Transnationalization in international migration: implications for the study of citizenship and culture

Thomas Faist Abstract

Terms such as transnational social spaces, transnational social elds or transnationalism usually refer to sustained ties of persons, networks and organizations across the borders across multiple nation-states, ranging from little to highly institutionalized forms. However, there are two large conceptual gaps in the study of transnational social spaces arising out of international migration and refugee ows. First, terms such as transnational social spaces and transnational communities are often used synonymously, as if transnational community were the only form or type of transnational social space. This analysis outlines the primary mechanisms operative in transnationalization: reciprocity in small groups, exchange in circuits and solidarity in communities. These mechanisms correspond to distinct types of transnational social spaces transnational kinship groups, transnational circuits and transnational communities. Second, the implications of transnationalization for citizenship and culture have not been systematically explored. The concept of border-crossing expansion of social ties also helps to enrich our understanding of immigrant integration in the political and cultural realms. There is an elective af nity between the three broad concepts to explain and describe immigrant adaptation: assimilation, ethnic pluralism and bordercrossing expansion of social space, on the one hand, and the concepts used to describe citizenship and culture, on the other hand. In the political realm the concepts are national, multicultural and transnational citizenship; and in the cultural sphere, acculturation, cultural retention and transnational syncretism. Keywords: Transnationalism; international migration; assimilation; ethnic pluralism; diaspora.

Whether we talk of transnational social spaces, transnational social elds, transnationalism or transnational social formations in international migration systems, we usually refer to sustained ties of persons, networks and organizations across the borders across multiple nation-states, ranging from little to highly institutionalized forms. We do not mean
Ethnic and Racial Studies Volume 23 Number 2 March 2000 pp. 189222 2000 Taylor & Francis Ltd ISSN 0141-9870 print/ISSN 1466-4356 online

190 Thomas Faist occasional and eeting contacts between migrants and relatively immobile people in the countries of immigration and the countries of emigration. Transnational social spaces and the other names we have given these phenomena are characterized by a high density of interstitial ties on informal or formal, that is to say, institutional levels. However, when we look at the basic implications drawn from this common denominator, it is really astounding that there are still two large conceptual gaps among others in the study of transnational social spaces arising out of international migration and refugee ows. First, terms such as transnational social spaces and transnational communities are often used synonymously, as if transnational community were the only form or type of transnational social space (Portes 1996). Certainly, remittances owing in transnational families between rstgeneration migrants within reciprocally-organized households is a phenomenon quite different from centuries old diaspora communities that span part of the globe such as the Jewish diaspora. In turn, these two forms also probably differ from the transnational circuits of exchange in which Chinese, Lebanese or Indian businesspeople have ourished in various parts of the globe for decades. Moreover, the sustained transnationalization of migrant ties is often called transnationalism (for a recent example, see Portes, Guarnizo and Landolt 1999): It is not clear, however, whether it is the ideology of transmigrants who engage in local resistances of the informal economy and grass-roots activism (Smith and Guarnizo 1998), or the conscious-tainting efforts of rulers who try to hold on to expatriates (Basch et al. 1994), or the worldview of researchers who investigate the associated phenomena (Glick Schiller and Fouron 1998), or all three. We therefore need to conceptualize more clearly the different types of phenomena subsumed under the heading of transnational social spaces. So my rst intention is to provide a step towards a systematic typology of transnational social spaces arising out of international South-North migration. In addition, I identify several factors conducive to the formation of durable transnational social spaces that exist beyond the rst generation of migrants. Second, since concepts termed transnational have become catch-all phrases for sustained border-crossing ties, it is not surprising to nd claims that transnationalization has far-reaching consequences in how we think about immigrant adaptation, global civil society, communities, culture and citizenship. For example, some authors have perceptively argued that transnational ties may work against melting into the majority core predicted by classical assimilation theories thus leading to more cultural autonomy and even cultural hybrid identities (see, for example, Clifford 1994). Nevertheless, there has been no systematic conceptualization of different forms of immigrant adaptation, comparing the transnational concept with older ones. My second goal therefore is to show the usefulness of a concept of border-crossing expansion of social space

Citizenship and culture

191

as a third and somewhat apocryphal gospel in addition to the canonical theories of immigrant assimilation and ethnic pluralism which have overemphasized the container aspects of politics and culture. While economic aspects of transnationalization have received a great deal of attention both in the rst wave of transnationalist studies on multinational companies in the early 1970s (Keohane and Nye 1977) and in the initial phase of interest in small business and later migrant transnationalism (see, for example, Harvey 1989, pp. 14759), the implications for concepts such as citizenship and culture are always alluded to but never spelt out in comparison with their competitors. The concept of transnational social spaces In this part I advance two propositions: (1) The concept of transnational spaces covers diverse phenomena such as transnational small groups, transnational circuits and transnational communities. Each of these is characterized by a primary mechanism of integration: reciprocity in small groups, exchange in circuits and solidarity in communities. (2) Factors conducive to the formation of transnational social spaces not only include favourable technological variables, troubled nationstate formation and contentious minority policies in the developing world, and restrictions such as socio-economic discrimination. Instead, political opportunities such as multicultural rights may also advance border-crossing webs of ties. Towards a de nition and a typology of transnational social spaces Transnational social spaces are combinations of ties, positions in networks and organizations, and networks of organizations that reach across the borders of multiple states. These spaces denote dynamic social processes, not static notions of ties and positions. Cultural, political and economic processes in transnational social spaces involve the accumulation, use and effects of various sorts of capital, their volume and convertibility: economic capital, human capital, such as educational credentials, skills and know-how, and social capital, mainly resources inherent in or transmitted through social and symbolic ties. The reality of transnational social spaces indicates, rst, that migration and remigration may not be de nite, irrevocable and irreversible decisions transnational lives in themselves may become a strategy of survival and betterment. Also, transnational webs include relatively immobile persons and collectives. Second, even those migrants and refugees who have settled for a considerable time outside their country of origin, frequently entertain strong transnational links. Third, these links can be of

192 Thomas Faist a more informal nature, such as intra-household or family ties, or they can be institutionalized, such as political parties entertaining branches in various countries of immigration and emigration. The transnational social spaces inhabited by immigrants and refugees and immobile residents in both countries thus supplement the international space of sovereign nation-states. Transnational social spaces are constituted by the various forms of resources or capital of spatially mobile and immobile persons, on the one hand, and the regulations imposed by nation-states and various other opportunities and constraints, on the other; for example, state-controlled immigration and refugee policies, and institutions in ethnic communities. Transnational social spaces are delimited by pentatonic relationships between the government of the immigration state, civil society organizations in the country of immigration, the rulers of the country of emigration (sometimes viewed as an external homeland), civil society groups in the emigration state, and the transnational group migrants and/or refugee groups, or national, religious and ethnic minorities. For example, quite a few countries of emigration have recently reacted to transnationalization and the wishes of immigrants for border-crossing recognition in adapting their citizenship rules and allowing for dual citizenship, and trying to sustain the ow of remittances and create investment avenues for citizens and their children from abroad. This, among other things, has forced the government and public in the immigration states to consider whether or not to tolerate dual citizenship. There is a marked difference between the concepts of globalization and transnational social spaces, that is to say, transnationalization: transnationalization overlaps globalization but typically has a more limited purview. Whereas global processes are largely decentred from speci c nation-state territories and take place in a world context above and below states, transnational processes are anchored in and span two or more nation-states, involving actors from the spheres of both state and civil society. Also, transnationalization differs from denationalization. The latter term has denoted the fact that the stateless and many minorities (in post-World War I Europe) had no recourse to governments to represent and protect them (Arendt 1973, p. 269). There are three types of resources within social and symbolic ties that allow individuals to cooperate in networks, groups and organizations. They also serve to connect individuals to networks and organizations through af liations. Technically speaking, they lower transaction costs, the expenses monitoring and sanctioning obligations and contracts. We can differentiate the following forms of transactions: (1) Social exchange in the form of mutual obligations and expectations of the actors, associated with speci c social ties and based on exchanges and services rendered in the past (Coleman 1990, pp. 3069).

Citizenship and culture

193

These obligations and expectations can be an outcome of instrumental activity, for example, the tit-for-tat principle. Often, reciprocity is involved. It is reciprocity as exchange. (2) Reciprocity as a social norm: what one party receives from the other requires some return (Gouldner 1960, p. 160). (3) Solidarity with others in a group who share similar positions such as kinship or local community membership, or who can be reached only through symbolic bonds, for example, membership in otherwise anonymous national collectives. It is an expressive form of social transaction. The most important form of solidarity is collective representations (Durkheim 1965, p. 471). These are shared ideas, beliefs, evaluations and symbols. Collective representations can be expressed in some sort of collective identity, we-feeling or weconsciousness, and refers to a social unit of action. In its ideal-typical form these are cultural communities, such as families, ethnic groups, religious parishes, congregations, communities and nations. Solidarity can also be institutionalized: citizenship, for example, is an institutionalized form of ties between a citizen and a state; often in short supply among migrant newcomers. There are three main bene ts to be derived from resources inherent in transactions. In general, it helps members of networks or groups to get access to more economic, human and social capital. This crucially depends on the number of persons in a network or collective who are prepared or obliged to help one when called upon to do so, that is, the number of social and symbolic ties available (Bourdieu 1983, pp. 1905). Information is also a bene t. In general, the information bene ts of a large, diverse network tend to be greater than the information bene ts of a small and socially homogeneous network. Moreover, the higher the stock of resources in ties, the more control can be exerted in monitoring and sanctioning other actors. The basic idea is that the extent matters to which any particular person (or collective) is an important link in the indirect ties to others in controlling the ow of information, authority, power and other resources. While the bene ts of capital and information potentially accrue to all participants involved, control is usually only available and bene cial for those who hold or are close to positions of authority, whether legitimately in kinship groups, communities and organizations in such states, or illegitimately in Ma a-style organizations. Transactions based on exchange, the norm of reciprocity and solidarity, have desirable and undesirable effects. On the one hand, the mechanisms and bene ts of social capital allow cooperation; on the other, they can restrict the degrees of freedom of individuals involved in signi cant ways. While the norm of reciprocity tends to enhance cooperation, it can also lead to revenge and retaliation. Take the case of two immigrant groups who have carried con icts abroad, for example among

194 Thomas Faist some groups of Algerians in France, who not only hold differing views on the role of religion within secular states but have also imported violent con icts into the new country of settlement. Solidarity may not only help to pool energies among kinship members when building a business, it can also encourage envy and sti e entrepreneurship when the pro ts are constantly split instead of being used to establish funds for reinvestment. Resources inherent in social and symbolic ties have two important characteristics: First, it is dif cult to transfer them from one country to another: they are primarily local assets. This is less true of symbolic than it is of social ties, because the former do not depend on face-to-face or indirect contact through other persons. Among local assets we may nd diverse ties such as attachment to kin and friends, a language that is familiar, communities such as a church congregation that offers spiritual nourishment, and an ethnic group or a nation with a distinct cultural-ideological outlook. These assets remain local unless forced into motion not only by macrostructural factors such as (international) labour recruitment or civil wars, but also by the evolution and presence of mechanisms inherent in the manifold ties connecting potential movers, stayers and larger communities and organizations. Thus, in addition to political regulations of international migration, this is one of the main causes for the relatively low, though increasing, rates of international mobility (Faist 2000: chs 4 and 5). However, if transnational networks and chain migration emerge in the course of international migration, the transferability of ties carrying obligations, reciprocity, solidarity, information and control increases. Second, these various resources are crucial mechanisms for applying other forms of capital. They provide transmission belts that bridge collectives and networks in distinct and separate nation-states. Resources inherent in social and symbolic ties are necessary for mobilizing other forms of capital, especially among those short of economic capital. And often, immigrants need social ties to established immigrants or brokers to nd work. When transnational social spaces emerge out of migratory ows, even the return to the country of origin may not be permanent, since many older migrants temporarily migrate again in the opposite direction in order to secure their medical needs in the countries in which they once worked and where some of their children or other kin still live. These forms of recurrent migration would not be possible without intrakinship obligations and reciprocity. Thus, mechanisms such as reciprocity and solidarity are crucial in the formation of a circular ow of goods and persons between countries, and ful l contingent bridging functions. What needs to be described is the type of transnational social spaces, ranging from reciprocal ties within kinship systems, to exchange relationships among business persons and transnational communities. There are at least three forms of transnational social spaces that need to be distinguished: transnational reciprocity in small groups (usually kinship

Citizenship and culture

195

collectives), transnational exchange in circuits; and solidarity within transnational communities (see Figure 1). While we can nd all forms of social capital in all kinds of transnational social spaces, each type of space is characterized by a dominant mechanism of integration: reciprocity in small groups, exchange in circuits, solidarity in communities. Reciprocity in transnational kinship groups is typical of many rstgeneration labour migrants and refugees. Reciprocity can be seen, for example, in remitters sending back money to members of their kinship group in the country of origin; especially where territorial exit is part of a strategy which includes economic survival or betterment for migrants and for those who stay behind migration as a sort of informal risk insurance. In those cases the migrants remit money to those who run domestic affairs in the country of origin. Often, seasonal, recurrent and eventual return migration is part of the strategy. This mechanism usually operates only until family reuni cation or members of the rst generation have all died.
Figure 1. Three types of transnational social spaces arising from international migration and ight
Types of transnational Primary resources social spaces in ties Transnational kinship groups Reciprocity: what one party receives from the other requires some return Main characteristic Typical examples

Upholding the social norm of equivalence

Remittances of household or family members from country of immigration to country of emigration: e.g., contract workers Trading networks, e.g., Chinese, Lebanese and Indian business people

Transnational circuits

Exchange: mutual obligations and expectations of the actors; outcome of instrumental activity (e.g. the tit-for-tat principle) Solidarity: shared ideas, beliefs, evaluations and symbols; expressed in some sort of collective identity

Exploitation of insider advantages: language; strong and weak social ties in peer networks

Transnational communities

Mobilization of collective representations within (abstract) symbolic ties: religion, nationality, ethnicity

Diasporas: e.g., Jews, Armenians, Palestinians, Kurds; frontier regions: e.g., Mexico-US; Mediterranean

196 Thomas Faist Transnational circuits are characterized by a constant circulation of goods, people, and information transversing the borders of emigration and immigration states (Rouse 1991) along the principle of exchange, that is to say, instrumental reciprocity. Economic entrepreneurs often make use of insider advantages knowing the language, having friends and acquaintances abroad to establish a foothold. They typically develop in a context in which we often nd successful socio-economic adaptation to conditions in the receiving country, or successful reintegration in the emigration country. For an extreme case of circularity, take the hypermobile Chinese businessmen in North America. These astronauts establish a business, say, in Singapore, but locate their families in Los Angeles, New York or Toronto to maximize the educational opportunities for their children or as a safe haven in the event of political unrest. The astronauts constantly move between two places (Cohen 1997, p. 93). Other entrepreneurs and their dependants are rmly rooted in either the country of emigration, or of immigration, or yet another country, and use it as a base from which to carry out entrepreneurial activities elsewhere. Transnational communities characterize situations in which international movers and stayers are connected by dense and strong social and symbolic ties over time and across space to patterns of networks and circuits in two countries based upon solidarity. Communities [that is, Gemeinschaft] encompasses all forms of relationship which are characterized by a high degree of personal intimacy, emotional depth, moral commitment, social cohesion and continuity in time (Nisbet 1966, p. 47). For transnational communities to emerge, solidary ties need to reach beyond narrow kinship systems. Such communities without propinquity, in which community and spatial proximity are partially decoupled, do not necessarily require individual persons living in two worlds simultaneously or between cultures in a total global village of deterritorialized space. What is required, however, is that communities without propinquity link through reciprocity and solidarity to achieve a high degree of social cohesion, and a common repertoire of symbolic and collective representations. Transnational communities can emerge on different levels of aggregation. The most fundamental and widespread are village communities in emigration and immigration countries that connect through extensive forms of solidarity over longer periods of time (see, for example, Engelbrektsson 1978). Frequently investment of those abroad or of returnees in private and public projects exempli es this kind of support. Transnational communities can also consist of larger aggregates, primarily held together by symbolic ties of common ethnicity or even nationhood. For example, refugees such as Kurds from Turkey who have pursued nationbuilding or political opposition projects in their home countries typically try to develop and entertain dense transnational ties.

Citizenship and culture

197

Diasporas tend to constitute a speci c type of transnational community. History abounds with examples of diasporas. The Jewish experience is usually the rst to come to mind as a prototype for diaspora formation; and could be extended to include African Americans, Armenians and Palestinians. In diasporas, a group has suffered some kind of traumatic event which leads to the dispersal of its members, and there is a vision and memory of a lost or an imagined homeland still to be established, often accompanied by a refusal of the receiving society to recognize in full the cultural distinctiveness of the immigrants. Diasporas frequently include a complete cross-section of community members who are dispersed in many diverse regions of the world (Safran 1991). It is inappropriate to apply the term diaspora to settlers and labour migrants because they have not undergone traumatic experiences, nor can it be said that most of the members of these groups yearn to return to their lost homeland. This type of transnational community could evoke solidarities that may be inconsistent, sometimes even contradicting the allegiances demanded by the territorial nation-states involved. This is most often the case where diasporas are connected to nationalist projects. Especially in cases of war between the countries of emigration and immigration, the charge of dual loyalty and disloyalty has arisen (Sheffer 1986, p. 8). Diasporas can only be called transnational communities, if the members also develop some signi cant social and symbolic ties to the receiving country. If they do not, we can speak of exile. For instance, the political exile is a person who, after persecution and ight, yearns to return to his home country. Some temporary labour migrants clearly intending to return home can also be regarded as exiles. Exile communities are single-mindedly drawn to the former homeland; although the intentions, especially among labour migrants, may change. This goal is of such overriding importance that no substantial ties to the new and supposedly temporary country of settlement can develop. To prosper, diasporas do not necessarily need concrete social ties. It is possible that the memory of a homeland manifests itself primarily in symbolic ties. This has been the case for the Jewish diaspora for centuries after the destruction of the Second Temple. More than a thousand years later, some authors have characterized the relationship of diasporic Jews with those in Israel as mishpachah, literally meaning family (Jacobson 1995, p. 236). The difference between diasporas and other forms of transnational communities becomes clear when we compare the Jewish diaspora before the establishment of the state of Israel with global communities such as Chinese entrepreneurs and traders in many countries of Southeast Asia, Africa and the Americas. Jews experienced dispersal in a traumatic fashion, and at a time when assimilation seemed inevitable in many European countries the horrors of the Holocaust rekindled their

198 Thomas Faist consciousness of kind. By contrast, many Chinese went abroad as settlers during the nineteenth century, and later experienced xenophobia. The vision of the Chinese was, at rst, much less oriented towards the ancestral homeland, and lacked components of exile. It was only later that Chinese in Southeast Asia became united as a result of discrimination. In addition, they gained increasing awareness of unity through the revolution against Manchu rule and the resistance of the Japanese invasion of their homeland. The much later rise of nationalism throughout Southeast Asia and the attacks against their economic position by the longerestablished ethnic groups further intensi ed their collective identity of being Chinese abroad. Another distinct form of transnational communities are groups with collective identities in frontier regions. In the South-North context we can think of groups and networks in spaces characterized but not delimited by contiguous nation-state borders. Around and along these borders regular and sustained interstitial transactions arise. Immigration need not be the key factor driving the development of frontier regions. For example, people on either side of the frontier usually occupy a contiguous geographical space which is interlinked by those who commute to work on the other side of the border. Prominent examples are the US-Mexican frontier region, with nowadays intensi ed economic exchange under the umbrella of NAFTA and the Western Mediterranean zone of the Iberian Peninsula, France and Italy, on the one hand, and North African states such as Morocco, Tunisia and Algeria, on the other. While we do not yet see a particular and unique collective identity of frontier peoples in the North American or the Euro-Mediterranean region, both border spaces carry the potential for sustained transactions going beyond economic links to political cooperation and cultural commonalties. If the more than thousand-year-old experiences of the frontier regions of the former Holy Roman Empire offer any guide to the future think of the now threecountry Oberrhein region (Alsace of France, Baden of Germany, and the Basel region of Switzerland) , it is that the common history gains new momentum through efforts at supranational economic integration. Factors contributing to the formation of transnational social spaces The technological breakthrough in long-distance communication and travel which occurred in the nineteenth century may have accelerated the emergence of transnational social spaces. New and improved methods of communication and travel, such as trans-oceanic steamship passages and telegraphic communication, set the necessary but not suf cient stage for the development of transnational ties. Since then the ongoing communication and transport revolution has considerably cut the cost of bridging long geographical distances, and this has sharply accelerated since World War II.

Citizenship and culture

199

We then have to distinguish processes of transnationalization in the economic sphere, on the one hand, and in the political and cultural realm, on the other. For economic transnational spaces to develop, transnational networks of businesspeople plus bene cial conditions to invest economic capital in the emigration country, such as lower production costs, may suf ce. Although individual transnational entrepreneurs obviously bene t from social and symbolic ties between emigration and immigration countries for example, ties through friends and kinship systems , economic activities do not need to be strongly embedded in these systems over extended periods of time through solidarity. Exchange- and reciprocity-based resources are suf cient. This situation is quite different from the formation of transnational communities built around political or religious projects that last beyond the rst generation of migrants. Here, the main catalysts are, rst, strong ties of migrants and refugees to the country of origin and the country of immigration over an extended period of time. Social ties and symbolic ties need to ourish social connections, language, religion and cultural norms. Second, these ties and corresponding resources are not only embedded in migration ows but in other linkages as well, such as trade and mass communications. Third, juridical and political regulations, such as domestic and international regimes, may allow to varying degrees for the movement of people and tolerate or repress political and religious activities of immigrants and refugees in either countries of emigration or immigration. In other words, transnational communities must be embedded in larger political and economic international structures. A necessary prerequisite for international migration to occur in the rst place are prior exchanges in the economic (for example, foreign investments), political (for example, military cooperation or domination), or cultural (for example, colonial education systems) dimensions. This is why activities in transnational social spaces do not create such transnational linkages ex nihilo, but usually evolve within pre-existing linkages, building new ones, and challenging existing arrangements such as citizenship and notions of acculturation. We now need to specify some of the ceteris paribus conditions within the countries of immigration and emigration. First, the factor most conducive to transnationalization of politics and culture in the emigration countries has been contentious minority politics relating to ethnicity and religion, often associated with the building of edgling nation-states. These emigration country con icts sometimes tend to be exported to the countries of immigration. Examples abound, ranging from Indian Sikhs in Great Britain, Canada and the United States (Tatla 1999) to Kurds in Germany, The Netherlands, Belgium and Sweden (Falk 1998). Second, in the country of immigration, serious obstacles to socioeconomic integration and/or a denial of acculturation or cultural recognition are extremely conducive to the transnationalization of political

200 Thomas Faist and cultural activities. The two dif culties, economic and cultural, may go hand in hand, or may proceed separately. For example, some groups may be denied opportunities for cultural assimilation or recognition while they are well integrated socio-economically. This used to be true for Chinese in the white settler colonies until the 1940s. In other cases, partial socio-economic exclusion and a perception on the part of substantial groups among the newcomers that their cultural recognition is blocked, can go hand in hand, as the examples of some labour migrant groups in Western Europe suggest; for example, Surinamese in The Netherlands or Caribbeans in the United Kingdom. Third, if the countries of immigration are liberal democracies that do not assimilate immigrants by force, immigrant minorities have a greater chance to uphold cultural distinctiveness and ties to the countries of origin. The import of symbols and the propensity for community formation around multicultural claims is more likely, the more liberal or tolerant the political regime. In particular, multicultural policies of the countries of settlement are conducive to upholding immigrants transnational ties: for example, people from the Caribbeans in the United Kingdom (Goulbourne 1991). Put differently, not only repressive policies and discrimination advance immigrant transnationalization. On the contrary, opportunities to exercise multicultural rights and a liberal political environment can also further transnational activities and a border-crossing collective consciousness. Political and cultural transnationalization As I have suggested, international migration is not a discrete event constituted by a permanent move from one nation-state to another. Rather, it is a multi-dimensional economic, political, cultural and demographic process that encapsulates various links between two or more settings and manifold ties of movers and stayers between them. Transnational social spaces emerge. In some cases these ties and the unfolding transnational social spaces even extend beyond the rst generation phenomena such as diasporas have attested to this for centuries. Here, I wish to make two further propositions: (1) The trajectories of immigrant adaptation envisaged by the canonical concepts of assimilation and ethnic pluralism theories hold in certain cases. Other phenomena, such as continuing transnational ties and linkages, need to be categorized in a new and separate conceptual niche. Assimilation and ethnic pluralism are insuf cient because they espouse a container concept of space adaptation of immigrants within nation-states is considered to be a process not signi cantly in uenced by border-crossing transactions. However, since growing transnationalization contributes to the plurality of avenues open to

Citizenship and culture

201

labour migrants and refugees in various nation-states, but is nevertheless always tied to speci c places, the concept of border-crossing expansion of space enriches our understanding of adaptation. (2) There is an elective af nity between the three broad concepts to explain and describe immigrant adaptation: assimilation, ethnic pluralism and border-crossing expansion of social space, on the one hand, and the concepts used to describe citizenship and culture, on the other. In the political realm the concepts are: national, multicultural and transnational citizenship, and in the culture sphere: acculturation, cultural retention and transnational syncretism. Treated as ideal-typical concepts each of the notions captures an important part of immigrant membership (see Figure 2). Political transnationalization: national, multicultural and transnational citizenship Citizenship has been historically closely linked with the evolution of nation-states. Modern states are based on the congruity of territory that is de ned by borders, recognized by neighbouring states and other members of the international system of states. A further characteristic is state sovereignty, which means the priority over all other political institutions ensconced within the demarcated territory. Within this purview,
Figure 2. Three concepts for the analysis of immigrant adaptation in the receiving countries
Approach Assimilation Ethnic pluralism Border-crossing expansion of social space

Realms of adaptation Main prediction Political Melting into the core Pluralization national citizenship: unitary national political culture multicultural citizenship: common elements of political culture include recognition of cultural differences cultural retention: practices maintained in a new context; collective identities transplanted from emigration country Transnationalization dual state membership: elements of political culture from various states can be complementary

Cultural

acculturation: full-scale adaptation of values and behaviour to the nation-states core

transnational syncretism: diffusion of culture and emergence of new types mixed identities

202 Thomas Faist residents have citizenship status when they are accepted with all rights and duties. Some critics have seen the institutionalization of transnational ties in dual state membership as a grave challenge. For them the political issue is this: The more transnational or multifocal ties immigrants entertain, the greater their ambivalence towards the receiving polity; the weaker the roots in the nation-state of settlement, the stronger the incentives to form a transnational community; the bolder the claim to a diaspora, the greater the tendency on the part of natives to question the allegiance of the newcomers, and, nally, the weaker the inclination of immigrants to adapt to the immigration country. In short, dual state membership hinders immigrant adaptation in the country of immigration, encourages populism on the part of the majority groups, and leads to divided loyalties among immigrants. Other critics have argued that dual state membership reduces nationality to holding a passport and thus devalues citizenship (for many, see Isensee 1974). These criticisms warrant a closer look at the nature of dual state membership, compared to other concepts of citizenship. Dual state membership comes in two forms. The rst is dual citizenship. A person holds passports of two nation-states and has full rights and duties in both, although one citizenship is usually resting. Only the citizenship of the actual country of residence is operative. The second is dual nationality. Dual nationality is different from dual citizenship in that the rights under the former are more restricted than under the latter. For example, holders of Declaration of Mexican Nationality IDs are not able to vote or hold political of ce in Mexico, or to serve in the Mexican Armed Forces. The bene ts of Mexican nationality include the right to buy and sell land free of the restrictions imposed on aliens and to receive better treatment under investment and inheritance laws in Mexico, to attend public schools and universities as Mexicans, and to access other Mexican government services and jobs. Other major countries of immigration also changed their laws to allow for dual citizenship, and increased the rights of expatriates (Cebecioglu 1995). Pressured by the German government, the Turkish side had to relinquish granting dual citizenship and introduced a sort of dual nationality with the so-called pink card in late 1997. The dimensions of citizenship Citizenship in a state is an institutionalized form of solidarity. It constitutes an expression of full and formal membership. Citizenship forms a continuing series of reciprocal transactions, between a citizen and a state; whether or not we derive it from a contract between state and citizen (Hobbes 1962), or between citizens who are authors of their constitutions (Rousseau 1973 and Kant 1984). States and citizens can claim a set of

Citizenship and culture

203

mutually enforceable rights and duties. Citizenship also connotes the public representation of ties between members and corresponding nation-states. It is based on the perception of common belonging to a state or a nation or both and it confers the identity citizen. Citizenship is in short supply among virtually all newcomers to a polity. Access of newcomers such as migrants to citizenship can be analysed along two axes, the sort of membership vertical: legal-constitutional and politicalinstitutional; and the dimension of citizenship horizontal: reciprocal state-citizen ties and the public recognition of these ties (belonging) (Figure 3). In the rst cell (I), we can use a threefold analytical distinction to delineate the various degrees of rights from entry into the territory and few rights (aliens) to permanent residency (denizens) and full membership status with associated rights and duties (citizens). The question raised is to what kind of rights do distinct categories of immigrants and refugees have access? And on what basis? In the second cell (II) the main question is: what kind of institutions and policies are necessary to implement and enforce the rights and duties that correspond to the legal status? And what kind of reciprocal duty can states expect from members that can be differentiated into the categories of aliens, denizens and citizens? Cells three and four (III & IV) are most interesting when we look at the transnationalization of immigrant life. In the third cell, we face the question: under what conditions should newcomers be allowed to naturalize, to acquire the corresponding nationality in the country in which they reside? And, of particular interest here: are full members of a nation-state allowed to hold more than one nationality, perhaps dual or triple nationality? While national citizenship envisages an assimilation of immigrants to a unitary political culture, transnational citizenship trusts the compatibility of citizens loyalties to multiple states. In the fourth cell the question arises: what kind of status should be recognized
Figure 3. Citizenship: dimensions and realm of membership
Dimension of citizenship Reciprocal state-citizen ties Realm of membership legal-constitutional rights and duties of citizens (I) political-institutional access to rights and political participation; enforcing duties (II) nationality/supra-nationality/ dual nationality (III) recognition of identities other than nation: religion, language, cultural customs and practices, etc. (IV) Public recognition of ties

204 Thomas Faist

religious, cultural, ethnic? Multicultural citizenship operates like national citizenship in the realm of a unitary state. However, it demands

the recognition of culture for nations, ethnic, religious and other groups living in this state. The questions in cell III and IV are related in the following manner. Can societal groups claim a certain kind of legal status such as dual nationality or dual citizenship because this is conducive to uphold a certain way of life in transnational social spaces? We now turn rst to the elective af nities between the concepts of immigrant adaptation with distinct models of citizenship. Each of the three concepts of immigrant adaptation corresponds to distinct, albeit at times overlapping, understandings of citizenship: assimilation to a unitary political culture in a single nation-state, ethnic pluralism as the recognition of distinct cultures to multicultural citizenship, and border-crossing expansion of social space as enriching individual and collective identities to dual citizenship and dual nationality. These three forms of formal citizenship relate somewhat differently to the nation-states congruity assumption of one people to one territory and one cultural space. In addition, forms of political membership in and between nation-states are to be observed, such as participation in homeland-oriented associations, or truly transnational organizations that connect countries of origin with several host countries; nowadays exempli ed by Muslim organizations in Europe, Australia and North America, or diaspora communities striving to establish nation-states, carved out of their former home country. Assimilation and national citizenship Assimilation is akin to the idea of rather exclusive citizenship in a single nation-state, national citizenship. Assimilation theory sees a gradual adaptation of immigrants not only socio-economically but also culturally and behaviourally. Gradually immigrants do away with the cultural baggage transported from the emigration country. As immigrants continue to embark upon the member-ships of the perhaps multiple rivers and riverains of the receiving country, the logical end point is single nation-state citizenship, characterized by a dominant and unitary political cultural core. It does not really matter here that assimilation theorists have not envisaged the political realm explicitly. What is important is that they have not seen the necessity to consider the implications of transnational ties beyond the rst generation, with exceptions pertaining mainly to folkloric expressions. The best known and still most prevalent pattern of full membership acquisition by newcomers is insertion into the citizenship of a single nation-state which is the state of settlement. States regulate access to a single citizenship by various procedures. Two obvious juxtapositional modes are ius sanguinis vs. ius soli (legal notions) and ius imperium vs.

Citizenship and culture

205

ius republica (not legal notions). First, in some receiving countries, the empire had a national as well as a global reference. Take Britain and The Netherlands. All subjects were theoretically free to travel to any other part of the empire. As a consequence of this ius imperium, in the early periods of large-scale immigration after World War II, these colonial powers admitted considerable numbers of immigrants, nominally with equal rights as the domestic citizens. This stood in marked contrast to ius republica, where in countries such as the US, naturalization proceeded on the basis of the constitution. Needless to say, many in the latter countries found themselves outside the charmed circle of the republics citizenship, as African Americans until the 1960s attest. A second and venerable distinction ranges from the ideal types of ius soli (territory principle: citizenship accorded upon birth in a country, independent of the parents citizenship) combined with ius sanguinis (blood principle: citizenship granted as a result of parents or ancestral citizenship), on the one hand, to ius sanguinis without ius soli, on the other hand (de Rahm 1990). The US is probably the case with the strongest elements of ius soli, followed by countries such as France. Germany (until 1999), Greece and Italy have typi ed cases in which ius sanguinis rules supreme. The unique religious-national narrative in Israel, embodied in the Law of Return, represents the clearest case of this form of descent-based citizenship. In order to complement the picture of access to citizenship we need a third principle, not legal concepts such as ius sanguinis and ius soli, but nevertheless a useful distinction. Virtually all countries have regulations which we could call a sort of ius domicili, that allows denizens nally to acquire full citizenship. Based on social and symbolic ties immigrants have developed since their arrival and their economic contributions, ius domicili speci es the conditions that newcomers have to ful l when applying for citizenship, after having lived for periods of mostly two to eight years in the country of residence. Among the most common criteria for the admission of newcomers are uninterrupted residence and work history for some years, a regular income, suf cient living space, no criminal record and mastery of the dominant language in the nation-state. In recent years, some countries have altered their regulations somewhat, so that various of the dimensions just mentioned have begun to blur. For example, Germany has complemented her dominant ius sanguinis with a more liberal ius domicili since 1991 foreign citizens who have lived and gone to school in Germany may claim a German passport between the ages of 16 and 21 years. Elements of ius soli have also found entry: beginning in 2000 children whose parents have settled in Germany automatically receive German citizenship in addition to their parents nationality. By contrast, the 1981 British Nationality Act changed access to citizenship to being mainly a matter of descent. This Act abolished ius soli whereby individuals acquired citizenship simply by the fact of birth on British soil, and promulgated ius sanguinis.

206 Thomas Faist Yet there are some common elements which apply to all these cases. Liberal-democratic nation-states as welfare states, characterized by high degrees of regulation and redistribution which require institutionalized solidarity, organize their political order on the basis of an egalitarian and homogenized citizenship, including, among other things, free and universal suffrage and access to social rights. The rst to recognize this pattern, T.H. Marshall, ingeniously analysed membership in terms of rights accorded to members of a nation-state who share a sense of belonging. For him, citizenship connotes a bundle of rights and a few duties. The device of bundling entitlements is the territorially delimited state in which most permanent residents share a common nationality. In Marshalls words (c)itizenship requires . . . a direct sense of community membership based on loyalty to a civilization which is a common possession (Marshall 1964, p. 92). The notion that solidary policies of redistribution and regulation demand a common understanding of who is a member and who is eligible for rights and services carries a lot of weight. Migration then raises a distinct challenge to this state of affairs because the world economy tends to favour open exchange, while the notion of citizenship demands certain territorial and communal protections. This is most clearly the case when we look at those nation-states that are highly developed welfare states, in which citizens have something to lose. These states usually have stricter external and internal controls but also relatively easy access to rights once inside. A comparison of Sweden and the US bears out this claim. Researchers who have taken up Marshalls approach are mostly concerned with how immigration is changing the notion of citizenship in which rights, duties and a sense of belonging have hitherto been inextricably linked to some sort of common culture (cf. Brubaker 1992). Marshalls reference to some sort of common bonding and shared customs as a basis for the recognition of equal status for all members remains valid. The questions then are: how much of a common consensus is necessary in multi-ethnic and multi-religious polities and what are the decisive elements of such a commonality? What kind of rights and how many should be granted to non-citizens? It is clear that immigration tends to further the unbundling of rights tied to formal citizenship in liberal-democratic welfare states: Even non-citizens are entitled not only to civil but also to certain social rights in national welfare states; and in some countries, such as The Netherlands and Sweden, permanent residents who are non-citizens are even allowed to exercise political rights, such as voting in local elections. There is also an international extension of the national model, postnational citizenship (Soysal 1994). It is part of the academic musings that we are witnessing the dawning of a post-national era. It deals with international in uences upon single nation-state citizenship rights. Basically, it tries to show that human rights have indeed come closer to citizen

Citizenship and culture

207

rights. While it was still possible to claim in the late 1940s that the right to citizenship transcended the rights of citizens because the persons without statehood found themselves without any recourse to legal claims (see the cogent analysis by Arendt 1949), post-nationals assert that liberal-democratic nation-states have come increasingly to respect human rights of persons, irrespective of citizenship over the past few decades. What post-national citizenship suggests is that supranational institutions and discourses have an impact on nation-state citizenship. According to this view, the life-chances of immigrants in Western democracies have been primarily shaped not by belonging to a speci c national community and the corresponding citizenship. Rather, human, civil and social rights are governed by universal discourses, embedded in international agreements, consensus and nation-state constitutions. Human rights can be interpreted as a part of world-culture (Meyer et al. 1997). Migrants take advantage of opportunities created by the growing prevalence of human and civil rights discourse of a world polity to move around and settle. Unfortunately, there are virtually no supranational institutions conferring the status of citizenship except the EU in a weak, albeit not totally declamatory, but steadily expanding form. Yet, EU citizenship does not cover third-country citizens. Another obvious problem with this approach is that we do not know exactly how universal norms and discourses shape policy and practices on the ground. A competing and much simpler explanation holds that virtually all nation-states have enshrined civil rights in their constitution. And as we know, the inclusion of immigrants into welfare states does not mainly depend on nationhood, at least not in their legal form. Access to full social rights is primarily tied to residence and not to nationality or citizenship. Therefore, the observation that membership and life-chances of immigrants do not so much depend on full citizenship can easily be explained as a consequence of Rechtsstaat (civil rights) and welfare state principles when border-crossing and settlement of persons occur. The point made by the post-nationals is more accurate when related to wider notions of membership, not to citizenship in the nation-state. Membership in nation-state polities is less often tied to formal citizenship but to rights arising from settlement and socialization. And the granting of citizens rights is not coterminous with formal membership in the nation-state. Take the concept of denizenship as being located between alien status and full citizenship (Hammar 1990). Denizens are permanent residents who practically hold the full set of rights accorded to citizens, except for voting rights at the national level. It is the recognition of social ties and economic contributions of long-term resident aliens. In sum, the tie between denizens and the state is not as thick as that between citizens and the state, but not as thin as between aliens and the state. In essence, the triad citizen-denizen-alien re ects the old Greek

208 Thomas Faist distinction between politai citizen, katoikoi resident alien, now called denizen, and xenoi foreigner. Yet, to speak meaningfully about the membership of immigrants living in transnational social spaces, we have to go beyond nation-states and institutions of international society and allow for signi cant transnational ties. Ethnic pluralism and multicultural citizenship In essence, the proponents of multicultural citizenship have revived the notion of ethnic pluralism in the political sphere. Like the proponents of assimilation theory, they regard adaptation exclusively in the container space of nation-states. This is also true for a normative and liberal-democratic version of ethnic pluralism, usually referred to as multiculturalism. There are basically two types of multiculturalism, a passive and an active one. Passive multiculturalism means that immigrants and minorities can express their cultural difference in the private realm. However, the public realm is organized along principles of universalism and equal rights for all (Rex 1991). Here, only active multiculturalism is relevant: Active multiculturalists argue that, taking freedom and equality seriously as preconditions for participation in public life, individuals need to be assured of a secure cultural background. The supportive framework of cultural groups constitutes such a context of choice (Kymlicka 1995). Drawing normative and policy implications, the active multiculturalists then go on to postulate that this context can only be maintained by granting special rights to ethnic and religious groups; an assumption that engenders Herculean tasks to defend, both intellectually and policy-wise. Special rights deemed necessary range from rights to political autonomy for indigenous groups to comparatively uncontroversial assurances for religious practices. Proponents of multiculturalism propose cultural rights to accommodate the cultural identities and practices of immigrant groups. These rights are meant to keep alive and strengthen the intragroup social and symbolic ties. Among these rights are the following: (1) voting rights for permanent immigrant residents; (2) af rmative action programmes that aim at increasing the representation of visible minorities in major educational and economic institutions; (3) revised work schedules to accommodate the religious holidays of immigrant groups; (4) bilingual education programmes for the children of immigrants, so that their earliest years of education are conducted partly in their mother tongue, as a transitional phase to secondary and post-secondary education in the dominant receiving country language; and (5) minority group schools such as Muslim schools. In this view, reciprocity and solidarity provide the basis for collective identities that foster common and publicly declared narratives. What is noteworthy about all these conceptions of citizenship so far, the Marshallian and the multicultural types, is that they clearly point to the central role of nation-state institutions in the process of conferring

Citizenship and culture

209

membership status. After all, only nation-states can grant formal and institutional status and ultimately secure human rights. However, there is a serious shortcoming. Both canonical conceptualizations treat immigrant adaptation exclusively in the realm of a nation-state devoid of signi cant transnational ties that the people in the respective places and spaces entertain. There is no room for meaningful transnational ties criss-crossing nation-state borders which in uence the daily lives of immigrants. Border-crossing expansion of social space and dual state membership Activities in transnational social spaces suggest that membership is multilayered. The natural equivalent to political activities and identities spanning nation-state borders in the formal political sphere is dual state membership. This form of transnational citizenship does not deny the existence or relevance of borders and nation-states. It simply recognizes the increasing possibility of membership in two states. Dual state membership refers to the fact of being a citizen in two states; less fullyedged forms could mean being a citizen in one state and a settled immigrant with a sort of denizenship status in another. At a minimum, it tolerates immigrants close ties with the country of emigration. The features of dual state membership become clear when we introduce the emigration country governments interests, attitudes and policies towards their expatriates. All emigration countries are interested in economic bene ts such as remittances and investments, and in political control of emigrants abroad. Two patterns emerge. Some governments have used their expatriates abroad as a captive group to exercise maximum control. For example, the mainland Chinese government has insisted upon ius sanguinis for its citizens in South East Asia since the 1950s. This meant moving from overseas Chinese to Chinese living overseas (Nonini and Ong 1997, p. 9). The Peoples Republic of China made a statement about controlling her citizens abroad. After all, competition has existed between Taiwan and mainland China for overseas Chinese. But the Peoples Republic has gradually lost control. Quite contrary to this stance, other emigration country governments, for example, Mexico, have used the instrument of dual state membership to keep the ties of expatriates alive. The Mexican government has been using their immigrants as a support for conducting business at home and abroad (Smith 1999). In Europe, the Italian government opted for a homelandoriented approach for its guestworkers during the 1960s and 1970s in Germany, led by expectations of returning migrants who would reestablish themselves in the Italian South and contribute to the development of this economically backward region (Schmitter Heisler 1984). Given the asymmetric relationships between countries of emigration and immigration, the position of the latter proves decisive. If it allows dual citizenship or dual nationality, the emigration states usually allow it as well. This has been a seminal trend. Earlier in this century, a new

210 Thomas Faist US-American citizen forfeited her citizenship if she voted in foreign elections, or held public of ce in another nation-state. Nowadays, the laws have not changed but the US does not check upon dual state membership, and many countries of origin have rushed to allow it. Other countries such as the UK, France and The Netherlands have tolerated dual citizenship. Indeed, in many countries of settlement a signi cant proportion of newcomers who get naturalized currently keep their former citizenship as well. Around half of the worlds countries currently recognize dual citizenship or dual nationality (Traces No. 3, 1998). From the immigrants point of view, dual state membership constitutes a deliberate strategy to protect various rights in multiple states. This strategy has become more prominent, not least because many nation-states have liberalized their citizenship laws. Also, from an international point of view, this development nds support. While in 1963 European nation-states still overwhelmingly supported the Convention to avoid multiple nationality, a majority of European nation-states now support a new 1997 Convention that explicitly allows dual citizenship. Even in a country such as Germany that has not signed the Convention, there are about 2 million Germans with a second passport (Migration und Bevlkerung 1/1998, p. 2). Of course, this preliminary sketch cannot do justice to the important fact that the countries of immigration and emigration not only offer very different conditions for establishing dual citizenship but also for speci c citizenship rights accompanying transnational lives. Thus, states which are both immigration countries and advanced welfare states, differ importantly in opportunities for immigrants to partake in social rights. It is to be expected that more highly institutionalized welfare states such as Germany attract more people who transmigrate to participate in social rights than in more residual welfare states such as the United States. For example, there is a high proportion of older Turkish migrants in transnational kinship groups who commute between Turkey and Germany. Clearly, ties to family in both countries is one of the reasons. Another may be that these elderly migrants still have access to German health insurance services and do not want to lose them by remaining in Turkey for too long. Ironically, dual citizenship may not only increase the short-term mobility of older migrants but may also be more cost-ef cient for immigration welfare states because the migrants themselves have increased options to take care of themselves in places most appropriate to their living conditions. Moreover, dual state membership can facilitate economic cooperation in transnational circuits; for example, by easing investments. Cultural transnationalization: acculturation, cultural retention and the strengthening of transnationally-induced syncretism Symbolic ties in cultures transmit meaningful ways of life across the full range of human activities and, as such, include both public and private

Citizenship and culture

211

realms. As to the development of immigrant culture(s), we can discern three ideal-typical views: the acculturation thesis of the assimilation perspective, the proposition of cultural retention of ethnic pluralism; and the emergence of syncretist cultural practices and meanings, as suggested by the concept of border-crossing expansion of social space. The main problem of the strong versions of the acculturation and ethnic retention perspectives is that they espouse a container concept of culture. They do not pay suf cient attention to phenomena such as cultural diffusion and syncretism. Cultural cross-overs are most likely to ourish in transnational communities, such as village and regional communities Landsmannschaften in German and hem eri in Turkish diasporas and frontier regions. For assimilation theory, immigrant culture is a sort of baggage brought from the old world. It mainly considers adaptation of immigrants to core culture(s). Everything beyond folkloric expressions is considered a transitory phenomenon. At rst sight, cultural pluralism contradicts this view because it emphasizes cultural retention among immigrant groups. Nevertheless, this is also a rather bounded view of culture, since it does not pay attention to hybrid cultural practices and cultural syncretism. Akin to an acculturation perspective that gives analytical priority to cultural core(s), cultural pluralism focuses on cultural retention at the fringes, the margins. In other words, assimilation theory surveys the main river, whereas ethnic pluralism investigates the tributaries and channels. Without doubt, both perspectives have merit, have captured important trends in the past, and correspond to crucial aspects of present-day cultural adjustment of immigrants. We should also try to include the whole river valley with the main rivers and their tributaries to get a more complete picture (Conzen 1991). Nevertheless, we have to go one step further, since the cultural diversity around us evolving out of international migration and transnational social spaces has been increasing. At the root of these phenomena lies the mobility not only of persons but also of cultural practices, meanings and symbols (Hannerz 1996, p. 64). Hence the need to supplement the two canonical views to take into account diffusion and syncretism. The concept of border-crossing expansion of social space tries to capture how immigrant cultural syncretism connects to ongoing transnationalization. Acculturation and assimilation: cultural life in a nation-state container One prominent version of assimilation theory suggests that immigrant adjustment means melting of immigrants into the core culture. In the most sophisticated version (Gordon 1964) the process of assimilation starts with acculturation. In this view, acculturation is often, although not always, followed by structural assimilation, the entry of immigrants into primary groups of the immigration country. The last step concerns again

212 Thomas Faist the cultural realm, the identi cational assimilation and thus the individual and collective identities of immigrants. This indicates that cultural adaptation and meanings accompany the process of immigrant adjustment all along. The nal result is, more or less, overall cultural submergence. Some versions of assimilation theory make certain amendments to non-linear processes of adaptation. For example, ethnicity survives or is reinvented while ethnic cultures disappear. One often overlooked claim of assimilation perspectives is that persistent collective identity in the second generation does not altogether jettison the adaptive process. In essence, ethnicity expressed as collective identity can be preserved or invented for reasons that may have little to do with inherited culture (Gans 1979). Unlike language, which changes in a linear fashion the longer you stay, the better you tend to speak it collective self-identities vary signi cantly over time. Here, we are not confronted with linear developments but with reactive developments. Some research on ethnicity among immigrants suggests that it is best dealt with as an emergent category, which arises under conditions reinforcing the maintenance of kinship and friendship networks (Yancey et al. 1976). Transnationalization has spurred this trend: With the help of new media and interlocutors immigrants forge new symbolic ties to putative ancestors abroad, sometimes in countries other than the emigration and immigration states. For example, since the late 1980s, Hmong immigrants from Laos now in the US have discovered their roots in interior China in a people called Miao. Even though the languages of the two peoples are vastly diffeent the Hmong consider the Miao a pure version of their clouded past (Schein 1998). The Hmong have fostered new social and symbolic ties with the help of mechanisms such as long-distance travel and videos on the Chinese Miao marketed in the US. It is too early to say whether this exchange, aided by modern technologies, will result in a Miao Hmong transnational community. This last example already suggests patterns of immigrant acculturation that deviate from the main path envisaged by assimilation theory. Groups with a sense of being discriminated against, such as the Hmong mentioned above, may not turn to acculturate to the core culture but may look for examples abroad to nd their place. Or, inner-city groups characterized by low human capital, weak social cohesion, and poverty, such as new arrivals in the US from Haiti and the West Indies have much more in common with African-American cultural practices than with the so-called Anglo mainstream (Portes and Zhou 1994). Since assimilation theory assumes that immigrants discard their old countrys cultural baggage or dissolve it into the mainstream, it does not pay suf cient attention to cultural diffusion and syncretism. The view of culture in the acculturation perspective is one of tight boundedness. Although diffusion is possible, it is of minor analytical importance

Citizenship and culture

213

because the nation-state as a container for a societal culture acts as an assimilator for newcomers. However, we should be more careful. Growing transnationalization may change the rules of the game, even in the cultural sphere. There is some sparse evidence, that even groups such as second-generation German-Turks engage in transnational syncretism (Caglar 1995). Cultural pluralism: from the transfer and retention of culture to culture as a context of choice The early versions of cultural pluralism claimed that immigrants, after experiencing discrimination and (partial) rejection in the country of immigration, would turn back to their cultures of origin, those of the emigration countries. This process of reorientation, called dissimilation, would lead to distinct national cultures existing side by side in the immigration country. The initial insistence that immigrants are able to reconstruct autonomous cultural worlds as separate nationalities has yielded gradually to a more nuanced understanding of the selective function and character of adaptation (Bodnar 1985). Empirically, the softer versions of ethnic pluralism have seen immigrants becoming ethnics over time. This latter nding is consistent with assimilation theory which provides, as we have seen, for the fact that a groups collective identity can exist without a strong cohesion of cultural meanings as social bonds. In other words, symbolic ties may refer to collective identity without necessarily being part of a rather coherent system of practices and meanings of a whole way of life. Ethnic pluralism should not be confused with multiculturalism. Nevertheless, a normative version of multiculturalism based on liberalism seeks to justify rights undergirding minority cultures. In this view, culture is important because it constitutes, as mentioned before, a context of choice (Kymlicka 1995). Cultural traditions, symbols and practices allegedly form a cultural repertoire that enables minorities to participate in liberal democracies on a competitive basis. In order to derive rights for distinct categories of minorities, this perspective distinguishes national minorities and immigrant categories. National minorities do have a national culture that has to be supported by limited rights to self-government. But most immigrants do not, because they have chosen to come voluntarily. Even if their migration is involuntary, such as that of refugees, they are often oriented towards their emigration country. Nevertheless, immigrant groups should have rights, polyethnic rights, such as the right for Jews and Muslims to be exempt from Sunday closing laws or Sikhs allowed to wear a turban instead of a helmet when riding a motorcycle (Kymlicka 1995, p. 101). In sum, this extension of cultural pluralism says that national minorities do have a national and thus a complete societal culture, while immigrant minorities do not. Although the distinction proposed is fraught with many empirical and normative problems it cannot accommodate

214 Thomas Faist the many hard cases in between such as slaves and conquered minorities (for pertinent criticism, see Young 1997) it offers a sophisticated and graduated concept of multiculturalism. But even in this latter version of cultural pluralism two main points of criticism remain. It overemphasizes cultural retention among minorities, and underemphasizes the impact of transnationalization on immigrant cultural adjustment. Cultural retention usually goes hand in hand with adaptation of new elements. Assimilation theory has painted perhaps too strong a picture in that immigrants supposedly get rid of their cultural baggage. It is equally unlikely, however, that immigrant cultures develop without some modi cation unless rigid seclusion prevails (for example, Hutterites). Immigrant culture can thus never be identical with country of origin culture. Much evidence points to the thesis that cultural practices and meanings do not simply disappear quickly, reduced to folkloric functions. Going even further, because of eased transnational exchange of meanings across social and symbolic ties, there is nowadays a higher potential that old patterns are transferred and go into a synthesis with new ones. As in the Turkish-German example, Islamic organizations such as Milli Gr have gradually sought to re-adapt to German patterns. For example, not only do they attempt to be recognized as a religious organization with a special status, a quasi-public institution (Krperschaft des ffentlichen Rechts). They have also developed new ideas to reach second-generation Turks in Germany and thereby eased naturalization in addition to dual citizenship. Newer cultural pluralist approaches make room for the fact that the character of a (minority) culture can change as a result of its members choices. For instance, cultural diffusion can enrich the opportunities for expression of meanings. But, endemic in the conception of culture as a very bounded concept, if a culture is not a societal or national culture, it will be reduced to ever-increasing marginalization (Kymlicka 1995, p. 80). As we have seen, this strong assumption has to be questioned because of the ever-growing transnationalization of cultural repertoires. Beyond the container concept of culture: transnational syncretism Canonical assimilation and cultural pluralist views provide only a thin veneer allowing for cultural syncretism in order to achieve the desired nal results of acculturation or retention. These views give short shrift to the dynamic nature of all cultures. Both theoretical traditions have described plausible and long-term real-world outcomes in the past. But in the case of contemporary immigrants in the North, there are many phenomena, even if only transitory ones, that elude such neat categorizations. They disregard the syncretist practices, mixed languages and hyphenated collective identities.

Citizenship and culture

215

Overall, there is a surprising similarity between strong versions of assimilation and ethnic pluralism. They overemphasize culture as a xed and essential phenomenon; assimilation theory does so with core cultures and ethnic pluralism with minority cultures. This container concept sees culture as essentially territorial, based on a shared language and somewhat static. In this view culture stems from a learning process that is, in the main, tightly localized. This is culture in the sense of a culture; the culture of a social group. Moreover, they involve common institutions and practices. Such cultures are linked with processes of modernization, such as the build-up of educational systems in nation-states (Gellner 1983). In an extreme version, it imbues a hypostasized notion of places as bounded and unchanging spaces with a xed meaning, identi ed with rather strong communities (for a similar characterization, see Nederveen Pieterse 1994, pp. 1767). Clearly, the container concept of culture has to be widened, as tolerant assimilationists and ethnic pluralists have already suggested, although they have not brought in the effects of transnationalization suf ciently. The canonical concepts of immigrant adjustment have sensitized researchers and the public to issues of ethnic and national pluralism, while neglecting the comparable pluralism of space. This has not only been true for aspects of local culture but also for those of wider transnational diffusion. Therefore, an alternative conceptualization views culture as relating to elements of a more general human software; the tool kit version of culture applies here (Swidler 1986). This dynamic notion of culture has been implicit in theories of evolution and diffusion, in which culture is also viewed as a translocal or even a transnational learning process. Fluidity and not xity, spatiality and not locality mark this notion. Immigrant culture cannot be seen as baggage or a template, not as something to be guratively packed and unpacked, uprooted (assimilationists) and transplanted (cultural pluralists). Instead, an analytical approach looks for structures of meaning engendered by and expressed in private and public behaviours, images, institutions, languages (see also Geertz 1973, pp. 330). These structures of meaning are inherent in social and symbolic ties. Such ties and their content do not vanish or merge imperceptibly nor can they be retained easily under new circumstances, unless the transactions with surrounding groups cease. Certainly, the ongoing and spreading transnationalization of meanings and symbols through social and symbolic ties in transnational social spaces helps to keep up multifold transactions transversing borders. Under propitious conditions such as modern technologies (satellite or cable TV, instant mass communication, personal communication bridging long distances through telephone and fax, mass affordable short-term long-distance travel), liberal state policies (polyethnic rights and anti-discrimination policies), changing emigration state policies (reaching out to migrants living abroad for remittances, investment and political support), and

216 Thomas Faist immigrant capacities to mobilize resources (organizational, social and human capital) transnational syncretism of culture nds a fertile breeding ground. Eventually, these two very broad understandings of culture are compatible: for to nd expression, dynamic-syncretist culture de nitely needs territorial boundaries implicit in models of container culture. Immigrant cultures cannot exist in a deterritorialized space. Spatially hypermobile individuals are the exception rather than the rule. Nevertheless, many immigrants in transnational kinship groups who have a focus on one country and not an equally strong one on another (or others), entertain transnational links, be they social ties or symbolic ties. Strong common strands of culture are still necessary and possible within nation-states. Among other things, common national, that is to say, societal cultures are necessary because a highly educated and trained workforce constitutes a functional prerequisite of a modern economy. In addition, a kind of diffuse solidarity is essential for modern welfare states, and equality of opportunity for all residents depends on common understandings of legitimate principles of justice, rights and redistribution. In short, without a pervasive nation-state culture, even transnational communities would face no prospects for successful adjustment. These considerations not only apply to nation-state formation but also to the emergence of immigrant and ethnic communities across nationstate borders. Usually, community formation is the product of transferred capital, differential treatment and subsequent organization on the part of newcomers to overcome perceived disadvantages and discrimination or to exploit new opportunities. We should not forget that many perceptions of discrimination need a climate of toleration to be expressed in the public realm. Only in liberal democracies tolerant towards cultural difference does discrimination become an issue that leads to successful transnational syncretism and transnational political organization. No multiculturalism, no transnationalism. Although multicultural rights do not necessarily encourage an enduring transnationalization of migrant ties, they do advance the expansion of border-crossing spaces. For sustained transnationalization including whole communities the content of symbolic ties has to go beyond multicultural orientations focused on one nation-state. Syncretist content becomes important for forging and upholding transnationally-oriented networks and organizations. This not only means that container views of culture have to be modi ed, it also implies that the unrealistic image of a deterritorialized and global culture has to be cast aside. This view suggests that we are all migrants now. Even those who do not move spatially are bound to experience diffusion due to the migration of cultures (Waldron 1995). While globalized professionals, intellectuals, artists and entrepreneurs may cherish this image, it is utterly unrealistic for the majority of migrants and for those who are relatively immobile.

Citizenship and culture

217

To think of transnationally enriched syncretism as another layer of immigrants insertion processes in addition to acculturation and cultural retention is to use an understanding of culture as a whole way of (immigrant) lives, one that emphasizes their translocal aspects without occluding the fact that cultures are still overwhelmingly nationally bounded and have mainstreams. Even eighty years ago in the US, during the heyday of Americanization drives, a one-sided melting into the core of immigrants was unrealistic, as Randolph Bourne wisely remarked in 1916: No Americanization will ful ll this vision which does not recognize the uniqueness of this trans-nationalism of ours . . . America is coming to be, not a nationality but a trans-nationality, a weaving back and forth, with the other lands, of many threads of all sizes and colors (Bourne 1996, p. 107). Leaving aside the unrealistic image of immigrants as nations and America as a nationally pluralist state, the quotes main thrust, cultural pluralization in the wake of immigration, is relevant for todays world. Migrant and migration networks have brought forth the dual characteristic of migrant resources as both local assets and border-crossing transmission belts. Again, as in the political sphere, it should be clear that the concepts of acculturation, ethnic retention and transnational syncretism are ideal types. When we start applying them to real-world cases, we need to be aware of country-speci c differences affecting the cultural development of immigrant groups and the extremely complicated interactions between old and new immigrants, sometimes from the same country of origin. Let us consider Mexican immigrants in the US. In the 1980s and 1990s, for example, a developing we-consciousness among Mexican immigrants has been not only a by-product of immigration and, often, continuing transnational ties, but has been heavily in uenced by con icts between recent Mexican immigrants and settled Mexican-Americans with the latter group sometimes looking back at several generations of settlement in the US. While there have been few demands for multicultural rights on the part of Mexican immigrants, all of the categories have shown various mixtures of tendencies towards acculturation and transnational syncretism. Not surprisingly, acculturation generally is more pronounced among Mexican-Americans, and transnational syncretism, such as participation in country of origin festivals, could be observed among more recent Mexican immigrants in the US. Conclusion: transnational ties as shifting bridges We can conclude that the notion of a singular political or cultural trajectory envisaged by the canonical theories of assimilation and ethnic

218 Thomas Faist pluralism, and container concepts of immigrant adaptation have to be questioned. Since the factors conducive to the formation and maintenance of transnational social spaces eased technological means of communication, incomplete nation-state formation in many countries of emigration, discrimination and multiculturalism in the countries of immigration show no signs of abiding but rather of further spreading, the concept of border-crossing expansion of social space has become more important to grasp issues of transnational membership. Associated phenomena such as dual state membership and transnational syncretism are not located on a magic carpet of a deterritorialized space of ows. It only makes sense when rmly tied to speci c spaces in different nationstates. It is not a notion above nation-states but a combination of both the inside and the between. In other words, the national and transnational dimensions of dual state citizenship and syncretism are not like Russian dolls with no interlinks other than a level of analysis differences. To speak of the bridging function of transnational ties makes no sense without doors through which persons may both enter and leave. In contrast, the dominant conception of citizenship in political theory and membership in cultural studies has been rather archaic. For example, one widespread image draws on walls protecting the essence of political communities, a core of cherished practices, beliefs and rights. Yet, the true civilizational achievements are not walls, or windows for that matter, but doors. The doors delineate territorially bounded states and, in an emerging way, supra-national institutional structures such as the European Union. The match between bridges and territories is for ever shifting, as are the positions of the doors. The considerations presented so far relate to the normative aspects of membership: dual nationality and dual citizenship. Speci cally, how can legal statuses linked to dual state membership be morally justi ed? One strategy could parallel the ways to reason about multicultural citizenship (Gerdes 2000). The underlying thought in arguing for multicultural citizenship is that differentiated citizenship (Young 1989) is necessary to ensure recognition and, under certain circumstances, advance equality of opportunity. The concept of multicultural citizenship critiques universal and culture-blind conceptions of citizenship in saying that discriminated minorities will remain con ned to an inferior social, economic and political position unless they receive special rights which compensate for exclusion. The ultimate rationale is that culture constitutes a basis for recognition and that special rights are necessary to empower minority groups to partake in the full rights and duties of the polity. The proposal of multicultural citizenship has drawn a lot of valid criticism pertaining to issues such as intra-group democracy and whether a unitary group culture is really a necessary basis for societal and political participation. However, it is a rst step towards a fuller discussion of whether dual state membership is normatively desirable or not. This discussion can extend in three ways. First, we can think of dual

Citizenship and culture

219

state membership as re ecting the transnational ties of rst-generation immigrants. There is a great wealth of empirical evidence that they entertain border-crossing social and symbolic ties over their life span (for an introduction, see Gmelch 1980). As we de ned above, citizenship is a morally demanding and institutionalized form of solidarity. Ultimately, it is a highly regulated form of a key mechanism of social capital. The fundamental characteristics of social capital therefore apply. It is primarily a local asset and can have a border-crossing function only when transmission belts are available such as migrant networks or transnational organizations, such as human rights groups. As such, dual state membership is not likely to contribute to the export of con icts from the emigration to the immigration countries, nor does it in itself further processes of democratization in the emigration states. All these processes have occurred without multiple formal membership. What is crucial, are the transmission belts of interstitial networks and organizations. Second, from the point of view of the states involved, dual loyalties to different states and to transnational communities will usually not present a problem for the states involved, unless in situations of war. However, these are exceedingly rare between countries of emigration and immigration along the South-North axis. Generally, we have to differentiate between emigration and immigration states. The former have an interest in keeping ties to emigrants for the sake of remittances or investments. The latter, as already mentioned, have started, openly or tacitly, to tolerate multiple memberships in ever increasing numbers. Third, dual state membership directly impinges upon the public recognition of ties and not only upon state-citizen ties (see Figure 3). The majority of rst-generation immigrants who participate in transnational kinship groups, circles and communities, and who are interested in acquiring citizenship of the immigration country favour dual state membership, viewing it as a welcome recognition of their multiple attach ments (K l 1994, p. 75; en and Karakasoglu 1994). This is because issues central to transnationals such as inheritance laws can be changed accordingly without bestowing full citizenship. Instead, dual state citizenship pertains to the aspects of belonging and recognition. Its main purpose is to acknowledge the symbolic ties reaching back to the countries of origin. References
ARENDT, HANNAH 1949 Es gibt nur ein einziges Menschenrecht, in Dolf Sternberger (ed), Die Wandlung, Vol. IV. Heidelberg, pp. 754770 1973 [1959] The Origins of Totalitarianism, New York: Harcourt Brace Jovanovich BASCH, LINDA, GLICK SCHILLER, NINA and SZANTON BLANC, CRISTINA 1994 Nations Unbound. Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nation-States, Langhorne, PA: Gordon and Breach BODNAR, JOHN 1985 The Transplanted. A History of Immigrants in Urban America, Bloomington, IN: Indiana University Press

220 Thomas Faist

BOURDIEU, PIERRE 1983 konomisches Kapital, kulturelles Kapital, soziales Kapital, in Reinhard Kreckel (ed.), Soziale Ungleichheiten (Soziale Welt, Sonderheft 2), pp. 183198 BOURNE, RANDOLPH 1996 [1916] Trans-national America, in Werner Sollors (ed.), Theories of Ethnicity. A Classical Reader, London: Macmillan, pp. 93108 BRUBAKER, WILLIAM ROGERS 1992 Citizenship and Nationhood in France and Germany, Cambridge, MA: Harvard University Press CAGLAR, AY E 1995 German Turks in Berlin: social exclusion and strategies for social mobility, New Community, vol. 21, no. 3, pp. 30923 CEBECIOGLU, TARIK 1995 nderungen im Staatsangehrigkeitsrecht der Trkei, Das Standesamt, vol. 48, pp. 2345 CLIFFORD, JAMES 1994 Diasporas, Cultural Anthropology, vol. 9, no. 3, pp. 30238 COHEN, ROBIN 1997 Global Diasporas. An Introduction, London: University College of London (UCL) Press COLEMAN, JAMES S. 1990 Foundations of Social Theory, Cambridge, MA: The Belknap Press of Harvard University Press CONZEN, KATHLEEN NEILS 1991 Mainstreams and side channels: The localization of immigrant cultures, Journal of American Ethnic History, vol. 11, no. 1, pp. 520 DURKHEIM, MILE 1965 [1912] The Elementary Forms of Religious Life, translated by J.W. Swain, New York: The Free Press ENGELBREKTSSON, ULLA-BRITT 1978 The Force of Tradition: Turkish Migrants at Home and Abroad, Gteborg: Acta Universitatis Gothoburgensis FAIST, THOMAS 2000 The Volume and Dynamics of International Migration and Transnational Social Spaces, Oxford: Oxford University Press FALK, SVENJA 1998 Dimensionen kurdischer Ethnizitt und Politisierung: eine Fallstudie ethnischer Gruppenbildung in der Bundesrepublik Deutschland, Baden-Baden: Nomos GANS, HERBERT J., 1979 Symbolic ethnicity: the future of ethnic groups and cultures in America, Ethnic and Racial Studies, vol. 2, no. 1, pp. 120 GEERTZ, CLIFFORD 1973 The Interpretation of Cultures, New York: Basic Books GELLNER, ERNEST 1983 Nations and Nationalism, Ithaca, NY: Cornell University Press GERDES, JRGEN 2000 Multikulturalismus und doppelte Staatsangehrigkeit, in Thomas Faist (ed.), Dazwischen und doch verortet. Transstaatliche Rume in und zwischen Deutschland und der Trkei, Bielefeld: transcript GLICK SCHILLER, NINA and FOURON, GEORGES 1998 Transnational lives and national identities: the identity politics of Haitian immigrants, in Michael Peter Smith and Luis Eduardo Guarnizo (eds), Transnationalism from Below, New Brunswick: Transaction Publishers, pp. 13061 GMELCH, GEORGE 1980 Return migration, Annual Review of Anthropology, vol. 9, pp. 135159 GORDON, MILTON 1964 Assimilation in American Life, New York: Oxford University Press GOULBOURNE, HARRY 1991 The offence of the West Indian. Political leadership and the communal option, in Pnina Werbner and Muhammad Anwar (eds), Black and Ethnic Leaderships in Britain. The Cultural Dimensions of Political Action, London: Routledge, pp. 296322 GOULDNER, ALVIN W. 1960 The norm of reciprocity: a preliminary statement, American Sociological Review, vol. 25, no. 2, pp. 16178 HAMMAR, TOMAS 1990. Democracy and the Nation-State. Aliens, Denizens and Citizens in a World of International Migration, Aldershot: Gower HANNERZ, ULF 1996 Transnational Connections. Culture, People, Places, New York: Routledge HARVEY, DAVID 1989 The Condition of Postmodernity, Oxford: Blackwell HOBBES, THOMAS 1962 edn Leviathan or the Matter, Form and Power of a Commonwealth Ecclesiastical and Civil, London: Collier-Macmillan

Citizenship and culture

221

ISENSEE, JOSEF 1974 Die staatsrechtliche Stellung der Auslnder in der Bundesrepublik Deutschland, Verffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer, vol. 32, Berlin: de Gruyter, pp. 15174 JACOBSON, MATTHEW FRYE 1995 Special Sorrows. The Diasporic Imagination of Irish, Polish, and Jewish Immigrants in the United States, Cambridge, MA: Harvard University Press KANT, IMMANUEL 1984 edn Zum ewigen Frieden, Stuttgart: Reclam KEOHANE, ROBERT O. and NYE, JOSEPH S. 1977 Power and Interdependence. World Politics in Transition, Boston, MA: Little, Brown KILI, MEMET 1994 Deutsch-trkische Doppelstaatsangehrigkeit? Das Standesamt, vol. 47, pp. 738 KYMLICKA, WILL 1995 Multicultural Citizenship. A Liberal Theory of Minority Rights, Oxford: Clarendon Press MARSHALL, T.H. 1964 Class, Citizenship and Social Development. Essays by T.H. Marshall, New York: Anchor Books MEYER, JOHN W., BOLI, JOHN, THOMAS, GEORGE M. and RAMIREZ, FRANCISCO O. 1997 World society and the nation-state, American Journal of Sociology, vol. 103, no. 1, pp. 14481 NEDERVEEN PIETERSE, JAN 1994 Globalisation as hybridisation, International Sociology, vol. 9, no. 2, pp. 16184 NISBET, ROBERT 1966 The Sociological Tradition, New York: Basic Books NONINI, DONALD M. and ONG, AIHWA 1997 Introduction, in Aihwa Ong and Donald M. Nonini (eds), Underground Empires. The Cultural Politics of Modern Chinese Transnationalism. London: Routledge, pp. 336 PORTES, ALEJANDRO 1996 Transnational communities: their emergence and signi cance in the contemporary world system, in Roberto Patricio Korzeniewicz and William C. Smith (eds), Latin America in the World Economy, Westport, CT: Praeger, pp. 15168 PORTES, ALEJANDRO and ZHOU, MIN 1994 Should immigrants assimilate?, The Public Interest, no. 116 (Summer), pp. 1833 PORTES, ALEJANDRO, GUARNIZO, LUIS E. and LANDOLT, PATRICIA 1999 Introduction, Special Issue: Transnational Communities, Ethnic and Racial Studies, vol. 22, no. 2, pp. 21737 DE RAHM, GRARD 1990 Naturalisation: the politics of citizenship acquisition, in Zig Layton-Henry (ed.), The Political Rights of Migrant Workers in Western Europe, London: Sage, pp. 15885 REX, JOHN 1991 The political sociology of a multi-cultural society, European Journal of Intercultural Studies, vol. 2, no. 1, pp. 719 ROUSE, ROGER 1991 Mexican migration and the social space of postmodernism, Diaspora, vol. 1, no. 1, pp. 823 ROUSSEAU, JEAN-JACQUES 1973, 4th edn, The Social Contract and Discourses, London: Dent, Everymans Library SAFRAN, WILLIAM 1991 Diasporas in modern societies: myths of homeland and return, Diaspora, vol. 1, no. 1, pp. 8395 SCHEIN, LOUISA 1998 Forged transnationality and oppositional cosmopolitanism, in Michael Peter Smith and Luis Eduardo Guarnizo (eds), Transnationalism from Below, New Brunswick: Transaction Publishers, pp. 291313 EN, FARUK and KARAKASOGLU, YASEMIN 1994 Einstellungen zum kommunalen Wahlrecht, zu Parteien und zu der doppelten Staatsangehrigeit bei Auslndern in der Bundesrepublik Deutschland, Die Zeit, no. 39/94 (23 September 1994), p. 6 SHEFFER, GABRIEL 1986 A new eld of study: modern diasporas in international politics, in Gabriel Sheffer (ed.), Modern Diasporas in International Politics, London: Croom Helm, pp. 115 SCHMITTER HEISLER, BARBARA 1984 Sending countries and the politics of emigration and destination, International Migration Review, vol. 19, no. 3, pp. 46984

222 Thomas Faist

SMITH, MICHAEL PETER and GUARNIZO, LUIS EDUARDO (eds), Transnationalism from Below, New Brunswick: Transaction Publishers SMITH, ROBERT 1999 Re ections on migration, the state and the construction, durability and newness of transnational life, in Ludger Pries (ed.), Migration and Transnational Social Spaces, Ashgate: Aldershot, pp. 187219 SOYSAL, YASEMIN N. 1994 The Limits of Citizenship, Chicago, IL: University of Chicago Press SWIDLER, ANN 1986 Culture in action: symbols and strategies, American Sociological Review, vol. 51, pp. 27388 TATLA, DARSHAN SINGH 1999 The Sikh Diaspora. The Search for Statehood, London: University College London (UCL) Press TRACES 1998 Newsletter No. 3 issued by the programme on Transnational Communities, University of Oxford WALDRON, JEREMY 1995 Minority cultures and the cosmopolitan alternative, University of Michigan Journal of Law Reform, vol. 25, pp. 75193 YANCEY, WILLIAM L., ERICKSEN, EUGENE P. and JULIANI, N. 1976 Emergent ethnicity: a review and reformulation, American Sociological Review, vol. 41, no. 3, pp. 391403 YOUNG, IRIS MARION 1989 Policy and group difference. A critique of the Ideal of universal citizenship, Ethics, vol. 99, no. 2, pp. 25074 1997 A multicultural continuum: a critique of Will Kymlickas ethnic-nation dichotomy, Constellations, vol. 4, no. 1, pp. 4853

THOMAS FAIST is Assistant Professor (Privatdozent) at the Institute for Intercultural and International Studies, University of Bremen. ADDRESS: InIIS, University of Bremen, Postfach 33 04 40, D-28334 Bremen, Germany.