
4 tries of immigration’ is a dubious rhetoric on any empirical level,
5 not least from a historical point of view. In Europe, we are talking
6 about tightly bounded and culturally specific nation-states dealing in
7 the post-war period with an unexpected but still not very large – in-
8 flux of highly diverse immigrant settlers, at a time when, for other
9 international reasons, their sense of nationhood is insecure or in de-
10 cline. It is a problematic very different to those faced by the US or
11 Australia, whose histories and sense of nationhood have always been
12 built on immigration. Europe, rather, faces a problematic where the
13 continuity of nation-building is perhaps a much more significant fact
14 than the multicultural hybridity that is sometimes sought for in these
15 other, newer ‘model’ nations. A great deal of revisionist effort has
16 gone into reconstructing certain European nations as undiscovered
17 immigration nations (e.g., Noiriel, 1991). Although widely accepted,
18 it is an effort which in fact empties significance out of other empiri-
19 cal attempts to problematize integration as a limited process of cul-
20 tural change, combining multicultural adaptation with national re-
21 invention. Instead, it rather lamely gestures European survey-based
22 culturally neutral researchers back towards the most culturally-neutral model available:
23 model that of classic American assimilation research, which charts the prog-
24 ress of different immigrant ethnic groups towards some ideal-typical
25 absorption into the suburban middle class – a process where the per-
26 vasively national orientation of American assimilation is never even
27 put into question, and where the nation-building effect here stays
28 invisible (see also Brubaker, 2001). The spectacular resurgence of
29 American patriotism in its crudest forms post ‘9/11’ has at least clari-
30 fied how deeply nationalistic ideas of American unity and America’s
31 global role in fact are.
32 Operationalizing this particular normative frame for immigrant
33 integration which recasts European societies as immigration nations
34 in the idealized, immigrant American mould – has been done in dis-
35 tinctive national ways. On the face of it, the French offer the purest
36 instance of a self-styled universalist country of immigration, not least
37 after the assiduous reconstruction of this idea by historians and so-
38 ciologists in the 1980s. Establishing this as the normative frame for
39 new progressive policies was relatively straightforward. But, in em-
40 pirical terms, the formal prohibition in official survey data on intro-
41 ducing any sub-categorization of the population by ethnicity (i.e., in
42 the data produced by the national statistics office, INSEE), left gran-
43 diose declarations about the continued success of the French republi-

can model bereft of evidence for these claims. For how else could the 1
sociological integration of different cultural groups in France in fact 2
be measured? A study which reintroduced some sub-classification 3
of the population by ethnicity was, in other words, needed to show 4
that ethnicity in fact did not matter. The nation-sustaining argument 5
about integration was in a sense generating its own contradictions, 6
that would then need resolution by a new scientific approach. This, 7
then, was the background to the ambitious study by INED, headed by 8
Michèle Tribalat, that still represents the state-of-the-art in integra- 9
tion research in France (Tribalat, 1995; Tribalat et al., 1996). Sample 10
ethnic groups of different national origin – tracked down by ethno- 11
graphic investigation, using the census only indirectly – were com- 12
pared to a control group of non-immigrant origin French on ques- 13
tions of cultural behaviour, language use, housing concentration, 14
political participation, and so on. The strongly French socialization 15
of most groups observed – the Turkish and Chinese being the two 16
outliers – in fact offered strong evidence for continued ‘assimilation’ 17
in France, as Tribalat preferred to call it. The mere introduction of 18
ethnicity into the survey, however, brought desperately controversial 19
public reactions from other commentators, such as Hervé Le Bras 20
(1998); and this despite the fact that it led to such conventionally 21
‘French’ results. 22
Systematic cross-ethnic comparative work is much more highly Germany: strong 23
developed in Germany, which has strong national surveys of data by national-origin data 24
national-origin available, such as the socioeconomic panel commis- 25
sioned annually by the Deutsches Institut für Wirtschaft, which pro- 26
vides data on ethnicity, language, identity questions and participa- 27
tion (an example of such work being Diehl et al., 1999). Progressive 28
researchers here are even more sensitive to the de-categorization of 29
foreigners and the positive idea of Germany as a country of immi- 30
gration. There have been advantages to such research in the fact it 31
has had to be diverted away from the ideologically dominated dis- 32
cussions on citizenship and naturalization, where progress has been 33
more difficult. German research is thus more likely to concentrate 34
on conceptualizing integration in technical socioeconomic terms: 35
in terms of participation in the welfare state, and in differences be- 36
tween federal or city level contexts. One consequence is the possibil- 37
ity of internal comparisons of integration geographically within the 38
nation, something of which there is no trace in France and Britain. 39
German research, however, does not escape the pervasively nation- 40
centred frame which dominates its political debates. Negative evi- 41
dence of non-integration – such as ethnic concentration or the fail- 42
ure of second and third generations to speak German – tends to get 43

392 adrian favell
1 constructed as evidence of segregation or marginalization, in con-

2 trast with more successful state-centred integration or assimilation.
3 These closed typologies of immigrant trajectories – which reinforce
4 the idea of full national integration as the ideal – can be found in
5 research going on in all kinds of countries (Nauck and Schönpflug,
6 1997; see also the closed scheme of claims-making laid out as an
7 introduction by Koopmans and Statham, 2000).
8 Britain In Britain, meanwhile, the ‘ethnic’ self-identification question in
9 its census is clearly out of sync with its European neighbours. It in-
10 dicates a conceptual history that has always looked for its normative
11 inspiration to American race relations of the 1960s, and has always
12 defined Britain more narrowly as a country of postcolonial immigra-
13 tion only. For all the masses of data provided about the select group
14 of post-colonial racial and national groups recognized in the census,
15 the framework has come to have serious limitations over time. The
16 categories themselves have become highly politicized, putting into
17 practice a variable geometry that has sought to respond to the emerg-
18 ing demands of new and increasingly diverse migrant groups who
19 recognize that the census categories are a fundamental source of rec-
20 ognition, as well as legal coverage and public funding. Basic black
21 and white distinctions, for example, have now fallen away into a
22 broader recognition of Asian groups. Other new migrants in Britain,
23 however, find themselves lost between the generic ‘white’ and ‘other’
24 boxes. Indeed, with Jewish and Irish anti-discrimination campaign-
25 ers forcing open the pandora’s box of whiteness (the all important
26 control group) in the census of 2001, it is quite likely that the sharp
27 ‘minority’ ethnic groupings that have been the core and inspiration
28 of British race research may in future begin to crumble.
29 Obviously, the sources of minority data, and the qualitative evi-
30 dence it also provides about nuances in ethnic self-identification,
31 have created a boon for identities type work in Britain, much of it
32 now pursued under the banner of ‘new ethnicities’. There are nu-
33 merous studies in which individuals are ethnographically studied
34 playing with or resisting (unsurprisingly) their given ‘ethnic minor-
35 ity’ category (Back, 1995; Anthias and Yuval-Davis, 1993; Modood et
36 al., 1994). Such work can often be an ideal vehicle for articulating
37 ethnic ‘voices’ themselves. But structural work about the social mo-
38 bility of such groups is hampered by the crude comparison forced by
39 the data between racially designated ethnic groups and the generic
40 ‘white’ block of the host population; this, inevitably it seems, leads
41 research to claim ethnic success as rooted in minority group solidar-
42 ity, but ethnic failure as rooted in majority group racial discrimina-
43 tion. In this frame, too, there is no way of assessing the continued

impact of nation-building assimilation – via evidence on cultural be- 1

haviour, etc – on ethnic groups, despite the self-evident Britishness 2
of many of these well-established minority groups. Nor is it easy in 3
this frame to cross-check for class, gender or regional factors, par- 4
ticularly if these might lead to the declining salience of race-based 5
explanations. In some of the best recent work on social mobility and 6
ethnic identities, transnational behaviour and sources of social suc- 7
cess are still surprisingly downplayed against the interpretation that 8
ethnic minority success is further proof of vibrant British multicul- 9
turalism (Modood et al., 1997). Britain celebrates with some pride 10
its longstanding role in Europe as the leading country of post-war 11
immigration; yet has until very recently refused officially to see itself 12
as a country of new immigration. Within this paradoxical picture, 13
well-integrated and recognized ethnic minorities have a status and 14
advantage denied to the many other new migrant groups now found 15
in the country. 16
In the nation-state centred version of integration research in the 17
larger European countries, there is something odd about the fact 18
that the status and success of immigrants gets measured entirely in 19
terms of a social mobility relative to norms of integration into the 20
nation-society, or average national social mobility paths; yet it is in- 21
creasingly normal to think of elites in the same country becoming 22
increasingly transnational in their roles, networks and trajectories. 23
The exclusive destiny of full integration into host nation states may 24
however not be the norm for immigrants in the future. Already, in 25
other smaller European nations, a rather different picture is emerg- 26
ing. New migration countries such as Italy, Spain, and Portugal are 27
actually going through the process of formulating their own uncer- 28
tain national conceptions of integration at a very different histori- 29
cal moment compared to larger nations who continue to offer their 30
models. As well as being countries that are more geographically ex- 31
posed to migration, they are, moreover, countries with weaker state 32
penetration of society or the market. In these less structured situa- 33
tions, the normative imperative of full national integration begins 34
to lessen, if new non-nation-centred structures of social integration non-nation-centred 35
begin to emerge. A similar consequence follows from research on in- structures of social 36
tegration into a non-unified or multi-levelled state such as Belgium. integration 37
In seeking to avoid the inevitability of nation-state centred visions of 38
integration apparently forced on research by the kind of data avail- 39
able and the kind of concepts we work with, studying these smaller 40
or newer integration scenarios may indeed offer a way forward out of 41
the current paradigm. 42
43

394 adrian favell
1 Beyond the integration paradigm?

2
3 The clear message from the critical survey of current integration re-
4 search in Europe offered here is that better research would be re-
5 search that sets out to be more autonomous academically, and more
6 thoroughly comparative in its intent. Academics need to escape their
7 role of underwriting nation-building efforts directed towards small
8 immigrant populations that have provoked a renewed symbolic effort
9 to imagine (inclusive) western nation-state cultures. A much higher
10 degree of self-consciousness is needed about the way contextual fac-
11 tors determine the intellectual content of research itself.
12 How might this be done? I will conclude with a discussion of
13 some of the newer insights provided by the way scholars of transna-
14 tionalism have approached the problem (e.g., Portes, 1996; Basch et
15 al., 1994; Smith and Guarnizo, 1998). Scholars of transnationalism
16 have sought – for exactly the kinds of reasons I spell out in this paper
17 – to expunge ‘integration’ from their terms of research. By definition,
18 they do not wish to be underwriting the nation-state in a world which
19 they see as increasingly transnational or global. Methodologically,
20 too, their bottom up, ethnographic drive suits a style of work which
21 draws large conclusions from the study of cases likely to be seen as
22 exceptional, or indeed deviant from the conventional integration-fo-
23 cused perspective. For sure, it is this too which may account for the
24 often excessively celebratory tone of transnational studies. Seeking a
25 new kind of liberation, some studies fall into the longstanding prob-
26 lem that has distorted much radical ethnic and racial studies: the
27 transfer of sympathy for the experiences, difficulties, and sometimes
28 plight of migrants and ethnic minorities, into visions of these groups
29 as some sort of heroic new ‘proletariat’. Although the ‘search for a
30 new world’ – and the slogan ‘globalization from below’ – is the rather
31 romantic packaging chosen in the work of Portes, Castells et al., this
32 should not deflect us from the key insights of their work. Its ma-
33 jor advance has been the empirical uncovering of trans-state, trans-
34 nation economic and cultural networks of transactions (and protean
35 forms of social organization) among new and developing migrant
36 groups. These networks are clearly generating sources of collective
37 social power outside of territorial state structures familiar from our
38 conventional understanding of the world of nations. Whereas Portes
39 principally recognizes the source of transnational power as the global
40 market, others might point to Islam or Hispanic culture, or indeed
41 informal sources informal (‘illegal’) sources of these same powers (see Cohen, 1997;
42 of power Phizacklea, 1998).
43 The other crucial aspect of Portes’s work, however, is its insistence

on linking emergent transnational forms with classic integration 1

questions. The exploration of the notion of ‘segmented assimila- 2
tion’ in the US, has pointed towards the new structural relationship 3
between the transnational ‘survival’ strategies resorted to by mi- 4
grant groups and the unappealing ‘downward assimilation’ offered downward 5
to them by the host societies’ state and societal structures (Portes, assimilation 6
1995). European examples of this have been the similar emergence 7
of community resilience against the negative socioeconomic con- 8
ditions they found themselves in, or the strongly assimilatory host 9
reception. The results have been the paradoxical innovations of the 10
informal economy or inner city Islam in many European cities. The 11
integration path may indeed prove to be, in Kloosterman and Rath’s 12
terms, ‘a long and winding road’ (Kloosterman et al., 1998). As the 13
Dutch state, for example, seems ever tighter in its heavily legislated 14
attempt to discover, encompass, regularize and normalize the spon- 15
taneous economic activities of new migrants, so there has seemed 16
to be an ever-growing over-flow of undisciplined, self-organized 17
informal activities in the country (Engbersen, 1996). The very best 18
continental European work has focused on precisely this issue of in- 19
formality or non-institutionalized forms of social organization; often 20
focusing, unsurprisingly, on those groups identified in conventional 21
integration research as the ethnic cases which fit worst into the kinds 22
of automatically integrating schemes set up, for example, by French 23
and British research (Fennema and Tillie, 1999; Bousetta, 1997; 24
Phalet et al., 2000). It is not surprising that this work has invariably 25
focused on either Turkish or Moroccan groups in various countries: 26
two newer, non-colonial migrant groups that have displayed some of 27
the most pronounced ‘transnational’, non-integrating social trajecto- 28
ries in Europe. 29
Systematizing these deviant tendencies in research without sim- 30
ply reproducing the nation-state-society as the container unit has 31
proven a lot more difficult. One might point to the Polanyi-inspired 32
way forward in recent work by Faist (2000) or Kesteloot (2000). 33
In this they offer schemes of transnational or local integration in 34
economic and community structures which cross-cut with national, 35
citizenship-centred forms. Empirical anthropologists, too, have pro- 36
vided some of the best recent work about immigrant and ethnic self- 37
organization in urban contexts (Werbner, 1999; Baumann, 1996). 38
Whether it is the bustling migrant markets of old Antwerp or East 39
Amsterdam, or the mosque-centred inner city Islam of Turks and 40
Moroccans in Brussels, there is clearly a need to recognize these city- 41
embedded activities as emergent forms of social organization – and 42
hence social power largely unstructured or not incorporated (in for- 43

396 adrian favell
1 mal or informal terms) by the state. The somewhat anarchical multi-

2 culturalism of some European cities now points towards a new type
3 of multi-ethnic culture in Europe, rather different to the multicul-
4 tural citizenship shaped by integrating nation-states. It is not egali-
5 tarian, it is not anchored in rights, and it is certainly not conflict free;
6 but it is, for better or for worse, much less disciplined by the nation-
7 building pressures hidden in top-down policies of ‘integration’.
8 deterritorialized Interestingly, however, even this kind of multicultural challenge
9 nationalities to dominant European nation-state-centred cultures tends to still be
10 anchored in deterritorialized ‘nationalities’: the persistence of impor-
11 tant political and social links with the ‘homeland’, as both a concrete
12 and symbolic reference. This fact – which is certainly the case with
13 Turks and Moroccans in Europe – indicates a limit to these forms
14 of transnationalism outside of their European context. Viewed from
15 here they are not really transnational at all, but rather examples of de-
16 territorialized nation-state building, familiar perhaps from the older
17 diasporic histories of countries like Ireland, Italy or Greece. What
18 there is precious little evidence of across Europe is the kind of radi-
19 cal diasporic multicultural forms, beloved of British cultural stud-
20 ies writers: the ‘black’ Atlantic diaspora or ‘black’ Asian pan-ethnic
21 groups (see Gilroy, 1987; Brah, 1993; Hall, 1988). Such diasporas
22 would indeed constitute a more radical challenge to the present day
23 international system, still fixed upon relations between nation-states
24 in the western and developing world to the south and east. But their
25 absence betrays just how British these writers in fact are; reflecting
26 – in their archetypal radical responses to frustrations encountered in
27 the ethnic categories of the liberal multicultural race relations frame-
28 work – the everyday activist struggles of British race politics.
29 As these overwhelmingly national sources for transnational ideas
30 suggest, we should be wary of seeing transnationalism as an end
31 to the integration paradigm. Rather, transnationalism in Europe has
32 to be seen as a growing empirical exception to the familiar nation-
33 centred pattern of integration across the continent. This remains
34 the dominant focus for policy actors and migrant activists alike.
35 Transnationalism points towards the new sources of power accessed
36 by migrant groups when they begin to organize themselves and their
37 activities in ways not already organized for them by an integrating
38 nation-state. By setting these forms against the continuity of nation-
39 state centred patterns of integration, we may be able to understand
40 how and why new spaces in the empire of the state are beginning
41 to develop. What transnationalists should not do is leap beyond this
42 into claims of an emerging international or global structure, in which
43 all these nation-state challenging phenomena add up to a new global

framework of governance, at which level a new kind of incorporation 1

will be achieved (Soysal’s supra-national human rights regime, for 2
example). To do so is to project the same old normative nation-state- 3
building impulses onto an emerging international situation charac- 4
terized rather by its market and culture led undermining of tradition- 5
al nation-state powers. It means, in other words, to reinvent the state 6
by the back door at global level. There are, of course, political actors 7
who dream of a postnational state at European, even global level; but 8
the factual capture of this ideal by the far more powerful realpolitik of 9
everyday international relations, simply turns these efforts back into 10
a paradoxical ‘rescue of the nation-state’, to borrow Alan Milward’s 11
(1992) famous phrase. 12
In many ways, the continued focus on integration as the cen- 13
tral idea in postimmigration policy debates across Europe, is itself 14
a choice of rhetoric designed explicitly to rescue the nation-state. 15
European policy makers and commentators have begun to formu- 16
late more constructive visions of a multicultural future that will be 17
able to contain and structure within the nation-state the many new 18
forms of immigration and multiculturalism beginning to spring up 19
across the continent. As I have argued, these visions – and the aca- 20
demic research which has provided the knowledge to substantiate 21
their claims – have continued to work within a nation-state centred 22
paradigm, even when they claim to be transcending it. An awareness 23
of transnational phenomena, as well as a better consciousness of the 24
pervasive way work has been structured by a nation-state centred 25
epistemology, may enable migration and ethnic studies researchers 26
to escape in their analyses the normative constraints of the integra- 27
tion paradigm. But it is vital in looking for new concepts and tools to 28
describe the changing relations of state and society across the conti- 29
nent, that we also continue to recognize the extraordinary continuity 30
and resilience of the nation-state-society as the dominant principle of nation-state-society 31
social organization in Europe. 32
33
34
Notes 35
36
1 Published in ‘The multicultural challenge’, Comparative Social Research, 22, 37
2003, pp. 13-42, reprinted with permission from Elsevier Ltd. 38
2 A more extended discussion and survey can be found in Favell (2001).
Responding to this piece, Banton (2001) dismisses the use of ‘integration’ –
39
‘a treacherous mathematical metaphor’ – in any sociological studies on the 40
subject. His vision is to purify sociological research on ethnic and race rela- 41
tions of these pervasive ordinary language concepts. Though a valid scientific 42
response to the dilemma of using such terms, it forecloses the possibility in 43

398 adrian favell
our research of reflexively accounting for why such terms are so predomi-
1 nant in policy discussions and academic research alike. See also related dis-
2 cussions in Bommes (1998) and Wimmer and Glick Schiller (2002).
3 3 The number of quasi-academic policy studies on integration funded by such
4 organizations in recent years has been remarkable. The Council of Europe’s
Committee on Migration has produced a number of reports on gender and
5
religious issues, labour markets, and social and political participation, as well
6 as an outstanding conceptual framework for research by Bauböck (1994).
7 The ILO has pursued work on integration in labour markets (Doomernik
8 1998), and the OSCE has been linking minority rights and integration.
9 Among NGOs in Brussels, there is the highly active Migration Policy Group,
10 who have produced major cross-national studies of policies and policy think-
ing on integration (MPG 1996; Vermeulen 1997). Finally, charitable transat-
11
lantic organizations have also joined the trend. The Carnegie Endowment’s
12 massively ambitious ‘Comparative Citizenship Project’ identified political
13 and social integration as two key areas of concern (Aleinikoff and Klusmeyer,
14 2002), and the Canadian-led Metropolis project focused on migrants in cit-
15 ies has sponsored several major studies (i.e., Cross and Waldinger, 1997;
Vertovec, 1997). These various studies are some of the most ambitious com-
16
parative international projects to be found. Here, I mention but a sample.
17 4 For example, there was the creation by the left wing government of Italy in
18 1999 of a ‘Commissione per l’integrazione’ under the leadership of political
19 sociologist Giovanna Zincone. This was explicitly intended to counter the in-
20 creasingly salient use of negative anti-immigration rhetoric by Berlusconi’s
21 right wing coalition. In Denmark, again under pressure from the right, the
government passed an ‘Act on the Integration of Aliens in Denmark’ in July
22
1998, followed by much public discussion and further reports on continu-
23 ing integration problems. In Austria, the turn to integration (see Waldrauch
24 and Hofinger, 1997) has been formulated by the opposition as a response to
25 specifically exclusionary government attitudes.
26 5 In the report of the Commission on Multi-Ethnic Britain (2000), which
27 involved some of these more radical commentators alongside more main-
stream figures, ‘integration’ was the organizing concept that dared not speak
28 its name. However, the Commission’s chair, Bhikhu Parekh, has frequently
29 written about the concept in his own work (Parekh, 2000).
30
31
32
33 References
34
35 Alba, R. and Nee, V. (1997), ‘Rethinking Assimilation Theory for
36 a New Era of Immigration’, International Migration Review, 31
37 (Winter), pp. 826-74.
38 Aleinikoff, A. and Klusmeyer, D. (2002), Citizenship Policies for an
39 Age of Migration, Carnegie Endowment, Washington, DC.
40 Alibhai-Brown, Y. (2000), Who Do We Think We Are? Penguin,
41 London.
42 Alund, A. and Schierup, C-U. (1991), Paradoxes of Multiculturalism,
43 Avebury, Aldershot.

Anthias, F. and Yuval-Davis, N. (1993), Racialized Boundaries: Race, 1

Nation, Gender, Class and the Anti-Racist Struggle, Routledge, 2
London. 3
Back, L. (1995), New Ethnicities and Urban Culture: Racisms and 4
Multiculture in Young Lives, UCL Press, London. 5
Banton, M. (2001), ‘National Integration in France and Britain’, 6
Journal of Ethnic and Migration Studies, 27 (1), pp. 151-68. 7
Basch, L., Schiller, N.G. and Szanton-Blanc, C. (1994), Nations 8
Unbound: Transnational Projects, Post-Colonial Predicaments and 9
Deterritorialized Nation·States, Gordon and Breach, Amsterdam. 10
Bauböck, R. (1994), The Integration of Immigrants. CMDG-Report, 11
The Council of Europe, Strasbourg. 12
Baumann, G. (1996), Contesting Culture: Ethnicity and Community in 13
West London, Cambridge UP, New York. 14
Blommaert, J. and Verschueren, J. (1998), Debating Diversity, 15
Routledge, London. 16
Bommes, M. (1998), ‘Migration, Nation State and Welfare State: 17
A Theoretical Challenge for Sociological Migration Research’. 18
Paper presented to the European Forum on Migration, European 19
University Institute, Florence, 16 February 1998. 20
Bousetta, H. (1997), ‘Citizenship and Political Participation in 21
France and the Netherlands: Reflections on Two Local Cases’, New 22
Community, 23 (2), pp. 215-32. 23
Bousetta, H. (2000), ‘Political Dynamics in the City. Citizenship, 24
Ethnic Mobilisation and Socio-political Participation: Four Case 25
Studies’, in S. Body-Gendrot and M. Martiniello (eds.), Minorities 26
in European Cities: The Dynamics of Social Integration and Social 27
Exclusion at the Neighbourhood Level, Macmillan, London. 28
Brah, A. (1993), Cartographies of Diaspora: Contesting Identities, 29
Routledge, London. 30
Brubaker, R. (ed.) (1989), Immigration and the Politics of Citizenship 31
in Western Europe, UP of America, New York. 32
Brubaker, R. (1992), Citizenship and Nationhood in France and 33
Germany, Harvard UP, Cambridge, MA. 34
Brubaker, R. (1995), Nationalism Reframed: Nationhood and the 35
National Question in the New Europe, Cambridge UP, Cambridge. 36
Brubaker, R. (2001), ‘The Return of Assimilation? Changing 37
Perspectives on Immigration and Its Sequels in France, Germany, 38
and the United States’, Ethnic and Racial Studies, 24 (4), pp. 531-48. 39
Castles, S. (1995), ‘How Nation-States Respond to Immigration and 40
Ethnic Diversity’, New Community, 21 (3), pp. 293-308. 41
Castles, S., Korac, M., Vasta, E., and Vertovec, S. (2002), Integration: 42
Mapping the Field, Home Office, London. 43

400 adrian favell
1 Çinar, D., Hofinger, C, and Waldrauch, H. (1995), Integrationsindex.

2 Zur rechtlichen Integration von Ausländerinnen in ausgewählten
3 europäischen Ländern (Political Science Series No. 25), Institute
4 for Advanced Study, Vienna.
5 Cohen, R. (1997), Global Diasporas, UCL Press, London.
6 Cohen, R. and Kennedy, P. (2000), Global Sociology, Macmillan,
7 London.
8 Commission on Multi-Ethnic Britain (2000), The Future of Multi-
9 Ethnic Britain: The Parekh Report, Runnymede, London.
10 Costa-Lascoux, J. (1989), De l’immigré au citoyen, La documentation
11 française, Paris.
12 Cross, M. and Waldinger, R. (1997), ‘Economic Integration and
13 Labour Market Change: A Review and a Reappraisal’. Metropolis
14 Discussion Paper, www.international.metropolis.net
15 Diehl, C., Urbahn, J., and Esser, H. (1999), Die soziale und politische
16 Partizipation von Zuwandern in der Bundesrepublik Deutschland,
17 Forschungsinstitut der Friedrich-Ebert-Stiftung, Bonn.
18 Doomernik, J. (1998), ‘The Effectiveness of Integration Policies
19 Towards Immigrants and Their Descendants in France, Germany
20 and the Netherlands’, International Migration Papers, 27,
21 International Labour Organisation, Geneva.
22 Edmonston, B. and Passel, J.S. (eds), (1994), Immigration and
23 Ethnicity: The Integration of America’s Newest Arrivals, The Urban
24 Institute Press, Washington, DC.
25 Engbersen, G. (1996), ‘The Unknown City’, Berkeley Journal of
26 Sociology, 40, pp. 87-111.
27 Erikson, R. and Goldthorpe, J. (1992), The Constant Flux, Clarendon,
28 Oxford.
29 Esser, H. (1999), ‘Inklusion, Integration und thnische Schichtung’,
30 Journal flir Konflikt und Gewaltforschung, 1 (1), pp. 5-34.
31 Faist, T. (2000), The Volume and Dynamics of International Migration
32 and Transnational Social Spaces, Oxford UP, Oxford.
33 Favell, A. (1998), Philosophies of Integration: Immigration and the Idea
34 of Citizenship in France and Britain, Macmillan, London.
35 Favell, A. (2001), ‘Integration Policy and Integration Research in
36 Europe: A Review and Critique’ in A. Aleinikoff and D. Klusmeyer
37 (eds.) Citizenship Today: Global Perspectives and Practices,
38 Brookings Institute/Carnegie Endowment for International Peace,
39 Washington DC, pp. 249-99.
40 Fennema, M. and Tillie, J. (1999), ‘Political Participation and Political
41 Trust in Amsterdam: Civic Communities and Ethnic Networks’,
42 Journal of Ethnic and Migration Studies, 25 (4), pp. 703-26.
43 Fermin, A. (1999), ‘Inburgeringsbeleid en burgerschap’,
Migrantenstudien, 15 (2), pp. 99-112.

Fielding, A.J. (1995), ‘Migration and Social Change: A Longitudinal 1

Study of the Social Mobility of “Immigrants” in England and 2
Wales’, European Journal of Population, 11, pp. 107-21. 3
Forbes, I. and Mead, G. (1992), Measure For Measure: A Comparative 4
Analysis of Measure to Combat Racial Discrimination in the Member 5
Countries of the European Community, Employment Dept., 6
Sheffield. 7
Gellner, E. (1983), Nations and Nationalism, Blackwell, Oxford. 8
Giddens, A. (1990), The Consequences of Modernity, Polity Press, 9
Cambridge. 10
Gilroy, P. (1987), There Ain’t No Black in the Union Jack, Hutchinson, 11
London. 12
Gilroy, P. (2000), Between Camps: Nation, Culture and the Allure of 13
Race, Penguin, London. 14
Hall, S. (1988), ‘New ethnicities’, in J. Donald and A. Rattansi (eds.), 15
Race, Culture and Difference, Sage, London, pp. 252-9. 16
Hammar, T. (ed.) (1985), European Immigration Policy: A Comparative 17
Study, Cambridge UP, Cambridge. 18
Hansen, R. and Weil, P. (eds.) (2000), Towards a European Nationality: 19
Citizenship, Immigration and Nationality Law in the EU, Macmillan, 20
London. 21
Haut Conseil à I’Intégration (1993), L’intégration à la française, La 22
documentation française, Paris. 23
Heckmann, F. and Schnapper, D. (eds.) (2003), The Integration of 24
Immigrants in European Societies: National Differences and Trends of 25
Convergence, Lucius and Lucius, Stuttgart. 26
Ireland, P. (1994), The Policy Challenge of Ethnic Diversity, Harvard 27
UP, Cambridge, MA. 28
Joppke, C. (1999), Immigration and the Nation State: The United 29
States, Germany and Great Britain, Oxford UP, Oxford. 30
Kesteloot, C. (2000), ‘Segregation and Economic Integration of 31
Immigrants in Brussels’, in S. Body-Gendrot and M. Martiniello 32
(eds.), Minorities in European Cities: The Dynamics of Social 33
Integration and Social Exclusion at the Neighbourhood Level, 34
Macmillan, London. 35
Kloosterman, R., van der Leun, J., and Rath, J. (1998), ‘Across the 36
Border: Immigrants’ Economic Opportunities, Social Capital and 37
Informal Business Activities’, Journal of Ethnic and Migration 38
Studies, 24 (2), pp. 249-68. 39
Koopmans, R. and Statham, P. (eds.) (2000), Challenging Immigration 40
and Ethnic Relations Politics: Comparative European Perspectives, 41
Oxford UP, Oxford. 42
Kymlicka, W. (1995), Multicultural Citizenship, Oxford UP, Oxford. 43

402 adrian favell
1 Le Bras, H. (1998), Le démon des origins, L’edition de l’aube, Paris.

2 Lesthaege, R. (ed.) (2000), Communities and Generations: Turkish and
3 Moroccan Populations in Belgium, VUB Press, Brussels.
4 Mahnig, H. (1998), Integrationspolitik in Großbritannien, Frankreich,
5 Deutschland und den Niederlanden: Eine vergleichende Analyse,
6 Forschungsberichte des Schweizer Forums für Migrationstudien,
7 10, Neuchâtel.
8 Mann, M. (1993), The Sources of Social Power (2 vols), Cambridge
9 UP, Cambridge. Migration Policy Group (MPG) (1996), The
10 Comparative Approaches to Societal Integration Project, Brussels,
11 http://www.migpolgroup.com
12 Milward, A. (1992), The European Rescue of the Nation State,
13 Routledge, London.
14 Modood, T., Beishon, S., and Virdee, S. (1994), Changing Ethnic
15 Identities, Policy Studies Institute, London.
16 Modood, T. et al. (1997), Ethnic Minorities in Britain: Diversity and
17 Disadvantage, Policy Studies Institute, London.
18 Nauck, B. and Schönpflug, U. (eds.) (1997), Familien in verschiedenen
19 Kulturen, Enke, Stuttgart.
20 Noiriel, G. (1991), Le creuset français, Seuil, Paris.
21 Papastergiadis, N. (1999), The Turbulence of Migration: Globalization,
22 Deterritorialization and Hybridity, Polity Press, Cambridge.
23 Parekh, B. (2000), Rethinking Multiculturalism: Cultural Diversity and
24 Political Theory, Macmillan, London.
25 Phalet, K. and Swyngedouw, M. (1999), ‘Integratie ter discussie’, in
26 Swyngedouw et al. (eds.), Minderheden in Brussel, pp. 19-40.
27 Phalet, K., van Lotringen, C., and Entzinger, H. (2000), Islam in de
28 multiculturele samenleving, ERCOMER, Utrecht.
29 Phizacklea, A. (1998), ‘Migration and Globalization: A Feminist
30 Perspective’, in K. Koser and H. Lutz, The New Migration in Europe:
31 Social Constructions and Social Realities, Macmillan, London, pp.
32 21-38.
33 Portes, A. (ed.) (1995), The Economic Sociology of Immigration, Russell
34 Sage Foundation, New York.
35 Portes, A. (1996), ‘Transnational Communities: Their Emergence
36 and Their Significance in the Contemporary World-System’, in
37 R.P. Korzeniewicz and W.C. Smith (eds.), Latin America in the
38 World Economy, Greenwood Press, Westport, CT, pp.151-68.
39 Rath, J. (ed.) 2000, Immigrant Businesses: The Economic, Political and
40 Social Environment, Macmillan, London.
41 Rex, J. (1991), ‘The Political Sociology of a Multicultural Society’,
42 European Journal for Intercultural Studies, 2 (1).
43 Rose, E.J.B. et al. (1969), Colour and Citizenship: A Report on British

Race Relations, Oxford UP for the Institute of Race Relations, 1

London. 2
Schierup, C.-U. (1993), På Kulturens Slagmark: Mindretal og Størretal 3
om Danmark, Sydjysk Universitetsforlag, Esbjerg. 4
Schnapper, D. (1991), La France de l’intégration, Gallimard, Paris. 5
Simon, P. (1997), ‘La statistique des origines: l’ethnicité et la 6
‘race’ dans les recensements aux-Etats-Unis, Canada et Grande- 7
Brétagne’, Sociétés Contemporaines, 26, pp. 11-44. 8
Smith, M.P. and Guarnizo, L. (eds) (1998), Transnationalism from 9
Below, Transaction publishers, New Brunswick, NJ. 10
Soininen, M. (1999), ‘The “Swedish Model” as an Institutional 11
Framework for Immigrant Membership Rights’, Journal of Ethnic 12
and Migration Studies, 25 (2), pp. 685-702. 13
SOPEMI (1998), Trends in International Migration. Annual report of 14
continuous reporting system on migration, OECD, Rome. 15
Soysal, Y. (1994), Limits of Citizenship: Migrants and Post-National 16
Membership in Europe, Chicago UP, Chicago. 17
Swyngedouw, M., Phalet, K., and Deschouwer, K. (eds.) (1999), 18
Minderheden in Brussel, VUB Press, Brussels. 19
Todd, E. (1994), Le destin des immigrés: assimilation et ségrégation dans 20
les démocraties occidentales, Seuil, Paris. 21
Togeby, L. (1999), ‘Migrants at the Polls: An Analysis of Immigrant 22
and Refugee Participation in Danish Local Elections’, Journal of 23
Ethnic and Migration Studies, 25 (4), pp. 665-84. 24
Torpey, J. (2000), The Invention of the Passport, Cambridge UP, 25
Cambridge. 26
Tribalat, M. (1995), Faire France: une enquête sur les immigrés et leurs 27
enfants, La découverte, Paris. 28
Tribalat, M. et al. (1996), De l’immigration à l‘assimilation: une enquête 29
sur la population étrangère en France, INED, Paris. 30
Urry, J. (2000), Sociology Beyond Societies: Mobilities for the 21st 31
Century, Routledge, London. 32
Veenman, J. (1998), Buitenspel: Over langdurige werkloosheid onder 33
ethnische minderheden, Assen, Van Gorum. 34
Vermeulen, H. (ed.) (1997), Immigrant Policy for a Multicultural 35
Society: A Comparative Study of Integration, Language and Religious 36
Policy in Five Western European Countries, Migration Policy Group/ 37
IMES, Brussels. 38
Vertovec, S. (1997), ‘Social Cohesion and Tolerance’, in Key Issues for 39
Research and Policy on Migrants in Cities. Metropolis Discussion 40
Paper, www.intemational.metropolis.net 41
Waldrauch, H. and Hofinger, C. (1997), ‘An Index to Measure the 42
Legal Obstacles to the Integration of Migrants’, New Community, 43
23 (2), pp. 271-86.

404 adrian favell
1 Waldrauch, H. (ed.) (2001), Die Integration von Einwanderern: Ein

2 Index der Rechtlichen Diskriminierung, Campus, Frankfurt.
3 Weil, P. (1991), La France et ses étrangèrs: l’aventure d’une politique de
4 l’immigration, Calmann-Lévy, Paris.
5 Weil, P. (1996), ‘Nationalities and Citizenships: The Lessons of
6 the French Experience for Germany and Europe’, in D. Cesarani
7 and M. Fulbrook (eds.), Citizenship, Nationality and Migration in
8 Europe, Routledge, London, pp. 74-87.
9 Werbner, P. (1999), ‘What Colour Success? Distorting Value in
10 Studies of Ethnic Entrepreneurship’, Sociological Review, 47 (3),
11 pp. 548-79.
12 Wimmer, A. and Glick Schiller, N. (2002), ‘Methodological
13 Nationalism and Beyond: Nation-State Building, Migration and
14 the Social Sciences’, Global Networks, 2 (4), pp. 301-34.
15 Wrench, J. (1996), Preventing Racism at the Workplace, European
16 Foundation for the Improvement of Living and Working
17 Conditions, Dublin.
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
Part III 4
5
6
7
Conceptual issues 8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
marge tekst 36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Introduction to ethnic groups and boundaries: 3
4
the social organization of cultural difference
5
6
Fredrik Barth 7
8
9
10
11
12
In 1969, the anthropologist Fredrik Barth published a collection of ground-
13
breaking essays entitled Ethnic Groups and Boundaries: The Social Organization
14
of Cultural Difference. This collection criticised the orthodoxy of the time that 15
conceived of ethnic groups as tribes or people who are able to maintain their 16
individual cultural traits despite the ignorance of their neighbours. It is in 17
geographical and social isolation that one can find ethnic groups in their pu- 18
rity. Barth argued convincingly against the suggestion that splendid isolation 19
is the critical factor in sustaining cultural diversity. He took the innovative po- 20
sition that ethnic identity is basically a social identity that emerges in interac- 21
tion with others. Ethnic groups are categories of ascription and identification 22
taken on by the actors themselves, having the capacity to organise interac- 23
tion between people. To observe these processes the focus of investigation 24
should be shifted from separate groups’ internal constitutions and histories 25
to ethnic boundaries and boundary maintenance. 26
27
28
This collection of essays addresses itself in the problems of ethnic 29
groups and their persistence. This is a theme of great, but neglected, 30
importance to social anthropology. Practically all anthropological rea- 31
soning rests on the premise that cultural variation is discontinuous: 32
that there are aggregates of people who essentially share a common 33
culture, and interconnected differences that distinguish each such 34
discrete culture from all others. Since culture is nothing but a way 35
to describe human behaviour, it would follow that there are discrete 36
groups of people, i.e. ethnic units, to correspond to each culture. The 37
differences between cultures, and their historic boundaries and con- ethnic group 38
nections, have been given much attention; the constitution of eth- boundaries 39
nic groups, and the nature of the boundaries between them, have 40
not been correspondingly investigated. Social anthropologists have 41
largely avoided these problems by using a highly abstracted concept 42
of ‘society’ to represent the encompassing social system within which 43

408 fredrik barth
1 smaller, concrete groups and units may be analysed. But this leaves
2 untouched the empirical characteristics and boundaries of ethnic
3 groups, and the important theoretical issues which an investigation
4 of them raises.
5 Though the naïve assumption that each tribe and people has
6 maintained its culture through a bellicose ignorance of its neigh-
7 bours is no longer entertained, the simplistic view that geographical
8 and social isolation have been the critical factors in sustaining cul-
9 tural diversity persists. An empirical investigation of the character of
10 ethnic boundaries, as documented in the following essays, produces
11 two discoveries which are hardly unexpected, but which demonstrate
12 the inadequacy of this view. First, it is clear that boundaries persist
13 despite a flow of personnel across them. In other words, categorical
14 ethnic distinctions do not depend on an absence of mobility, con-
15 tact and information, but do entail social processes of exclusion and
16 incorporation whereby discrete categories are maintained despite
17 changing participation and membership in the course of individual
18 life histories. Secondly, one finds that stable, persisting, and often
19 vitality important social relations are maintained across such bound-
20 aries, and are frequently based precisely on the dichotomized ethnic
21 statuses. In other words, ethnic distinctions do not depend on an
22 absence of social interaction and acceptance, but are quite to the con-
23 trary often the very foundations on which embracing social systems
24 are built. Interaction in such a social system does not lead to its liq-
25 uidation through change and acculturation; cultural differences can
26 persist despite inter-ethnic contact and interdependence.
27
28
29 General approach
30
31 There is clearly an important field here in need of rethinking. What
32 is required is a combined theoretical and empirical attack: we need to
33 investigate closely the empirical facts of a variety of cases, and fit our
34 concepts to these empirical facts so that they elucidate them as sim-
35 ply and adequately as possible, and allow us to explore their implica-
36 tions. In the following essays, each author takes up a case with which
37 he is intimately familiar from his own fieldwork, and tries to apply a
38 common set of concepts to its analysis. The main theoretical depar-
39 ture consists of several interconnected parts. First, we give primary
40 emphasis to the fact that ethnic groups are categories of ascription
41 and identification by the actors themselves, and thus have the char-
42 acteristic of organizing interaction between people. We attempt to
43 relate other characteristics of ethnic groups to this primary feature.

part iii – conceptual issues 409
Second, the essays all apply a generative viewpoint to the analysis: 1

rather than working through a typology of forms of ethnic groups 2
and relations, we attempt to explore the different processes that seem 3
to be involved in generating and maintaining ethnic groups. Third, 4
to observe these processes we shift the focus of investigation from 5
internal constitution and history of separate groups to ethnic bound- 6
aries and boundary maintenance. Each of these points needs some 7
elaboration. 8
9
10
Ethnic group defined 11
12
The term ethnic group is generally understood in anthropological 13
literature (cf. e.g. Narroll 1964) to designate a population which: 14
1. is largely biologically self-perpetuating 15
2. shares fundamental cultural values, realized in overt unity in cul- 16
tural forms 17
3. makes up a field of communication and interaction 18
4. has a membership which identifies itself, and is identified by oth- 19
ers, as constituting a category distinguishable from other catego- 20
ries of the same order. 21
This ideal type definition is not so far removed in content from the ideal type model 22
traditional proposition that a race = a culture = a language and that a 23
society = a unit which rejects or discriminates against others. Yet, in 24
its modified form it is close enough to many empirical ethnographic 25
situations, at least as they appear and have been reported, so that this 26
meaning continues to serve the purposes of most anthropologists. 27
My quarrel is not so much with the substance of these characteris- 28
tics, though as I shall show we can profit from a certain change of 29
emphasis; my main objection is that such a formulation prevents 30
us from understanding the phenomenon of ethnic groups and their 31
place in human society and culture. This is because it begs all the 32
critical questions: while purporting to give an ideal type model of a 33
recurring empirical form, it implies a preconceived view of what are 34
the significant factors in the genesis, structure, and fundion of such 35
groups. 36
Most critically, it allows us to assume that boundary maintenance 37
is unproblematical and follows from the isolation which the itemized 38
characteristics imply: racial difference, cultural difference, social sep- 39
aration and language barriers, spontaneous and organized enmity. 40
This also limits the range of factors that we use to explain cultural di- 41
versity: we are led to imagine each group developing its cultural and 42
social form in relative isolation, mainly in response to local ecologic 43

410 fredrik barth
1 factors, through a history of adaptation by invention and selective

2 borrowing. This history has produced a world of separate peoples,
3 each with their culture and each organized in a society which can
4 legitimately be isolated for description as an island to itself.
5
6
7 Ethnic groups as culture-bearing units
8
9 Rather than discussing the adequacy of this version of culture history
10 for other than pelagic islands, let us look at some of the logical flaws
11 in the viewpoint. Among the characteristics listed above, the sharing
12 of a common culture is generally given central importance. In my
13 view, much can be gained by regarding this very important feature
14 implications as an implication or result, rather than a primary and definitional
15 characteristic of ethnic group organization. If one chooses to regard
16 the culture-bearing aspect of ethnic groups as their primary charac-
17 teristic, this has far-reaching implications. One is led to identify and
18 distinguish ethnic groups by the morphological characteristics of the
19 cultures of which they are the bearers. This entails a prejudged view-
20 point both on (1) the nature of continuity in time of such units, and
21 (2) the locus of the factors which determine the form of the units.
22 1. Given the emphasis on the culture-bearing aspect, the classifica-
23 tion of persons and local groups as members of an ethnic group
24 must depend on their exhibiting the particular traits of the cul-
25 ture. This is something that can be judged objectively by the eth-
26 nographic observer, in the culture-area tradition, regardless of
27 the categories and prejudices of the actors. Differences between
28 groups become differences in trait inventories; the attention is
29 drawn to the analysis of cultures, not of ethnic organization.
30 The dynamic relationship between groups will then be depicted
31 in acculturation studies of the kind that have been attracting de-
32 creasing interest in anthropology, though their theoretical inad-
33 equacies have never been seriously discussed. Since the histori-
34 cal provenance of any assemblage of culture traits is diverse, the
35 viewpoint also gives scope for an ‘ethnohistory’ which chronicles
36 cultural accretion and change, and seeks to explain why certain
37 items were borrowed. However, what is the unit whose continuity
38 in time is depicted in such studies? Paradoxically, it must include
39 cultures in the past which would dearly be excluded in the pres-
40 ent because of differences in form – differences of precisely the
41 kind that are diagnostic in synchronic differentiation of ethnic
42 units. The interconnection between ‘ethnic group’ and ‘culture’ is
43 certainly not clarified through this confusion.

2. The overt cultural forms which can be itemized as traits exhibit 1

the effects of ecology. By this I do not mean to refer to the fact 2
that they reflect a history of adaptation to environment; in a more 3
immediate way they also reflect the external circumstances to 4
which actors must accommodate themselves. The same group of 5
people, with unchanged values and ideas, would surely pursue 6
different patterns of life and institutionalize different forms of 7
behaviour when faced with the different opportunities offered in 8
different environments? Likewise, we must expect to find that one 9
ethnic group, spread over a territory with varying ecologic circum- 10
stances, will exhibit regional diversities of overt institutionalized 11
behaviour which do not reflect differences in cultural orientation. 12
How should they then be classified if overt institutional forms 13
are diagnostic? A case in point is the distributions and diversity 14
of Pathan local social systems, discussed below (pp, 117 ff.). By Pathan local 15
basic Pathan values, a Southern Pathan from the homogeneous, social systems 16
lineage-organized mountain areas, can only find the behaviour of 17
Pathans in Swat so different from, and reprehensible in terms of, 18
their own values that they declare their northern brothers ‘no lon- 19
ger Pathan’. Indeed, by ‘objective’ criteria, their overt pattern of 20
organization seems much closer to that of Panjabis. But I found 21
it possible, by explaining the circumstances in the north, to make 22
Southern Pathans agree that these were indeed Pathans too, and 23
grudgingly to admit that under those circumstances they might 24
indeed themselves act in the same way. It is thus inadequate to 25
regard overt institutional forms as constituting the cultural fea- 26
tures which at any time distinguish an ethnic group – these overt 27
forms are determined by ecology as well as by transmitted cul- 28
ture. Nor can it be claimed that every such diversification within 29
a group represents a first step in the direction of subdivision and 30
multiplication of units. We have wellknown documented cases 31
of one ethnic group, also at a relatively simple level of economic 32
organization, occupying several different ecologic niches and yet 33
retaining basic cultural and ethnic unity over long periods [cf., 34
e.g., inland and coastal Chuckchee (Bogoras 1904-9) or reindeer, 35
river, and coast Lapps (Gjessing, 1954]. 36
In one of the following essays, Bjorn (pp. 74 ff.) argues cogently on 37
this point with reference to central Norwegian mountain farmers. 38
He shows how their participation and self-evaluation in terms of 39
general Norwegian values secures them continued membership in 40
the larger ethnic group, despite the highly characteristic and devi- 41
ant patterns of activity which the local ecology imposes on them. 42
To analyse such cases, we need a viewpoint that does not confuse 43

412 fredrik barth
1 the effects of ecologic circumstances on behaviour with those of cul-

2 tural tradition, but which makes it possible to separate these factors
3 and investigate the non-ecological cultural and social components
4 creating diversity.
5
6
7 Ethnic groups as an organizational type
8
9 By concentrating on what is socially effective, ethnic groups are seen
10 as a form of social organization. The critical feature then becomes
11 item (4) in the list on p. 409 the characteristic of self-ascription and
12 ethnic ascription ascription by others. A categorical ascription is an ethnic ascription
13 when it classifies a person in terms of his basic, most general iden-
14 tity, presumptively determined by his origin and background. To the
15 extent that actors use ethnic identities to categorize themselves and
16 others for purposes of interaction, they form ethnic groups in this
17 organizational sense.
18 It is important to recognize that although ethnic categories take
19 cultural differences into account, we can assume no simple one-to-
20 one relationship between ethnic units and cultural similarities and
21 differences. The features that are taken into account are not the sum
22 of ‘objective’ differences, but only those which the actors themselves
23 regard as significant. Not only do ecologic variations mark and exag-
24 gerate differences; some cultural features are used by the actors as
25 signals and emblems of differences, others are ignored, and in some
26 relationships radical differences are played down and denied. The
27 cultural contents of ethnic dichotomies would seem analytically to
28 be of two orders: (i) overt signals or signs – the diacritical features
29 that people look for and exhibit to show identity, often such features
30 as dress, language, house-form, or general style of life, and (ii) basic
31 value orientations: the standards of morality and excellence by which
32 performance is judged. Since belonging to an ethnic category im-
33 plies being a certain kind of person, having that basic identity, it also
34 implies a claim to be judged, and to judge oneself, by those standards
35 that are relevant to that identity. Neither of these kinds of cultural
36 ‘contents’ follows from a descriptive list of cultural features or cul-
37 tural differences; one cannot predict from first principles which fea-
38 tures will be emphasized and made organizationally relevant by the
39 actors. In other words, ethnic categories provide an organizational
40 vessel that may be given varying amounts and forms of content in
41 different socio-cultural systems. They may be of great relevance to
42 behaviour, but they need not be; they may pervade all social life, or
43 they may be relevant only in limited sectors of activity. There is thus

an obvious scope for ethnographic and comparative descriptions of 1

different forms of ethnic organization. 2
The emphasis on ascription as the critical feature of ethnic groups 3
also solves the two conceptual difficulties that were discussed above. 4
1. When defined as an ascriptive and exclusive group, the nature of 5
continuity of ethnic units is clear: it depends on the maintenance 6
of a boundary. The cultural features that signal the boundary 7
may change, and the cultural characteristics of the members may 8
likewise be transformed, indeed, even the organizational form of 9
the group may change – yet the fact of continuing dichotomi- 10
zation between members and outsiders allows us to specify the 11
nature of continuity, and investigate the changing cultural form 12
and content. 13
2. Socially relevant factors alone become diagnostic for member- 14
ship, not the overt, ‘objective’ differences which are generated 15
by other factors. It makes no difference how dissimilar members 16
may be in their overt behaviour – if they say they are A, in contrast 17
to another cognate category B, they are willing to be treated and 18
let their own behaviour be interpreted and judged as A’s and not 19
as B’s; in other words, they declare their allegiance to the shared 20
culture of A’s, The effects of this, as compared to other factors 21
influencing actual behaviour, can then be made the object of in- 22
vestigation. 23
24
25
The boundaries of ethnic groups 26
27
The critical focus of investigation from this point of view becomes 28
the ethnic boundary that defines the group, not the cultural stuff that 29
it encloses. The boundaries to which we must give our attention are 30
of course social boundaries, though they may have territorial coun- social boundaries 31
terparts. If a group maintains its identity when members interact 32
with others, this entails criteria for determining membership and 33
ways of signalling membership and exclusion. Ethnic groups are not 34
merely or necessarily based on the occupation of exclusive territories; 35
and the different ways in which they are maintained, not only by a 36
once-and-for-all recruitment but by continual expression and valida- 37
tion, need to be analysed. 38
What is more, the ethnic boundary canalizes social life – it entails 39
a frequently quite complex organization of behaviour and social rela- 40
tions. The identification of another person as a fellow member of an 41
ethnic group implies a sharing of criteria for evaluation and judge- 42
ment. It thus entails the assumption that the two are fundamentally 43

414 fredrik barth
1 ‘playing the same game’, and this means that there is between them
2 a potential for diversification and expansion of their social relation-
3 ship to cover eventually all different sectors and domains of activity.
4 On the other hand, a dichotomization of others as strangers, as mem-
5 bers of another ethnic group, implies a recognition of limitations on
6 shared understandings, differences in criteria for judgement of value
7 and performance, and a restriction of interaction to sectors of as-
8 sumed common understanding and mutual interest.
9 This makes it possible to understand one final form of boundary
10 maintenance whereby cultural units and boundaries persist. Entailed
11 in ethnic boundary maintenance are also situations of social contact
12 between persons of different cultures: ethnic groups only persist as
13 significant units if they imply marked difference in behaviour, i.e.
14 persisting cultural persisting cultural differences. Yet where persons of different culture
15 differences interact, one would expect these differences to be reduced, since in-
16 teraction both requires and generates a congruence of codes and val-
17 ues – in other words, a similarity or community of culture (cf. Barth
18 1966, for my argumentation on this point). Thus the persistence
19 of ethnic groups in contact implies not only criteria and signals for
20 identification, but also a structuring of interaction which allows the
21 persistence of cultural differences. The organizational feature which,
22 I would argue, must be general for all inter-ethnic relations is a sys-
23 tematic set of rules governing inter-ethnic social encounters. In an
24 organized social life, what can be made relevant to interaction in any
25 particular social situation is prescribed (Goffman 1959). If people
26 agree about these prescriptions, their agreement on codes and values
27 need not extend beyond that which is relevant to the social situa-
28 tions in which they interact. Stable inter-ethnic relations presuppose
29 such a structuring of interaction: a set of prescriptions governing
30 situations of contact, and allowing for articulation in some sectors
31 or domains of activity, and a set of proscriptions on social situations
32 preventing inter-ethnic interaction in other sectors, and thus insulat-
33 ing parts of the cultures from confrontation and modification.
34
35
36 Poly-ethnic social systems
37
38 This of course is what Furnivall (1944) so dearly depicted in his anal-
39 ysis of plural society: a poly-ethnic society integrated in the market
40 place, under the control of a state system dominated by one of the
41 groups, but leaving large areas of cultural diversity in the religious
42 and domestic sectors of activity.
43 What has not been adequately appreciated by later anthropolo-

gists is the possible variety of sectors of articulation and separation, 1

and the variety of poly-ethnic systems which this entails. We know variety of 2
of some of the Melanesian trade systems in objects belonging to the poly-ethnic systems 3
highprestige sphere of the economy, and even some of the etiquette 4
and prescriptions governing the exchange situation and insulating 5
it from other activities. We have information on various traditional 6
polycentric systems from S.E. Asia (discussed below, Izikowitzs, pp. 7
135 ff.) integrated both in the prestige trade sphere and in quasi-feu- 8
dal political structures. Some regions of S.W. Asia show forms based 9
on a more fully monetized market economy, while political integra- 10
tion is polycentric in character. There is also the ritual and productive 11
cooperation and political integration of the Indian caste system to be 12
considered, where perhaps only kinship and domestic life remain as 13
a proscribed sector and a wellspring for cultural diversity. Nothing 14
can be gained by lumping these various systems under the increas- 15
ingly vague label of ‘plural’ society, whereas an investigation of the 16
varieties of structure can shed a great deal of light on social and cul- 17
tural forms. 18
What can be referred to as articulation and separation on the 19
macro-level corresponds to systematic sets of role constraints on the 20
micro-level. Common to all these systems is the principle that ethnic 21
identity implies a series of constraints on the kinds of roles an indi- 22
vidual is allowed to play, and the partners he may choose for differ- 23
ent kinds of transactions.1 In other words, regarded as a status, eth- 24
nic identity is superordinate to most other statuses, and defines the 25
permissible constellations of statuses, or social personalities, which 26
an individual with that identity may assume. In this respect ethnic 27
identity is similar to sex and rank, in that it constrains the incumbent 28
in all his activities, not only in some defined social situations.2 One 29
might thus also say that it is imperative, in that it cannot be disre- 30
garded and temporarily set aside by other definitions of the situa- 31
tion. The constraints on a person’s behaviour which spring from his 32
ethnic identity thus tend to be absolute and, in complex poly-ethnic 33
societies, quite comprehensive; and the component moral and social 34
conventions are made further resistant to change by being joined in 35
stereotyped clusters as characteristics of one single identity. 36
37
38
The associations of identities and value standards 39
40
The analysis of interactional and organizational features of inter- 41
ethnic relations has suffered from a lack of attention to problems 42
of boundary maintenance. This is perhaps because anthropologists 43

416 fredrik barth
1 have reasoned from a misleading idea of the prototype inter-ethnic

2 situation. One has tended to think in terms of different peoples, with
3 different histories and cultures, coming together and accommodat-
4 ing themselves to each other, generally in a colonial setting. To visu-
5 basic requirements alize the basic requirements for the coexistence of ethnic diversity, I
6 would suggest that we rather ask ourselves what is needed to make
7 ethnic distinctions emerge in an area. The organizational require-
8 ments are clearly, first, a categorization of population sectors in ex-
9 clusive and imperative status categories, and second, an acceptance
10 of the principle that standards applied to one such category can be
11 different from that applied to another. Though this alone does not
12 explain why cultural differences emerge, it does allow us to see how
13 they persist. Each category can then be associated with a separate
14 range of value standards. The greater the differences between these
15 value orientations are, the more constraints on inter-ethnic interac-
16 tion do they entail: the statuses and situations in the total social sys-
17 tem involving behaviour which is discrepant with a person’s value
18 orientations must be avoided, since such behaviour on his part will
19 be negatively sanctioned. Moreover, because identities are signalled
20 as well as embraced, new forms of behaviour will tend to be dichoto-
21 mized: one would expect the role constraints to operate in such a way
22 that persons would be reluctant to act in new ways from a fear that
23 such behaviour might be inappropriate for a person of their iden-
24 tity, and swift to classify forms of activity as associated with one or
25 another cluster of ethnic characteristics. Just as dichotomizations of
26 male versus female work seem to proliferate in some societies, so
27 also the existence of basic ethnic categories would seem to be a factor
28 encouraging the proliferation of cultural differentiae.
29 In such systems, the sanctions producing adherence to group-
30 specific values are not only exercised by those who share the identity.
31 Again, other imperative statuses afford a parallel: just as both sexes
32 ridicule the male who is feminine, and all classes punish the pro-
33 letarian who puts on airs, so also can members of an ethnic group
34 in a poly-ethnic society act to maintain dichotomies and differences.
35 Where social identities are organized and allocated by such princi-
36 ples, there will thus be a tendency towards canalization and stan-
37 dardization of interaction and the emergence of boundaries which
38 maintain and generate ethnic diversity within larger, encompassing
39 social systems.
40
41
42
43

Interdependence of ethnic groups 1

2
The positive bond that connects several ethnic groups in an encom- 3
passing social system depends on the complementarity of the groups complementarity 4
with respect to some of their characteristic cultural features. Such 5
complementarity can give rise to interdependence or symbiosis, and 6
constitutes the areas of articulation referred to above; while in the 7
fields where there is no complementarity there can be no basis for 8
organization on ethnic lines – there will either be no interaction, or 9
interaction without reference to ethnic identity. 10
Social systems differ greatly in the extent to which ethnic iden- 11
tity, as an imperative status, constrains the person in the variety of 12
statuses and roles he may assume. Where the distinguishing values 13
connected with ethnic identity are relevant only to a few kinds of 14
activities, the social organization based on it will be similarly lim- 15
ited. Complex polyethnic systems, on the other hand, clearly entail 16
the existence of extensively relevant value differences and multiple 17
constraints on status combinations and social participation. In such 18
systems, the boundary maintaining mechanisms must be highly 19
effective, for the following reasons: (i) the complexity is based on 20
the existence of important, complementary cultural differences; (ii) 21
these differences must be generally standardized within the ethnic 22
group – i.e. the status cluster, or social person, of every member of 23
a group must be highly stereotyped – so that inter-ethnic interaction 24
can be based on ethnic identities; and (iii) the cultural characteristics 25
of each ethnic group must be stable, so that the complementary dif- 26
ferences on which the systems rest can persist in the face of close 27
inter-ethnic contact. Where these conditions obtain, ethnic groups 28
can make stable and symbiotic adaptations to each other: other eth- 29
nic groups in the region become a part of the natural environment; 30
the sectors of articulation provide areas that can be exploited, while 31
the other sectors of activity of other groups are largely irrelevant from 32
the point of view of members of anyone group. 33
34
35
Ecologic perspective 36
37
Such interdependences can partly be analysed from the point of view 38
of cultural ecology, and the sectors of activity where other popula- 39
tions with other cultures articulate may be thought of as niches to 40
which the group is adapted. This ecologic interdependence may take ecologic 41
several different forms, for which one may construct a rough typol- interdependence 42
ogy. Where two or more ethnic groups are in contact, their adapta- 43
tions may entail the following forms:

418 fredrik barth
1 (1) They may occupy clearly distinct niches in the natural environ-
2 ment and be in minimal competition for resources. In this case
3 their interdependence will be limited despite co-residence in the
4 area, and the articulation will tend to be mainly through trade,
5 and perhaps in a ceremonial-ritual sector.
6 (2) They may monopolize separate territories, in which case they are
7 in competition for resources and their articulation will involve
8 politics along the border, and possibly other sectors.
9 (3) They may provide important goods and services for each other,
10 i.e. occupy reciprocal and therefore different niches but in close
11 interdependence. If they do not articulate very closely in the po-
12 litical sector, this entails a classical symbiotic situation and a va-
13 riety of possible fields of articulation. If they also compete and
14 accommodate through differential monopolization of the means
15 of production, this entails a close political and economic articula-
16 tion, with open possibilities for other forms of interdependence
17 as well.
18 These alternatives refer to stable situations. But very commonly, one
19 will also find a fourth main form: where two or more interspersed
20 groups are in fact in at least partial competition within the same
21 niche. With time one would expect one such group to displace the
22 other, or an accommodation involving an increasing complementar-
23 ity and interdependence to develop.
24 From the anthropological literature one can doubtless think of
25 type cases for most of these situations. However, if one looks care-
26 mixed situations fully at most empirical cases, one will find fairly mixed situations
27 obtaining, and only quite gross simplifications can reduce them to
28 simple types. I have tried elsewhere (Barth 1964b) to illustrate this
29 for an area of Baluchistan, and expect that it is generally true that an
30 ethnic group, on the different boundaries of its distribution and in
31 its different accommodations, exhibits several of these forms in its
32 relations to other groups.
33
34
35 Demographic perspective
36
37 These variables, however, only go part of the way in describing the
38 adaptation of a group. While showing the qualitative (and ideally
39 quantitative) structure of the niches occupied by a group, one can-
40 not ignore the problems of number and balance in its adaptation.
41 Whenever a population is dependent on its exploitation of a niche
42 in nature, this implies an upper limit on the size it may attain corre-
43 sponding to the carrying capacity of that niche; and any stable adapta-

tion entails a control on population size. If, on the other hand, two 1
populations are ecologically interdependent, as two ethnic groups in 2
a symbiotic relationship, this means that any variation in the size of 3
one must have important effects on the other. In the analysis of any 4
poly-ethnic system for which we assert any degree of time depth, we 5
must therefore be able to explain the processes whereby the sizes 6
of the interdependent ethnic groups are balanced. The demographic 7
balances involved are thus quite complex, since a group’s adaptation 8
to a niche in nature is affected by its absolute size, while a group’s 9
adaptation to a niche constituted by another ethnic group is affected 10
by its relative size. 11
The demographic problems in an analysis of ethnic inter-relations 12
in a region thus centre on the forms of recruitment to ethnic groups recruitment 13
and the question of how, if at all, their rates are sensitive to pressures 14
on the different niches which each group exploits. These factors are 15
highly critical for the stability of any poly-ethnic system, and it might 16
look as if any population change would prove destructive. This does 17
not necessarily seem to follow, as documented e.g. in the essay by 18
Siverts (pp. 101 ff.), but in most situations the poly-ethnic systems we 19
observe do entail quite complex processes of population movement 20
and adjustment. It becomes clear that a number of factors other than 21
human fertility and mortality affect the balance of numbers. From 22
the point of view of any one territory, there are the factors of individ- 23
ual and group movements: emigration that relieves pressure, immi- 24
gration that maintains one or several co-resident groups as outpost 25
settlements of larger population reservoirs elsewhere. Migration and 26
conquest play an intermittent role in redistributing populations and 27
changing their relations. But the most interesting and often critical 28
role is played by another set of processes that effect changes of the 29
identity of individuals and groups. After all, the human material that 30
is organized in an ethnic group is not immutable, and though the 31
social mechanisms discussed so far tend to maintain dichotomies 32
and boundaries, they do not imply ‘stasis’ for the human material 33
they organize: boundaries may persist despite what may figuratively 34
be called the ‘osmosis’ of personnel through them. 35
This perspective leads to an important clarification of the condi- 36
tions for complex poly-ethnic systems. Though the emergence and 37
persistence of such systems would seem to depend on a relatively 38
high stability in the cultural features associated with ethnic groups 39
– i.e. a high degree or rigidity in the interactional boundaries – they 40
do not imply a similar rigidity in the patterns of recruitment or as- 41
cription to ethnic groups: on the contrary, the ethnic inter-relations 42
that we observe frequently entail a variety of processes which effect 43

420 fredrik barth
1 changes in individual and group identity and modify the other demo-
2 graphic factors that obtain in the situation. Examples of stable and
3 persisting ethnic boundaries that are crossed by a flow of personnel
4 are clearly far more common than the ethnographic literature would
5 lead us to believe. Different processes of such crossing are exempli-
6 fied in these essays, and the conditions which cause them are shown
7 to be various. We may look briefly at some of them.
8
9
10 Factors in identity change
11
12 The Yao described by Kandre (1967b) are one of the many hill peoples
13 Yao assimilation on the southern fringe of the Chinese area. The Yao are organized for
14 productive purposes in extended family households, aligned in clans
15 and in villages. Household leadership is very clear, while commu-
16 nity and region are autochthonously acephalous, and variously tied to
17 poly-ethnic political domains. Identity and distinctions are expressed
18 in complex ritual idioms, prominently involving ancestor worship.
19 Yet this group shows the drastic incorporation rate of 10% non-Yao
20 becoming Yao in each generation (Kandre 1967a: 594). Change of
21 membership takes place individually, mostly with children, where
22 it involves purchase of the person by a Yao houseleader, adoption to
23 kinship status, and full ritual assimilation. Occasionally, change of
24 ethnic membership is also achieved by men through uxorilocal mar-
25 riage; Chinese men are the acceptable parties to such arrangements.
26 The conditions for this form of assimilation are clearly twofold:
27 first, the presence of cultural mechanisms to implement the incorpo-
28 ration, including ideas of obligations to ancestors, compensation by
29 payment, etc., and secondly, the incentive of obvious advantages to
30 the assimilating household and leader. These have to do with the role
31 of households as productive units and agro-managerial techniques
32 that imply an optimal size of 6-8 working persons, and the pattern of
33 intra-community competition between household leaders in the field
34 of wealth and influence.
35 Movements across the southern and northern boundaries of the
36 Pathan area (cf. pp. 123 ff.) illustrate quite other forms and condi-
37 tions. Southern Pathans become Baluch and not vice versa; this
38 transformation can take place with individuals but more readily with
39 whole households or small groups of households: it involves loss of
40 position in the rigid genealogical and territorial segmentary system
41 of Pathans and incorporation through clientage contract into the hi-
42 erarchical, centralized system of the Baluch. Acceptance in the re-
43 ceiving group is conditional on the ambition and opportunism of

Baluch political leaders. On the other hand, Pathans in the north 1

have, after an analogous loss of position in their native system, set- 2
tled in and often conquered new territories in Kohistan. The effect 3
in due course has been a reclassification of the settling communities 4
among the congeries of locally diverse Kohistani tribes and groups. 5
Perhaps the most striking case is that from Darfur provided by 6
Haaland (pp, 58 ff.), which shows members of the hoe-agricultural 7
Fur of the Sudan changing their identity to that of nomadic cattle the Fur case 8
Arabs. This process is conditional on a very specific economic cir- 9
cumstance: the absence of investment opportunities (or capital in the 10
village economy of the Fur in contrast to the possibilities among the 11
nomads. Accumulated capital, and the opportunities for its manage- 12
ment and increase, provide the incentive for Fur households to aban- 13
don their fields and villages and change to the life of the neighbour- 14
ing Baggara, incidentally also joining one of the loose but nominally 15
centralized Baggara political units if the change has been economi- 16
cally completely successful. 17
These processes that induce a flow of personnel across ethnic 18
boundaries will of necessity affect the demographic balance between 19
different ethnic groups. Whether they are such that they contribute 20
to stability in this balance is an entirely different question. To do so, 21
they would have to be sensitive to changes in the pressure on eco- 22
logic niches in a feed-back pattern. This does not regularly seem to 23
be the case. The assimilation of non-Yao seems further to increase 24
the rate of Yao growth and expansion at the expense of other groups, 25
and can be recognized as one, albeit minor, factor furthering the 26
progressive Sinization process whereby cultural and ethnic diversity 27
has steadily been reduced over vast areas. The rate of assimilation of 28
Pathans by Baluch tribes is no doubt sensitive to population pressure 29
in Pathan areas, but simultaneously sustains an imbalance whereby 30
Baluch tribes spread northward despite higher population pressures 31
in the northern areas. Kohistani assimilation relieves population 32
pressure in Pathan area while maintaining a geographically stable 33
boundary. Nomadization of the Fur replenishes the Baggara, who are 34
elsewhere becoming sedentarized. The rate, however, does not corre- 35
late with pressure on Fur lands – since nomadization is conditional 36
on accumulated wealth, its rate probably decreases as Fur popula- 37
tion pressure increases. The Fur case also demonstrates the inherent 38
instability of some of these processes. and how limited changes can 39
have drastic results: with the agricultural innovation of orchards over 40
the last ten years, new investment opportunities are provided which 41
will probably greatly reduce, or perhaps for a while even reverse, the 42
nomadization process. 43

422 fredrik barth
1 Thus, though the processes that induce change of identity are

2 important to the understanding of most cases of ethnic interdepen-
3 dence, they need not be conducive to population stability. In general,
4 however, one can argue that whenever ethnic relations are stable
5 over long periods, and particularly where the interdependence is
6 close, one can expect to find an approximate demographic balance.
7 The analysis of the different factors involved in this balance is an
8 important part of the analysis of the ethnic inter-relations in the area.
9
10
11 The persistence of cultural boundaries
12
13 In the preceding discussion of ethnic boundary maintenance and
14 interchange of personnel there is one very important problem that
15 I have left aside. We have seen various examples of how individuals
16 and small groups, because of specific economic and political circum-
17 stances in their former position and among the assimilating group,
18 may change their locality, their subsistence pattern, their political al-
19 legiance and form, or their household membership. This still does
20 not fully explain why such changes lead to categorical changes of
21 ethnic identity, leaving the dichotomized ethnic groups unaffected
22 (other than in numbers) by the interchange of personnel. In the case
23 of adoption and incorporation of mostly immature and in any case
24 isolated single individuals into pre-established households, as among
25 the Yao, such complete cultural assimilation is understandable: here
26 every new person becomes totally immersed in a Yao pattern of rela-
27 tionships and expectations. In the other examples, it is less clear why
28 this total change of identity takes place. One cannot argue that it fol-
29 lows from a universally imputable rule of cultural integration, so that
30 the practice of the politics of one group or the assumption of its pat-
31 tern of ecologic adaptation in subsistence and economy, entails the
32 the Pathan case adoption also of its other parts and forms. Indeed, the Pathan case
33 (Ferdinand 1967) directly falsifies this argument, in that the bound-
34 aries of the Pathan ethnic group crosscuts ecologic and political
35 units. Using self-identification as the critical criterion of ethnic iden-
36 tity, it should thus be perfectly possible for a small group of Pathans
37 to assume the political obligations of membership in a Baluch tribe,
38 or the agricultural and husbandry practices of Kohistanis, and yet
39 continue to call themselves Pathans. By the same token one might
40 expect nomadization among the Fur to lead to the emergence of a
41 nomadic section of the Fur, similar in subsistence to the Baggara but
42 different from them in other cultural features, and in ethnic label.
43 Quite clearly, this is precisely what has happened in many his-

torical situations. In cases where it does not happen we see the or- 1
ganizing and canalizing effects of ethnic distinctions. To explore the 2
factors responsible for the difference, let us first look at the specific 3
explanations for the changes of identity that have been advanced in 4
the examples discussed above. 5
In the case of Pathan borderlands, influence and security in the 6
segmentary and anarchic societies of this region derive from a man’s 7
previous actions, or rather from the respect that he obtains from these 8
acts as judged by accepted standards of evaluation. The main fora for 9
exhibiting Pathan virtues are the tribal council, and stages for the 10
display of hospitality. But the villager in Kohistan has a standard of 11
living where the hospitality he can provide can hardly compete with 12
that of the conquered serfs of neighbouring Pathans, while the client 13
of a Baluch leader cannot speak in any tribal council. To maintain 14
Pathan identity in these situations, to declare oneself in the running 15
as a competitor by Pathan value standards, is to condemn oneself 16
in advance to utter failure in performance. By assuming Kohistani 17
or Baluch identity, however, a man may, by the same performance, 18
score quite high on the scales that then become relevant. The incen- 19
tives to a change in identity are thus inherent in the change in cir- 20
cumstances. 21
Different circumstances obviously favour different performances. performances 22
Since ethnic identity is associated with a culturally specific set of val- 23
ue standards, it follows that there are circumstances where such an 24
identity can be moderately successfully realized, and limits beyond 25
which such success is precluded. I will argue that ethnic identities 26
will not be retained beyond these limits, because allegiance to basic 27
value standards will not be sustained where one’s own comparative 28
performance is utterly inadequate.3 The two components in this rela- 29
tive measure of success are, first, the performance of others and, sec- 30
ondly, the alternatives open to oneself. I am not making an appeal 31
to ecologic adaptation. Ecologic feasibility, and fitness in relation to 32
the natural environment, matter only in so far as they set a limit in 33
terms of sheer physical survival, which is very rarely approached by 34
ethnic groups. What matters is how well the others, with whom one 35
interacts and to whom one is compared, manage to perform, and 36
what alternative identities and sets of standards are available to the 37
individual. 38
39
40
41
42
43

424 fredrik barth
1 Ethnic identity and tangible assets

2
3 The boundary-maintaining factors in the Fur are not immediately il-
4 luminated by this argument. Haaland (pp. 65 f.) discusses the evalu-
5 ation of the nomad’s life by Fur standards and finds the balance be-
6 tween advantages and disadvantages inconclusive. To ascertain the
7 comparability of this case, we need to look more generally at all the
8 factors that affect the behaviour in question. The materials derive
9 from grossly different ethnographic contexts and so a number of fac-
10 tors are varied simultaneously.
11 productive resources The individual’s relation to productive resources stands out as the
12 significant contrast between the two regions. In the Middle East, the
13 means of production are conventionally held as private or corporate,
14 defined and transferable property. A man can obtain them through a
15 specific and restricted transaction, such as purchase or lease; even in
16 conquest the rights that are obtained are standard, delimited rights.
17 In Darfur, on the other hand, as in much of the Sudanic belt, the pre-
18 vailing conventions are different. Land for cultivation is allocated, as
19 needed, to members of a local community. The distinction between
20 owner and cultivator, so important in the social structure of most
21 Middle Eastern communities, cannot be made because ownership
22 does not involve separable, absolute, and transferable rights. Access
23 to the means of production in a Fur village is therefore conditional
24 only on inclusion in the village community – i.e. on Fur ethnic iden-
25 tity. Similarly, grazing rights are not allocated and monopolized, even
26 as between Baggara tribes. Though groups and tribes tend to use the
27 same routes and areas every year, and may at times try in an ad hoc
28 way to keep out others from an area they wish to use, they normally
29 intermix and have no defined and absolute prerogatives. Access to
30 grazing is thus an automatic aspect of practising husbandry, and en-
31 tails being a Baggara.
32 The gross mechanisms of boundary maintenance in Darfur are
33 thus quite simple: a man has access to the critical means of produc-
34 tion by virtue of practising a certain subsistence; this entails a whole
35 style of life, and all these characteristics are subsumed under the eth-
36 nic labels Fur and Baggara. In the Middle East, on the other hand,
37 men can obtain control over means of production through a transac-
38 tion that does not involve their other activities; ethnic identity is then
39 not necessarily affected and this opens the way for diversification.
40 Thus nomad, peasant, and city dweller can belong to the same eth-
41 nic group in the Middle East; where ethnic boundaries persist they
42 depend on more subtle and specific mechanisms, mainly connected
43 with the unfeasibility of certain status and role combinations.

Ethnic groups and stratification 1

2
Where one ethnic group has control of the means of production uti- 3
lized by another group, a relationship of inequality and stratification 4
obtains. Thus Fur and Baggara do not make up a stratified system, 5
since they utilize different niches and have access to them indepen- 6
dently of each other, whereas in some parts of the Pathan area one 7
finds stratification based on the control of land, Pathans being land- 8
owners, and other groups cultivating as serfs. In more general terms, 9
one may say that stratified poly-ethnic systems exist where groups 10
are characterized by differential control of assets that are valued by all 11
groups in the system. The cultures of the component ethnic groups 12
in such systems are thus integrated in a special way: they share cer- 13
tain general value orientations and scales, on the basis of which they 14
can arrive at judgements of hierarchy. hierarchy 15
Obversely, a system of stratification does not entail the existence 16
of ethnic groups. Leach (1967) argues convincingly that social class- 17
es are distinguished by different sub-cultures, indeed, that this is a 18
more basic characteristic than their hierarchical ordering. However, 19
in many systems of stratification we are not dealing with bounded 20
strata at all: the stratification is based simply on the notion of scales 21
and the recognition of an ego-centered level of ‘people who are just 22
like us’ versus those more select and those more vulgar. In such sys- 23
tems, cultural differences, whatever they are, grade into each other, 24
and nothing like a social organization of ethnic groups emerges. 25
Secondly, most systems of stratification allow, or indeed entail, mo- 26
bility based on evaluation by the scales that define the hierarchy. 27
Thus a moderate failure in the ‘R’ sector of the hierarchy makes you 28
a ‘C’, etc. Ethnic groups are not open to this kind of penetration: the 29
ascription of ethnic identity is based on other and more restrictive 30
criteria. This is most clearly illustrated by Knutsson’s analysis of the 31
Galla in the context of Ethiopian society (pp. 86 ff.) – a social system 32
where ‘Whole ethnic groups are stratified with respect to their posi- 33
tions of privilege and disability within the stale. Yet the attainment 34
of a governorship does not make an Amhara of a Galla, nor does 35
estrangement as an outlaw entail loss of Galla identity. 36
From this perspective, the Indian caste system would appear to 37
be a special case of a stratified poly-ethnic system. The boundaries of 38
castes are defined by ethnic criteria: thus individual failures in per- 39
formance lead to out-casting and not to down-casting. The process 40
whereby the hierarchical system incorporates new ethnic groups is 41
demonstrated in the sanscritization of tribals: their acceptance of the 42
critical value scales defining their position in the hierarchy of ritual 43

426 fredrik barth
1 purity and pollution is the only change of values that is necessary

2 for a people to become an Indian caste. An analysis of the different
3 processes of boundary maintenance involved in different inter-caste
4 relations and in different regional variants of the caste system would,
5 I believe, illuminate many features of this system.
6 The preceding discussion has brought out a somewhat anoma-
7 lous general feature of ethnic identity as a status: ascription4 is not
8 conditional on the control of any specific assets, but rests on criteria
9 of origin and commitment; whereas performance in the status, the ad-
10 equate acting out of the roles required to realize the identity, in many
11 systems does require such assets. By contrast, in a bureaucratic office
12 the incumbent is provided with those assets that are required for the
13 performance of the role; while kinship positions, which are ascribed
14 without reference to a person’s assets, likewise are not conditional on
15 performance – you remain a father even if you fail to feed your child.
16 distribution of assets Thus where ethnic groups are interrelated in a stratified system,
17 this requires the presence of special processes that maintain dif-
18 ferential control of assets. To schematize: a basic premise of ethnic
19 group organization is that every A can act roles 1, 2 and 3. If actors
20 agree on this, the premise is self-fulfilling, unless acting in these
21 roles requires assets that are distributed in a discrepant pattern. If
22 these assets are obtained or lost in ways independent of being an A,
23 and sought and avoided without reference to one’s identity as an A,
24 the premise will be falsified: some A’s become unable to act in the
25 expected roles. Most systems of stratification are maintained by the
26 solution that in such cases, the person is no longer an A. In the case
27 of ethnic identity, the solution on the contrary is the recognition that
28 every A no longer can or will act in roles 1 and 2. The persistence
29 of stratified poly-ethnic systems thus entails the presence of factors
30 that generate and maintain a categorically different distribution of
31 assets: state controls, as in some modern plural and racist systems;
32 marked differences in evaluation that canalize the efforts of actors
33 in different directions, as in systems with polluting occupations; or
34 differences in culture that generate marked differences in political
35 organization, economic organization, or individual skills.
36
37
38 The problem of variation
39
40 Despite such processes, however, the ethnic label subsumes a num-
41 ber of simultaneous characteristics which no doubt cluster statisti-
42 cally, but which are not absolutely interdependent and connected.
43 Thus there will be variations between members, some showing

many and some showing few characteristics. Particularly where peo- 1

ple change their identity, this creates ambiguity since ethnic mem- 2
bership is at once a question of source of origin as well as of cur- 3
rent identity. Indeed, Haaland was taken out to see ‘Fur who live in 4
nomad camps’, and I have heard members of Baluch tribal sections 5
explain that they are ‘really Pathan’. What is then left of the bound- 6
ary maintenance and the categorical dichotomy, when the actual dis- 7
tinctions are blurred in this way? Rather than despair at the failure 8
of typological schematism, one can legitimately note that people do 9
employ ethnic labels and that there are in many parts of the world 10
most spectacular differences whereby forms of behaviour cluster so 11
that whole actors tend to fall into such categories in terms of their 12
objective: behaviour. What is surprising is not the existence of some 13
actors that fall between these categories, and of some regions in the 14
world where whole persons do not tend to sort themselves out in this 15
way, but the fact that variations tend to cluster at all. We can then be 16
concerned not to perfect a typology, but to discover the processes that 17
bring about such clustering. 18
An alternative mode of approach in anthropology has been to di- anthropology 19
chotomize the ethnographic material in terms of ideal versus actual 20
or conceptual versus empirical, and then concentrate on the con- 21
sistencies (the ‘structure’) of the ideal, conceptual part of the data, 22
employing some vague notion of norms and individual deviance to 23
account for the actual, statistical patterns. It is of course perfectly fea- 24
sible to distinguish between a people’s model of their social system 25
and their aggregate pattern of pragmatic behaviour, and indeed quite 26
necessary not to confuse the two. But the fertile problems in social 27
anthropology are concerned with how the two are interconnected, 28
and it does not follow that this is best elucidated by dichotomizing 29
and confronting them as total systems. In these essays we have tried 30
to build the analysis on a lower level of interconnection between 31
status and behaviour. I would argue that people’s categories are for 32
acting, and are significantly affected by interaction rather than con- 33
templation. In showing the connection between ethnic labels and the 34
maintenance of cultural diversity, I am therefore concerned primar- 35
ily to show how, under varying circumstances, certain constellations 36
of categorization and value orientation have a self-fulfilling charac- 37
ter, how others will tend to be falsified by experience, while others 38
again are incapable of consummation in interaction. Ethnic bound- 39
aries can emerge and persist only in the former situation, whereas 40
they should dissolve or be absent in the latter situations. With such 41
a feedback from people’s experiences to the categories they employ, 42
simple ethnic dichotomies can be retained, and their stereotyped 43

428 fredrik barth
1 behavioural differential reinforced, despite a considerable objective

2 variation. This is so because actors struggle to maintain conventional
3 definitions of the situation in social encounters through selective
4 perception, tact, and sanctions, and because of difficulties in finding
5 other, more adequate codifications of experience. Revision only takes
6 place where the categorization is grossly inadequate – not merely be-
7 cause it is untrue in any objective sense, but because it is consistently
8 unrewarding to act upon, within the domain where the actor makes
9 it relevant. So the dichotomy of Fur villagers and Baggara nomads is
10 maintained despite the patent presence of a nomadic camp of Fur
11 in the neighbourhood: the fad that those nomads speak Fur and
12 have kinship connections with villagers somewhere does not change
13 the social situation in which the villager interacts with them – it
14 simply makes the standard transactions of buying milk, allocating
15 camp sites, or obtaining manure, which one would have with other
16 Baggara, flow a bit more smoothly. But a dichotomy between Pathan
17 landowners and non-Pathan labourers can no longer be maintained
18 where non-Pathans obtain land and embarrass Pathans by refusing
19 to respond with the respect which their imputed position as menials
20 would have sanctioned.
21
22
23 Minorities, pariahs, and organizational characteristics of
24 the periphery
25
26 In some social systems, ethnic groups co-reside though no major
27 aspect of structure is based on ethnic inter-relations. These are gen-
28 erally referred to as societies with minorities, and the analysis of the
29 minority situation involves a special variant of inter-ethnic relations.
30 I think in most cases, such situations have come about as a result of
31 external historical events; the cultural differentiae have not sprung
32 from the local organizational context – rather, a pre-established cul-
33 tural contrast is brought into conjunction with a pre-established so-
34 cial system, and is made relevant to life there in a diversity of ways.
35 minorities An extreme form of minority position, illustrating some but not
36 all features of minorities, is that of pariah groups. These are groups
37 actively rejected by the host population because of behaviour or char-
38 acteristics positively condemned, though often useful in some spe-
39 cific, practical way. European pariah groups of recent centuries (ex-
40 ecutioners, dealers in horseflesh and -leather, collectors of nightsoil,
41 gypsies, etc.) exemplify most features: as breakers of basic taboos
42 they were rejected by the larger society. Their identity imposed a
43 definition on social situations which gave very little scope for interac-

tion with persons in the majority population, and simultaneously as 1

an imperative status represented an inescapable disability that pre- 2
vented them from assuming the normal statuses involved in other 3
definitions of the situation of interaction. Despite these formidable 4
barriers, such groups do not seem to have developed the internal 5
complexity that would lead us to regard them as full-fledged ethnic 6
groups; only the culturally foreign gypsies5 clearly constitute such a 7
group. 8
The boundaries of pariah groups are most strongly maintained 9
by the excluding host population, and they are often forced to make the excluding host 10
use of easily noticeable diacritica to advertise their identity (though population 11
since this identity is often the basis for a highly insecure livelihood, 12
such over-communication may sometimes also serve the pariah in- 13
dividual’s competitive interests). Where pariahs attempt to pass into 14
the larger society, the culture of the host population is generally well 15
known; thus the problem is reduced to a question of escaping the 16
stigmata of disability by dissociating with the pariah community and 17
faking another origin. 18
Many minority situations have a trace of this active rejection by 19
the host population. But the general feature of all minority situations 20
lies in the organization of activities and interaction: In the total social 21
system, all sectors of activity are organized by statuses open to mem- 22
bers of the majority group, while the status system of the minority 23
has only relevance to relations within the minority and only to some 24
sectors of activity, and does not provide a basis for action in other 25
sectors, equally valued in the minority culture. There is thus a dis- 26
parity between values and organizational facilities: prized goals are 27
outside the field organized by the minority’s culture and categories. 28
Though such systems contain several ethnic groups, interaction be- 29
tween members of the different groups of this kind does not spring 30
from the complementarity of ethnic identities; it takes place entirely 31
within the framework of the dominant, majority group’s statuses 32
and institutions, where identity as a minority member gives no basis 33
for action, though it may in varying degrees represent a disability in 34
assuming the operative statuses. Eidheim’s paper gives a very clear 35
analysis of this situation, as it obtains among Coast Lapps. 36
But in a different way, one may say that in such a poly-ethnic sys- 37
tem, the contrastive cultural characteristics of the component groups 38
are located in the non-articulating sectors of life. For the minority, 39
these sectors constitute a ‘backstage’ where the characteristics that 40
are stigmatic in terms of the dominant majority culture can covertly 41
be made the objects of transaction. 42
43

430 fredrik barth
1 The present-day minority situation of Lapps has been brought

2 about by recent external circumstances. Formerly, the important con-
3 text of interaction was the local situation, where two ethnic groups
4 with sufficient knowledge of each other’s culture maintained a rela-
5 tively limited, partly symbiotic relationship based in their respective
6 identities. With the fuller integration of Norwegian society, bring-
7 ing the northern periphery into the nation-wide system, the rate of
8 cultural change increased drastically. The population of Northern
9 Norway became increasingly dependent on the institutional system
10 of the larger society, and social life among Norwegians in Northern
11 Norway was increasingly organized to pursue activities and obtain
12 benefits within the wider system. This system has not, until very re-
13 cently, taken ethnic identity into account in its structure, and until a
14 decade ago there was practically no place in it where one could partic-
15 ipate as a Lapp. Lapps as Norwegian citizens, on the other hand, are
16 perfectly free to participate, though under the dual disability of pe-
17 ripheral location and inadequate command of Norwegian language
18 and culture. This situation has elsewhere, in the inland regions of
19 Finnmark, given scope for Lappish innovators with a political pro-
20 gram based on the ideal of ethnic pluralism (cf. Eidheim 1967), but
21 they have gained no following in the Coast Lapp area here discussed
22 by Eidheim. For these Lapps, rather, the relevance of Lappish sta-
23 tuses and conventions decreases in sector after sector (cf. Eidheim
24 1966), while the relative inadequacy of performance in the widest
25 system brings about frustrations and a crisis of identity.
26
27
28 Culture contact and change
29
30 This is a very widespread process under present conditions as depen-
31 dence on the products and institutions of industrial societies spreads
32 in all parts of the world. The important thing to recognize is that a
33 drastic reduction of cultural differences between ethnic groups does
34 not correlate in any simple way with a reduction in the organizational
35 relevance of ethnic identities, or a breakdown in boundary-maintain-
36 ing processes. This is demonstrated in much of the case material.
37 We can best analyse the interconnection by looking at the agents
38 of change: what strategies are open and attractive to them, and what
39 are the organizational implications of different choices on their part?
40 The agents in this case are the persons normally referred to some-
41 new elites what ethno-centrically as the new elites: the persons in the less in-
42 dustrialized groups with greater contact and more dependence on
43 the goods and organizations of industrialized societies. In their pur-

suit of participation in wider social systems to obtain new forms of 1

value they can choose between the following basic strategies: (i) they basic strategies 2
may attempt to pass and become incorporated in the pre-established 3
industrial society and cultural group; (ii) they may accept a ‘minority’ 4
status, accommodate to and seek to reduce their minority disabilities 5
by encapsulating all cultural differentiae in sectors of non-articula- 6
tion, while participating in the larger system of the industrialized 7
group in the other sectors of activity; (iii) they may choose to empha- 8
size ethnic identity, using it to develop new positions and patterns to 9
organize activities in those sectors formerly not found in their soci- 10
ety, or inadequately developed for the new purposes. If the cultural 11
innovators are successful in the first strategy, their ethnic group will 12
be denuded of its source of internal diversification and will probably 13
remain as a culturally conservative, low-articulating ethnic group 14
with low rank in the larger social sytem. A general acceptance of the 15
second strategy will prevent the emergence of a clearly dichotomiz- 16
ing polyethnic organization, and – in view of the diversity of indus- 17
trial society and consequent variation and multiplicity of fields of ar- 18
ticulation – probably lead to an eventual assimilation of the minority. 19
The third strategy generates many of the interesting movements that 20
can be observed today, from nativism to new states. 21
I am unable to review the variables that affect which basic strategy 22
will be adopted, which concrete form it may take, and what its degree 23
of success and cumulative implications may be. Such factors range 24
from the number of ethnic groups in the system to features of the eco- 25
logic regime and details of the constituent cultures, and are illustrated 26
in most of the concrete analyses of the following essays. It may be of 27
interest to note some of the forms in which ethnic identity is made 28
organizationally relevant to new sectors in the current situation. 29
Firstly, the innovators may choose to emphasize one level of iden- 30
tity among the several provided by the traditional social organization. 31
Tribe, caste, language group, region or state all have features that 32
make them a potentially adequate primary ethnic identity for group 33
reference, and the outcome will depend on the readiness with which 34
others can be led to embrace these identities, and the cold tactical 35
facts. Thus, though tribalism may rally the broadest support in many 36
African areas, the resultant groups seem unable to stand up against 37
the sanctioning apparatus even of a relatively rudimentary state orga- 38
nization. 39
Secondly, the mode of organization of the ethnic group varies, as 40
does the inter-ethnic articulation that is sought. The fact that contem- 41
porary forms are prominently political does not make them any less 42
ethnic in character. Such political movements constitute new ways 43

432 fredrik barth
1 of making cultural differences organizationally relevant (Kleivan

2 1967), and new ways of articulating the dichotomized ethnic groups.
3 The proliferation of ethnically based pressure groups, political par-
4 ties, and visions of independent statehood, as well as the multitude
5 of subpolitical advancement associations (Sommerfelt 1967) show
6 the importance of these new forms. In other areas, cult-movements
7 or mission-introduced sects are used to dichotomize and articulate
8 groups in new ways. It is striking that these new patterns are so rare-
9 ly concerned with the economic sector of activities, which is so major
10 a factor in the culture contact situation, apart from the forms of state
11 socialism adopted by some of the new nations. By contrast, the tradi-
12 tional complex poly-ethnic systems have been prominently based on
13 articulation in this sector, through occupational differentiation and
14 articulation at the market place in many regions of Asia and Middle
15 America, or most elaborately, through agrarian production in South
16 Asia. Today, contending ethnic groups not infrequently become dif-
17 ferentiated with respect to educational level and attempt to control
18 or monopolize educational facilities for this purpose (Sommerfelt
19 1967), but this is not so much with a view to occupational differen-
20 tiation as because of the obvious connection between bureaucratic
21 competence and opportunities for political advancement. One may
22 speculate that an articulation entailing complex differentiation of
23 skills, and sanctioned by the constant dependence on livelihood, will
24 have far greater strength and stability than one based on revocable
25 political affiliation and sanctioned by the exercise of force and politi-
26 cal fiat, and that these new forms of poly-ethnic systems are probably
27 inherently more turbulent and unstable than the older forms.
28 When political groups articulate their opposition in terms of eth-
29 political nic criteria, the direction of cultural change is also affected. A po-
30 confrontation litical confrontation can only be implemented by making the groups
31 similar and thereby comparable, and this will have effect on every
32 new sector of activity which is made politically relevant. Opposed par-
33 ties thus tend to become structurally similar, and differentiated only
34 by a few clear diacritica. Where ethnic groups are organized in politi-
35 cal confrontation in this way, the process of opposition will therefore
36 lead to a reduction of the cultural differences between them.
37 For this reason, much of the activity of political innovators is con-
38 cerned with the codification of idioms: the selection of signals for
39 identity and the assertion of value for these cultural diacritica, and
40 the suppression or denial of relevance for other differentiae. The is-
41 sue as to which new cultural forms are compatible with the native
42 ethnic identity is often hotly contended, but is generally settled in
43 favour of syncretism for the reasons noted above. But a great amount

of attention may be paid to the revival of select traditional culture 1

traits, and to the establishment of historical traditions to justify and 2
glorify the idioms and the identity. 3
The interconnection between the diacritica that are chosen for 4
emphasis, the boundaries that are defined, and the differentiating 5
values that are espoused, constitute a fascinating field for study.6 6
Clearly, a number of factors are relevant. Idioms vary in their appro- 7
priateness for different kinds of units. They are unequally adequate 8
for the innovator’s purposes, both as means to mobilize support 9
and as supports in the strategy of confrontation with other groups. 10
Their stratificational implications both within and between groups 11
are important: they entail different sources and distributions of in- 12
fluence within the group, and different claims to recognition from 13
other groups through suppression or glorification of different forms 14
of social stigmata. Clearly, there is no simple connection between 15
the ideological basis of a movement and the idioms chosen; yet both 16
have implications for subsequent boundary maintenance, and the 17
course of further change. 18
19
20
Variations in the setting for ethnic relations 21
22
These modern variants for poly-ethnic organization emerge in a 23
world of bureaucratic administration, developed communications, 24
and progressive urbanization. Clearly, under radically different cir- 25
cumstances, the critical factors in the definition and maintenance 26
of ethnic boundaries would be different. In basing ourselves on lim- 27
ited and contemporary data, we are faced with difficulties in general- 28
izing about ethnic processes, since major variables may be ignored 29
because they are not exhibited in the cases at our disposal. There 30
can be little doubt that social anthropologists have tended to regard 31
the rather special situation of colonial peace and external administra- 32
tion, which has formed the backdrop of most of the influential mono- 33
graphs, as if this were representative of conditions at most times and 34
places. This may have biased the interpretation both of pre-colonial 35
systems and of contemporary, emergent forms. The attempt in these 36
essays to cover regionally very diverse cases is not alone an adequate 37
defence against such bias, and the issue needs to be faced directly. 38
Colonial regimes are quite extreme in the extent to which the ad- colonial regimes 39
ministration and its rules are divorced from locally based social life. 40
Under such a regime, individuals hold certain rights to protection 41
uniformly through large population aggregates and regions, far be- 42
yond the reach of their own social relationships and institutions. This 43

434 fredrik barth
1 allows physical proximity and opportunities for contact between per-

2 sons of different ethnic groups regardless of the absence of shared
3 understandings between them, and thus clearly removes one of the
4 constraints that normally operate on inter-ethnic relations. In such
5 situations, interaction can develop and proliferate – indeed, only
6 those forms of interaction that are directly inhibited by other factors
7 will be absent and remain as sectors of non-articulation. Thus ethnic
8 boundaries in such situations represent a positive organization of so-
9 cial relations around differentiated and complementary values, and
10 cultural differences will tend to be reduced with time and approach
11 the required minimum.
12 political regimes In most political regimes, however, where there is less security
13 and people live under a greater threat of arbitrariness and violence
14 outside their primary community, the insecurity itself acts as a con-
15 straint on inter-ethnic contacts. In this situation, many forms of
16 interaction between members of different ethnic groups may fail
17 to develop, even though a potential complementarity of interests
18 obtains. Forms of interaction may be blocked because of a lack of
19 trust or a lack of opportunities to consummate transactions. What is
20 more, there are also internal sanctions in such communities which
21 tend to enhance overt conformity within and cultural differences
22 between communities. If a person is dependent for his security on
23 the voluntary and spontaneous support of his own community, self-
24 identification as a member of this community needs to be explicitly
25 expessed and confirmed: and any behaviour which is deviant from
26 the standard may be interpreted as a weakening of the identity, and
27 thereby of the bases of security. In such situations, fortuitous histori-
28 cal differences in culture between different communities will tend
29 to perpetuate themselves without any positive organizational basis:
30 many of the observable cultural differentiae may thus be of very lim-
31 ited relevance to the ethnic organization.
32 The processes whereby ethnic units maintain themselves are thus
33 clearly affected, but not fundamentally changed, by the variable of
34 regional security. This can also be shown by an inspection of the
35 cases analysed in these essays, which represent a fair range from the
36 colonial to the poly-centric, up to relatively anarchic situations. It is
37 important, however, to recognize that this background variable may
38 change very rapidly with time, and in the projection of long-range
39 processes this is a serious difficulty. Thus in the Fur case, we observe
40 a situation of externally maintained peace and very small-scale local
41 political activity, and can form a picture of inter-ethnic processes and
42 even rates in this setting. But we know that over the last few genera-
43 tions, the situation has varied from one of Baggara-Fur confrontation

under an expansive Fur sultanate to a nearly total anarchy in Turkish 1

and Mabdi times; and it is very difficult to estimate the efforts of 2
these variations on the processes of nomadization and assimilation, 3
and arrive at any long-range projection of rates and trends. 4
5
6
Ethnic groups and cultural evolution 7
8
The perspective and analysis presented here have relevance to the 9
theme of cultural evolution. No doubt human history is a story of 10
the development of emergent forms, both of cultures and societies. 11
The issue in anthropology has been how this history can best be de- 12
picted, and what kinds of analyses are adequate to discover general 13
principles in the courses of change. Evolutionary analysis in the rig- 14
orous sense of the biological fields has based its method on the con- 15
struction of phyletic lines. This method presumes the existence of 16
units where the boundaries and the boundary-maintaining processes 17
can be described, and thus where the continuity can be specified. 18
Concretely, phyletic lines are meaningful because specific boundar- 19
ies prevent the interchange of genetic material; and so one can insist 20
that the reproductive isolate is the unit, and that it has maintained an 21
identity undisturbed by the changes in the morphological character- 22
istics of the species. 23
I have argued that boundaries are also maintained between ethnic 24
units, and that consequently it is possible to specify the nature of 25
continuity and persistence of such units. These essays try to show 26
that ethnic boundaries are maintained in each case by a limited set 27
of cultural features. The persistence of the unit then depends on the 28
persistence of these cultural differentiae, while continuity can also be changes in cultural 29
specified through the changes of the unit brought about by changes differentiae 30
in the boundary-defining cultural differentiae. 31
However, most of the cultural matter that at any time is associ- 32
ated with a human population is not constrained by this boundary; 33
it can vary, be learnt, and change without any critical relation to the 34
boundary maintenance of the ethnic group. So when one traces the 35
history of a ethnic group through time, one is not simultaneously, 36
in the same sense, tracing the history of ‘a culture’: the elements of 37
the present culture of that ethnic group have not sprung from the 38
particular set that constituted the group’s culture at a previous time, 39
whereas the group has a continual organizational existence with 40
boundaries (criteria of membership) that despite modifications have 41
marked off a continuing unit. 42
Without being able to specify the boundaries of cultures, it is not 43

436 fredrik barth
1 possible to construct phyletic lines in the more rigorous evolution-

2 ary sense. But from the analysis that has been argued here, it should
3 be possible to do so for ethnic groups, and thus in a sense for those
4 aspects of culture which have this organizational anchoring.
5
6
7 Notes
8
9 1 The emphatic ideological denial of the primacy of ethnic identity (and rank)
10 which characterises the universal religions that have arisen in the Middle
East is understandable in this perspective, since practically any movement
11 for social or ethical reform in the poly-ethnic societies of that region would
12 clash with conventions and standards of ethnic character.
13 2 The difference between ethnic groups and social strata, which seems prob-
14 lematical at this stage of the argument, will be taken up below.
15 3 I am here concerned only with individual failure to maintain identity, where
most members do so successfully, and not with the broader questions of
16
cultural vitality and anomie.
17 4 As opposed to presumptive classification in passing social encounters – I am
18 thinking of the person in his normal social context where others have a con-
19 siderable amount of previous information about him, not of the possibilities
20 afforded occasionally for mispresenting one’s identity towards strangers.
21 5 The condemned behaviour which gives pariah position to the gypsies is com-
pound, but rests prominently on their wandering life, originally in contrast
22 to the serf bondage of Europe, later in their flagrant violation of puritan eth-
23 ics of responsibility, toil and morality.
24 6 To my knowledge, Mitchell’s essay on the Kalda dance (Mitchell 1956) is the
25 first and still the most penetrating study on this topic.
26
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
The theory of race relation: 3
4
a Weberian approach
5
6
John Rex 7
8
9
10
11
12
The contribution of John Rex to the sociology of ethnic and race relations in
13
Britain, as well as throughout the Continent, is immense. Nowadays, we often 14
hear how ethnic and migration studies in Europe presents a lacuna: its lack 15
of connection with the development of general social and political theories. 16
In this article, first published under UNESCO’s auspices in 1980 in a book 17
entitled Sociological Theories: Race and Colonialism, Rex clarifies the specific- 18
ity of the Weberian approach to sociology. He does this with intelligence and 19
sophistication, also identifying its particular contribution to the sociology of 20
race relations. By doing so, Rex links the sociology of race relations to general 21
sociological theory in a very rigorous way. This article illustrates the necessity 22
of such theoretically oriented work for the further development of ethnic and 23
migration studies. 24
25
26
It would be foolish to suggest that any one school of sociology held 27
a monopoly of wisdom in the field of race relations theory. Equally 28
it would be misleading to suggest that any of the great founders of 29
sociological theory in the nineteenth and early twentieth century had 30
dealt directly with the problem of race relations. Nonetheless Max Weber 31
Weber is at least one of the founders of the discipline of sociology; 32
there is what one might call a specifically Weberian style of sociologi- 33
cal thinking; and there can be little doubt that the scope of Weber’s 34
comparative studies, in terms of both time and place, make it in- 35
evitable that his work should throw at least an indirect light on the 36
structure of the relationships between racial and ethnic groups. 37
The following distinctions between schools of sociological think- 38
ing might perhaps be briefly made in order to clarify the specificity 39
of the Weberian approach to sociology. They distinguish it from the 40
tradition of French Positivism running from Comte to Durkheim 41
and from that of Marxism, which, although it is far more than a so- 42
ciology, is an approach to the study of nature, culture and society 43

438 john rex
1 which has considerable implications for sociology. They also distin-

2 guish the approach of Weber from that of a number of contempo-
3 rary schools of sociology, some of which represent out-growths of
4 the Weberian approach, which emphasize certain implicit aspects
5 of Weber’s thought. Thus one has in modern sociology Positivist
6 Empiricism as represented in American sociology by such authors
7 as Lazarsfeld and Blalock, the highly systematic general theory of
8 Talcott Parsons commonly called Structural-Functionalism, and
9 the various reactions to these schools which may loosely be called
10 phenomenological, including the seminal work of Schutz, the trend
11 known as Symbolic Interactionism deriving from the work of Mead,
12 and the growing trend of Ethnomethodology. And while these are
13 schools from which Weberian sociology has to be distinguished in
14 America there are also distinctions to be made within European social
15 thought. There, a variety of approaches to human affairs have arisen
16 from the Phenomenology of Husserl, through the intermingling of
17 phenomenological themes with those of existentialism, through the
18 critique of the Logical Positivism of the Vienna Circle which culmi-
19 nates in the Critical Theory of the Frankfurt School and later in the
20 more complex theories of Habermas, through the new development
21 of Structuralism in the social sciences deriving from the work of Levi-
22 Strauss, through the development of Orthodox Marxism within the
23 Communist movement and through the differing critiques of this
24 orthodoxy represented by the work of Lukacs, Gramsci, Althusser
25 and related writers.
26 I cannot hope to list systematically the differences of concept,
27 style and method which distinguish the work of Weber from these
28 trends taken one by one. But I may point to certain salient features of
29 Weberianism which in part indicate its concern with a core of prob-
30 lems common to all sociological endeavour and in part indicate the
31 specific restrictions which this approach places on the sociological
32 enterprise.
33 The common core of all sociology is to be found in its concern
34 with social relations and, underlying this, with social action. Talcott
35 the structure Parsons in fact emphasized this in entitling his greatest book The
36 of social action structure of social action and underlined the point when he published
37 The social system by saying that he would again have used the earlier
38 title if it had not been pre-empted by the earlier work.1 What this defi-
39 nition of sociology’s core concern excludes is any kind of reduction-
40 ism which reduces social facts to epiphenomena consequential upon
41 biological or psychological causes as well as any kind of reification
42 of social facts which suggests that they are ‘things’ or that they are
43 to be seen as consequent upon the working of social systems. Such

a concern is common to the early Marx, especially in the Theses on 1

Feuerbach, to Simmel and to Weber, and is also evident in the prac- 2
tice of Emile Durkheim, if not in his systematic statements of his 3
methodological position. It is also to be found in the concept of social 4
structure, employed by the English anthropologist Radcliffe Brown 5
and specifically rejected by Levi-Strauss. 2 6

It is also common to all the founders of sociology that they em- 7
ployed their concept of structure of social relations to distinguish the 8
specificity of the modern capitalist industrial order and to contrast modern capitalist 9
that order with pre-capitalist and putative post-capitalist formations. order 10
Quite obviously, Marx and Weber equally recognize that the modern 11
capitalist order is a distinctive one not paralleled in the ancient world, 12
even though they differ in their precise concepts of the distinction, 13
and both attempt some sort of morphology and theory of the succes- 14
sion of social types. Again, both Durkheim and Simmel as well as 15
Tonnies are concerned with the distinctiveness of the new industrial 16
society brought into being by the industrial revolution even if they do 17
not discuss this society as being essentially capitalist. All these writ- 18
ers have something to say about the nature of pre-historic society, 19
about ancient civilizations, about the medieval or ‘feudal’ world, and 20
about the possibilities of a socialist form of industrialism. 21
22
Weber’s distinctiveness lies in his peculiar sensitivity to the tension 23
between the notion of the ‘facticity’ of the social world on the one 24
hand and its availability to human control on the other. He always 25
insists that group concepts must be used only as a shorthand capable 26
of being explicated in terms of more fundamental units, namely, 27
the theoretical actors, whose taking account of one another’s actions 28
serves to constitute the group’s structure. Yet he by no means sug- 29
gests that social structure can be reduced to the motivations of indi- 30
vidual empirical human beings. Structures are seen as arising from 31
the continuity in time of interlocking patterns of interaction and, 32
though these may be changed by intervention at strategic points, 33
it would be utopian in the extreme to suggest that these structures 34
were always and everywhere open to change by redefinition. This is 35
what distinguishes Weber from all rigid forms of determinism on 36
the one hand, and from any kind of subjectivism and utopianism 37
on the other. He would be equally opposed to any orthodox or neo- 38
Marxist concept of a deterministic science of social formations, and 39
to the kind of subjectivism common in the more vulgarized forms 40
of phenomenological sociology which suggest that social reality is 41
a matter of labels which can readily be replaced by alternative ones. 42
More profoundly than this, it must also be said that Weber’s 43

440 john rex
1 thinking about the nature of sociology and the vocation of the so-
2 ciologist is deeply impregnated with the philosophy of Kant. What
3 this means is that he recognizes the value of social science, but also
4 sees its limitations. Social science does not yield ultimate truth. It in-
5 volves looking at the world of social structure in a quasi-phenomenal
6 way as organized in terms of humanly-imposed categories of action
7 and social relations, and in using our knowledge of structure as a
8 guide to practical action. But our findings about such structures have
9 no ontological significance as they are frequently thought to have by
10 writers in the Hegelian and Marxist traditions. It is also possible for
11 men – appreciating, as the result of their sociological investigations,
12 the intractability of the social world – to evaluate that world, to make
13 their own judgements about it and, indeed, guided by freely-formed
14 value judgements, to select certain limited ranges of determinate so-
15 cial reality for investigation in order that they might be better con-
16 trolled. It is his awareness both of the intractability of an increasingly
17 organized and bureaucratized world as well as his belief in the ines-
18 capable moral responsibility of men for their actions that produces in
19 Weber a stance of intellectual heroism quite unlike that to be found
20 in any other sociology. In particular, it is opposed to any post-Kan-
21 tian tendency which asserts either the union of science and ethics
22 in ontology, or the irrelevance or impropriety of value judgements
23 about the findings of social science. Weber would not be at home,
24 therefore, either with those Hegelian Marxists who insist upon the
25 unity of the observed world and the observing subject, or with mod-
26 ern Positivist Empiricism which accords reality only to statements
27 capable of being verified or disproved.
28 race relations All this may seem a far cry from the study of race relations. Yet
29 it is precisely Weber’s stands on these issues which makes his so-
30 ciology so relevant to the problems of understanding race relations
31 today. For while Weber did make some empirical contribution to the
32 analysis of structures closely connected with race relations problems,
33 the most important point to notice about the relevance of his work to
34 the study of race relations is that he shows us that while it is possible
35 to follow through long chains of ‘causality’ in our study of ethnic and
36 racial structures, it is also possible to evaluate those structures and
37 to suggest points at which the institutionalized actions which un-
38 derly structures may most effectively be altered so as to bring about
39 a different social outcome. This is a point which was grasped lucidly
40 and simply by Gunnar Myrdal in his study of North American race
41 relations. Myrdal saw that, as a sociologist, he would be saying not
42 merely that such-and-such was the case, but that is was necessarily
43 the case, and that this kind of assertion imperiously posed the ques-
tion ‘Necessary from what (or whose) point of view?’3


So far as Weber’s direct empirical contribution to the study of race 1
and ethnic relations is concerned, it is necessary, in the first place, to 2
make some distinction between his overt political interventions and 3
his more academic work in economic history. In neither case does he 4
give any support to racist interpretations of events, but the former are 5
more likely to be tinged with a political stance of nationalism which 6
is missing in the more detached comparative historical work. Much 7
of Weber’s earliest work was concerned with ancient and medieval 8
history but, when he turned to modern problems, he did so asking 9
what light this history threw on problems confronting the German 10
nation of his day. Thus, some of his earliest work for the Verein fur 11
Sozialpolitik was concerned with locating factors which might under- 12
mine the solidarity of the German nation. He studied the conditions 13
of agricultural workers in East Germany, influenced by his belief that 14
feudal and seignorial social relations represented some amelioration 15
of the conditions of labour when these were contrasted with those of 16
slaves in the latifundia of the late Roman empire. But these essen- 17
tially paternalistic relations were being undermined in East Germany 18
through the penetration of the system by the forces of unrestrained 19
capitalism. Weber’s response to this situation was to deplore the es- 20
tablishment of master-servant relations based purely on what Marx 21
called ‘the callous cash nexus’, but then to embark upon a prolonged 22
series of studies which sought to show that a modernizing capitalist 23
system could be based upon ethical values, albeit values of a pecu- 24
liarly introverted and individualistic kind4. 25
Superficially this study of immigrant labour would suggest that 26
Weber was merely a crude nationalist who was opposed to the im- 27
migration of Polish workers simply because such workers did not fit 28
in with the social structures or cultures of Germany. This conclusion 29
would be wrong, however, as Weber’s more systematic writings on 30
the question of race and ethnicity show. These are to be found in 31
the early chapters of his Economy and society which occur in the final Economy and 32
published version of that work as Part Two of Volume One, being society 33
preceded by four chapters of a systematic kind written later which 34
make no reference to the question of race or ethnicity5. 35
In this, Weber’s most explicit discussion of race and ethnicity, it 36
is clear that Weber does not regard bonds based on ethnicity alone 37
as significant bases for the structuration of society, at least so far as 38
its economic operations are concerned. There is a renewed reference 39
to the Poles in East Germany in the last part, but the discussion is 40
discursive and, if anything, Weber underplays the degree to which 41
they are or feel themselves to be segregated from German society. 42
The main theoretical point made in this chapter taken as a whole is 43

442 john rex
1 simply that, while race or ethnicity taken by itself is a weak structur-

2 ing bond within the economy, when there is need for a group to close
3 a social relationship and, thereby, also access to economic opportuni-
4 ties, race in the sense of difference of perceived physical appearance
5 is one, but only one, possible basis for the exclusion of individuals
6 from the closed relationship.
7 closure of The closure of relationships is the crucial factor for Weber.
8 relationships Human beings normally close relationships in order to ensure eco-
9 nomic opportunities for an in-group at the expense of an out-group.
10 Closure of relationships in this way is the most fundamental catego-
11 ry of economic sociology and is prior to ‘appropriation’, which is the
12 basis of property rights. Thus it is clear that race relations or ethnic
13 group relations may have a crucial function for economic life and
14 the making available of a group for exploitation. But race and ethnic-
15 ity need not necessarily be salient in this way, the only functional
16 necessity of the economic system being the division of a society into
17 closed property-owning groups and those who do not have access to
18 the same opportunities as the property-owners for the acquisition
19 of utilities. Here, Weber is obviously very close to Marx, and clearly
20 transcends the simple nationalism which might be read into his writ-
21 ings on East Germany.
22 Marx A good illustration of Weber’s economic or, in a loose sense,
23 ‘Marxist’ interpretation of ethnic difference and segregation is to be
24 found in his remarks on intermarriage between black and white in
25 the United States. Thus he tells us:
26
27 ‘Serious research on the sexual attraction and repulsion between dif-
28 ferent ethnic groups is only incipient, but there is not the slightest
29 doubt that racial factors, that means common descent, influence the
30 incidence of sexual relations and of marriage, sometimes decisively.
31 However the existence of several million mulattoes in the United
32 States speaks clearly against the assumption of a “natural” antipa-
33 thy, even among quite different races. Apart from the laws against
34 biracial marriage in the Southern States, sexual relations are now
35 abhorred by both sides, but this development began only with eman-
36 cipation and resulted from the Negroes’ demand for equal civil right.
37 Hence this abhorrence on the part of Whites is socially determined
38 by the previously sketched tendency toward the monopolization of
39 social power and honour, a tendency which happens in this case to
40 be linked to race.’6
41
42 Earlier, in his more general chapter on ‘The economic relationships
43 of organized groups’, he writes:

‘When the number of competitors increases in relation to the profit 1

span the participants become interested in curbing competition. 2
Usually one group of competitors takes some externally identifiable 3
characteristic of another group of (actual or potential) competitors 4
– race, language, religion, local or social origin, descent, residence 5
etc. – as a pretext for attempting their exclusion. It does not mat- 6
ter which characteristic is chosen in the individual case: whatever 7
suggests itself most easily is seized upon. Such group action may 8
provoke a corresponding reaction on the part of those against whom 9
it is directed.’7 10
11
These passages make it clear that Weber shares with liberals on the 12
race question, as he does with Marxists, the belief that racial or ethnic 13
exclusiveness is not effective as an intractable force in itself in creat- 14
ing racial separation and conflict. And with Marxists he would be 15
predisposed to look for its origin in the attempt to close off economic 16
opportunities by one group as against another. 17
It by no means follows from this that even the most intense forms 18
of closure and exploitation of the excluded, as in the case of slavery, slavery 19
necessarily and always lead to racial exclusiveness and conflict. This 20
is clear from Weber’s extended discussion of slavery in the ancient 21
world8 which he sees as an important and distinctive institution so 22
far as the working of the economy is concerned but not, as in the 23
North American case, leading to the closure of relationships on racial 24
grounds. 25
The question of slavery is so closely related to that of race rela- 26
tions in contemporary sociology, however, that it is necessary to see 27
what Weber has to say about this institution and its relationship to 28
the mode of production. The most important question here is that 29
of the relationship between slavery and capitalism. Marx, it will be 30
remembered, had characterized the ancient mode of production as 31
being based upon slavery, just as the ‘feudal’ mode was based upon 32
serfdom, and the capitalist mode upon free wage-labour. What then 33
was Weber’s attitude to the question of the existence of capitalism in 34
the ancient world, to the compatibility of slavery with such capitalism 35
as existed and the relationship between both slavery and capitalism 36
and the closure of social relationships? 37
The basis of Weber’s position on capitalism in the ancient world 38
is that it existed as a possible mode of want provision but that it was 39
not the typical mode. Capitalism did exist, but there was no capitalist 40
system. 41
Perhaps the most crucial distinction of all in Weber’s systematic 42
sociology is the distinction between ‘oikos’ and capitalist enterprise. 43

444 john rex
1 The term ‘oikos’ is taken over by Weber from Rodbertus and Karl
2 Bucher. It refers for Weber to an ideal type of provisioning and of
3 production within a closed group where all production, or nearly all
4 of it, is for consumption within the group and in which the provision
5 of needs is normally arranged within the group. The enterprise on
6 the other hand is directed towards some kind of commercial trans-
7 action and involves the counting of funds before and after a project
8 with a view to profit.
9 In the ancient world capitalism as a system resting upon capital-
10 ist enterprises never finally gained the upper hand over the oikos
11 economy. As Weber says in his Agrarverhältnisse im Altertum such a
12 system excludes:
13
14 ‘All manorial charges levied in rural areas on subject groups like the
15 various tributes – rents, dues and services – extracted from peasants
16 in the early Middle Ages... neither the land owned nor the people
17 subjected can be regarded as capital; title to both depended not on
18 purchase in the open market but on traditional ties.’
19
20 and although ‘There also existed in antiquity the commercial practice
21 of dividing estates and leasing them out’ in this case ‘the land is used
22 as a source of rent and capitalist enterprise is absent’.9
23
24 On the other hand capitalist investment with a view to profit does
25 take place. It finds an outlet in the following limited range of func-
26 tions:
27 ‘(1) government contracts for partial or total collection of taxes and
28 public works; (2) mines; (3) sea trade... (4) plantations; (5) banking
29 and related activities; (6) mortgages; (7) overland trade; (8) leasing
30 out slaves... (9) capitalist exploitation of slaves skilled in a craft.’10
31
32 In one way or another all of these forms of capitalist investment in-
33 volve high risk and the possible use of force. They are non-peaceful
34 ‘adventurer forms which Weber sometimes describes as ‘adventurer capitalism’
35 capitalism’ or ‘booty capitalism’. They never involve the systematic and continu-
36 ous use of labour in a rationally-planned enterprise with a view to
37 making profit through market opportunities.11
38
Thus there is an important difference between the typology of
39 modes of production suggested by Weber and that suggested by Marx.
40 For Marx the economy of the ancient world rests upon slavery and
41 hence represents a non-capitalist mode of production. For Weber, on
42 the other hand, it is possible to suggest that there are certain areas of
43 activity which take a capitalist form, and that this form of capitalism

differs from the prevailing in the modern Western capitalist system. 1

Thus in some ways Weber established a more direct linkage between capitalism and 2
capitalism and slavery than Marx. But he is still eager to point out the slavery linked 3
differences between capitalism as it operates in a number of settings, 4
including that of the slave plantation, and the way in which it oper- 5
ates within a fully-fledged capitalist system. In order to elaborate this, 6
it is necessary to consider further Weber’s view both of the institu- 7
tion of slavery and that of the plantation. 8
It should be noted that Weber does not say that slavery is itself a 9
capitalist institution, only that in the three forms mentioned it may 10
provide a field for capitalist investment. Clearly, in its most elemen- 11
tary form, it is an institution of the oikos, and probably the most fre- 12
quently-found type of slavery is the domestic institution. The house 13
slave is a part of his master’s household and participates in the larger 14
economy only indirectly through his master. It is also possible for a 15
large domain to attach to itself manufacturing and processing estab- 16
lishments, independent of any relationship to the market, and em- 17
ploying slave labour. 18
When slave labour is used to make a profit through the provisions 19
of goods and services for sale in the market, slavery comes to be part 20
of a larger capitalist system, whether the unit which profits from slave 21
labour and its product is a household or a deliberately created enter- 22
prise. Weber makes the point, however, in his study of ancient agri- 23
cultural systems, that slave labour is not essential to these purposes, 24
and that from the point of view of the attainment of the full purposes 25
of capitalism, it is not particularly efficient. He disagrees with L.M. 26
Hartmann in his discussion of the Roman Republic as follows: 27
28
‘L.M. Hartmann observed that in antiquity slavery was necessary be- 29
cause of the burden of army service which was borne by citizens. 30
That is in part correct... However it is also true that such a generaliza- 31
tion cannot explain that which is characteristic of Roman society, the 32
development of large plantations worked by slaves, nor indeed can 33
one deduce from it the necessity of slavery. The situation demanded 34
that yeoman citizens be able to leave their lands to serve the State in 35
politics and war and this need could have been met by other forms of 36
unfree labour: serfdom, share-cropping, helotry and so on.’12 37
38
And argues that the actual emergence of slave plantations was the 39
consequence of a particular social and class situation which emerged 40
after the defeat of the Gracchi. As to the inefficiency of slavery Weber 41
tells us that: 42
43

446 john rex
1 ‘when low sales cause suspension of production not only does capital
2 invested in slaves bring no interest – as is true of capital invested in
3 machines – but also slaves literally “eat up” additional amounts. The
4 result is to slow down capital turnover and capital formation’.13
5
6 There is, moreover,
7
8 ‘a large risk in investing capital in slave labour. This was due first of
9 all to the fact that slave mortality was very high and entirely unpre-
10 dictable causing capital loss to the owner’.
11
12 Slavery provides ‘no basis for reliable cost accounting, the necessary
13 condition for large industrial enterprises based on the division of la-
14 bour’, and:
15
16 ‘Another limitation on the truly capitalist exploitation of slaves as a
17 means of production was the fact that the slave market depended for
18 supply on succesful wars. For full capitalist exploitation of the work
19 force was possible only if the slaves had no families, in fact as well
20 as in law; in other words they were kept in barracks which, however,
21 made reproduction of slaves impossible. For the cost of maintaining
22 women and rearing children would have been a dead ballast on work-
23 ing capital.’14
24
25 For Weber, therefore, even though the slave plantation was a char-
26 acteristic form of ancient capitalist enterprise, it was not compatible
27 with the logic of the more advanced and systematic form of capital-
28 ism which emerged in Europe. Precisely in so far as the modern capi-
29 talist mode of production gained the ascendancy, slavery was likely to
30 free labour be superseded by free labour.
31 So far as plantations are concerned, Weber did in his last lectures
32 see them as essentially a capitalist phenomenon.15 The characteris-
33 tic agricultural unit in most societies is the manor in which serfs
34 hold their land subject to their paying dues to a lord. When such
35 a unit begins to respond to market forces and production switches
36 primarily to the market, the manor is likely to undergo two forms of
37 development. Either the land is divided up between individual farm-
38 ers who keep stock or raise crops for sale while the landlord ‘farms
39 rent’ (this is what Weber calls an estate system), or the workforce is
40 deprived of all freedom and put to work in labour-intensive forms of
41 horticultural production. Mining, which may originally be organized
42 to provide precious metals for the lord’s treasury, or to be yielded
43 up to the lord’s monopoly of external trade, may undergo a parallel

development, miners being engaged in labour-intensive extractive 1

operations paralleling the horticultural occupations of the plantation 2
slave. 3
It seems that, by the time Weber came to give his lectures on 4
‘General economic history’, he was beginning to see the plantation 5
not simply as a backward form of capitalism, but as a capitalist form 6
which transcended feudal forms, particularly in agriculture. This is 7
a question of some importance, as we shall see when we turn our 8
attention to problems of slavery and race relations in the modern 9
period. 10
Apart from the question of the nature of slavery and of the plan- 11
tation as social and economic forms, there is one other theme in 12
Weber which is of importance for contemporary debates about slav- 13
ery and race relations. This is the question of the role of religious and 14
other ideological factors in shaping socio-economic systems. There is 15
no need here to rehearse the well-known debate which Weber initi- 16
ated about the relationship between the Protestant ethic and the capi- 17
talist spirit. What one should notice, however, is that according to 18
Weber there was both a congruence between Calvinist teaching and Calvinist teaching 19
capitalist social institutions, and an incompatibility between slave la- 20
bour and capitalism. It would seem natural to deduce from this that 21
Weber’s theories would imply that slavery would be most prevalent 22
in those Christian countries which had not undergone anything like 23
a Calvinist reformation, and that it would be displaced where there 24
was Protestantism. This is certainly one of Weber’s theoretical be- 25
liefs which is at odds with much opinion about North and South 26
American slavery. 27
Weber wrote comparative economic history and also attempted, at 28
the end of his life, to develop a set of type-concepts in terms of which 29
any socio-economic formation could be analyzed. Thus, although he 30
wrote about the institutions of the ancient and medieval world in 31
Europe, about Chinese, Indian and Jewish civilizations, and about 32
advanced forms of industrial and commercial capitalism, the nature 33
of his concepts is such that they can very fruitfully be applied to mod- 34
ern empires and to the colonial social systems which they generate. 35
It might be said that Weber believed that history repeats itself in bits. 36
In so far as this is so, one might be able to understand the problems 37
of colonialism, of colonial labour exploitation and of race relations in 38
terms of the welding together of ‘bits’ already well-developed for the 39
purposes of the analysis of ancient empires to form new patterns. 40
It is sometimes said that the use of type concepts is defective be- 41
cause it suggests a bitty, mosaic view of historical societies, whereas 42
the nature of these societies is not one which exhibits the mechani- 43

448 john rex
1 cal unity of a mosaic, but rather the unity of a developing organism.

2 There may be truth in this, and certainly Weber’s special skill lay
3 in his ability to take out the bits of social systems, to look at them,
4 and to see that they were not new. But it is also true that, however
5 much Weber may have refused to consider any organic approach, his
6 studies of the ancient world especially do not suggest simply a con-
7 geries of bits. There is a great deal of inter-connectedness between
8 one institution and another, and any careful reading of these writ-
9 ings leaves an impression of an over-all pattern in which the various
10 ancient social types are well distinguished in spirit from modern in-
11 dustrial capitalism.
12 I may now turn to the application of concepts which, after all,
13 were not primarily directed to the study of race relations and see
14 whether they may not be fruitfully applied in those areas of academic
15 interest which centre on modern problems of race relations. The core
16 of this debate concerns the question of slavery and race relations in
17 the Americas the Americas, and with this I shall deal first, but over and above this
18 it will be necessary to look for the over-all pattern or developing sys-
19 tem within which modern race relations problems occur.
20 The crude lines of the debate about slavery turned, in the first
21 place, on an argument amongst North American writers as to wheth-
22 er slavery was in fact the moral scandal which post-bellum politi-
23 cians made it out to be and, secondly, on whether the Latin-American
24 countries had produced a more humane form of slavery and, if so,
25 why?16
26
First, I should deal with the question of the effect of ideology on
27 differing systems. Here the suggestion has sometimes been made
28 that, far from Calvinism, capitalism and free labour being systemati-
29 cally linked with one another, there is a strong relationship between
30 the presence of Protestant religion and the harsher forms of slavery
31 and racism. Racism, some would argue, is a Protestant phenomenon
32 while Catholicism encourages relations between master and man
33 and between dominant and subordinate peoples of a humane if pa-
34 ternalistic kind.
35 In fact it is easy to show that this simple theory generalizes too
36 much. Regardless of religious confession, legal systems or national
37 ideology, there are economic circumstances which will facilitiate the
38 development of a particular rationalized system of plantation pro-
39 duction – what has sometimes been called a factory of the field –
40 and other circumstances which will hinder it. But any simple theory
41 of economic determinism of this kind also generalizes too much. It
42 would be far better to say that particular economic circumstances of
43 demand in the market do favour particular kinds of development,

but that such developments are also affected by ideological traditions 1

and also by historical lag. 2
Perhaps the really interesting thing about the plantation as a so- social institution 3
cial institution is precisely its relationship to other related agrarian 4
forms. Weber was almost certainly wrong about the association of 5
Protestantism and free labour, but he did see that the plantation was 6
a social form which was continuous with various other forms such 7
as the manor, the estate and independent peasant production. It may 8
well be, indeed, that Weber’s approach is more useful in the study of 9
plantation societies than any other, precisely because it uses typology 10
and, by so doing, relates the plantation to other agrarian forms. 11
Once one understands that the plantation may be thought of as an 12
evolutionary development which arises out of the manor, it comes as 13
no surprise that particular plantation systems lie somewhere along 14
a continuum between a pure ‘factory of the field’, using disciplined 15
unfree labour, and a quasi-manorial system in which at least a very 16
large part of the worker’s product is used either to feed himself or 17
to contribute to his ‘lord’s’ table. Nor should one be surprised if, in 18
times when the plantation finds that demand for its product slack- 19
ens, it survives as a kind of a manor with other than purely economic 20
and market-oriented functions or, on the other hand, when what was 21
once a manor becomes more and more like a plantation in times of 22
economic boom. 23
It is interesting to note here that both the manorial polar type of 24
plantation, and the factory of the field type, use slave labour. This 25
would suggest that, whether or not the settler conquerors set out to 26
be seigneurs or lords, the mere fact that slaves were available meant 27
that they were likely to be employed in preference to other forms of 28
labour. In fact, it would seem that workers captured or purchased 29
and transported from their native soil could be exploited more ef- 30
fectively and efficiently than either wage workers moving from the 31
metropolis to the colonies, or the true natives whose ties to their own 32
native soil made them less exploitable. One should not jump to the 33
conclusion, however, that where there is conquest there will always 34
be the same economic forms, for a slave’s subsequent destiny would 35
depend both upon the social and legal structure in which he found 36
himself (e.g. in a manor or a factory of the field), upon the way in 37
which pressures were placed upon that particular productive institu- 38
tion by changing market forces. 39
The debate about slave plantation systems should, however, be 40
made more comprehensive by being extended in a number of dif- 41
ferent directions. In the first place it should be noted that the planta- 42
tion, whether in its more manorial or more capitalist factory form, is 43

450 john rex
1 only one of a number of related types of productive and exploitative

2 systems. Others include the Spanish ecomienda, under which native
3 people were assigned to conquerors for their use as workers, without
4 necessarily being tied to a place; latifundia, or large estates which
5 engulfed some peasants and made then subject to labour service or
6 to rent payments; minifundia, or small peasant holdings whose econ-
7 omy was such that peasants were forced into debt and into migration
8 to towns or to plantations; ‘reductions’ in which native people were
9 in part separated, protected and educated, and in part made available
10 for exploitation; and also all possible intermediate forms.
11 If one regards the Americas as a kind of historical laboratory for
12 the analysis of labour and productive systems, it is interesting to no-
13 tice that the plantation system throve with imported African labour,
14 whereas the estate system was widespread in areas where Spanish
15 conquerors subdued an Amerindian population. Miscegenation was
16 more frequent in the latter areas and in some of them the domi-
17 nant group was subject over a few generations to considerable mes-
18 tization. But one cannot conclude that harsh conditions, restriction
19 of intermarriage, and a descent rule which made the offspring of
20 miscegenation part of the subordinate group always occurred where
21 African slave labour was employed in plantations. There was in fact a
22 considerable difference between plantation systems, and the kinds of
23 contact and intermarriage between the races which they allowed. It is
24 different plantation this which led to the all-too-crude statement of the differences which
25 systems existed between North and South American slavery. The real prob-
26 lem, however, is to discover the conditions under which different
27 patterns of slavery and different patterns of race relations emerged.
28 Even if one confines oneself to the question of different types of
29 slavery, however, the problem is not an easy one. A ‘type of slav-
30 ery’ includes all kinds of economic relationship, legal and customary
31 forms, as well as ideological factors, and it may well be that varia-
32 tion in one respect does not necessarily imply variation in another.
33 Moreover, if one tries to pursue a causal analysis in terms of depen-
34 dent and independent variables, one quickly finds that it is very dif-
35 ficult indeed to be sure whether a particular factor, say a particular
36 law or custom, should be counted as an independent or a dependent
37 variable. In a causal analysis of this kind one often finds that par-
38 ticular features tend to crop up in both columns. It may well be that,
39 at this point, Weber’s resolve in his methodological introduction to
40 Economy and society to pursue both causal analysis and analysis in
41 terms of meaning breaks down, and that the best that one could hope
42 for is a hermeneutic analysis of the form of life exhibited in each
43 particular variant of plantation society.

Thus it would seem that a comparative analysis of plantation 1

slavery itself involves grave difficulty, and any more far-reaching at- 2
tempt to compare modern forms with those which prevailed in the 3
ancient world is even more difficult. Nevertheless, the bitty nature 4
of Weber’s analysis here has its merits. It is clear from his consid- 5
eration of slavery in the ancient world that he would not be content 6
with comparisons between units loosely defined as slave plantation 7
systems. Such units can be broken down into a number of separate 8
institutions and, when they are, much more fruitful comparisons 9
become possible. It is interesting here to compare what would be 10
involved in a Weberian theoretical and comparative analysis of slave 11
systems with the best work currently being carried out by Genovese 12
and others. Penetrating though these studies are, one can usefully 13
ask whether the Marxian concepts which Genovese derives from 14
Hobsbawn are not too general and imprecise.17 Weber’s analysis of 15
the agrarian systems of antiquity appears far more precise and de- 16
tailed in the concepts which it uses than does the Genovese analy- 17
sis of modern slave systems. In the last analysis, Genovese’s merits 18
are those of a historian and Weber’s those of a theoretical sociolo- 19
gist. These empirical problems could be even better illuminated by a 20
scholar who combined the merits of both of them. 21
The debate to which I have been referring confines itself to 22
the economic relations between European settlers, Amerindians 23
and African slaves in largely agriculturally-based societies in the 24
Americas. One should now also note that the basic patterns of the 25
plantation system of the factory of the field type are reproduced with 26
other forms of unfree labour, that the labour is imported from other 27
countries and cultures, and that it may be applied, not only in ag- 28
riculture, but in mining and to some extent in the manufacturing 29
industry. 30
It has already been seen that, Weber’s reservations about the ulti- 31
mate incompatibility of capitalism and slave labour notwithstanding, 32
slave plantations did for long periods exist as highly efficient capital- 33
ist undertakings. One can say this even if one does not accept the 34
full thesis of Fogel and Engermann in their recent work Time on the 35
Cross.18 Indeed, one can find in the literature examples not merely 36
of capitalist efficiency in plantations, but of detailed cost account- efficiency 37
ing coupled with a good dash of the Protestant ethic. One particular 38
example of this is to be found in Pares account of a plantation-own- 39
ing family, the Pinneys on the Island of Nevis.19 When this plan- 40
tation passes into the hands of the generation most possessed by 41
the Protestant spirit one finds that detailed accounting is applied not 42
merely to the plantation as a capitalist enterprise but to the disposi- 43

452 john rex
1 tion of human resources within the family household. Nevertheless,

2 there are difficulties in making slave labour fully efficient in chang-
3 ing conditions and some of these difficulties are circumvented in the
4 labour forms which follow the emancipation movement.
5 The emancipation of slaves did not of course occur solely for ide-
6 alistic reasons. Its idealism was able to gain a hearing only because
7 there were more efficient ways of carrying on production. In the case
8 of the sugar plantations of Guyana, as well as in large areas in which
9 the plantation system flourished in the East, slavery was superseded
10 by the indentured labour system. This was not designed as a moral
11 improvement on slave labour, but as a system which would in fact
12 enable the employer to enjoy the advantages of slavery with none of
13 its disadvantages. Thus, the labourers were formally more free than
14 slaves and the term of their employment was limited, but this meant
15 a much greater calculability of costs was possible for the employer.
16 At the end of the contract he had no incalculable obligation to the
17 worker, and the worker indeed often found that his right to return
18 to freedom at home was theoretical rather than real. He would thus
19 be available at first in semi-slave conditions, and then as a depen-
20 dent and impoverished tenant or share-cropper. Such was the fate
21 of Indian ‘coolie’ workers in Fiji, Malaya, Assam, Ceylon, Mauritius
22 and Natal, as well as in Guyana, where they literally took the place of
23 slaves.20
24
Even more important though for the sociology of colonial society
25 shorter-term has been the use of shorter-term contract labour in Africa. Its ideal
26 contract labour form is to be found in South Africa and in the mines, where impov-
27 erished workers from the overcrowded rural reserves are separated
28 from their families in barracks called compounds, are paid wages
29 appropriate only to such barrack conditions, and are returned at the
30 end of a nine-month contract period to the reserves which employers
31 cynically describe as a kind of social security system. As I have said
32 elsewhere, the South African labour system is probably the most effi-
33 cient system for the capitalist exploitation of labour yet devised, rest-
34 ing as it does on the three institutions of the rural reserve, the mining
35 compound and the controlled urban ‘location’.21 And if such a system
36 exists empirically in an almost ideal typical form in mining, it is ap-
37 plied also in its essentials in manufacturing industry and in settler
38 agriculture. As one famous settler in East Africa put the matter: ‘We
39 have stolen his land. Now we must steal his limbs... Compulsory la-
40 bour is the corollary of our occupation of the country.’22 In the same
41 territory, a Government commission nicely grasped the problem of
42 African labour conditions when it discussed the housing of urban
43 workers not in terms of housing units but in terms of ‘bedspaces’.23


It is clear from these examples that one cannot discuss the colo- 1
nial societies established by the European powers in terms of a slave 2
and a post-slave period. And one is compelled to ask, as Weber did, 3
whether the most significant aspect of the problem is the existence of 4
slavery, since slavery is one means of achieving ends which may also 5
be achieved through a variety of alternative forms of unfree labour. 6
Nevertheless, there is a problem of very great importance to the his- 7
tory of race relations which arises from slave emancipation. 8
It should be noticed that, thus far, I have not discussed ‘racism’ as ‘racism’ 9
an element in economic situations. Weber himself finds it possible 10
to discuss similar problems in the ancient world by referring only to 11
the distinction between closed and open social relationships. Thus, 12
a group which monopolizes any type of economic opportunity by ex- 13
cluding outsiders tends to find some rationalization and justification 14
of its actions by drawing attention to certain observable characteris- 15
tics of the excluded group. But since Weber is so much concerned 16
with the central problem for him of the rationalization of capitalist 17
individualism in Calvinism, he has little to say about the kind of ide- 18
ologies which might justify the exploitation of one group by another. 19
The striking thing indeed about his account of Roman plantation 20
slavery is that he does not see the system as justified in terms of any 21
elaborate ideology about racial or any other group difference. 22
Was it then the case that slavery and other harsh forms of political 23
oppression and economic exploitation existed in the Roman Empire 24
without the phenomenon of racism making an appearance, but that 25
slavery in the modern period was associated from the first with rac- 26
ism? This is by no means an easy question to answer and the lines of 27
the debate are often very confused indeed. Much depends upon what 28
is meant by racism. 29
It should be clear that nearly every group in modern times which 30
was engaged in colonial conquest and exploitation found justification 31
for its practice in abusive accounts of the exploited group. Charles 32
Boxer, for example, has demonstrated24 that, however much the 33
Portuguese might be Latin and Catholic, their settlers are on record, 34
in Church as well as secular contexts, as abusing the native people of 35
the Portuguese Empire. From such evidence many liberal scholars 36
over-react by saying that one colonialist is much the same as another 37
and that, whatever their culture and religion, they are all in the end 38
not merely exploiters and oppressors but racists. Against this, one has 39
to set the long record of influential clerics, particularly in the Spanish 40
territories, arguing against the exploitation through enslavement of 41
the native peoples of America. 42
There can be little doubt that in the period of slavery and other 43

454 john rex
1 early forms of violent colonial conquest and oppression many racist

2 relation to property ideas were canvassed. What is perhaps more important, however, is
3 that the structure of society depended upon the relations which men
4 had to property. Whether they freely owned or were tied to their land
5 with an obligation to labour services, whether they owned other work-
6 ers, or whether they were themselves property, so long as there was
7 a law which upheld such forms of inequality and sanctions to back
8 it, the inequalities in the social structure did not depend upon the
9 system of racist beliefs which were canvassed. The interesting thing
10 from our point of view is that when the social order could no longer be
11 buttressed by legal sanctions it had to depend upon the inculcation in
12 the minds of both exploiters and exploited of a belief in the superior-
13 ity of the exploiters and the inferiority of the exploited. Thus it can be
14 argued that the doctrine of equality of economic opportunity and that
15 of racial superiority and inferiority are complements of one another.
16 Racism serves to bridge the gap between theory and practice.
17 This is not, of course, to say that the use of force ceases with slave
18 emancipation. In some countries like South Africa it is systematically
19 mobilized on a political level to ensure continued white supremacy.
20 But it is to say that when inequality, exploitation and oppression are
21 challenged by economic liberalism, they have to be opposed by doc-
22 trines which explain the exceptions to the rule. While it is admitted
23 that all men are equal, some men are deemed to be more equal than
24 others.
25 Doctrines of racial equality and inequality, and practices asso-
26 ciated with them, are already worked out in slave societies before
27 emancipation, for in all there are some free men or men whose af-
28 filiation to either the plantation-owners or the slaves is ambiguous.
29 These include freed slaves, the offspring of miscegenation, and poor
30 whites whose standard of life approximates to that of ‘coloureds’ even
31 though they share the skin colour and other characteristics of the rul-
32 ing group. The way in which these groups are related, in terms of
33 status and life-changes to the plantation-owners and slaves, prepares
34 the way for social stratification after emancipation. One possibility is
35 that, from the point of view of status ideology, the crucial line will be
36 that between whites and poor whites. Another is that there may be a
37 status ordering of the society which overlaps with a racial or colour
38 ordering, so that there is a continuous status gradation in which,
39 roughly speaking, white or lightly-coloured people are at the top and
40 completely black people at the bottom. Finally, there is the possibility
41 that there will be three estates of White, Coloured and Black.
42 Any incipient ordering of groups and individuals in terms of a
43 status order of this kind has to bear a greater strain when the legal

basis of inequality provided by slavery, as well as by legal estate sys- 1

tems, breaks down. In this situation everyone is theoretically equal, 2
but a new status structure emerges in which race or colour is a cru- 3
cial indicator of a man’s position. Even if it is at odds with the actual 4
social relations of production prevailing in the new social order such 5
a system will have some influence. But frequently, the status system 6
and the economic order move into line when the society as a whole 7
is deeply dependent upon production for the market, because high- 8
status, lightly-coloured people tend to close off economic opportuni- 9
ties to those lower down the scale and because money, education, 10
and economic position are all said to ‘whiten’ the individual who 11
possesses them. Moreover, since post-emancipation society is often 12
economically stagnant, the status order may be the main structuring 13
fact in the social order. In Barbados shortly before independence, for 14
example, over-all stagnation and poverty were coupled, particularly 15
among whiter, higher-status people, with a continuous preoccupa- 16
tion with matters of coulour and status. 17
Weber would certainly have had much to say about the structure 18
of colonial societies had he turned his attention to them because, 19
although he shared with Marx an interest in social class (which he 20
saw as a matter of groups with common or differentiated interests 21
in a market situation), he supplemented this with a theory of sta- status group 22
tus groups, differentiated in terms of styles of life and a consequent 23
differential apportionment of honour.25 He also envisaged the pos- 24
sibility that a status group distinguished by its specific life-style 25
might come to exercise hegemony over the society as a whole. Thus 26
the Mandarins had imposed their way of life on ancient China, the 27
Brahmins on India and the bourgeoisie (considered here as a status 28
group rather than a class) on Western European society. He would 29
therefore have had no difficulty in understanding a situation in which 30
an ethnic group achieved hegemony. He might only have added to 31
this a Marxian type of scepticism, suggesting that the claim made in 32
terms of style of life was, in part at least, a cover for the closure of 33
economic opportunities. 34
Weber might, it is true, have had more difficulty in understanding 35
a social order in which there was not so much status domination by 36
a particular group as a status grading of individuals. Oliver Cromwell 37
Cox makes a useful contribution here in distinguishing what – using 38
the term in a peculiar sense – he calls ‘social class’ as distinct from 39
caste, estate and political class.26 For Cox, this social class is a con- 40
ceptual system in terms of which individuals rate themselves against 41
others rather than a closed form of social grouping. 42
So far, however, the focus in dealing with colonial societies in 43

456 john rex
1 which race relations problems emerge has been a narrow one. I have
2 dealt basically with the relationship between exploiting owners, com-
3 ing to the colony from the metropolis, and the exploited workers or
4 peasants, who are either colonial natives or imported slaves, together
5 with those who have some relationship of descent with either of these
6 two groups. This much can be comprehended in terms of a fairly
7 simple model of an economic order, coupled with a status system
8 which, in essence, has two poles. One of these is represented by the
9 group of owners coming from the metropolis. The other is that of the
10 major group of workers, slaves or peasants of colonial or imported
11 origin. But few societies are as simple as this, because pluralism can
12 be found among the exploited as well as among the colonialists, and
13 there are other groups who have no ethnic and economic affiliation
14 with either colonizers or colonized.
15 pluralism First, one should notice the pluralism which comes from the di-
16 vision between workers in plantations, mines and factories on the
17 one hand, and peasants on the other. Some societies are dominantly
18 plantation societies and some are predominantly peasant societies,
19 but in the former there are likely to be a minority of peasants, and in
20 the latter a minority of workers in plantations, mines and factories.
21 Very often the minority and the majority will be ethnically distinct. In
22 a society which offers little scope for independent subsistence farm-
23 ing, ‘peasants’ will either be forced into being migrant labourers in
24 plantations or towns, or they will be pushed to the margins of the
25 society to carry on their segregated way of life in conditions which are
26 ultimately insupportable. On the other hand, in societies in which
27 there is a predominantly peasant population, urban industrial work
28 and mining, as well as work on occasional plantations, may be car-
29 ried on by specially imported workers or by ethnic minorities. In
30 both of these cases we have ethnic and occupational differentiation
31 combined with differences in status.
32 The second extremely important alternative is that in which slaves
33 of one race or ethnicity are replaced by indentured workers from an-
34 other. In this case, cultural pluralism amongst the working people of
35 the colony coexists with an ambiguity as to the relative status of the
36 two groups. Guyana, for example, is what some think to be the classic
37 case of a plural society, in which there are both Indians descended
38 from indentured workers, and the descendants of African slaves. It
39 is true, of course, that because of their differing history, these groups
40 have their own distinct sets of domestic institutions and that they do
41 not therefore amalgamate culturally, and it is also true that it is hard
42 to place the two groups in terms of status in a horizontal sense, since
43 one enjoys the advantage of having been recruited on theoretically

less arduous terms, but the other group has now very largely left the 1
plantation for the town and is culturally closer to the ruling group. 2
This is an important type of intergroup situation in many colonial 3
territories, and it is sometimes complicated by the fact that one of the 4
two colonial groups is in a permanent majority (this is not the case in 5
Guyana). 6
M.G. Smith has used the model of this particular intergroup re- 7
lationship as a general model for colonial societies in the Caribbean 8
and in Africa.27 He argues that it is usually better to conceive colonial 9
societies as plural rather than as horizontally stratified, that the dif- 10
ferent plural segments have no institution in common save the po- 11
litical one, and that the society as a whole is held together by the con- 12
trol of the political institution by one segment. This theory is by no 13
means accepted here, since we have made it clear that there are both 14
status orders and economic systems which bind groups and individuals 15
in colonial societies together, and that what we are dealing with here is 16
pluralism amongst the exploited workers and their descendants only. 17
Nevertheless there are circumstances in which such groups contend 18
for political power, when the political domination of the colonialists 19
is withdrawn, and the struggle may not merely be two-sided, as in 20
Guyana. It may have three sides, as in Malaya, or conceivably even 21
more. In these circumstances, it may well be that a struggle for pow- political power 22
er becomes the central structuring theme in the post-colonial world. struggle 23
But it would be misleading, even in these cases where such an ethnic 24
political struggle is evident, and still more so in others where it is not, 25
to underestimate the binding force of economic institutions which 26
are by no means necessarily displaced with the coming of political 27
independence. What Smith seems to have done, at least in his earlier 28
writing, is to over-emphasize the importance of one structuring fea- 29
ture of one kind of society and suggest that it is the basis for a general 30
theory. 31
Divisions among the exploited workers, such as those we have 32
been discussing, are by no means the only other structuring features 33
of ethnically-plural colonial societies, for such societies quite com- 34
monly also include a number of other elements. The most important 35
of these are the pariah traders and the settlers, though we should also 36
give some consideration to two other groups from the metropolis, 37
namely, the missionaries and the governmental administrators who 38
remain, to some extent, culturally and socially as well as functionally 39
separate from other colonialists. Such distinctions would fit naturally 40
into a Weberian sociology of the colonies, since Weber recognized 41
that functional differentiation, not necessarily of a simple economic 42
kind, did in fact generate what he loosely called class struggles in the 43
Roman Republic and the Roman Empire.

458 john rex
1 It seems to be almost universally the case that colonial societies

2 provide subsidiary economic opportunities which major colonial
3 owner-producers are unable, or unwilling, to exploit. Theoretically,
4 there could be a pure plantation or manorial economy which sat-
5 isfied all its own needs internally – a complete oikos economy.
6 Normally, however, trade in food, household equipment and other
7 commercial items falls to groups of outsiders, who have no aspira-
8 tion to political control, or to direct exploitation of the colonial work-
9 ers through production, and are prepared to live solely through trade.
10 Such groups tend to be despised by the major colonial proprietors
11 and to be regarded with suspicion by the workers themselves. Very
12 easily, in times of difficulty, they can become a scapegoat group. In
13 those cases where the scale of their economic activities is large the
14 community which grows up around such trade may become a not
15 insignificant structural element of the society. This is true, for in-
16 stance, of the Indian traders in East Africa as it is of the Chinese in
17 Malaysia. Weber would have recognized the similarities here with
18 the role of the Jews in medieval Europe and of the metics in Ancient
19 Greece.
20 Quite commonly these groups of traders have an ethnic affiliation
21 with groups of immigrant peasants and workers in the colonial terri-
22 tory. This is the case of Indian traders in Guyana or Natal, and also of
23 the Chinese traders in Malaysia. The existence of such a community
24 increases the power of the traders in political terms, and they often
25 have political skills to offer to workers of their own ethnicity. One
26 consequence of all this is that the potentiality for class politics even
27 amongst an ethnically separate group of workers is undermined by
28 cultural unity the formation of strong ties of cultural unity and of clientage across
29 class boundaries. This adds to the appearance of pluralism in the
30 society, but also affects the group image which a particular group
31 may have. Thus, attitudes towards all Indians in East Africa may be
32 influenced by the image of the Indian trader or moneylender. It is
33 perhaps not too much to say that, next to the tension between the
34 main owning and exploiting groups, this is the major source of racial
35 tension in colonial societies.
36 So far, however, I have not spoken of the group who give the char-
37 acteristic shape to one major type of colonial society, namely, the
38 ‘white settlers’. This category should not be taken to include the own-
39 ers of the major means of production, who appear to the settlers as
40 a kind of plantocracy which often operates on an absentee basis. Nor
41 should it be taken to include administrators and missionaries. It does
42 include farmers, who are either able to occupy land on cheap terms
43 following military conquest or participate directly in that conquest. It

is characteristic of these settlers that they see the colonial situation 1

as providing them with opportunities for economic acquisition and 2
for status not available to them in the metropolitan country. This is 3
obviously true of the farmer settler who very quickly behaves like a 4
European lord on his manor. But it is also true of the worker, who 5
recognizes that he might do easier manual work aided by a native 6
assistant, and that he will be paid more for his skills because of his 7
ethnic affiliation to the plantocracy and the administrators. If there 8
are sufficient of these settlers a separate settler economy is likely to 9
develop, separate both from that of the plantocracy and from that of 10
the metropolitan country. There may be some divisions of interest 11
among these settlers but, in general, alliances of a political kind will 12
be possible among farmers, capitalists and urban workers. Again, 13
class affiliations are transcended by ethnic ones although, in this 14
case, the interlocking of interests of the separate groups of workers 15
makes it possible for them to be seen simply as the settler interest for 16
most purposes. 17
In some cases the settlers eventually won political power for 18
themselves, and were able to develop wholly new nations capable 19
of standing up, and more, to the European nation States. This was 20
most clearly the case in the United States of America where two 21
crucial wars were fought, one against the mother country, and the 22
other against the colonial plantocracy. Something similar also hap- 23
pened in South Africa and in the settler territories of Latin America. 24
Where numbers were smaller, however, as in Algeria, and in East 25
and Central Africa, the settlers fought unavailingly to succeed to po- 26
litical power in the post-colonial period and were forced, in the long 27
run, to emigrate or to form such alliances as were necessary for their 28
survival. 29
If the American settlers defeated the plantocracy and the South 30
African settlers the Uitlanders who opened up the Rand, they, how- 31
ever, by no means put an end to racist theory and racialist practice. 32
Indeed, they had more reason in some ways than the planters pro- 33
tected by law and custom, for adopting racialist and exclusivist prac- 34
tice; Van den Berghe is perhaps correct here when he speaks of such 35
situations as competitive rather than paternalist.28 Certainly the New 36
Jerusalem which the settlers were seeking could be most obviously 37
obtained only by finding ways of excluding competitors. Thus, the 38
very essence of white settlerdom lay in its capacity to monopolize monopolization 39
jobs, land, commercial, industrial and domestic property. In some 40
ways, settler society therefore produced the extreme example of racial 41
domination. 42
The presence of missionaries in colonial territories is a recur- 43

460 john rex
1 rent phenomenon, but it is ultimately a mistake to seek in any of the

2 religions produced by the Christian Reformation, or by traditional
3 Catholicism in Europe, the ideological source of inspiration for colo-
4 nial practice relating to the equality or inequality which should pre-
5 vail between individuals or groups. In fact, these new societies had
6 to work out their own ideologies, and what the missionaries were
7 often doing was to provide a setting in which new and appropriate
8 ultimate value concepts could be worked out. Differential theologi-
9 cal and ethical starting points did no doubt make some difference
10 to what occurred, but what gave them their life was their capacity to
11 accommodate wholly new patterns of social relations.
12 Finally, one should notice that the cadre of colonial administra-
13 tors had their peculiar role to play. They did not represent the interest
14 of particular owners or particular groups. Rather they were there to
15 ensure that the normal range of colonial operations could be carried
16 on without the metropolitan government being dragged into unnec-
17 essary conflict. Thus, even though they, like the missionaries, did
18 often side with the interests of the plantocracy or the settlers, they
19 were to some extent bound to continue to hold the ring so that the
20 colonial game might be played. As a consequence they sometimes
21 seemed to be a kind of estate or caste or class apart, separated from
22 their kinsmen who had more direct economic interests than they did.
23 How then shall we view the new colonial societies brought into
24 being by the expansion of Europe overseas from the fifteenth century
25 onwards. Are they caste societies? Are they estate systems? Can they
26 be understood in Marxian terms, as based either directly on a class
27 struggle, or as resulting from a changing series of relationships con-
28 sequential upon European need for the accumulation of capital, for
29 raw materials, for markets or for the export of capital? In fact they
30 are none of these purely and simply. Nor do they represent some
31 new form of colonial stratification system distinct from Indian caste,
32 medieval European estates or modern European class struggle. The
33 truth is that they were not simple determinate economical systems
34 capable of being understood through the use of some simple theoret-
35 composite entities ical key. They were composite entities, very like the Roman Empire
36 in their structural complexity and diversity. They have, of course, in
37 various ways, been affected by the changes and developments at the
38 centre of European capitalism, but whether they are needed or not
39 by that capitalism, these societies have come into being and stay in
40 being. It is within them, moreover, that the main problems of race
41 relations in the modern world have their origin. These problems de-
42 rive, as Weber saw, in the first place from the closure of relationships
43 to protect and enhance economic opportunities. But long after the

original closures and their accompanying privileges had ceased to be 1

significant such problems and conflicts continued. The black descen- 2
dant of a slave in a settler or a metropolitan society finds themself 3
still marked with the stigma of slavery; the African peasant in South 4
Africa finds himself kept in reserves for the purposes of ultra-exploi- 5
tation, and excluded from social and economic and political partici- 6
pation on grounds of race; middlemen minorities of traders now find 7
themselves the scapegoats for the ills of colonialism, and are threat- 8
ened with expulsion and exclusion; and, in the wake of political colo- 9
nialism, ethnic communities vie with each other in the struggle for 10
political power. The diverse problems of race relations are now at the 11
head of the world’s political agenda. They are all better understood in 12
terms of the sort of sociological theory of colonialism which has been 13
sketched here, and which applies some of the concepts and the style 14
of sociological analysis to the modern colonial situation that Weber 15
applied in his comparative and historical studies.29 16

So far, however, I have dealt only with problems of race relations 17
between groups who constitute colonial societies. But the full under- 18
standing of race relations problems in the contemporary world must 19
also include, over and beyond this, the study of the relations between 20
these complex colonial structures, on the one hand, and the metro- metropolitan 21
politan economy on the other. This should include the study of the economy 22
relationship between metropolitan societies and colonies, and the 23
neo-colonialist period which often follows political independence, on 24
the one hand, and the migration of workers from the colonies to the 25
metropolis on the other. 26
In considering the relations of metropolitan capitalism with co- 27
lonial territories, there is a difference of emphasis between Weber’s 28
approach and that of Marxists and neo-Marxists. The latter would see 29
colonial societies as, successively, the source of primitive accumula- 30
tion, the source of raw materials, an area for the expansion of markets 31
and, finally, a means for the export of capital. More recently, under 32
the influence of A.G. Frank30, Marxism has seen the so-called process 33
of economic and social development in the Third World as a pro- 34
cess of the development of under-development. A number of other 35
scholars have suggested that, in the study of the modern world, there 36
is only one unit within which studies can be adequately organized, 37
and that this is a single world capitalist system31. The emphasis of all 38
these studies moves the traditional locus of sociological interest from 39
the study of structures of social relations and groups to the study of 40
political economy. It is, of course, a part of the Marxist theory to sug- 41
gest that the dynamics of economic change and revolution are to be 42
found in a process of class struggle, but it is well known that, after 43

462 john rex
1 he turned his attention to the theory of capitalist crisis, Marx did not
2 adequately return to the problem of the structuration of classes. It is
3 certainly true of neo-Marxist sociology of development that it fails to
4 deal adequately with this theory. If any consideration is given to the
5 question of the social formations which will lead to the overthrow of
6 capitalism, the assumption appears to be that the process of capital-
7 ist exploitation in the Third World will go on until the world system
8 of capitalism is ended by the action of the urban industrial working
9 class in the most advanced countries.
10 Now, there are some respects in which Weberian sociology is
11 lacking when it comes to the study of the economics of imperial-
12 ism. Weber writes as an economic historian and sociologist rather
13 than as a political economist, with the result that his ideas on the
14 accumulation of capital, the search for raw materials, the process of
15 capital export, and so on, can only be gleaned from remarks which
16 he makes en passant about particular historical episodes. In these he
17 Marxism seems to adopt a quite cynical practical Marxism, taking as the main
18 assumption that men seek profit and booty where they can. On the
19 other hand, what is striking about Weber’s work on the ancient world
20 is that he describes these processes of conquest and capitalism as be-
21 ing far more inhuman than those which occur in circumstances of
22 advanced capitalism. The key to his thinking here lies in the notions
23 of non-peaceful adventurer and booty capitalism.
24 On the level of the study of social structure rather than that of
25 political economy, these ideas are of some importance. The economic
26 institutions which arose in the course of European imperialism in-
27 volved not simply logical and necessary developments arising out of
28 the capitalist system but a regression to the economic forms of boo-
29 ty capitalism which Weber had studied in the Roman Empire. The
30 Marxian tendency to see these institutions as mercantilist, feudal, or
31 in some other way at odds with capitalism, misses the point here. The
32 crude processes of conquest and exploitation in Latin America, Africa
33 and Asia are capitalist processes, but they belong as structures un-
34 der the heading of booty capitalism. Characteristically, the major eco-
35 nomic institution for colonial development is the chartered company
36 which permits it to gather the revenue within a territory, to govern it,
37 and to pursue monopolistic trading activities within it. This involves
38 a licence to use force against the population, and to find labour for
39 economic enterprises, not through the labour market, but by some
40 non-peaceful means. This is a high-risk capitalism, as Weber pointed
41 out, but it is also a capitalism which is capable of unrestrained ex-
42 ploitation. Thus Elkins is essentially right when he speaks of North-
43 American slave plantations as working according to the dynamics of

unopposed capitalism32. But it is precisely this absence of restraints 1

on the use of force which distinguishes booty capitalism from the 2
routine capitalism of advanced industrial societies. 3
Use of some such term as booty capitalism to describe a set of booty capitalism 4
institutions of economic exploitation gives us an ideal type, in terms 5
of which typical patterns of colonial social structure can be under- 6
stood. We should expect that where capitalism of this kind prevails 7
we shall have chartered companies with a licence to gather revenue, 8
to govern, and to engage in monopolistic trade; we shall have large- 9
scale estates, with servile labour provided by squatters; we shall have 10
plantation systems; and we shall have the complex of institutions 11
which characterize migrant labour in Southern Africa. On the other 12
hand, we should expect a dualism in socio-economic structures as 13
such societies develop. New economic institutions will arise in the 14
towns, or among new metropolitan entrants to the colonial economy, 15
and these will have more of the characteristics of routine capitalism 16
based upon the calculation of market opportunity. 17
It is true that Weber, committed as he was in economics to mod- 18
ern marginal utility theories, did not see the routine forms of capital- 19
ism as problematic, and that he probably over-emphasized the differ- 20
ence between the ancient imperial forms of capitalism and those of 21
modern Europe. Nevertheless, there are differences here, most nota- 22
bly in the kinds of political structures and movements which are the 23
consequence of booty capitalism. Moreover, it cannot be assumed 24
that all the main features of this earlier socio-economic form have 25
now been eliminated. They constitute the very centre of some of the 26
most advanced capitalist economies in formerly colonial territories 27
such as South Africa, with the result that the political sociology of 28
these territories cannot be comprehended in terms of a simple politi- 29
cal sociology of modern capitalism. Some Marxists have indeed rec- 30
ognized a similar point when they notice that the concept of ‘primi- 31
tive accumulation’ has to be extended to take account of ‘permanent 32
primitive accumulation’. 33
The simple political sociology to which I refer assumes that the 34
major line of political development in a capitalist society is towards 35
a polarization of classes and to the emergence of a united working 36
class which, according to the Marxist alternative, carries through the 37
revolution against capitalism and, according to the liberal and so- 38
cial democratic alternative, establishes a new social contract for the 39
working class within a welfare State. These alternatives may be avail- 40
able for the workers within the metropolitan economy and within 41
the modernized sectors of the colonial economy, but they are not 42
available to the workers within the booty economy which still pre- 43

464 john rex
1 vails in many areas. In these circumstances, it is not to be expected

2 that all exploited workers will unite to defend themselves within an
3 urban and trade-union based movement. Rather the workers who
4 have experienced the political reality of booty capitalism will grapple
5 with the problem of power, and may even find it necessary to oppose
6 the Marxist or social-democratic movement in the more advanced
7 sectors of capitalism, just as that movement may well act against it as
8 it becomes compromised in the imperialist development of the me-
9 tropolis. These reflections are borne out by the emergence in recent
10 theory of the Third years of a theory of the Third World revolution which sets itself up
11 World revolution against Marxism. It is expressed particularly sharply in the writings
12 of such writers as Fanon33, Debray34 and Segal35. It cannot be said that
13 this theory itself has Weberian origins. But once the sharp distinc-
14 tion between the institutions of booty capitalism and those of normal
15 capitalism is understood, the possibility of two parallel revolutions
16 can very well be envisaged. And the greater violence of the colonial
17 revolution is likely to produce a need for simple definitions of in-
18 group and out-group which are readily provided by racist theories.
19 Thus, soldiers from the metropolitan countries, sent to repress the
20 colonial revolution, define their enemies racially, and Segal, at least,
21 goes beyond Fanon in seeing the revolution of the Third World as a
22 whole, not as simply a struggle of ‘the wretched of the earth’, but as
23 race war.
24 Similar considerations arise in connexion with the study of co-
25 lonial migrant workers and their families in the metropolitan coun-
26 tries and, again, if Weber did not write about these problems directly,
27 certain of his positions on the structure of the advanced capitalist
28 societies suggest what his approach would have been. On the one
29 hand he was very conscious – as one engaged in German politics in
30 the late nineteenth and early twentieth century – of the importance
31 of the German Social Democratic movement and of the emergence
32 of a Welfare State. On the other, his general notion of the closure of
33 social relationship to protect economic opportunities would have led
34 him to understand that that movement, and its related organizations,
35 would not readily be sympathetic to immigration by less privileged
36 and potentially more docile workers.
37 In fact all of the advanced industrial societies have found it neces-
38 sary to look for supplementary sources of labour, particularly to fill
39 vacancies in arduous, inconvenient, or dirty work. This is in part due
40 to the fact that employers consciously look for docile and cheap la-
41 bour, and that the most obvious place to look for this labour is in the
42 colonial world and in the more backward European countries. But
43 it may also be the case that rising standards among workers in the

advanced countries simply make them unwilling to take on such jobs 1

at any price, and that any advanced economy would need immigrant 2
labour, at least until jobs of this kind could be eliminated by techno- 3
logical advances. 4
From the point of view of the immigrant worker, the ideal situ- 5
ation is one in which he gains entry to the ‘closed relationship’ that 6
trade unions and Social Democratic parties have established. From 7
the point of view of the metropolitan worker, however, there is some 8
ambiguity: if the workers are admitted he would prefer them to be 9
inside his organizations rather than as ‘unfair’ competitors outside; 10
conversely, even if the workers join unions there is a danger of over- 11
supply. Everywhere where workers face this competition, one might 12
expect racial grounds to be offered, among others, for exclusion but, 13
in the case of those countries with a direct experience of the colonial 14
situation of the kind we have been discussing, the attraction of racist 15
definitions may be overwhelming. Thus, although such a develop- 16
ment is not inevitable and can be prevented by democratic political 17
planning and by education, it must be expected that every advanced 18
capitalist country employing immigrant labour from formerly colo- 19
nial territories is likely to see the emergence of racist movements. racist movements 20
Even though there may be some passage of immigrant workers and 21
their children into full participation in working-class organizations, 22
it is likely that the obstacles to entry to such organizations, and to the 23
acquisition of trade union and welfare privileges, may well lead the 24
immigrants themselves to define their conflict with society in racial 25
terms. Naturally enough, too, some of them will see their struggle 26
as immigrant workers as part of the revolution of the Third World. 27
Political ideas of this kind are already widespread in the United 28
States and in Britain, and they represent for the immigrant worker36 29
an historical equivalent to the ideas of Marxist and other forms of 30
socialism among workers in the metropolises of earlier periods. 31
In setting out these ideas I am of course going beyond the con- 32
cepts, theories, and areas of concern which Max Weber actually 33
had in his own work. But we are bound to do this in describing a 34
Weberian approach to problems of race and ethnic relations which 35
have become more acute since his day. What may certainly be said, 36
however, is this: Weber, like Marx, would not have ascribed racism, 37
and its practical correlate of racialism, to some simple factor of in- 38
compatibility of cultures or of natural antipathy to the unknown. Nor 39
would he have attributed them to some unexplained factor of psycho- 40
logical prejudice. He was fully aware and, as we have seen, actually 41
said that racial definitions of social groups were related to the pur- 42
suit of economic interest in closed social relations. Here again, there 43

466 john rex
1 is much in common between Marxian and Weberian approaches.

2 Where Weber differed from Marx was in his more detailed analy-
3 sis of the kinds of social structure, organization, and process which
4 were to be found in different historical periods. In his analysis of the
5 economic institutions of the ancient world especially, we see the way
6 in which he showed, in some detail, the kinds of capitalism which
7 might operate. It is the reproduction of such structures in European
8 empires of the last five centuries which have been productive of the
9 specific kind of problems which we call race relations problems, and
10 it is in the systematic analysis of these structures that we will find a
11 characteristically Weberian approach to the study of race relations.
12
13
14 Notes
15
16 1 Parsons, T. The social system, London, Tavistock, 1952, p. ix.
17 2 See e.g.: Levi-Strauss, C. The scope of anthropology. London, Jonathan Cape,
1967.
18 3 Myrdal, G. Value in social theory. London, Routledge & Kegan Paul, 1958.
19 Chapter VII.
20 4 See: Bendix, R. Max Weber: an intellectual portrait. London, Heinemann,
21 1962; Weber, M. Economy and society. Vol. I. New York, Bedminster Press,
22 1968 (Introduction by G. Roth); Gerth, H.; Mills, C.W. From Max Weber.
London, Oxford University Press, 1946. p. 363-95.
23
5 Weber, M. op. cit. p. 339-97.
24 6 Ibid. p. 375-86.
25 7 Ibid., p. 341-2.
26 8 See especially: Weber, M. The agrarian sociology of ancient civilizations.
27 London, New Left Book, 1976. p. 53-60.
28 9 Ibid., p. 49.
10 Ibid., p. 51.
29 11 Weber, M. Economy and society. op. cit. p. 164-5. See also: Weber, M. General
30 economic history. New York, Collier Books, Macmillan.
31 12 Weber, M. The agrarian sociology... op. cit. p. 319.
32 13 Ibid., p. 53.
33 14 Ibid., p. 54.
15 Weber, M. General economic history. op. cit.
34
16 A good introduction to this debate with a select bibliography is to be found
35 in: Foner, Laura; Genovese, E.D. Slavery in the New World. Englewood
36 Cliffs, N.J., Prentice-Hall, 1969.
37 17 Genovese, E.D. The world the slaveholders made. London, Allen Lane, 1969
38 (especially Chapters 1-2). Surprisingly Genovese sees Weber’s particular
39 contribution as lying in his study of ideological factors, ignoring the direct
contribution which he made to the analysis of the institution of plantation
40 slavery.
41 18 Fogel, R.W.; Engermann, S.L. Time on the Cross. Boston, Little, 1974. (2
42 vols.).
43 19 Pares, R. A West Indian fortune. London, Longmans, 1950.

20 Tinker, H. A new system of slavery. London, Oxford University Press, 1974. 1

(Institute of Race Relations.) 2
21 Rex, J. The plural society: the South African case. Race (London), vol. XII, 3
no. 3, 1971, p. 401-13; Rex, J. The compound, the reserve and the location –
4
the essential institutions of South African labour exploitation. South African
labour bulletin, vol. 1, no. 4, April 1971, p. 4-17; Van Onselen, C. Chibaro, 5
African mine labour in Rhodesia 1900-1913. London, Pluto Press, 1976. 6
22 Woddis, J. Africa – the roots of revolt. London, Laurence & Wishart, 1960. p. 7
64. 8
23 Ibid., p. 143.
9
24 Boxer, C.R. Race relations in the Portuguese Colonial Empire, 1415-1825. Oxford,
Clarendon Press, 1963. 10
25 Weber, M. Economy and society. Vol. 2, op. cit. p. 926-39. 11
26 Cox, O. Caste, class and race. New York, Monthly Review Press, 1959. p. 143- 12
52. 13
27 Smith, M.G. The plural society in the British West Indies. Berkeley, University 14
of California Press, 1965; Smith, M.G.; Kuper, L. Pluralism in Africa.
15
Berkeley, University of California Press, 1969 (Chapters 2, 4 and 5).
28 Van den Berghe, P. Race and racism: a comparative perspective. New York, 16
Wiley, 1967. p. 29 ff. 17
29 For a more extended discussion of these problems in the context of Latin 18
America and the Caribbean, see: Rex, J. New nations and ethnic minorities: 19
comparative and theoretical questions. To be published by Unesco in a sym-
20
posium on inter-ethnic relations in the Caribbean and Latin America, 1977;
Rex. J. Race relations in sociological theory. London, Weidenfeld & Nicolson, 21
1970. 22
30 Frank, A.G. The sociology of underdevelopment and the underdevelopment of 23
sociology. London, Pluto Press, 1971. See also: Oxaal, I.; Barnett, T.; Booth, 24
D. Beyond the sociology of development. London, Routledge & Kegan Paul, 25
1975.
26
31 Wallerstein, I. The modern world system. New York, Academic Press, 1974.
32 Elkins, S. Slavery. New York, Grossap & Dunlop, 1959. 27
33 Fanon, F. The wretched of the earth. London, Penguin Books, 1965. 28
34 Debray, R. The revolution in the revolution. New York, Grove Press, 1967. 29
35 Segal, R. The race war. London, Jonathan Cape, 1966. 30
36 The situation of the American black is seen here as equivalent to that of a
31
colonial migrant.
32
33
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Contextualizing feminism: 3
4
gender, ethnic and class divisions
5
6
Floya Anthias and Nira Yuval-Davis 7
8
9
10
11
12
The rise of ethnic studies in Europe coincided with a particular interest in
13
exploring the interrelationship of race and class. This scholarship was pro-
14
foundly theoretical, albeit embedded in a wider political struggle for equality. 15
According to sociologists Floya Anthias and Nira Yuvall-Davis, the plight of 16
black women was largely ignored within the literature of women’s and femi- 17
nist studies as well as the wider feminist movement. Black feminism, which 18
developed in response to this alleged ignorance, defined black women as 19
suffering from the ‘triple oppression’ of race, gender and class. Anthias and 20
Yuval-Davis dismissed this attitude for both theoretical and political reasons. 21
They suggested that such features could not be enmeshed in each other. 22
Moreover, they felt that the position of black women could not be reduced 23
so simplistically in opposition to white women. In an influential article pub- 24
lished by the journal Feminist Review in 1983, the authors set out to systemati- 25
cally address the issue of ethnic and gender divisions without reducing them 26
to some form of class division. 27
28
29
30
Introduction 31
32
‘Sisterhood is powerful.’ ‘Sisterhood’ can also be misleading unless Sisterhood 33
contextualized. Black, minority and migrant women have been on 34
the whole invisible within the feminist movement in Britain and 35
within the literature on women’s or feminist studies. 36
This paper attempts to explore the issue of the interrelationship 37
of ethnic and gender divisions.1 Not only is such an attempt long 38
overdue theoretically but it also raises political issues which must be 39
central to feminist struggle. 40
Our analysis serves to problematize the notion of ‘sisterhood’ and 41
the implicit feminist assumption that there exists a commonality of 42
interests and/or goals amongst all women. Rather we argue that ev- 43

470 floya anthias and nira yuval-davis
1 ery feminist struggle has a specific ethnic (as well as class) context.
2 Although the notion of the ‘ethnic’ will be considered later in the
3 paper we note here that for us it primarily relates to the exclusionary/
4 inclusionary boundaries of collectivities formed round the notion of
5 a common origin.2 The ‘ethnic’ context of feminist struggles has
6 been systematically ignored (except in relation to various minorities,
7 especially ‘black’) and we suggest this has helped to perpetuate both
8 political and theoretical inadequacies within feminist and socialist
9 analyses.
10 the black The black feminist movement has grown partly as a response to
11 feminist movement the invisibility of black women and to the racism of the white feminist
12 movement. Recently several books have appeared, mostly American
13 which discuss black women and feminism. Bell Hooks puts her case
14 against white feminism clearly when she states:
15
16 In much of the literature written by white women on the ‘woman
17 question’ from the nineteenth century to the present day, authors
18 will refer to ‘white men’ but use the word ‘woman’ when they really
19 mean ‘white woman’. Concurrently, the term ‘blacks’ is often made
20 synonymous with black men (1981: 140).
21
22 In addition she points out that there has been a constant comparison
23 of the plight of ‘women’ and ‘blacks’ working with these racist/sexist
24 assumptions and which has diverted attention from the specificity
25 of the oppression of black women. We share this critique of white
26 feminism which is found within the black feminist movement in
27 Britain also. However we want to broaden out the frame of reference
28 of the existing debate. Within black feminism the most dominant ap-
29 proach defines black women as suffering from the ‘triple oppression’
30 of race, gender and class. This approach is inadequate, however, both
31 theoretically and politically. Race, gender and class cannot be tagged
32 on to each other mechanically for, as concrete social relations, they
33 are enmeshed in each other and the particular intersections involved
34 produce specific effects. The need for the study of the intersection
35 of these divisions has been recognized recently by black feminists.3
36 We also suggest, however, that the issue of the interrelationship
37 of the different social divisions cannot focus only on black versus
38 white women’s position. This has the theoretical effect of singling
39 out ‘racism’ as applicable only to ‘black’ women and focusses then on
40 the colour rather than on the structural location of ethnic groups as
41 determinants of their social relations. In addition an exclusive focus
42 on ‘racism’ fails to address the diversity of ethnic experiences which
43 derive from other factors like economic or political position. The no-

tion of ‘black women’ as delineating the boundaries of the alternative 1

feminist movement to white feminism leaves non-British non-black 2
women (like us – a Greek-Cypriot and an Israeli-Jew) unaccounted 3
for politically. Although we recognize the impetus behind the black 4
women’s movement and the need for its autonomous organization, 5
black feminism can be too wide or too narrow a category for specific 6
feminist struggles. On the one hand, there are struggles which con- 7
cern all migrant women, like those against immigration laws, and on 8
the other hand there are struggles which might concern only Sikh 9
Indian women for instance. 10
For these reasons, our paper will use the notion of ethnic divi- ethnic divisions 11
sions rather than the black/white division as a more comprehensive 12
conceptual category for struggling against racism. One of our tasks 13
will be to consider the links between the concepts of racism and eth- 14
nicity as well as attempting to relate ethnic divisions to those of gen- 15
der and class. 16
The marxist tradition of analysis which has informed much of 17
socialist-feminist analysis has been partly responsible for the invisi- 18
bility of ethnic divisions (as well as the feminist tradition itself which 19
assumes unitary and biological roots to ‘women’). Contemporary 20
marxist analysis has indeed recognized the importance of relating 21
ethnic to class divisions and gender to class divisions but there has 22
been little attempt to link ethnic and gender divisions to each other. 23
In addition Marxism has had difficulty in analysing ethnic or gen- 24
der divisions without reducing them to some form of class division. 25
Because of the significance of this tradition of analysis for us we shall 26
present a critique of Marxism as a necessary preliminary to develop- 27
ing our own position. 28
We shall then present an exploratory framework for analysing the 29
interrelationship of ethnic and gender divisions. We shall briefly ex- 30
amine these divisions within two central areas of feminist analysis, 31
employment and reproduction. The paper will conclude by consider- 32
ing some of the implications of the analysis presented for the west- 33
ern/Third World feminist debate. 34
35
36
Ethnic and gender divisions and marxism 37
38
As already noted Marxism has particular difficulties in analysing non- Marxism 39
class social divisions. The marxist concept of the mode of production 40
is based on an abstract model of relations that does not signal the 41
concrete groups of people within it. It does however establish a firm 42
grounding for class divisions in as much as the concept of class is 43

1 hierarchically incorporated within a systematic theory whose central

2 concept is that of mode of production. But ethnic divisions and gen-
3 der divisions cannot be situated within this theory for they are not es-
4 sential constituents of it – the theoretical basis for them is missing.4
5 The abstract level of analysis in Marx’s Capital presents problems for
6 class the analysis of concrete social relations including those of class. In
7 some versions of Marxism found in economistic approaches, classes
8 as concrete groups of people are reduced to the workings of the econ-
9 omy or the ‘needs’ of capital. We do not accept the depiction of class
10 in concrete analysis as reducible to its own dynamics as found within
11 the sphere of the economy. Indeed much recent analysis has treated
12 classes as concrete historical groupings whose actual practices are
13 not reducible to mode of production effects. We would take issue
14 with a reductionist position that sees a necessary relationship be-
15 tween, for example, class determination and political/class position.
16 Particularly we reject this not only because of the usual reasons given
17 by Marxists, i.e. the separate effectivity of the ideological and political
18 realms, but also because we consider the intersection between class,
19 ethnic and gender divisions as important in the development of par-
20 ticular forms of political consciousness and action.
21 Unlike the analysis of class which finds a theoretical basis in
22 Marxism despite the difficulties encountered in concrete analysis,
23 different problems are presented in the analysis of gender and eth-
24 nic divisions. When these categories are used by Marxists they often
25 involve very common-sense usages since Marxism has not system-
26 atically concerned itself with them as theoretical constructs. This
27 has led to very unclear and unspecific usages and shifts in meaning
28 from, for example, identifying gender with a biological constituent
29 and at other times seeing it as a social construct or race as histori-
30 cally produced and yet as basically organized around the ascriptive
31 characteristic of ‘blackness’.
32 Because of Marxism’s failure to specifically deal with gender and
33 ethnic divisions, marxist-feminists and marxist anti-racists have at-
34 tempted to ground them within economic relations, although marx-
35 ist-feminists particularly have sought to do so in a non-class reduc-
36 tionist way.
37 Ethnic and gender groups have been seen as structured by the
38 ‘needs’ of capital for migrant labour or cheap labour. The reserve
39 army of labour debate is an example of this.5 In addition there has
40 been a tendency to reduce these groups to fundamentally class group-
41 ings. For example we have seen attempts to theorize black people in
42 Britain as a class fraction, or an underclass and migrants in Europe
43 as a ‘class stratum’ of the working class. This approach empirically

fails to note the differentiation within the ethnic or migrant catego- 1

ry, both in terms of ethnicity and gender and in terms of economic, 2
political and ideological location. In addition this reduction to class 3
can only present gender and ethnic identities as some form of ‘false- 4
consciousness’ – as illusionary. For example some attempts to theo- 5
rize ethnicity have seen it as a form of incipient class consciousness 6
whose essential project develops into that of class.6 (Interestingly 7
the notion of women as a class is mostly systematically presented by 8
Delphy (1977) from a radical-feminist position.) 9
The marxist theorization of the state, ranging from the classical 10
marxist tradition of Engels, Lenin and Rosa Luxemburg to more re- 11
cent developments (instrumental, coordinator functional and state 12
derivation approaches) presents a different problem for the analysis 13
of ethnic divisions.7 14

Marxist theories of the state have tended to identify the boundar- 15
ies of the national collectivity with that of the relations of production. 16
This is found in Marx’s own assumption concerning the overlapping 17
of the boundary between civil and political society. In Marx’s words 18
‘In the state the whole civil society of an epoch is epitomized’. For 19
Marxists, on the whole, the rise of the nation-state is actively bound- 20
ed by the relations of production and conditions of class conflict. For 21
example the classical analysis of Engels of the emergence of the state 22
depicts it as a result of society’s entanglement in insoluble class an- 23
tagonisms (Engels, 1972). Thus marxist analyses have been sensitive 24
to differential access to power of different classes but not to other 25
forms of differential access based on gender or ethnic, national or 26
racial divisions.8 These assumptions are not seriously challenged by 27
the various recent marxist theorizations of the state. 28
Our view is that it is not sufficient to assert as Schermerhorn 29
(1970) does that each nation-state in the modern world contains sub- 30
sections or sub-systems. It is also the case that in almost all social 31
formations there are sections of the population that are to varying de- 32
grees excluded from political participation and representation. This 33
exclusion operates at least partially in a different manner from the exclusion 34
exclusion of ‘classes’ of the dominant national or ethnic group. For 35
example, the new Nationality Bill in Britain presents exclusion not 36
on the basis of class (as does legislation concerning private property 37
for example) but on the basis of ethnicity and gender. 38
A further problem within some marxist literature is the sugges- 39
tion that internal ethnic divisions are ideological in the sense of ‘false’ 40
or non-real. The attempt to theorize a distinction between historical 41
(i.e. real) and non-historical (i.e. non-real) nations assumed that if an 42
ethnic minority was able to obtain a separate and independent state, 43

1 then it was based on a real and historical origin and other minorities
2 were non-historical and only ‘ideological’.9
3 three divisions All three divisions have an organizational, experiential and repre-
4 sentational form, are historically produced and therefore changeable,
5 are affected by and affect each other and the economic, political and
6 ideological relations in which they are inserted. Relations of power
7 are usually found within each division and thus often the existence of
8 dominant and subordinate partners. They are all therefore framed in
9 relation to each other within relations of domination. They may thus
10 involve political mobilization, exclusion from particular resources
11 and struggles over them, claims to political representation and the
12 formation of concrete interests and goals which may shift over time.
13 It is not a question therefore of one being more ‘real’ than the others
14 or a question of which is the most important. However it is clear that
15 the three divisions prioritize different spheres of social relations and
16 will have different effects which it may be possible to specify in con-
17 crete analysis. However we suggest that each division exists within
18 the context of the others and that any concrete analysis has to take
19 this into account.
20 Firstly, we shall briefly comment on these divisions, clarifying the
21 sense in which we use them and noting some of the main differences
22 amongst them. Secondly, we shall begin to situate them in relation
23 to each other in the spheres of employment and reproduction, two
24 central areas of feminist analyses. We shall particularly note the links
25 between gender and ethnic divisions since this has rarely been con-
26 sidered.
27
28
29 Class, gender and ethnic divisions
30
31 As socialists working within a broadly marxist-informed analysis we
32 see class divisions as grounded in the different relations of groups
33 to the means of production which provides what has been called a
34 group’s class determination. However class mobilization cannot be
35 read from class determination for class goals are constructed through
36 a variety of different mechanisms with ideological practices having
37 a central role in this. Concrete class groupings may be composed
38 of both men and women, of black and white and different cultures
39 and ethnic identities. These concrete groupings are constructed his-
40 torically. At times there may be a coincidence of class and gender
41 or ethnic position (and at other times there maybe cross cuttings).
42 For example, some fractions of the working class may be primarily
43 composed of women or black people. This may reflect economic, po-

litical and ideological processes but may also be structured through 1

struggle and negotiation between the groups themselves and in rela- 2
tion to the state. Classes are not homogeneous ethnically, culturally 3
or in terms of gender in most cases but class fractions may constitute 4
some kind of homogeneity. 5
Gender divisions relate to the organization of sexual difference gender divisions 6
and biological reproduction and establish forms of representation 7
around these, although their concrete contents will include notions of 8
the appropriateness of wage-labour, education and so on to men and 9
to women. Usually sexual difference and biological reproduction (the 10
ontological basis of gender) are represented as having necessary so- 11
cial effects (from say ‘sexual intercourse’ to ‘class position’). Gender 12
divisions thus usually work with a notion of a ‘natural’ relationship 13
between social effects and sexual differences/biological reproduc- 14
tion. We do not accept such a depiction nor that biological reproduc- 15
tion is an equivalent material basis for gender to that of production for 16
class. Indeed the attempt to discover a feminist materialism in the 17
social relations of reproduction fails precisely in the attempt to super- 18
impose a materialist project onto a different object and reproduce its 19
terms of reference.10 Finally the end result is indeed to reduce these 20
social relations to their material base (biology) just as within marxist 21
materialism the reduction is to ‘mode of production’. 22
Rather we reject both biological reductionism and class reduction- 23
ism. We are suggesting that there is an object of discursive reference 24
in the sphere of gender divisions which relates to groups of subjects 25
defined by their sexual/biological difference as opposed to groups of 26
subjects defined by their economic production difference as in class. 27
Gender divisions are ‘ideological’ to the extent that they do not have a 28
basis in reproduction, but reproduction is represented as their basis. 29
However, the ideological nature of gender divisions does not mean 30
they do not exist nor that they do not have social origins and social 31
effects or involve material practices. 32
Unlike class and gender divisions, ethnic divisions are difficult to 33
ground in some separate sphere of relations. This makes the various 34
marxist and sociological attempts to try to find systematic concep- 35
tual differences between national/ethnic and racial groupings even 36
more problematic. This attempt is never successful because it is im- 37
possible to systematically ascribe particular and different realms to 38
them. Migration, conquest and colonization have developed a vast 39
heterogeneous body of historical cases. 40
The only general basis on which we can theorize what can broadly 41
be conceived as ‘ethnic’ phenomena in all their diversity are as vari- 42
ous forms of ideological construct which divide people into different 43

1 collectivities or communities. This will involve exclusionary/inclu-

2 sionary boundaries which form the collectivity. In other words al-
3 though the constructs are ideological, they involve real material prac-
4 tices and therefore origins and effects. Whether the boundaries are
5 those of a tribe, a nation or a linguistic or cultural minority, they will
6 tend to focus themselves around the myth of common origin (wheth-
7 er biological, cultural or historical). Although sometimes there will
8 be other means of joining the collectivity than being born into it (like
9 religious conversion or naturalization), group membership is con-
10 sidered as the ‘natural’ right of being born into it. The salience of the
11 collectivity and the social relations involved can vary greatly.
12 ethnicity Ethnicity is not only a question of ethnic identity. This latter does
13 not exhaust the category of the ‘ethnic’ nor does it necessarily oc-
14 cur. Ethnicity may be constructed outside the group by the material
15 conditions of the group and its social representation by other groups.
16 However in practice ethnic identity and often solidarity may occur
17 either as a pre-requisite for the group or as an effect of its mate-
18 rial, political or ideological placement. In addition ethnicity involves
19 struggle, negotiation and the use of ethnic resources for the counter-
20 ing of disadvantages or perpetuation of advantages. Conditions of
21 reproduction of the ethnic group as well as its transformation are
22 related to the divisions of gender and class. For example, class homo-
23 geneity within the ethnic group will produce a greater cohesion of
24 interests and goals.
25 The concept of ethnicity has too often been identified in Britain
26 with the Ethnic School tradition which tends to concentrate on issues
27 of culture or identity and has come under a great deal of justified at-
28 tack for ignoring racism and the structural disadvantages of minority
29 ethnic groups.11 However our use of the term ethnicity has as a central
30 element exclusion/inclusion practices and the relations of power of
31 dominance/subordination that are aspects of these. Majority groups
32 possess an ethnicity as well as minority groups. Ethnicity and racism
33 share both the categories of exclusion and power but racism is a spe-
34 cific form of exclusion. Racist discourse posits an essential biologi-
35 cal determination to culture but its referent may be any group that
36 has been ‘socially’ constructed as having a different ‘origin’, whether
37 cultural, biological or historical. It can be ‘Jewish’, ‘black’, ‘foreign’,
38 ‘migrant’, ‘minority’. In other words any group that has been located
39 in ethnic terms can be subjected to ‘racism’ as a form of exclusion.
40 The ‘Racist’ category is more deterministic than the mere ‘ethnic’
41 category.
42 Concerning the difference between ethnic and national groups, it
43 is often a question of the different goals and achievements of the col-

lectivity. The nationalist project is more strictly political for its claims 1
will necessarily include rights to separate political representation 2
or to territory (as in the case of Palestinians and Jews in Israel and 3
Turkish-Cypriots and Greek-Cypriots in Cyprus). 4
We consider that gender and ethnic divisions particularly are un- 5
derpinned by a notion of a ‘natural’ relation. In gender divisions it is a ‘natural’ relation 6
found in the positing of necessary social effects to sexual difference 7
and biological reproduction and in ethnic divisions by assumptions 8
concerning the ‘natural’ boundaries of collectivities or the ‘natural- 9
ness’ of culture. In capitalist societies like Britain very often the ‘nat- 10
ural’ ideological elements of gender and ethnic divisions are used to 11
‘naturalize’ unequal class divisions. Gender and ethnic divisions are 12
used as legitimizors in two major ways. 13
In patriarchal white societies it is perceived as ‘natural’ that men 14
will occupy a higher economic position in the labour market than 15
women and white people than black people. For example notions 16
of women’s sexual difference (more ‘submissive’, ‘feminine’, ‘intui- 17
tive’, ‘expressive’, ‘dextrous’) and their ‘essential mothering role’ are 18
used and are often manipulated for economically justifying (explain- 19
ing) women’s position (at times by women themselves). Racism and 20
ethnicity also have a role in justifying the economic/class subordi- 21
nation of black people. For example arguments about the cultural 22
choices of ethnic groups – and racial stereotypes about Asian men 23
(money-seeking) and Afro-Caribbean men (work idle) – are used to 24
account for their economic position. The second way in which the 25
‘natural’ elements of gender and ethnic divisions are used is as ral- 26
lying points for political struggle against class inequality as well as 27
gender and ethnic inequalities. This is the case in most anti-imperi- 28
alist struggles where notions of national identity are used. The black 29
power movement has often used racial/ethnic identification partly 30
as a counter to existing racial stereotypes and oppressions (for ex- 31
ample in black nationalism the identification with Africa and in black 32
power the ‘black is beautiful’ rhetoric and more recently, culturalist 33
and religious revivals such as Rastafarianism). As regards gender, 34
feminists have used women’s ‘nature’ as a rallying point, particularly 35
with reference to the positive values of women’s culture and ‘nature’. 36
However, using ethnic and gender categories in this way as rallying 37
points for political mobilization in class-related struggles can present 38
a problem for class unity. 39
As well as ethnic and gender divisions being used for class goals, 40
class divisions can provide the material conditions for ethnic and 41
gender groups, for these will give unequal access to economic re- 42
sources. State practices may exclude class, ethnic and gender group- 43

1 ings in different ways, structure their relationship to each other and

2 give differential political power to different groups. Therefore when
3 we analyse specific historical cases these divisions often cannot be
4 separated.
5 We have suggested that the ‘natural’ ideological aspects of ethnic
6 and gender groupings inform class relations. In addition we would
7 suggest that ethnic and gender divisions are more socially immu-
8 table. Whereas it is possible theoretically for subjects to change class
9 position (although empirically it may be difficult), it is not so for gen-
10 der or ethnic position (especially for the ‘racial’ category). Gender
11 position is fixed (apart from transexuals) and generally one is ‘born’
12 into one’s ethnic position. In particular cases, women can become
13 ‘honorary’ men (when men are not available for example to do ‘male’
14 work as in war) or religious conversion can occur. But the major
15 mechanism is ascriptive for both ethnic and gender divisions.
16
17
18 The relations between gender and ethnic divisions
19
20 intermeshed We suggested above all that three divisions are intermeshed in such
21 divisions a way that we cannot see them as additive or prioritize abstractly any-
22 one of them. Each division presents ideological and organizational
23 principles within which the others operate, although in different his-
24 torical contexts and different social arenas their role will differ. The
25 fusion of gender and class and ethnicity and class will also operate in
26 the relationship between gender and ethnic divisions.
27 For example if we consider the household we will find gender
28 divisions will differ according to ethnicity. Ethnically specific defi-
29 nitions of women’s and men’s roles underlie the sexual division of
30 labour in the family. Such aspects as mothering, housework, sexual
31 obligations, obedience and submissiveness to male commands (and
32 indeed to other members of the family) will differ according to eth-
33 nicity (as well as class of course). We would suggest that ethnic di-
34 visions are particularly important in the internal gender divisions
35 within the household and family therefore, although state practices
36 will affect them.
37 If we consider the sphere of employment – the more public or
38 external sexual division of labour – this will be affected particularly
39 by the gender divisions of the majority ethnic group. Values and in-
40 stitutionalized practices about women’s ‘nature’ and ‘role’ present
41 constraints to men and women from minority/subordinate ethnic
42 groups despite their own gender ideologies.
43 Another link between ethnic and gender divisions is found in the

way in which the boundary of ethnicity depends on gender. The defi- 1

nition of membership within the ethnic group often depends on per- 2
forming gender attributes correctly. Both identity and institutional 3
arrangements of ethnic groups incorporate gender roles and specify 4
appropriate relations between sexes such as, for example, who can 5
marry them. A Greek-Cypriot girl of the second generation is re- 6
garded as ‘Kypraia’ usually when she conforms to rules about sexu- 7
ally appropriate behaviour – otherwise she becomes excluded. The 8
definition of boundaries is far from being an internal practice alone. 9
If we consider racial stereotypes we can see the centrality of gender 10
roles; for example stereotypes about the ‘dominant’ Asian father and 11
the ‘dominant’ black mother, or stereotypes about black men and 12
women as sexual ‘studs’. These all indicate the reliance on gender 13
attributes for specifying ethnic difference. We want to briefly suggest 14
some more specific links between ethnic and gender divisions in em- 15
ployment and reproduction. 16
17
18
Employment 19
20
The internal gender divisions of an ethnic group will also affect the 21
participation of men and women of the group in the labour mar- 22
ket. Men and women of a specific ethnic group will tend to hold 23
particular but different positions in the labour market; for example 24
Afro-Caribbean men in the construction industry and on the buses, 25
Afro-Caribbean women as service workers in manufacturing and as 26
nurses, Asian men in textile firms and Asian women as outwork- a sexually and 27
ers in small-scale dress-making factories. A sexually differentiated ethnically 28
labour market will structure the placement of subjects according to differentiated 29
sex but ethnic divisions will determine their subordination within labour market 30
them so, for example, black and white women may both be subordi- 31
nate within a sexually differentiated labour market but black women 32
will be subordinated to white women within this. 33
We would suggest that within western societies, gender divisions 34
are more important for women than ethnic divisions in terms of la- 35
bour market subordination. In employment terms, migrant or ethnic 36
women are usually closer to the female population as a whole than to 37
ethnic men in the type of wage-labour performed. Black and migrant 38
women are already so disadvantaged by their gender in employment 39
that it is difficult to show the effects of ethnic discrimination for 40
them. When examining the position of ethnic minority men in the 41
labour market, the effect of their ethnic position is much more vis- 42
ible. This may lead to a situation where for example Afro-Caribbean 43

1 or Asian women have at times had greater ease in finding employ-

2 ment – as cheap labour in ‘women’s work’, whether it be nursing,
3 assembly-line or clerical work – than the men.
4 But the interrelationship between ethnic and gender divisions in
5 employment goes beyond the mere differentiation in employment
6 of ethnic subjects according to their gender. This additional dimen-
7 sion however is even less stressed in the literature on ethnic and
8 race relations. The economic and social advancement of a migrant
9 group may depend partly on the possibility of using the household
10 and in particular the women within it as a labour resource. The ex-
11 tent to which migrant ethnic men have become incorporated into
12 wider social production and the form this takes may also depend on
13 the use of migrant women’s labour overall. Men from different mi-
14 grant/ethnic groups have been incorporated differently economical-
15 ly. Afro-Caribbean men for example are in the ‘vanguard’ of British
16 industry in large-scale production (Hall et al, 1978:349). Asian and
17 Cypriot men on the other hand have had a greater tendency to go
18 into small-scale entrepreneurial concerns and into the service sector
19 of the economy. In particular, entrepreneurial concerns both within
20 the formal and hidden economy depend on the exploitation of female
21 wage-labour and in particular on kinship and migrant labour. Ethnic
22 and familial bonds serve to allow the even greater exploitation of fe-
23 male labour (Anthias, 1983). The different form of the family and
24 gender ideologies may partly explain the differences between Afro-
25 Caribbean employment patterns and those of Asians and Cypriots.
26
27
28 Reproduction
29
30 We want now to turn to the area of reproduction and briefly consider
31 gender, ethnic and it as a focus for the interrelation of gender, ethnic and class divisions.
32 class divisions The concept of reproduction itself is a problematic one. This part-
33 ly derives from the inconsistent and heterogeneous treatment it has
34 received in the literature.12 Edholm et al (1977:103) suggest that the
35 notion of reproduction might be read as assuming that ‘social sys-
36 tems exist to maintain themselves through time (to reproduce them-
37 selves) and secondly, that all levels of the system must be maintained
38 through time in the same way’. This assumption indeed, would have
39 all the pitfalls of the functionalist approach to social analysis. The
40 reproduction of people and collectivities is directly shaped by the his-
41 torical and social context in which it takes place. Nor is it an homoge-
42 neous process, and contradictions and conflicts are found not only in
43 the reproduction of various entities that partially overlap each other
but also in the form of the reproduction process itself.

Women not only reproduce the future human and labour power 1
and the future citizens of the state but also ethnic and national col- 2
lectivities. As in other aspects of the gender division of labour, the 3
ethnic and class position of women will affect their role in the repro- 4
duction process. Questions concerning who can actually reproduce 5
the collectivity and under what conditions are often important here. 6
Such things as the legitimacy of marriage, the appropriate religious 7
conviction and so on are often preconditions for the legitimate repro- 8
duction of the nation or collectivity. The actual degree and form of 9
control exercised by men of ethnic collectivities over their women can control 10
vary. In the Muslim world for example and in Britain under the old 11
nationality law, the ethnic, religious or national position of women 12
was immaterial. In other cases, like in the Jewish case, the mother’s 13
origin is the most important one in delineating the boundaries of the 14
collectivity, and this determined the reproduction of the Jewish ‘na- 15
tion’ (Yuval-Davis, 1980). This clearly does not mean such women 16
have greater freedom but only that they are subject to a different set 17
of controls. 18
As in other areas, the links between gender divisions and ethnic 19
divisions can be and often are subject to the intervention of the state. 20
For example, in Israel even secular people have to marry with a reli- 21
gious ceremony and according to traditional religious rules, in order 22
for their marriage to be recognized by law. In the most extreme cas- 23
es, the way the collectivity is constituted by state legislation virtually 24
prevents inter-marriage between collectivities. In Egypt, for instance, 25
while a Christian man can convert to Islam, Muslim women are pre- 26
vent from marrying Christian Copts – if they do, they are no longer 27
part of the Muslim community nor are they recognized as part of the 28
Christian community and they virtually lose their legal status. The 29
state may treat women from dominant and subordinate ethnic col- 30
lectivities differently. For example, the new nationality law in Britain 31
has given autonomous national reproduction rights to white British 32
women, while totally witholding them from many others, mostly 33
black women. 34
This differential treatment does not relate only to ideological or 35
legal control of reproduction. The infamous contraceptive injection 36
Depo-Provera has been given in Britain and elsewhere virtually exclu- 37
sively to black and very poor women, and a study found more birth 38
control leaflets in family planning clinics in Asian languages than 39
in English (see Brent Community Council, 1981). In Israel, Jewish 40
families (under the label of being ‘relatives of Israeli soldiers’) re- 41
ceive higher child allowances than Arab ones, as part of an elaborate 42
policy of encouraging Jewish population growth and discouraging 43

1 that of Arabs. Indeed the Beveridge Report in Britain justified the

2 establishment of child allowances in order to combat the danger of
3 the disappearance of the British race (1942: 154).
4 political tool On the other hand, reproduction can become a political tool at
5 the hand of oppressed ethnic minorities. A common Palestinian say-
6 ing is that ‘The Israelis beat us at the borders and we beat them at
7 the bedrooms’ – Palestinian women, like Jewish ones (and with a
8 higher rate of success due to various material and ideological factors)
9 are under pressure by their collectivity, although not by the state, to
10 reproduce and enlarge it. It is a fact, for example, that no Palestinian
11 children in Lebanon were allowed (unlike Vietnamese children un-
12 der similar circumstances) to be adopted by non-Palestinians – all
13 the children are looked on as future Palestinian liberation fighters. In
14 other words, the control of reproduction can be used both as a subor-
15 dinating strategy – by dominant groups against minority groups – as
16 well as a ‘management’ strategy by ethnic collectivities themselves.
17 We started the section by pointing out that the process of repro-
18 duction of human subjects, as well as of collectivities is never uni-
19 tary. We want to emphasize that this is the case also concerning the
20 participation of women themselves in the control of reproduction.
21 We can point out that virtually everywhere, the interests of the na-
22 tion or the ethnic group are seen as those of its male subjects, and
23 the interests of ‘the state’ are endowed with those of a male ethnic
24 class and not just a class which is ‘neutral’ in terms of ethnicity and
25 gender. However, very often women participate directly in the power
26 struggle between their ethnic collectivity and other collectivities and
27 the state, including by voluntarily engaging in an intensive repro-
28 ductive ‘demographic’ race. At the same time women of dominant
29 ethnic groups are often in a position to control the reproductive role
30 of women of other ethnic groups by state welfare and legal policies,
31 as well as to use them as servants and child minders in order to ease
32 part of their own reproductive burden.
33 This last point leads us to consider the political implications of
34 the above discussion concerning feminist politics and the common-
35 ality of feminist goals.
36
37
38 Political Implications
39
40 As mentioned in the introduction to this paper, our interest in the
41 subject is far from being merely academic. It originates from our
42 own frustration in trying to find a political milieu in which ethnic
43 divisions will be seen as an essential consideration, rather than as

non-existent or as an immovable bloc to feminist politics. 1

The theoretical part of this paper pointed out how misleading it 2
is to consider gender relations without contextualizing them within 3
ethnic and class divisions. Once we take the full implications of this 4
into account, the mystification of the popular notion of ‘sisterhood’ 5
becomes apparent. As we pointed out there can be no unitary cat- 6
egory of ‘women’. The subordination of women to men, collectivities 7
and the state operates in many different ways in different historical 8
contexts. Moreover, very often women themselves participate in the 9
process of subordinating and exploiting other women. 10
One major form of women’s oppression in history has been 11
their invisibility, their being ‘hidden from history’. The invisibility 12
of women other than those who belong to the dominant ethnic col- 13
lectivity in Britain within feminist analysis has been as oppressive. 14
Except for black feminists who fought their own case in isolation, 15
minority women have been virtually absent in all feminist analysis. 16
Anthropological and historical differences in the situation of women 17
have been explored, but only in order to highlight the social basis of 18
gender relations in contemporary Britain. The heterogeneous ethnic 19
character of the latter has never been fully considered. 20
Recently there have been some signs of a developing awareness 21
of the need to take into account ethnic diversity. Earlier writing by 22
socialist-feminists like Michèle Barrett (1980) and Elizabeth Wilson 23
(1977) on women in Britain had completely ignored minority, mi- 24
grant, ethnic or black women. In the introduction to their latest 25
books however (Barrett and McIntosh, 1982; Wilson, 1983) they ac- 26
knowledge that they do not deal with ‘ethnic’ women or families. 27
This recognition is clearly no substitute for an attempt to situate eth- 28
nic divisions when analysing ‘the family’ in Britain. 29
On the political level some concessions have been made within 30
the last few months to the black feminist movement. For example, black feminism 31
the inclusion of black women in the Spare Rib Collective and on the 32
Women’s Committee of the Greater London Council are unprece- 33
dented and very important political achievements. However, these 34
concessions to black feminists are not a substitute for a coherent 35
self-critique and analysis of the white feminist movement in contex- 36
tualizing its own ethnic interests. 37
When we talk about the need of white feminists in Britain to rec- 38
ognize their own ethnicity, we are relating to questions as basic as 39
what we actually mean when we talk about ‘feminist issues’. Can we 40
automatically assume, as has been done by western feminist move- 41
ments, that issues like abortion, the depiction of the family as the 42
site of female oppression, the fight for legal equality with men and 43

1 against sex discrimination and so on are the feminist issues? Maxine

2 Molyneux (1983) has recently argued that what separates Third World
3 and western women is not so much the specific cultural or historical
4 contexts in which they are engaged but differences of a theoretical
5 and political nature.
6 Different theoretical and political positions exist, of course, as
7 Maxine claims, both in the West and in the Third World. But femi-
8 nist goals cannot be the same in different historical contexts. For
9 instance, the family may not be the major site for women’s oppres-
10 sion when families are kept apart by occupying or colonizing forces
11 (as in Lebanon or South Africa), abortion may not be the major is-
12 sue when forced sterilizations are carried out, nor is legal equality
13 for women the first priority in polygamic societies where there is
14 no independent autonomous mode of existence open to women
15 whose husbands marry other younger and more fertile women. In
16 their paper on the South African women’s movement, Judy Kimble
17 and Elaine Unterhalter (1982) suggest that ‘the analysis and objec-
18 tive of western feminism cannot be applied abstractly and univer-
19 sally’. Western feminist struggles cannot be seen as dealing with ‘the
20 feminist issues’ but with culturally and historically specific issues
21 relevant mainly to middle class white women who have their own (in-
22 visible to them?) ethnicity. Judy and Elaine stress an essential point.
23 However, it seems that in their search for an alternative perspective,
24 they go to the other extreme and end up in fact with a circular argu-
25 ment – that feminist struggles in the context of national liberation
26 movements are to be found in what the women in these movements
27 do. In other words, once we stop perceiving western white feminism
28 as providing the ultimate criteria for defining the contents of femi-
29 nism, we are faced with the problem of how to politically evaluate
30 various women’s struggles.
31 The beginning of a possible approach might be found in an article
32 by Gail Omvedt (1978) in which she suggests that there is a differ-
33 feminist struggles entiation between ‘women’ struggles and ‘feminist’ struggles, in as
34 much as the latter are those that challenge rather than use traditional
35 gender divisions within the context of national or ethnic struggles.
36 We would add, however, that the challenge has to be, in our opin-
37 ion, directed to both women’s and men’s work. All too often, in na-
38 tional liberation struggles, as in other periods of social crisis, women
39 are called upon to fulfill men’s jobs, as men are otherwise engaged
40 at the front (as in war). This expansion in women’s roles is seen too
41 often as an act of women’s liberation rather than as another facet of
42 women’s work. When the crisis is over, women are often assigned
43 again to the more exclusively feminine spheres of women, to the

surprise, as well as disappointment, of all those who have seen in 1

the mere participation of women in the ‘struggle’ (whether in the 2
Israeli Kibbutz, Algiers or Vietnam), a feminist achievement. We 3
claim therefore that the challenge has to be to the actual notion of the 4
sexual division of labour rather than only to its specific boundaries. 5
This is far from being simple, because so many, if not all ethnic 6
cultures, as we have noted before, have as central the construction of 7
a specific form of gender division. It is too easy to pose the question, 8
as many anti-imperialist and anti-racist feminists do, as if the origin 9
and site of their oppression is only constructed from above, by white 10
male sexism. 11
Ethnic and gender liberation struggles and solidarities can cut 12
across each other and be divisive. We do not believe that there is one 13
‘right’ line to be taken in all circumstances. The focus or project of 14
each struggle ought to decide which of the divisions we prioritize and 15
the extent to which separate, as opposed to unified, struggle is neces- 16
sary. Political struggles, however, which are formulated on an ethnic 17
or sexual essence, we see as reactionary. Nor do we see it as a viable 18
political option for women of subordinate collectivities to focus all 19
their struggle against the sexism of dominant majority men. 20
The direct conclusion from our analysis in this paper is that any 21
political struggle in relation to any of the divisions considered in this 22
paper, i.e. class, ethnic and gender, has to be waged in the context context 23
of the others. Feminist struggle in Britain today cannot be perceived 24
as an homogeneous struggle, for the participation and oppression 25
of women, both in the family and at the work site, are not homoge- 26
neous. White middle class feminists have to recognize the particular- 27
ity of their own experiences, not only in relation to the Third World 28
but also in relation to different ethnic and class grouping in Britain 29
and integrate this recognition into their daily politics and struggles. 30
Only on this basis can a valid sisterhood be constructed among wom- 31
en in Britain. 32
33
34
Notes 35
36
Floya Anthias and Nira Yuval-Davis lecture in Sociology at Thames 37
Polytechnic, London. They are currently engaged on a research proj- 38
ect on ethnic and gender divisions in Greenwich and Woolwich, 39
Southeast London. 40
41
42
43

1 1 Our analysis in this paper has benefitted much from discussions with
2 and feedback from our colleagues in the Sociology Division at Thames
3 Polytechnic who are working with us on the Ethnic and Gender Division
Project and we would like to thank them all. We should also like to thank
4
all those who participated in the Gender and Ethnic Divisions seminars ar-
5 ranged by the Sociology Division. Additionally we would like to thank the Sex
6 and Class Group of the CSE, and the Feminist Review Collective, especially
7 Annie Whitehead and Lesley Caldwell, for their insightful comments after
8 reading the first draft of our paper.
2 The term ‘ethnic’ and ‘ethnicity’ have come under a great deal of attack re-
9
cently for mystifying racist social relations. However, as we argue later, we
10 do not use these concepts within a mainstream sociological tradition. For a
11 critique of these terms see for example E. Lawrence (1982).
12 3 In a series of seminars organized by the Thames Polytechnic Sociology
13 Division on Gender and Ethnic Divisions, Valerie Amos, Pratibha Parmar
14 and Amina Mama all presented analyses that stressed the importance of
studying the way in which the fusion of ethnic, gender and class divisions
15
for black women gave a specificity to their oppression.
16 4 For the problems of theorizing gender divisions using a marxist framework
17 see H. Hartmann (1979). For problems of theorizing race in Marxism see
18 particularly J. Gabriel and G. Ben-Tovim (1978).
19 5 See V. Beechey (1977) for an attempt to apply the concept to women. See S.
Castles and G. Kosack (1972) for an analysis of migrants as a reserve army.
20
For a critique of such attempts see F. Anthias (1980).
21 6 For critical reviews of this position see J. Kahn (1981) and J.S. Saul (1979).
22 7 For a review of marxist theories of the State see Bob Jessop (1982).
23 8 Socialist-feminist analysis of course is an exception to this. For example see
24 the work of E. Wilson (1977).
25 9 For example H.B. Davis (1973:31) states ‘Engels was using the theory of “his-
toryless peoples” according to which peoples that have never formed a state
26
in the past cannot be expected to form a viable state in the future’.
27 10 This approach is found for example in Z. Eisenstein (1979).
28 11 For a critique see J. Bourne and A. Sivanandan (1980).
29 12 See M. Mackintosh (1981), F. Edholm et al. (1977) and N. Yuval-Davis (1982).
30
31
32 References
33
34 ANTHIAS, F. (1980) ‘Women and the Reserve Army of Labour’
35 Capital & Class No. 10.
36 ANTHIAS, F. (1983) ‘Sexual Divisions and Ethnic Adaptation’ in
37 PHIZACKLEA (1983).
38 BARRETT, M. (1980) Women’s Oppression Today London: Verso.
39 BARRETT, M. and McINTOSH, M. (1982) The Anti-social Family
40 London: Verso.
41 BEECHEY, V. (1977) ‘Some Notes on Female Wage Labour in the
42 Capitalist Mode of Production’ Capital & Class No. 3.
43 BOURNE, J. and SIVANANDAN, A. (1980) ‘Cheerleaders and

Ombudsmen: the Sociology of Race Relations in Britain’ Race and 1

Class Vol. XXI, No. 4. 2
BRENT COMMUNITY COUNCIL (1981) Black People and the Health 3
Service. 4
CASTLES, S. and KOSACK, G. (1972) ‘The Function of Labour 5
Immigration in Western European Capitalism’ New Left Review 6
No. 73. 7
CENTRE FOR CONTEMPORARY CULTURAL STUDIES (1982) 8
The Empire Strikes Back London: Hutchinson. 9
DAVIS, H.B. (1973) Nationalism and Socialism New York: Monthly 10
Review Press. 11
DELPHY, C. (1977) The Main Enemy London: Women’s Research 12
and Resources Centre. 13
EDHOLM, F., HARRIS, O. and YOUNG, K. (1977) ‘Conceptualizing 14
Women’ Critique of Anthropology Vol. 3, Nos. 9-10. 15
EISENSTEIN, Z. (1979) editor Capitalist Patriarchy and the Case for 16
Socialist Feminism New York: Monthly Review Press. 17
ENGELS, F. (1972) The Origin of the Family, Private Property and the 18
State London: Lawrence and Wishart. 19
GABRIEL, J. and BEN-TOVIM, G. (1978) ‘Marxism and the Concept 20
of Racism’ Economy and Society Vol. 7, No. 2. 21
HALL, S. et al. (1978) Policing the Crisis: Mugging, the State and Law 22
and Order London: Macmillan. 23
HARTMANN, H. (1979) ‘The Unhappy Marriage of Marxism and 24
Feminism: Towards a More Progressive Union’ Capital & Class 25
No. 8. 26
HM Government (1942) Social Insurance and Allied Services (The 27
Beveridge Report) Cmd 6404 London: HMSO. 28
HOOKS, B. (1981) Ain’t I a Woman? South End Press. 29
JESSOP, B. (1982) The Capitalist State Oxford: Manin Robertson. 30
KAHN, J. (1981) ‘Explaining Ethnicity’ Critique of Anthropology Vol. 31
4, No. 16, Spring. 32
KIMBLE, J, and UNTERHALTER, E. (1982) ‘“We opened the road 33
for you, you must go forward” ANC Women’s Struggles 1912- 34
1982’ Feminist Review No. 12. 35
LAWRENCE, E. (1982) ‘In the abundance of water the fool is 36
thirsty: sociology and black “pathology” ’ in CENTRE FOR 37
CONTEMPORARY CULTURAL STUDIES (1982). 38
MACKINTOSH, M. (1981) ‘Gender and Economics’ in YOUNG, 39
WOLKOWITZ and McCULLOGH (1981). 40
MOLYNEUX, M. (1983) ‘First and Third World Feminism: Solidarity 41
and Conflict’ Paper presented to Socialist Society Conference on 42
the Family, March 1983. 43

1 OMVEDT, G. (1978) ‘Women and rural revolt in India’ Journal of

2 Peasant Studies Vol 5, No. 3.
3 PHIZACKLEA. A. (1983) editor, One Way Ticket London: Routledge
4 and Kegan Paul (forthcoming).
5 SAUL, J.S. (1979) ‘The dialectic of class and tribe’ Race and Class Vol.
6 XX, No. 4.
7 SCHERMERHORN, R. (1970) Comparative Ethnic Relations New
8 York: Random House.
9 WILSON, E. (1977) Women and the Welfare State London: Tavistock
10 Publications.
11 WILSON, E. (1983) What is to be done about violence against women?
12 London: Penguin.
13 YOUNG, K., WOLKOWITZ, C. and McCULLOGH, R. (1981) editors,
14 Of Marriage and the Market London: CSE Books.
15 YUVAL-DAVIS, N. (1980) ‘The bearers of the collective: Women and
16 religious legislation in Israel’ Feminist Review No. 4.
17 YUVAL-DAVIS, N. (1982) ‘National Reproduction: Sexism, Racism
18 and the State’ Unpublished paper presented to BSA Conference,
19 April 1982.
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Varieties of Marxist conceptions of ‘race’, class 3
and the state: a critical analysis 4
5
6
John Solomos 7
8
9
10
11
12
For a number of years, Marxist analysis of international migration and the
13
interrelationship of race, class and the state were commonplace. Many ‘clas-
14
sic’ analyses, however, were dogged by economic determinism and theoreti- 15
cal abstraction. They failed to appreciate such complex non-class forms of 16
division and oppression, including ethnic and religious loyalty, gender, rac- 17
ism and sexism. But Marxism was not composed of a unified set of dogmas. 18
A growing number of authors responded to the theoretical imperfections by 19
exploring these phenomena from a more critical position. Sociologist John 20
Solomos’ article is a concise overview of the state of the art in the British 21
neo-Marxist debate. All neo-Marxist theoretical approaches agree that there 22
is no ‘race relations problem’ as such or, at least, that there is no problem 23
of racism that can be thought of as separate from the structural features of 24
capitalist society. The approaches differ with regard to the role of the state. 25
This article also questions to what extent racial and ethnic categorisations 26
are autonomous from economic and class determinations. 27
28
29
30
1. Introduction 31
32
It is a commonplace that the reliance of Marxist theory on the pivotal 33
concepts of mode of production and class, along with the preoccupa- 34
tion with general models of historical development, has precluded 35
Marxists from making a significant contribution to the study of racial 36
and ethnic divisions within capitalist society.1 The relative absence of 37
a substantive discussion of these questions within the texts of clas- 38
sical Marxism seems to add weight to the assertion made by Frank 39
Parkin that, as a form of social analysis, Marxism is incapable of deal- Parkin 40
ing with such divisions short of subsuming them under more gen- 41
eral social relations (production- or class-based) or treating them as a 42
kind of superstructural phenomenon (Parkin 1979a and b). 43

490 john solomos
1 This commonplace assertion seems to be contradicted, however,

2 by the increased interest among a number of Marxist theorists in clar-
3 ifying the complex forms of non-class (even if class-related) forms of
4 division and oppression that are characteristic of late capitalist societ-
5 ies, including racial and ethnic divisions, but also gender, national,
6 regional, religious and locality-based divisions.2 Indeed, over the last
7 decade in particular, a wide variety of Marxist conceptualisations of
8 race, class and the state have emerged, including a substantial body
9 of theoretical studies which attempt to develop a more precise and
10 systematic understanding of racism in capitalist society as rooted in
11 the dominant social relations and power structures (Genovese 1971,
12 Nikolinakos 1973, Hall 1977, 1980b, Gabriel and Ben-Tovim 1978,
13 Sivanandan 1982, Miles 1982, Centre for Contemporary Cultural
14 Studies 1982, Brittan and Maynard 1984). In addition a number of
15 Marxist-inspired historical and empirical studies of specific forms of
16 racist structures in different societies have been published over the
17 years, including the USA (Reich 1981, Fox-Genovese and Genovese
18 1983, Marable 1984) and South Africa (Wolpe 1980, Burawoy 1981).
19 The existence of these theoretical and empirical studies does not,
20 of course, mean that the criticisms of writers such as Frank Parkin
21 can be dismissed. Many of the problems which they highlight within
22 Marxist discourse, especially economic determinism and theoreti-
23 cal abstraction, are still to be found in much of the mainstream of
24 Marxism, which continues to treat racism as little more than an ir-
25 ritant to the smoother structures of historical materialism (see Ben-
26 Tovim et al. in this volume for more discussion of this point). Racism
27 an inadequately remains an inadequately theorised concept within the terms of both
28 theorised concept sociological and Marxist theory. The remainder of this paper will,
29 first, discuss some of the most important attempts to develop a criti-
30 cal understanding of the interrelationship between race, class and
31 the state in contemporary capitalism. Second, I shall attempt to de-
32 velop an alternative framework for analysing racism which builds
33 upon the strengths of recent contributions, particularly in relation to
34 the need to ground a theory of racism in the broader framework of
35 political economy. The paper concludes with a few remarks about the
36 implications of Marxist analyses of racism and the state for political
37 practice, particularly in relation to anti-racist struggles.
38
39
40 2. Origins and foundations
41
42 It will be helpful to clear away some preliminary points before pro-
43 ceeding. Although this paper addresses the question of a Marxist

analysis of ‘race’ and racism in capitalist society, it would be quite 1

mistaken to think of contemporary Marxism either as unified or as 2
composed of a unified set of dogmas. This is an assumption that is unified set of 3
too often made in the race relations literature, on the grounds that dogmas 4
the substantial difference between a Marxist approach to race and 5
other approaches lies in the reliance by Marxists on an economic 6
determinist explanation for the emergence and reproduction of rac- 7
ism. Consider the following remarks from Frank Parkin’s critique of 8
Marxism and its analysis of class: 9
10
On current evidence one could be forgiven for concluding that the 11
preferred Marxist response to the fact of racial or communal strife 12
is to ignore it. Not one of the various reformulations of class the- 13
ory... makes any serious attempt to consider how the division be- 14
tween blacks and whites, Catholics and Protestants, Flemings and 15
Walloons, Francophones and Anglophones, or between indigenous 16
and immigrant workers affects their general analysis. It is especially 17
difficult to see what kind of explanation could in any case be expect- 18
ed from those formulations which draw heavily upon the concep- 19
tual storehouse of political economy. Notions such as the mode of 20
production make their claims to explanatory power precisely on the 21
grounds of their indifference to the nature of the human material 22
whose activities they determine. To introduce questions such as the 23
ethnic composition of the workplace is to clutter up the analysis by 24
laying stress upon the quality of social actors, a conception diametri- 25
cally opposed to the notion of human agents as träger or ‘embodi- 26
ments’ of systemic forces. (Parkin 1979b, p. XXX) 27
28
As a statement in support of the thesis that it is impossible to com- 29
bine a Marxist analytic framework with a serious analysis of racial 30
and/or ethnic divisions this passage suffers from several problems. 31
First, it takes only a limited degree of knowledge about recent Marxist 32
debates to see that Parkin’s main assertion, that the explanatory pow- 33
er of the concept of mode of production depends on an indifference 34
to the role of social actors, is contradicted by the vast body of litera- 35
ture (on class, the state, the labour process and political economy) 36
which has attempted to argue the centrality of human agency to any 37
rounded Marxist explanatory model.3 More than this, the thrust of re- 38
cent Marxist writings on class and the state has been informed by the 39
need to take on board the insights derived from feminism, and this 40
has further broadened the parameters of what Parkin calls the ‘con- 41
ceptual storehouse of political economy’ (Sargent 1981, Gilroy 1982). 42
More fundamentally, perhaps, there is little to support Parkin’s 43

492 john solomos
1 assertion that ‘the preferred Marxist response to the fact of racial or

2 communal strife is to ignore it’. On the contrary, a sizeable and grow-
3 ing body of theory and research in the area of race and ethnic rela-
4 tions is based on or draws some inspiration from Marxism. While it
5 may be true that much of the recent debate about class and the state
6 does not say much that is of direct relevance to the question of race,
7 it is strictly speaking not the case that recent Marxist writings ignore
8 divisions within classes or the role of non-class political organisa-
9 tion. The substance of the work of authors such as Nicos Poulantzas,
10 Manuel Castells, Guglielmo Carchedi and Erik Olin Wright recognis-
11 es the reality of such divisions and the role that they play in processes
12 of class formation and in political struggles.4 What is even more clear
13 from these debates is that it is quite mistaken to see Marxism as a
14 monolithic set of assertions or to assimilate it wholesale into some
15 notion of economic determinism or class reductionism. Rather, it is
16 best viewed today as consisting of a spectrum of competing schools
17 of thought ranging from economic determinism to more sophisticat-
18 ed explanatory models which fully recognise the centrality of human
19 agency and collective action (Wright 1980).
20 This view of Marxism as heterogeneous contradicts the oft-stated
21 assertion (which Parkin repeats) that the Marxist approach to racial
22 and ethnic divisions can be identified according to the basic prin-
23 ciples of reducing ‘race’ to class, and the explanation of the origins of
24 racism as co-terminous with the rise of capitalism. Such a view of the
25 Marxist contribution to the study of racism is seemingly supported
26 the class/race model by the close association between the class/race model developed by
27 Oliver C. Cox in his study of Caste, Class and Race (first published in
28 1948) and some more contemporary contributions to the analysis of
29 racism (Sivanandan 1982). Although this is not the place to develop
30 a critical discussion of Cox’s analysis of class and ‘race’, it is impor-
31 tant to point out that his work is by no means seen by contemporary
32 Marxists as an adequate analysis of the complex historical determi-
33 nants of racism or of the relationship between racism and capitalist
34 social relations (see e.g. Gabriel and Ben-Tovim 1978, Miles 1980).
35 Moreover, as Eugene Genovese (1971) has pointed out, Cox’s work
36 was very much the product of his time, in that he was familiar with
37 a Marxism that had not yet been influenced by the work of Gramsci
38 and other ‘Western’ Marxists or by the experience of racial conflict
39 that took place during the 1960s.
40 If Parkin’s dismissive attitude towards Marxism does not hold on
41 the grounds which he suggests, this is not to say that a coherent and
42 fully fledged analysis of racism has been produced from a Marxist
43 perspective. Far from it. Cox’s study, though not self-consciously

written as a Marxist analysis, is still widely considered as the Marxist Cox 1

analysis of this question (see Banton, this volume), largely because 2
it is the most substantive study which attempts to utilise concepts 3
such as class and exploitation in order to explain the role of ‘race’ and 4
racism in capitalist societies. Other studies written from a Marxist 5
perspective have tended to limit their analysis to abstract theoretical 6
exegesis, or to analyse the experience of one particular society in iso- 7
lation. Cox’s attempt to combine theory with a comparative analysis 8
of racism thus stands out as a unique contribution, whose status as a 9
classic sociological analysis is acknowledged by even his most severe 10
critics. 11
There can be no question here of attempting critically to analyse 12
the contribution of Cox to a Marxist analysis of racial and ethnic divi- 13
sions, which is a theme in any case of other papers in this volume 14
and of a growing debate within Marxist circles (Gabriel and Ben- 15
Tovim 1978, Miles 1980, 1982). It needs to be pointed out, however, 16
that the model of Marxism with which Cox was familiar was based 17
on the conceptual baggage of ‘base’ and ‘superstructure’ and an in- 18
strumental view of the state as the agent of the capitalist class (Cox 19
1948, p. 321). This adherence to such views runs counter to the main 20
tendency of contemporary Marxist analysis, which in fact has evolved 21
a number of competing schools of thought, and whose central con- 22
cern is to question the tenability of the classical base-superstructure 23
model as a conceptual framework (P. Anderson 1983). In relation to 24
the question of class, for example, Adam Przeworski has pointed out 25
that the traditional separation between the economic definition of 26
classes and the political and ideological determinants of class-forma- 27
tion is in fact quite misleading when it comes to the concrete analy- 28
sis of the contradictions that arise either within or between social 29
classes. Przeworski argues, and here he expresses a view shared by 30
most neo-Marxist writers, that it is not possible to separate the ‘objec- 31
tive’ analysis of class from the totality of economic, ideological and 32
political relations which organise, disorganise and reorganise social 33
classes as a result of class struggles and historical transformations 34
(Przeworski 1977; but see also Wright 1980). 35
It would be quite mistaken, therefore, to see recent Marxist writ- 36
ings on the question of race and class as deriving from Cox as such. 37
In some cases Cox’s work does form one starting point, but only one 38
among many. It can be argued that equally important influences on 39
recent Marxist writings on ‘race’ are the works of neo-Marxist writ- 40
ers such as Louis Althusser and Nicos Poulantzas, the criticisms lev- 41
elled at economistic Marxism by such writers as John Rex and Edna 42
Bonacich, and the works of feminist writers. All of these influences 43

494 john solomos
1 are evident in the approaches discussed below, although this does

2 not mean that they do not also rely on the conceptual apparatus of
3 classical Marxism and to some extent on the pioneering work of Cox
4 and others (Gabriel and Ben-Tovim 1978). The argument developed
5 in this paper, therefore, will be that there is not one approach to the
6 question of race and class from within the Marxist tradition but rath-
7 er several approaches. The equation of a Marxist approach with the
8 work of Cox, or with a simple form of economic and class reduction-
9 ism, is both mistaken and woefully out of date in the context of re-
10 cent debates about the nature of the state, class and racism. In order
11 to substantiate this point I would like to move on to a critical analysis
12 of three of the most important Marxist approaches to race, class and
13 the state.
14
15
16 3. Neo-Marxist approaches to ‘race’, class and the state
17
18 Within the broad spectrum of recent Marxist or Marxisant approach-
19 es to ‘race’, class and the state it is possible to detect a wide variety
20 of theoretical models, historical analyses and political arguments.
21 Even though using similar theoretical reference points, either to
22 classic texts by Marx and Engels or to the works of more contempo-
23 rary Marxist thinkers such as Nicos Poulantzas, a number of fairly
24 distinct schools of thought have emerged over the last decade. Each
25 of these schools lays a claim to the work of Marx, either as a source
26 of inspiration or more directly as a general theoretical framework
27 within which any analysis of racism in capitalist society must be lo-
28 cated. The complexity of recent debates cannot be adequately anal-
29 ysed within the limits of this paper, but for heuristic purposes I shall
30 discuss three important models that constitute various dimensions
31 three models of recent Marxist debates on ‘race’, class and the state: the relative au-
32 tonomy model, the autonomy model and the migrant labour model.5
33 There can be no question here of attempting a general survey of all
34 the literature that could be classified as falling into these models.
35 Rather the limited objective of this paper is to raise some theoretical
36 problems concerning all three approaches and to make some sugges-
37 tions for an alternative formulation.
38
39 (a) Relative autonomy model
40 Within the last decade, one of the most important and influential
41 redefinitions of the Marxist analysis of ‘race’ and racism has been
42 developed by a number of studies originating from the Birmingham
43 CCCS Centre for Contemporary Cultural Studies (CCCS).6 The works

which have emanated from CCCS over this period are heterogeneous 1
in approach, substantive issues and political inclination but are uni- 2
fied through a common concern with developing an analysis of rac- 3
ism which fully accepts its relative autonomy from class-based social 4
relations and its historical specificity in relation to the laws of mo- 5
tion of capitalist development. Although it would be unwise to label 6
this body of work as a ‘school’ of thought with a coherent and fully 7
worked out framework of analysis, there does seem to be some justi- 8
fication in Brittan and Maynard’s view (1984) that there is a distinct 9
CCCS approach to such issues as racism, sexism and more generally 10
intra-class divisions. Moreover, the theoretical and political contro- 11
versy which surrounded the publication of The Empire Strikes Back in The Empire 12
1982 has resulted in a number of critical articles which question both Strikes Back 13
the theoretical and the political linkages between recent CCCS texts 14
and Marxism (Young 1983, Miles 1984a). 15
The origins of the Centre’s concern with racism can be dated 16
back to the early 1970s, when a number of research students and 17
its then Director, Stuart Hall, became involved in a project which 18
was concerned with explaining the development of ‘moral panics’ 19
about the involvement of young blacks in a specific form of street 20
crime, namely mugging.7 The context of this study was the environ- 21
ment of cities such as Birmingham, where sizeable black communi- 22
ties had grown up and established their own specific community, 23
cultural and political practices. This in turn led to the development 24
of ideological and political responses from within local communities, 25
from the local state and its agencies and from the institutions of the 26
central state. The research carried out by the CCCS team, which was 27
eventually published in 1978 as Policing the Crisis (Hall et al. 1978), Policing the Crisis 28
took as its central concerns the processes by which ‘race’ came to be 29
defined as a ‘social problem’ and the construction of ‘race’ as a politi- 30
cal issue which required state intervention from both the central and 31
the local state. There is no space here to discuss the rich and complex 32
analysis which Hall and his associates developed of this period or the 33
subsequent discussion of these issues by other authors.8 Suffice it to 34
say that the concrete historical analysis on which Policing the Crisis is 35
based provided a materialist basis for what has subsequently become 36
known as the ‘CCCS approach’ to ‘race’ and class and has continued 37
to exert a deep influence on the work of younger researchers at the 38
Centre. This is best exemplified by the jointly produced volume of 39
the CCCS Race and Politics Group, The Empire Strikes Back: Race and 40
Racism in 70s Britain (1982). 41
Before moving on to discuss the more recent work of the Centre, 42
however, it is important to understand the core concepts developed 43

496 john solomos
1 by Stuart Hall and his colleagues in the earlier phase. A purchase

2 on the distinctiveness of this approach can be gained through Hall’s
3 programmatic statement of his position in a paper significantly titled
4 ‘Race, Articulation and Societies Structured in Dominance’ (1980b),
5 which had been widely read and discussed even before it was pub-
6 lished. Hall’s starting point is clear enough, in that he attempts to
7 develop an analytic framework which locates racism in historically
8 specific social relations while allowing for a degree of autonomy of
9 the ‘racial aspects’ of society. He makes this clear when he argues
10 that:
11
12 There is as yet no adequate theory of racism which is capable of deal-
13 ing with both the economic and the structural features of such societ-
14 ies, while at the same time giving a historically concrete and socio-
15 logically specific account of distinctive racial aspects. (Hall, 1980b,
16 p. 336)
17
18 From this starting point he engages in a dialogue with a number
19 Rex of sociological analyses of ‘race’, particularly the work of John Rex,
20 and with the analyses of class ideology and the state which devel-
21 oped under the influence of Althusserian Marxism. At the core of
22 this dialogue are two fundamentally important questions. The most
23 important of these focuses on the relationship between racism and
24 the structural features of capitalist society and asks ‘How does rac-
25 ism function within capitalist social relations and how is it produced/
26 reproduced?’ The second question points to a related but more con-
27 crete set of concerns about how racism is actually constituted in spe-
28 cific societies or institutions, asking ‘How does racism influence the
29 ways in which class, political, gender and other social relationships
30 are actually experienced?’ While the concerns of Hall and his associ-
31 ates in Policing the Crisis are somewhat different from those of the
32 authors of The Empire Strikes Back, for example in relation to the
33 analysis of black youth cultures and the role of the state, they gen-
34 erally agree on the importance of locating the relative autonomy of
35 racism at a macro-level and on the centrality of racism in relations of
36 power and domination in post-war Britain (Hall 1980a, CCCS 1982,
37 chapters 1 and 8).
38 Hall’s reconceptualisation of racism hinges upon a reappraisal
39 both of Marxist concepts and of some aspects of the work of sociolo-
40 gists of ‘race’. In relation to the first he is particularly concerned to
41 draw out the implications of the reconceptualisation of ideology and
42 the state in contemporary Marxism for the analysis of racism. The
43 bulk of his main theoretical paper on the subject begins by support-

ing John Rex’s critique of reductionist Marxist analyses of racism, 1

particularly in relation to South Africa, but then goes on to argue 2
that the emergence of a critical theoretical paradigm within Marxism 3
allows for a more adequate analysis of racism within the context of 4
Marxist theoretical and historical research. Drawing upon studies of 5
imperialism, dependency theory, the state and ideology, he argues: 6
7
A new theoretical paradigm [has emerged], which takes its funda- 8
mental orientation from the problematic of Marx’s, but which seeks, 9
by various theoretical means, to overcome certain of the limitations 10
– economism, reductionism, ‘a priorism’, a lack of historical speci- 11
ficity – which beset certain traditional appropriations of Marxism, 12
which still disfigure the contributions to this field by otherwise dis- 13
tinguished writers, and which have left Marxism vulnerable and ex- 14
posed to effective criticism by many different variants of economic 15
monism and sociological pluralism. (Hall 1980b, p. 336) 16
17
While conceding the criticisms made by Rex (1973, 1983c) and others 18
of a simplistic ‘Marxist’ analysis of racism, Hall wants also to argue 19
that a more critical and multi-dimensional materialist analysis of the critical multi- 20
phenomenon is possible. dimensional 21
In establishing this possibility he himself suggests three prin- materialist analysis 22
ciples as the starting point for a critical Marxist analysis of racism. 23
First, he rejects the idea that racism is a general feature of all human 24
societies, arguing that what actually exist are historically specific rac- 25
isms. Though there may be features common to all racially structured 26
societies, it is necessary to understand what produces these features 27
in each specific historical situation before one can develop a compar- 28
ative analysis of racism. The second principle is that, although rac- 29
ism cannot be reduced to other social relations, one cannot explain 30
racism in abstraction from them. Racism has a relative autonomy 31
from other relations, whether they be economic, political or ideologi- 32
cal. This relative autonomy means that there is no one-way correspon- 33
dence between racism and specific economic or other forms of so- 34
cial relations. Third, Hall criticises a dichotomous view of ‘race’ and 35
‘class’, arguing that in a ‘racially structured’ society it is impossible 36
to understand them through discrete modes of analysis. ‘Race’ has 37
a concrete impact on the class consciousness and organisation of all 38
classes and class factions. But ‘class’ in turn has a reciprocal relation- 39
ship with ‘race’, and it is the articulation between the two which is 40
crucial, not their separateness (Hall 1980b, pp. 336-42). 41
Hall’s own writings on this subject have been fairly limited and 42
programmatic so far, and have moved little beyond the three prin- 43

498 john solomos
1 ciples suggested above. They have been influential, however, in the

2 development of subsequent Marxist studies of racism, partly through
3 the popularised and revised form of his ideas which can be found in
4 The Empire Strikes Back, produced collectively by the Race and Politics
5 Group of CCCS in 1982. Although written at a distance from some
6 of the concerns to be found in the Centre’s earlier work on ‘race’ and
7 from Hall’s theoretical sources, this volume took as its starting point
8 a theme already made familiar by Hall and his colleagues, namely
9 that the political construction of ‘race’ as a problem in contemporary
10 Britain represents an integral aspect of how the British state is at-
11 tempting to manage the current ‘organic crisis of British capitalism’
12 (CCCS 1982, chapters 1 and 8). Drawing particularly upon the work
13 of a number of authors who have attempted to reconceptualise the
14 role of the state in relation to racism (e.g. Carchedi, Sivanandan, and
15 Castells), the authors of this work attempt to rework Hall’s earlier
16 studies and to provide a more concrete analysis of the relation of
17 ‘race’ to British decline during the 1970s.
18 three divergencies The Empire Strikes Back can be said to mark a change from the
19 previous works of the Centre on ‘race’ in at least three senses. First,
20 it argues that previous sociological and Marxist accounts of race rela-
21 tions represent a body of work which has done little to further our
22 knowledge of racism and which can even be seen as reproducing
23 ethnocentric or common-sense views of ‘race’ (CCCS 1982, chapters
24 2 and 8). This mode of critique is in fact quite different from Hall’s
25 critical, but by no means unsympathetic, treatment of the works of
26 ‘sociologists of race’ and their relationship to Marxism. In addition
27 it links up with a more fundamental line of critique emanating from
28 authors such as Cedric Robinson (1983), who sees the central con-
29 cepts of Marxism as ‘Eurocentric’ and fairly limited in their applica-
30 bility to ‘racially structured’ societies.
31 The second divergence relates to the greater emphasis placed on
32 the role of ‘state racism’, or the role of state activity in reproducing
33 racism. While elements of this analysis can be traced back to the work
34 of Hall and his associates (Hall et al. 1978), there is a sharper focus
35 in The Empire Strikes Back on the concrete ways in which the state in-
36 tervened to manage ‘race’ throughout the 1970s, in ways which were
37 detrimental to the interests of black communities. This is achieved
38 at a general level through an analysis of the growth of ‘authoritarian
39 statism’ and ‘popular racism’ within the context of deep-seated crisis:
40
41 The parallel growth of repressive state structures and new racisms
42 has to be located in a non-reductionist manner, within the dynam-
43 ics of both the international crisis of the capitalist world economy,

and the deep-seated structural crisis of the British social formation. 1

(CCCS 1982, p. 9) 2
3
It is also achieved through an emphasis on the ways in which racism 4
structures different areas of social life, notably education, policing, 5
youth policy and also the position of black women in the labour mar- 6
ket (Solomos et al. 1982). 7
This in turn links up to a third area, namely the attempt to recon- 8
ceptualise the complex relationship between ‘class’ and ‘race’. In the 9
concluding chapter of the book, Paul Gilroy mounts a sustained cri- 10
tique of both Marxist and sociological analysis of ‘race’ for failing to 11
deal adequately with the autonomy of ‘race’ from ‘class’. In so doing 12
he questions the view of the working class as a continuous historical 13
subject, particularly since such a view cannot deal adequately with 14
the ways in which blacks can constitute themselves as an autono- 15
mous social force in politics or with the existence of ‘racially demar- 16
cated class factions’ (Gilroy 1982, p. 284). The theoretical basis of 17
this critique can be traced back to the work of Hall, although it also 18
draws some of its inspiration from previous studies at the Centre of 19
working-class culture (Hall et al. 1980) and from the more recent de- 20
bates about class theory within Marxism (see e.g. Przeworski 1977). 21
This is exemplified by the combination in Gilroy’s work of a model of 22
determination which gives class struggle as opposed to class structure a class struggle 23
degree of determinacy, and a view of black workers as racially struc- 24
tured. The difficulties which this position entails are made explicit 25
when Gilroy argues: 26
27
The class character of black struggles is not a result of the fact that 28
blacks are predominantly proletarian, though this is true. It is estab- 29
lished in the fact that their struggles for civil rights, freedom from 30
state harassment, or as waged workers, are instances of the process 31
by which the working class is constituted politically, organised in 32
politics. (Gilroy 1982, p. 302) 33
34
Referring specifically to those excluded from employment, par- 35
ticularly the young black unemployed, he posits that there may be 36
various types of struggles which mobilise them politically, not all of 37
which bear a direct relationship to ‘objective’ conditions. It follows 38
that ‘the privileged place of economic classes in the Marxist theory of 39
history is not to be equated with an a priori assertion of their political 40
primacy in every historical moment’ (Gilroy 1982, p. 303). 41
It is also of some relevance to note, in relation to the above point, 42
that The Empire Strikes Back includes some of the most sustained 43

500 john solomos
1 treatments of the place of gender in the dialectic of ‘race’ and class

2 (see the chapters by Hazel Carby and Pratibha Parmar). Along with
3 the work of Annie Phizacklea on migrant women (Phizacklea 1983),
4 it constitutes an isolated attempt in this field seriously to analyse the
5 role of gender in the articulation of racist ideologies.
6 Perhaps the most notable absence from the Centre’s work on rac-
7 absence of ism is a serious analysis of the political economy of racism. Apart
8 political economy from a rudimentary and limited study by Green (1979), and some
9 minor references in both Policing the Crisis and The Empire Strikes
10 Back, this remains a serious gap in the Centre’s work. It becomes
11 particularly critical in the context of the oft-repeated criticisms made
12 of mainstream sociological studies of ‘race’ for not taking account of
13 the broad economic and social determinants of racism. The empha-
14 sis on the relative autonomy of racism seems to have led to a neglect
15 of the economic context of racial structuration, or at the least to a de-
16 emphasis on the role played by the ‘economic’ in the narrow sense.
17 This is a point that will be discussed later, particularly in relation to
18 the migrant labour model.
19 There are a number of other aspects of the Centre’s work on rac-
20 ism which can be fruitfully discussed (see Freedman 1983-4). Here I
21 have tried to highlight the broad contours of the contribution it has
22 made to a Marxist analysis of racism and to mention some of the
23 ambiguities and tensions that arise. Before taking up the problems
24 to which this model gives rise, it is necessary to outline the other two
25 models.
26
27 (b) Autonomy model
28 Recently some Marxist theorists have argued that there is a need to
29 go beyond the notion that racism is a ‘relatively autonomous’ social
30 phenomenon and to break more definitely from the economic and
31 class-reductionist elements in Marxist theory. Thus a major theme
32 in the influential writings of John Gabriel and Gideon Ben-Tovim,9
33 who have developed a theoretical perspective which specifically em-
34 phasises this point, is that the bulk of neo-Marxist theory on racism is
35 still based on implicit, if not explicit, economic and class-reductionist
36 assumptions. They are particularly critical of the ‘relative autonomy’
37 model, which they see as defective from both a theoretical and a po-
38 litical perspective. From a theoretical angle they see the dichotomy
39 between capitalist social relations and ‘race’ as merely another way of
40 reproducing a more sophisticated form of class-reductionism, under
41 the guise of the nebulous concept of relative autonomy. This in turn
42 is seen as supporting a deterministic analysis of political struggles
43 against racism and thus allowing little room for anti-racist politi-

cal strategies to be effective rather than symbolic (Gabriel and Ben- 1

Tovim 1979, Ben-Tovim et al. 1981a). 2
Contrary to the bulk of recent Marxist writings on racism, which 3
take capitalist social relations and class relations as a starting point, 4
Gabriel and Ben-Tovim argue that racism can best be understood 5
as the product of contemporary and historical struggles which are 6
by no means reducible to wider sets of economic or social relations. 7
This leads them to take as their starting point the various struggles, 8
local and national, political and ideological, which go into the social 9
construction of ‘race’ in specific situations (Gabriel and Ben-Tovim 10
1978). Yet it would be too simplistic to see their position as one which 11
holds that racism is not in some way related to wider social relations. 12
A number of their papers on anti-racist struggles do in fact show how 13
wider structural constraints do play a role in limiting the effective- 14
ness of such struggles (e.g. Ben-Tovim et al. 1981a). What they do 15
argue, however, is that there is no way of determining what these 16
limits are, outside of specific struggles and historical situations. 17
The consequences of this position are that there is no a priori rea- 18
son to see racism as the product of class or economic relations, and 19
that the only way to overcome the traditional dilemma in relation to 20
the ‘base/superstructure’ model is to eschew any attempt to analyse 21
racism outside of its own ideological conditions of existence (Gabriel 22
and Ben-Tovim 1978). In opposition to the preoccupation of the 23
CCCS studies with the linkages between ‘race’ and class, and more 24
concretely with the articulation between capitalist crisis and the de- 25
velopment of racism, Gabriel and Ben-Tovim suggest that the start- 26
ing point of a Marxist analysis should be the ideological and political ideological and 27
practices which work autonomously to produce racism. Rejecting all political practices 28
forms of reductionism they argue that: 29
30
Racism has its own autonomous formation, its own contradictory 31
determinations, its own complex mode of theoretical and ideological 32
production, as well as its repercussions for the class struggle at the 33
levels of the economy and the state. (Gabriel and Ben-Tovim 1978, 34
p. 146) 35
36
In this view, the ideological level is primary, since it is only after the 37
ideological production of racist ideologies that they intervene at the 38
level of the economy and of political practice. In effect, Gabriel and 39
Ben-Tovim attempt to push beyond the constraints of the relative au- 40
tonomy model by questioning the viability of any attempt to situate 41
‘race’ in terms of class. The ‘autonomy’ of racism lies precisely in its 42
irreducibility to any other set of social relations, since any attempt to 43

502 john solomos
1 account for racism in terms of external relations entails a reduction-

2 ist argument (Ben-Tovim et al., this volume).
3 Moving a step beyond this formal critique of reductionist
4 Marxism, supporters of the autonomy model would also argue that
5 their analysis provides a more relevant guide to the complex political
6 racist politics realities of racist politics and anti-racist struggles (Ben-Tovim et al.
7 1981a). Starting from the position that the state as an institution is
8 not monolithic but the site of constant struggles, compromises and
9 administrative decisions, they argue that the most important task of
10 research on ‘race’ is to highlight the political and ideological context
11 in which anti-racist struggles occur. Referring to the need for strug-
12 gles to change institutionalised racism as a ‘long march through the
13 institutions’, with the overall objective of bringing about ‘positive
14 and democratically based political and policy changes to secure the
15 elimination of racial discrimination and disadvantage’ (p. 178), they
16 question the usefulness of the notion of relative autonomy when con-
17 fronted with the complexity of political struggles against racism.
18 This last point is important in understanding the coherence of
19 the analysis developed by Gabriel and Ben-Tovim, since they self-
20 consciously see their theoretical work as linking up with political
21 practice. There is no space here to discuss the detailed and rich anal-
22 ysis they have made of the political context of anti-racist struggles
23 in Liverpool and Wolverhampton. Suffice it to say that the develop-
24 ment of their approach, from the early formal critique of traditional
25 Marxist views of ‘race’ and class to their more recent preoccupation
26 with the local politics and racism, reflects their actual political in-
27 volvement in anti-racist politics.
28 Another way of making this point is that although they would
29 agree with Hall that ‘race’ and ‘class’ form part of a complex dialec-
30 tical relation in contemporary capitalism, they would question the
31 usefulness of interpreting this relationship in terms of the ‘relative
32 autonomy’ of racism. Ultimately they see a contradiction in arguing
33 that racist ideologies have a certain autonomy from material rela-
34 tions, while also holding on to the principle that it is these relations
35 which determine ‘in the last instance’ the degree of autonomy. More
36 fundamentally, they seem to be arguing that even the work of Hall
37 and his associates, with its explicit disavowal of determinism, sup-
38 ports an implicit base/superstructure model.
39 Given their insistence on the irreducibility of ‘race’ to class, and
40 the political conclusions they draw from this position, it may not
41 be surprising that Gabriel and Ben-Tovim do not spend much time
42 discussing the degree of determinancy which state power and class
43 relations have in relation to racial structuration. Their version of the

‘extended’ as opposed to the ‘monolithic’ state does have a rather 1

pluralistic ring about it, at least as regards their discussion of the 2
role of race relations legislation and the role of the local state. Their 3
dual strategy of attrition against racism, both within and outside the 4
state apparatuses, is predicated upon the premise of the primacy of 5
struggle over all other levels of determinancy (Ben-Tovim et al. 1981a, 6
and Ben-Tovim et al., this volume). But this seems to push the agen- 7
cy versus structure argument in the direction of a voluntarist theory 8
of political change, and one which ignores the centrality of the dis- 9
tinction between the appearance and the reality of political struggles 10
(Connolly 1981). Moreover, there seems to be a heavy emphasis in 11
their approach on the importance of policy-oriented research as a 12
tool for anti-racist struggles. The lack of policy-relevance is one of the 13
weaknesses they highlight in other Marxist approaches in this field. 14
The ambiguities of the autonomy model relate as much to po- 15
litical issues as to straightforward theoretical questions. The work 16
of Gabriel and Ben-Tovim can also be read, however, as a theoreti- 17
cal innovation in the sense that it breaks quite fundamentally with 18
the main concern of other Marxists working in this field, namely 19
the search for a non-reductionist and historically specific analysis of 20
racism. For Gabriel and Ben-Tovim the search for a more plausible 21
model of determination leads into a cul-de-sac, and they have re- 22
sponded by rejecting all forms of determination outside of struggle. 23
24
25
(c) Migrant labour model 26
The third explanatory model which has been used by recent Marxist 27
writers, especially by Robert Miles and Annie Phizacklea,10 takes a 28
radically different starting point from the other two approaches out- 29
lined above. Arguing on the basis of a critical reinterpretation of 30
classical and neo-Marxist theories of class, the state and ideology, 31
Miles and Phizacklea construct a theoretical model of racism which 32
prioritises the ‘political economy of migrant labour’ as opposed to the ‘political 33
what they call the ‘race relations problematic’ (Miles 1980, 1982, economy of 34
Phizacklea and Miles 1980, Phizacklea 1984). The substance of migrant labour’ 35
the difference between this approach and the previous two is that 36
throughout their work Miles and Phizacklea seek to prioritise the 37
role that class and production relations play in the reproduction of 38
racism. This position has recently been clearly stated by Phizacklea, 39
who argues: 40
41
If social scientists continue to use the term ‘race’... because people 42
act as though ‘race’ exists, then they are guilty of conferring analytical 43

504 john solomos
1 status on what is nothing more than an ideological construction. Our

2 object of analysis cannot be ‘race in itself’, but the development of
3 racism as an ideology within specific historical and material contexts.
4 (Phizacklea 1984, p. 200)
5
6 This quotation is from an article which bears the title ‘A Sociology
7 of Migration or “Race Relations”?’. In a similar vein Miles argues
8 that the work of some Marxists (notably that of Sivanandan and the
9 authors of The Empire Strikes Back) shares a common terrain with the
10 ‘race relations’ problematic of John Rex, because they both ‘attribute
11 the ideological notion of “race” with a descriptive and explanatory
12 importance’ (Miles, 1984a, p. 218).
13 Central to this position is the notion that racism can only be un-
14 derstood by analysing it in relation to the basic structural features of
15 capitalism. This is linked to a related point, which has been repeat-
16 edly made by Miles and Phizacklea, in relation both to the sociology
17 of race relations and to other Marxist studies of racism. Their work
18 carefully eschews any reference to ‘race’ except in inverted commas,
19 because they see ‘race’ as itself an ideological category which requires
20 explanation and which therefore cannot be used for either analytical
21 or explanatory purposes (Miles 1982, 1984a, Phizacklea 1984). The
22 reason for their insistence on the distinction between ‘race’ and rac-
23 ism becomes clear through their reliance on what they call ‘the pro-
24 racial categorisation cess of racialisation’ or ‘racial categorisation’ (Miles 1982, pp. 153-67,
25 Phizacklea 1984). Broadly speaking, this concept posits that ‘race’ is
26 a social construction which attributes meanings to certain patterns of
27 phenotypical variation. This process of attributing meaning to ‘race’
28 results in a reification of real social relations into ideological catego-
29 ries and leads to the commonsense acceptance that ‘race’ is an objec-
30 tive determinant of the behaviour of black workers or other racially
31 defined social categories. As evidence of this confusion Miles and
32 Phizacklea cite the example of how black workers are not analysed
33 in terms of the social relations of production but as a ‘race apart’
34 (Phizacklea and Miles 1980, Miles 1982), the ways in which politi-
35 cians and governments have utilised the category of ‘race’ in order to
36 obfuscate the reality of racism (Miles and Phizacklea 1984), and at a
37 more concrete level the way in which the use of the idea of ‘race’ to
38 interpret the 1958 riots deflected attention away from the actions of
39 racists against blacks and from the role of the state (Miles 1984b).
40 Precisely because they conceptualise ‘race’ as an ideological reifica-
41 tion, and one which can do little to challenge common-sense images
42 of ‘race’, they suggest two main programmatic conclusions: (a) that
43 ‘race’ cannot be the object of analysis in itself, since it is a social

construction which requires explanation; (b) that the object of analy- 1

sis should be the process of ‘racialisation’ or ‘racial categorisation’, 2
which takes place within the context of specific economic, political 3
and ideological relations. 4
In rejecting the descriptive or analytical value of ‘race’ as a con- 5
cept Miles and Phizacklea insist on the importance of racism, and 6
the discriminatory practices which it produces, as the crucial factor 7
in the formation of what they call a racialised fraction of the work- 8
ing class, and of other classes (Phizacklea and Miles 1980, Miles 9
1984a, pp. 229-30). This has been interpreted as a way of reiterating 10
the role of class determination as opposed to ‘race’, or the use of 11
an economistic version of Marxism to analyse the position of black 12
workers in Britain (Gilroy 1982). In addition, it has been argued that 13
the emphasis that Miles and Phizacklea put on ‘class’ as opposed to 14
‘race’ serves to underplay the role that black struggles play in unify- 15
ing people who ostensibly occupy different class positions (Parmar 16
1982). 17
In rejecting these criticisms Miles has recently attempted to clar- 18
ify the starting point of his work, and its relationship to the work of 19
CCCS and Gabriel and Ben-Tovim (Miles 1984a). Rejecting the view 20
that his work, along with that of Annie Phizacklea, asserts the prima- 21
cy of class over ‘race’, he goes on to argue that his model is grounded 22
in the notion that internal and external class relations are shaped by 23
a complex totality of economic, political and ideological processes. 24
As regards the role of racism in this complex totality he develops a 25
definition of racialisation which differentiates between the economic redefinition of 26
and the political/ideological determinants. Miles explains: racialisation 27
28
The ‘race’/class dichotomy is a false construction. Alternatively, I 29
suggest that the reproduction of class relations involves the determi- 30
nation of internal and external class boundaries by economic, politi- 31
cal and ideological processes. One of the central political and ideo- 32
logical processes in contemporary capitalist societies is the process 33
of racialisation... but this cannot, in itself, over-ride the effects of the 34
relations of production. Hence, the totality of ‘black’ people in Britain 35
cannot be adequately analysed as a ‘race’ outside or in opposition to 36
class relations. Rather, the process by which they are racialised, and 37
react to that racialisation (both of which are political and ideological 38
processes), always occurs in a particular historical and structural con- 39
text, one in which the social relations of production provide the nec- 40
essary and initial framework within which racism has its effects. The 41
outcome may be the formation of racialised class fractions. (Miles 42
1984a, p. 233) 43

506 john solomos
1 What is important about this redefinition is that it (1) locates racism

2 as a process of ideological construction, and (2) prioritises the ‘ef-
3 fects’ of the relations of production. The substantial difference be-
4 tween the migrant labour model and the two previous models lies
5 precisely in the emphasis it places on the ways in which migrant la-
6 bour is included or excluded in terms of the relations of production.
7 In the later works associated with this approach the model of
8 ‘racialisation’ gains an added dimension through comparative refer-
9 ences to the experience of migrant labour in other advanced capitalist
10 societies. This is seen as providing added proof as to the limited na-
11 ture of the race relations approach (Phizacklea 1984). Another area
12 in which Miles and Phizacklea have shown a growing interest is the
13 role of political discourse and ideologies, particularly in relation to
14 the construction of ‘immigration’ and ‘race relations’ as a political
15 problem (Miles and Phizacklea 1984).
16 Miles’s critique of the approach encapsulated in The Empire
17 Strikes Back is a succinct statement of this difference of approach.
18 Distinguishing between the liberal and the radical sociology of ‘race’
19 relations, he identifies the work of John Rex as representing the for-
20 mer and the work of CCCS (1982) as representing the latter. He does
21 so on the ground that both liberal and radical sociologists of ‘race’
22 share the same terrain, i.e. they both hold that ‘race’ is a real political
23 phenomenon with its own effects and determinate relationships, but
24 they are distinguished by the latter’s attachment to Marxism (Miles
25 1984a, p. 218). As a starting point, therefore, Miles argues that while
26 all variants of the sociology of ‘race’ accept the equivalence of class
27 and ‘race’ as analytic concepts, the Marxist position should be that
28 production relations provide the historical and structural context
29 within which racialisation occurs. Although he accepts that in some
30 respects the CCCS authors question the validity of ‘race’ as an ana-
31 lytic concept, he makes the point that this critique is undermined by
32 ‘silence on their emphasis on the importance of cultural as opposed to produc-
33 production tion relations. It is this ‘silence on production relations’ that leads
34 relations’ the CCCS authors, according to Miles, to ignore the material and
35 political basis of racism within the working class (Miles 1984a, pp.
36 228-30).
37 Both of these issues are of some significance, since they highlight
38 a point often repeated by Miles and Phizacklea in their empirical
39 research, namely that blacks are not a ‘race’ apart which has to be
40 related to class but ‘persons whose forms of political struggle can be
41 understood in terms of racialisation within a particular set of produc-
42 tion (class) relations’ (Miles 1984a, p. 230). At any particular time
43 racism can have an autonomous impact, but its effects will be limited
by the wider sets of capitalist social relations.

The migrant labour model diverges drastically from the work of 1

Gabriel and Ben-Tovim, although perhaps less so from the work of 2
Hall or the authors of The Empire Strikes Back. Although it is clearly 3
arguing against a simple reductionism, to the economic or other lev- 4
els, it also consciously avoids the ‘silence on production relations’ 5
which it sees as characteristic of the CCCS school. What is at issue 6
in the migrant labour model is not ‘race’ as such but the racialisation 7
of a specific migrant population in the historical context of post-1945 8
Britain. 9
10
11
4. A critique and an alternative framework 12
13
As argued above, the basic problem confronting any Marxist (and per- 14
haps non-Marxist) account of the complex relations between ‘race’, implications 15
class and the state is to be found in the very nature of ‘racism’ in 16
contemporary capitalist societies. From the brief survey of the com- 17
peting approaches to this question in neo-Marxist discourse it should 18
be clear that there are at least two problems which seem to defy reso- 19
lution. First, the question of the ‘relative autonomy’ or ‘autonomy’ of 20
racial and ethnic categorisations from economic and class determi- 21
nation. Second, the role of the state and the political institutions of 22
capital societies in the reproduction of racism, including the complex 23
role of state intervention in many countries to control immigration, 24
to manage ‘race relations’ and, more broadly, to ‘integrate’ racial and 25
ethnic groupings into the ‘wider society’. Finally, it must be remem- 26
bered that few Marxist writers have ventured beyond theoretical and 27
macro-level analysis, resulting in a mode of analysis that points to 28
contradictions and struggle but says little about the concrete histori- 29
cal and contemporary experience of racism at the level of everyday 30
life and human agency.11 This has meant a notable failure to push 31
Marxist analysis beyond the theoretical understanding of racism 32
towards the practical understanding of how to overcome it, a point 33
noted elsewhere in this volume of Ben-Tovim and his co-authors. 34
Before venturing into a discussion of these implications, how- 35
ever, I want to reiterate that it is far too simplistic to see Marxism as 36
essentially a determinist theory of social development, whether from 37
an economic or a class perspective. Given the wide currency which 38
is still given to such a view of Marxism within the race relations lit- 39
erature (see e.g. Jeffcoate 1984), and the tendency to search for an 40
essentialist theory of racism in some Marxist writings, it may be as 41
well to note that numerous schools of thought within Marxism have 42
been established precisely in opposition to a determinist interpreta- 43

508 john solomos
1 tion of Marxist theory. Many of the most challenging Marxist studies

2 of the state, ideology, social class and specific historical events over
3 the last two decades have attempted to develop an analytic framework
4 and empirical analyses which question deterministic models of poli-
5 tics and society (Wright 1980, Anderson 1983, Jessop 1982, 1983).
6 Moreover, there is by now a sizeable body of empirical and historical
7 studies which have relied on Marxist analytic concepts in order to
8 analyse specific aspects of advanced industrial societies (Anderson
9 1983, Burawoy and Skocpol 1983). Taken together these two bodies
10 of literature bear ample witness to the vitality and complexity of neo-
11 Marxist theory Marxist theory and to the futility of trying to construct an analytic
12 of racism framework of racism which is acceptable to all Marxists. What fol-
13 lows therefore are some suggestions which are meant to draw to-
14 gether strands of argument which were developed in the previous
15 sections and to open up questions for debate.
16 Now, if the arguments developed above are accepted, one must
17 ask what kind of theory of racism is possible within a Marxist frame-
18 work if each kind of racism has to be analysed in relation to its his-
19 torical and socio-political context. Bearing in mind the critical obser-
20 vations about the three analytic models discussed above, I want to
21 draw briefly on a point first made by Stuart Hall and his colleagues
22 and recently taken up by a number of other authors, namely that in
23 post-1945 Britain:
24
25 Race is intrinsic to the manner in which the black labouring classes
26 are complexly constituted... Race enters into the way black labour,
27 male and female, is distributed as economic agents on the level of
28 economic practice – and the class struggles which result from it into
29 the way the fractions of the black labouring class are constituted as
30 a set of political forces in the ‘theatre of politics’ – and the political
31 struggle which results; and in the manner in which the class is ar-
32 ticulated as the collective and individual ‘subjects’ of emergent ide-
33 ologies and forms of consciousness – and the struggle over ideology,
34 culture and consciousness which results. This gives the matter of
35 race and racism a theoretical as well as a practical centrality to all
36 the relations and practices which affect black labour. The constitution
37 of this class fraction as a class, and the class relations which inscribe it,
38 function as race relations. The two are inseparable. Race is the modality
39 in which class is lived. It is also the medium in which class relations are
40 experienced. (Hall et al. 1978, p. 394, emphasis added)
41
42 This reconceptualisation of the class-‘race’ dialectic is certainly awk-
43 ward, and represents a programmatic statement rather than a fully

worked-out framework of analysis. But it has the merit of focusing on 1

racism as a specific social relation and on the need to analyse the his- 2
torical conditions which make distinctions based on ‘race’ and ethnic 3
origins an important issue in a specific society. In addition it serves 4
to highlight the weakness of the accusation that Marxist accounts of 5
the ‘race’-class dialectic are necessarily deterministic. There is, how- 6
ever, a degree of obfuscation in the argument that ‘race’ and class 7
relations are inseparable, since this tells us little about the specificity 8
of either, or of the historical processes which produce this complex 9
structure in dominance.12 In the end the approach suggested by Hall, 10
and by subsequent CCCS work, does little to show the specificity of 11
racism, or to analyse the ‘work’ which racism accomplishes (Hall et 12
al. 1978, p. 338). It merely suggests ways of reworking the categories 13
of Marxist analysis in such a way as to account for the complex real- 14
ity of racial categorisation in contemporary capitalism, and it does 15
not tackle thorny problems in the definition of ‘relative autonomy’. 16
It has thus been criticised for being too abstract and ahistorical in its 17
analysis of the role of black labour in Britain and of migrant labour 18
more generally. 19
In considering this problem the work of Gabriel and Ben-Tovim 20
suggests the most straightforward resolution. Arguing that the choice 21
between ‘determinism’ and relative autonomy is a false one, they go 22
on to reject the whole idea of a ‘society structured in dominance’ be- 23
cause they see it as introducing a base/superstructure model by the 24
back door. By implication they argue that the central question is not 25
the relationship between racism and the wider social totality but the 26
conceptualisation of racism as the object of struggle in historically historically defined 27
defined conditions (Ben-Tovim et al., this volume). Another resolu- conditions 28
tion is suggested by the work of Miles and Phizacklea. They reject 29
the problematic of ‘race in itself’ and concentrate their analysis on 30
the development and reproduction of racism as an ideology based 31
on specific political, economic and ideological relations (Miles and 32
Phizacklea 1984, Phizacklea 1984). In essence this second approach 33
sees attempts to analyse the ‘interrelationship between race and 34
class’ as based on the false premise that these two categories have 35
the same analytical significance, while in fact racism is but one of the 36
means which transform the positions occupied by class fractions in 37
capitalist societies (Miles 1984a, pp. 228-9). 38
The work of the autonomy and migrant labour schools, like that 39
of CCCS and Sivanandan, does indeed raise the questions which re- 40
main unclear in much of the Marxist discussion of ‘race’ and class. 41
But they all do so within fairly limited parameters, and they have by 42
no means exhausted the potential for a more rigorous formulation 43

510 john solomos
1 of the theoretical problems confronting a Marxist analysis of rac-

2 ism. What follows are some tentative suggestions about how to build
3 upon and move beyond the parameters of recent debates.
4 It is not my intention to develop a fully fledged alternative frame-
5 work for dealing with the issues raised in the previous discussion.
6 Rather, the limited objective here is to draw out some of the im-
7 plications of the criticisms made above for a critical analysis of the
8 dialectic of ‘race’, class and state. I want to concentrate, particularly,
9 on the problems which arise in trying to utilise a Marxist analytic
10 framework for explaining racism, by outlining a conceptual model
11 three propositions which holds that: (a) there is no problem of ‘race relations’ which can
12 be thought of separately from the structural (economic, political and
13 ideological) features of capitalist society; (b) there can be no general
14 Marxist theory of racism, since each historical situation needs to be
15 analysed in its own specificity; and (c) ‘racial’ and ‘ethnic’ divisions
16 cannot be reduced to or seen as completely determined by the struc-
17 tural contradictions of capitalist societies.
18 In broad outline these three propositions are meant to establish
19 the interconnectedness of racism with wider social relations, while
20 allowing for a degree of autonomy and discontinuity. This in itself
21 does not take us very far in establishing the actual nature of disconti-
22 nuities in an empirical sense, and indeed this is perhaps impossible
23 without comparative and national studies of different kinds of rac-
24 ism. But it seems to me to be important that the three propositions
25 remain interlinked, because short of this it is only possible to achieve
26 a one-dimensional analysis of racism and not the dialectical and dy-
27 namic approach which Bonacich (1980) rightly identifies as the basic
28 feature of Marxist approaches to ‘race’.
29 Nevertheless, it should be clear from the above discussion that all
30 three propositions are essentially contested among Marxists. While
31 propositions (b) and (c) can be said to have a wide currency in one
32 form or another, there is much dispute about (a), whether at a macro-
33 level or through specific debates about the relationship of ‘race’ and
34 class. Gabriel and Ben-Tovim would dispute the relevance of point
35 (a) in relation to a concrete analysis of racist ideologies. The problem
36 remains, however, that economic and social conditions do play a role
37 in structuring racism as an ideology and as a set of practices in spe-
38 cific institutions. If this is accepted, and to some extent even Gabriel
39 and Ben-Tovim accept that there are limits on the effectiveness of
40 struggles against racism, then the question arises of how one con-
41 ceptualises the relationship between ideologies and social structures.
42 Is it simply a question of an eclectic combination of autonomous lev-
43 els in a specific situation? Or do economic, political and ideological

relations exercise some determining influence on the expression of 1

racist ideologies? 2
The fundamental problem with abandoning the relative autonomy 3
model is that of avoiding the trap of a simple pluralism, which sees 4
‘race’ and class relationships as completely separate. This is why it 5
seems to me that it is important to insist on the complexity of ‘deter- the complexity of 6
mination in the last instance’, while accepting that there is some form determinism 7
of determination of racism by other social relations. For example, 8
within the context of economic decline and political crisis-manage- 9
ment during the post-war period, can one really talk of the complete 10
autonomy of racism? Or can one separate out the political meanings 11
which are attached to ‘race’ today from the actions of successive gov- 12
ernments in defining and redefining the immigration/‘race’ issue 13
during the last four decades? Or can one understand the long-term 14
patterns of inclusion and exclusion of black workers in the labour 15
market without an analysis of the restructuring of British industry 16
during this period? 17
For these reasons alone there are grounds for questioning wheth- 18
er a pluralistic version of Marxism is any more adequate in analys- 19
ing the contradictions of racial structuration than pluralistic theories 20
have been in their analysis of capitalist societies (Meiksins Wood 21
1983, Connolly 1981). Additionally, however, there seems to be little 22
possibility that the autonomy model can capture the complexity of 23
power relations or adequately analyse the historical context in which 24
racism has become entrenched, in different societies and at different 25
times, at all levels of the social formation. In this sense I am less wor- 26
ried about the distance Gabriel and Ben-Tovim have travelled from 27
classical Marxism than about the fact that their model does not seem 28
to be able to analyse the development of racism except through the 29
ever-present concept of ‘struggle’, which is not located in any social 30
context. 31
Perhaps one way of dealing with the issue of determinism may be 32
through a strict application of proposition (b), namely that there can 33
be no general theory of racism. It is precisely on this point that there 34
hinges the possibility of further advance in Marxist theory, since it 35
focuses attention on the contexts in which racist ideologies develop 36
and are transformed, or on what Gilroy has called ‘the construction, 37
mobilisation, and pertinence of different forms of racist ideology and 38
structuration in specific historical circumstances’ (Gilroy 1982, p. 39
281). But the application of this position has led to the emergence of 40
more problems, since few Marxists have actually analysed processes 41
of racial structuration at the level of actual societies. The example of 42
South Africa is one which has attracted most attention (Wolpe 1980, 43

512 john solomos
1 Burawoy 1981), along with some aspects of racism in contemporary

2 Britain. This had led to a tendency to produce more refined concepts
3 without approaching the more thorny questions relating to their ap-
4 plicability to actual concrete situations (Rex 1981).
5 The unsettling nature of the encounter between contemporary
6 Marxism and ‘race’ is far from reaching a conclusion. This is reflect-
7 ed in the numerous either/or kind of formulations which have been
8 summarised and criticised above; for example do we talk of ‘race’
9 or class, ‘race’ or racism, autonomy or relative autonomy, ‘race’ or
10 Marxism and class ‘migrant labour’? This type of debate is prominent in the early stages
11 of theoretical discussion, when there is uncertainty about the exact
12 nature of differences and agreements across the main contestants.
13 The further development of debate, however, would require greater
14 specification of the social relations of racism in specific societies, and
15 its interconnections with class and non-class aspects of social real-
16 ity (Resnick and Wolff, 1982). Once the question is defined in this
17 way it also becomes clear that, although it is important in a specifi-
18 cally Marxist framework to establish some degree of determination,
19 Marxian theory is also radically anti-determinist.
20 It needs to be said that there are numerous aspects of recent de-
21 bates which have not been fully covered in the above discussion. All
22 three theoretical models, for example, are closely linked to differing
23 assessments of the role of the state, of politics and of the possibil-
24 ity of anti-racist struggles. The role of autonomous black political
25 struggles in relation to class-based political action remains a central
26 area of dispute, as does the issue of the role of state intervention in
27 the area of ‘race relations’. Many of these issues are also the object of
28 lively discussion outside of Marxism (Rex 1981). These are questions,
29 however, which need to be addressed separately, since they relate to
30 more specific assessments of the political economy of contemporary
31 Britain.
32
33
34 5. Conclusion
35
36 This paper has tried to locate the position of ‘race’ in Marxist discourse
37 and to assess the adequacy of the various theoretical approaches to
38 its study in capitalist societies. While much of the recent literature
39 written from the various perspectives analysed above hardly merits
40 the designation of a ‘Marxist theory of race and ethnic relations,’ it
41 clearly represents a large and growing body of work. I have tried to
42 argue that Marxist theories of ‘race’ are heterogeneous in approach,
43 though it can be argued that they are unified through a common

concern with (a) the material and ideological basis of racism and 1
racial oppression, however it may be defined, and (b) the role that 2
racism plays in structuring the entire social, political and economic 3
structures of societies. In other words, the basic level of agreement 4
between the various Marxist approaches is that they accept that there 5
is no race relations problem as such, that there is no problem of rac- 6
ism which can be thought of as separate from the structural features 7
of capitalist society. 8
Equally important, however, are the differences in approach 9
which have become evident over the last decade within the broad 10
spectrum of Marxist writings on racism. It is in this context that 11
we can best appreciate the studies discussed above. Whatever their 12
theoretical deficiencies and analytic weaknesses, the overall effect of 13
Marxist contributions in this area has been to redefine the problem of redefine problem 14
‘race’ in capitalist society in a way that makes theoretical and political of racism 15
debate more open and challenging. They have focused attention on 16
the history and contemporary reality of racism in capitalist society, 17
and its complex economic, political and ideological preconditions. 18
By questioning the adequacy of both traditional Marxist and non- 19
Marxist treatments of racism, and by emphasising the need for link- 20
ing theoretical analysis to anti-racist politics, these studies have in 21
their different ways helped reinstate the idea that racism is no mere 22
epiphenomenon but a social construct resulting from the complex 23
social relationships and economic and political structures of capital- 24
ist societies (Hall et al. 1980, Freedman 1983-4, Miles 1984a). 25
But the interest of these studies is not restricted to the field of 26
Marxist theory and politics. For the problems with which they have 27
been grappling occur in similar forms in non-Marxist social and 28
political theory. For although the basic starting point of Marxist ap- 29
proaches to this question may be said to differ markedly from the 30
various non-Marxist approaches, there can be little doubt that many 31
of the substantive analytical problems are actually quite similar. This 32
is not to say that the specific theoretical and analytical divergences 33
between the two sets of approaches are not important, for they clearly 34
are. What is at issue, however, is the adequacy of the explanations 35
they offer about the role of racism in contemporary capitalist societ- 36
ies, the role of the state in reproducing or countering racist practices, 37
and the adequacy of the political conclusions they draw about how 38
to overcome racism. Because the Marxist approaches have focused 39
on the social relations that produce and reproduce racism, they have 40
touched upon issues which are of concern to non-Marxist theorists, 41
namely the origins of racist ideologies and institutions and the role 42
of political power relations. In so doing, recent Marxist analyses may 43

514 john solomos
1 well open up possibilities for broadening out the debates about ‘race’,
2 class and the state in potentially fruitful directions.
3 Perhaps in the long run this will be seen as one of the main
4 achievements of recent Marxist debates on racism. The kinds of
5 question which they raise about theory and anti-racist politics open
6 up the possibility for reflective discussions of the role of racism in
7 contemporary societies and the strategies for overcoming it. The
8 theoretical and political selfconsciousness which the approaches dis-
9 cussed above show are a fundamental challenge to both traditional
10 Marxism and rival problematics within the social sciences, and one
11 which deserves to be taken up across a variety of disciplines. In ad-
12 dition, however, they have provided an extra impetus to attempts to
13 link academic research to questions of practice, particularly in rela-
14 tion to political struggles against racism. In so doing they have posed
15 questions beyond the limits of traditional Marxist class analysis and
16 have pointed to the need for a deeper analysis of non-class forms of
17 domination.
18 If this brief sketch of the content of recent Marxist debates on
19 ‘race’, class and the state is accurate, there are many questions about
20 the specificity of racism which have been inadequately theorised.
21 But recent debates have at least opened up the possibility of a more
22 dynamic and accessible Marxist contribution to the analysis of rac-
23 ism. Whether this possibility is realised depends on the success of
24 attempts to broaden the horizons of current Marxian conceptions of
25 the dynamics of advanced capitalism. Along with gender, racism re-
26 mains one of the key axes on which this reconceptualisation has to
27 take place, both at the level of theory and at that of practice.
28
29
30 Notes
31
32 1 Apart from the work of Parkin, which is discussed below, see Forsythe 1979,
33 Stone 1977, Bonacich 1980, Brotz 1983, Banton 1983.
2 It is not possible to discuss these issues specifically in the context of this
34 paper, but valuable and provocative overviews of all of them can be found in
35 Wright 1980, Sargent 1981, Resnick and Wolff 1982 and Cottrell 1984.
36 3 The dialectic of ‘agency’ and structure in Marxist thinking is usefully dis-
37 cussed in Gintis and Bowles (1981), where it is argued that there are usually
38 two opposing tendencies in Marxist writing, one based on a commitment
to structural determination and another committed to a notion of practice.
39
They themselves suggest a resolution in terms of a unified conception of
40 structure and practice.
41 4 A useful and challenging discussion of the political context of their analysis
42 can be found in Jessop 1982. But see also Meiksins Wood 1983.
43 5 This threefold classification is imposed and reflects an assessment of the
main tendency in each body of work. There are no doubt other models which

can be usefully discussed, or other points on which these three approaches 1

could be sub-divided. Nevertheless for the purposes of this presentation this 2
classification seemed most appropriate. See also Bonacich 1980, G. Morgan 3
1981 and Omi and Winant 1983.
4
6 A somewhat broader overview of the Centre’s work on this can be found in
Freedman 1983-4. On the work of the Centre more generally see the edited 5
volume, Culture, Media and Language, by Hall et al. 1980, and Johnson 1983. 6
7 The concern with racism can be traced back further in terms of Hall’s own 7
work, but the impact of ‘race’ on the Centre’s project dates from this period 8
and therefore predates Hall’s more theoretical studies of racism and social
9
relations.
8 A fuller discussion of this point can be found in Solomos et al. 1982. 10
9 Throughout this paper I refer to the work of Gabriel and Ben-Tovim, though 11
in fact much of their work has been carried out with a number of other re- 12
searchers associated with their work in Wolverhampton and Liverpool. On 13
the theoretical origins of the criticisms which this model develops in relation 14
to relative autonomy see Cutler et al. 1977-8, Hindess 1984, and more gener-
15
ally the work associated with Barry Hindess and Paul Hirst.
10 In a recent paper Phizacklea argues that there are links between this position 16
and the broader tradition of the sociology of migration which has developed 17
in both Europe and the USA (Phizacklea 1984). 18
11 The relative absence of historical awareness and specificity from much of 19
the Marxist debate on racism has been noted, from rather different angles,
20
by Rex 1981, Bonacich 1981a and 1981b, Miles 1982, Robinson 1983, and
Brittan and Maynard 1984. What is surprising, however, is that despite this 21
awareness few attempts have been made to redress the balance and develop 22
historically based analyses of racist ideologies and practices. 23
12 This is a problem discussed from a different perspective by G. Morgan 1981 24
and Green 1979. For an interesting American perspective see Omi and 25
Winant 1983.
26
27
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Racism, migration and the state in Western 3
4
Europe: a case for comparative analysis*
5
6
Frank Bovenkerk, Robert Miles and Gilles Verbunt 7
8
9
10
11
12
In the 1980s, the criminologist Frank Bovenkerk and the sociologists Robert
13
Miles and Gilles Verbunt embarked on an ambitious project. They sought to 14
compare post-war migration to Western Europe and the political and ideo- 15
logical responses that this migration elicited. The project was undertaken by 16
developing a theoretical framework that was broad enough to encompass the 17
historical specificity of – and between – particular cases, while still permit- 18
ting a general explanation that was sensitive to the specificity and variation. 19
Bovenkerk, Miles and Verbunt’s framework revolves around the formation of 20
the nation state and highlights the state’s role in the reproduction of the na- 21
tion as imagined community. These processes, they suggest, are embedded 22
in a more universal process of the regulation of scarcity. Critical here are the 23
inclusion and exclusion of people from the hierarchy of political, economic 24
and ideological positions in the nationstate. 25
26
27
28
Introduction 29
30
Given the still vivid memory of the holocaust, and following a long 31
period of affluence and relative social order after the Second World 32
War, it was widely believed within Western Europe during the 1960s 33
and 1970s that racism, and related ideologies, had been permanently 34
eliminated. But to the surprise of many, the ‘race’ myth (in old or the ‘race’ myth 35
new forms) has gained renewed support. It is being suggested once 36
more that the origin of long-standing and emergent economic and 37
cultural problems lies in the presence of groups who do not ‘belong’ 38
to the nation-state by virtue of biological, social and/or cultural char- 39
acteristics that they are thought to possess inherently. Although the 40
complex of ideas is spread more widely, they have found formal po- 41
litical expression in all countries of Western Europe in different ways 42
and at different times. 43

518 frank bovenkerk, robert miles and gilles verbunt
1 Taking the example of the far-right political agitation and the re-
2 surgence of neo-fascist groups, these were first evident in Britain and
3 Switzerland (National Front, the Schwarzenbach referendum) in the
4 1960s, other countries such as Holland and France (Centrumpartij,
5 Front National) followed in the 1970s, and last have been Belgium
6 and the Federal Republic of Germany in the 1980s (Vlaams Blok,
7 Republikaner). Immigrants, or more accurately, certain categories
8 of immigrants, have become the main targets of the hostility either
9 generated or fostered by this agitation and these political organisa-
10 tions. Our primary interest in this paper is with the interrelation-
11 ship between the common appearance of this hostility and agitation
12 throughout Western Europe and the differentiation in the mode and
13 timing of expression in the constituent countries. Further, it pro-
14 vides an outline agenda for research.
15
16
17 Common themes, dissimilar expressions
18
19 There are remarkable similarities in the manifestation of hostility to
20 certain migrant groups in the various countries of Western Europe.
21 There are demands to stop or ‘control’ immigration; a movement to
22 send migrants back (‘to assist in their repatriation’); a demand for
23 the withdrawal of political and social rights; a quest for repressive
24 measures to curb ‘ethnic’ and ‘racial’ crime, and so on. At the same
25 time, there are considerable differences between these countries
26 a variety of with respect to the content, timing and progress of various forms of
27 anti-immigrant anti-immigrant sentiment.
28 sentiments For example, in Britain, its surface content seems to be domi-
29 nated by a discourse on ‘race’ as a biological entity, by conceptions
30 of unassimilability on the grounds of cultural or national origin as in
31 Switzerland, France, Belgium and the Federal Republic of Germany,
32 and on perceptions of social undesirability as in the Netherlands.
33 However, a more detailed analysis of the ideological content of this
34 agitation is necessary, and may reveal a more deep-seated arrange-
35 ment of similar themes and content, even if the specific interrela-
36 tionship varies from country to country.
37 Concerning timing, political opposition to certain categories of
38 migrant began earlier in Britain where serious political consideration
39 was given (although in secret) to stopping ‘coloured’ immigration
40 (but not immigration from, for example, the Republic of Ireland) as
41 soon as it began in the late 1940s. Controls were eventually intro-
42 duced in 1962. Politicians justified their decisions on the grounds
43 that they would improve ‘race relations’ (e.g. Joshi and Carter 1984;

MacDonald 1987). The first political concerns in France were ex- 1

pressed publicly in the late 1960s, but the debate on immigration 2
control and, more generally, the politicisation of the migrant pres- 3
ence, did not begin until the early 1970s (Freeman 1979; Verbunt 4
1985). In the Netherlands, following the expression of public and 5
political concern over political terrorism during the mid-1970s by a 6
small Moluccan group, the introduction of an extensive social policy 7
programme for ‘minorities’ was legitimated by an accompanying 8
decision to seriously curtail further immigration in the early 1980s 9
(Groenendijk 1988). 10
There are also differences in the progress of increasing elec- 11
toral support for right-wing and neo-fascist political parties. Racist 12
voting recently increased significantly in the Federal Republic of 13
Germany and in Belgium, but seems to have declined or stabilised 14
so far in England and France following its growth in the 1970s and 15
early 1980s respectively (e.g. Fielding 1981; Ogden 1987). It has re- 16
emerged in the late 1980s, after apparent dormancy, in Switzerland 17
and the Netherlands (e.g. Donselaar and Praag 1983). Also marked 18
is the difference in the penetration of political racism into the formal 19
political system and, especially significant is the variety of ways in 20
which the major established and governing political parties have re- 21
acted, ranging from rejection to the incorporation of anti-immigrant 22
themes in their agitation and propaganda. 23
The generality of the phenomenon suggests common causes and 24
we believe that, in so far as there are, they are to be found in impor- 25
tant changes within the capitalist mode of production and in political 26
strategies to respond to and reverse the economic crises of the early 27
1970s. But the extent of diversity is equally impressive: the specific 28
relationship between economic, political and ideological dynamics 29
clearly varies from one country to another. It follows that arguments 30
which advance a simple, linear determination in which racism or 31
other forms of anti-immigrant sentiment are explained as a func- 32
tional product of a particular economic development such as the eco- 33
nomic crisis (Castles and Kosack 1973) have, at best, only a limited 34
utility. Because there is considerable variation in the nature, extent 35
and pace of the politicisation of the migrant presence, an explana- historical 36
tion must also be historically specific if one wishes to grasp both the explanations 37
complex whole and the nature of the reaction in each country. 38
This paper constitutes a first, preliminary step in effecting this 39
task. We begin to construct a theoretical framework on the founda- 40
tion of a set of assumptions which are transhistorical, but in order 41
to offer explanations which take account of historical specificity. The 42
Marxist tradition, characterised by historical materialism, has been 43

1 an important source of inspiration for this perspective. For example,

2 Marx argued concerning production:
3
4 However, all epochs of production have certain common traits, com-
5 mon characteristics. Production in general is an abstraction, but a ra-
6 tional abstraction in so far as it really brings out and fixes the com-
7 mon element and thus saves us repetition. Still this general category,
8 this common element sifted out by comparison, is itself segmented
9 many times over and splits into different determinations. Some de-
10 terminations belong to all epochs, others only to a few. (Marx 1973:
11 85)
12
13 Thus, a central analytical assumption is that certain relations are
14 specificity found in all social formations, but they always exist only in specific,
15 historically constituted forms, and within that specificity lies differ-
16 entiation and variability which also requires explanation.
17 More recently, Sayer has sought to retrieve and highlight the his-
18 torical dimension of Marx’s theory in the context of the influence of
19 structuralism during the 1960s and 1970s (Sayer 1983, 1987). In the
20 light of this, we consider it to be necessary to formulate a theoreti-
21 cal framework which identifies the relevant historical dynamics and
22 processes both generally and in each specific instance. Hence, the
23 guiding principle is the dialectic between historical generality and
24 specificity.
25 Thus, it is only by recognising the generality of certain processes
26 which characterise the historical development of the capitalist mode
27 of production (e.g. capital accumulation, nation-state formation and
28 reproduction, labour shortage, migration etc.) alongside the equally
29 important search for the historically specific forms that these take
30 that a full explanation can be found. For example, the expression of
31 anti-immigrant feeling is a generality, but it takes a particular form
32 in each country. For instance, the degree to which these expressions
33 are ‘racialised’, i.e. defined in terms of a discourse of ‘race’, may
34 vary. By following this dialectical method, we seek to avoid not only
35 the rather futile debate about whether there is more racism in one
36 country compared with another, but also the conclusion that because
37 there is racism and economic crisis in each country there is little or
38 nothing more that requires explanation.
39 For reasons given in the following section, we shall concentrate
40 our analysis on what we believe to be the all-important, or even deci-
41 sive, influence of the state. It is our aim to explain the general char-
42 acter of these phenomena, and their specific manifestations and de-
43 velopment. The methodological instrument is comparative analysis

and, because our subject is considerably complex, we wish to hold 1

constant as many potential explanada as possible (compare with 2
Przeworski and Teune’s ‘most similar systems approach’, 1970) in 3
order to maximise the theoretical scope of the analysis. However, it 4
would be presumptuous to announce that we are the first to use com- 5
parative analysis. Elsewhere (Bovenkerk, Miles and Verbunt 1990) 6
we have reviewed critically the major comparative works on migra- 7
tion and racism in Europe that focus on the role of the state. 8
The specific, empirical object of analysis is state responses to mi- state responses 9
gration, and to the ideological and political reaction to migration, in 10
three countries in Western Europe since 1945: Britain, France and 11
the Netherlands. These three countries have been chosen because 12
they were all prominent participants in the historical emergence of 13
capitalism in Europe, a process that included the creation of nation- 14
states as political units and of colonial empires. They all remain 15
amongst the group of most ‘advanced’ capitalist societies, sharing 16
a mode of production which is now characterised by an interdepen- 17
dence of capital accumulation and rapid technological change, a ‘wel- 18
fare state’ and a form of representative government based upon uni- 19
versal suffrage. These three countries are to be distinguished from 20
other West European countries, most notably in this context by their 21
colonial histories which are evident in the contemporary period in 22
the form of the settlement of colonial migrants (although they share 23
the experience of other migration movements). In the light of histori- 24
cal and cultural explanations for racism, which place particular ex- 25
planatory significance upon the colonial enterprise and experience, 26
the relevance of this factor is self-evident. 27
Furthermore, since 1945, all three countries have witnessed four 28
analytically distinct (though, logically not mutually exclusive) migra- 29
tion movements that have been the object of different forms of state 30
regulation: (1) of owners of wealth, along with managerial and tech- 31
nical staff of international companies; (2) of (industrial) workers; (3) 32
of colonial subjects; and (4) of refugees. In all three countries, the 33
combined numbers of resulting settlers are very small, being around 34
5 to 7 per cent of the total population. 35
36
37
The significance of the state 38
39
Against the background of (a) changes in production relations and 40
in the political power structure on a world scale that has led inter alia 41
to decolonisation and (b) the crisis of accumulation that developed 42
from the end of the 1960s and that fundamentally restructured both 43

1 migration and capital flows, our interest lies in analysing the interre-
2 lationship between two political and ideological processes. These are,
3 first, the manner in which processes of migration and consequent
4 settlement of a significant proportion of migrants within Western
5 Europe have been structured by direct and active state intervention.
6 Second, we examine the extent to which the nature and content of
7 the political and ideological reaction to migration and settlement has
8 also been shaped by state involvement.
9 It should be emphasised that neither the process of migration,
10 nor theories of migration, are the object of our study here. We focus
11 on state reactions to migration and, in terms of methodology, the
12 state intervention ‘stimulus’ to state intervention is considered to be sufficiently similar
13 to be conceived of as constant. This procedure is warranted on the
14 level of abstraction that we have identified. Thus, although owners
15 of capital and company management may originate from quite dif-
16 ferent countries (United States of America, Canada, Japan etc.), they
17 belong to the same ‘class of migration’ by virtue of their function in
18 the spatial restructuring of production relations.
19 For example, colonial migrants to Britain originate mainly from
20 the British Caribbean, the Indian sub-continent and East African
21 countries; those who came to the Netherlands left Indonesia and the
22 Dutch Caribbean; French colonial immigrants originated from North
23 Africa and the French Caribbean. Although their migration histo-
24 ries may differ, all these groups took part in a migration movement
25 that has been closely linked to the decolonisation process and that
26 has depended upon special political links with the ‘mother country’.
27 Those who have come as migrant labourers proper have responded
28 to labour shortages that have been produced by a certain state in
29 the post-war accumulation process. They may come from Turkey,
30 Eastern Europe or Italy etc., but they also belong to a same ‘migration
31 class’ of workers. Finally, political refugees who have gained access
32 in small numbers in the three countries under study tend to origi-
33 nate from the same background. All three countries have migrants
34 from, for example, Hungary, Vietnam, Chile and Sri Lanka. It should
35 be clear that these four categories need not be mutually exclusive
36 (for instance, both Britain and France have imported migrant labour
37 from their colonies) but they can be analytically separated in so far as
38 they constitute distinct categories to which the state has reacted in all
39 three instances.
40 We use the concept of state to refer to an institutional complex
41 which comprises minimally government, bureaucratic administra-
42 tion, judiciary, police and military forces. These collectively claim and
43 use power to structure a particular ensemble of economic, social and

political relations within a specified spatial unit and to mediate the 1

impact of exterior forces upon that unit. Historically, within capitalist 2
societies, the state bureaucracy has expanded to include additional 3
apparatuses concerned with the provision of education and medi- 4
cal care, with the redistribution of material resources in the form of 5
welfare and unemployment cash payments, and with the provision 6
of a collective material infrastructure for commodity production and 7
exchange. 8
This definition could be understood as representing the state as a 9
monolithic unity. In fact, state activity and intervention consist of a 10
specific action or a complex of actions on the part of a person or per- 11
sons within one or more of these institutions. The possibility of con- 12
testation both within and between different institutions is assumed 13
rather than a purposive, consistent strategy followed by a complex 14
but cohesive unit. Furthermore, if the state is conceived as, in part, a 15
reflection of the contradictions within the social formation, it follows 16
that state activity can be the outcome of internal struggle and com- 17
promise. 18
If we take the function of the state to be to guarantee and safe- 19
guard the reproduction of the dominant mode of production, then 20
this role entails the organisation of not only economic but also po- 21
litical and ideological practices, and the regulation of structural and 22
conjunctural contradictions. Moreover, economic relations can never 23
be divorced absolutely from political and ideological relations. For 24
example, the operation of the market has certain political conditions 25
of existence: in so far as the functioning of the market is dependent 26
upon exchange regulated by contracts enforceable by law, and in so 27
far as resistance must be suppressed, then the state (as a political 28
institution) is an essential relation of production, a condition of exis- 29
tence of the market (Corrigan et al. 1980; Rueschemeyer and Evans 30
1985). Thus, within the capitalist mode of production, the role of the 31
state should not be conceived in a narrow economistic manner but 32
also in terms of the reproduction of certain essential political and 33
ideological conditions and relations. 34
There is an empirical and a theoretical reason for our emphasis 35
on state intervention. The development of capitalism is paralleled 36
historically by an expansion in the size and complexity of the state 37
apparatus, and by an increase in its power to regulate the range and state power 38
scope of actions of individuals and classes. A large part of our empiri- 39
cal motivation to study state intervention is our contention that its 40
influence on the social, economic and political position of migrants 41
is far greater than has been recognised by scholars who have studied 42
the migration process so far. The same holds true for the develop- 43

1 ment of anti-immigrant ideologies and related practices: rather than

2 studying the extent of ‘prejudice’ among (segments of) the indige-
3 nous populations or measuring the incidence of discriminatory prac-
4 tices, we seek to show that the way in which the state regulates migra-
5 tion processes, and consequent political and ideological processes, is
6 in the end decisive.
7 Part of the explanation for this belated recognition may be the
8 fact that much research on migration and ‘prejudice’ has been fund-
9 ed by state institutions themselves. Governments that have socially
10 constructed problems of ‘ethnic minority formation’ or ‘racial intol-
11 erance’ and that spend money on scientific research to document
12 such phenomena, can hardly be expected to invite scholars to analyse
13 their own preconceptions. This may be more true in some countries
14 (the Netherlands, Sweden, Federal Republic of Germany) than oth-
15 ers (France, Britain). This shows that the organisation of scientific
16 research on these matters is an integral part of state activity and it
17 should therefore be part of our comparative research project.
18 Additionally, the ‘welfare state’ constitutes the apex of state power
19 thus far within Western-type democracies. This has given rise to re-
20 newed theoretical interest in the role and activities of the state on the
21 part of Marxist and Weberian scholars (e.g. Anderson 1979; Evans
22 et al. 1985) and has led to heated political discussion between social
23 democrats, socialists and neoconservatives about the nature and rel-
24 evance of state intervention (e.g. Keane 1984, 1988: 1-30).
25 The escalation of intervention to regulate international migra-
26 tion (Plender 1988: 61-93) provides an excellent illustration of in-
27 immigration creasing state power. Immigration controls are, in historical terms,
28 controls a very recent phenomenon. Broadly speaking, the era of politically
29 unrestricted migration and entry in Western Europe ends only at the
30 beginning of the twentieth century. For example, although signifi-
31 cant state controls were first introduced in Britain in an Aliens Act of
32 1793 in an attempt to restrict the entry of refugees from the French
33 Revolution, the provisions in the Act were weakened by legislation in
34 1824 and 1826. Consequently, the nineteenth century is now gener-
35 ally regarded as lacking effective state controls over migration, and
36 restrictive state controls begin with the Aliens Act of 1905 (Plender
37 1972: 39-50; MacDonald 1987: 7).
38 Concerning the Netherlands, foreigners were freely admitted
39 until the first half of the nineteenth century and a law of 1798 ex-
40 plicitly granted a number of freedoms. The first Aliens Act was en-
41 acted in 1849. However, as it was rarely used in practice, it was not
42 until two bills were passed in 1918 and 1920 that an effective sys-
43 tem of immigration control was established. It should be noted that

this legislation was not so much motivated by worries about unre- 1

stricted immigration as it was by concern about the revolutionary 2
events throughout Europe in 1848 and the failure of revolutionary 3
movements in Germany and the Netherlands in 1918 (Swart 1978; 4
Lucassen and Penninx 1985). 5
The first measures in France date from 1893 (Withol de Wenden 6
1988: 24-28). They were intended to register the number of foreign- 7
ers resident in France, and their places of work and residence, in 8
order to provide protection for native French workers. Controls over 9
entry into France were institutionalised in 1906. It should be noted 10
that, during this period, private companies, especially those involved 11
in mining and metal production, determined and organised immi- 12
gration rather than the government. Direct state intervention began 13
in 1914, and thereafter immigration was promoted in order to com- 14
pensate for the loss of the male work force in the First World War. 15
These brief remarks on the history of immigration controls pro- 16
vide a preliminary illustration of the reasons why we have chosen to 17
place so much emphasis upon state intervention. Given that, since 18
the seventeenth century, the world has been increasingly divided spa- 19
tially into nation-states where, since the nineteenth century, these 20
separate populations have been constructed legally and ideologically 21
by the legal categories of nationality and citizenship (Plender 1988: 22
4-6, 9), states constitute the institutions which regulate international 23
spatial movements of people. Citizens of other nation-states are pre- 24
vented, permitted or encouraged to cross national boundaries as a 25
result of decisions by governments exercising political sovereignty 26
within specified territories. Moreover, it is within the jurisdiction of 27
the states that conditions of continued residence (or return) of those 28
who are not citizens are determined. 29
All this highlights the gatekeeper role of the state. This role has the gatekeeper role 30
become increasingly significant as capitalist expansion has taken as 31
one of its forms the export of industrial capital which has, in turn, 32
intensified the longer-term development of an international labour 33
market. This has been facilitated by technical development, which 34
has helped to create the possibility of fast and efficient long-distance 35
transport, and by the increasing awareness of enormous differences 36
in wealth and compliance with human rights which has motivated 37
people to seek refuge in other lands and continents. Both processes 38
have been overdetermined by the development of worldwide com- 39
munication systems. 40
Second, the economic and social circumstances of the population 41
living within the boundary of the nation are no less determined by 42
state decisions. Education, housing, welfare and other aspects of re- 43

1 production have become, to varying degrees, state responsibilities,

2 and the government has become one of the largest single employers.
3 the collective Here we refer to the welfare or collective consumption role of the state.
4 consumption role Just as people are divided into citizens, noncitizens and other legal
5 categories by decisions of the state functioning as gatekeeper, the col-
6 lective consumption function also implies the distribution of scarce
7 resources and services, which can also only be effected by establish-
8 ing criteria of eligibility. These criteria may or may not include some
9 or all of the population of immigrant origin. In this respect, a critical
10 issue arises in circumstances where migrants settle in one nation-
11 state, but retain a legal status as a national or citizen of another.
12 Because migrants are legally aliens in their country of residence, their
13 access to rights and resources is usually restricted in comparison to
14 those allocated to nationals.
15 A third role of the state apparatus is to maintain social order and
16 the law and order sustain a democratic legitimation. This constitutes the law and order
17 role role of the state. The exercise of this function impinges upon popula-
18 tions of migrant origin in different ways. When immigration is fol-
19 lowed by resistance and conflict, either by sections of the indigenous
20 population opposing the migrant presence or by migrants resisting
21 racism and discrimination, state intervention will be required to reg-
22 ulate the ensuing disorder. The nature of this regulation may have
23 enormous consequences for the quality of life of the immigrants (for
24 example, Commission for Racial Equality 1986). To take another
25 striking example of the importance of state intervention, the state
26 has considerable powers to signify certain activities as illegal or to
27 persecute specific criminal offences, that is to criminalise designated
28 activities and groups, in the process of maintaining social order. If
29 such groups happen to be of migrant origin (and they often are),
30 criminalisation may have far-reaching effects on the wider social rep-
31 resentation of that group at large (Hall et al. 1978; CCCS 1982).
32
33
34 A theoretical background
35
36 Against this background, we proceed to sketch a transhistorical
37 theoretical framework which will lead us to a preliminary agenda
38 of comparative research. The general processes will be deducted on
39 the basis of political economy theory and a related conception of the
40 nation-state.
41 The reproduction of all forms of social organisation depends
42 upon, first, production of the means of human existence and, sec-
43 ond, the maintenance of mechanisms to regulate (relative and ab-

solute) scarcity in relation to socially defined human needs, that is, 1

processes of distribution. Relations of production and distribution are 2
therefore essential relations in all modes of production and in all 3
social formations. Whilst we acknowledge that (because there can be 4
no distribution without prior production) production is the superor- 5
dinate element, ‘the point of departure’ as Marx (1973: 89, also 94) 6
put it, and that ‘production predominates not only over itself ... but 7
over the other moments as well’ (Marx 1973: 99), our problematic 8
here leads us to focus upon the regulation of scarcity and therefore 9
upon processes of distribution by which scarce goods are allocated. 10
This follows from our focus not on the determinants of migration 11
per se, but on state responses to migration, a process that we conceive 12
broadly to encompass various dimensions of inclusion and exclusion 13
relating to the mobility of human beings across national boundaries 14
and their temporary or permanent settlement within nation-states 15
other than that of their birth. 16
The regulation of scarcity implies decisions about who are to re- 17
ceive or share and who are not. This is expressed in the twin concepts 18
of inclusion and exclusion. Hence, effective regulation requires the inclusion and 19
creation of a hierarchy by which people are organised into distinct exclusion 20
collectivities in order to effect the uneven distribution of scarce re- 21
sources. This requires the interrelated processes of signification and 22
categorisation, whereby certain characteristics are chosen to effect and 23
legitimate a process of differentiation. These characteristics are then 24
utilised to typify individuals and sort them in groups (Miles 1989). 25
This process of allocation includes people in so far as they are placed 26
in privileged positions and in so far as they receive scarce resources; 27
and it excludes them in so far as they are placed in a disadvantaged 28
position and as they are denied resources. 29
The regulation of scarcity is an economic issue in so far as hu- 30
man material needs must be met in order to guarantee social re- 31
production. But it is not only an economic process. It is a political 32
question in so far as human choices are made as to whom and to 33
what available resources are to be allocated. And it is an ideological 34
question in so far as cultural and biological characteristics (real and 35
imagined), in combination with economic position, are signified and 36
reified as criteria of differentiation in the process of the allocation of 37
resources. Hence, when analysing the capitalist mode of production 38
(or any other), we do not conceive of scarcity exclusively in terms of 39
the distribution of forces of production or concrete commodities: it 40
is a condition that is also evidenced, for example, in the processes by 41
which individuals are distributed to different economic positions, in 42
relation to the acquisition of juridical status and citizenship, and in 43

1 the determination of eligibility for access to state-allocated resources.

2 The concepts of inclusion and exclusion are central to under-
3 class formation standing the processes of class formation and reproduction. For ex-
4 and reproduction ample, within the capitalist mode of production, a large proportion of
5 the population are excluded from access to the means of production.
6 Historically, these have been made scarce by the social processes of
7 dispossession and the concentration of ownership in few hands (that
8 is, by historically concrete acts of exclusion). While, abstractly, this is
9 understood to result in the formation of two classes (the bourgeoisie
10 and the proletariat), in reality this dichotomy is better understood
11 as a more complex hierarchy. Thus, there are important gradations
12 within the ranks of those who own and control the means of produc-
13 tion (e.g. ‘big’ capital and ‘small’ capital, finance and manufactur-
14 ing capital), while those who have only their labour power to sell are
15 able to do so with different (socially produced) abilities and skills
16 (e.g. manual as compared with non-manual labour, the possession of
17 trade skills as compared with no formal skills, etc.). There are, in ad-
18 dition, those who are excluded from access to wage labour, and those
19 who utilise their own labour power to exploit the means of produc-
20 tion on a small scale (i.e. the petite bourgeoisie).
21 The processes of class formation and reproduction have two main
22 dimensions which are analytically distinct (although in reality closely
23 interrelated), the formation and reproduction of the positions in the
24 structure and the distribution of people to occupy those positions.
25 The distinction is embodied in Marx’s claim that:
26
27 The individual comes into the world possessing neither capital nor
28 land. Social distribution assigns him at birth to wage labour. But this
29 situation of being assigned is itself a consequence of the existence
30 of capital and land property as independent agents of production.
31 (1973: 96)
32
33 The positions in the structure are established by the mode of produc-
34 tion (although the mode of production is not a natural given but the
35 result of previous class struggles). The process of ‘social distribution’
36 is characterised by inclusionary and exclusionary dimensions.
37 For example, with respect to the hierarchy of positions within the
38 proletariat, the outcome of the distribution process depends not only
39 upon which positions are available in what quantity (e.g. determined
40 by the circumstances of the labour market) but also upon individu-
41 al capacities such as physical strength, linguistic skills and the way
42 in which individuals have been prepared for these positions by, for
43 example, the institutions of the family and the educational system.

There are, in other words, many dimensions of human differentia- 1

tion (both real and imagined) which serve as pre-selectors or criteria 2
of pre-emption, or which mediate between human potentials and ac- 3
cess to positions and resources. 4
The complex processes of class formation and reproduction with- 5
in the capitalist mode of production are based on these processes 6
of inclusion and exclusion. And while they do not encompass the 7
totality of inclusionary and exclusionary processes, they do constitute 8
the foundation upon which all others rest. This is because social life 9
is only possible if human material needs are met, and this requires 10
some process of production (of food, shelter etc.): the social relations 11
which are organised in the process of establishing and reproducing a 12
system of production are therefore prior to and so constitutive of (but 13
do not necessarily determine) all other social relations. Furthermore, 14
the additional dimensions and consequences of inclusion and exclu- 15
sion cannot be detached absolutely from the collectivities of class be- 16
cause those who are their object are not thereby displaced from the 17
occupation of any position relative to the means of production. 18
Nevertheless, there are other dimensions of human differentia- dimensions 19
tion that have been signified in the allocation of people to structural of human 20
positions relative to production in the context of scarcity. Sexual dif- differentiation 21
ference is one and a gendered division of labour (understood in rela- 22
tion to waged labour but also in relation to unpaid domestic labour) is 23
one of its results. Phenotypical characteristics (often signified by the 24
idea of ‘race’ which is a central element in discourse within Europe, 25
North America, Australia and South Africa) are also widely signified 26
to both exclude certain groups of people from access to wage labour 27
positions when these have been scarce, and to include other groups 28
of people when recruiting in situations of labour shortage. This ra- 29
cialisation of the process of class formation gives rise to a racialised 30
labour market (Miles 1989). Other aspects of human differentiation 31
include age, physical capacity, subculture or way of life, religion, and 32
language, all of which are associated with segregation on the labour 33
market. Some of these properties are valued positively, others nega- 34
tively. 35
What is true for the labour market holds for access to all other 36
scarce resources, including those of a political and ideological charac- 37
ter (e.g. citizenship, access to the media, protection against physical 38
attack, the issue of residence permits). The allocation or distribution 39
process comprises the totality of human decisions on access to scarce 40
resources that are based on varying combinations of evaluated prop- 41
erties within a given social context. Within these combinations, in 42
which not all dimensions of evaluation need to be present, some of 43

1 them override others. A particular distribution process can be char-

2 acterised by the hierarchy of properties that have been weighted.
3 Class formation (in association with processes of gendering, ra-
4 cialisation and other forms of differentiation and exclusion) within
5 the capitalist mode of production has occurred historically within
6 distinctive spatial and political units that are called nation-states.
7 Within these, certain cultural characteristics (e.g. a specific language,
8 legal system, religion) have been constructed and signified by the
9 dominant class as universal attributes of the ‘nation’. This ideologi-
10 cal process implies the establishment of criteria which serve as a fur-
11 ther measure of inclusion and exclusion, that is, as a measure of
12 ‘belonging’ to the nation.
13 From the late eighteenth century, these cultural characteristics
14 have been interpreted as given and natural. Moreover, certain bio-
15 logical characteristics have been typified as indicators of the exis-
16 tence of the ‘nation’. Both serve to constitute an imagined community
17 (Anderson 1983), whereby a specific collection of people that nor-
18 mally would not know each other personally nevertheless believe that
19 they share a common identity, and therefore a common heritage and
20 future. Hence, members of all social classes, including the proletari-
21 at, within a territorial boundary tend to consider themselves as shar-
22 ing something essential with each other. This sense of identification
23 has been reinforced by the foundation of forms of political represen-
24 tation, the creation of citizenship and a complex of state institutions,
25 such as schools, that educate all those defined as belonging to the
26 nationalisation nation. We refer to this as a process of nationalisation in the sense
27 explicated by Nairn (1988: 281).
28 The creation of national identity around specific characteristics
29 serves not only as an inclusionary process within the nation-state.
30 It defines by implication Others inside and outside the nation-state.
31 Thus, the boundaries of the nation-state have been marked not only
32 by the specific form given to state institutions, but also by the signifi-
33 cation by the dominant class of cultural symbols which exclude those
34 with a different cultural profile. It follows that competition between
35 each national bourgeoisie was not only economic but also cultural in
36 form for each believed that it was the agent of a distinct and supe-
37 rior civilisation. The formation of nation-states is therefore the con-
38 sequence of a combination of ideological signification and struggles
39 by culturally specific dominant classes to gain and retain access to
40 scarce resources within a defined space by representing certain char-
41 acteristics as signifying a collective interest.
42 It follows that the nation-state has a political and ideological real-
43 ity which is dependent upon an international process of inclusion

and exclusion. As a result, nationality generally has become a highly 1

significant juridical status and each specific instance has a peculiar 2
quality of scarcity. In a world divided into nation-states, nationality 3
has become a juridical status that most people acquire at birth and 4
that is based on descent (jus sanguinis) and that is linked to residence 5
in that nation-state. Moreover, it is a juridical status which, in as- 6
sociation with the notion of citizenship, usually carries with it rights 7
to specific forms of economic and political participation within the 8
nation-state. 9
Its significance, and its scarcity, only become apparent when peo- 10
ple cross the boundary of the nation-state and take up residence in 11
another. Such mobility is not automatically followed by the acquisi- 12
tion of a new juridical status. Access to this status is state-regulated 13
(qua gatekeeper) which imposes certain conditions which, in turn, 14
ensure that the quality of scarcity is retained. 15
16
17
A preliminary research agenda 18
19
So far we have identified four categories of migrants that are present 20
in the three countries studied and we have distinguished three forms 21
of state regulation that impinge upon the management of migration 22
flows. Further, we have argued that mechanisms of inclusion and ex- 23
clusion should be the central focus in seeking a comparative under- 24
standing of state responses to both migration and the political and 25
ideological responses to migration. We are not only or not primarily 26
concerned with actual policy measures and practices, but rather with 27
the ideological constructions that lie behind, or are embedded in, 28
these policies and practices. Our comparative research effort would 29
concentrate on identifying the real or assumed properties of catego- 30
ries of people that are subject to inclusion and exclusion. 31
Through deductive reasoning, we have left the space to study his- deductive reasoning 32
torical specificity. In seeking to study exclusionary and inclusionary 33
processes in three different nation-states, we therefore do not as- 34
sume a priori that this is effected by the signification of one factor 35
(for example, skin colour) in all three cases. In other words, it may 36
not be racism per se that is the main mechanism to keep people out 37
of, or in an inferior position within, the nation-state. For example, 38
in Britain much of the official and public discourse dealing with and 39
responding to post-1945 migration has been dominated by the no- 40
tions of ‘race’ and colour. But French discourse on the same matter 41
seems much less concerned with ‘race’ as such, and tends to refer to 42
culture and religion. The discourse in the Netherlands is about eth- 43

1 nic minorities and reflects concerns about social undesirability and

2 therefore we wish to investigate empirically the value of the concept
3 of minorisation (Rath 1988). The degree of exclusion and the mecha-
4 nisms by which exclusion is effected may not differ a great deal from
5 one country to another, but the discourse does. This reflects sepa-
6 rate national traditions and sensitivities and this needs to be clearly
7 grasped conceptually and analytically.
8 Concerning the access of migrants to the three countries, a first
9 question would be: on what conditions have the four categories been
10 admitted by the state? But it is equally important to ascertain who has
11 been denied access. Thereafter, one should investigate the grounds
12 upon which people have been accepted or refused entry? This re-
13 quires a detailed study of government sources (both public and pri-
14 vate) in order to analyse the ideological content and discourse em-
15 ployed in effecting and legitimating differential migration controls.
16 research questions This research is now in progress within our research group.
17 As for the management of the migrants’ presence, a first priority
18 would be to study the meaning and the history of the various words
19 that have been employed to identify ‘these people’, following their
20 permanent settlement in the three countries. A general term for
21 larger categories of migrants is omnipresent in Europe. Instead of
22 differentiating by referring to Pakistani British, Dutch Surinamers
23 or French people of Algerian origin, there is always a generic term:
24 travailleurs immigrés, ethnic minorities, foreign workers, (‘coloured’)
25 immigrants etc. Explaining the origin and meaning of these different
26 terms constitutes a significant task within our problematic.
27 Furthermore, it seems strategic to select a single dimension of
28 legal status which may be considered as an act of inclusion within
29 the nation-state. Hence, that issue would be closely connected to
30 conceptions of citizenship and national belonging. For example, two
31 case studies would be particularly appropriate, one on the debate on
32 citizenship and the other on the debate of voting rights for foreign
33 citizens. Again, the specific measures themselves would be of less
34 interest to us (and not only because they have been listed already,
35 see Brubaker 1989 and Layton-Henry 1989), because we are more
36 interested in the ideological construction of the pluriform reality that
37 is hidden in the discourse.
38 A further central dimension of the state’s role in managing the
39 presence of migrants concerns its reaction to their own political ac-
40 tivity. Specifically, it would be strategic to study forms of political
41 action around issues that are signified as potentially challenging or
42 subverting the unity implied of the nation-state as an ‘imagined com-
43 munity’. For example, within all three countries, there is now a sig-

nificant Muslim presence as a result of post-1945 migration flows, 1

and their faith is considered to be alien to European nation-states 2
wherein the Christian religion continues to be signified as a central 3
dimension of the official national identity. All Muslims demand spe- 4
cific rights (to open mosques, ritual slaughter, segregation in schools 5
etc.) and meet resistance. We might then compare the ways in which 6
the state of our three countries has dealt with Muslim demands. 7
The migrant presence has also become the object of hostile politi- exclusionary acts 8
cal activity from within the ‘indigenous’ population. Certain forms 9
of such action constitute a challenge to ‘law and order’, to which the 10
state must necessarily respond. For example, in all three countries 11
extreme right-wing and neofascist parties have organised racist cam- 12
paigns against the migrant presence and have sought to gain elec- 13
toral support for a range of racist demands, including ‘repatriation’. 14
We might therefore ask how the state has responded to these exclu- 15
sionary political actions in each of the three countries? Second, and 16
not unrelated to the first instance, certain migrant populations have 17
become the object of violence, which might be considered to be one 18
of the ultimate acts of exclusion. Again, we might consider to what 19
extent the state has formally acknowledged such exclusionary acts 20
by defining it as a ‘law and order’ problem, along with the measures 21
taken to prevent such violence. 22
23
24
Conclusion 25
26
Within each of the Western European nation-states, it has been pub- 27
licly recognised to varying degrees that the post-1945 experience of 28
migration has been paralleled by the expression of hostility and re- 29
sistance which has commonly taken a racist form. Furthermore, so- 30
cial scientists of various disciplinary backgrounds have recognised 31
the importance and value of a comparative analysis of these migra- 32
tions and their political and ideological consequences, and there is 33
now an escalating literature devoted to such research. Elsewhere, we 34
have critically evaluated an important part of this body of literature, 35
concluding that the comparative method employed in most cases 36
has been significantly flawed (Bovenkerk, Miles and Verbunt 1990). 37
Adequate comparative analysis presumes a conceptual framework 38
which is formulated at a level of generality which encompasses the 39
historical specificity of, and variation between, particular instances, 40
yet which permits a general explanation which is sensitive to that 41
specificity and variation. The general theoretical approach outlined 42
above, and the illustrative research agenda, are offered in the light of 43
this critique and objective.

1 • This paper is the momentary outcome of a discussion that has continued for
2 a period of more than three years and that has involved the present authors,
3 a group of young researchers and established scholars in three countries. Its
aim is to formulate a design for an ambitious comparative research project
4
based on insights grounded in political economy theory. Whilst bearing sole
5 responsibility for the content of this paper, the authors wish to acknowledge
6 the assistance of Paula Cleary, Moustapha Diop, Han Entzinger, Marjan
7 van Hunnik, Francien Keers, Jan Rath, Marel Rietman, John Schuster and
8 Jeanne Singer-Kerel who have at various stages participated in a series of
seminars in Utrecht and Paris at which the arguments set out here were
9
discussed.
10
11
12
13 References
14
15 ANDERSON, B. 1983. Imagined Communities. London: Verso.
16 ANDERSON, P. 1979. Lineages of the Absolutist State. London:
17 Verso.
18 BOVENKERK, F., MILES, R. and VERBUNT, G. Forthcoming 1991.
19 ‘Comparative Studies of Migration and Racism in Western Europe:
20 A Critical Appraisal’. International Migration Review.
21 BRUBAKER, W.R. ed. 1989./mmigralion and the Politics of Citizenship
22 in Europe and North America. Lanham: University Press of America.
23 CASTLES, S. and KOSACK, G. 1973. Immigrant Workers and Class
24 Structure in Western Europe. London: Oxford University Press.
25 CCCS 1982. The Empire Strikes Back. London: Hutchinson.
26 Commission for Racial Equality 1986. Immigration Control Procedures:
27 Report of a Formal Investigation. London: Commission for Racial
28 Equality.
29 CORRIGAN, P. et al. 1980. ‘The State as a Relation of Production’, in
30 Corrigan, P. (ed.), Capitalism, State Formation and Marxist Theory:
31 Historical Investigations. London: Quartet Books.
32 DONSELAAR, J. and PRAAG, C. 1983. Stemmen op de Centrumpartij:
33 De Opkomst van Antivreemdelingenpartijen in Nederland. Leiden:
34 C.O.M.T. (Rijksuniversiteit Leiden).
35 EVANS, P.B., RUESCHEMEYER, D. and SKOCPOL, T. 1985.
36 Bringing the State Back In. Cambridge: Cambridge University
37 Press.
38 FIELDING, N. 1981. The National Front. London: Routledge and
39 Kegan Paul.
40 FREEMAN, G. 1979. Immigrant Labour and Racial Conflict in
41 Industrial Societies: the French and British Experience, 1945-1975.
42 Princeton: Princeton University Press.
43 GROENENDIJK, K. 1988. ‘Migratiebeheersing, controle en discrim-

inatiebestrijding: de dubbelzinnigheid van het overheidsbeleid’. 1

Migrantenrecht 2. 2
HALL, S. et al. 1978. Policing the Crisis. London: Hutchinson. 3
JOSHI, S. and CARTER, B. 1984. ‘The Role of Labour in the Creation 4
of a Racist Britain’. Race and Class 25 (3). 5
KEANE, J. 1984. Public Life and Late Capitalism: Toward a Socialist 6
Theory of Democracy. Cambridge: Cambridge University Press. 7
KEANE, J. 1988. Democracy and Civil Society: On the Predicaments of 8
European Socialism, the Prospects for Democracy, and the Problem of 9
Controlling Social and Political Power. London: 10
Verso. 11
LAYTON-HENRY, Z. 1989. The Political Rights of Migrant Workers in 12
Western Europe. London: Sage. 13
LUCASSEN, J. and PENNINX, R. 1985. Nieuwkomers: Immigranten en 14
hun nakomelingen in Nederland 1550-1985. Amsterdam: Meulenhoff 15
Informatief. 16
MACDONALD, I. 1987. Immigration Law and Practice. London: 17
Butterworths. 18
MARX, K. 1973. Grundrisse. Harmondsworth: Penguin. 19
MILES, R. 1989. Racism. London: Routledge. 20
NAIRN, T. 1988. The Enchanted Glass: Britain and its Monarchy. 21
London: Radius. 22
OGDEN, P. 1987. ‘Immigration, Cities and the Geography of the 23
National Front in France’, in Glebe, G. and O’Loughlin, J. (eds.), 24
Foreign Minorities in Continental European Cities. Wiesbaden: 25
Steiner Verlag. 26
PLENDER, R. 1972. International Migration Law. Leiden: Sijhoff. 27
PLENDER, R. 1988. International Migration Law. Dordrecht: 28
Martinus Nijhoff. 29
PRZEWORSKI, A. and TEUNE, H. 1970. The Logic of Comparative 30
Social Inquiry. New York: Wiley. 31
RATH, J. 1988. ‘Minorisation in the Netherlands: the Political 32
Participation of Immigrants’. Paper presented to Intercongress 33
Meeting of the Research Committee on Migration of the 34
International Sociological Association, University of Utrecht, 30 35
March – 1 April. 36
RUESCHEMEYER, D. and EVANS, P.B. 1985. ‘The State and 37
Economic Transformation: Toward an Analysis of the Conditions 38
Underlying Effective Intervention’, in Evans, P.B., Rueschemeyer, 39
D. and Skocpol, T. (eds.), Bringing the State Back in. Cambridge: 40
SAYER, D. 1983. Marx’s Method: Ideology, Science and Critique in 42
‘Capital’. Brighton: Harvester Press. 43

1 SAYER, D. 1987. The Violence of Abstraction: The Analytic Foundations

2 of Historical Materialism. Oxford: Basil Blackwell.
3 SWART, A.H. 1978. De toelating en uitzetting van vreemdelingen.
4 Deventer: Kluwer. VERBUNT, G. 1985. ‘France’, in Hammar,
5 T. (ed.), European Immigration Policy: A Comparative Study.
6 Cambridge: Cambridge University Press.
7 WIHTOL DE WENDEN, C. 1988. Les immigrés et la politique. Paris:
8 Presses de la Fondation Nationale des Sciences Politiques.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Migration, racism and ‘postmodern’ capitalism1 3
4
Robert Miles and Victor Satzewich 5
6
7
8
9
10
11
12
This article was part of an ongoing debate on the nature of racism. The soci- 13
ologists Robert Miles and Vic Satzewich positioned themselves against the 14
more vulgar Marxist, functionalist explanations of racism. According to their 15
explanations, racism is not an independent phenomenon, but the product of 16
the divide-and-rule policy of the bourgeoisie and its agent – the state. Miles 17
and Satzewich argued, however, that racism did not originate as a conspiracy 18
by capitalists. The ruling class gained no benefit from conflicts within the 19
working class. Neither the capitalists nor the state had an interest in stir- 20
ring up working-class racism. According to Miles and Satzewich, the work- 21
ing class was fragmented long before there was any immigration. They also 22
opposed the assumption that the development of racism had been linear, as 23
racism is a far from homogeneous phenomenon. 24
25
26
27
Introduction 28
29
In these ‘new times’, it has become de riguer to undertake a re-ex- 30
amination of the theories that Marxists have been using to analyse 31
contemporary capitalism, its ‘laws of motion’ and its future develop- 32
ment. Certainly, there has been a major reorganization of the capital- 33
ist accumulation process over the past decade or more and that this 34
has had significant implications for, inter alia, international migra- 35
tion flows. The general assumption is that, with the ending of the 36
expansionary boom in the early 1970s, the era of large-scale labour 37
migration from the ‘periphery’ to the ‘centre’ of the world economic 38
system also terminated (e.g. Salt 1987: 241). If this assumption were 39
to be correct, it might be concluded that there is no longer any object capitalism and 40
for a theory of the interplay between capitalism and migration. migration 41
But is the assumption correct? There are commonsense and anec- 42
dotal reasons to question not only the assumption that labour migra- 43

538 robert miles and victor satzewich
1 tion has ceased but also the assumption that the only migration flows
2 that occur are composed of persons seeking to sell their labour power
3 for a wage. For example, during the 1980s, the British government
4 has had to respond to the attempt on the part of refugees from Sri
5 Lanka to seek asylum in Britain and with a continuing movement of
6 refugees from Vietnam to Hong Kong, one of Britain’s last remain-
7 ing colonies. And political events in China in 1989 raised again the
8 question of why it is that UK passport holders in Hong Kong do not
9 have the right to migrate and settle in Britain.
10 To take another recent example, during the 1980s, there have
11 been large-scale movements of population from East Germany,
12 Poland and the Soviet Union into West Germany, movements sup-
13 ported and encouraged by the West German state which simultane-
14 ously sought to deny entry to political refugees from Sri Lanka on the
15 grounds that the ‘boat is already too full’. This evidence points not
16 only to the continuing reality of international migration flows as an
17 empirical phenomenon but also to a qualitative theoretical problem.
18 Because these ‘refugee’ migrations do not have their origin, at least
19 not in any direct form, in the capital accumulation process, then they
20 cannot be conceptualized within theories of migration which priori-
21 tize that process as the determinant force.
22 There is also academic evidence to consider. The ‘postmodern’
23 world capitalist system is characterized by the domination of multi-
24 national companies and a new international division of labour. And,
25 as theorists of the latter have emphasized (e.g. Fröbel et al. 1980),
26 mass commodity production has not ceased, but rather has to a sig-
27 nificant degree been relocated in Export Processing Zones in the ‘pe-
28 ripheries’ of the world economic system. And, as Sassen (1988) has
29 shown, this process of capital export has stimulated a new phase of
30 migration and proletarianization within those peripheries as well as
31 to the United States.
32 This leads us to suggest that the European experience of the
33 nineteenth century has not so much been overtaken by a new ep-
34 transfer of mass och but is being extended to spatial locations which previously es-
35 commodity caped the interplay of migration and proletarianization (cf. Warren
36 production 1980). Indeed, the partial transfer of mass commodity production to
37 these new spatial locations is a crucial precondition for the processes
38 that the postmodernists constantly refer to. For example, new infor-
39 mation technologies and the computer age could not exist without
40 the ‘nimble fingers’ of migrant and recently proletarianized ‘Third
41 World’ women assembling micro-processors (Lim 1978, Safa 1981).
42 The ‘brave new world’ of Western Europe is therefore dependent
43 upon the continuation elsewhere of the separation of the direct pro-

ducer from the means of production and their spatial mobility to find 1
a capitalist wishing to purchase and exploit labour power for a wage. 2
There is also a ‘hidden side’ to these processes. Discussion of the 3
export of capital often takes on a reified character if it focuses only 4
upon the movement of sums of money. As Marx constantly reiter- 5
ated, the concept of capital refers not so much to a thing but to a so- 6
cial relation (e.g. 1976: 932) between two classes which is mediated 7
by things. Consequently, the export of capital involves not only the 8
movement of money but also the ‘agents’ of capital, understood to re- 9
fer to both those who own and control capital directly and those who 10
manage in various ways the use of capital. There has been an undue 11
silence about the migration of such people within Marxist and non- 12
Marxist theories of migration, a silence which becomes even more 13
inappropriate in the context of the increasing mobility of capital with- 14
in the world capitalist system. 15
Hence, the intention of this paper is to offer some critical reflec- 16
tions on the development of a Marxist theory of migration. However, Marxist theory 17
our objective is not to formulate a new ‘postmodern’ theory of mi- of migration 18
gration. Rather, we argue that the apparent difficulties facing po- 19
litical economy explanations of migration when interpreting recent 20
evidence of migration flows arise largely from their inadequacies in 21
explaining migration in pre-‘postmodern’ capitalism. 22
23
24
Marxism and migration theory 25
26
The Marxist tradition has made a significant contribution to the de- 27
bate about the interrelationship between the development of capi- 28
talism and migration flows, both within and between nation states, 29
and has provided a foundation for the development of an alternative 30
approach to the dominant position of the sociology of ‘race/ethnic re- 31
lations’ in Britain (Miles 1982). Central to work within this tradition 32
has been the contribution of Stephen Castles and his various collabo- 33
rators. Castles and Kosack’s Immigrant Workers and Class Structure Castles and Kosack 34
in Western Europe (1973) is widely regarded as the classic statement 35
of the Marxist analysis of the interrelationship between migration 36
and capitalism (see also History Task Force 1979). Similarly, while 37
part of the more recent work of Castles, especially Here for Good: 38
Western Europe’s New Ethnic Minorities (1984), arguably departs from 39
the Marxist approach, this book has already taken its place as an im- 40
portant work within the tradition of political economy. 41
Indeed, there is much that is of continuing value in this work, as 42
we reiterate below. It rejects those theoretical traditions which anal- 43

1 yse migration in terms of individual decision making in the context

2 of a variety of discrete ‘push’ and ‘pull’ factors (Jackson 1986: 13-16)
3 in favour of an examination of the significance and dynamics of mi-
4 gration in the material and structural processes of capital accumula-
5 tion and uneven development. This analysis has therefore become
6 an important corrective to orthodox studies of migration which focus
7 primarily upon individuals who migrate and their ‘problems’ of ‘ad-
8 aptation, assimilation and integration’ (Bolaria 1984: 219).
9 Nevertheless, aspects of the work of Castles and Kosack have been
10 criticized, although many of these writers remain sympathetic to its
11 broader underlying assumptions. For example, Lever-Tracy (1983)
12 and Miles (1986) have questioned their use of the concept of reserve
13 army of labour to describe the structural position of foreign-born
14 workers within Western Europe; Burawoy (1976) has queried their
15 assumption that migrant labour is ‘cheap labour’ and their instru-
16 mentalist view of the state; Böhning (1984) has criticized their ten-
17 dency to over-generalize and blur important differences in patterns
18 of migration both within and between nation states; and Miles (1982,
19 1986) and Phizacklea and Miles (1980: 11-12) have questioned their
20 functionalist analysis of the relationship between racism and migra-
21 tion and their conception of the impact of migration on the class
22 structure of Western Europe since 1945 and 1973. In this paper, we
23 draw upon and extend certain of these themes.
24 In order to do so, we summarize briefly the main arguments of
25 Castles and his collaborators. The initial stimulus to migration into
26 post-1945 Western Europe is located in the interrelated processes of
27 capital accumulation and uneven development which create reserve
28 armies of labour within the periphery of the world system. Pressure
29 to emigrate is regarded as an expression of inequality among nation
30 states and between the centre and the periphery of the world capital-
31 ist system. The tendency to import labour is regarded as a cyclical
32 expression of the uneven expansion of capital accumulation among
33 economic sectors, among nation states and within the world econo-
34 my (e.g. Petras 1980).
35 During the early post-war years, the process of capital accumula-
36 tion resulted in an increase in the demand for unskilled and semi-
37 skilled labour. Unskilled and semi-skilled positions were vacated by
38 the indigenous male working class, a proportion of whom found
39 better paying work in more skilled sectors of production. Western
40 European capital responded to this trend by mobilizing internal
41 reserves of labour. These included, amongst others, women who
42 had previously worked solely within the domestic unit and the la-
43 tent reserves of rural agricultural commodity producers. In most

cases, however, these internal reserves were exhausted during the 1

early 1950s or soon after, and were insufficient to fill all of the emer- 2
gent vacancies (Castles et al. 1984: 25; see also Castles 1985: 519). 3
The respective states, and employers within various social forma- 4
tions, responded to the continued demand for unskilled labour, in 5
some cases by implementing new labour saving technologies and 6
in others through the recruitment of foreign-born labour (see also 7
Kindelberger 1967; Sassen-Koob 1978). 8
Two main sources of labour were identified: the colonial and exco- labour migration 9
lonial formations in Asia, Africa and the Caribbean, and the spatially 10
proximate social formations of the Mediterranean periphery. They 11
suggest that those who migrated to Western Europe were primarily 12
young single males and females who were displaced by the penetra- 13
tion of capitalist relations of production in agricultural sectors of pe- 14
ripheral formations. As such, the majority of post-war migration to 15
Western Europe was made up, according to Castles et al. (1984: 25) 16
of a movement of a latent reserve army of labour which was spatially 17
located in the periphery of the world capitalist system. 18
During the initial phase of mass labour migration, which lasted 19
from around 1950 to 1973, the state is accorded a peripheral, instru- 20
mental role in the process of migration. State intervention, in the 21
form of the formulation, articulation and administration of an ‘im- 22
migration policy’ was only developed several years after the migra- 23
tory process began: ‘Government immigration policies have come 24
after the event, to control and direct existing movements rather than 25
to determine them from the outset’ (Castles and Kosack 1973: 26; see 26
also Castles et al. 1984: 6). Thus, before 1973, migration is defined 27
as a relatively spontaneous reaction to labour demand (Castles et al. 28
1984: 2; Castles and Kosack 1973: 25). But the minimal nature of 29
state intervention which did occur was structured solely by the inter- 30
ests of employers. According to Castles, ‘When recruitment started 31
in the late 1950’s state migration policies were concerned only with 32
short-term fulfillment of capital’s labour requirements’ (1985: 522). 33
The years 1973/74 were a turning point in the history of labour 34
migration to post-war Western Europe. Each Western European 35
state, with the exception of Britain where legislation had been passed 36
in 1962, placed restrictions on the entry of migrants who were seek- 37
ing work. According to Castles et al (1984: 28-9), this decision was 38
the outcome of a number of conjunctural economic, political and 39
ideological factors. The immediate economic reason for the partial 40
ban on primary labour migration was the ‘oil crisis’ and the accom- 41
panying economic recession. 42
However, they suggest (1984: 29; cf. Cohen 1987: 140-3) that the 43

1 restrictions were also motivated by political and ideological consider-

2 ations. The repeated renewal of labour contracts and the accompany-
3 ing process of family reunification, without corresponding increases
4 in state expenditure on services in demand by foreign-born workers
5 and their families, meant increasing ‘immigrant’ competition for
6 education, housing, health and social services with the indigenous
7 population. The competition for scarce resources resulted in increas-
8 ing conflicts between the foreign-born and indigenous populations,
9 and constituted a threat to the social order of the labour importing
10 nation states. Furthermore, foreign workers were becoming increas-
11 ingly militant both politically and on the shop-floor. Their presence
12 became defined as a threat to the long term stability of the social or-
13 der, and limits were placed on the scope for their use as a docile and
14 manipulable labour force (Castles et al. 1984: 30).
15 The restrictions imposed on migration, then, signalled the emer-
16 gence of a set of qualitatively new political priorities and concerns on
17 the part of the respective states. Whereas prior to 1973, the state and
18 capital both defined foreign-born labour in strictly economic terms,
19 its value lying in its relative ‘cheapness’ and in its contribution to
20 industrial production, after 1973 political and ideological consid-
21 erations about the future stability of the nation state pushed these
22 strictly economic factors into the background (Castles et al 1984: 29-
23 32). The central assertion therefore is that state intervention and the
24 expression of political and ideological concern were evident only af-
25 ter the migration was underway.
26 In the remainder of the paper, we shall argue that this theoreti-
27 political and cal approach, because of its economism, has been blind to the sig-
28 ideological nificance of the political and ideological determinants of migration
29 determinants flows and of the significance of the migration of not only skilled non-
30 manual labour but also of the owners of capital and their ‘agents’.
31 Moreover, it ignores the considerable evidence of the state’s concern
32 about the political and ideological implications of migration long
33 before the economic crisis of 1973/74. It is ironic that this error in
34 Castle’s work mirrors a feature of a great deal of non-Marxist writ-
35 ing on the history of migration to Britain in so far as it asserts that
36 the period before 1962 was an ‘age of innocence’, a period of lais-
37 sez-faire, wherein the state played no role in relation to migration
38 flows (Deakin 1970: 47; Carter, Harris and Joshi 1987: 335; Miles and
39 Solomos 1987: 88-9).
40 Consequently, when viewed from the perspective of the late 1980s,
41 what has changed in the period since the late 1940s has been relative
42 rather than absolute in character. That is to say, there is nothing new
43 about refugee migrations or about the migration of the bourgeoisie

and skilled non-manual labour into European social formations but 1

rather that these migrations have become relatively more significant 2
than in the past. Moreover, there is nothing new about state interven- 3
tion to control migration flows, only a change in the visibility and/or 4
extent of the state’s intervention. 5
6
7
Post-1945 migration flows 8
9
During the 1980s, political and public attention has increasingly fo- 10
cused on refugee migrations (for example, the flight of people from refugee migrations 11
Vietnam, Chile, Iran and Sri Lanka) into Western Europe, and in- 12
creasing academic attention is being devoted to such migrations 13
(e.g. Kay 1987). The state in most Western European countries have 14
been anxious to, at the very least, reduce these migration flows, usu- 15
ally claiming that many of the refugees are bogus because their mi- 16
gration has been motivated by the desire to find paid employment 17
(i.e. that they are economic migrants rather than political refugees) 18
(Cleary 1989). When interpreted in the light of Castles’ theoretical 19
analysis of post-1945 migration, these migrations seem to constitute 20
novel events which cannot be explained easily within the tradition 21
of the political economy of migration because such theories reduce 22
migration (Zolberg 1983a: 4): 23
24
to a unidimensional process of uneven economic exchange between 25
states of origin and destination. As such, all appears to have been 26
said when migration has been identified as another variety of exploi- 27
tation, a process into which every policy variation is made to fit. 28
29
But as a more comprehensive overview of both pre- and post-1945 30
European migration history reveals (e.g. Zolberg 1983b; Marrus 31
1985), there is little novelty, certainly not in any absolute sense, in 32
recent refugee migrations. While there are significant problems in 33
defining refugees and refugee migrations (e.g. Zolberg 1983a: 19-22; 34
1983b 25-7), there is no doubt that they constitute a large proportion 35
of international migration in the twentieth century (Beijer 1969). 36
For example, and considering the British case, refugee migrations 37
have been of considerable numerical and political significance over 38
the past one hundred years, the most prominent being those of Jews 39
from Eastern Europe and, later, from Germany (Miles and Solomos 40
1987; Holmes 1988). Since 1945, although the 1962 Commonwealth 41
Immigrants Act and the 1965 White Paper on Commonwealth 42
Immigration effectively brought to an end what was largely an ec- 43

1 onomically induced migration from the Caribbean and the Indian

2 subcontinent, this legislation did not prohibit migration flows per
3 se. The most notable, subsequent migrations were those from East
4 Africa in the late 1960s and again in the early 1970s (Bachu 1985),
5 migrations of people who held UK passports but whose right to enter
6 Britain was withdrawn in 1968 in response to the migration from
7 Kenya which began to increase in the second half of 1967.
8 The origin of these migrations lay in colonial history and in post-
9 independence political developments in Kenya and Uganda, devel-
10 opments which led to a large proportion of the population of South
11 Asian origin in both of these countries concluding that they faced an
12 uncertain economic and political future in East Africa in the context
13 of policies of Africanization. In the case of Uganda, there was a for-
14 mal expulsion of this population. The migration to Britain therefore
15 resulted from a political process in the course of which they were
16 defined, and defined themselves, as not belonging to the post-colo-
17 nial and newly emergent nation-states of Kenya and Uganda. In a
18 loose sense, therefore, they might all be considered to be politically
19 induced migrants, although the Ugandan Asians were certainly, in
20 a formal sense, political refugees. Without doubt, it is impossible to
21 explain these migrations from East Africa in terms of a response to
22 labour shortages in the British economy (Bachu 1985: 2-3).
23 The apparent novelty of the refugee migrations of the 1980s is
24 further reduced when one considers the immediate post-1945 period
25 of European, and especially West German (Herbert 1986: 179-86),
26 history. In the decade beginning in 1945, approximately 20 million
27 people either fled or were expelled, transferred or exchanged within
28 Europe. Of this massive migration of population, the largest compo-
29 nent consisted of around 13 million people of German origin (usu-
30 ally known as volkdeutsche) who were officially returned, or who were
31 expelled or fled to Germany (Schechtman 1962: 363; Marrus 1985:
32 330). The official transfers of population arose from a decision of
33 Tripartite the Tripartite Conference in Potsdam in 1945 concerning the pop-
34 Conference in ulation of German origin in Czechoslovakia, Poland and Hungary
35 Potsdam (Schechtman 1962: 36).
36 In the immediate aftermath of the Second World War, there was
37 considerable concern about the economic and political consequenc-
38 es of such a large transfer of population into West Germany in the
39 context of the widespread destruction of the economy and the social
40 infrastructure. However, this proved to be misplaced in the light of
41 the subsequent development of the post-war economy, sustained in
42 part by the import of capital from the United States. In 1946, the
43 index of industrial production in West Germany stood at 34 (1936 =

100) but it had regained its pre-war level by 1950, and by 1956, it was 1
twice the 1936 level. Gross National Product increased from $23.1 2
billion in 1950 to $52.6 billion in 1958 and industrial production 3
tripled between 1949 and 1959. Integral to this successful expansion 4
of capitalism was an increase in the labour force from 13.6 million 5
persons in 1949 to 19.6 million in 1959 while unemployment fell 6
from 8.8 per cent of manpower in 1949 to 1 per cent in 1959. A 7
key component of this increased labour force consisted of the mi- 8
grant population of German origin which was transferred or expelled 9
from Eastern and south-eastern Europe in the course of redrawing 10
the boundaries of nation states and negotiating respective spheres of 11
influence between Western and Eastern political blocs (Schechtman 12
1962: 315-7; Marrus 1985:330-1). 13
But it was not only the migration of expellees and refugees from 14
Eastern Europe that added to the population and the labour force 15
of West Germany. In addition, in the post-war ‘Cold War’ period, 16
there was a continuing migration of refugees from Eastern Europe, 17
and especially East Germany, into West Germany. This movement of 18
population was composed largely of people who emigrated illegally 19
or in violation of government policy rather than of people formally 20
expelled from the territory in which they were living. After 1948, this 21
refugee migration to the West became increasingly difficult with the 22
formation of Communist governments in Eastern Europe and with 23
the subsequent sealing of the borders, although the peculiar situa- 24
tion of Berlin continued to permit relatively easy access to the West 25
for those who lived in the Eastern sector of the city or who could 26
enter that sector. 27
In 1949, formal procedures for granting political asylum were es- political asylum 28
tablished in West Germany, permitting a more precise enumeration 29
of the flow of refugees into the country. In 1950, 197,000 people 30
entered West Germany as refugees, the figures for the following 31
three years being 165,000, 182,000 and 331,000 respectively. By 32
1952, there were nearly 200,000 anti-Communist refugees living 33
in various camps and centres in Berlin and West Germany. This 34
exodus continued for a decade more from 1951, during which time 35
approximately 3.5 million people entered West Germany from East 36
Germany. This migration was finally terminated in 1961 with the 37
building of the Berlin Wall (Marrus 1985: 354-5; Esser and Korte 38
1985: 169; Bade 1987: 151). 39
A similar relationship between refugee migration and the expan- 40
sion of the national labour force can be found in the work of the 41
International Refugee Organization (IRO). The IRO was established 42
by the United Nations in December 1946 to organize the repatria- 43

1 tion of the remaining refugees, about a third of whom remained in

2 Displaced Persons camps in Italy, Germany and Austria. About half
3 of those in the camps were of Polish origin, and there were also sig-
4 nificant numbers of Balts, Ukranians and Yugoslavs. When the IRO
5 began work in mid-1947, it became responsible for approximately
6 1.5 million refugees in total, and in the period up to the end of 1951,
7 it resettled 1,039,150 people. More than 75 per cent were resettled
8 in the United States, Australia, Canada and Israel, while European
9 nation states accepted around 170,000 people. Of the latter, 86,000
10 entered Britain, 38,000 went to France and 22,000 were resettled in
11 Belgium (Marrus 1985: 344-5).
12 A major IRO objective was to link resettlement with post-war eco-
13 nomic reconstruction in Europe, and even where resettlement oc-
14 curred outside Europe, an attempt was made to marry the economic
15 skills and experience of the refugees with the stated economic de-
16 mands of the receiving countries. Thus, a policy of refugee resettle-
17 ment became intertwined with a labour migration policy, with the
18 IRO thereby functioning as a form of international labour exchange.
19 This role was further illustrated by the fact that, by mid-1949, the
20 majority of the 175,000 people remaining under the supervision of
21 the IRO were people whose age, health or physical condition made
22 an economically conditioned resettlement elsewhere in the world dif-
23 ficult (Marrus 1985: 345).
24 This same interrelationship was evident in the initiative in 1946
25 of the British post-war Labour Government to resolve specific labour
26 shortages in key sectors of the economy by the recruitment of refu-
27 gees from the Displaced Persons camps in Germany and Austria.
28 Between 1947 and 1949, some 91,000 men and women arrived in
29 Britain as European Volunteer Workers and were placed in agricul-
30 ture, in the mining and textile industries, and in hospital and domes-
31 tic service, under a set of restrictive conditions which expressed the
32 contradiction between what was in one sense a labour recruitment
33 refugee resettlement scheme and, in another, a refugee resettlement programme (Kay and
34 programme Miles 1988, 1990).
35 Castles and his various collaborators have noted the occurrence
36 of these ‘refugee’ migrations (e.g. 1984: 25), but have not sought to
37 explain them. Indeed, within the context of their theoretical frame-
38 work, they are difficult to explain because they were not immediately
39 determined by the capital accumulation process. Yet, at least in the
40 case of the immediate post-war refugee migrations, an articulation
41 of political and economic relations determines their occurrence, as
42 Castles acknowledges implicitly when he notes that the German state
43 only initiated a formal ‘guestworker’ recruitment system when the
flow of political refugees began to decline.

But the silence about the migration of refugees within and into 1
Europe is not the only silence within the political economy of migra- 2
tion. It also fails to acknowledge the migration of skilled non-man- 3
ual labour and of capitalists. Assessing the scale and significance of 4
these migrations within Western Europe since 1945 is difficult be- 5
cause of the limited academic attention devoted to the phenomenon 6
and of the absence of adequate official statistics. Both factors reflect 7
a broader determinant. These migration flows are accepted as part 8
of the normal and necessary working of the capitalist economy and 9
are therefore not considered worthy of identification within the po- 10
litical system as being the source of alleged political and ideological 11
problems for the indigenous population. Because such migrations 12
have not been identified by leading politicians and by sections of the 13
working class as problematic, there is no ‘problem’ to document and, 14
therefore, no need to investigate the occurrence, scale and determi- 15
nants of such migrations. 16
Throughout Western Europe since 1945, there has been a close 17
relationship between the migration of semi- and unskilled labour- 18
ers seeking to enter manual wage labour and the movement of in- 19
digenous labour into a variety of semi- and highly skilled non-man- 20
ual jobs, especially in the tertiary or service sector of the economy 21
(Böhning 1984). Thus, this specific migration has served to fill posi- 22
tions in the hierarchy of wage labour vacated by indigenous workers 23
as a result of the availability of better paid jobs, or jobs involving 24
more attractive work and/or conditions. But this form of internal oc- internal 25
cupational mobility has not been sufficient to ensure that all salaried, occupational 26
non-manual positions in tile economy have been filled by indigenous mobility 27
labour. 28
First, in the age of the transnational company, staff are trans- 29
ferred from one branch to another in a different part of the world 30
as an alternative to the recruitment and employment of indigenous 31
labour, partly in order to create a career structure for non-manual, 32
salaried staff and partly to obviate local labour shortages of highly 33
skilled non-manual labour. Second, there is an international labour 34
market for various forms of highly specialized, skilled non-manual 35
labour. In situations of scarcity, such labour can be highly mobile 36
in response to high salaries offered in different parts of the world, 37
especially where the training of such labour is very expensive and/ 38
or where it is only required for relatively short periods of time (as 39
for example in the case of oil exploration). In these latter circum- 40
stances, the recruitment of non-manual workers from outside the 41
nation-state as a form of migrant labour might be chosen as the most 42
suitable solution to the need to recruit skilled non-manual labour. 43

1 Here, when referring to skilled non-manual, we are concerned

2 with various categories of managerial, technical and professional
3 staff. The significance of the international migration of such staff
4 employed in transnational companies has been discussed by Salt
5 (1983, 1984, 1988) who has shown that companies with branches
6 and subsidiaries in several countries constitute a form of closed la-
7 bour market within which managerial and technical staff are able to
8 circulate. Thus, while remaining in employment in the same com-
9 pany, they migrate from one country to another for limited periods
10 of time while moving up the career hierarchy. Hence, this interna-
11 tional migration of labour is synonymous with the pursuit of self-
12 advancement within the transnational company. The implication of
13 his argument is that as the number and scale of transnational com-
14 panies have increased since 1945, then so has the significance of this
15 form of international migration. More specifically, on the basis of the
16 limited date available, Salt ‘has concluded that company managers
17 and various professionals are a major component of the migration
18 between advanced capitalist nation states (1984: 638, 645-6).
19 The migration of managerial. technical and professional staff
20 occurs not only within the advanced capitalist sector of the world
21 economic system, but also between that sector and other sectors un-
22 dergoing capitalist development (Findlay 1987). Included here are
23 multinational firms employees of multinational firms operating in regions such as the
24 Middle East as well as individuals with specialist, non-manual skills
25 engaged on short-term contracts by recruitment agencies instructed
26 by both local and international companies operating in the region
27 (Salt 1988: 390).
28 In contrast with the international migration of semi- and un-
29 skilled manual labourers, considerably greater formal assistance
30 to facilitate migration and temporary settlement is provided by the
31 companies involved, including the direct and indirect provision of
32 the reproduction costs of the managerial or professional migrant
33 worker. International companies employ staff to negotiate and ob-
34 tain work and residence permits where necessary, and either em-
35 ploy agencies to obtain accommodation or purchase property to let,
36 often at subsidised rents, to the migrant manager or professional
37 (Salt 1984: 648-9). One of the consequences is a high degree of resi-
38 dential segregation on the part of highly skilled non-manual labour.
39 Where those persons providing this labour power possess a distinct
40 cultural heritage (as, for example, in the case of Japanese managers
41 employed in transnational firms in Europe), the result is what might
42 be considered to be a form of ‘ethnic segregation’ (Findlay 1987:
43 8,16) similar yet distinct from that which has usually preoccupied
academic investigators.

Sassen’s recent attempt to ‘capture analytically and empirically 1

the articulation of the process of labour migration with fundamental 2
processes in the contemporary phase of the world economy’ (1988: 3
186), in which she argues that the internationalization of production 4
is a key dimension, makes no mention of the migration of manage- 5
rial, technical and professional staff of transnational companies. Yet, 6
she highlights the importance of the growth of Export Processing Export Processing 7
Zones and describes in some detail the increasing scale of foreign Zones 8
investment by companies based in the advanced capitalist sector of 9
the world economic system, a trend encouraged by the adoption of 10
export-led industrialization by the states in the ‘peripheries’ of that 11
system (1988: 98-105). But her interest in analysing this globaliza- 12
tion of production is exclusively with the interrelated processes of 13
proletarianization, feminization and the international emigration of 14
semi- and unskilled manual labour that it stimulates, and so she ig- 15
nores the migration of managerial, technical and professional staff 16
that accompanies the migration of capital and who organize and pro- 17
mote these processes. 18
In addition to the relative silence about the fact and extent of such 19
migrations, there is also the issue of the nature of the explanation to 20
be offered for them. Salt initially questioned the applicability to ‘post- 21
industrial societies’ of theories which were generated in the context 22
of a transition from rural, subsistence production to urban, indus- 23
trial production, and he offered an alternative analytical framework 24
which focused on the nature of modern labour markets ‘in which 25
specialist skills and training mean that the workforce is segmented 26
into self-contained noncompeting groups’ (1984: 634). 27
As Salt’s more recent writing indicates (1988, also Salt and 28
Findlay 1989), this is a rather narrow basis on which to explain the 29
phenomenon because it disembodies the labour market for manage- 30
rial, technical and professional staff from the wider operations and 31
development of national and transnational capital. It is only on the 32
basis of the successful operation of the transnational company with- 33
in the capital accumulation process on a world scale that a career 34
structure can be available to company managerial and technical staff. 35
Understood simplistically, the migration of these ‘mental labourers’ 36
is a concomitant of the migration of capital, and so the factors deter- 37
mining the migration of capital structure the international mobility 38
of highly skilled managerial and technical staff. 39
Amongst the highest ranks of managerial and professional staff 40
employed by international companies, one finds a stratum of the 41
capitalist class proper, those managers who also own and control 42
capital in some combination. But this concept also refers to people 43

1 whose ownership and control of capital means that they personally

2 run companies of various sizes. In the context of a world economic
3 system, especially where regional economies are of increasing sig-
4 nificance, such capitalists also migrate in the search for the best spa-
5 tial location for investment, to develop markets, to obtain supplies of
6 raw materials etc. Again, it is difficult to assess the nature and extent
7 of this migration because the statistical measurement of migration
8 is oriented primarily to assessing and controlling the migration of
9 actual or potential wage labourers.
10 A very partial assessment of the significance of the migration of
11 the capitalist class and of the managerial and technical staff of multi-
12 national capital is possible in relation to Britain using the immigra-
13 tion statistics collected by the Home Office. In Table 1, we document
14 the number of foreign people of non-EEC origin allowed to enter
15 Britain Britain annually between 1974 and 1986 because they were in pos-
16 session of a work permit or because they sought to pursue business
17 activity.
18 Concerning the first category, during the 1970s, the work permit
19 system allowed the entry of semi- and unskilled manual labour for,
20 for example, hotel and catering work, but this option was removed
21 in 1979 and, since 1980, work permits have been issued primarily to
22 professional, administrative and technical staff under very restrictive
23 procedures (Macdonald 1987: 194-203; Salt 1988: 391). Thus, Table I
24 shows that, through the 1980s, the number of managerial and tech-
25 nical staff (a large proportion of whom were probably in the employ
26 of non-EEC based multinational companies operating in Britain) al-
27 lowed to enter Britain remained at a fairly constant level of around
28 12,000 people. One can reasonably assume that the number of such
29 skilled, non-manual staff allowed entry into Britain from other coun-
30 tries is considerably higher than this because employees of compa-
31 nies based in other EEC countries are excluded from these statistics.
32
33 1 Source: Control of Immigration Statistics, United Kingdom, 1974-
34 1986, London: HMSO, 1975-87.
35 2 Statistics exclude EEC nationals but include Commonwealth citi-
36 zens and non-EEC foreign nationals.
37 3 Business passengers include all those given leave to enter for less
38 than 12 months for business purposes.
39 4 Work permit holders given leave to enter includes those holding
40 permits for 12 months and less than 12 months.
41
42 The other category, those given leave to enter Britain for business
43 purposes, includes two groups of people who unfortunately (but sig-

nificantly) are not separately aggregated. The first are those who en- 1
ter Britain to undertake self-employment or to establish a business, 2
or to join or take over an existing business. As with the work permit 3
system, there is a long history to the development of the regulations 4
governing the entry of people seeking to establish themselves as capi- 5
talists. The most recent, important changes occurred in 1980, as a re- 6
sult of which those currently seeking to enter Britain for the purpose 7
of self-employment or to establish, to join or to take over a business 8
must have a minimum investment of £150,000 and must demon- 9
strate that this investment will create new, full-time employment for 10
persons already settled in Britain. The intention of these conditions 11
is to prevent the entry of petit bourgeois capitalists (especially from 12
the Indian subcontinent and Cyprus) intending to establish small 13
shops, restaurants and manufacturing activities dependent on family 14
labour (Macdonald 1987: 204). 15
The second group includes those people who are engaged on a 16
temporary business visit. The criteria employed by immigration of- 17
ficials to permit entry for such purposes are probably detailed in the 18
Immigration Rules which are not published. One assumes that a 19
large proportion of those permitted entry are capitalists or company 20
managers and technical staff (including sales staff) who are engaged 21
on official company business and who may stay for a few days or 22
weeks (but perhaps months) at the most. How many become perma- 23
nent settlers is impossible to determine from these statistics alone. 24
Again, when assessing the significance of the scale of this migration 25
into Britain, it should be noted that the statistics exclude nationals of 26
EEC countries. 27
Table 1 shows that the number of aliens (including Commonwealth 28
citizens) granted entry to Britain for business purposes has almost 29
doubled between the mid-1970s and the mid-1980s, and that, dur- 30
ing the 1980s, the British state has permitted around or in excess of 31
three-quarters of a million of aliens to enter Britain annually for such 32
purposes. If numbers really are a crucial indicator of the problematic 33
consequences of a migrant presence, this is a figure that we might 34
have expected to generate considerable political alarm and contro- political alarm 35
versy within Britain. The fact that it has not done so, along with the 36
fact that academic interest in migration on such a scale has been very 37
limited, is therefore highly significant. 38
The silence about these migrations within the writing of Castles 39
and other Marxist theorists is puzzling because, in principle, an 40
explanation for them can be found with reference to the capital ac- 41
cumulation process. The increasing concentration and centraliza- 42
tion of capital within the world economic system is associated with 43

1 capital export, and hence with the increased mobility of the owners
2 and agents of capital. Perhaps the explanation for the silence is to
3 be found in the fact that a research focus on such migrations seems
4 more likely to reveal privilege rather than exploitation, although this
5 is not an especially good reason to ignore processes which are inte-
6 gral to our understanding of the expansion of the capitalist mode of
7 production on a world scale. Yet, even if we were to explore such a
8 theoretical explanation, it would leave unanalysed the mediating role
9 of the state in all international migration flows.
10
11
12 Migration and the role of the state
13
14 In the previous section of the paper, we have identified certain mi-
15 grations which have been ignored and/or which cannot be easily ex-
16 plained by a capital accumulation theory of migration. We now turn
17 to an evaluation of Castles’ claim that, in post-1945 Europe, the state
18 was an absent force in relation to the pre-1973 development of migra-
19 tion flows. We criticize this interpretation and highlight the central
20 autonomous state and partially autonomous role of the state in the regulation of post-
21 1945 migration flows by considering migration to Britain between
22 1945 and the early 1950s (Isaac 1954). These migration flows were
23 composed largely of people from the Irish Republic, from Poland,
24 from the Baltic states, and from the Ukraine, although there were
25 also small movements of people from north west Europe. Hereafter,
26 we ignore the migration of people from the Republic of Ireland
27 because, by virtue of defining citizens of the Republic as in effect
28 British citizens, the state was not able to control this migration. But
29 these other populations, by virtue of being defined in law as aliens,
30 were subject to state regulation on entry and after taking up resi-
31 dence in Britain (Miles 1989b).
32 The British state intervened in several ways to structure these mi-
33 grations. First, because the migrants were aliens, the state necessar-
34 ily provided the political/legal framework for members of the Polish
35 armed forces and their dependents to remain and settle in Britain,
36 and for Displaced Persons to enter Britain (Miles and Solomos 1987:
37 85-6). Second, the state actively recruited and screened the Displaced
38 Persons, not only to identify war criminals and fascist sympathis-
39 ers, but also to ensure that they would be productive workers.
40 Representatives of the Ministry of Labour ‘handpicked’ those grant-
41 ed permission to enter Britain from amongst those who volunteered
42 ‘with preference going to “men of labouring type who are hardy and
43 of good physical standard”... and those prepared to leave behind their

dependents until further arrangements could be made’ (Harris 1987: 1

60). Identical concerns were evident in the recruitment of women 2
(Kay and Miles 1988: 219-21). 3
Third, the state intervened in the process of migration by initially 4
imposing conditions on the rights of Displaced Persons selected to displaced persons 5
enter Britain. In the first instance, they were granted entry for a pe- 6
riod of one year and their chance of remaining in the country was 7
conditional in part on their behaving ‘as a worthy member of the 8
British community’ (Tannahill 1958: 123-8; Kay and Miles 1988: 221- 9
3). However, their status as refugees placed limits on the power of 10
the state to require them to leave Britain because there was nowhere 11
for them to go to, other than the Displaced Persons camps that they 12
had been recruited from. In effect, therefore, these migrants were 13
eventually to become permanent settlers (Layton-Henry 1984: 19; 14
Kay and Miles 1988). Fourth, in the case of the Poles, the state inter- 15
vened to encourage the retention of a distinct cultural identity and 16
community structure by providing money and other resources for 17
the running of three Polish hospitals, educational facilities and hos- 18
tel accommodation (Miles and Solomos 1987: 86). 19
Fifth, the British state intervened to structure the migrants’ entry 20
into, and place within, the labour market. The entry of Displaced 21
Persons was dependent upon their signing a contract of employment 22
which stipulated that they accept work selected by the Minister of 23
Labour and that they could change employment only with the per- 24
mission of the Minister (Miles and Solomos 1987: 87; Kay and Miles 25
1988: 222-3). They were therefore unable to freely dispose of their la- 26
bour power as a commodity and so constituted a form of unfree wage 27
labour (Miles 1987b: 32-4). Thus, they occupied a qualitatively dif- 28
ferent position in production relations when compared with British 29
workers (as well as migrants from British colonial societies) who 30
were not subject to such restrictions. 31
Indeed, it was just this characteristic that made them especially 32
desirable from the point of view of British employers and the British 33
state. According to a government inter-departmental working party, 34
the recruitment of EVWs (cited in Harris 1987: 61), 35
36
enables the Department [of Labour] both to put these foreign work- 37
ers into specific jobs and to keep them in those jobs. The sanction 38
that lies at hand to guard against noncompliance with these landing 39
conditions is deportation of the workers concerned to the ‘Displaced 40
Persons’ camps in Europe, and this sanction has from the very begin- 41
ning proved to be an extremely effective one. Besides being kept out 42
of ‘inessential’ industries, European Volunteer Workers who have 43

1 been brought into this country could not for any length of time re-
2 main unemployed at public expense.
3
4 These restrictions were not however permanent: in 1951, the govern-
5 ment lifted all conditions of the sale of their labour power once the
6 individual had been resident in Britain for three years (Kay and Miles
7 1988: 229-30).
8 The decision to impose these conditions reveals a dialectic of eco-
9 nomic, political and ideological rationality. In part, the British state
10 sought to ensure that specific industries which were short of labour
11 had a ‘guaranteed’ workforce for at least three years. But it was also
12 the case that their formal legal status as aliens facilitated the state’s
13 decision to constitute these migrants as a form of unfree wage labour
14 (Freeman and Spencer 1980: 63-4). In other words, it was politically
15 possible to deny certain basic rights to non-citizens, a denial that was
16 very difficult to legitimate in the case of migrants who were British
17 citizens.
18 European refugees had been exhausted as a source of labour be-
19 fore the end of the 1940s but the demand for labour remained high.
20 British employers continued to rely on Irish labour, along with a
21 limited recruitment under contract from elsewhere in Europe, es-
22 pecially Italy (Miles and Phizacklea 1984: 24; Duffield 1988: 11-14)
23 but they were unable to fill all the emergent vacancies from these
24 sources. This was the key domestic economic determinant of the
25 increased migration from British colonial and ex-colonial societies
26 which began in the late 1940s and which expanded during the 1950s,
27 although it was facilitated by the legal status of such people as British
28 citizens. Unlike with the migrations considered above, the state was,
29 recruitment with one exception, not directly involved in recruitment. Rather, in
30 the context of a long tradition of emigration from the Caribbean (e.g.
31 Thomas-Hope 1986; Petras 1988), this migration was largely ‘spon-
32 taneous’ and took the form of an informal chain migration, although
33 some employers did recruit labour directly in Barbados (Sivanandan
34 1982: 102; Layton-Henry 1984: 23).
35 Although the state did not actively recruit labour on a significant
36 scale in the 1950s, it was not an absent force (Carter, Harris and
37 Joshi 1987). The state in fact intervened in two important ways. First,
38 the absence of state-organized recruitment of labour from the colo-
39 nies and ex-colonies was in itself a form of intervention. The decision
40 not to recruit such labour was a conscious decision which racialized
41 the migrants in such a way that they were deemed unacceptable.
42 According to a prominent civil servant (cited in Joshi and Carter
43 1984: 59):

Whatever may be the policy about British citizenship, I do not think 1

any scheme for the importation of colonials for permanent settle- 2
ment should be embarked upon without the full understanding that 3
this means that a coloured element will be brought in for permanent 4
absorption into our population. 5
6
Second, between 1948 and the mid-1950s, the British state inter- 7
vened directly in the process of colonial and ex-colonial migration 8
using various covert administrative measures, some illegal (Carter, 9
Harris and Joshi 1987: 336-7). The measures varied according to the 10
Commonwealth or colonial status of the potential entrants. In the 11
case of the West Indies, for example, Carter, Harris and Joshi have 12
documented that (1987: 336): 13
14
Governors were asked to tamper with shipping lists and schedules to 15
place migrant workers at the back of the queue; to cordon off ports 16
to prevent passport-holding stowaways from boarding ships and to 17
delay the issue of passports to migrants. This last measure was also 18
adopted by India and Pakistan where the ... Governments refused 19
passports if migrants had no firm prospect of establishing them- 20
selves. Police reports were carried out at the request of the Home 21
Office to establish the basis of these prospects. 22
23
Thus, well before the Commonwealth Immigrants Act of 1962, the Commonwealth 24
state actively, although surreptitiously, intervened in the process of Immigrants 25
migration. The 1962 Act, and the subsequent legislation, are now Act of 1962 26
well documented and need not be recounted here (Layton-Henry 27
1984; Miles and Phizacklea 1984). These interventions of the state 28
were the result of a complex of political and ideological processes, 29
and cannot in any direct way be attributed to the economic needs of 30
capital as is implied by the work of Castles and his various collabo- 31
rators (see also Miles 1985). For example, the state decided not to 32
recruit colonial British subjects in part precisely because their formal 33
status as British citizens meant that they could not be treated differ- 34
ently from the rest of the British working class and could not be dis- 35
ciplined with the sanction of deportation. Thus, it was because they 36
could not be allocated a position in production relations as unfree 37
wage labour that worked against their recruitment by the state. The 38
civil servant cited above put the matter in these terms in 1948 (Joshi 39
and Carter 1984: 59): 40
41
Unlike ex-prisoners of war or other aliens, I assume there could be 42
no authority for deporting coloured British subjects if they felt they 43

1 wished to stay here and take their chance. If there were any assurance
2 that these people could in fact be sent away when they had served
3 their purpose, this proposition might be less unacceptable.
4
5 In addition, the state did not actively recruit in the Caribbean and
6 other colonial and ex-colonial formations because of its concern over
7 the creation of a ‘race relations’ problem in Britain (Carter, Harris
8 and Joshi 1987: 345). State officials, along with the media and sec-
9 tions of the working class, constructed the ‘imagined community’
10 which constituted the English/British nation in terms of the idea
11 of ‘race’ (Miles 1987a: 38-40). ‘Coloured’ people were defined as an
12 ‘alien race’ whose presence constituted a threat to the ‘British way of
13 life’.
14 In the light of all this evidence, what Castles et al. identify as the
15 state intervention beginning of the period of the British state’s intervention in the pro-
16 cess of migration, the Commonwealth Immigrants Act of 1962, was
17 really the formal culmination of a process of state intervention (and
18 racialisation) which began soon after the end of the war. The state
19 did not restrict colonial and ex-colonial migration through a formally
20 codified and publically visible ‘immigration policy’, in part because of
21 an apparent continued commitment to the idea of a ‘free and equal’
22 Commonwealth of nation-states. Similarly, the British state inter-
23 vened in the process of migration through the recruitment, control
24 and provision of settlement assistance to Eastern European refugees,
25 practices which suggest that from the British state’s point of view,
26 not all of those people who were born outside of the spatial boundar-
27 ies of the British nation-state were defined as equally suitable sources
28 of wage labour.
29 More generally, comparative research on post-1945 migration to
30 Europe reveals important differences in the form and consequences
31 of state intervention to regulate migration flows (e.g. Freeman 1979;
32 Edye 1987). One of the more general conclusions to emerge from
33 these studies is that the state has not acted consistently in an instru-
34 mentalist fashion, seeking to serve only the interests of capital. This
35 is in part because there is no single interest of capital in relation to
36 labour migration and, in part, because the partially autonomous exis-
37 tence of the state is grounded in the necessity to mediate not only the
38 competing demands of different fractions of capital but also to guar-
39 antee the political and ideological conditions for the reproduction of
40 the capitalist mode of production within a given national territory.
41 For example, there was conflict between Die Bundesvereinigung
42 der Deutschen Arbeitgeberverbaede (BDA), the West German employ-
43 ers’ federation, and the German state from the early 1970s. When

the German government halted the recruitment of foreign labour 1

in 1973, partly in response to concerns about the social and politi- 2
cal consequences of the increasing presence of foreign workers in 3
Germany, the BDA was strongly opposed. It objected to the sudden 4
implementation of the ban, arguing that at least those workers who 5
were already in the process of recruitment should have been allowed 6
to enter Germany. It also proclaimed the need for greater flexibility, 7
especially with respect to the employment of seasonal labour under 8
contract, and argued that, because of the central role of foreign la- 9
bour in the economy, it would be necessary for the foreseeable fu- 10
ture. During the mid-1970s, the BDA reported that certain employ- 11
ers were unable to recruit the foreign labour that they needed. The 12
Deutscher Hotel- und Gaststaettenverband (DEHOGA) was especial- 13
ly active in its opposition to the recruitment ban of 1973 (Edye 1987: 14
88-90, 92-6). 15
Consequently, in a world which has been divided into nation- 16
states, the state is necessarily involved as a determinant actor, with 17
its own distinct (although not autonomous) interests, in mediating 18
international migration flows. Concomitant with the formation of 19
the nation-state has been the creation of nationality as a legal concept nationality 20
and the creation and policing of national boundaries which consti- 21
tute a political barrier to human spatial mobility because each hu- 22
man being has been attributed with a legal status ‘tying’ him or her 23
to that nation state. And within those boundaries, a central role of 24
the state is to reproduce social order, a process that entails both he- 25
gemonic practices and direct force. Hegemonic practices include the 26
reproduction of national identity, which includes a process of self- 27
identification with the state itself. This is constantly secured by the 28
identification of different Others outside and beyond the nation state, 29
and it can therefore be threatened by the migration of representatives 30
of these Others. 31
Constantly assured that its ‘own’ national community is a supe- 32
rior economic and cultural force, it is not surprising that sections of 33
the working class within a nation state may become involved in agita- 34
tion against a migrant presence, a process that is further encouraged 35
by material scarcity. The process of international migration, especial- 36
ly when ‘required’ by the need to expand the size of the labour force, 37
may therefore constitute a contradictory phenomenon in so far as 38
the state has to regulate both the entry of capitalists, non-manual and 39
manual labour as well as refugees and the social disorder that can 40
arise from agitation against that presence. We have shown above that 41
concerns about social order were prominent in the evaluation of the 42
British state concerning migration in the immediate post-war period 43
and did not emerge only after the migration flows had ‘matured’.

1 Conclusion
2
3 The tradition of the political economy or migration, as represented
4 in the work of Castles and his various collaborators, and despite its
5 real strengths, has failed to highlight and explain certain significant
6 international migrations, and to adequately specify the significance
7 of the role of the state in organizing and regulating international mi-
8 gration flows since 1945. Thus, insofar as there is reason to believe
9 that the migration of refugees, of skilled non-manual labour and of
10 the bourgeoisie itself will continue during the epoch of ‘postmodern’
11 capitalism, we need to elaborate upon, in order to develop, this theo-
12 retical tradition, not only to explain the present and the future, but
13 also the past.
14 In order to do this, it is necessary to highlight the role of both the
15 nation as a spatial and political unit, and of the state as an institu-
16 tional complex in the analysis of international migration processes
17 (Miles 1987b: 181-6). In this respect, we reinforce the point made by
18 Zolberg that (1983a: 4):
19
20 it is the political organisation of contemporary world space into mu-
21 tually exclusive and legally sovereign territorial states which delin-
22 eates the specificity of international migration as a distinctive process
23 and hence as an object of theoretical reflection.
24
25 But in so doing, we do not abstract political relations from econom-
26 ic and political relations in the way that he appears to do. This is
27 because the rise of the nation state was dialectically related to the
28 emergence of the capitalist mode of production, a mode of produc-
29 tion that had profound implications for spatial mobility as a result of
30 the new social relations of production. We therefore seek to explain
31 historical international migration flows in relation to the historical articulation
32 articulation between the process of capital accumulation and the reproduction of
33 the nation state, an articulation which is mediated by the state, the
34 role of which is to guarantee the reproduction of the dominant mode
35 of production and, hence, the nation state itself.
36 In this context, refugee migrations appear far less anomalous. If,
37 following Zolberg, such migrations result from the historical process
38 of nation-state formation (l983b: 30), then that is a process medi-
39 ated by the emergence of the world economic system as structured
40 by the capitalist mode of production. Furthermore, it is mediated by
41 the interventions of national states as they seek to, in turn, mediate
42 the articulation between the reproduction of the mode of production
43 and the nation-state which provides the political and spatial context

for the interrelated activities of capital and labour. This is well illus- 1
trated by the example previously cited of the migrations of Kenyan 2
and Ugandan Asians to Britain, both of which resulted from a pro- 3
cess of nation-state formation after decolonization in the context 4
of the promotion of capitalist development by means of, inter alia, 5
Africanization. And the process of nation-state formation is still far 6
from complete in Africa, not to mention the Middle East, Southern 7
Africa and the Indian subcontinent. all spatial contexts which gener- 8
ate large refugee migrations. 9
But the articulations and mediations referred to can be, in specific 10
conjunctures. even more precise in advancing the interests of capital. 11
The massive movements of refugees within Europe immediately af- 12
ter the Second World War arose not only from the consequences of 13
fascism, but also from the processes of redrawing the boundaries of 14
nation-states and of giving a new political content to nation-state for- 15
mation in Eastern Europe. Yet the resolution of the problem of state- 16
less populations resident within Western Europe was to a significant 17
degree facilitated by the new phase of capitalist accumulation that 18
was launched in the late 1940s, a process that required significant 19
additions to the size of the national labour forces. In this conjunc- 20
ture, a large proportion of a relatively surplus (refugee) population 21
was constituted in the realm of political relations, and drawn into the 22
nation states of Western Europe in order to fill vacant positions in 23
the hierarchy of wage labour (Kay and Miles 1988: 231). The concept 24
of refugee-worker is helpful in conceptualizing this process (Kay and refugee-worker 25
Miles 1988: 215). Thus, what is new about the refugee crisis of the 26
1980s is not its existence per se, but rather that the scope for this new 27
surplus population to be drawn into the capitalist world economy is, 28
comparatively, considerably constrained. 29
A further context in which to examine the articulation and media- 30
tion of the capitalist mode of production. the nation-state as a spatial- 31
political structure and the state as an institutional complex, is the 32
interplay of different international migration flows. A more accurate 33
comprehension of these migrations would focus not only on the mi- 34
gration of semi- and unskilled manual labourers from the ‘periphery’ 35
to the ‘centre’ of the world economy, but also upon the migration of 36
skilled managers, technicians and professionals within the ‘centre’ 37
and from the ‘centre’ to the ‘periphery’. Arguably, the latter has be- 38
come more important than the former in the past fifteen years, at 39
least in Europe, in the light of the fact that the export of capital has 40
to a significant extent replaced the import of manual labour. Such 41
a focus returns us centrally to the task of analysing the process of 42
capital accumulation on a world scale, not in order to understand the 43

1 migration of labour but rather the migration of capital.

2 But we make a mistake if we counterpose the movement of sums
3 of money (to represent capital export) to the movement of peasants
4 and proletarians (to represent labour import) in order to understand
5 this dynamic. For not only are capital (in the sense of money) flows
6 from one spatial location to another dependent upon the permission
7 of the respective states, but so is the movement of those who own
8 those sums of money as well as those who manage, control and ser-
9 vice their utilisation. All these people have a national status, and their
10 international mobility is therefore also subject to the controls estab-
11 lished by the states which control entry into the space and the social
12 relations that they administer. We might be well advised to examine
13 much more closely the privileged terms on which such people are
14 able to move from one nation-state to another (alongside the scale of
15 such mobility) because, in so far as the ownership of capital serves
16 to mediate restrictions on entry and settlement (i.e. mediate the ties
17 of national status), this is the outcome of a political decision. In turn
18 this might tell us rather more about the patterns, processes and con-
19 sequences of ‘flexible accumulation’ than musing about the aesthet-
20 ics of the architecture of the offices of the multinational firms that
21 such people use to exercise their control over us all.
22 This is also a dimension of the central task of contextualizing the
23 racism impact of racism (Miles 1989: 134). Specifically, the British discus-
24 sion about immigration control has centred almost exclusively on the
25 manner in which it has been determined by racism and has thereby
26 ignored its class character. We have shown above that British immi-
27 gration controls facilitate the entry of persons engaged in business
28 activity, persons with capital in excess of £150,000 and managerial,
29 technical and professional employees of international companies.
30 Consequently, with respect to British immigration control, the ex-
31 clusion of ‘black’ migrants seeking to enter wage labour, along with
32 those seeking to establish themselves as a petite bourgeoisie, is one
33 pole of a dialectic in which there is a simultaneous inclusion of mem-
34 bers and agents of the capitalist class.
35 The silence about this latter migration is of considerable signifi-
36 cance in relation to the expression of racism within contemporary
37 Western European nation states. The articulation of that racism,
38 whether by the state itself or by sections of the working class, focuses
39 upon the presence and cultural consequences of those who migrated
40 with the purpose of selling their labour power. And there is no doubt
41 that these migrations have led to a range of cultural transformations
42 within Western Europe. But we might also begin to consider the cul-
43 tural transformations consequent upon the migration of capital, cap-

italists and their various ‘agents’. Thus, alongside the cultural trans- 1
formations initiated by, for example, the increased Muslim presence 2
in Western Europe, we might also begin to consider the transforma- 3
tions of Western European nation states that constitute the processes 4
of Americanization and ‘Japanization’. 5
To cite a single example, Western European city centres have been 6
transformed by, inter alia, the establishment of a range of ‘fast food’ 7
outlets. Both the companies themselves, and the commodity that 8
they supply, constitute a major cultural transformation about which 9
there is a comparative silence in comparison with, for example, agi- 10
tation against the appearance of a mosque. Thus, the reasons why 11
certain cultural transformations consequent upon certain migrations 12
become the object of racist agitation while others consequent upon 13
other migrations are ignored is an important academic and political 14
question which leads us to consider both the nature of racism and 15
the role of the state in its reproduction. And as a prelude to such a 16
study, we need first to undertake an analysis of the migration of the 17
bourgeoisie and its agents to assess the material foundation for these ‘hidden’ cultural 18
‘hidden’ cultural transformations. transformations 19
20
21
Note 22
23
1 This is a revised version of a paper prepared for a conference on ‘Racism and 24
the Post-Modern City’ held at the University of Warwick, 29-31 March 1989. 25
We acknowledge the useful comments by conference participants and by the
Editors of Economy and Society but we take full responsibility for this draft.
26
27
28
29
References 30
31
Bachu, P. (1985) Twice Migrants: East African Sikh Settlers in Britain, 32
London: Tavistock. 33
Bade, K.J. (1987) ‘Transatlantic Emigration and Continental 34
Immigration: the German Experience Past and Present’, in K.J. 35
Bade (ed.), Population, Labour and Migration in 19th and 20th 36
Century Germany, Leamington Spa: Berg. 37
Beijer, G.J. (1969) ‘Modern Patterns of International Migratory 38
Movements’, in J.A. Jackson (ed.), Migration, Cambridge: 39
Böhning, R. (1984) Studies in International Labour Migration, London: 41
Macmillan. 42
Bolaria, S. (1984) ‘On the Study of Race Relations’, in J. Fry (ed.), 43
Contradictions in Canadian Society, Toronto: J. Wiley & Sons.

1 Burawoy, M. (1976) ‘The Functions and Reproduction of Migrant

2 Labour’, American Journal of Sociology, 81: 1051-87.
3 Carter, B., Harris, C. and Joshi, S. (1987) ‘The 1951-55 Conservative
4 Government and the Racialisation of Black Immigration’,
5 Immigrants and Minorities, 6: 335-47.
6 Castles, S. (1985) ‘Guests Who Stayed: The Debate on Foreigners
7 Policy in the German Federal Republic’, International Migration
8 Review, 19: 517-34.
9 –, and Kosack, G. (1973) Immigrant Workers and Class Structure in
10 Western Europe, London: Oxford University Press.
11 –, Booth, H. and Wallace, T. (1984) Here for Good: Western Europe’s
12 New Ethnic Minorities, London: Pluto Press.
13 Cleary, P. (1989) Human Rights and the Admission of Refugees: the
14 British Case, Glasgow University: mimeo.
15 Cohen, R. (1987) The New Helots: Migrants in the International Division
16 of Labour, Farnborough: Avebury.
17 Deakin, N. (1970) Colour, Citizenship and British Society, London:
18 Panther Books.
19 Duffield, M. (1988) Black Radicalism and the Politics of De-
20 industrialisation: The Hidden History of Indian Foundry Workers,
21 Farnborough: Avebury.
22 Edye, D. (l987) Immigrant Labour and Government Policy: the Cases
23 of the Federal Republic of Germany and France, Aldershot: Gower.
24 Esser, H. and Korte, H. (1985) ‘Federal Republic of Germany’, in T.
25 Hammer (ed.), European Immigration Policy: A Comparative Study,
26 Cambridge: Cambridge University Press.
27 Findlay, A. (1987) The Development Implications of Skilled International
28 Migration for Host and Sender Societies, Glasgow: University of
29 Glasgow Department of Geography Occasional Paper no. 20.
30 Freeman, G. (1979) Immigrant Labour and Racial Conflict in Industrial
31 Societies: The French and British Experience, 1945-1975, Princeton:
32 Princeton University Press.
33 Freeman, M. and Spencer, S. (1980) ‘Immigration Control, Black
34 Workers and the Economy’, British Journal of Law and Society, 6:
35 53-81.
36 Frobel, Heinrichs, J. and Kreye, O. (1980) The New International
37 Division of Labour, Cambridge: Cambridge University Press.
38 Harris, C. (1987) ‘British Capitalism, Migration and Relative Surplus
39 Population: A Synopsis’, Migration, 1: 47-90.
40 Herbet, U. (1986) Geschichte der Ausländer-beschäftigung in Deutschland
41 1880 bis 1980: Saisonarbeiter, Zwangsarbeiter, Gastarbeiter, Bonn:
42 Verlag J.H.W. Dietz Nachf.
43 History Task Force (l979) Labor Migration/ under Capitalism: The
Puerto Rican Experience, New York: Monthly Review Press.

Holmes, C. (1988) John Bull’s Island: Immigration and British Society, 1

1871-1971, London: Macmillan. 2
Isaac, J. (1954) British Post-War Migration, Cambridge: Cambridge 3
University Press. 4
Jackson, J.A. (1986) Migration, London: Longman. 5
Joshi, S. and Carter, B. (1984) ‘The Role of Labour in the Creation of 6
a Racist Britain’, Race and Class, 25: 53-70. 7
Kay, D. (1987) Chileans in Exile: Private Struggles, Public Lives, London: 8
Macmillan. 9
–, and Miles, R. (1988) ‘Refugees or Migrant Workers? The Case of 10
the European Volunteer Workers in Britain (1946-1951)’, Journal 11
of Refugee Studies, 1: 214-36. 12
and – (1990), Refugees Or Migrant Workers?, London: Routledge 13
(forthcoming) . 14
Kindelberger, C. (1967) Europe’s Postwar Growth, London: Oxford 15
University Press. 16
Layton-Henry, Z. (1984) The Politics of Race in Brtain, London: Allen 17
& Unwin. 18
Lever-Tracy, C. (1983) ‘Immigrant Workers and Postwar Capitalism: 19
In Reserve or Core Troops in the Front Line?’, Politics alld Society, 20
12: 127-57. 21
Lim, L.Y.C. (1978) Women in Export Processing Zones, New York: 22
United Nations Industrial Development Organisation. 23
Macdonald, I.A. (1987) Immigration Law and Practice in the United 24
Kingdom, London: Butterworths. 25
Marrus, M. R. (1985) The Unwanted: European refugees in the Twentieth 26
Century, New York: Oxford University Press. 27
Marx, K. (1976) Capital, Vol. I; Harmondsworth: Penguin. 28
Miles, R. (1982) Racism and Migrant Labour: A Critical Text, London: 29
Routledge & Kegan Paul. 30
– (1985) The Relative Autonomy of Ideology: Labour Migration and 31
Racism in Britain, Paris: Groupe de Reserche et d’Analyse des 32
Migrations Internationales, Document de Travail, no. 7. 33
– (1986) ‘Labour Migration, Racism and Capital Accumulation in 34
Western Europe Since 1945: An Overview’, Capital and Class, 28: 35
49-86. 36
– (1987a) ‘Recent Marxist Theories of Nationalism and the Issue 37
of Racism’, British Journal of Sociology, 38: 24-43. 38
– (1987b) Capitalism and Unfree Labour: Anomaly or Necessity?, 39
London: Tavistock. 40
– (1989a) Racism, London: Routledge. 41
– (1989b) ‘Nationality, Citizenship and Immigration in Britains 42
Journal of Law and Society, 1989: ??. 43

1 – and Phizacklea, A. (1984) White Man’s Country: Racism in British

2 Politics, London: Pluto Press.
3 – and Solomos, J. (1987) ‘Migration and the State in Britain:
4 A Historical Overview’, in C. Husband (ed.), ‘Race in Britain:
5 Continuity and Change, London: Hutchinson.
6 Petras, E. (1980) ‘The Role of National Boundaries in a Cross
7 National Labour Market’, International Journal of Urban and
8 Regional Research, 4: 157-95.
9 – (1988) Jamaican Labor Migration: White Capital and Black Labor,
10 1850-1930, Boulder: Westview Press.
11 Phizacklea, A. and Miles, R. (1980) Labour and Racism, London:
12 Routledge & Kegan Paul.
13 Safa, H.I. (1981) ‘Runaway Shops and Female Employment: the
14 Search for Cheap Labour’, Signs, 7: 418-33.
15 Salt, J. (1983) ‘International Labor Migration in Western Europe:
16 A Geographical Review’, in M.M. Kritz, C.B. Keeley and S.M.
17 Tomasi (eds), Global Trends in Migration: Theory and Research
18 on International Population Movements, New York: Center for
19 Migrarion Studies.
20 – (1984) ‘High Level Manpower Movements in Northwest Europe
21 and the Role of Careers: An Explanatory Framework’, International
22 Migration Review,17:633-52.
23 – (1987) ‘Contemporary Trends in International Migration Study’,
24 International Migration, 25: 241-51.
25 – (1988), ‘Highly-skilled International Migrants, Careers and
26 Internal Labour Markets’, Geoforum, 19: 387-99.
27 – and Findlay, S. (1989), ‘International Migration of Highly
28 Skilled Manpower: Theoretical and Developmental Issues’, in R.
29 Appleyard (ed.), The Impact International Migration on Developing
30 Countries, Paris: OECD.
31 Sassen, S. (1988) The Mobility of Labour and Capital: A Study in
32 International Investment and Labour Flow, Cambridge: Cambridge
33 University Press.
34 Sassen-Koob, S. (1978) ‘The International Circulation of Resources
35 and Development: The Case of Migrant Labour’, Development and
36 Change, 9: 509-45.
37 Schechtman, J.B. (1962) Postwar Population Transfers in Europe 1945-
38 1955, Philadelphia: University of Pennsylvania Press.
39 Sivanandan, A. (1982) A Different Hunger, London: Pluto Press.
40 Tannahill, J. (1958) European Volunteer Workers in Britain, Manchester:
41 Manchester University Press.
42 Thomas-Hope, E.M. (1986) ‘Caribbean Diaspora - the Inheritance
43 of Slavery: Migrations from the Commonwealth Caribbean” in

C. Brock (ed.), The Caribbean in Europe: Aspects of the West Indian 1

Experience in Britain, France and the Netherlands, London: Frank 2
Cass. 3
Warren, B. (1980) Imperialism: Pioneer of Capitalism, London: Verso. 4
Zolberg, A. (1983a) ‘International Migrations in Political Perspective’ 5
in M.M. Kritz, C.B. Keeley and S.M. Tomasi (eds.), Global Trends 6
in Migration: Theory and research on International Population 7
Movements, New York: Center for Migration Studies. 8
– (1983b) ‘The Formation of New States as a Refugee Generating 9
Process’, Annuals of the American Academy of Political and Social 10
Science, 467: 24-38. 11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Class racism 3
4
Etienne Balibar 5
6
7
8
9
10
11
12
In the 1980s, the concept of racism was high on the academic agenda. The
13
philosopher Etienne Balibar published extensively on the issue. At that time,
14
many others treated this concept in a one-dimensional, monocausal sense. 15
They argued that racism was intrinsically rooted in the colonial project, often 16
starting from the assumption that the only – or anyway, the most important – 17
racism is that with black people as its object. Balibar was one of the first to 18
demonstrate how the problematisation of some categories of non-black na- 19
tives – e.g. the ‘labouring classes’ or the ‘dangerous classes’ – has remark- 20
able congruence with those of some categories of blacks. By showing that 21
particular sections of the (native, white) working class can also be victims 22
of racism – ‘class racism’ in this case – it became clear that the history of 23
colonialism is not a sufficiently adequate starting point for theoretical discus- 24
sion about the nature and significance of racism in Europe. In the same vein, 25
Balibar gave suggestions for a new theoretical approach to racism. 26
27
28
Academic analyses of racism, though according chief importance to 29
the study of racist theories, none the less argue that ‘sociological’ rac- sociological racism 30
ism is a popular phenomenon. Given this supposition, the develop- 31
ment of racism within the working class (which, to committed social- 32
ists and communists, seems counter to the natural order of things) 33
comes to be seen as the effect of a tendency allegedly inherent in 34
the masses. Institutional racism finds itself projected into the very 35
construction of that psycho-sociological category that is ‘the masses’. 36
We must therefore attempt to analyse the process of displacement 37
which, moving from classes to masses, presents these latter both as 38
the privileged subjects of racism and its favoured objects. 39
Can one say that a social class, by its situation and its ideology 40
(not to mention its identity), is predisposed to racist attitudes and be- 41
haviour? This question has mainly been debated in connection with 42
the rise of Nazism, first speculatively and then later by taking vari- 43

568 etienne balibar
1 ous empirical indicators.1 The result is quite paradoxical since there

2 is hardly a social class on which suspicion has not fallen, though
3 a marked predilection has been shown for the ‘petty bourgeoisie’.
4 But this is a notoriously ambiguous concept, which is more an ex-
5 pression of the aporias of a class analysis conceived as a dividing
6 up of the population into mutually exclusive slices. As with every
7 question of origins in which a political charge is concealed, it makes
8 sense to turn the question around: not to look for the foundations
9 of the racism which invades everyday life (or the movement which
10 provides the vehicle for it) in the nature of the ‘petty bourgeoisie’,
11 but to attempt to understand how the development of racism causes
12 a ‘petty bourgeois’ mass to emerge out of a diversity of material situ-
13 class bases of racism ations. For the misconceived question of the class bases of racism,
14 we shall thus substitute a more crucial and complex question, which
15 that former question is in part intended to mask: that of the relations
16 between racism, as a supplement to nationalism, and the irreducibil-
17 ity of class conflict in society. We shall find it necessary to ask how
18 the development of racism displaces class conflict or, rather, in what
19 way class conflict is always already transformed by a social relation
20 in which there is an inbuilt tendency to racism; and also, conversely,
21 how the fact that the nationalist alternative to the class struggle spe-
22 cifically takes the form of racism may be considered as the index of
23 the irreconcilable character of that struggle. This does not of course
24 mean that it is not crucial to examine how, in a given conjuncture,
25 the class conditions [la condition de classe] (made up of the material
26 conditions of existence and labour, though also of ideological tradi-
27 tions and practical relationships to politics) determine the effects of
28 racism in society: the frequency and forms of the ‘acting out’ of rac-
29 ism, the discourse which expresses it and the membership of orga-
30 nized racist movements.
31 The traces of a constant overdetermination of racism by the class
32 struggle are as universally detectable in its history as the nationalist
33 determination, and everywhere they are connected with the core of
34 meaning of its phantasies and practices. This suffices to demonstrate
35 that we are dealing here with a determination that is much more
36 concrete and decisive than the generalities dear to the sociologists
37 of ‘modernity’. It is wholly inadequate to see racism (or the national-
38 ism-racism dyad) either as one of the paradoxical expressions of the
39 individualism or egalitarianism which are supposed to characterize
40 modern societies (following the old dichotomy of ‘closed’, ‘hierarchi-
41 cal’ societies and ‘open,’ ‘mobile’ societies) or a defensive reaction
42 against that individualism, seen as expressing nostalgia for a social
43 order based on the existence of a ‘community’.2 Individualism only

exists in the concrete forms of market competition (including the 1

competition between labour powers) in unstable equilibrium with 2
association between individuals under the constraints of the class 3
struggle. Egalitarianism only exists in the contradictory forms of po- 4
litical democracy (where that democracy exists), the ‘welfare state’ 5
(where that exists), the polarization of conditions of existence, cul- 6
tural segregation and reformist or revolutionary utopias. It is these 7
determinations, and not mere anthropological figures, which confer 8
an ‘economic’ dimension upon racism. 9
Nevertheless, the heterogeneity of the historical forms of the rela- heterogeneity 10
tionship between racism and the class struggle poses a problem. This 11
ranges from the way in which anti-Semitism developed into a bogus 12
‘anti-capitalism’ around the theme of ‘Jewish money’ to the way in 13
which racial stigma and class hatred are combined today in the cat- 14
egory of immigration. Each of these configurations is irreducible (as 15
are the corresponding conjunctures), which make it impossible to 16
define any simple relationship of ‘expression’ (or, equally, of substi- 17
tution) between racism and class struggle. 18
In the manipulation of anti-Semitism as an anti-capitalist delu- 19
sion, which chiefly occurred between 1870 and 1945 (which is, we 20
should note, the key period of confrontation between the European 21
bourgeois states and organized proletarian internationalism), we 22
find not only the designation of a scapegoat as an object of prole- 23
tarian revolt, the exploitation of divisions within the proletariat and 24
the projective representation of the ills of an abstract social system 25
through the imaginary personification of those who control it (even 26
though this mechanism is essential to the functioning of racism).3 27
We also find the ‘fusion’ of the two historical narratives which are 28
capable of acting as metaphors for each other: on the one hand, the 29
narrative of the formation of nations at the expense of the lost unity 30
of ‘Christian Europe’ and, on the other, that of the conflict between 31
national independence and the internationalization of capitalist eco- 32
nomic relations, which brought with it the attendant threat of an in- 33
ternationalization of the class struggle. This is why the Jew, as an in- 34
ternally excluded element common to all nations but also, negatively, 35
by virtue of the theological hatred to which he is subject, as witness to 36
the love that is supposed to unite the ‘Christian peoples’, may, in the 37
imaginary, be identified with the ‘cosmopolitanism of capital’ which 38
threatens the national independence of every country while at the 39
same time re-activating the trace of the lost unity.4 40

The figure is quite different when anti-immigrant racism achieves 41
a maximum of identification between class situation and ethnic ori- 42
gin (the real bases for which have always existed in the inter-regional, 43

570 etienne balibar
1 international or intercontinental mobility of the working class;’

2 this has at times been a mass phenomenon, at times residual, but
3 it has never been eliminated and is one of the specifically proletar-
4 ian characteristics of its condition). Racism combines this identifi-
5 antagonistic social cation with a deliberate confusion of antagonistic social functions:
6 functions thus the themes of the ‘invasion’ of French society by North Africans
7 or of immigration being responsible for unemployment are con-
8 nected with that of the money of the oil sheikhs who are buying up
9 ‘our’ businesses, ‘our’ housing stock or ‘our’ seaside resorts. And
10 this partly explains why the Algerians, Tunisians or Moroccans have
11 to be referred to generically as ‘Arab’ (not to mention the fact that
12 this signifier, which functions as a veritable ‘switch word’, also con-
13 nects together these themes and those of terrorism, Islam and so on).
14 Other configurations should not, however, be forgotten, including
15 those which are the product of an inversion of terms: for example,
16 the theme of the ‘proletarian nation’, which was perhaps invented
17 in the 1920s by Japanese nationalism5 and was destined to play a
18 crucial role in the crystallization of Nazism, which cannot be left out
19 of consideration when one looks at the ways in which it has recently
20 reappeared.
21 The complexity of these configurations also explains why it is im-
22 possible to hold purely and simply to the idea of racism being used
23 against ‘class consciousness’ (as though this latter would necessar-
24 ily emerge naturally from the class condition, unless it were blocked,
25 misappropriated or de-natured by racism), whereas we accept as an
26 indispensable working hypothesis that ‘class’ and ‘race’ constitute the
27 two antinomic poles of a permanent dialectic, which is at the heart of
28 modern representations of history. Moreover, we suspect that the in-
29 strumentalist, conspiracy-theory visions of racism within the labour
30 movement or among its theorists (we know what high price was to
31 be paid for these: it is tremendously to the credit of Wilhelm Reich
32 that he was one of the first to foresee this), along with the mechanis-
33 tic visions which see in racism the ‘reflection’ of a particular class
34 condition, have also largely the function of denying the presence of
35 nationalism in the working class and its organizations or, in other
36 words, denying the internal conflict between nationalism and class
37 ideology on which the mass struggle against racism (as well as the
38 revolutionary struggle against capitalism) depends. It is the evolution
39 of this internal conflict I should like to illustrate by discussing here
40 some historical aspects of ‘class racism’.
41 Several historians of racism (Leon Poliakov, Michele Duchet and
42 Madeleine Réberioux, Colette Guillaumin, Eric Williams on modern
43 slavery, and others) have laid emphasis upon the fact that the modern

notion of race, in so far as it is invested in a discourse of contempt 1

and discrimination and serves to split humanity up into a ‘super- 2
humanity’ and a ‘sub-humanity’, did not initially have a national (or 3
ethnic), but a class signification. Or rather (since the point is to rep- 4
resent the inequality of social classes as inequalities of nature) a caste 5
signification.6 From this point of view, it has a twofold origin: first, in 6
the aristocratic representation of the hereditary nobility as a superior 7
‘race’ (that is, in fact, the mythic narrative by which an aristocracy, 8
whose domination is already coming under threat, assures itself of 9
the legitimacy of its political privileges and idealizes the dubious 10
continuity of its genealogy); and second, in the slave owners’ repre- 11
sentation of those populations subject to the slave trade as inferior 12
‘races’, ever predestined for servitude and incapable of producing an 13
autonomous civilization. Hence the discourse of blood, skin colour 14
and cross-breeding. It is only retrospectively that the notion of race race 15
was ‘ethnicized’, so that it could be integrated into the nationalist 16
complex, the jumping-off point for its successive subsequent meta- 17
morphoses. Thus it is clear that, from the very outset, racist repre- 18
sentations of history stand in relation to the class struggle. But this 19
fact only takes on its full significance if we examine the way in which 20
the notion of race has evolved, and the impact of nationalism upon it 21
from the earliest figures of ‘class racism’ onwards – in other words, 22
if we examine its political determination. 23
The aristocracy did not initially conceive and present itself in 24
terms of the category of ‘race’: this is a discourse which developed 25
at a late stage, the function of which is clearly defensive (as can be 26
seen from the example of France with the myth of ‘blue blood’ and 27
the ‘Frankish’ or ‘Germanic’ origin of the hereditary nobility), and 28
which developed when the absolute monarchy centralized the state 29
at the expense of the feudal lords and began to ‘create’ within its 30
bosom a new administrative and financial aristocracy which was 31
bourgeois in origin, thus marking a decisive step in the formation 32
of the nation-state. Even more interesting is the case of Spain in the 33
Classical Age, as analysed by Poliakov: the persecution of the Jews 34
after the Reconquista, one of the indispensable mechanisms in the 35
establishment of Catholicism as state religion, is also the trace of 36
the ‘multinational’ culture against which Hispanization (or rather 37
Castilianization) was carried out. It is therefore intimately linked to 38
the formation of this prototype of European nationalism. Yet it took 39
on an even more ambivalent meaning when it gave rise to the ‘stat- 40
utes of the purity of the blood’ (limpieza de sangre) which the whole 41
discourse of European and American racism was to inherit: a prod- 42
uct of the disavowal of the original interbreeding with the Moors and 43

572 etienne balibar
1 the Jews, the hereditary definition of the raza (and the correspond-
2 ing procedures for establishing who could be accorded a certificate
3 of purity) serves in effect both to isolate an internal aristocracy and
4 to confer upon the whole of the ‘Spanish people’ a fictive nobility, to
5 make it a ‘people of masters’ at the point when, by terror, genocide,
6 slavery and enforced Christianization, it was conquering and domi-
7 nating the largest of the colonial empires. In this exemplary line of
8 development, class racism was already transformed into nationalist
9 racism, though it did not, in the process, disappear.8
10 What is, however, much more decisive for the matter in hand
11 is the overturning of values we see occurring from the first half of
12 the nineteenth century onwards. Aristocratic racism (the prototype
13 of what analysts today call ‘self-referential racism’, which begins by
14 elevating the group which controls the discourse to the status of a
15 ‘race’ – hence the importance of its imperialist legacy in the colonial
16 context: however lowly their origins and no matter how vulgar their
17 interests or their manners, the British in India and the French in
18 Africa would all see themselves as members of a modern nobility) is
19 already indirectly related to the primitive accumulation of capital, if
20 only by its function in the colonizing nations. The industrial revolu-
21 tion, at the same time as it creates specifically capitalist relations of
22 new racism production, gives rise to the new racism of the bourgeois era (histori-
23 cally speaking, the first ‘neoracism’): the one which has as its target
24 the proletariat in its dual status as exploited population (one might
25 even say super-exploited, before the beginnings of the social state)
26 and politically threatening population.
27 Louis Chevalier has described the relevant network of significa-
28 tions in detail.9 It is at this point, with regard to the ‘race of labour-
29 ers’ that the notion of race becomes detached from its historico-
30 theological connotations to enter the field of equivalences between
31 sociology, psychology, imaginary biology and the pathology of the
32 ‘social body’. The reader will recognize here the obsessive themes of
33 police/detective, medical and philanthropic literature, and hence of
34 literature in general (of which it is one of the fundamental dramatic
35 mechanisms and one of the political keys of social ‘realism’). For the
36 first time those aspects typical of every procedure of racialization of
37 a social group right down to our own day are condensed in a single
38 discourse: material and spiritual poverty, criminality, congenital vice
39 (alcoholism, drugs), physical and moral defects, dirtiness, sexual
40 promiscuity and the specific diseases which threaten humanity with
41 ‘degeneracy’. And there is a characteristic oscillation in the presen-
42 tation of these themes: either the workers themselves constitute a
43 degenerate race or it is their presence and contact with them or in-

deed their condition itself which constitute a crucible of degeneracy 1

for the ‘race’ of citizens and nationals. Through these themes, there 2
forms the phantasmatic equation of ‘labouring classes’ with ‘danger- 3
ous classes’, the fusion of a socioeconomic category with an anthro- 4
pological and moral category, which will serve to underpin all the 5
variants of sociobiological (and also psychiatric) determinism, by tak- 6
ing pseudoscientific credentials from the Darwinian theory of evolu- 7
tion, comparative anatomy and crowd psychology, but particularly by 8
becoming invested in a tightly knit network of institutions of social 9
surveillance and control.10 10

Now this class racism is indissociable from fundamental histori- 11
cal processes which have developed unequally right down to the pres- 12
ent day. I can only mention these briefly here. First, class racism is 13
connected with a political problem that is crucial for the constitution 14
of the nation-state. The ‘bourgeois revolutions’ – and in particular the 15
French Revolution, by its radical juridical egalitarianism – had raised 16
the question of the political rights of the masses in an irreversible 17
manner. This was to be the object of one and a half centuries of social 18
struggles. The idea of a difference in nature between individuals had 19
become juridically and morally contradictory, if not inconceivable. 20
It was, however, politically indispensable, so long as the ‘dangerous 21
classes’ (who posed a threat to the established social order, property 22
and the power of the ‘elites’) had to be excluded by force and by legal 23
means from political ‘competence’ and confined to the margins of 24
the polity – as long, that is, as it was important to deny them citizen- 25
ship by showing, and by being oneself persuaded, that they consti- 26
tutionally ‘lacked’ the qualities of fully fledged or normal humanity. 27
Two anthropologies clashed here: that of equality of birth and that of 28
a hereditary inequality which made it possible to re-naturalize social 29
antagonisms. 30
Now, this operation was overdetermined from the start by na- national ideology 31
tional ideology. Disraeli11 (who showed himself, elsewhere, to be a 32
surprising imperialist theorist of the ‘superiority of the Jews’ over 33
the Anglo-Saxon ‘superior race’ itself) admirably summed this up 34
when he explained that the problem of contemporary states was the 35
tendency for a single social formation to split into ‘two nations’. In 36
so doing, he indicated the path which might be taken by the domi- 37
nant classes when confronted with the progressive organization of 38
the class struggle: first divide the mass of the ‘poor’ (in particular by 39
according the qualities of national authenticity, sound health, mo- 40
rality and racial integrity, which were precisely the opposite of the 41
industrial pathology, to the peasants and the ‘traditional’ artisans); 42
then progressively displace the markers of dangerousness and he- 43

574 etienne balibar
1 redity from the ‘labouring classes’ as a whole on to foreigners, and

2 in particular immigrants and colonial subjects, at the same time as
3 the introduction of universal suffrage is moving the boundary line
4 between ‘citizens’ and ‘subjects’ to the frontiers of nationality. In this
5 process, however, there was always a characteristic lag between what
6 was supposed to happen and the actual situation (even in countries
7 like France, where the national population was not institutionally
8 segregated and was subject to no original apartheid, except if one
9 extends one’s purview to take in the whole of the imperial territory):
10 class racism against the popular classes continued to exist (and, at
11 the same time, these classes remained particularly susceptible to ra-
12 cial stigmatization, and remained extremely ambivalent in their atti-
13 tude towards racism). Which brings us to another permanent aspect
14 of class racism.
15 I am referring to what must properly be called the institutional
16 institutional racialization of manual labour. It would be easy to find distant origins
17 racialization of for this, origins as old as class society itself. In this regard, there is
18 manual labour no significant difference between the way contempt for work and
19 the manual worker was expressed among the philosophical elites of
20 slave-owning Greece and the way a man like Taylor could, in 1909,
21 describe the natural predisposition of certain individuals for the ex-
22 hausting, dirty, repetitive tasks which required physical strength, but
23 no intelligence or initiative (the ‘man of the type of the ox’ of the
24 Principles of Scientific Management: paradoxically, an inveterate pro-
25 pensity for ‘systematic soldiering’ is also attributed to this same man:
26 this is why he needs a ‘man to stand over him’ before he can work
27 in conformity with his nature).12 However, the industrial revolution
28 and capitalist wage labour here effect a displacement. What is now
29 the object of contempt – and in turn fuels fears – is no longer manual
30 labour pure and simple (we shall, by contrast, see this theoretically
31 idealized – in the context of paternalistic, archaizing ideologies – in
32 the form of ‘craft work’), but mechanized physical work, which has
33 become ‘the appendage of the machine’ and therefore subject to a
34 violence that is both physical and symbolic without immediate prec-
35 edent (which we know, moreover, does not disappear with the new
36 phases of the industrial revolution, but is rather perpetuated both
37 in ‘modernized’ and ‘intellectualized’ forms – as well as in ‘archaic’
38 forms in a great many sectors of production).
39 This process modifies the status of the human body (the human
40 status of the body): it creates body-men, men whose body is a ma-
41 chine-body, that is fragmented and dominated, and used to perform
42 one isolable function or gesture, being both destroyed in its integrity
43 and fetishized, atrophied and hypertrophied in its ‘useful’ organs.

Like all violence, this is inseparable from a resistance and also from 1
a sense of guilt. The quantity of ‘normal’ work can only be recog- 2
nized and extracted from the worker’s body retrospectively, once its 3
limits have been fixed by struggle: the rule is overexploitation, the 4
tendential destruction of the organism (which will be metaphorized 5
as ‘degeneracy’) and, at the very least, excess in the repression of the 6
intellectual functions involved in work. This is an unbearable process 7
for the worker, but one which is no more ‘acceptable’, without ideo- 8
logical and phantasmatic elaboration, for the worker’s masters: the 9
fact that there are body-men means that there are men without bodies. 10
That the body-men are men with fragmented and mutilated bodies 11
(if only by their ‘separation’ from intelligence) means that the indi- 12
viduals of each of these types have to be equipped with a superbody, 13
and that sport and ostentatious virility have to be developed, if the 14
threat hanging over the human race is to be fended off…13 15

Only this historical situation, these specific social relations make 16
it possible fully to understand the process of aestheticization (and aestheticization 17
therefore of sexualization, in fetishist mode) of the body which char- 18
acterizes all the variants of modern racism, by giving rise either to 19
the stigmatization of the ‘physical marks’ of racial inferiority or to 20
the idealization of the ‘human type’ of the superior race. They cast 21
light upon the true meaning of the recourse to biology in the history 22
of racist theories, which has nothing whatever to do with the influ- 23
ence of scientific discoveries, but is, rather, a metaphor for – and 24
an idealization of – the somatic phantasm. Academic biology, and 25
many other theoretical discourses, can fulfil this function, provided 26
they are articulated to the visibility of the body, its ways of being and 27
behaving, its limbs and its emblematic organs. We should here, in 28
accordance with the hypotheses formulated elsewhere regarding 29
neo-racism and its link with the recent ways in which intellectual 30
labour has been broken down into isolated operations, extend the in- 31
vestigation by describing the ‘somatization’ of intellectual capacities, 32
and hence their racialization, a process visible everywhere – from the 33
instrumentalization of IQ to the aestheticization of the executive as 34
decision maker, intellectual and athlete.14 35

But there is yet another determining aspect in the constitution of 36
class racism. The working class is a population that is both hetero- 37
geneous and fluctuating, its ‘boundaries’ being by definition impre- 38
cise, since they depend on ceaseless transformations of the labour 39
process and movements of capital. Unlike aristocratic castes, or even 40
the leading fractions of the bourgeoisie, it is not a social caste. What 41
class racism (and, a fortiori, nationalist class racism, as in the case of 42
immigrants) tends to produce is, however, the equivalent of a caste 43

576 etienne balibar
1 closure at least for one part of the working class. More precisely, it
2 is maximum possible closure where social mobility is concerned,
3 combined with maximum possible openness as regards the flows of
4 proletarianization.
5 Let us put things another way. The logic of capitalist accumula-
6 tion involves two contradictory aspects here: on the one hand, mobi-
7 lizing or permanently de-stabilizing the conditions of life and work,
8 in such a way as to ensure competition on the labour market, draw
9 new labour power continually from the ‘industrial reserve army’ and
10 maintain a relative over-population; on the other hand, stabilizing
11 collectivities of workers over long periods (over several generations),
12 to ‘educate’ them for work and ‘bond’ them to companies (and also
13 to bring into play the mechanism of correspondence between a ‘pa-
14 ternalist’ political hegemony and a worker ‘familialism’). On the one
15 hand, class condition, which relates purely to the wage relation, has
16 nothing to do with antecedents or descendants; ultimately, even the
17 notion of ‘class belonging’ is devoid of any practical meaning; all that
18 counts is class situation, hic et nunc. On the other hand, at least a sec-
19 tion of the workers have to be the sons of workers, a social heredity has
20 to be created.15 But with this, in practice, the capacities for resistance
21 and organization also increase.
22 It was in response to these contradictory demands that the de-
23 mographic and immigration policies and policies of urban segrega-
24 tion, which were set in place both by employers and the state from
25 the middle of the nineteenth century onwards – policies which D.
26 anthroponomic Bertaux has termed anthroponomic practicesl6 – were born. These
27 practices have two sides to them: a paternalistic aspect (itself closely connected
28 to nationalist propaganda) and a disciplinary aspect, an aspect of ‘so-
29 cial warfare’ against the savage masses and an aspect of ‘civilizing’
30 (in all senses of the term) these same masses. This dual nature we
31 can still see perfectly illustrated today in the combined social and
32 police approach to the ‘suburbs’ and ‘ghettos’. It is not by chance that
33 the current racist complex grafts itself on to the ‘population problem’
34 (with its series of connotations: birth rate, depopulation and over-
35 population, ‘interbreeding’, urbanization, social housing, public
36 health, unemployment) and focuses preferentially on the question of
37 the second generation of what are here improperly called ‘immigrants’
38 with the object of finding out whether they will carry on as the pre-
39 vious generation (the ‘immigrant workers’ properly so-called) – the
40 danger being that they will develop a much greater degree of social
41 combativeness, combining class demands with cultural demands; or
42 whether they will add to the number of ‘declassed’ individuals, occu-
43 pying an unstable position between subproletarianization and ‘exit’

from the working class. This is the main issue for class racism, both 1
for the dominant class and for the popular classes themselves: to 2
mark with generic signs populations which are collectively destined 3
for capitalist exploitation – or which have to be held in reserve for 4
it – at the very moment when the economic process is tearing them 5
away from the direct control of the system (or, quite simply, by mass 6
unemployment, is rendering the previous controls inoperative). The 7
problem is to keep ‘in their place’, from generation to generation, 8
those who have no fixed place; and for this, it is necessary that they 9
have a genealogy. And also to unify in the imaginary the contradic- 10
tory imperatives of nomadism and social heredity, the domestication 11
of generations and the disqualification of resistances. 12
If these remarks are well founded, then they may throw some 13
light on what are themselves the contradictory aspects of what I shall 14
not hesitate to call the ‘self-racialization’ of the working class. There self-racialization of 15
is here a whole spectrum of social experiences and ideological forms the working class 16
we might mention: from the organization of collectivities of workers 17
around symbols of ethnic or national origin to the way in which a 18
certain workerism, centred on criteria of class origins (and, conse- 19
quently, on the institution of the working-class family, on the bond 20
which only the family establishes between the ‘individual’ and ‘his 21
class’) and the over-valorization of work (and, consequently, the vi- 22
rility which it alone confers), reproduces, within the ambit of ‘class 23
consciousness’, some part of the set of representations of the ‘race 24
of workers’.17 ‘Admittedly, the radical forms of workerism, at least in 25
France, were produced more by intellectuals and political apparatuses 26
aiming to ‘represent’ the working class (from Proudhon down to the 27
Communist Party) than by the workers themselves. The fact remains 28
that they correspond to a tendency on the part of the working class to 29
form itself into a closed ‘body’, to preserve gains that have been made 30
and traditions of struggle and to turn back against bourgeois society 31
the signifiers of class racism. It is from this reactive origin that the 32
ambivalence characterizing workerism derives: the desire to escape 33
from the condition of exploitation and the rejection of the contempt 34
to which it is subject. Absolutely nowhere is this ambivalence more 35
evident than in its relation to nationalism and to xenophobia. To the 36
extent that in practice they reject official nationalism (when they do 37
reject it), the workers produce in outline a political alternative to the 38
perversion of class struggles. To the extent, however, that they proj- 39
ect on to foreigners their fears and resentment, despair and defiance, 40
it is not only that they are fighting competition; in addition, and much 41
more profoundly, they are trying to escape their own exploitation. 42
It is a hatred of themselves, as proletarians – in so far as they are in 43

578 etienne balibar
1 danger of being drawn back into the mill of proletarianization – that

2 they are showing.
3 To sum up, just as there is a constant relation of reciprocal de-
4 termination between nationalism and racism, there is a relation of
5 reciprocal reciprocal determination between ‘class racism’ and ‘ethnic racism’
6 determination and these two determinations are not independent. Each produces its ef-
7 fects, to some extent, in the field of the other and under constraints
8 imposed by the other. Have we, in retracing this overdetermination
9 in its broad outline (and in trying to show how it illuminates the con-
10 crete manifestations of racism and the constitution of its theoretical
11 discourse), answered the questions we posed at the beginning of this
12 chapter? It would be more accurate to say that we have reformulated
13 them. What has elsewhere been called the excess which, by compari-
14 son with nationalism, is constitutive of racism turns out at the same
15 time to be a shortfall as far as the class struggle is concerned. But,
16 though that excess is linked to the fact that nationalism is formed in
17 opposition to the class struggle (even though it utilizes its dynamic),
18 and that shortfall is linked to the fact that the class struggle finds
19 itself repressed by nationalism, the two do not compensate one another;
20 their effects tend, rather, to be combined. The important thing is
21 not to decide whether nationalism is first and foremost a means of
22 imagining and pursuing the unity of state and society, which then
23 runs up against the contradictions of the class struggle, or whether it
24 is primarily a reaction to the obstacles which the class struggle puts
25 in the way of national unity. By contrast, it is crucially important to
26 note that, in the historical field where both an unbridgeable gap be-
27 tween state and nation and endlessly re-emerging class antagonisms
28 are to be found, nationalism necessarily takes the form of racism, at
29 times in competition with other forms (linguistic nationalism, for
30 example) and at times in combination with them, and that it thus
31 becomes engaged in a perpetual headlong flight forward. Even when
32 racism remains latent, or present only in a minority of individual
33 consciousnesses, it is already that internal excess of nationalism
34 which betrays, in both senses of the word, its articulation to the class
35 struggle. Hence the ever recurring paradox of nationalism: the re-
36 gressive imagining of a nation-state where the individuals would by
37 their nature be ‘at home’, because they would be ‘among their own’
38 (their own kind), and the rendering of that state uninhabitable; the
39 endeavour to produce a unified community in the face of ‘external’
40 enemies and the endless rediscovery that the enemy is ‘within’, iden-
41 tifiable by signs which are merely the phantasmatic elaboration of its
42 divisions. Such a society is in a real sense a politically alienated soci-
43 ety. But are not all contemporary societies, to some degree, grappling
with their own political alienation?

Notes 1
2
1 Pierre Ayçoberry, The Nazi Question; An Essay on the Interpretation of National 3
Socialism (1922-73), transl. R. Hurley, Routledge & Kegan Paul, London 1981. 4
2 See the theorizations of Karl Popper, The Open Society and Its Enemies (2
vols), 5th edn (revised), Routledge & Kegan Paul, London 1966; and, more
5
recently, of Louis Dumont, Essays on Individualism: Modern Ideology in 6
Anthropological Perspective; University of Chicago Press, Chicago 1986. 7
3 The personification of capital, a social relation, begins with the very figure of 8
the capitalist. But this is never sufficient in itself for arousing an emotional 9
reaction. This is why, following the logic of ‘excess’, other real-imaginary
10
traits accumulate: life-style, lineage (the ‘200 families’*), foreign origins, se-
cret strategies, racial plots (the Jewish plan for ‘world domination’), etc. The 11
fact that, specifically in the case of the Jews, this personification is worked 12
up in combination with a process of fetishization of money is clearly not ac- 13
cidental. 14
* The idea that 200 families held most of the wealth of France and used it to 15
exert political power was current in France in the 1930s, being quoted by
Daladier at the Radical Congress of 1934. It seems probable that the figure
16
200 derived from the number of shareholders allowed to attend the annual 17
meeting of the Bank of France. 18
4 Matters are further complicated by the fact that the lost unity of ‘Christian’ 19
Europe, a mythic figuration of the’ ‘origins of its civilization’, is thus repre- 20
sented in the register of race at the point when that same Europe is embark-
21
ing upon its mission of ‘civilizing the world’, i.e. submitting the world to its
domination, by way of fierce competition between nations. 22
5 Cf. Benedict Anderson, Imagined Communities, Verso, London 1983, pp. 92- 23
3. 24
6 L. Poliakov, The History of Anti-semitism (4 vols), transl. R. Howard, Routledge 25
& Kegan Paul, London 1974; M. Duchet & M. Réberioux, ‘Préhistoire et his- 26
toire du racisme’, in P. de Commarond and C. Duchet, eds, Racisme et soci-
27
eté, Maspero, Paris 1969; C. Guillaumin, L’idéologie raciste. Genèse et langage
actuel, Mouton, Paris-The Hague 1972; ‘Caractères spécifiques de l’idéologie 28
raciste’, Cahiers internationaux de sociologie, vol. LIII, 1972; ‘Les ambigu- 29
ïtés de la catégorie taxinomique “race”’, in L. Poliakov ed., Hommes et bêtes: 30
Entretiens sur le racisme (I), Mouton, Paris-The Hague 1975; Eric Williams, 31
Capitalism and Slavery, University of North Carolina Press, Chapel Hill 1944.
32
7 And one which substitutes itself, in the French case, for the ‘ideology of the
three orders’, a basically theological and juridical ideology, which is, by con- 33
trast, expressive of the organic place occupied by the nobility in the building 34
of the State (‘feudalism’ properly so-called). 35
8 L. Poliakov, History of Anti-Semitism, vol. 2, pp. 222-32. 36
9 Louis Chevalier, Labouring Classes and Dangerous Classes in Paris during the 37
First Half of the Nineteenth Century. transl. F. Jellinek. Routledge & Kegan
38
Paul. London 1973.
10 Cf. G. Netchine, ‘L’individuel et le collect if dans les représentations psy- 39
chologiques de la diversité des êtres humains au XIXe siècle’, in L. Poliakov. 40
ed., Ni juif ni grec: Entretiens sur le racisme (II), Mouton, Paris-The Hague 41
1978; L. Murard and P. Zylberman. Le Petit Travailleur infatigable ou le 42
prolétaire régénéré. Villes-usines, habitat et intimités au XIXe siècle, Editions
43
Recherches, Fontenay-sous-Bois 1976.

580 etienne balibar
1 11 Cf. H. Arendt. ‘Antisemitism’, Part One of The Origins of Totalitarianism.

2 Andre Deutsch, London 1986, pp.68-79; L. Poliakov. History of Anti-semitism.
3 vol. 3. pp. 328-37: Karl Polanyi, ‘Appendix II: Disraeli’s “Two Nations” and
the problem of colored races’, The Great Transformation. Beacon Press,
4
Boston 1957. pp. 290-94.
5 12 Frederick Winslow Taylor, Principles of Scientific Management, 1911. See the
6 commentaries by Robert Linhart. Lenine, les paysans. Taylor, Seuil, Paris
7 1976; and Benjamin Coriat, L’Atelier et le chronomètre. Christian Bourgeois.
8 Paris 1979. See also my study, ‘Sur le concept de la division du travail manu-
el et intellectuel’ in Jean Belkhir et al., L’Intellectuel, l’intelligentsia et les manu-
9
els, Anthropos, Paris 1983.
10 13 Clearly, the ‘bestiality’ of the slave has been a continual problem, from
11 Aristotle and his contemporaries down to the modern slave trade (the hy-
12 persexualization to which it is subject is a sufficient indication of this): but
13 the industrial revolution brought about a new paradox: the ‘bestial’ body of
14 the worker is decreasingly animal and increasingly technicized and there-
fore humanized. It is the panic fear of a super-humanization of man (in his
15
body and his intelligence which is ‘objectivized’ by cognitive sciences and the
16 corresponding techniques of selection and training), rather than his sub-hu-
17 manization – or, in any case, the reversibility of these two – which discharges
18 itself in phantasies of animality and these are projected for preference on to
19 the worker whose status as an ‘outsider’ [étranger] confers upon him at the
same time the attributes of an ‘other male’, a ‘rival’.
20
14 See chapters I and 3 above.
21 15 Not only in the sense of individual filiation, but in the sense of a ‘population’
22 tending towards the practice of endogamy; not only in the sense of a trans-
23 mission of skills (mediated by schooling, apprenticeship and industrial dis-
24 cipline) but in the sense of a ‘collective ethic’, constructed in institutions and
25 through subjective identification. Alongside the works already cited. see J.-P.
de Gaudemar. La Mobilisation générale. Editions du Champ Urbain, Paris
26
1979.
27 16 Daniel Bertaux, Destins personnels et structure de classe, PUF, Paris 1977.
28 17 C.G. Noiriel. Longwy: Immigrés et prolétaires. 1880-1980, PUF, Paris 1985:
29 J. Fremontier, La Vie en bleu: Voyage en culture ouvrière, Fayard, Paris
30 1980; Françoise Duroux, ‘La Famille des ouvriers: mythe au politique?’, un-
published thesis, Université de Paris VII, 1982.
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
The ghetto and the ethnic enclave 3
4
Ceri Peach 5
6
7
8
9
10
11
12
This article is about the social process of assimilation and the spatial pattern
13
of concentration and dispersal. Ethnic concentrations are often associated 14
with segregation and regarded as a barrier for integration. The geographer 15
Ceri Peach, however, argues that this common-sense take on spatial concen- 16
tration is rooted in the early American literature on segregation. Peach gives 17
a critical review of this theoretical literature, arguing that it fails to make a 18
clear distinction between the ghetto and the ethnic enclave and presents 19
an alternative model for understanding spatial concentration. He also pres- 20
ents clear definitions and concrete operationalisations for the concepts of 21
the ghetto and the enclave, and discusses their theoretical implications. 22
This article is a good example of the kind of scholarship that does not take 23
common-sense notions for granted, but critically questions theoretical and 24
methodological issues. 25
26
27
This chapter examines the reasons behind the political fear of ethnic 28
concentrations. It asks the question: is segregation always bad? It ex- segregation 29
amines how a confusion of terminology in the early American analy- 30
sis of ethnic and racial segregation has produced a malignant effect 31
on the literature, politics and policies affecting the ways in which 32
minorities are accommodated in west European cities. 33
The paper is, in a way, a piece of intellectual archaeology, but 34
it has a potent message for our current understanding of segrega- 35
tion and policies of minority accommodation. The key point is that 36
there is a major difference between the ghetto and the ethnic en- 37
clave. However, American sociology for a long time failed to make 38
this distinction and, worse still, linked the ghetto to the enclave to 39
the suburb as the first of three spatial stages on the inevitable process 40
of ethnic assimilation. The problem of intellectual archaeology is to 41
separate (1) theory and methodology on the one hand from (2) mod- 42
els and application on the other. The theory and methodology devel- 43

582 ceri peach
1 oped in the literature are correct; the model of application has been
2 far too restrictive. The truth of the matter is that there are two basic
3 models of minority incorporation: the assimilationist (melting pot)
4 and the multiculturalist (mosaic). The enclave is part of the assimi-
5 lationist model; the ghetto is confined to the multiculturalist model.
6 Multiculturalism is a necessary condition for the ghetto, but it is not
7 a sufficient condition.
8 The central theory in the study of the spatial patterns of ethnic
9 residential segregation is that there is a direct relationship between
10 the social process of assimilation and the spatial pattern of dispersal
11 (Park 1926; Duncan & Lieberson 1959; Peach 1975; Massey 1984).
12 assimilation This view is, I believe, correct. The problem is that assimilation is not
13 the only model for ethnic accommodation and integration.
14 However, taking the theory first, seventy five years ago, Robert
15 Park argued:
16
17 “It is because social relations are so frequently and so inevitably cor-
18 related with spatial relations; because physical distances, so frequent-
19 ly are, or seem to be, the indexes of social distances, that statistics
20 have any meaning whatsoever for sociology” (Park 1926: 18).
21
22 From this observation developed one of the most fruitful theories
23 of the Chicago School of the 1920s and 1930s and one of the few
24 examples of cumulative social science. The theory equated the statis-
25 tical levels of residential segregation of minority ethnic populations
26 to their levels of assimilation to the wider society. High levels of seg-
27 regation were equated with non-assimilation; low levels with high
28 levels of assimilation.
29 The key process involved was social interaction; cultural behavior
30 was modified according to whether one interacted more with one’s
31 own ethnic group or with the charter population. This interaction
32 was controlled by proximity to, and intermingling with, the respec-
33 tive groups. Residential isolation was hypothesized to minimize
34 social interaction with outsiders while promoting social interaction
35 within the group. Within-group interaction was hypothesized to re-
36 inforce the group’s identity, language maintenance and in-marriage.
37
38
39 Interpretation and operationalization
40
41 Although the general proposition of the relationship between resi-
42 dential segregation and social assimilation was clear, there were
43 problems of operationalization. Assimilation was difficult to de-

fine. Books were written on the topic (for example, Gordon 1964). 1
However, Lieberson (1963) provides us with a helpful definition: “as- 2
similation has taken place when it is no longer possible to predict 3
anything about an individual or a group on the basis of their ethnic 4
origins than it is for any member of the population as a whole.” 5
Operationalization meant taking multidimensional comparisons multidimensionality 6
of the minority population in relation to the target of the core soci- 7
ety. Structural assimilation, or the large scale entry into prime group 8
(close friendship circles) of the core society was regarded by Gordon 9
as the key step (Gordon 1964: 81). Thereafter, intermarriage and oth- 10
er identificational changes were seen by Gordon to follow inevitably. 11
Thus operationally, assimilation was treated as a multi-dimensional 12
phenomenon and its progress was measured by examining longitu- 13
dinal change of its many variables (Gordon 1964). Acquisition of the 14
English language, socio-economic status, out-marriage, citizenship 15
were some of the variables examined. 16
Segregation also proved problematic to operationalize, largely be- 17
cause of the different ways in which it was conceptualized (Peach 18
1981). Residential segregation is also a multidimensional phenom- 19
enon. A large number of different techniques, differing not only in 20
mathematical formula but in conceptualization of segregation it- 21
self have been suggested (Peach 1981). Massey and Denton (1993) 22
have suggested a battery of five measures to measure what they have 23
termed the hyper segregation of African Americans. However, a re- 24
view paper by Duncan and Duncan (1955) effectively concentrated 25
most subsequent work on the Index of Dissimilarity (id). id mea- 26
sures the percentage of a population which would have to shift its 27
area of residence in order to replicate the distribution of the popula- 28
tion with which it is being compared. id is a measure of uneven- 29
ness with similar characteristics and values to the economists’ Gini 30
Index. Lieberson’s P* (Lieberson 1981), a measure of isolation, has 31
also come into more general use since the 1980s. Unlike id it is an 32
asymmetric measure. It recognizes that the degree of exposure of a 33
small group to a large group is different from the exposure of the 34
large group to the small group. Unlike id its use has tended to be 35
descriptive rather than analytical in correlation regressions. 36
37
38
Segregation and interaction 39
40
Duncan and Lieberson demonstrated for Chicago in the 1930s and 41
1950s that there was an inverse relationship between the level of seg- 42
regation of foreign national groups and the percent of the group able 43

584 ceri peach
1 to speak English. They also showed that high degrees of out-mar-

2 riage correlated with low levels of segregation (Duncan & Lieberson
3 1959; Lieberson 1963: 156-158). Their argument was taken further
4 by Peach (1980a; 1980b) who demonstrated that Kennedy’s (1944;
5 1952) triple melting pot (Protestants, Catholics and Jews) in New
6 Haven, Connecticut, did not exist. The Irish Poles and Italians in the
7 supposed Catholic melting pot were all highly segregated from each
8 other. Intermarriage rates between these groups were lower than
9 statistically expected while Irish intermarriage with the (‘Protestant’)
10 British, Germans and Scandinavians, from whom they had low lev-
11 els of segregation, were higher than statistically expected. Residential
12 segregation was the clearest predictor of group intermarriage.
13 social interaction Thus, residential mixing was hypothesized as the key to social in-
14 teraction. If residential mixing is limited to one’s own ethnic group,
15 then the values and taken-for-granted nature of the group’s beliefs
16 will be reinforced. If mixing takes place with outsiders, then taken-
17 for-granted values, language and expected marriage partner choice is
18 likely to become modified. Residential mixing is a necessary but not
19 sufficient condition for social interaction. However, where residen-
20 tial mixing takes place, it is likely to promote social interaction.
21 The hypothesis formed itself into what we may conceive of as a
22 simple three stage cycle. The first generation of immigrants clustered
23 together in high concentrations and high segregation in the central
24 city. There they were unassimilated, few spoke English; overwhelm-
25 ingly they married their own ethnic group. The second generation
26 moved a little away from their inner city port of entry; they were less
27 segregated; a higher proportion spoke English; a greater proportion
28 married out. The third generation suburbanized, spoke English and
29 intermarried fully. They were assimilated.
30
31
32 But assimilation is not the only model
33
34 However, one should not assume that assimilation was the desired
35 outcome for all groups. On the one hand, social assimilation is en-
36 hanced by residential dispersal, while on the other hand, residential
37 segregation has the opposite effect. Therefore, a group wishing to as-
38 similate will tend to disperse, whilst for a group wishing to maintain
39 its ethnic identity, clustering is an important strategy. It is also true
40 that a group that disperses tends to assimilate whether or not the
41 group as a whole is in favor of assimilation or not.
42 There are thus two basic ways in which minorities are accommo-
43 dated into a wider society: assimilation and integration. Assimilation

argues for the disappearance of difference either through conform- 1

ing to a dominant structure (as in Anglo conformism) or through 2
merging (as in the melting pot). Integration or plurality or multicul- 3
turalism means accommodation while maintaining a separate iden- 4
tity. Integration is often economic while maintaining social closure. 5
The two models will thus be expected to produce different spa- 6
tial outcomes. Assimilation requires spatial diffusion. The mi- 7
nority and majority become socially and residentially intermixed. 8
Multiculturalism or integration (as opposed to assimilation) posits 9
a plural society in which social encapsulation and residential con- 10
centrations and separation, through higher degrees of segregation, 11
remain. 12
13
14
Not all plurality is voluntary 15
16
However, the ghetto model may come about from totally different 17
causes. It may be either voluntarily embraced or negatively enforced 18
(Boal 1981). A hegemonic group wishing to separate itself from its 19
perceived inferiors will attempt to enforce segregation upon the low- 20
er group (Massey & Denton 1993; Lemon 1991). 21
There are therefore two diametrically different reasons for ethnic 22
segregation. Ethnic segregation may be either voluntarily adopted as 23
a strategy for group survival or else it may be negatively imposed imposition 24
upon a weaker group. 25
While there are two different models of accommodation, key 26
points of the interpretation of the levels of social integration repre- 27
sented by the degree of spatial segregation of groups from one anoth- 28
er remain the same. Low levels of segregation indicate high degrees 29
of social interaction; high levels of segregation represent low degrees 30
of social interaction. Thus interpreting the probable outcomes of giv- 31
en levels of segregation, it is not critical to know whether those levels 32
are the net result of positive or negative forces. 33
34
35
The problem with the Chicago School 36
37
The central problem with the Chicago School was that while it cor- 38
rectly conceptualized the relationship between spatial pattern and 39
social process, it failed to recognize that the unidirectional transition 40
from the highly concentrated inner city to suburban dispersal was 41
not an inevitable process nor was it the only process. The Chicago 42
School did not distinguish between the melting pot and the mosaic. 43

586 ceri peach
1 They did not distinguish between the assimilationist and the plural-
2 istic models. They did not distinguish between the ghetto and the
3 differentiation enclave. The ghetto and the immigrant colony were conceptualized
4 as interchangeable terms.
5
6 “The Chinatowns, the Little Sicilians, and the other so-called ‘ghet-
7 tos’ with which students of urban life are familiar are special types
8 of a more general species of natural area which the conditions and
9 tendencies of city life inevitably produce (...) the keener, the more en-
10 ergetic and the more ambitious very soon emerge from their ghettos
11 and immigrant colonies and move into an area of second immigrant
12 settlement, or perhaps into a cosmopolitan area in which the mem-
13 bers of several immigrant and racial groups live side by side” (Park
14 1926: 9).
15
16 Worse still, not only did the Chicago School fail to distinguish be-
17 tween the ghetto and the enclave, but it believed that the ghetto was
18 a stage within the melting pot model. It saw the ghetto as the first
19 stage of three generational progression of (1) ghetto, (2) enclave, (3)
20 suburb. In this fundamental misunderstanding the Chicago School
21 falsified the ethnic history of long settled groups, misunderstood the
22 processes affecting African Americans and mis-forecast their future
23 in American cities.
24 For the Chicago School, the terms “ghetto” and “enclave” were
25 not problematized. Furthermore, it was assumed that the outward
26 movement of minority ethnic populations away from the inner city
27 equated to dispersal. A series of researchers in the lower foothills
28 of the Chicago School busied themselves demonstrating the unstop-
29 pable outward diffusion of minority groups from their inner city seg-
30 regated ports of entry to their inevitable suburban diffusion (Cressey
31 1938; Ford 1950; Kiang 1968). However, while they demonstrated
32 the progressive shift of the center of gravity of ethnic groups away
33 from the cbd over time, in the case of African Americans, outward
34 movement did not always equate to dispersal. The ghetto moved out
35 with them like the tongue of a glacier.
36 Diagrammatically, Figure 1 shows the outward movement of the
37 centre of gravity of the ethnic populations of Chicago, over time,
38 from the cbd. The diagram also shows the decrease in the degree
39 of concentration as the suburbanizing process continues. The ex-
40 pected relationship between segregation, measured by the index of
41 dissimilarity and assimilation is shown in Figure 2. The combined
42 relationship between outward movement, decreasing segregation
43 and increasing assimilation is represented in Figure 3.

However, for the African American population, the level of segre- 1

gation remained obstinately fixed on the high side of the assimilation 2
diagram, even when the center of gravity of the group showed move- 3
ment away from the central city (Figure 4). 4
5
6
The flaws of the three generational model 7
8
While the basic hypothesis of the equation of high segregation with 9
non-assimilation remains valid, interpretations of the model were 10
flawed by the mistaken belief that the hypothesis had universal valid- universal validity 11
ity, that all groups would conform to this three generational cycle. 12
Even in the 1950s it was confidently declared by the Head of the 13
Chicago School, Philip Hauser, that African Americans would inevi- 14
tably follow this model. 15
16
“The Negro migrant to the city will, without question, follow the 17
same pattern of mobility blazed by the successive waves of immi- 18
grants who settled our central cities. Just as the immigrant under- 19
went a process of ‘Americanization’ the immigrant Negro is under- 20
going a process of ‘urbanization’. The Negro is already rising and 21
will continue to rise on the social-economic scale as measured by 22
education, occupation, income and the amenities of urban existence. 23
Furthermore, the Negro, in time, will diffuse through the metropoli- 24
tan area and occupy outlying suburban as well as central city areas” 25
(Hauser 1958: 65). 26
27
This view was deeply mistaken. It equated upward mobility with spa- 28
tial diffusion. It regarded the process of ghetto formation and disper- 29
sal as the same as the three generation process of other immigrant 30
groups. It regarded time as the independent variable for ghetto dis- 31
solution. It was wrong on all counts. 32
In reality, the African American ghetto was different in kind from 33
the ethnic enclave of the European and other ethnics. Park’s casual 34
equation of Chinatowns, Little Sicily’s and other so-called “ghettos” 35
with the black ghetto (Park 1926: 9) was deeply flawed. The black 36
ghetto was dually segregated; nearly all urban African Americans 37
lived in such areas; almost the whole population in such areas was 38
black. The enclaves, on the other hand, were dually dilute. Only a mi- 39
nority of ethnic lived in areas which were associated with them. Very 40
rarely did they form even a majority of the population of what were 41
42
43

588 ceri peach
1 Figure 1 Outward movement of minority groups, Chicago, 1890-1914

2
3 60 60
56 56
4 52 52
5 48 48
44 44 1890
6 40 40 1940
7 36 36
Percent
Percent
32 32
8 28 28
24 24
9 20 20
10 16 16
12 12
11 8 8
12 4 4
0 0
13 I II III IV V VI VII VIII XI X I II III IV V VI VII VIII XI X
Mile Zones Mile Zones
14 DISTRIBUTION OF PERSONS DISTRIBUTION OF PERSONS
15 BORN IN CHICAGO BORN IN IRELAND
16 60 60
17 56 56
52 52
18 48 48
44 44
19 40 40
20 36 36
Percent
Percent
32 32
21 28 28
22 24 24
20 20
23 16 16
24 12 12
8 8
25 4 4
0 0
26 I II III IV V VI VII VIII XI X I II III IV V VI VII VIII XI X
27 Mile Zones Mile Zones
DISTRIBUTION OF PERSONS DISTRIBUTION OF PERSONS
28 BORN IN GERMANY BORN IN POLAND
29 60 60
30 56 56
52 52
31 48 48
32 44 44
40 40
33 36 36
Percent
Percent
34 32 32
28 28
35 24 24
20 20
36 16 16
37 12 12
8 8
38 4 4
39 0 0
I II III IV V VI VII VIII XI X I II III IV V VI VII VIII XI X
40 Mile Zones Mile Zones
41 DISTRIBUTION OF PERSONS DISTRIBUTION OF NEGROES
BORN IN ITALY
42
43 Source: Ford 1950

Figure 2 Hypothesized relationship between segregation (Index of 1

Dissimilarity) and assimilation over time 2
3
Exepected Relationship between Residential assimilation model 4
Segregation and Social Assimilation; 5
Assimilation Model
6
7
Start Finish 8
9
10
11
12
100 0 13
High Segregation Low Segregation 14
Non Assimilation Total Assimilation 15
16
City of Chicago, 1930, 1950 and 1960 17
Indices of Dissimilarity for Foreign-born Whites 18
born in Poland
19
20
1930 1950
1960 21
(58) (45)
(38)
22
23
1990 24
(19)
25
26
27
100 0
28
29
supposedly “their” areas. Thomas Philpott’s (1978) book The Slum 30
and the Ghetto, hammered the point home (Table 1). 31
It can be seen that while 92.7% of the black population lived in 32
the black ghetto and the African American population formed 81.5% 33
of the population of the black ghetto, only 3% of the Irish lived in 34
Irish areas and they formed only one third of the population of Irish 35
areas. The two most concentrated white groups were the Italians and 36
the Poles. Just under half of the Italians lived in Italian areas and they 37
formed just under half of the population of Italian areas. The Poles 38
were a little more concentrated: 61% lived in Polish areas and they 39
formed just over half of the population of Polish areas. 40
41
42
43

590 ceri peach
1 Figure 3 Hypothesized relationship between decreases in segregation

2 and group outward movement from inner cities
3
4 100
5
6 First Ch
ang
% ethinic group
generation eo
7 INNER CITY ver
8 tim
Second e
9 generation
10 Third
11 generation
12 SUBURB
13 0
14 Distance from centre
15
16 Change over time
17 Second
50 generation
18 First 70 30 Third
19 generation generation
20
21
22
23
24 100 0
Degree of segregation, index of dissimilarity
25
High segregtion Low segregtion
26 Non-assimilation High assimilation
27 Non-English-speaking English-speaking
28 High in-marriage Low in-marriage
Wearing of traditional dress
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

Figure 4 Failure of the African American levels of segregation (id) to 1

decrease over time, Chicago, 1930-1990 2
3
Black-White Segregation Chicago 1930 – 1990 4
(Ward/tract level) 5
6
7
1930 (76) 8
1980 (80) 9
1990 (85) 10
1970 (92) 11
12
100 0 13
14
15
Table 1 ‘Ghettoization’ of ethnic groups, Chicago, 1930 16
17
Group Group’s Group’s Total Percentage Group’s 18
City “Ghetto” “Ghetto” of group percentage 19
Population Population Population “Ghettoized” “Ghetto”
20
Population
21
Irish 169,568 4,993 14,595 2.9 33.8 22
German 377,975 53,821 169,649 14.2 31.7 23
Swedish 140,913 21,581 88,749 15.3 24.3
24
25
Russian 169,736 63,416 149,208 37.4 42.5 26
Czech 122,089 53,301 169,550 43.7 31.4 27
Italian 181,861 90,407 195,736 49.7 46.2
28
29
Polish 401,316 248,024 457,146 61.0 54.3 30
Negro 233,903 216,846 266,051 92.7 81.5 31
32
Source: Philpott 1978: 141, Table 7 33
34
However, even these levels of concentration were different in 35
kind rather than different in degree from the situation of African 36
Americans. All the European minorities lived in mixed areas. Hardly 37
any of the blacks did. While white ethnic enclaves dissolved over 38
time, black ghettos intensified and expanded territorially in a com- 39
pact form. Even in 1990, the massive concentration of the African 40
American population into black areas of Chicago continued (Table 41
2). Two thirds of the African American population were living in ar- 42
eas which were 90% or more black; 82% were in areas that were 50% 43
or more black.

592 ceri peach
1 Table 2 Percentage of the African American population of Chicago

2 PMSA, living in tracts of a given black percentage, 1990
3
4 Black Percentage Black Population Percentage of the total Black
5 of Tract living in such tracts Population of Chicago
6 in such tracts
7 100% 111,804 8.4
8 99% or more 381,347 28.7
9
10 90% or more 884,725 66.5
11 50% or more 1,087,600 81.7
12 30% or more 1,163,969 87.5
13
Total Black 1,330,636
14
Population
15
16 Source: Based on data from GeoLytics Census cd +Maps us Census 1990 data
17
18 If one compares the Chicago situation in 1930 and 1990 with London
19 in 1991 (Table 3) the difference in kind rather than degree between
20 the situation of blacks in the US and Britain is vividly illustrated.
21 The column heading “Percentage of Group Ghettoized” simply cop-
22 ies Philpott’s category, but refers to the proportion of a group living
23 in areas arbitrarily defined as those where they form 30% + of an
24 enumeration district (block).
25 the ghetto-enclave By failing to distinguish between the ghetto and the ethnic en-
26 link clave, the two distinct phenomena were linked together as the first
27 two stages of the three generational model: ghetto – enclave – sub-
28 urb. From here it was an easy step to envisage groups occupying
29 these three positions as occupying places on an escalator. Those at
30 the bottom of the staircase, in the ghetto, were new arrivals; those
31 at the top, in the suburbs had been on the staircase longest and had
32 reached their destination. Those who were half way up had previ-
33 ously been at the bottom and were now on their way to the top.
34 From this conceptualization, it became easy to see time/space
35 substitutions in the three generational model. If, for the sake of ar-
36 gument, the Irish were suburbanized and the Poles were still in an
37 enclave and the African Americans in the ghetto, then it became pos-
38 sible to argue that a generation previously, the Irish were in the en-
39 clave and two generations ago, they were in the ghetto. The African-
40 Americans and the Poles were envisaged as representing the first
41 two stages of the Irish past. In the same way, the contemporarily
42 suburbanized Irish, predicted the Polish future in the next genera-
43 tion and the African American future in two generations. This, af-

ter all was what Hauser (1958) was predicting. However, while the 1
Polish/Irish time/space substitution was correct, the Irish future did 2
not exist for the black population. Nor did the contemporaneously 3
ghettoized black situation represent the Irish past. No other group 4
had experienced the hyper segregation of the African Americans. 5
6
Table 3 ‘Ghettoization’ of ethnic groups at ed level in Greater 7
London, 30% cutoff 8
9
Group Group’s Group’s Total Percentage Group’s 10
City “Ghetto” “Ghetto” of group percentage 11
Population Population Population “Ghettoized” “Ghetto”
Population
12
13
Non-white 1,346,119 721,873 1,589,476 53.6 45.4
14
Black 290,968 7,755 22,545 2.6 34.4 15
Caribbean
16
Black African 163,635 3,176 8,899 2.0 35.6
17
Black Other 80,613 . . . . 18
Indian 347,091 88,887 202,135 25.6 44.0 19
20
Pakistani 87,816 1,182 3,359 1.4 35.2
21
Bangladeshi 85,738 28,280 55,500 33.0 51.0 22
Chinese 56,579 38 111 0.0 34.2 23
Other Asian 112,807 176 572 0.2 30.8 24
25
Other Other 120,872 209 530 0.2 39.4
26
Irish born 256,470 1023 2,574 0.4 39.8 27
28
Source: Peach 1996 29
30
While Hauser in 1958 could confidently predict the inevitability of 31
black diffusion and assimilation, seven years later the whole optimis- 32
tic edifice collapsed with the publication of Karl and Alma Taeuber’s 33
book Negroes in Cities (1965). Using the first large scale availability of Negroes in Cities 34
census block data from the 1960 census, the Taeubers demonstrated 35
the overwhelming segregation of African Americans in American 36
cities. On a scale from 0 (no segregation) to 100 (total segregation) 37
the Taeubers showed that the mean segregation index was 86.2 for 38
the 207 cities for which block level data were available in 1960. They 39
showed that the index was high in all regions (1965: 37), that it was 40
high irrespective of whether city populations were large or small, 41
whether the non-white population was large or small, whether the 42
non-white percentage was high or low. They showed that indexes had 43

594 ceri peach
1 been high in the past and had remained high. Hauser’s comforting
2 expectation of decreasing segregation with time was a delusion.
3 The Taeubers also dealt a death blow to another American dream.
4 segregation This was that economic progress would dissolve racial segregation.
5 by income Using Lieberson’s (1963) technique of indirect standardization, they
6 calculated how segregated the black population of Chicago would be
7 from whites, if income differences were the only variable affecting
8 their distribution. This is achieved by applying the percentage that
9 African Americans form of each income band in the city population to
10 the appropriate number of person in each income band in each tract
11 in the city. For example, if blacks formed 10% of the middle income
12 group in Chicago, then 10% of the middle income group would be
13 expected to be black, wherever the middle income group lived and so
14 forth. Having calculated the “expected” distribution of black and white
15 in the city, the degree of segregation between the two groups could be
16 calculated and compared with the observed level of segregation. On
17 this basis, the observed level of segregation in Chicago in 1960 was
18 83 and the “expected” index was 10. In other words, only 10/83 or 12%
19 of the observed level of segregation could be attributed to differences
20 in income (Taeuber & Taeuber 1964). Blacks were segregated from
21 whites because black, not because they were poorer than whites.
22 Subsequent work by Massey and Denton (1993: 86) showed
23 that the intervening years since Taeuber and Taeuber’s work (pace
24 William Julius Wilson 1978) had not produced a decline in the signif-
25 icance of race. Massey and Denton demonstrated that irrespective of
26 income level, poor black were segregated from poor whites, middle
27 income blacks from middle income whites and rich blacks from rich
28 whites by the same massive amounts, with indexes over 80 almost
29 without exception (Table 4).
30
31 Table 4 Segregation by income in thirty metropolitan areas with the
32 largest black populations, 1970-1980
33
34 Income Category
35 Metropolitan area Under $2,500 $25,000-$27,500 $50,000 +
36
Northern areas
37
38 Boston 85.1 83.9 89.1
39 Buffalo 85.2 80.0 90.0
40
Chicago 91.1 85.8 86.3
41
42 Cincinnati 81.7 70.9 74.2
43

Table 4 Continued 1
2
Income Category 3
Metropolitan area Under $2,500 $25,000-$27,500 $50,000 + 4
5
Cleveland 91.6 87.1 86.4
6
Columbus 80.3 74.6 83.4 7
Detroit 88.6 85.0 86.4 8
9
Gary-Hammond-E.Chicago 90.6 89.5 90.9
10
Indianapolis 80.8 76.6 80.0 11
Kansas City 86.1 79.3 84.2 12
13
Los Angeles-Long Beach 85.4 79.8 78.9
14
Milwaukee 91.3 87.9 86.3 15
New York 86.2 81.2 78.6 16
17
Newark 85.8 79.0 77.5
18
Philadelphia 84.9 78.6 81.9 19
Pittsburgh 82.1 80.6 87.9 20
21
St. Louis 87.3 78.4 83.2
22
San Francisco-Oakland 79.9 73.7 72.1 23
Average 85.8 80.7 83.2 24
25
Southern Areas
26
Atlanta 82.2 77.3 78.2 27
Baltimore 82.4 72.3 76.8 28
29
Birmingham 46.1 40.8 45.2
30
Dallas-Ft. Worth 83.1 74.4 82.4 31
Greensboro-Winston Salem 63.2 55.1 70.8 32
33
Houston 73.8 65.5 72.7
34
Memphis 73.8 66.8 69.8 35
Miami 81.6 78.4 76.5 36
37
New Orleans 75.8 63.1 77.8
38
Norfolk-Virginia Beach 70.1 63.3 72.4 39
Tampa-St. Petersburg 81.8 76.0 85.7 40
41
Washington, D.C. 79.2 67.0 65.4
42
Average 74.4 66.7 72.8 43
Source: Massey &: Denton 1993 (Table 4.1, p. 86)

596 ceri peach
1 Thus we arrive at the realization that there is not one model of

2 American minority integration but two: the assimilationist and the
3 pluralist, the enclave and the ghetto. The great error has been to force
4 the pluralist model of African American segregation into the assimi-
5 lationist framework and to graft on the contemporary ghetto model
6 onto the historical European settlement patterns in cities. The ghetto
7 was different in kind; the ghetto was distinct from the ethnic enclave.
8 We can summarise some of the differences (Table 5).
9
10 Table 5 Summary of differences between the African American
11 ghetto and the ethnic enclave
12
13 African American Ghetto Ethnic Enclave
14 Dually segregated: Large majority of Dually dilute: only a minority of the
15 blacks are in it; large majority in it are group are in it; they form only a mi-
16 black nority of the population of the area
17 associated with the group
18 Negative Positive
19 Enforced Voluntary
20
21 Expanding Residual
22 Real Symbolic
23 Threatening Touristic
24
25 Permanent Temporary
26
27
28 Is all high segregation for negative reasons?
29
30 However, because the disproving of the universality of the Chicago
31 School’s three generational model was demonstrated through the ex-
32 ample of the African American ghetto, another error was created.
33 This error was the belief that all high levels of segregation were pro-
34 negative duced by negative discrimination.
35 discrimination The reason for this belief is not hard to find. First, the expecta-
36 tion of decreasing levels of segregation over time led to the belief
37 that high segregation was an early and primitive feature of minority
38 settlement. Secondly, nearly all of the available examples of high seg-
39 regation related to groups which were disadvantaged. Thirdly, the
40 key minority group, the Jews, on whom the plural model of socioeco-
41 nomic-progress-but-continuing-ethnic-segregation could be tested,
42 were not counted in the US census as either a national origin group
43 nor as an ethnic group. “Russian-born” was treated by some analysts

as a surrogate for Jewish origin (Lieberson & Waters 1988: 10-11) but 1
of course not all Jews were of Russian origin nor were all of Russian 2
origin Jewish. Nathan Kantrowitz (1969) hinted at segregation as a 3
viable strategy for groups that wished to maintain their ethnic identi- 4
ty, but in a fairly oblique way, arguing only that decreases in the level 5
of European segregation in American cities should not be expected 6
to continue for ever. 7
However, while US government identified the Jewish population as 8
a religious rather than ethnic group and therefore desisted the census 9
from enumerating them, the Canadian census harbored no such deli- 10
cacy. The Canadian census counts the Jewish population as both a reli- 11
gious and as an ethnic group. The levels of Jewish residential segrega- 12
tion in Canadian cities is markedly high (Table 6). In terms of the Index Index of 13
of Dissimilarity, Jewish segregation is as high as African American seg- Dissimilarity 14
regation in American cities. In Toronto and Montreal, which in 1991 15
contained the two largest concentrations of the Jewish population of 16
Canada, the ids were 75 and 82 respectively. The Canadian Jewish pop- 17
ulation is extremely successful on a socio-economic scale and although 18
anti Semitism exists in Canada, there is no indication that the levels 19
of Jewish segregation noted in the table are not the result of positive 20
wishes for association (Darroch & Marston 1972; Hiebert 1995). 21
22
Table 6 Indices of dissimilarity for the Jewish ethnic population of 23
Toronto, Montreal, Vancouver and Winnipeg, 1991 24
25
Toronto Montreal Vancouver Winnipeg Calgary 26
Jewish id 75.0 81.9 56.8 71.6 58.2 27
28
Jewish %
3.0 2.46 0.68 1.84 0.56 29
Pop
Jewish 30
114,735 76,780 10,930 11,980 4,240 31
Pop
Total Pop 3,893,046 3,127,242 1,602,502 652,364 754,033
32
33
Source: Author’s calculation from StatsCanada data 34
35
Perhaps, even more interesting about the Jewish patterns of segrega- 36
tion is the suggestion that it has come about accompanied not only 37
by upward social mobility but by suburbanization as well. 38
However, the high indices of dissimilarity for the Jewish population 39
in Toronto and Montreal are similar to those for African Americans 40
south of the border, the Jewish population lives in enclaves rather 41
than ghettos on the black model. The highest percentage that the reli- 42
gious Jewish population formed of any Toronto tract was 70% in 1991 43

598 ceri peach
1 and only 2% of the population lived there. Only a third of the Jewish
2 population lived in areas in which they formed a majority of the tract
3 population and all of these tracts held a mixed (i.e. non Jewish) popu-
4 lation as well (Table 7). In Montreal, the highest percentage which
5 the Jewish religious population formed of any tract was 90%. Like
6 Toronto, a third of the Montreal religious Jewish population lived in
7 tracts where they formed a majority of the population.
8 Europe In London, although we do not have ethnic census data, it is appar-
9 ent from other sources that the Jewish population which originally set-
10 tled in the working class East End at the end of the nineteenth century,
11 suburbanized, notably to the north western outer fringes of the city
12 during the twentieth century, but remained concentrated (Newman
13 1985; Waterman & Kosmin 1986a, 1986b). Such patterns of ethnic
14 pluralism may be referred to as relocating enclaves (see Figure 5)·
15 There is also evidence from European experience that some afflu-
16 ent minority ethnic populations manifest high levels of segregation.
17 Glebe’s work on the Japanese in Dusseldorf (1986) and White’s work
18 on the Japanese in London (1998) both indicate id’s in the seventies.
19 These groups differ, of course from settled minorities in that they
20 are largely composed of sojourners who are seconded by their firms
21 for a period of years. Such concentrations may be thought of as para-
22 chuted communities (see Figure 5).
23 On the other hand, there is considerable evidence that levels of
24 segregation in European cities do not approach the levels observed
25 for African Americans and that the European experience is closer to
26 the ethnic enclave model than to the ghetto (Amersfoort 1974, 1978,
27 1980, 1982, 1987; Amersfoort & Cortie 1973, 1994; Friedrichs 1998;
28 Giffinger 1998; Kempen & Özüekren 1998; Kempen & Van Weesep
29 1998; Kesteloot & Cortie 1998; Musterd et al. 1998).
30
31 Table 7 Percentage of the Religious Jewish Population of Toronto
32 living in tracts where they formed a given percentage of the
33 population
34
35 Jewish percentage Jewish population % of total Toronto Jewish
36 of tract population living in such tracts population in such tracts
37 70+ 3,135 2.1
38
60-69 20,470 13.5
39
40 50-59 29,300 19.4
41 40-49 14,955 9.9
42
Toronto 151,115 100.0 *
43
* = 3,9% of total Toronto population of 3,893,046

Summary of types of enclaves and ghettos 1

2
We may now summarize a variety of enclaves and ghettos in a dia- 3
gram. 4
5
Model 1 is the traditional assimilation-diffusion model of the Chicago 6
three generational schemas. This is the most widespread and gen- 7
eral type. Settlement begins in the inner city; the second generation 8
moves out a little and becomes more assimilated; the third genera- 9
tion is suburbanized diffused and totally assimilated. Even in its ear- 10
ly days, the center is not the exclusive preserve of one group. 11
12
Model 2 is the American Ghetto Model. It is involuntary and plural 13
(nonassimilatory). It starts in the inner city, but with almost exclusive 14
concentration of the minority. Nearly all blacks are in it; nearly all in 15
it are black. It expands outwards in a segment shape over time, but 16
remains dually exclusive. 17
18
Figure 5 Diagrammatic representation of different spatial models of 19
assimilation and multi-culturalism 20
21
22
Model 1 Assimilation - Diffusion 23
(Examples: Europeans in US cities)
24
25
26
Model 2 Involuntary Plural (High Segregation) 27
(Examples: Chicago South Side Black Ghetto) 28
29
30
Model 3a Voluntary Plural in Situ
Persistent Ethnic Enclave, moderately 31
high segregation 32
(Examples: Turks in Berlin, Pakistanis 33
in Birmingham) 34
35
Model 3b Voluntary Plural: Relocation 36
(Example: London Jewish model) 37
38
39
Model 3c PARACHUTED SUBURBAN 40
Instant suburbanization, Affluent immigrants 41
(Examples: London Indian, Düsseldorf Japanese,
London Japanese, Brussels American) 42
43

600 ceri peach
1 Model 3a is what we may call the voluntary plural, in situ, persistent

2 enclave. It is the San Francisco or New York Chinatown model. A
3 high proportion of the population of the areas may be of a given eth-
4 nic group, but the Chinese population in Chinatown forms only a
5 minority (often a small minority) of the total Chinese population of
6 the city. It is a symbolic or touristic center. It is an institutional or
7 market center. It may remain and persist over time, but it is not the
8 exclusive center of the ethnic group.
9
10 Model 3b is the voluntary plural relocated model. The Jewish inner
11 city location, which relocates en masse to the suburbs is the key exem-
12 plar. The London Jewish shift from the East End to the northwest-
13 ern suburbs is the best studied example. Although segregation levels
14 measured by the Index of Dissimilarity may be high, the areas are
15 not the exclusive preserve of the Jewish population, but are mixed.
16 Nor are all Jews living in such areas.
17
18 Model 3c is what I have termed the Parachuted Suburban model.
19 These are concentrated areas of affluent often transitory sojourners.
20 The Japanese in London and Dusseldorf or the Hong Kong Chinese
21 in Vancouver are good examples.
22
23
24 Discussion and conclusion
25
26 The United States has had an unparalleled, successful history of as-
27 similating minorities. Buoyed by this success, the theorization of this
28 process has been cast into a single model, which I have character-
29 ized as the three generational model. However, it has ignored the
30 multicultural or plural model and worse still, tried to make this es-
31 sentially contrasting model part of the assimilation model itself. Put
32 brick-in-the-pond crudely, the assimilation model is a brick-in-the-pond model. The
33 model group starts concentrated, segregated and unassimilated in the inner
34 city; it speaks a foreign language; it marries its own kind; it is unas-
35 similated. The second generation ripples out a little, mixes more with
36 the charter group, learns English and begins to marry out. The third
37 generation replicates the socioeconomic structure of the population
38 as a whole; it speak English; it is highly intermarried; it is suburban-
39 ized and assimilated.
40 The theory states that there is a direct relationship between the
41 degree of residential, spatial segregation and the degree of social dis-
42 tance: high spatial segregation, high social distance between groups;
43 low segregation, low social distance (and high degree of social inter-
action, marriage et cetera).

The methodology of using the Index of Dissimilarity to measure 1

the degrees of spatial segregation is correct. High indices of dissimi- 2
larity give an excellent and consistent measure of group social dis- 3
tances. 4
The problem lies in assuming that there is only one model. The 5
single model argument is for an inevitable, unidirectional change single model 6
from high to low segregation over time. This model does work for argument 7
a large number of groups in a large number of situations. But, it is 8
not the only model. The African American population does not fit 9
into this model, nor does the Jewish population. For a long time, at- 10
tempts were made to interpret the African American experience in 11
terms of the single model when the evidence pointed in a totally dif- 12
ferent direction. Black segregation was high and remained high. The 13
Jewish pattern escaped notice because the data were not collected in 14
the US. But, if we can extrapolate Canadian experience, where such 15
data exist, the Canadian data show an unmistakable pattern of high 16
and long-lasting segregation. But Jewish concentration does not con- 17
stitute a ghetto, but a voluntary enclave. 18
Because the us sociological analysis operated for so long on the 19
single model, it had a massively distorting effect on both historical 20
analysis and on contemporary policy. Historically, it was assumed 21
that all groups were previously as segregated as the contemporary 22
African American population. There was a mythological back-projec- 23
tion of current levels of black segregation onto the nineteenth centu- 24
ry history of European immigration. The ghetto came to be seen as a 25
stage through which all immigrant groups went. Hence the Chicago 26
School references to Irish ghettos, German ghettos et cetera. Since 27
there was only one model it was assumed that it was only a matter 28
of time before African Americans would diffuse through the urban 29
system and assimilate like the Irish and Germans. This process for 30
different groups was viewed as a time/space substitution, with old 31
groups representing the future positions of new groups and new 32
groups representing the past position of old groups. 33
However, the black ghetto was different in kind from the degree of 34
segregation experienced by other groups. It was massively more con- 35
centrated and dually segregated: nearly all blacks were in the ghetto 36
and nearly all of the ghetto population was black. The black ghetto 37
did not dissolve with time. The Jewish high levels of concentration, 38
also failed to dissipate over time. The precise locations did change. 39
There was movement from the inner city to the suburbs, but it was a 40
movement en masse: a relocation rather than a diffusion. Unlike the 41
black segregation, however, these concentrations were voluntaristic 42
and by no means as dually segregated as the black experience. Not all 43

602 ceri peach
1 of the Jewish population lived in Jewish areas, nor was the popula-
2 tion of Jewish areas all Jewish. Both the African American and the
3 Jewish populations were following plural rather than assimilatory
4 models. The assimilation model was not the only one.
5 Britain If we look at the contemporary experience of Britain, we can see
6 both the plural and the assimilatory models in existence. The Black
7 Caribbean population has followed the assimilatory trajectory. Its
8 levels of segregation in London have fallen, census by census since
9 1961. The areas of greatest concentration have experienced the great-
10 est losses of Caribbean population. The movement has followed
11 the classic pattern of outward movement from the center towards
12 the periphery. However, when we look at the Indian, Pakistani and
13 Bangladeshi populations, changes in population have tended to rein-
14 force rather than reduce existing areas of concentration.
15 Both the assimilation and the multicultural models equate disper-
16 sal, diffusion and low segregation with assimilation. However, the
17 dominant model of the Chicago School considered the assimilation
18 model to be the only one and considered its process to be inevitable.
19 It recognized the existence of the ghetto, but did not distinguish it
20 from the enclave. It conceptualized the ghetto as the first stage of the
21 sequence of the three generational model. It incorporated its very
22 antithesis as part of the model itself.
23 The failure to distinguish between the ghetto and the enclave
24 has had a pernicious effect on the understanding of ethnic areas in
25 American cities. First, it has conceptualized the ghetto as a temporary
26 phenomenon. In reality the ghetto has become permanent. Secondly,
27 it envisioned socio-economic improvement as the mechanism for
28 the dissolution of the ghetto; in reality, rich African Americans are
29 as segregated from rich whites as poor blacks are from poor whites.
30 Economic differences are not unimportant but they do not explain
31 black segregation. Thirdly, it encouraged academics to identify the
32 ghetto as a product of wealth difference rather than race (Harvey
33 1973: 120-152; Wacquant 1997; Peach 1998). Fourthly it has falsified
34 our view of ethnic history in the United States by envisioning a ghet-
35 toized past for the early years of all groups; it has led to the assump-
36 tion that Irish, Italian and other ethnic enclaves were homogeneously
37 made up of the Irish, Italians or whatever. They never were. Fifthly it
38 encouraged the belief that the African American ghetto would dis-
39 solve in a “natural” and inevitable way. Sixthly, it encouraged the
40 belief that all segregation was bad and negatively superimposed on
41 groups. In reality, for those groups who choose it and for whom it is
42 not enforced, concentration has many benefits. However, we need to
43 be able to recognize the difference between the chosen enclave and
the enforced ghetto.

References 1
2
Amersfoort, H. van 3
1974 Immigratie en Minderheidsvorming: een analyse van de 4
Nederlandse situatie 1945-1973. Alphen a/d Rijn: Samsom. 5
1978 “Migrant workers, circular migration and development.” 6
Tijdschrift voor Sociale en Economische Geografie 69: 17-26. 7
1982 Immigration And The Formation Of Minority Groups: The Dutch 8
Experience 1945-1975. Cambridge: Cambridge University Press. 9
Amersfoort H. van & C. Cortie 10
1994 Social Polarisation in a Welfare State? Immigrants in Dutch 11
Cities. Paper presented to an international conference on mi- 12
gration, social exclusion and the European city. 13
Amersfoort, H. van & L. de Klerk 14
1987 “The dynamics of immigrant settlement: Surinamese, Turks 15
and Moroccans in Amsterdam 1973-1983.” In: G. Glebe & J. 16
O’Loughlin (eds), Foreign Minorities in Continental European 17
Cities, 199-222. Stuttgart: Franz Steiner Verlag. 18
Amersfoort, J.M.M. van 19
1980 “Woonsegregatie, gettovorming en de overheid.” In: P.W. 20
Blau & C. Pastor (eds), Soort bij Soort: Beschouwingen over 21
Ruimtelijke Segregatie als Maatschappelijk Probleem. Deventer: 22
Van Loghum Slaterus. 23
Amersfoort, J .M.M. & C. Cortie 24
1973 “Het patroon van de Surinaamse vestiging in Amsterdam in de 25
periode 1968 t/m 1970.” Tijdschrift voor Sociale en Economische 26
Geografie 63: 283-294. 27
Boal, FW. 28
1981 “Residential segregation, Ethnic Mixing and Resource 29
Conflict.” In: C. Peach, V. Robinson & S. Smith (eds), Ethnic 30
Segregation in Cities, 235-251. London: Croom Helm. 31
Cressey, P.F. 32
1938 “Population Succession in Chicago: 1898-1930.” American 33
Journal of Sociology 44: 59-69. 34
Darroch, A.G. & W.G. Marston 35
1972 “Ethnic Differentiation: Ecological Aspects of a 36
Multidimensional Concept.” In: A.H. Richmond (ed.), 37
Readings in Race Relations, 107-128. Oxford: Pergamon Press. 38
Ford, R.G. 39
1950 “Population Succession in Chicago.” American Journal of 40
Sociology 56: 151-160. 41
Friedrichs, J. 42
1998 “Ethnic Segregation in Cologne, Germany, 1984-1994.” Urban 43
Studies 35: 1745-1764.

604 ceri peach
1 Giffinger, R.
2 1998 “Segregation in Vienna: Impact of Market Barriers and Rent
3 Regulations 1.” Urban Studies 35: 1791-1812.
4 Glebe, G.
5 1986 “Segregation and intra-urban mobility of a high-status ethnic
6 group: the case of the Japanese in Düsseldorf.” Ethnic and
7 Racial Studies 9: 432-441.
8 Gordon, M.M.
9 1964 Assimilation in American Life. New York: Oxford University
10 Press.
11 Harvey, D.
12 1973 Social Justice and the City. London: Edward Arnold.
13 Hauser, P.M.
14 1958 “On the impact of urbanism on social organization, human
15 nature and political order.” Confluence 7(1): 57-69.
16 Hiebert, D.
17 1995 “The social geography of Toronto in 1931: a study of residen-
18 tial differentiation and social structure.” Journal of Historical
19 Geography 21(1): 55-74.
20 Kantrowitz, N.
21 1969 “Ethnic and racial segregation in the New York Metropolis.”
22 American Journal of Sociology 74: 685-695.
23 Kempen, R. van & A.S. Özüekren
24 1998 “Ethnic segregation in Cities: New forms and explanations in
25 a Dynamic World.” Urban Studies 35: 1631-1656.
26 Kempen, R van & J. van Weesep
27 1998 “Ethnic Residential Patterns in Dutch Cities: Backgrounds,
28 Shifts and Consequences.” Urban Studies 35: 1813-1834.
29 Kennedy, R.J.R.
30 1944 “Single or Triple Melting Pot? Intermarriage in New Haven,
31 1870-1940.” American Journal of Sociology 49: 331-339.
32 1952 “Single or Triple Melting Pot? Intermarriage in New Haven,
33 1870-1950.” American Journal of Sociology 58: 56-59.
34 Kesteloot, C. & C. Cortie
35 1998 “Housing Turks and Moroccans in Brussels and Amsterdam:
36 The Difference Between Public And Private Markets.” Urban
37 Studies 35: 1835-1854.
38 Kiang, Y.-C.
39 1968 “The distribution of ethnic groups in Chicago.” American
40 Journal of Sociology 74: 292-295.
41 Lieberson, S.
42 1963 Ethnic Patterns in American Cities. New York: Free Press of
43 Glencoe.

Lieberson, S. & M. Waters 1

1988 From Many Strands: Ethnic and Racial Groups in Contemporary 2
America. New York: Russell Sage Foundation. 3
Massey, D.S. 4
1984 “Ethnic residential segregation: A theoretical synthesis and 5
empirical review.” Sociology and Sociological Research 69(3): 6
315-350. 7
Massey, D.S. & N. Denton 8
1993 American Apartheid. Cambridge (MA): Harvard University 9
Press. 10
Musterd, S., W. Ostendorf & M. Breebart 11
1998 Multi-ethnic Metropolis: Patterns and Policies. Dordrecht: Kluwer 12
Academic Publishers. 13
Park, R.E. 14
1926 “The Urban Community as a Special (Sic) Pattern And A 15
Moral Order.” In: E.E. Burgess (ed.), The Urban Community. 16
Chicago: University of Chicago Press (reprinted in C. Peach 17
(ed.), Urban Social Segregation. London: Longman, 1975). 18
Lemon, A. (ed.) 19
1991 Homes Apart: South Africa’s Segregated Cities. Bloomington 20
(Indiana): Indiana University Press. 21
Newman, D. 22
1985 “Integration and ethnic spatial concentration: The changing 23
distribution of the Anglo-Jewish community.” Transactions, 24
Institute of British Geographers ns 10: 360-370. 25
Peach, C. 26
1980a “Ethnic segregation and intermarriage.” Annals of the 27
Association of American Geographers 70(3): 371-381. 28
1980b “Which triple melting pot?” Ethnic and Racial Studies 3(1): 1-16. 29
1981 “Conflicting interpretations of segregation.” In: P. Jackson & 30
S.J. Smith (eds), Social Interaction and Ethnic Segregation, 19-34 31
(Special Publication of the Institute of British Geographers). 32
London: Academic Press. 33
1998 “Loic Wacquant’s Three Pernicious Premises in the Study 34
of the American Ghetto.” International Journal of Urban and 35
Regional Research 22(3): 507-510. 36
Philpott, T. 37
1978 The Slum and the Ghetto. New York: Oxford University Press. 38
Taeuber, K. E. & A. Taeuber 39
1964 “The Negro as an Immigrant Group.” American Journal of 40
Sociology 69(4): 374-382. 41
1965 Negroes in Cities, Chicago, Aldine. 42
Wacquant, L. 43

606 ceri peach
1 1997 “Three Pernicious Premises In The Study Of The American

2 Ghetto.” International Journal of Urban and Regional Research
3 21(2): 341-353.
4 Waterman, S. & B.A. Kosmin
5 1986a “Residential patterns and processes: a study of Jews in three
6 London boroughs.” Transactions, Institute of British Geographers
7 ns 13: 79-95.
8 1986b “Mapping an underenumerated ethnic population: Jews in
9 London.” Ethnic and Racial Studies 9(4): 484-501.
10 White, P.
11 199 “The Settlement Patterns of Developed World Migrants in
12 London.” Urban Studies 35(10): 1725-1744.
13 Wilson, W.J.
14 1978 The Declining Significance Of Race: Blacks And Changing
15 American Institutions. Chicago: University of Chicago Press.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
About the editors 3
4
5
6
7
8
9
10
Marco Martiniello is research director of the National Fund for 11
Scientific Research (FRS-FNRS) in Belgium and a professor of so- 12
ciology and politics at the University of Liège, where he serves as di- 13
rector of the Center for Ethnic and Migration Studies (CEDEM). He 14
is also a member of the IMISCOE Executive Board and president of 15
the International Sociological Association’s Research Committee on 16
Sociology of Migration. He has authored and edited numerous works 17
providing transatlantic comparative perspectives on migration, eth- 18
nicity, racism, multiculturalism and citizenship in the European 19
Union. Recent publications include The Transnational Political 20
Participation of Immigrants: A Transatlantic Perspective (Routledge 21
2009), Citizenship in European Cities: Immigrants, Local Politics and 22
Integration Policies (Ashgate 2004), Migration between States and 23
Markets (Ashgate 2004) and La nouvelle Europe migratoire: Pour une 24
politique proactive de l’immigration (Labor 2001). 25
26
Jan Rath is a professor of urban sociology at the University of 27
Amsterdam, where he serves as director of the Institute for 28
Migration and Ethnic Studies (IMES). He is also the European chair 29
of International Metropolis and a member of the IMISCOE Board of 30
Directors. An anthropologist and an urban studies specialist, he has 31
authored and edited numerous works on the sociology, politics and 32
economics of post-migration processes. Recent publications include 33
Tourism, Ethnic Diversity and the City (Routledge 2007), Immigrant 34
Entrepreneurs: Venturing Abroad in the Age of Globalization (Berg 35
2003), Unravelling the Rag Trade: Immigrant Entrepreneurship in Seven 36
World Cities (Berg/New York University Press 2002) and Immigrant 37
Businesses: The Economic, Political and Social Environment (Macmillan 38
2000). For more information see www.janrath.com. 39
40
41
42
43

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
List of Sources 3
4
5
6
7
8
9
10
Part I – The migration process 11
12
13
1 Castles, S. & G. Kosack (1972), ‘The function of labour immigra- 14
tion in Western European capitalism’, New Left Review 73(May- 15
June): 3-21. 16
17
2 Hammar, T. (1985), ‘Introduction to European immigration 18
policy: A comparative study’, European Immigration Policy: A 19
Comparative Study, 1-13. Cambridge: Cambridge University Press. 20
21
3 Faist, T. (1997), ‘The crucial meso-level’, in T. Hammar, G. 22
Brochmann, K. Tamas & T. Faist (eds.), International Migration, 23
Immobility and Development: Multidisciplinary Perspectives, 187- 24
217. Oxford/New York: Berg. 25
26
4 Vertovec, S. (1999), ‘Conceiving and researching transnational- 27
ism’, Ethnic and Racial Studies 22(2): 447-462. 28
29
5 King, R. (2002), ‘Towards a new map of European migration’, 30
International Journal of Population Geography 8(2): 89-106. 31
32
6 Guiraudon, V. (2003), ‘The constitution of a European immigra- 33
tion policy domain: A political sociology approach’, Journal of 34
European Policy 10(2): 263-282. 35
36
7 Sayad, A. (2004), ‘Immigration and “state thought”’, in A. Sayad 37
(ed.), The Suffering of the Immigrant, 278-293. Cambridge: Polity 38
Press. 39
40
41
42
43

610 list of sources
1 Part II – Modes of incorporation

2
3 8 Van Amersfoort, H. (1982), ‘Minority as a sociological concept’,
4 Immigration and the Formation of Minority Groups: The Dutch
5 Experience 1945-1975, 10-30. Cambridge: Cambridge University
6 Press.
7
8 9 Modood, T. (1983), ‘“Black”, racial equality and Asian identity’,
9 New Community 14(3): 397-405.
10
11 10 Brubaker, W. R. (1989), ‘Introduction to immigration, immigra-
12 tion and the politics of citizenship in Europe and North America’,
13 in W. R. Brubaker, Immigration and the Politics of Citizenship in
14 Europe and North America, 1-22. Lanham: University Press of
15 America.
16
17 11 Martiniello, M. (1993), ‘Ethnic leadership, ethnic communities’
18 political powerlessness and the state in Belgium’, Ethnic and
19 Racial Studies 16(2): 236-255.
20
21 12 Wieviorka, M. (1994), ‘Racism in Europe: Unity and diversity’, in
22 A. Rattansi & S. Westwood (eds.), Racism, Modernity and Identity:
23 On the Western Front, 173-188. Cambridge: Polity Press.
24
25 13 Bauböck, R. (1994), ‘Changing the boundaries of citizenship: The
26 inclusion of immigrants in democratic polities’, in R. Bauböck
27 (ed.), From Aliens to Citizens: Redefining the Status of Immigrants in
28 Europe, 199-232. Aldershot: Avebury.
29
30 14 Kloosterman, R. J. van der Leun & J. Rath (1999), ‘Mixed embed-
31 dedness: (In)formal economic activities and immigrant business-
32 es in the Netherlands’, International Journal of Urban and Regional
33 Research 23(2): 252-267.
34
35 15 Simon, P. (2000), ‘The mosaic pattern: Cohabitation between
36 ethnic groups in Belleville, Paris’, in S. Body-Gendrot & M.
37 Martiniello (eds.), Minorities in European Cities: The Dynamics
38 of Social Integration and Social Exclusion at the Neighbourhood
39 Level, 100-115. Houndsmill, Basingstoke, Hampshire/New York:
40 Macmillan/St. Martin’s Press.
41
42 16 Bousetta, H. (2000), ‘Political dynamics in the city: Three case
43 studies’, in S. Body-Gendrot & M. Martiniello (eds.), Minorities

list of sources 611
in European Cities: The Dynamics of Social Integration and Social 1

Exclusion at the Neighbourhood Level, 129-144. Houndsmill, 2
Basingstoke, Hampshire/New York: Macmillan/St. Martin’s 3
Press. 4
5
17 Favell, A. (2003), ‘Integration and nations: The nation-state and 6
research on immigrants in Western Europe’, Comparative Social 7
Research 22: 13-42. 8
9
10
Part III – Conceptual issues 11
12
18 Barth, F. (1969) ‘Introduction to ethnic groups and boundaries: 13
The social organization of cultural difference’, in F. Barth (ed.), 14
Ethnic Groups and Boundaries: The Social Organization of Culture 15
Difference, 9-38. London: George Allen & Unwin. 16
17
19 Rex, J. (1980), ‘The theory of race relations: A Weberian ap- 18
proach’, in UNESCO, Sociological Theories: Race and Colonialism, 19
117-142. Paris: UNESCO. 20
21
20 Anthias, F. & N. Yuval-Davis (1983), ‘Contextualizing feminism: 22
Gender, ethnic and class divisions’, Feminist Review 15(Winter): 23
62-75. 24
25
21 Solomos, J. (1986), ‘Varieties of Marxist conceptions of “race”, 26
class and the state: A critical analysis’, in J. Rex & D. Mason 27
(eds.), Theories of Race and Ethnic Relations, 84-109. Cambridge: 28
30
22 Bovenkerk, F., R. Miles & G. Verbunt (1990), ‘Racism, migration 31
and the state in Western Europe: A case for comparative analysis’, 32
International Sociology 5(4): 475-490. 33
34
23 Miles, R. & V. Satzewich (1990), ‘Migration, racism and “post- 35
modern” capitalism’, Economy and Society 19(3): 334-358. 36
37
24 Balibar, E. (1991), ‘Class racism’, in E. Balibar & I. Wallerstein 38
(eds.), Race, Nation, Class: Ambiguous Identities, 204-216. London: 39
Verso. 40
41
25 Peach, C. (2005), ‘The ghetto and the ethnic enclave’, in D. 42
P. Varady (ed.), Desegregating the City: Ghettos, Enclaves and 43
Inequalities, 31-48. Albany: State University of New York Press.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Index 3
4
5
6
7
8
9
10
A Aristotle, 279 11
Ackerman, B. A., 278 Asia, 7, 26, 49, 50, 216, 432, 462, 541 12
Africa, 7, 26, 50, 65, 78, 82, 122, 203, Asia-Pacific, 104
204, 216, 452, 457, 462, 477, 541, Assam, 452
13
559, 572 Assimilation Model, 589 14
Africanization, 544, 559 Associatie Marokkaanse Migranten 15
African Island of Réunion, 26 Utrecht (AMMU), 366 16
Agrarverhältnisse in Altertum, 444 Association de Mineurs Marocains du
Alba, R. D., 240, 376 Nord, 364 17
Albania, 112, 123 Association de Travailleurs Marocains 18
Alger, C., 98 en France (ATMF), 364 19
Algeria, 48, 89, 97, 342, 459 Atlanta, 595 20
Algiers, 485 Atlantic, 220
Alibhai-Brown, Y., 376, 379 Atlas of International Migration, 116 21
Aliens Act of 1905, 524 Australia, 36, 295, 389, 390, 529, 546 22
Aliens Act of 1973, 524 Austria, 25, 46, 149, 272, 284, 372, 376, 23
Althusser, L., 438, 493 546 24
Althusserian Marxism, 496 Autonomy model, 494, 500, 511
Alund, A., 380 25
Amendement 14th 1868, 225, 226 B 26
America/Americas the, 14, 68, 100, 111, Bachelard, G., 9 27
203, 210, 267, 378, 438, 448, 450, Bachu, P., 544 28
451, 453 Back, L., 392
Americanization, 561, 587 Baden-Württemberg, 29 29
Amersfoort, H. van, 183, 598 Baetsen, P., 327 30
Amicales, 366 Baggara, 421, 422, 424, 425 31
Amsterdam, 143, 145, 148-152, 159, 315, Bagley, C., 195 32
318, 326 Bakker, E. S. J., 326-328
Amsterdam Treaty, 150, 154, 158 Balibar, E., 266, 567 33
Ancient Greece, 458 Baltic states, 552 34
Andall, J., 125 Baltimore, 595 35
Anderson, B., 350, 530 Baluch tribe, 422 36
Anderson, P., 493, 508, 524 Baluchistan, 418
Antarctica, 284 Bamyeh, M. A., 91 37
Anthias, F., 124, 238, 392, 469, 480 Banfield, E., 87 38
Antwerp, 395 Bangladesh, 112, 121, 199 39
Anwar, M., 117 Banton, M., 389, 493 40
Appadurai, A., 95, 96, 99-101 Barbados, 455, 554
Arab Gulf, 103 Barcelona, 128 41
Arendt, H., 279, 300 Barkan, E., 262 42
Aristide, J. B., 100 Barker, M., 268 43

614 index
1 Barrett, G., 316, 319, 324 Bosnia, 103

2 Barret, M., 483 Bosscher, A., 152
Barth, F., 241, 350, 359, 407, 414, 418 Boston, 594
3 Barthes, R., 350 Bouamama, S., 364, 365
4 Basch, L., 92, 94, 100, 394 Boulevard de Belleville, 345, 346, 348
5 Base/superstructure model, 501 Bourdieu, P., 9, 73, 143, 156, 168
6 Basic Law of the Federal Republic, 222 Bousetta, H., 355, 380, 395
Basic value-expectancy model, 62 Bovenkerk, F., 517, 521, 533
7 Bastenier, A., 248 Box, C., 453
8 Baubock, R., 275 Boyd, M., 65
9 Baumann, Z., 104, 395 Brah, A., 396
10 Baumgartner, F., 144 Brandt, W., 34
Beck, U., 98 Bras, H. le, 391
11 Beijer, G. J., 543 Brass, P., 239, 241
12 Belgian law 1974, 363 Breckenridge, C., 95, 96
13 Belgium, 25, 46, 237, 238, 239, 244-251, Brent, 209
14 253-255, 273, 301, 355, 357, 372, 375, Brent Community Council, 481
387, 393, 518, 519, 546 Breton, R., 250
15 Belleville, 339, 341-345, 347-353 Britain, 25-30, 33, 34, 36, 46, 47, 49-51,
16 Belleville model, 352 57, 103, 104, 187, 202, 204, 206-
17 Belleville cohabitation model, 348 209, 216, 220, 224, 225, 322, 372,
18 Belleville quarter, 341 375, 376, 378, 379, 381, 383, 387,
Bellevilleuse La, 351 388, 391-393, 437, 465, 469, 470,
19 Bendix, R., 186 472, 473, 476, 477, 481-483, 485,
20 Ben-Tovim, G., 490, 493, 494, 500-503, 498, 505, 507-509, 512, 517, 521,
21 505, 507, 508, 510, 511 522, 524, 531, 538, 539, 541, 542,
22 Berger, J., 131 544, 546, 550-554, 556, 559, 592,
Berger, M., 9 602
23 Berghe van den, P. L., 192-194, 459, 545 Brittan, A., 490, 495
24 Berlin, 545, 599 BBC (Britisch Broadcasting
25 Berlin Wall, 158, 27 Corporation), 201, 207
26 Bertaux, D., 576 British Caribbean, 191, 522
Beveridge Report, 482 British Commission for Racial Equality,
27 Bhabha, H, 99 154
28 Bigo, D., 142 British Unions, 37
29 Big Overseas Experience (BOE), 127 Brochman, G., 59
30 Birmingham, 494, 495, 595, 599 Brouwer, J. W. de, 148
Bjorn, 411 Brown, R., 439
31 Black feminist movement, 470, 483 Brubaker, W. R., 215, 379, 387, 390, 532
32 Black Section movement, 201 Bruinsma, F., 322
33 Blalock, H. M., 438 Brussels, 128, 145, 153, 155, 157, 241,
34 Blankenburg, E., 322 248, 375, 381, 395
Blommaert, J., 367 Bucher, K., 444
35 Bloom, L. 184 Budapest process, 149
36 Blotevogel, H. H., 122 Buffalo, 594
37 Boal, F. W., 585 Buller, H., 128
38 Bodnar, J., 70 Bundesvereinigung der Deutschen
Body-Gendrot, S., 316, 324 Arbeitgeberverbaede (BDA), 556,
39 Bogoras, W., 411 557
40 Böhning, R., 540, 547 Burawoy, M., 490, 508, 512, 540
41 Bolaria, S., 540 Burgers, J., 367
42 Bolivia, 103 Burgers Report, 367
Bonacich, E., 244, 493, 510 Burgess, E., 353
43

index 615
Burt, R. S., 66 Chicago PMSA, 592 1

Butcher Vocational Training, 329, 330 Chicago three generational schemas, 2
599
C Chicago School, 11, 14, 340, 582, 585-
3
Caces, F., 85 587, 596, 601, 602 4
Calgary, 597 Chile, 224, 522, 543 5
Callovi, G., 152 China, 103, 455 6
Calvinism, 448, 453 Chinatown, 586, 587, 600
Campani, G., 124 Christianization, 572 7
Canada, 7, 103, 202, 215, 216, 221, 226, Christian Reformation, 460 8
227, 389, 522, 546, 597 Choldin, H. M., 66 9
Cape Verdeans, 125 Chow, R., 96 10
Capital, 472 Chuckchee, 411
Carby, H., 500 Çinar, D., 382 11
Carchedi, G., 492, 498 Cincinnati, 594 12
Carens, J., 216, 228, 229, 233 Citroen, 30 13
Caribbean/Caribbean Islands, 26, 68, City of Light, 339 14
89, 191, 216, 225, 457, 541, 544, Civil Rights Act 1866, 225
554, 556 Civil Service Commission, 210 15
Carnegie Endowment for International Civil War, 226 16
Peace or Metropolis, 375 Clark, K., 187 17
Carpentier, A., 339 Classical Age, 571 18
Carter, B., 518, 542, 554-556 Class/race model, 492
Carter, T., 205 Cleary, P., 543 19
Cassarino, J. P., 323 Cleveland, 595 20
Caste, Class and Race, 492 Clifford, J., 94 21
Castells, M., 92-94, 96, 98, 100, 320, Coast Lapp area, 430 22
394, 492, 498 Code of Hygiene for Islamic Butchers,
Castilianization, 571 328 23
Castles, S., 7, 21, 104, 114, 115, 375, 379, Cohen, R., 66, 93, 95, 97, 99, 114, 119, 24
519, 539-543, 546, 551, 552, 555, 556, 121, 124, 377, 394 25
558 Cohn-Bendit, D., 153 26
Catholicism, 448, 460, 571 Coing, H., 343
CBD (Central Business District), 586 Cold War, 68, 129, 147, 148, 545 27
Census in France 1968, 29 Collective consumption role, 526 28
Centraal Bureau voor de Statistiek Columbus, 595 29
(CBS), 317, 318 Commission Directorate for 30
Central Africa, 459 Employment and Social Affairs, 152
Central European Initiative, 149 Committee of Experts for Identity 31
Centre of working-class culture, 499 Documents and the Movement of 32
Centre for Contemporary Cultural Persons, 149 33
Studies (CCCS), 490, 494-496, Committee for Islamic Butchers, 330 34
498, 499, 501, 505-507, 509, 526 Commission for Racial Equality (CRE),
Centrum Partij, 518 208-210, 526 35
Certeau de, M., 341 Commodity Board for Cattle, Meat and 36
Ceylon, 452 Eggs, 326 37
Chairs of Committees, 205 Commonwealth, 27 38
Chamber of Commerce, 329, 330 Commonwealth Immigrants Act of
Chambers, I., 118 1962, 543, 555, 556 39
Charbonnages de France, 364 Communist movement, 438 40
Chevalier, L., 572 Communist Party, 577 41
Chicago, 583, 586, 588, 589, 591, 592, Community Relations Council, 208 42
594 Comte, A., 437
43

616 index
1 CFDT (Confédération Française Denton, N., 583, 585, 594, 595

2 Démocratique du Travail), 37 Department of Education and Science’s
CGT labour federation (Confédération survey, 210
3 Générale du Travail), 33, 37, 38, 364 Depo-Provera, 481
4 Connecticut, 584 Deschouwer, K., 386
5 Connell, J., 96 Detroit, 595
6 Connolly, W.E. 503, 511 Deutsche Institut für Wirtschaft, 391
Conseil consultatif des Immigrés de la Deutscher Hotel-und
7 Ville (CCILg), 361-363 Gaststaettenverband (DEHOGA),
8 Conseil consultatif pour les Populations 557
9 d’origine étrangère (CCPOE), 362 Deutsche Volksunions, 270
10 Conspiracy-theory, 570 DGB (Confederation of German Trade
Continental Europe, 11, 15 Unions), 38, 272
11 Control of Immigration Statistics, 550 Dicken, P., 96
12 Cornelius, W., 7, 85 Diehl, C., 386, 391
13 Corrigan, P., 523 Di Rupo, E., 249
14 Cortie, C., 598 Displaces Persons camp, 546, 552, 553
Costa del Sol, 116, 129 Disraeli, B. 573
15 Costa-Lascoux, J., 375 Dissanayake, W., 95
16 Council of Europe Committee of Dobson, J., 133
17 Experts on the Legal Aspects of Donselaar, J., 519
18 Territorial Asylum, Refugees, and Douglas, J. D., 206
Stateless People, 149 Dublin agreement, 149
19 Council Social Affairs, 155 Dublin Convention 1990, 145
20 Courbage, Y., 353 Durkheim, E., 437, 439
21 Couronnes metro station, 346 Dutch Caribbean, 522
22 Cox, O. C., 184, 492-494 Dutch Ministry of Economic Affairs,
Craignos les, 363, 364 325
23 Cram, L., 153 Dutch National Bureau against Racism,
24 Cressey, P. F., 586 154
25 Criminal Law, 321 Duchet, M., 570
26 Critical Theory, 438 Duffield, M., 554
Cromwell Cox, O., 455 Duncan, O.D. 582-584
27 Currency Reform 1949, 26 Dusseldorf, 598
28 Cwerner, S. B., 126 Dworkin, R., 300
29 Cyprus, 477, 551
30 Czechoslovakia, 544 E
East Africa, 452, 458, 459, 544
31 D East Amsterdam, 395
32 Dallas-Ft. Worth, 595 East Europe/Eastern Europe, 23, 47, 48,
33 Darfur, 421, 424 125, 216, 226, 277, 387, 522, 543,
34 Darroch, A. G., 597 545, 559
Darwinian theory of evolution, 573 East Germany, 26, 47, 272, 441, 442,
35 Dassetto, F., 248 538, 545
36 DaVanzo, J. S., 63 East Indies, 184
37 Dawa mosque, El, 367 Ecolo, 361
38 Dayton Agreement 1995, 123 Ecomienda, 450
Deakin, N., 542 Economic and Social Research Council
39 Debray, R., 464 (ESRC), 91, 102, 104
40 Delors, J., 157 Economic Control Service, 329, 330
41 Delphy, C., 473 Economic crisis of 1973/1974, 542
42 Demokratische Partei Deutschland, 270 Economist, The, 97, 322
Denmark, 25, 46, 151, 223, 303, 372, Economy and society, 441, 450
43 376, 383, 388

index 617
Ecuador, 103 European Council of Ministers 1

Edholm, F., 480 Tampere, 375 2
Edmonston, B., 376 European Court of Justice (ECJ), 143,
Edye, D., 556, 557 145, 150, 152, 156, 158
3
Egypt, 97, 481 European Economic Community (EEC), 4
Eidheim, H., 429, 430 27, 51, 145 5
Elias, N., 343, 351 European Economic Community (EEC) 6
Elkins, S., 462 countries, 550, 551
El Salvador, 100 European Immigration Policy: A 7
Empire Strikes Back: Race and Racism Comparative Study, 45 8
in 70s Britain, The, 495, 496, 498- European Monetary Union (EMU), 142, 9
500, 506, 507 153 10
Employment and Social Affairs DG, 153 European Parliament, 298
Engbersen, G., 320, 395 European Political Union, 248 11
Engels, F., 22, 24, 473, 494 ESE, 157 12
Engermann, S. L., 451 European Services Forum (ESF), 157 13
England, 221, 224, 273, 519 European Union (EU), 115, 119, 122, 14
Epstein, R. A., 323 125-127, 131, 141-146, 148, 149, 151,
Equal Treatment Directive, 154 152, 154-159, 298, 299, 357, 375, 15
Erasmus, 127 379, 381, 383 16
Erasmus exchanges, 126 European Union Studies Association, 17
Erikson, R., 386 141 18
Eritrea, 103 European University Institute Florence,
Escriva, A., 125 237 19
Espace Intégration, 364 EURODAC (European Dactyloscopie), 20
Esping-Andersen, G., 323 145 21
Esser, H., 62, 376, 386, 545 European Voluntary Workers, 26, 37, 22
Ethnic and Racial Studies, 13, 91, 92, 546, 553
200, 237 Eurostat, 130 23
Ethnic Groups and Boundaries: The Evans, P. B., 523, 524 24
Social Organization of Cultural EVWs (European Voluntary Workers), 25
Differences, 407 553 26
Ethnic School, 476 Export Processing Zones, 538, 549
Ethnomethodology, 438 27
Europe, 7, 8, 10, 13-16, 26, 33, 45, 47, F 28
50, 68, 70, 78, 91, 103, 104, 111, 112, Fainstein, S., 320 29
114, 119-122, 124, 125, 128, 129, 131- Faist, T., 59, 65, 115, 377, 385, 395 30
133, 142, 144, 145, 151-153, 215, 217, Fanon, F., 464
220, 221, 230, 243, 259, 260, 263, Fargues, P., 353 31
265, 267, 270, 271, 289, 355, 369, Favell, A., 143, 153, 371, 375, 378, 380 32
371, 372, 375-381, 383, 384, 386, 387, Fawcett, J. T., 62, 66 33
389, 390, 393-397, 437, 446, 447, Federal Republic of Germany, 216, 518, 34
460, 463, 469, 472, 521, 525, 529, 519, 524
532, 544, 546-548, 553, 554, 559 Fédération des Associations des Jeunes 35
European Commission, 126, 150, 153, de Quartier (FAJQ), 364 36
156 Feirabend, J., 366 37
European Commission’s Seventh Feminist Review, 469 38
Framework Programme, 13 Fennema, M., 330, 395
European Community (EC), 45, 142, Ferdinand, K., 422 39
148, 149, 244, 247, 260 Ferguson, J., 93 40
European Consortium for Sociological Fermin, A., 376 41
Research, 386 Fielding, A. J., 388, 519 42
European Council/Council of Europe, Fiji, 452
298, 375, 379 43

618 index
1 Findlay, A. M., 125, 548, 549 Status of Immigrants in Europe,

2 Finland, 47, 50, 223, 303 275
Finnmark, 430 Front National, 270, 271, 518
3 First World War/World War I, 23, 199, Fur case, 421, 434
4 227, 302, 525 Furnivall, J. S., 190, 414
5 Flanders, 248, 249, 270, 273
6 Florida, R., 8 G
Fogel, R. W., 451 Gabriel, J., 490, 493, 494, 500-503,
7 Foner, N., 99, 101 505, 507, 509-511
8 Forbes, I., 383 Gans, H., 250, 349
9 FO (Force Ouvrière), 37 Garbage can model, 141, 144, 147
10 Ford, R., 125 Gardner, R. W., 64
Ford, R. G., 586, 588 Gary-Hammond-E.Chicago, 595
11 Ford of Cologne, 30 Gatekeeper role, 525
12 Fordism, 122 Geddes, A., 132, 153-155
13 Foreigners Law 1965, 27 Geisser, V., 365
14 Foreign Origin Populations’ Gellner, E., 267, 293, 294, 376
Consultative Councils (FOPCC), General Agreement on Trade in
15 251, 252 Services (GATS), 157
16 Foundation for Population, Migration Geneva, 38, 128
17 and Environment (PME/BMU), 13 Geneva Convention, 68
18 Fox, R., 253 Genovese, E. D., 490, 492
Fox-Genovese, E., 490 German Democratic Republic, 48
19 France, 24-31, 33, 37, 40, 46, 47, 49-51, German Federal Republic, 34
20 57, 65, 87, 89, 126, 128, 144, 148, German Foreigners Law of 1965, 33
21 149, 150, 173, 215, 216, 219, 221- German Marshall Fund, 216
22 224, 233, 259, 264, 265, 270, 273, German Reich, 24
301, 303, 341, 355-357, 360, 364, 365, Germany, 24, 25, 27-34, 36, 38, 39, 46,
23 372, 375, 379, 381, 383, 387, 391, 518, 47, 49-51, 57, 60, 89, 103, 116, 126,
24 519, 521, 522, 524, 525, 546, 571, 144, 146-150, 156, 216, 219, 222,
25 574, 577 270, 272, 284, 288, 303, 357, 360,
26 France Plus, 365 372, 375, 380, 381, 387, 391, 441,
Frank, A. G., 461 525, 543, 544, 546, 557, 588
27 Frankfurt, 128 German Social Democratic movement,
28 Frankfurt Bureau for Multicultural 464
29 Affairs, 153 Ghetto model/American Ghetto Model,
30 Frankfurt School, 438 585, 599
Fraser, F., 206 Giddens, A., 377
31 Freedman, M. 500, 513 Giffinger, R., 598
32 Freedom Party of Austria (FPO), 155, Gilbertson, G. A., 80
33 270 Gillespie, M., 96
34 Free labour movement policy, 27 Gilroy, P., 95, 206, 379, 396, 491, 499,
Freeman, G. P., 157, 324, 519, 554, 556 505, 511
35 Free Movement of Workers, Migrant, Gini Index, 583
36 Integration and Anti-racism, 152 Giullaumin, C., 570
37 French Caribbean, 522 Gjessing, G., 411
38 French Community government, 249 Glazer, N., 15
French Positivism, 437 Glebe, G., 598
39 French Revolution, 186, 230, 524, 573 Glick-Schiller, N., 89, 92, 94, 100, 115
40 Freyer, P., 206 Global Political Networks, 103
41 Friedrichs, J., 598 Goa, 68
42 Fröbel, H. J., 538 Goffman, E., 414
From Aliens to Citizens: Redefining the Golbert, R., 132
43

index 619
Goldring, L., 86, 100, 101 Hear van, N., 93, 98 1

Goldthorpe, J., 386 Heckmann, F., 381 2
Gordon, M. M., 193, 583 Heisenberg, D., 141
Gouldner, A. W., 74 Held, D., 277
3
Gramsci, A., 438, 492 Henry, L., 358 4
Granovetter, M. S., 80, 83, 323 Herbert, U., 544 5
Great Britain, 65, 102, 302 Here for Good: Western Europe’s New 6
Great Depression, 55 Ethnic Minorities, 539
Greece, 7, 47, 50, 117, 123, 149, 199, Hiebert, D., 597 7
263, 396, 574 High-Level Working Group (HLWG), 8
Green, D. P., 500 151 9
Greensboro-Winston Salem, 595 Hispanization, 571 10
Groenendijk, K., 519 History Task Force, 539
Groeneveld-Yayci, A., 328 Hobbes, T., 275, 281, 290, 291 11
Gross National Product, 545 Hobsbawm, E., 294, 451 12
Guarnizo, L. E., 92, 94, 97, 101, 316, Hoffmann-Nowotny, H. J., 64, 65 13
394 Hofinger, C., 382 14
Guillon, M., 340 Hoggart, K., 128
Guiraudon, V., 141, 147 Holland, 190, 192, 194, 195, 518 15
Gulf War, 122 Hollifield, J., 7, 141 16
Gupta, A., 93 Holmes, C., 543 17
Gurak, D. T., 80 Home Office, 550 18
Guyana, 452, 456-458 Hong Kong, 225, 538
Hooks, B., 470 19
H Houston, 595 20
Haaland, G., 421, 424, 425 Hoxha, E., 123 21
Habermas, J., 73, 438 Hreblay, V., 146 22
Hague The, 315, 318, 326 Hugo, G., 70
Hailbronner, K., 216, 228 Hungary, 522, 544 23
Haiti, 89, 100 Husserl, E., 438 24
Halder, J., 155 Huysmans, J., 147 25
Hall, S., 95, 129, 378, 396, 480, 490, 26
495-499, 502, 507-509, 513, 526 I
Hammar, T., 45, 59, 216, 229, 306, Iglicka, K., 126 27
356, 379 Iman mosque, El, 363 28
Handboek Minderheden, 330 IMISCOE (International Migration, 29
Handsworth, 204 Integration and Social Cohesion in 30
Handsworth Harambee organization, Europe), 16
204 Immigrant Workers and Class 31
Hannerz, U., 92, 100, 102 Structure in Europe, 21 32
Hansen, R., 380 Immigrant Workers and Class 33
Hargreaves, A. G., 357 Structure in Western Europe, 539 34
Harris, C., 542, 553-556 Immigrants Communal Consultative
Harris, J. R., 111 Councils (ICCC), 251, 252 35
Harris, M., 185, 186 Immigration and the Politics of 36
Hart-Celler Act 1965, 226 Citizenship in Europe and North 37
Hartmann, 445 America, 215 38
Harvey, D., 602 Immigration Act of 1971, 28, 33, 51
Hatton, T. J., 68 Immigration Rules, 551 39
Hauser, P., 587, 593 Independent The, 322 40
Haussmann, G. E., 341 Index of Dissimilarity (ID), 583, 589, 41
Häußermann, H., 316 597, 601 42
Haut Conseil à l’Intégration, 375 India, 47, 50, 89, 97, 103, 203, 217, 225,
455, 555, 572 43

620 index
1 Indian caste system, 425 Jansen, S., 132

2 Indian sub-continent, 522, 544, 559 Japan, 522
Indianapolis, 595 Japanization, 561
3 Indonesia, 47, 86, 522 Jeffcoate, R., 507
4 INED (Institut National Etudes Jenkins, R., 375
5 Demographiques), 391 Jenkins, S., 360
6 Information Age, 93 Jessop, B., 508
INSEE (Institut National de la Joly, D., 360
7 Statistique et des Etudes Jones, B., 144
8 Economiques), 390 Jong de, G. F., 62
9 Intergovernmental Conference (IGC), Jong de, W., 352
10 154, 158, 159 Joppke, C., 380
International Journal of Population Joshi, S., 518, 542, 554-556
11 Geography, 111 Journal of Ethnic and Migration
12 International Labor Organisation (ILO), Studies, 13
13 68, 375 Jura, 303
14 International Migration, Immobility Justice and Home Affairs (JHA), 148,
and Development: Multidisciplinary 151
15 Perspectives, 59
16 International Migration Review, 13 K
17 International non-governmental organi- Kaldor, M., 99
18 zations (INGOs), 98, 99, 143 Kandre, P., 420
International Organization for Kansas City, 595
19 Migration, 119 Kant, I., 440
20 International Red Cross, 98 Kantrowitz, N., 597
21 International Refugee Organization Kastoryano, R., 153
22 (IRO), 545, 546 Kay, D., 543, 546, 553, 554, 559
Iran, 34, 224, 543 Keane, J., 524
23 Iraq, 121, 152, 224 Kearney, M., 100
24 Ireland, 7, 47, 128, 216, 224, 358, 380, Kelsen, H., 168, 169
25 396, 588 Kempen, R. Van., 598
26 Irish Republic, 26, 303, 552 Kennedy, P., 377
Iron Curtain, 120, 126 Kenya, 544
27 Isaac, J., 552 Kepel, G., 346
28 ISI Web of Knowledge, 13 Kerneis, P., 157
29 Island of Nevis, 451 Kesteloot, C., 248, 395, 598
30 Israel, 288, 481, 546 Keyes, C., 240
Italy, 7, 24, 27, 47, 123, 125, 149, 195, Keynsian economics/Keynesian
31 247, 250, 272, 273, 376, 387, 393, Economic Theory, 23, 55
32 396, 522, 546, 554, 588 Kiang, Y. -C., 586
33 Itissam association el, 364 Kibbutz, 485
34 Itissam mosque, El, 363 Kimble, J., 484
Izikowitzs, K. G., 415 Kindelberger, C., 541
35 IDS (Index of Dissimilarity), 597 King, R., 96, 111, 117, 121, 122, 129, 130
36 Kingdon, J., 144, 154
37 J Kleivan, H., 432
38 Jacobson, M. F., 99 Kloosterman, R., 315, 316, 318, 319, 323,
Jackson, J., 210 324, 332, 395
39 Jackson, J. A., 540 KnowNothings, 226
40 Jacquemet, G., 342 Knutsson, K. E., 425
41 Jamaica, 225 Kockel, U., 121, 128
42 Jamieson, A., 124 Kohistan, 421
Jansen, C., 112, 113 Kohl, H. 147
43

index 621
KMAN (Komitee Marokkaanse Lisbon Agenda, 8 1

Arbeiders in Nederland), 366 Little Sicily, 587 2
Koopmans, R., 392 Live, Y.S. 347
Korte, H., 545 Liverpool, 502
3
Kosack, G., 21, 519, 539-541, 543 Lloyd, C., 15 4
Koser, K., 122, 125, 130 Lobkowicz, W. de, 148 5
Kosmin, B. A., 598 Locke, J., 290, 291, 293, 302, 304 6
Kraal, K., 9 Lodge, D., 131
Kriesberg, L., 98, 99 Logical Positivism, 438 7
Kritz, M. M., 60 London, 29, 128, 209, 316, 383, 550, 8
Kymlicka, W., 373, 388 592, 598, 600, 602 9
Kypraia, 479 Lord, C, 150 10
K4 committee, 148, 151 Lorenzo, P., 11, 12
Los Angeles, 7, 316 11
L Los Angeles-Long Beach, 595 12
Labour Party, 201 Lucassen, J., 525 13
Laczko, F., 122 Lukacs, G., 438 14
Lamy, P., 157 Lukes, S., 237, 243
Landolt, P., 92, 94, 101 Lutz, H., 122, 125, 130 15
Lapeyronnie, D., 15 Luxemburg, 49 16
Latifundia, 441, 450 Luxemburg, R., 473 17
Latin America, 7, 47, 49, 78, 459, 462 Lyon, 30 18
Lavenex, S., 151
Law and order role, 526 M 19
Law on Economic Criminal Offences, Maastricht, 148, 153 20
326 Maastricht Treaty 1992, 298 21
Layton-Henry, Z., 532, 553-555 MacDonald, I., 519, 524, 550, 551 22
Lazaridis, G., 123, 124, 125 Madrid, 8, 116
Lazarsfeld, P. F., 438 Magobunje, A. L., 65 23
Leach, M., 425 Mahler, S. J., 100 24
Leagues, 272 Mahnig, H., 375 25
Lebanon, 482, 484 Majorca, 129 26
Lee, E. S., 61, 111 Malaya, 452, 457
Lenin, V., 23, 24, 473 Malaysia, 458 27
Lemon, A. 585 Ma Mung, E., 316, 324 28
Le Pen, J. M., 222 Mann, M., 374 29
Lessinger, J., 97 Mangrove case, 33 30
Lesthaege, R., 386 Marable, M., 490
Leun van der, J., 315, 318, 324, 332 March, J., 141, 142, 144, 147 31
Lever-Tracy, C., 540 Marches de Beurs 1983, 364 32
Levi-Strauss, C., 268, 351, 438, 439 Marcinelle mine, 247 33
Lewis, B., 347 Marcus, G. E., 101 34
Lieberson, S., 582-584, 594, 597 Marquez, G. G., 201
Lieberson’s P, 583 Marrus, M. R., 543-546 35
Liechtenstein, 49 Marseille, 316 36
Liège, 355, 360-365, 367, 368 Marshall, D., 83 37
Lier van, R. A. J., 190-192 Marshall, T. H., 186, 187 38
Light, I., 317 Marston, W. G., 597
Lijphart, A.,193, 194 Martin, P., 7, 97 39
Lille, 363-365, 368 Martinelli, A., 317, 323 40
Lille Sud mosque, 364 Martiniello, M., 7, 15, 237, 243, 247- 41
Lim, L. Y. C., 538 249, 361 42
Limburg, 249 Marx, K., 22, 24, 239, 439, 442-445,
43

622 index
1 455, 462, 465, 466, 520, 527, 528, Ministry of Labour, 552
2 539 Ministry of Public Health and
Marxism, 461, 462, 464, 471, 472, 489- Environmental Protection, 326
3 492, 494-499, 502, 505-507, 511, Minorities in European Cities, 355
4 512, 514 Minorities in the New World, 185
5 Marxist theory, 461, 489, 490, 508, Mitchell, K., 97, 99, 101
6 511-513 Modood, T., 201, 386, 392, 393
Marxist theories of migration, 539 Molyneux, M., 484
7 Marxist theory of race and ethnic rela- Montreal, 597, 598
8 tions, 512 Moravcsik, A., 150
9 Marxist theory of racism, 510 Morelli, A., 247
10 Massey, D. S., 60, 84, 85, 117, 582, 583, Morley, D., 96
585, 594, 595 Morocco, 47, 121, 195, 326
11 Mauritius, 452 Mouahidin mosque, El, 363
12 May Events, 33 Moynihan, D., 15
13 Maynard, M., 490, 495 Musterd, S., 325, 598
14 Mazey, S., 154 Myrdal, G., 85, 251, 440
McGrew, A., 277
15 McHugh, K., 126 N
16 Mc Intosh, M., 483 Nagel, J., 359
17 Mead, G. H., 383, 438 Nairn, T., 530
18 Meditteranean, 111 Narroll, R., 407
Meditteranean caravanserai, 130 Natal, 452, 458
19 Med TV, 96 National Association of Asian Probation
20 Melanesian trade systems, 415 Staff, 210
21 Member of European Parliament National Association of Local
22 (MEP), 153 Government Officers (NALGO),
Memmi, A., 347 205
23 Memphis, 595 National Front, 518
24 Ménilmontant, 346 Nationality Bill, 473
25 Merriman, J., 432 National Union of Moroccan Students
26 Messamah, K., 353 (UNEM), 362
Mexican Bracero Program, 7 Nauck, B., 392
27 Mexico, 86, 89, 117 Nazism, 570
28 Mezzogiorno, 27 Near East, 50
29 Miami, 595 Nee, V., 376
30 Middle Ages, 444 Nelli, H. S., 250
Middle America, 432 Netherlands, 7, 46, 47, 49-51, 57, 65,
31 Middle East, 424, 548, 559 126, 144, 183, 303, 315, 317-321, 324-
32 Migrants’ Forum, 153, 154, 158 326, 330-332, 355, 356, 360, 366,
33 Migrant labour model, 494, 503 372, 376, 382, 387, 518, 519, 521,
34 Migration Policy Group (MPG), 154, 522, 524, 525, 531
157, 158, 381 Neuchâtel, 303
35 Miles, R., 490, 493, 495, 503-506, 509, Nevada, 320
36 513, 517, 521, 527, 529, 533, 539, 540, Newark, 595
37 542, 552-556, 558-560 New Community, 201
38 Miller, M., 7, 216, 356, 369 New Left Review, 21
Milward, A., 397 New Haven, 584
39 Milwaukee, 595 New Jerusalem, 459
40 Minifundia, 450 Newman, D., 598
41 Ministerial Order on Meat Inspection, New Opportunities for Research
42 326 Funding Co-operation in Europe
Ministry of Integration, 373 network (NORFACE), 13
43

index 623
New Orleans, 595 ÖzüekrenA. S., 598 1

New World, 11 2
New World Order post-1989, 119 P
New York/New York City, 7, 80, 81, 87, Pakistan, 26, 46, 47, 50, 97, 225, 555
3
89, 316, 595 Panafit, L., 363 4
New York Chinatown model, 600 Papastergiadis, N., 377 5
New Zealand, 302 Panjabis, 411 6
Nikolinakos, M., 490 Parachuted Suburban Model, 600
Noiriel, G., 10, 390 Pareto, W., 243 7
Nonini, D. M., 95, 98, 101 Paris, 29, 128, 316, 339-341, 345-346, 353 8
Non-Governmental Organizations Paris Commune, 342 9
(NGOs), 69, 142, 143, 145, 153-155, Park, R. E., 349, 582, 586, 587 10
157, 158, 375, 381 Parkin, F., 489-492
Non-resident Indians (NRIs), 97 Parmar, P., 500, 505 11
North Africa, 27, 49, 60, 216, 342, 346, Parsons, T., 75, 187, 438 12
522 Passel, J. S., 376 13
North America, 7, 104, 215, 217, 220, Pastore, F., 123 14
230, 289, 529 Pathan, 423
Northern Ireland, 151, 192 Pathan area, 420, 421, 425 15
Norfolk-Virginia Beach, 595 Pathan case, 422 16
Northern Norway, 430 Pathan local social systems, 411 17
Northwest Europe, 111, 552 Pathan values, 411 18
Norway, 25, 46, 223, 303 Pathan value standard, 423
Nozick, R., 291 Peach, C., 581-584, 593, 602 19
Pelagic islands, 410 20
O Penal Code, 171 21
Office of Population and Censuses Penninx, R., 9 22
Surveys (OPCS), 209, 210 Peters, L., 73
Ogden, P., 519 PEP study, 33 23
Ögelman, N., 234 Péraldi, M., 316 24
Oikos economy, 444 Père Lachaise cemetery, 346 25
Oil crisis of 1973, 9 Petersen, W. 185 26
Olsen, J., 141, 142, 144, 147 Petersen, W., 65
Olzak, S., 359 Petite, M., 149, 150 27
Omvedt, G., 484 Petras, E., 540, 554 28
O’Neill, O., 297 Pettigrew, T., 268 29
Ong, A., 95, 98, 101 Phalet, K., 385, 386, 395 30
Opel, 30 Phenomenology, 438
O’Reilly, K., 128 Philadelphia, 595 31
Organization for Economic Co- Philippines, 89, 97, 112 32
operation and Development Philpott, T., 589, 591, 592 33
(OECD), 130, 317 Phizacklea, A., 124, 394, 500, 503-506, 34
Organization for Security and Co- 509, 540, 554, 555
operation in Europe (OSCE), 149, Piore, M. J., 78 35
375 Pittsburgh, 595 36
OECD-SOPEMI report, 386 Platform Marokkaanse Jongeren 37
Oriol, M., 10, 14 Utrecht (PMJU), 366 38
Orthodox Marxism, 438 Plato, 275
Oswald, I., 316 Plender, R., 524, 525 39
Ottoman Empire, 353 Pohjola, A., 80 40
Oude Westen, Het, 352 Poland, 24, 47, 195, 538, 544, 552, 588, 41
Overseas Departments, 26 589 42
Oxford University, 102 Polanyi, K., 323
43

624 index
1 Poliakov, L., 570, 571 Renooy, P. H., 320

2 Policing the Crisis, 495, 496, 500 Republic of Ireland, 151, 518, 552
Polish Peasant in Europe and America Republikaner, 518
3 the, 70, 89 Resnick, S. A., 512
4 Pollack, M., 154 Revue européenne des migrations in-
5 Poly-ethnic social systems, 414 ternationales, 13
6 Portal, M., 146 Rex, J., 184, 191, 359, 360, 375, 437, 493,
Portes, A., 60, 65, 73, 92, 94, 98, 101, 496, 497, 506, 512
7 104, 115, 316, 320, 323, 328, 394, 395 Rhode, B., 121, 125
8 Portugal, 7, 27, 47, 263, 344, 393 Ribas Mateos, N., 124, 130
9 Portuguese Empire, 453 Richardson, J., 154
10 Positivist Empiricism, 438, 440 Roberts, B., 320, 323
Potsdam, 544 Robins, K., 96
11 Poulantzas, N., 492-494 Robinson, C., 498
12 Poutsma, E., 318, 319, 325 Rodbertus, K., 444
13 Powell, E., 35 Rokkan, S., 186
14 Praag, C., 519 Roland, C. 342
Prezworski, A., 493, 499, 521 Roman Catholic Church, 250
15 Pries, L., 79, 115 Roman Empire, 441, 453, 457, 460, 462
16 Principles of Scientific Management, Roman Republic, 445, 457
17 574 Rome, 116, 128, 247, 250
18 Protestantism, 447, 449 Romeo and Juliet, 249
Proudhon, P.J. 577 Rose, E. J. B., 187, 375
19 Prussia, 221 Rosenstein, C., 317
20 Pugliese, E., 132 Ross, G., 141, 157
21 Rotterdam, 315, 318, 322, 326, 352
22 Q Rousseau, J. J., 279, 293
Quebec, 227 Rudder de, A., 345
23 Rue de Belleville, 347, 348
24 R Rueschemeyer, D., 523
25 Race, Articulation and Societies Ruggie, J., 147
26 Structured in Dominance, 496 Ruhr, 29, 221
Race and Politics Group of CCCS, 495, Rush Portuguesa decision 27 March
27 498 1990, 156
28 Race Relations Act, 205 Ruzza, C., 153
29 Racial and cultural minorities, 184 Rwanda, 297
30 Racism and xenophobia, 260
Racism, Modernity and Identity, 259 S
31 Ramadan, 346 Sabbath, 347
32 Rastafarianism, 477 Safa, H. I., 538
33 Rath, J., 7, 9, 315, 316, 319-324, 326, 332, Safran, W., 93
34 366, 385, 395 Sahlins, M. D., 75
Ratnesarr, R., 122 Salt, J., 122, 125, 131, 537, 548-550
35 Raulin, A., 345, 347, 353 San Fransisco, 600
36 Ravenstein, E. G., 61, 111 San Fransisco-Oakland, 595
37 Réberioux, M., 570 Sanguinetti, A., 364
38 Reconquista, 571 Sargent, T. J., 491
Reich, M., 490 Sassen, A., 319
39 Reich, W., 570 Sassen, S., 538
40 Relative autonomy model, 494 Sassen-Koob, S., 316, 320, 541
41 Relative Deprivation theory, 83 Satzewich, V., 537
42 Renaissance, 260 Savona, E. U., 94
Renault, 30 Sayad, A., 10-12, 165, 344
43

index 625
Sayer, D., 520 Skocpol, T., 508 1

Segal, S., 464 SLG (Starting Line Group), 158 2
Sensenbrenner, J., 320 Slum and the Getto, The, 589
Schechtman, J. B., 544, 545 Smith, J., 99
3
Schengen, 143, 144, 149-151, 158, 159 Smith, M. G., 191-194, 457 4
Schengen agreement 1985, 146, 149, Smith, M. P., 94, 101, 394 5
152 Smith, R. C., 97, 101 6
Schengen Implementation Agreement Social Democratic parties, 465
(SIA), 146, 147 Social-Economic Council, 330
7
Schengen Information System, 145, 147 Social system the, 438 8
Schengenland, 116, 119 Sociology beyond Societies, 131 9
Schermerhorn, R. A., 196, 200, 239, Sociology of Migration or Race 10
473 Relations A, 504
Schierup, C.-U., 380, 383 Sociological Theories: Race and
11
Schmidtt, C., 281 Colonialism, 437 12
Schnapper, D., 379, 381 Socrates exchanges, 126 13
Schoenenberg, A., 360 Soininen, M., 376 14
Schönpflug, U., 392 Solidarité Arabe, 362
Schuck, P., 216, 229, 230 Solomos, J., 489, 499, 542, 543, 552, 553
15
Schumpeter, J., 282, 292 Sommerfelt, A., 432 16
Schuster, J., 321 SOPEMI, 130, 386 17
Schutz, A., 438 South Africa/Southern Africa, 184, 185, 18
Schwarzenbach Initiative, 35 432, 452, 454, 459, 461, 463, 484,
Schwarzenbach Referendum, 518 490, 497, 511, 529, 559
19
Scotland, 224, 273 Southeast Asia, 103, 190, 353, 415 20
Scotson, J. L., 343, 351 South-eastern Europe, 545 21
Scott, J. C., 71 Southern Europe, 23, 25, 26, 50, 60, 22
Seattle, 157 120, 124, 125, 216, 226
Second World War/World War II, 24, Southern Italy, 87
23
26, 68, 195, 249, 254, 262, 263, Southern Pathan, 411 24
284, 517, 559 Southern States, 442 25
Segal, A., 116 Southwest Asia, 415 26
Senegal, 112 Soviet Union, 116, 120, 217, 387, 538
Seville, 159 Soysal, Y., 373, 380, 397
27
Seville 2002 summit, 142, 159 Space TV Systems, 96 28
Shah, N. M., 82 Spain, 7, 24, 27, 47, 125, 128, 376, 393, 29
Sheffer, G., 93 263, 387, 571 30
Shell, 125 Spare Rib Collective, 483
Shohat, E., 96 Sparkbrook, 360
31
Shuttleworth, I., 121 Spencer, S., 554 32
Sicily, 27 Spivak, G., 96 33
Sieyès, A., 221 Sri Lanka, 522, 538, 543 34
Silj, A., 124 Stam, R., 96
Simmel, G., 439 Stares, P. B., 94
35
Simon, P., 316, 339, 342, 344, 388 Stark, O., 64 36
Simpson, G. E., 184, 190, 197 Starting Line Group (SLG), 154 37
Simpson Senator, 233 Statham, P., 392 38
Sinization, 421 Statistisches Bundesamt, 60
Sivanandan, A., 490, 492, 498, 509, Stein, J., 131
39
554 Stepick, A., 316 40
Siverts, H., 419 Stillwell, J., 133 41
Sjaastad, L. A., 111 St. Louis, 595 42
Sklair, L., 96, 97 Stone, J., 200
43

626 index
1 Stone, D., 144 Tillaart van den, H., 318, 319, 325
2 Structuralism, 438, 520 Tillie, J., 330, 395
Structural-Functionalism, 438 Tilly, C., 66, 88
3 Structure of social action the, 438 Time on the Cross, 451
4 Sudan, 421 Todaro, M. P., 111
5 Sudanic belt, 424 Todd, E., 379
6 Supreme Court, 233 Tölölyan, K., 93
Surinam, 190, 192 Tonnies, F., 439
7 Swart, A. H., 525 Toronto, 597, 598
8 Swat, 411 Torpey, J., 374, 377
9 Swedberg, R., 323 Toubon, J. C., 353
10 Sweden, 25, 46, 47, 50, 51, 57, 215, 216, Touraine, A., 264
219, 221, 223, 224, 302, 303, 376, Trade Commission, 156, 157
11 382, 524 TUC (Trades Union Congress), 37
12 Swiss unions, 37 Trading Association of Butchers, 327-
13 Switzerland, 24-29, 31-35, 38, 46, 49, 51, 330, 332
14 57, 149, 518, 519 Training on Commercial Practice, 329
Swyngedouw, M., 385, 386 Tränhardt, D., 130
15 Symbolic Interactionism, 438 Transnational Communities
16 Szanton-Blanc, C., 92, 94, 100 Programme, 100, 104
17 Transnational corporations (TNCs),
18 T 96, 103
Taeuber, A., 593, 594 Transnational Household Strategies,
19 Taeuber, K. E., 593, 594 103
20 Taguieff, P. A., 268 Transnational Religious Communities,
21 Taj Mahal, 202 104
22 Tamas, K., 59 Transnational Social Movement
Tampa-St. Petersburg, 595 Organizations (TSMOs), 99
23 Tampere 1999 summit, 142 Trevi group 1970, 146
24 Tannahill, J., 553 Tribalat, M., 386, 391
25 Tap, L. J., 326-328 Trieste, 126
26 Tapia, C., 344 Trinidad, 89
Targeted Social and Economic Research Tripartite Conference, 544
27 (TSER), 381 Tunderman, B., 329
28 Tarrius, A., 316, 320 Tunisia, 344, 345
29 Taylor, F.W. 574 Turk, A., 146
30 Teune, H., 521 Turkey, 27, 46, 47, 50, 82, 89, 145, 152,
Texture, 364, 365 195, 199, 224, 326, 522
31 Theory of Power, 237 Tyson, A., 155
32 Theory of the Third World revolution,
33 464 U
34 Thesis on Feuerbach, 439 Ueda, R., 302
Thistlethwaite, F., 68 Uganda, 544
35 Third Reich, 268 Ugur, E., 148
36 Third World, 23, 277, 461, 462, 464, Ukraine, 552
37 465, 484, 485, 538 UN Commission on Crime Prevention,
38 Thirld World feminist debate, 471 149
Thomas, B., 70 UNESCO (United Nations Educational,
39 Thomas, W., 70, 71, 75, 87, 89 Scientific and Cultural
40 Thomas-Hope, E. M., 554 Organisation), 437
41 Thompson, B., 344 UNHCR (United Nations High
42 Thompson, D., 122 Commissioner for Refugees), 69
Three generational model, 587, 596 UNICE (Union of Industrial and
43

index 627
Employers’ Confederations of Wakeman, F. E., 94 1

Europe), 157 Waldinger, R., 320 2
Unified Jewish Social Fund, 344 Waldrauch, H., 382
United Kingdom (UK), 9, 11, 89, 91, Wales, 224
3
114, 126, 127, 151, 203, 215, 225, 259, Wallerstein, I., 266 4
273, 550 Wallonia, 249 5
United Nations (UN), 28, 122, 284, 545 Walton, J., 65 6
United Nations Convention on the Walzer, M., 234, 301, 309
Reduction of Statelessness, 287 War of Independence, 225
7
United States (US/USA), 7, 11, 14, 15, Warnes, T., 129 8
71, 78, 79, 86, 89, 100, 117, 157, Warren, B., 538 9
192, 193, 202, 215, 216, 220, 221, Washington, D. C., 595 10
226, 227, 233, 301, 302, 324, 376, Waterman, S., 598
377, 389, 390, 395, 442, 459, 465, Waters, M.C. 597
11
490, 522, 538, 544, 546, 592, 596, Weber, M., 241, 437-450, 453, 455, 457, 12
600-602 461-466 13
United States Black Power Groups, 201 Weberianism, 438 14
United States Department of Defense Weesep, J. van., 598
(DoD), 94 Weil, P., 375, 380
15
United States Immigration and Wentholt, R., 321 16
Naturalization Service, 230 Werbner, P., 395 17
Universal Declaration of Human Rights West German employer’s federation, 18
1948, 287 556
University of Utrecht, 367 West German recession of 1966-67, 28
19
Unterhalter, E., 484 West Germany, 25, 27, 28, 47, 215, 222, 20
Urry, J., 126, 131, 377 223, 233, 538, 544, 545 21
Utrecht, 315, 318, 326, 355, 364-368 West Indies, 26, 47, 50, 555 22
West Midlands, 29
V Western Europe/West Europe, 8, 21, 23-
23
Valenci, L., 353 26, 30-32, 34, 37-40, 48-50, 53, 122, 24
Valk, I. van der, 366 237, 282, 297, 371, 374, 517, 518, 25
Vancouver, 597, 600 521, 522, 524, 538-543, 547, 559, 561 26
Veenman, J., 386 White, P., 598
Verbunt, G., 517, 519, 521, 533 White, P. E., 96, 111
27
Verein fur Sozialpolitik, 441 White Paper on Commonwealth 28
Vermeulen, H., 381 Immigration of 1965, 543 29
Verschueren, J., 376 Wickens, E., 124 30
Vertovec, S., 91, 93, 373 Wieviorka, M., 259, 260, 262, 264,
Vienna, 270 267, 269
31
Vienna Circle, 438 Williams, A. M., 128, 129 32
Vienna Club, 149 Williams, E., 570 33
Vienna Group, 149 Williamson, J. G., 68 34
Vietnam, 224, 485, 522, 538, 543 Williams, P., 94
Visco, I., 132 Wilpert, C., 360
35
Vlaams Blok, 254, 270, 518 Wilson, E., 483 36
Volkskrant de, 326-328 Wilson, R., 95 37
Volkswagen, 30 Wilson, W. J., 268, 594 38
Voskamp, J., 327 Winn, N., 150
Winnipeg, 597
39
W Wirth, L., 183-185, 189, 190, 196, 197 40
Wacquant, L., 602 Withol de Wenden, C., 356, 357, 525 41
Waffen SS, 155 Wolff, R. D., 512 42
Wagley, C., 185, 186 Wolpe, H., 490, 511
43

628 index
1 Wolverhampton, 502
2 Women’s Committee of the Greater
London Council, 483
3 Wood, M. E., 511
4 Woods, R. I., 111
5 World Bank, 97
6 Wrench, J., 383
Wright, E. O., 492, 493, 508
7
8 Y
9 Yearbook of International
10 Organizations, 99
Yinger, J. M., 184, 190, 197
11 Young, J. 495
12 Yugoslavia, 27, 47, 50, 116, 344, 387
13 Yuval-Davis, N., 392, 469, 481
14
Z
15 Zee TV, 96
16 Zhou, M., 320
17 Zinn, D. L., 123
18 Zlotnik, H., 60
Znaniecki, F. W., 70, 71, 75, 87, 89
19 Zolberg, A. R., 68, 543, 558
20 Zontini, E., 125
21
22 Other
9/11, 8, 390
23 3/11, 8
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

1
2
Other IMISCOE titles 3
4
5
6
7
8
9
10
IMISCOE Research 11
12
Rinus Penninx, Maria Berger, Karen Kraal, Eds. 13
The Dynamics of International Migration and Settlement in Europe: 14
A State of the Art 15
2006 (ISBN 978 90 5356 866 8) 16
(originally appearing in IMISCOE Joint Studies) 17
18
Leo Lucassen, David Feldman, Jochen Oltmer, Eds. 19
Paths of Integration: Migrants in Western Europe (1880-2004) 20
2006 (ISBN 978 90 5356 883 5) 21
22
Rainer Bauböck, Eva Ersbøll, Kees Groenendijk, Harald Waldrauch, Eds. 23
Acquisition and Loss of Nationality: Policies and Trends in 15 European 24
Countries, Volume 1: Comparative Analyses 25
2006 (ISBN 978 90 5356 920 7) 26
27
Rainer Bauböck, Eva Ersbøll, Kees Groenendijk, Harald Waldrauch, Eds. 28
Acquisition and Loss of Nationality: Policies and Trends in 15 European 29
Countries, Volume 2: Country Analyses 30
2006 (ISBN 978 90 5356 921 4) 31
32
Rainer Bauböck, Bernhard Perchinig, Wiebke Sievers, Eds. 33
Citizenship Policies in the New Europe 34
2007 (ISBN 978 90 5356 922 1) 35
36
Veit Bader 37
Secularism or Democracy? Associational Governance of Religious 38
Diversity 39
2007 (ISBN 978 90 5356 999 3) 40
41
42
43

1 Holger Kolb, Henrik Egbert, Eds.
2 Migrants and Markets: Perspectives from Economics and the Other
3 Social Sciences
4 2008 (ISNB 978 90 5356 684 8)
5
6 Ralph Grillo, Ed.
7 The Family in Question: Immigrant and Ethnic Minorities in
8 Multicultural Europe
9 2008 (ISBN 978 90 5356 869 9)
10
11 Corrado Bonifazi, Marek Okólski, Jeannette Schoorl, Patrick Simon, Eds.
12 International Migration in Europe: New Trends and New Methods of
13 Analysis
14 2008 (ISBN 978 90 5356 894 1)
15
16 Maurice Crul, Liesbeth Heering, Eds.
17 The Position of the Turkish and Moroccan Second Generation in
18 Amsterdam and Rotterdam: The TIES Study in the Netherlands
19 2008 (ISBN 978 90 8964 061 1)
20
21 Marlou Schrover, Joanne van der Leun, Leo Lucassen,
22 Chris Quispel, Eds.
23 Illegal Migration and Gender in a Global and Historical Perspective
24 2008 (ISBN 978 90 8964 047 5)
25
26 Gianluca P. Parolin
27 Citizenship in the Arab World: Kin, Religion and Nation-State
28 2009 (ISBN 978 90 8964 045 1)
29
30 Rainer Bauböck, Bernhard Perchinig, Wiebke Sievers, Eds.
31 Citizenship Policies in the New Europe: Expanded and Updated Edition
32 2009 (ISBN 978 90 8964 108 3)
33
34 Richard Black, Godfried Engbersen, Marek Okólski, Cristina Pantîru, Eds.
35 A Continent Moving West? EU Enlargement and Labour Migration from
36 Central and Eastern Europe
37 2010 (ISBN 978 90 8964 156 4)
38
39 Charles Westin, José Bastos, Janine Dahinden, Pedro Góis, Eds.
40 Identity Processes and Dynamics in Multi-Ethnic Europe
41 2010 (ISBN 978 90 8964 046 8)
42
43

Rainer Bauböck, Thomas Faist, Eds. 1
Diaspora and Transnationalism: Concepts, Theories and Methods 2
2010 (ISBN 978 90 8964 238 7) 3
4
Raivo Vetik, Jelena Helemäe, Eds. 5
The Russian Second Generation in Tallinn and Kohtla-Järve: The TIES 6
Study in Estonia 7
2010 (ISBN 978 90 8964 250 9) 8
2006 (ISBN 978 90 5356 878 1) 9
10
11
IMISCOE Reports 12
13
Rainer Bauböck, Ed. 14
Migration and Citizenship: Legal Status, Rights and Political 15
Participation 16
2006 (ISBN 978 90 5356 888 0) 17
18
Michael Jandl, Ed. 19
Innovative Concepts for Alternative Migration Policies: 20
Ten Innovative Approaches to the Challenges of Migration in the 21st 21
Century 22
2007 (ISBN 978 90 5356 990 0) 23
24
Jeroen Doomernik, Michael Jandl, Eds. 25
Modes of Migration Regulation and Control in Europe 26
2008 (ISBN 978 90 5356 689 3) 27
28
Michael Jandl, Christina Hollomey, Sandra Gendera, Anna Stepien, 29
Veronika Bilger 30
Migration and Irregular Work In Austria: A Case Study of the Structure 31
and Dynamics of Irregular Foreign Employment in Europe at the 32
Beginning of the 21st Century 33
2008 (ISBN 978 90 8964 053 6) 34
35
Heinz Fassmann, Ursula Reeger, Wiebke Sievers, Eds. 36
Statistics and Reality: Concepts and Measurements of Migration in 37
Europe 38
2009 (ISBN 978 90 8964 052 9) 39
40
Karen Kraal, Judith Roosblad, John Wrench, Eds. 41
Equal Opportunities and Ethnic Inequality in European Labour Markets 42
Discrimination, Gender and Policies of Diversity 43
2009 (ISBN 978 90 8964 126 7)

1 Maren Borkert, Tiziana Caponio, Eds.
2 The Local Dimension of Migration Policymaking
3 2010 (ISBN 978 90 8964 232 5)
4
5
6 IMISCOE Dissertations
7
8 Panos Arion Hatziprokopiou
9 Globalisation, Migration and Socio-Economic Change in Contemporary
10 Greece: Processes of Social Incorporation of Balkan Immigrants in
11 Thessaloniki
12 2006 (ISBN 978 90 5356 873 6)
13
14 Floris Vermeulen
15 The Immigrant Organising Process: Turkish Organisations in
16 Amsterdam and Berlin and Surinamese Organisations in Amsterdam,
17 1960-2000
18 2006 (ISBN 978 90 5356 875 0)
19
20 Anastasia Christou
21 Narratives of Place, Culture and Identity: Second-Generation
22 Greek-Americans Return ‘Home’
23 2006 (ISBN 978 90 5356 878 1)
24
25 Katja Ru š inovi ć
26 Dynamic Entrepreneurship: First and Second-Generation Immigrant
27 Entrepreneurs in Dutch Cities
28 2006 (ISBN 978 90 5356 972 6)
29
30 Ilse van Liempt
31 Navigating Borders: Inside Perspectives on the Process of Human
32 Smuggling into the Netherlands
33 2007 (ISBN 978 90 5356 930 6)
34
35 Myriam Cherti
36 Paradoxes of Social Capital: A Multi-Generational Study of Moroccans
37 in London
38 2008 (ISBN 978 90 5356 032 7)
39
40 Marc Helbling
41 Practising Citizenship and Heterogeneous Nationhood: Naturalisations
42 in Swiss Municipalities
43 2008 (ISBN 978 90 8964 034 5)

Jérôme Jamin 1
L’imaginaire du complot: Discours d’extrême droite en France et 2
aux Etats-Unis 3
2009 (ISBN 978 90 8964 048 2) 4
5
Inge Van Nieuwenhuyze 6
Getting by in Europe’s Urban Labour Markets: Senegambian Migrants’ 7
Strategies for Survival, Documentation and Mobility 8
2009 (ISBN 978 90 8964 050 5) 9
10
Nayla Moukarbel 11
Sri Lankan Housemaids in Lebanon: A Case of ‘Symbolic Violence’ and 12
‘Every Day Forms of Resistance’ 13
2009 (ISBN 978 90 8964 051 2) 14
15
John Davies 16
‘My Name Is Not Natasha’: How Albanian Women in France Use 17
Trafficking to Overcome Social Exclusion (1998-2001) 18
2009 (ISBN 978 90 5356 707 4) 19
20
Dennis Broeders 21
Breaking Down Anonymity: Digital Surveillance of Irregular Migrants 22
in Germany and the Netherlands 23
2009 (ISBN 978 90 8964 159 5) 24
25
Arjen Leerkes 26
Illegal Residence and Public Safety in the Netherlands 27
2009 (ISBN 978 90 8964 049 9) 28
29
Jennifer Leigh McGarrigle 30
Understanding Processes of Ethnic Concentration and Dispersal: 31
South Asian Residential Preferences in Glasgow 32
2009 (ISBN 978 90 5356 671 8) 33
34
João Sardinha 35
Immigrant Associations, Integration and Identity: Angolan, Brazilian 36
and Eastern European Communities in Portugal 37
2009 (ISBN 978 90 8964 036 9) 38
39
Elaine Bauer 40
The Creolisation of London Kinship: Mixed African-Caribbean and 41
White British Extended Families, 1950-2003 42
2010 (ISBN 978 90 8964 235 6) 43

1 Nahikari Irastorza
2 Born Entrepreneurs? Immigrant Self-Employment in Spain
3 2010 (ISBN 978 90 8964 243 1)
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

Selected Studies In: International Migration and Immigrant Incorporation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Selected Studies In: International Migration and Immigrant Incorporation

Uploaded by

Copyright:

Available Formats

imiscoe

martiniello & rath (eds.)

Selected Studies in International Migration and Immigrant Incorporation

“A collection of must-read, though sometimes hard-to-find, pieces that any scholar or

isbn 978 90 8964 160 1

amsterdam university press · www.aup.nl Amsterdam University Press

migration en ethnic deel 1.indd 1 04-03-10 15:56

migration en ethnic deel 1.indd 2 04-03-10 15:56

migration en ethnic deel 1.indd 3 04-03-10 15:56

migration en ethnic deel 1.indd 4 04-03-10 15:56

migration en ethnic deel 1.indd 5 04-03-10 15:56

1 14 Robert Kloosterman, Joanne van der Leun and Jan Rath

migration en ethnic deel 1.indd 6 04-03-10 15:56

migration en ethnic deel 1.indd 7 04-03-10 15:56

1 is amalgamated – have in recent years emerged as inevitable issues

migration en ethnic deel 1.indd 8 04-03-10 15:56

migration en ethnic deel 1.indd 9 04-03-10 15:56

1 not represent any exception. Sociologists have actually started cat-

migration en ethnic deel 1.indd 10 04-03-10 15:56

migration en ethnic deel 1.indd 11 04-03-10 15:56

1 quantitative nature – to policies already agreed upon. In other words,

migration en ethnic deel 1.indd 12 04-03-10 15:56

of academic prestige, migration and ethnic studies is still too often in 1

migration en ethnic deel 1.indd 13 04-03-10 15:56

1 ments developed in other social and national contexts. As observed

migration en ethnic deel 1.indd 14 04-03-10 15:56

These two problems of the theoretical and conceptual imports, es- 1

migration en ethnic deel 1.indd 15 04-03-10 15:56

migration en ethnic deel 1.indd 16 04-03-10 15:56

migration en ethnic deel 1.indd 17 04-03-10 15:56

1 Lloyd, C. (1995), ‘International comparison in the field of ethnic rela-

migration en ethnic deel 1.indd 18 04-03-10 15:56

migration en ethnic deel 1.indd 19 04-03-10 15:56

migration en ethnic deel 1.indd 20 04-03-10 15:56

migration en ethnic deel 1.indd 21 04-03-10 15:56

migration en ethnic deel 1.indd 22 04-03-10 15:56

capitalists through domination of the world market and imperialist 1

migration en ethnic deel 1.indd 23 04-03-10 15:56

1 addition to this economic function, the employment of immigrant

migration en ethnic deel 1.indd 24 04-03-10 15:56

migration en ethnic deel 1.indd 25 04-03-10 15:56

migration en ethnic deel 1.indd 26 04-03-10 15:56

ment. ‘Rotation’ of the foreign labour force was encouraged. In this 1

migration en ethnic deel 1.indd 27 04-03-10 15:56

1 labour supply, this measure contradicted the economic interests of

migration en ethnic deel 1.indd 28 04-03-10 15:56

figures do not show adequately the structural importance of immi- 1

migration en ethnic deel 1.indd 29 04-03-10 15:56

1 area which used to be regarded as the strongholds of the class-con-

migration en ethnic deel 1.indd 30 04-03-10 15:56

migration en ethnic deel 1.indd 31 04-03-10 15:56

1 ship to the means of production than that traditionally characteristic

migration en ethnic deel 1.indd 32 04-03-10 15:56

migration en ethnic deel 1.indd 33 04-03-10 15:56

1 political activity of foreigners can be forbidden if ‘important interests

migration en ethnic deel 1.indd 34 04-03-10 15:56

migration en ethnic deel 1.indd 35 04-03-10 15:56

1 ment through individual achievement and competition, rather than

migration en ethnic deel 1.indd 36 04-03-10 15:56

migration en ethnic deel 1.indd 37 04-03-10 15:56

migration en ethnic deel 1.indd 38 04-03-10 15:56

and having social, cultural and political functions. These organiza- 1

migration en ethnic deel 1.indd 39 04-03-10 15:56

1 Figure 7.1. Three levels of migration analysis