Theorizing transnational immigration: a critical review of current efforts

Peter Kivisto Abstract

During the past decade, transnationalism has entered the lexicon of migration scholars. As with other terms used in the study of immigration and ethnicity, this concept suffers from ambiguity as a result of competing de nitions that fail to specify the temporal and spatial parameters of the term and to adequately locate it vis--vis older concepts such as assimilation and cultural pluralism. This article offers a review and critique of the ways the term has come to be employed at the hands of key spokespersons that have articulated the most sustained theoretical rationales to date for transnationalism as a conceptual construct to account for new immigrant identities and communities. The conclusion of the essay offers in schematic form an alternative assessment of transnationalism that locates it as one potential subset of assimilation theory, rather than as an alternative to it. Keywords: Transnational immigration; nation-state; assimilation; pluralism; citizenship; globalization.

During the past decade, the concept of transnationalism has entered the lexicon of migration scholars, embraced by those who are attracted to its attempt to capture the distinctive and characteristic features of the new immigrant communities that have developed in the advanced industrial nations at the core of the capitalist world system (Roberts 1995; Glick Schiller 1997; Portes 1999a; Vertovec 1999; Faist 2000a; Urry 2000). It has arisen without an effort to connect it to and perhaps in some instances without a discernable awareness of social critic Randolph Bournes (1916) earlier depiction of the potential for forging a transnational America during the zenith of the last major wave of immigration in the United States. Rather, in its current guise, it has been portrayed as the migratory counterpart to other contemporary discussions of manifestations of transnationalism, particularly those related to corporate capitalism as a global phenomenon (Sklair 2001), the growth
Ethnic and Racial Studies Vol. 24 No. 4 July 2001 pp. 549577 2001 Taylor & Francis Ltd ISSN 0141-9870 print/1466-4356 online DOI: 10.1080/01419870120049789

550 Peter Kivisto of transnational NGOs, citizenship in a postnational era, and the increasing hybridization or creolization of popular culture (Soysal 1994; Appadurai 1996; Castells 1996, 1997; Hannerz 1996; Sassen 1996). The term has emerged and evolved at a time characterized by high levels of labour migration from economically less developed nations to the most developed and from similarly high levels of political refugees eeing con icts and instability in former communist and Third World nations (Castles and Miller 1993). The in ux of these new labour migrants and refugees is reshaping the ethnic mixes, not only of nations with long histories of immigration, the settler states of the United States, Canada and Australia, but also of states that have not been notable as immigrant receiving nations in the earlier phases of industrialization, those of Western Europe and, to a lesser extent, Japan. The high levels of migration, the new locales of settlement, changes in the nature of capitalist economies in a new industrial epoch, changes in the meaning and signi cance attached to the idea of citizenship, and the potency of a globalized popular culture have contributed to the conviction that what is novel about the present requires equally novel conceptual tools if we are to make sense of the impact of the new immigration both on the receiving and sending countries. However, as with other terms used in the study of immigration and ethnicity including the older concepts of assimilation and cultural pluralism that arose during an earlier period of mass immigration and were derived chie y from the American experience, as well as the more recent cross-culturally inspired concepts of multiculturalism, and globalization this concept suffers from ambiguity as a result of competing de nitions that fail to specify the temporal and spatial parameters of the term and to adequately locate it vis--vis the other concepts noted above. Steven Vertovec (1999, pp. 44956), who in his capacity as Director of the ESRCs Research Programme on Transnational Communities is an active promoter of the concept, points out several recurring themes that shape the ways the term is employed. He identi es six distinct, albeit potentially overlapping or intertwined, uses of the term: (1) as a social morphology focused on a new border spanning social formation; (2) as diasporic consciousness; (3) as a mode of cultural reproduction variously identi ed as syncretism, creolization, bricolage, cultural translation, and hybridity; (4) as an avenue of capital for transnational corporations {TNCs}, and in a smaller but signi cant way in the form of remittances sent by immigrants to family and friends in their homelands; (5) as a site of political engagement, both in terms of homeland politics and the politics of homeland governments vis--vis their migr communities, and in terms of the expanded role of international non-governmental organizations {INGOs}; and (6) as a recon guration of the notion of place from an emphasis on the local to the translocal. Note that not all of these uses pertain to immigration, but include the

Theorizing transnational immigration 551 signi cance of global corporate capitalism and INGOs. Whatever the merits of the idea of transnationalism in regard to these phenomena, the task undertaken herein is of more limited focus: to assess the concept as it relates to immigration, appreciating that in fact these are reciprocally interconnected phenomena. My purpose is twofold. First, I will review and critique the ways the term has come to be employed at the hands of key spokespersons that have articulated the most sustained theoretical rationales to date for transnationalism as a conceptual construct to account for new immigrant identities and communities. Second, based on the critique of the concept, I shall offer in schematic form an alternative assessment of transnationalism that locates it as one potential subset of assimilation theory, rather than as an alternative to it. Three versions of transnationalism Although transnationalism is used fairly widely by anthropologists, sociologists, political scientists, and geographers, for the purposes of this article, three versions of the concept will be discussed: (1) the earliest articulation by cultural anthropologists Nina Glick Schiller, Linda Basch, and Christina Szanton Blanc (1992, 1995; see also Basch, Glick Schiller and Szanton Blanc 1994 and Glick Schiller 1997); (2) the re nement of the term by sociologist Alejandro Portes (1996a, 1996b, 1998, 1999a, 1999b; see also Portes, Guarnizo and Landolt 1999) and its deployment by him and numerous colleagues in research on the second generation and beyond (Portes 1995; Portes and Zhou 1999); and (3) the most rigorously systematic theoretical articulation of the term by political scientist Thomas Faist (1998, 2000a, 2000b, 2000c). Transnationalism from the perspective of cultural anthropology Though one may quibble about the date that transnationalism made its rst appearance as an additional conceptual tool in the analytical arsenal of immigration scholars, a reasonable case can be made for a period between 1990 and 1994 as its debut years, for it was during these years that the rst articles appeared attempting to lay out the conceptual parameters de ning transnational immigrants and their communities. It was in 1990 that cultural anthropologists Nina Glick Schiller, Linda Basch and Christina Szanton Blanc organized a conference on transnationalism that was co-sponsored by the New York Academy of Sciences, the Wenner-Gren Foundation, and the Institute for the Study of Man. Two years later they edited a collection based on conference papers entitled Towards a Transnational Perspective on Migration (Glick Schiller, Basch and Szanton Blanc 1992) in which their introduction and lead article presented transnationalism as a novel analytic approach to understanding contemporary migration. Two years later

552 Peter Kivisto they again organized a conference, a Wenner-Gren sponsored symposium held in Mijas, Spain, on Transnationalism, Nation-Building, and Culture and in the same year published another book, Nations Unbound (Basch, Glick Schiller and Szanton Blanc 1994). These two publications constitute the rst theoretical articulations of the transnational perspective on contemporary migration. Glick Schiller and her colleagues, in the lead article to their 1992 collection, Transnationalism: A New Analytic Framework for Understanding Migration, make clear that they seek to articulate a new conceptual model for interpreting contemporary migration. They reiterate and in some minor ways amplify their position in the two introductory articles of Nations Unbound , Transnational Projects: A New Perspective and Theoretical Premises. Since much is repeated verbatim in the second book and there is no effort to revise or retreat from the arguments advanced in the 1992 article, though the earliest version is in some ways ampli ed, it is reasonable to treat these three articles in tandem as a uni ed statement of the authors position. Glick Schiller et al. make two initial points, one historical and the other theoretical. Historically, they contend that there is something qualitatively different about immigrants today compared to their late nineteenth and early twentieth centuries counterparts. Apparently agreeing with Oscar Handlins (1973) depiction of the latter group as uprooted a perspective shaped by the in uence of the Chicago School of Sociologys concepts of social and personal disorganization Glick Schiller et al. are prepared to view this earlier eras immigrants as having broken off all homeland social relations and cultural ties, and thereby locating themselves solely within the socio-cultural, economic, and political orbit of the receiving society. By contrast, they contend that todays immigrants are composed of those whose networks, activities and patterns of life encompass both their host and home societies. Their lives cut across national boundaries and bring two societies into a single social eld (Glick Schiller, et al. 1992, p.1; see also Glick Schiller 1997, p. 158). From this historical comparison, Glick Schiller et al. offer a rationale for a new analytic framework, making a case for the introduction of two new terms: transnationalism and transmigrants. The former refers to the process by which immigrants build social elds that link together their country of origin and their country of settlement, while the latter refers to the immigrants who build such social elds by maintaining a wide range of affective and instrumental social relationships spanning borders (Schiller et al. 1992, p. 1; Basch et al. 1994, p. 27). Implicitly, the introduction of these new concepts suggests that existing theoretical frameworks are not up to the task of analysing the new immigrants. Glick Schiller (1997, p. 158) makes this more explicit, when she contends that by embracing these concepts, scholars are discarding previous categorizations of return, circulatory, or permanent immigration.

Theorizing transnational immigration 553 To make their case, the authors offer three vignettes (Glick Schiller 1992, pp. 24). The rst involves a Haitian hometown association located in New York City. While the activities of the association clearly have something to do with immigrant adjustment, it has also initiated various projects in Haiti. This, Glick Schiller et al. contend, distinguishes contemporary mutual aid societies from those in the past, which they argue were solely designed to address the adjustment needs of the immigrants themselves (a more sustained treatment of the Haitian case is found in Basch et al. 1994, pp. 145224). The second example involves white-collar Grenadian immigrants being addressed by Grenadas Minister of Agriculture and Development. The point of this illustration is that the immigrants are at once both Grenadian nationals with an ability to in uence friends and relatives who have remained in the homeland and American ethnics capable of undertaking efforts to shape economic and political decisions in the host society (see Basch et al. 1994, pp. 49144). The third example looks to Filipinos and the Balikbayan box, a formalized and regulated form of remittance. The main point of this illustration is not made explicitly, but appears to be, not that remittances are new, but that homeland governments are increasingly likely to embrace their expatriate communities when such an embrace can be economically bene cial. From these illustrations, the authors make two main conceptual points. The rst, and the most original and crucial for their case, is that social science must become unbound. The problem with theories that operate as closed systems in which the unit of analysis is ultimately the nation-state (they point to structural functionalism in anthropology and Parsonian systems theory in sociology as the most in uential proponents of such a view) is that they fail to provide room for the wider eld of action occupied by contemporary immigrants. Such theoretical perspectives are faulted for their failure to provide an adequate basis for comprehending the dialectical interplay between homeland concerns and receiving nation realities and the impact this interplay has on immigrants. Thus, Glick Schiller et al. argue for the necessity of recasting theory from the nation to a global or world systems perspective. In this regard, they note that Wallersteins world systems theory is a useful corrective to closed systems, but in their view it suffers from a tendency towards economic reductionism. Nonetheless, they want to stress that transnationalism is the product of world capitalism (Glick Schiller et al. 1992, pp. 512; Basch et al. 1994, pp. 3034). In this regard, their discussion raises questions about their historical framework. They state that the world is currently bound together by a global capitalist system that produces economic dislocations making immigrants economically vulnerable. The result is a new and different phenomenon, . . . a new type of migrant experience . . . (Glick Schiller et al. 1992, pp. 89). While sympathetic to the discussions of transnationalism as cultural ows, seen in the work of gures such as Arjun Appadurai (1996)

554 Peter Kivisto and Ulf Hannerz (1996), the main thrust of the authors argument involves an articulation of a notion of transnational migration that focuses primarily on social relations. Finally, the authors point to the multiple and uid identities of contemporary transmigrants, but beyond offering a fairly commonplace social constructionist statement about the invented or imagined character of immigrant identities, they suggest that the manipulation of identities reveals a resistance on the part of transmigrants to the global political and economic situations that engulf them . . . (Glick Schiller et al. 1992, p. 11; see also Glick Schiller 1997, p. 164). This leads to the second conceptual point: the need to rethink received ideas regarding class, nationalism, ethnicity and race. Relying on the Gramscian idea of hegemony, the authors treat each of these aspects of identity as contested and pliable. In the light of the attention devoted to these issues, it is surprising that there is nothing particularly original about their discussion. Moreover, it fails to substantively clarify what they understand transnationalism to mean and to entail. The real signi cance of the discussion is that here they attempt to identify the shortcomings of assimilation and cultural pluralism, important in so far as this serves to establish the need for new conceptual tools to adequately comprehend contemporary immigration. Though crucial to their case, the critiques are remarkably cursory. Assimilation is reduced to the ideology of the melting pot, with its image of what it means to be an American in a land of economic opportunity. The main criticism is that assimilation theorists have tended to focus on ethnicity, and in the process have ignored race, which if true is less a criticism of the theory than of the way it has been put to use. Cultural pluralism is described as a view that focuses on the structural and cultural persistence of ethnic groups. Advocates of this theoretical tradition are faulted for their presumed tendency to ignore differences within ethnic groups. This again is a criticism of the ways the theory has been employed rather than a critique of its analytic utility. However, on the basis of these remarks, the authors conclude that it is necessary to create an alternative theory instead of attempting to revise or otherwise employ assimilation theory or cultural pluralism (Glick Schiller et al. 1992, pp. 1319). Thus, the crux of Glick Schiller et al.s argument boils down to the assertion that new times and new socio-historical circumstances demand a new theoretical paradigm. But how convincing are they? I would argue not very. In the rst place, they are off the mark historically, evidencing little awareness of the substantial body of scholarship that social historians and historical sociologists of the earlier wave of immigration have produced since the 1960s. In synthetic summaries of recent scholarship, it is quite clear that immigrants during the industrial era acculturated, but they did so in part on their own terms (Kivisto 1990; Morawska 1999; forthcoming). As part of this process, many groups continued to maintain an active interest and involvement in their homeland (see, for

Theorizing transnational immigration 555 example, Bodnar 1985; Daniels 1990; Fuchs 1990; Portes and Rumbaut 1990; Takaki 1993). From this substantial body of evidence, it would appear that the historical contrast Glick Schiller et al. make is overdrawn. The authors (1992, p. 8) locate the rise of a global capitalist system in the 1980s. Since they invoke Wallersteins world systems theory, one might have expected that they would have located the rise as beginning in the sixteenth century. However, even without this expansive view, it behoves the authors to differentiate the factors leading to the labour migrations of the nineteenth and early twentieth centuries compared to contemporary migrations. It seems altogether obvious that all of the labour migrants in both of these eras can appropriately be viewed as, to use John Bodnars (1985) felicitous phrase, children of capitalism. The exigencies of the capitalist market shaped the lives of past immigrants and shape the lives of present immigrants similarly, forcing individuals in both waves to live with considerable insecurity. Not surprisingly, in both time periods immigrant responses to capitalism varied considerably. Most, at least in the longer term, bought into the system rather than attempting to resist or subvert it. Delmos Jones (1992, p. 223) is correct to note that most of Glick Schiller et al.s transmigrants are actually seeking incorporation into the capitalist system rather than resisting it. And as John Bodnars (1985) synthetic portrait of the transplanted shows, the earlier immigrants too sought to nd those modes of incorporation most advantageous to themselves. Indeed, if there is one signi cant difference between these two waves of immigrants, it is that there was more evidence of resistance from socialists, communists, and anarchists with visions of an alternative to capitalism in the past than in the present (Rosenbaum 1973; Miller 1974; Kivisto 1984; Lipset and Marks 2000). Glick Schiller et al. are incorrect when they state that the mutual aid organizations of past immigrants differ from those of transmigrants because the former were solely concerned with immigrant adjustment. Immigrants in the past offer ample evidence of concern about and active involvement in homeland relations and issues. Remittances are nothing new, even if the contemporary Filipino case is characterized by a formal, governmentally regulated structure uncharacteristic of an earlier era and quite frankly of most remittances today. Immigrants in both periods have sent money and goods to relatives and friends in the homeland. They have been sources of information and support for individuals in their hometowns contemplating migration. At the political level, they have been actively involved in homeland politics (Jones 1976; DeConde 1992). Finally, the desire to return to ones homeland after obtaining enough capital to make an economically successful return is not new, with the Chinese birds of passage who ventured to the US in the middle of the nineteenth century constituting something of an ideal type (Lyman

556 Peter Kivisto 1974). Return migration is not a new phenomenon. Neither is circulatory migration, including seasonal migrations (Daniels 1990, p. 25). All of this casts into question the assertion that there is something qualitatively different about current immigrants compared to those of the past. Todays transmigrants do not look as different from yesterdays immigrants as Glick Schiller et al. presume them to be. Thus, if the terms transnationalism and transmigrants are to be embraced as useful concepts, it would appear that they should not be limited to present immigrants. On the other hand, should all immigrants today and yesterday be viewed as transnational? What about the immigrants from either era that sever family ties, care little if at all about homeland issues, and never return home? Would not it be more appropriate to consider which immigrants from both periods qualify as transmigrants and which ones do not? While a reasonable expectation, Glick Schiller et al. fail to provide criteria by which we could make such a distinction. Instead, we are left with a dichotomy between immigrants from the past and transmigrants of the present. Theoretically, the authors fail to make a convincing argument for the ef cacy of transnationalism as an alternative to assimilation or cultural pluralism. Key here is their failure or more explicitly, their refusal to wrestle with either of these as genuine theories, opting instead to see them as ideological expressions. While certainly both suffer from considerable ideological baggage, they are also theoretical constructs. In terms of the former, its long history in the social sciences from the often misconstrued formulations by Robert E. Park and the other members of the Chicago School of Sociology, through the comprehensive and balanced discussion in Milton Gordons (1964) classic work, to recent defences (Morawska 1994), developments (Gans 1979, 1992a), re nements (Portes 1995), reconceptualizations (Barkan 1995; Alba and Nee 1997) and applications (Gans 1992b) is entirely missing in Glick Schiller et al.s account. Of particular signi cance for this discussion is the extent to which assimilation theory can account for the types of phenomena that are described as transnational. When freed from an invariant teleology, assimilation theory has the potential for making sense not only of acculturation and incorporation, but also resistance, group survival, discrimination, ethnic con ict, and variation in outcomes. Gordon (1964) thought that he could account for cultural pluralism under the overarching theoretical umbrella of assimilation. Recent reconsiderations of assimilation have not necessarily agreed with his particular formulation, but they do agree with his domain assumption. The possibility that Glick Schiller et al. do not ponder is whether assimilation might similarly be capable of incorporating transnationalism. What we are left with is a brief on behalf of transnationalism that is problematic empirically and theoretically. Nonetheless, the work of

Theorizing transnational immigration 557 Glick Schiller et al. during this early phase in the development of the concept marks the beginning of a signi cant incursion of the idea of transnationalism into current scholarship on the contemporary age of migration. I should note that Glick Schillers (1999) subsequent revisiting of this earlier work serves as a corrective to some of these problematic features. Nonetheless, the general orientation of the original formulation has ltered into general thinking about the concept, and for that reason deserves careful scrutiny. Transnationalism as middle-range theory Perhaps the person most responsible for popularizing and expanding the use of the transnational perspective is Princeton sociologist Alejandro Portes (1996a, 1996b, 1998, 1999a; 1999b; Portes, Guarnizo and Landolt 1999). The focus of Portes research has been the new, post-1965, immigration to the United States. He has sought to expand our understanding of the role of ethnic enclave economies for these newcomers (Portes and Zhou 1999), offer a more complex notion of assimilative outcomes by introducing the idea of segmented assimilation (Portes 1995), and has urged a shift in research agendas from the immigrant generation to the second generation (Portes 1998). Though he has not done original research on the last great wave of immigration to the United States, as the book he co-authored with Rubn Rumbaut on Immigrant America (1990) reveals, he brings to his understanding of contemporary immigration a sophisticated comparative historical perspective that is lacking in the work of Glick Schiller et al. Portes has emerged as the pre-eminent scholar of the new immigration, and has served as something of a magnet for numerous social scientists engaged in empirical studies of new immigrant communities. Herein we explore only one aspect of Portes work: his advocacy on behalf of claims for the utility of employing the idea of transnationalism in analysing some new immigrant communities. His most de nitive statement to date is contained in two articles, a co-authored introduction and a solely authored conclusion, that framed a special issue of Ethnic and Racial Studies devoted to the study of transnationalism. In these two essays, Portes offers his most sustained assessment of what he means by the term and why he thinks it is a useful addition to the conceptual tool bag of scholars of immigration and ethnicity (Portes, Guarnizo and Landolt 1999; Portes 1999a). We focus primary attention on these essays, but also refer to his use of the term in a few articles that preceded them (Portes 1996a, 1996b, 1998). Before examining the two key articles, it is instructive to explore earlier manifestations of interest in transnational discourse that preceded these more fully developed conceptual statements. As with Glick Schiller and her colleagues, Portes has made use of vignettes to

558 Peter Kivisto illustrate the phenomenon. In fact, he has used some of the same examples of emerging transnational communities in at least four different articles (Portes 1996a, 1996b, 1998, 1999b): a Mexican enclave in Brooklyn working to create a safe water system for their Mexican village of origin; Dominican entrepreneurs who spent time in the United States but returned home to operate small businesses there, with economic support from the expatriate community; Otavalo traders from Ecuador who sell their ethnic wares in street markets in North America and Western Europe; and the new Chinatowns in suburban Monterey Park, California and Flushing, New York. It is worth examining each brie y in order to determine to what extent they might be construed as typical or atypical instances of contemporary immigrant communities. The Mexican example deals with immigrants remaining actively engaged in homeland concerns, and in this respect is not unlike the Haitian example used by Glick Schiller et al. In this case, the Mexican community in Brooklyn not only raised over $50,000 to improve the waterworks of their hometown of Ticuani, but a committee was actively engaged in the design of the plan in an ongoing way. Central to the viability of transnational communities, Portes contends, are improved communication and transportation networks. This made possible not only continuing oversight of the project from Brooklyn, but also the inspection of the materials and the actual installation of pipes by a committee that ew from New York to Mexico. Re ecting the desire to maintain homeland ties, a seal was designed that read, For the Progress of Ticuani: The Absent Ones, Always Present. Thus, rather than claim that earlier immigrants lacked a desire to be involved in homeland issues, Portes argument is that what makes the situation different today is that improved communication channels and transportation systems make it possible to act more readily on that desire. While probably true, two unanswered questions need to be raised lest technology alone is used to explain the presumed difference between past and present. First, how common are such organizations, not only for the Mexican-American community, but also for groups lacking the same kind of proximity to the United States? Second, will such active involvement in the homeland persist across generations, or is this, to a large extent, a rst-generation phenomenon? The Dominican example concerns the economic rather than the sociocultural. Hundreds of small business owners of factories, commercial enterprises, and nancial organizations in the Dominican Republic maintain close social relationships with the expatriate community in the United States, which serves as a vital source of investment capital and as a market for goods and services. Thus, though their operations are physically located in the Dominican Republic, the geographic eld of reference of these entrepreneurs is transnational. Portes has suggested that transnational communities are in a sense labours analog to the

Theorizing transnational immigration 559 multinational corporation (1996a, p. 74). This example, however, does not involve a global proletariat, but rather international small business or petite bourgeois operatives. While without a doubt this kind of business activity can be seen in many and perhaps even most of the new immigrant communities, the number of individuals involved in similar entrepreneurial ventures as a percentage of the total immigrant population remains an unanswered question, but we can assume it is relatively small. I pass over the example of the niche for authentic folk wares that the enterprising Otavalans have carved out for themselves in the major cities of the advanced industrial nations. Albeit fascinating, this example is clearly atypical. Drawing conclusions about transnational communities from the Otavalans would be like describing immigrant communities in Canada in the early twentieth century by using the Hutterites as their paradigmatic expression. The nal example looks at the birth of new Chinatowns outside of the older central city locales of the long-established Chinatowns. Robert E. Park (1950, p. 151) had predicated the demise of Chinatowns as a result of immigrant restriction; with the post-1965 resumption of immigration, older Chinatowns have been revitalized and new ones have emerged. The rather limited research to date on these new communities have tended to focus on the new patterns of accommodation to the United States that serve to differentiate immigration today from the past (Chen 1992; Fong 2000). Portes interest in these communities arises as a result of the role played by Taiwanese and Hong Kong nanced banks in these communities and the business astronauts who travel frequently between the US and Asia. The former is reminiscent of the rotating credit associations that have characterized Chinese immigrant communities since the nineteenth century (Takaki 1989, p. 241). The latter is a new phenomenon. The unanswered question about the former is how long these relatively small operations will continue to function. Part of the reason for their existence was the export of capital out of Hong Kong prior to the PRC takeover of the colony. It would be interesting to know how this has changed since the British handed over control of Hong Kong. More to the point, as Chinese immigrants and their children get acculturated and their credit ratings improve, will they continue to patronize Chinese banks, or will they opt for larger full-service American banks? And the unanswered question about the astronauts is the same as that posed about the Dominican entrepreneurs: how large a part of their community are they? With these examples in mind, we now turn to the above-mentioned Ethnic and Racial Studies framing essays, wherein Portes offers his most sustained articulation of his understanding of what transnationalism entails, why it is unique, and what its implications are for ethnic communities over time. At the outset, the distinctiveness of this phenomenon is described in historical comparative terms.

560 Peter Kivisto While back and forth movements by immigrants have always existed, they have not acquired until recently the critical mass and complexity necessary to speak of an emergent social eld. This eld is composed of a growing number of persons who live dual lives: speaking two languages, having homes in two countries, and making a living through continuous regular contact across national borders (Portes, et al. 1999, p. 217). Noting the varied and sometimes contradictory ways the term has been put to use, Portes and his colleagues seek to offer a middle-range theory of transnationalism capable of shaping a research agenda. They invoke Merton (1987) to suggest the circumstances in which it is appropriate to create a new concept: (1) if a signi cant percentage of immigrants are involved in the process; (2) if the activities they engage in exhibit persistence over time; and (3) if existing concepts fail to capture the content of these activities (Portes et al. 1999, pp. 2189). Assuming that each of these criteria has been met, on what would appear to be rather thin empirical grounds, they proceed to re ne and delimit the broad de nition inherited from Glick Schiller et al. and evident in much of the literature. This begins with a decision to abandon the term transmigrant, suggesting that the word immigrant suf ces and does not need to be supplemented or replaced. Next, they limit the use of the term transnationalism to include only those activities that involve continuity of social relationships across national borders over time. In other words, in contrast to Glick Schiller et al., only some, but not all, contemporary immigrants are transnationals. As an example of who would not count, they point to immigrants who on occasion send gifts to family and friends back home. Whether or not this would exclude Filipinos sending Balikbayan boxes is not obvious, since these are sent only periodically, but have acquired a formalized and regulated character. A nal form of de nitional delimitation identi es individuals and their support networks as the appropriate units of analysis, thereby excluding communities and more overarching structural units such as governments (Portes et al. 1999, pp. 21920). The authors separation of topics is designed to reinforce the idea that transnationalism as applied to immigration is transnationalism from below, in contrast to the transnationalism from above that is manifested by global corporations and governments (Guarnizo 1997; Smith and Guarnizo 1998). That being said, it is reasonable to raise a question about why communities are excluded, particularly since in preceding articles Portes has explicitly spoken about transnational communities. He argues for this exclusion, not so much in conceptual terms, but as a methodological strategy, based on his conviction that it is advisable at this stage in transnational research to concentrate on individuals and families. However, I would suggest

Theorizing transnational immigration 561 that the study of immigrants can never be simply the study of individuals and families, but must at all points take account of the corporate life within which individuals and families are embedded. In so far as this is the case, Portes strategy may be deemed problematic. Thus, transnational immigrant communities ought to be seen as necessary objects of investigation for those interested in manifestations of transnationalism from below. Portes and his colleagues (1999, p. 221) distinguish three different types of transnationalism: (1) economic, (2) political and (3) sociocultural. Economic transnationalism involves entrepreneurs whose network of suppliers, capital and markets crosses nation-state borders. Of interest here is that Portes (1996a, p. 74; 1996b, p. 151; see also Portes et al. 1999, p. 227) has made the point elsewhere that the idea that capital is global while labour is local is called into question by the advent of transnational communities, which he described as labours analog to the multinational corporation. However, in this de nition of economic transnationalism it is not labour migrants a global proletariat but mobile capitalist entrepreneurs that become the sole representation of this type of transnationalism. In addition, missing from this formulation are members of the professional middle class the so-called brain drain immigrants that have proved to be an important component of some contemporary immigrant communities. The implication of this de nition is that it severely limits the number of those immigrants whose economic activities can actually be considered transnational, given the fact that the majority of immigrants are labour migrants who would not be considered transnational. Political transnationalism is said to involve the political activities of party of cials, governmental functionaries, or community leaders whose main goals are the achievement of political power and in uence in the sending or receiving countries (Portes et al. 1999, p. 221). Curiously, the rst two roles pertain to people who are not actually immigrants, but rather are individuals interested in in uencing or controlling their expatriate communities or shaping various governmental decisions in the immigrant receiving nations in some way. Community leaders appears to refer to leaders in the immigrant communities, though they could presumably also include community leaders in the homeland. Why one would only consider leaders, and not include rank-and- le activists involved in homeland issues remains unclear. Moreover, proceeding in this way would seem to violate the stated intention of keeping transnationalism from below analytically distinct from transnationalism from above. The socio-cultural refers to activities oriented towards the reinforcement of a national identity abroad or the collective enjoyment of cultural events and goods (Portes et al. 1999, p. 221). The latter part of this sentence describes the sorts of activities that were prominent in past

562 Peter Kivisto migratory waves, evidence of a desire on the part of those immigrants to transplant, generally selectively, aspects of their cultural heritage. Though most characteristic of the rst and second generations, even later generations are often attracted to such activities. Indeed, Gans (1979) has suggested that these are precisely the kinds of activities that highly acculturated ethnics participate in as a means of preserving a nostalgic symbolic ethnicity. The rst part of the sentence is more ambiguous and dif cult to unpack. Are the authors using the term national identity as essentially a synonym for ethnic identity (as was common among social scientists and historians until the 1960s)? Or, do they actually mean that the new immigrants seek to remain tied to the nationality of their homeland, and thus either would refuse to become citizens of the receiving society or seek the possibility of dual citizenship? If the former is the case, this means that immigrants at present are little different from immigrants in the past. However, if the latter is the case, then something new is implied, perhaps something akin to Yasemin N. Soysals (1994) limits of citizenship thesis. The net result of this operational de nition is that no longer is it necessary to view present migration as necessarily qualitatively different from the preceding eras migrants. While conceived as more common today than in the past, the historically dubious dichotomous assumption underpinning Glick Schiller et al. is modi ed. Thus, the circulatory labour migrations of the preceding era seen for example in the US in the bracero program and in Germany with the annual movement of Poles in and out of the area east of the Elbe would be included. Nevertheless, in describing the necessary conditions for transnationalism, Portes and his colleagues point to the advances in transportation and communication technologies as making possible today a time-space compression unheard of in earlier times. Thus, they contend that the scope of transnationalism is far more extensive today than a century ago (Portes et al. 1999, pp. 22327). At the same time, as a cursory examination of the way the three types of transnational immigration are described suggests, it is clear that not all contemporary immigrants are transnationals. Portes et al. differentiate immigrants in terms of their access to the technological prerequisites for transnationalism. Those with higher levels of social capital would be more likely to forge transnational linkages than those with less capital. At the same time, proximity continues to count: groups with homelands closer to the receiving nation will be the most likely candidates for establishing transnational ties (Portes et al. 1999, p. 224). Given these stipulations, it would appear that those that can actually be de ned as transnational immigrants might in fact constitute a minority of todays total immigrant population. Despite the actual number of transnational immigrants, the authors point to the potential impact they might have on forging novel

Theorizing transnational immigration 563 boundary-breaking social elds by serving as the facilitators making possible the development and possible institutionalization over time of various translocal economic, political and cultural ties (Portes et al. 1999, pp. 22829). To the extent that this process occurs, Portes et al. contend that received versions of assimilation theory need to be revisited in order to consider the possibility of new modes of adaptation to the receiving country. They point to four possible outcomes, only one of which actually suggests transnationalism is both new and relatively permanent. The three that do not include the return home of transnational entrepreneurs with their families; the abandonment of transnationalism in favour of full assimilation into the receiving country; and the rejection of transnationalism by the children of transnational immigrants in favour of assimilation. The one that does entails the generational transmission of a transnational perspective and the knowledge needed to promote it. It is this latter prospect that they conclude may well become the most common accommodative path chosen by immigrants in the future (Portes et al. 1999, p. 229). In his conclusion, Portes (1999a, pp. 46466) treats transnationalism as one possible outcome out of several, and summarizes the factors that shape the choices individuals make. These include the forces contributing to migration, the extent to which homeland issues remain salient for immigrants, and the role of nativist hostility to newcomers. These factors, of course, are not new to todays immigrants; rather they can be and indeed have been employed to explain the differing modes of adaptation and acculturation among earlier immigrant groups. What is new is the role some homeland governments are playing in attempting to encourage ongoing connections with their expatriate communities. Immigrants a century ago confronted governments and cultural elites that tended to be overtly hostile; todays counterparts, in contrast, frequently nd their emigrants to be useful economically, and sometimes politically and culturally as well. Thus, rather than condemning their decision to exit or enticing them to return, they instead work to create relationships with the immigrants that are bene cial to the homeland. To the extent that immigrants perceive these relationships to be mutually bene cial, the potential for a transnational social eld arises. At the same time, Portes views assimilation as a powerful force, particularly for the second generation and beyond. In so far as this is the case, he presents transnationalism less as an alternative to assimilation than as an antidote to the tendency towards downward assimilation or in other words as a way of combating the consequences of what he has elsewhere referred to as segmented assimilation (Portes 1999a, pp. 44673; Portes 1995). Here Portes appears prepared to treat transnationalism as one variant of assimilation, in contrast to elsewhere where it is posed as an alternative to assimilation. Though offering a useful corrective to the most problematic features

564 Peter Kivisto of the Glick Schiller et al. formulation of transnationalism, this exegesis points to limitations in Portes position as well. In the end, it is not clear whether Portes views transnationalism as an alternative outcome to assimilation or as a new type or form of assimilation. Nor is it obvious, given the expected impact of acculturation on the second generation and beyond, why he thinks transnationalism as one of several possible outcomes might actually become normative. Despite these problems, one central feature of his position deserves further clari cation and development. This involves his notion of social elds, which can be seen as his way of unbounding the sociology of immigration. Needed is a more fully articulated discussion of what is meant by transnational social elds. Immigration and transnational social spaces By far the most sustained attempt at clarifying and developing the idea of the social eld has been that undertaken by Thomas Faist (2000a), though he recasts the term by speaking about transnational social spaces rather than social elds (see a related discussion in Beck 2000, p. 26). In a series of articles and papers he has developed the contours for a systematic theory of transnationalism predicated on the idea of the construction of border-crossing social spaces (Faist 1998; 2000b; 2000c). However, it is in The Volume and Dynamics of International Migration and Transnational Social Spaces (2000a) that his position achieves its most complete articulation, and thus it will be the focus of this analysis. The discussion of transnationalism occupies only the latter portion of the book. This is a re ection of Faists overarching theoretic intention: he seeks to offer a new model of migration that can account for two quandaries. First, why do so few people migrate (slightly more than 2 per cent of the worlds population according to some estimates) and why, more speci cally, are there so few people out of so many places? Second, why are so many migrants from so few places? The perspective Faist advances entails an effort to productively weave together network analysis and the concept of social capital, with particular emphasis given to the constructing of a meso-link between macro and micro approaches to migration. In wrestling with these two simple yet profoundly dif cult questions, transnationalism is treated as a piece of a much larger puzzle. In this respect, Faist is considerably more theoretically ambitious than either Glick Schiller and her colleagues or Portes (Faist 2000a, pp. 129). Since the focus of this paper is on the individual and collective implications of immigration, and not on the actual causes of immigration, we need not concern ourselves with all the facets of Faists theoretical model. Instead, the focus will be on those features that directly address the matter of whether or not it is useful to speak about new transnational modes of immigrant adaptation.

Theorizing transnational immigration 565 Faist locates his own contribution to theory construction within a tradition of scholarship devoted to migration, treating the relationship between his model and preceding ones as cumulative. In other words, rather than simply rejecting preceding theories of migration, he seeks to build on them while offering both needed correctives and expansions. More explicitly, he considers his transnational social space paradigm as an outgrowth of two earlier paradigms: the older push-pull model and the succeeding centre-periphery model. The former concentrated simply on internal factors in two nation states that together conspired to promote emigration from the homeland (push) and immigration to the receiving nation (pull). Thus, the focus was on migratory ows. The centre-periphery built upon this model by viewing the migratory process as a system and in so far as it is, it sought to identify those variables contributing to the creation and maintenance of the system. Faists new paradigm has a clear family resemblance to the centreperiphery model. What makes it distinctive is that the idea of transnational spaces treats the migratory system as a boundary-breaking process in which two (usually) or more nation-states are penetrated by and become a part of a singular new social space. The centre-periphery model added to the push-pull model the idea of a system that involved not only people but also goods and information. In similar fashion, the transnational social spaces paradigm includes the circulation of ideas, symbols, and material culture (Faist 2000a, p. 13). In a similar vein, the theory of transnational social relations is seen as yet another possible outcome of immigrant adaptation to the receiving country, one that supplements the canonical concepts of assimilation and ethnic pluralism (Faist 2000a, p. 29). This means that Faist considers both assimilation and ethnic pluralism as potential outcomes for contemporary migrants. Moreover, he disputes the claim that transnationalism is a new phenomenon that only applies to post-World War II South-North migrations. Instead, he insists that transnationalism clearly could be found in the migratory wave of a century ago to white settler states. However, he also concurs with Portes that the relative ease of long-distance travel and improved communication technologies means that transnationalism today has achieved the critical mass that makes it more extensive and consequential than its earlier manifestations (Faist 2000a, pp. 21112). A third type of complementarity involves the relationship between transnationalism and both world-systems theory and globalization. Faist contends that these three concepts should be seen as distinct, but overlapping. His understanding of world-systems theory locates it in the centre-periphery model. Though advocates of world-systems theory might disagree, his argument is that this paradigm is rooted in a perspective that looks solely at the symbiotic relationship between a particular nation from the periphery and a particular core nation. What

566 Peter Kivisto makes transnationalism different, he suggests, is that in contrast to world-systems theory it opens to consideration the possibility that migrants from a nation in the periphery may end up in several destinations that create a complex transnational space involving more than two nation-states. In this regard, it offers a broader vision than world-systems theory (Faist 2000a, p. 210). In contrast, transnationalism is seen as offering a more limited perspective than globalization theory. While the two processes can overlap, Faist distinguishes them by contending that globalization involves, referring to Marxs phrase from the Grundrisse (1973, p. 423), the annihilation of space and time. Perhaps a more appropriate characterization would have been to speak of this as David Harvey (1989) has as timespace compression or as Anthony Giddens (1984) has as time-space distanciation. For postmodern theorists such as Harvey and theorists of high modernity such as Giddens (1990, 1999), globalization refers to a process that makes possible the proliferation of social relations across greater distances, relationships that are becoming more intense and robust rather than stretched and attenuated (Waters 1995, p. 58). This would certainly appear to be what Faist and others mean by the ties forged by transnational immigrants. In this regard, the terms transnationalism and globalization might appear to be virtual synonyms (Robertson 1992; Bauman 1998). However, Faist (2000a, pp. 210211) rejects this conclusion by claiming that globalization tends to refer to processes that are decentred or deterritorialized, while transnational processes are anchored in and span two or more nation-states and thus are not denationalized. The linchpin of Faists thesis is the concept of transnational social spaces. It is his articulation of a systematic theory of such spaces that is the product of what Glick Schiller and her colleagues (1992, p. 5) identied as the desideratum of a social science unbound. Faist (1999, p. 15; Faist 2000a, p. 243) describes this as going beyond the container concept of space (see also, Beck 2000) that he thinks characterizes both assimilation theory and ethnic pluralism. Moreover, it can be seen as a parallel to what both Glick Schiller et al. and Portes describe, without much elaboration as social elds. While the intellectual lineage of social elds likely derives from Bourdieu, Faists use of the term space has been in uenced by the work of the Swedish school of time-geographers, particularly the work of Torsten Hgerstrand (1976). From this theoretical perspective space is not equivalent to place. Faist (2000a, pp. 4546) describes the difference in the following way: Space here does not only refer to physical features, but also to larger opportunity structures, the social life and the subjective images, values, and meanings that the speci c and limited place represents to migrants. Space is thus different from place in that it encompasses or

Theorizing transnational immigration 567 spans various territorial locations. It includes two or more places. Space has a social meaning that extends beyond simple territoriality; only with concrete social or symbolic ties does it gain meaning for potential migrants. Two things stand out that differentiate Faists position from that of Portes. First, he explicitly includes transnational immigrant communities in his framework and not simply individuals and families and secondly, he insists that such communities must be theoretically linked to what Portes describes as transnationalism from above, rather than treating transnationalisms from above and below as discrete parallel phenomena. Portes obviously sees transnationalism from above and transnationalism from below as interrelated phenomena, but his middlerange theory encourages, at the very least at the methodological level, an approach that does not lend itself to a fuller theoretical articulation of the nature of the interrelationship. Faist implicitly challenges this approach. In Faists earliest de nition, transnational social spaces were de ned as combinations of social and symbolic ties, positions in networks and organizations, and networks of organizations that can be found in at least two geographically and internationally distinct places (Faist 1998, p. 216). This differs little from subsequent formulations (Faist 1999, p. 4; Faist 2000a, p. 197). However, in identifying the component groups involved in these spaces, he amends his earlier characterization of a triadic relationship consisting of groups and institutions of the receiving nation, the homeland state, and the immigrant group itself to a pentagonic relationship. In this revised formulation, he separates the state from civil society in both the homeland and host society, thus adding four units of analysis to the immigrant group (Faist 1998, p. 217; Faist 2000a, p. 200). The social construction of enduring transnational social spaces requires the sustainability of various types of ties. Faist identi es three types of transnational social spaces, each with a characteristic tie. Kinship groups, he contends, are predicated on ties of reciprocity, as can be seen in the form of remittances. Transnational circuits, in contrast, require instrumental exchange ties such as those structuring trading networks. Finally, transnational communities are predicated on the solidarity derived from a shared conception of collective identity, and here, obviously, there is a clear parallel to ethnic communities located in one place rather than transnational space (Faist 2000a, pp. 202210). Under-theorized in this formulation is the affective element contained in each of these ties. Thus, while a norm of reciprocity may de ne the structured pattern of remittances sent to family members, this is not primarily an instrumental transaction, but one based on an emotional attachment to those left behind. It is a tie predicated on emotions related

568 Peter Kivisto to longing, the sorrow of absence, and the desire to remain bonded to ones family. As E.K. Francis (1947) pointed out long ago, while ethnic groups are secondary groups, they share with kinship groups collective attachments and interpersonal relationships that are emotionally charged. Moreover, kinship networks can appropriately be seen, not as distinct from the transnational ethnic community, but as a constituent part of it. But can the same affective character be said to characterize transnational trading circuits? Are these ties different: merely instrumental arrangements based on the pro t motive? Not when one considers the role that trust plays in economic exchange. Trust is an essential ingredient in combating economic risk, and in so far as ethnic af liations facilitate a requisite level of trust, they serve to underpin and make possible instrumental transactions. Once again, this would imply that transnational circuits are embedded in transnational communities, and can only be separated from them analytically. Thus, transnational immigrant social spaces require the creation of a new form of ethnic community. What makes it different from the more familiar form that typi ed immigrant enclaves in industrializing nations a century ago is that it is located in a space that encompasses two or more nation-states, a situation made possible by time-space compression. These are, in a well-selected phrase, communities without propinquity (Faist 2000a, p. 197). But, of course, in a fashion so were ethnic communities forged within the borders of particular nations. For example, European-origin immigrants to the United States forged a collective ethnic identity that linked fellow ethnics regardless of where in the US they lived. This type of collective identity was the product of an exit from the homeland and the migration to a particular place. In contrast, migrants in transnational spaces are seen as being engaged in a more uid and syncretistic process of adaptation. Metaphorically, assimilation is associated with the image of the uprooted (Handlin 1973), and cultural pluralism with the transplanted (Bodnar 1985). Faist (1998, p. 239) proposes as an appropriate transnational metaphorical alternative borrowing from the novelist Salman Rushdie the idea of translated people, writing that Migrants are continually engaged in translating languages, cultures, norms, and social and symbolic ties. In other words, transnational migrants forge their sense of identity and their community, not out of a loss or mere replication, but as something that is at once new and familiar a bricolage constructed of cultural elements from both the homeland and the receiving nation. Crucially for this thesis, it is not simply individuals living with one foot in two places that constitute the sole occupants of transnational communities. Rather, Faist (2000a, pp. 2078) writes: Transnational communities characterize situations in which international movers and dense and strong social and symbolic ties connect

Theorizing transnational immigration 569 stayers over time and across space to patterns of networks and circuits in two countries. . . . Such communities without propinquity do not necessarily require individual persons living in two worlds simultaneously or between cultures in a total global village of de-territorialized space. What is required, however, is that communities without propinquity link through exchange, reciprocity, and solidarity to achieve a high degree of social cohesion, and a common repertoire of symbolic and collective representations. Moreover, for a viable transnational community to be established and to sustain itself over time, a continual pattern of involvement with both governmental and civic institutions in the homeland and receiving country are essential (Faist 2000a, p. 208). This is due to the fact that transnational immigrants qua transnational immigrants are engaged in activities designed to de ne and enhance their position in the receiving nation, while simultaneously seeking to remain embedded in a participatory way in the everyday affairs of the homeland community. What is distinctive about this type of community, one assumes, is that over time the transnational social space thus carved out makes the dichotomous character of host society concerns versus homeland concerns if not irrelevant, at least less pronounced and at some level part of a transcendent structure of border-crossing social relations. Over time also means generational succession. The unanswerable question is whether or not transnational social spaces, to the extent that they have been created, will prove to be little more than a phenomenon describing the immigrant generation. Faist (2000a, p. 238), more forcefully than the previous theorists, puts on the table the matter of the durability of transnational ties with the passage of time, suggesting without elaborating, that we should not lose sight of the ever-present lures of cultural adaptation. . .. Unfortunately, he does not develop this point further. Faists analysis in conjunction with the two that preceded his is valuable in so far as it articulates, at least in general terms, the form, if not the content, of a potentially different mode of adjustment, acculturation and incorporation of immigrants into receiving nations compared to the heretofore more common mode in which with the passage of time immigrants sever their ties with the homeland. Faists insistence that transnationalism is not new, but can also be observed in the immigration wave that occurred between 1880 and 1920, is a well-taken corrective to the earlier arguments. Interestingly in this regard, in Glick Schillers (1999) recent revisiting of her earlier position, without explicitly engaging in self-critique, she too makes a sustained and compelling case for the existence of transnational immigration in this earlier period. Indeed, she rather pointedly challenges the claim, so prominent a feature of Portes thesis, that new technologies in communication and transportation

570 Peter Kivisto constitute the primary explanatory variable in accounting for differences between the present and past both in terms of the scope and depth of transnational linkages (Glick Schiller 1999, p. 114). In the end, one overriding question remains: have these three theoretical articulations of transnational immigration succeeded in constructing a model of immigrant adjustment that should be seen as a repudiation of or a complement to assimilation? In other words, have they, taken collectively, managed to make a compelling case for either the limited utility of assimilation theory or for its more general dismissal? The concluding section of this article suggests that they have not, and it proceeds to sketch out a case for incorporating transnationalism within the conceptual framework of assimilation theory. In addition, it will suggest some of the factors that may work for and against the social construction of enduring transnational ties. Towards a reconciliation of assimilation and transnationalism It has become a commonplace to preface any discussion of assimilation theory by noting that it has been the subject of such scathing criticism and repudiation that it has lost its heretofore-hegemonic status (Rumbaut 1999). This perception, for example, is seen in Richard Alba and Victor Nees (1997, p. 826) assertion that Assimilation has fallen into disrepute. It also led Nathan Glazer (1993) to pose the question, Is assimilation dead? While a fair characterization of the state of assimilation theory from the 1970s until the early 1990s, I believe it would be a mistake to conclude that it has been abandoned, and that the answer to Glazers question is yes. To the contrary, a convincing case can be made that during the past decade assimilation theory has made a remarkable comeback. Sociologists such as Alba and Nee (1997) and Gans (1979, 1992a, 1992b, 1999) along with historical sociologists such as Ewa Morawska (1994) and social historians such as Elliott Barkan (1995) have done much to effect the theorys rehabilitation, nding ways to criticize the primary alternatives to assimilation cultural pluralism and more recently multiculturalism while incorporating that which is valuable in those approaches into an assimilationist model. They have reexamined the classic statements on assimilation by such gures as Park and further developed Gordons (1964) seminal contribution. Without getting into the details, which are beyond the scope of this article, suf ce it to say that they do so by distinguishing assimilation from amalgamation and thus from the melting-pot model and by refusing to treat assimilation teleologically. This perspective accords particular importance to acculturation, and in so doing it makes possible the prospect that ethnic retention might persist over many generations, engaged in a bumpy-line dialectical relationship with acculturation (Gans 1992a, 1999). Another corrective to older versions of assimilation

Theorizing transnational immigration 571 theory involves an awareness that immigrants do not assimilate into a society that is xed and given, but rather one that is uid and subject to changes brought about by the presence of immigrants. One of the major shortcomings of earlier versions of assimilation theory was that they focused on the in uence of the host society on its immigrant populations, but failed to accord a similar amount of attention to the impact of immigration on the host societys cultural life and social institutions. New versions of assimilation as acculturation are prepared to redress this shortcoming. Finding these efforts compelling, I would like to propose that the relationship between transnationalism and assimilation ought to be seen in the same light as the relationship between assimilation and ethnic pluralism and multiculturalism. Rather than abandoning assimilation as the Glick Schiller et al. model proposes, or juxtaposing it to transnationalism as Faist does, it would appear most appropriate to consider transnationalism as one possible variant of assimilation. This is because at the moment that transnational immigrants are working to maintain homeland connections, they are also engaged in the process of acculturating to the host society. The social construction, or invention, of transnational social spaces involves major demands of time, resources, and energy, and unless the transnational community is composed of what Lewis Cosers (1974) has referred to as greedy institutions, the community competes with the civic society of the host society for them. Choices need to be made. Will immigrants opt to work on a water project back in their Mexican village of origin, or will they instead choose to participate in their neighbourhood block organization, the local PTA, or other organizations in the host society? Will they send remittances back home to their ageing relatives, or will they set money aside to pay for their childrens college education? Will political involvements focus on social reforms in the homeland or on urging the local city council to promote the enforcement of fair housing measures? These are the kinds of practical choices that immigrants are forced to make in their everyday lives. While these are not necessarily either/or choices, they do suggest the limitations to investing entirely in the transnational immigrant network. Two points are relevant here. First, place counts. Even in transnational social spaces, place continues to count. Contrary to the image of transnational immigrants living simultaneously in two worlds, in fact the vast majority is at any moment located primarily in one place. If the location where they spend most of their day-to-day lives is the receiving country, then over time the issues and concerns of that place will tend to take precedence over the more removed issues and concerns of the homeland. Political and economic crises in the homeland are stimuli for homelandfocused activities, but when those crises abate, immigrants will tend to focus their energies on the place where they nd themselves. Second, the

572 Peter Kivisto very act of participating in the life world of transnational communities can facilitate acculturation into the receiving country. Barbara Ballis Lal (1990) referred to this as the ethnicity paradox in her analysis of Parks understanding of the acculturative role that ethnic communities play in helping their members adjust and t into the host society. There is no reason to assume that this would be different for transnational communities. As Faist emphasized, the role of states is crucial in determining the fates of transnational communities. Of particular importance is their power in de ning the parameters of contemporary citizenship. Given the recent attention accorded to this topic, it is worth noting that over a quarter of a century ago Talcott Parsons pointed to the singular signi cance of citizenship in merging or uniting disparate ethnic groups without necessitating the eradication of particularistic identities. He wrote that, The most important new basis of inclusion in the societal community has been citizenship, developing in close association with the democratic revolution. Citizenship can be dissociated from ethnic membership, with its strong tendency towards nationalism and even racism, which provides a sharp ascriptive criterion of belonging (Parsons 1971, p. 92). The issues today, summarized very nicely by Christian Joppke (1999), focus on the extent to which states are willing to reconceptualize and modify inherited notions of citizenship, and on the changing role of the state in a world where transnational political institutions have acquired a heightened signi cance. Among the issues that need to be explored is the extent to which states will accept the idea of dual citizenship or will institutionalize new modes of political identity, such as that of the denizen (see, for example, Soysal 1994; 2000; see also Bourne 1916 for an early call for dual citizenship). These issues clearly speak to the ultimate viability of transnational social spaces, but at the moment it is dif cult to discern exactly where we are heading and what the implications of such changes might mean. In constructing a research agenda that is designed to assess the existence and the viability of transnational social spaces in the face of the inevitability of varying levels of acculturation, a number of issues need to be addressed. While not meant to be inclusive, among the factors that ought to be considered are the following: The costs associated with travel, which will vary depending on distance from the homeland and the socio-economic status of members of the ethnic group. Access to communication technologies particularly computers in both the receiving country and the homeland. The salience of homeland political issues or economic conditions versus the immediacy of similar concerns in the host society.

Theorizing transnational immigration 573 The role of discrimination in limiting access to the institutions of the receiving countrys civil society. The sustainability over time of involvement in institutions of the homelands civil society. The level of involvement over time in the institutions of the receiving countrys civil society. The impact of popular culture in hybridized forms on both sustaining transnational identities and on the process of acculturation. The role of nation-states both the homeland and receiving nation on migration policies and on de ning citizenship. Theorists of transnational immigration have attempted to provide a resource to help us better comprehend what Nikos Papastergiadis (2000) has called the turbulence of migration. Research will ultimately determine the extent to which transnational immigrant communities have actually developed out of contemporary South-North migrations. Likewise, research will determine whether transnationalism, to the extent that it actually exists, is only a phenomenon relevant to the immigrant generation or whether it will prove to be capable of persisting over time and across generations. Such research will be enhanced to the extent that it makes use of a theoretical perspective that locates the salience of transnational identities and communities within an assimilationist framework that is attentive to the lure of incorporation into the host society. Acknowledgements I would like to thank the following individuals for providing valuable comments on and criticisms of the article: Richard Alba, Herbert Gans, Thomas Faist, Stanford Lyman, Ewa Morawska and Alejandro Portes. Earlier versions were presented at seminars at the Institute for Intercultural Studies at the University of Bremen and the Race and Ethnicity Research Unit at South Bank University. I bene ted from the discussions at both venues, and thank my hosts, Thomas Faist in Bremen and Harry Goulbourne and John Solomos in London. References
ALBA, RICHARD D. and NEE, VICTOR 1997 Rethinking assimilation theory for a new era of immigration, International Migration Review, vol. 31 (Winter), pp. 82674 APPADURAI, ARJUN 1996 Modernity at Large: Cultural Dimensions of Globalization, Minneapolis: University of Minnesota Press BARKAN, ELLIOTT R. 1995 Race, religion and nationality in American society: a model of ethnicity from contact to assimilation, Journal of American Ethnic History, vol. 14, no. 2, pp. 3875 BASCH, LINDA, GLICK SCHILLER, NINA and SZANTON BLANC, CHRISTINA

574 Peter Kivisto

1994 Nations Unbound: Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nation-States, Basel, Switzerland: Gordon and Breach BAUMAN, ZYGMUNT 1998 Globalization: The Human Consequences, Oxford: Polity Press BECK, ULRICH 2000 What Is Globalization? Cambridge: Polity Press BODNAR, JOHN 1985 The Transplanted: A History of Immigrants in Urban America , Bloomington, IN: Indiana University Press BOURNE, RANDOLPH 1916 Trans-national America, The Atlantic Monthly, July, pp. 8697 CASTELLS, MANUEL 1996 The Rise of Network Society, vol. I of The Information Age: Economy, Culture, and Society, Oxford: Blackwell Publishers 1997 The Power of Identity, vol., II of The Information Age: Economy, Culture, and Society, Oxford: Blackwell Publishers CASTLES, STEPHEN and MILLER, MARK J. 1993 The Age of Migration: International Population Movements in the Modern World, New York: The Guilford Press CHEN, HSIANG-SHUI 1992 Chinatown No More: Taiwan Immigrants in Contemporary New York, Ithaca, NY: Cornell University Press COSER, LEWIS 1974 Greedy Institutions: Patterns of Undivided Commitment, New York: The Free Press DANIELS, ROGER 1990 Coming to America: A History of Immigration and Ethnicity in American Life, New York: HarperCollins DECONDE, ALEXANDER 1992 Ethnicity, Race, and American Foreign Policy: A History, Boston, MA: Northeastern University Press FAIST, THOMAS 1998 Transnational social spaces out of international migration: evolution, signi cance, and future prospects Arch. Europ. Sociol, vol. XXXIX, no. 2, pp. 21347 2000a The Volume and Dynamics of International Migration and Transnational Social Spaces, Oxford: Oxford University Press 2000b Social citizenship in the European Union. Institut fr Interkulturelle und Internationale Studien, Nr. 17/2000, Universitt Bremen 2000c Transnationalization in international migration: implications for the study of citizenship and culture, Ethnic and Racial Studies, vol. 23, no. 2, pp. 189222 FONG, TIMOTHY P. 2000 The rst suburban Chinatown: the remaking of Monterey Park, California, pp. 36980 in Peter Kivisto and Georganne Rundblad (eds), Multiculturalism in the United States: Current Issues, Contemporary Voices, Thousand Oaks, CA: Pine Forge Press FRANCIS, E. K. 1974 The nature of the ethnic group. American Journal of Sociology, vol. 52, no. 5, pp. 393400 FUCHS, LAWRENCE H. 1990 The American Kaleidoscope: Race, Ethnicity, and the Civic Culture, Hanover, NH: Wesleyan University Press 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 1992a Ethnic invention and acculturation: a bumpy-line approach. Journal of American Ethnic History, vol. 12, no. 1, pp. 4552 1992b Second-generation decline: scenarios for the economic and ethnic futures of the post-1965 American immigrants. Ethnic and Racial Studies, vol. 15, no. 2, pp. 17392 1999 Toward a reconciliation of assimilation and pluralism: the interplay of acculturation and ethnic retention. pp. 16171 in Charles Hirschman, Philip Kasinitz, and Josh DeWind (eds), The Handbook of International Migration: The American Experience, New York: Russell Sage Foundation GIDDENS, ANTHONY 1984 The Constitution of Society: Outline of the Theory of Structuration, Berkeley: University of California Press 1990 The Consequences of Modernity, Cambridge: Polity Press 1999 Globalization, Reith Lectures, BBC Online Network (http: //news.bbc.co. uk/reith99/week1.htm)

Theorizing transnational immigration 575

GLAZER, NATHAN 1993 Is assimilation dead?, Annals of the American Academy of Political and Social Science, 530, pp. 12236 GLICK SCHILLER, NINA 1997 The situation of transnational studies, Identities, vol. 4, no. 2, pp. 15566 1999 Transmigrants and nation-states: something old and something new in the U.S. immigrant experience, pp. 94119 in Charles Hirschman, Philip Kasinitz, and Josh DeWind (eds), The Handbook of International Migration: The American Experience, New York: Russell Sage Foundation GLICK SCHILLER, NINA, BASCH, LINDA and SZANTON BLANC, CHRISTINA 1992 Transnationalism: a new analytic framework for understanding migration, pp. 124 in Glick Schiller, Nina, Basch, Linda, and Szanton Blanc, Christina (eds), Towards a Transnational Perspective on Migration: Race, Class, Ethnicity, and Nationalism Reconsidered, New York: New York Academy of Sciences 1995 From immigrant to transmigrant: theorizing transnational migration, Anthropological Quarterly, vol. 68, no. 1, pp. 4863 GORDON, MILTON 1964 Assimilation in American Life: The Role of Race, Religion, and National Origins, New York: Oxford University Press GUARNIZO, LUIS E. 1997 The emergence of a transnational social formation and the mirage of return migration among Dominican transmigrants, Identities, vol. 4, no. 2, pp. 281322 HGERSTRAND, THORSTEN 1976 Innovation as a Spatial Process, Chicago: University of Chicago Press HANDLIN, OSCAR 1973 The Uprooted, Boston, MA: Little, Brown, and Company HANNERZ, ULF 1996 Transnational Connections: Culture, People, Places, London: Routledge HARVEY, DAVID 1989 The Condition of Postmodernity, Oxford: Blackwell Publishers JONES, DELMOS 1992 Comments on rethinking migration: a transnational perspective pp. 221225 in Glick Schiller, Nina, Basch, Linda and Szantoun-Blanc, Christina (eds), Towards a Transnational Perspective on Migration: Race, Class, Ethnicity and Nationalism Reconsidered, New York: NY Academy of Sciences JONES, MALDWYN 1976 The Old World Ties of American Ethnic Groups, London: Macmillan JOPPKE, CHRISTIAN 1999 How immigration is changing citizenship: a comparative view, Ethnic and Racial Studies, vol. 22, no. 4, pp. 62952 KIVISTO, PETER 1984 Immigrant Socialists in the United States: The Case of Finns and the Left, Rutherford, NJ: Fairleigh Dickinson University Press 1990 The transplanted then and now: the reorientation of immigration studies from the Chicago School to the new social history, Ethnic and Racial Studies, vol. 13, no. 4, pp. 45581 LAL, BARBARA BALLIS 1990 The Romance of Culture in an Urban Civilization: Robert E. Park on Race and Ethnic Relations in Cities, London: Routledge LIPSET, SEYMOUR MARTIN and GARY MARKS 2000 It Didnt Happen Here: Why Socialism Failed in the United States. New York: W.W. Norton LYMAN, STANFORD M. 1974 Chinese Americans, New York: Random House MARX, KARL 1973 {1859} Grundrisse: Foundations of the Critique of Political Economy, New York: Vintage Books MERTON, ROBERT K. 1987 Three fragments from a sociologists notebook: establishing the phenomena, speci ed ignorance, and strategic research materials, Annual Review of Sociology, vol. 13, pp. 128 MILLER, SALLY 1974 The Radical Immigrant, New York: Twayne MORAWSKA, EWA 1994 In defense of the assimilation model, Journal of American Ethnic History, vol. 13, no. 2, pp. 7687 1999 The Sociology and History of Immigration: Re ections of a Practitioner,

576 Peter Kivisto

European University Institute workshop on: Re ections on Migration Research, Florence, May 201 Forthcoming The New-Old Immigrants, Their Transnational Lives, and Ethnicization: A Comparison of 19th/20th and 20th/21st C. Situations, in Gary Gerstle and John Mollenhoff, (eds) Immigrants, Civic Culture, and Modes of Political Incorporation, New York: Russell Sage Foundation PAPASTERGIADIS, NIKOS 2000 The Turbulence of Migration, Cambridge: Polity Press PARK, ROBERT EZRA 1950 Race and Culture: Essays in the Sociology of Contemporary Man, New York: The Free Press PARSONS, TALCOTT 1971 The System of Modern Societies, Englewood Cliffs, NJ: Prentice-Hall PORTES, ALEJANDRO 1995 Children of immigrants: segmented assimilation, pp. 24880 in Alejandro Portes (ed), The Economic Sociology of Immigration, New York: Russell Sage Foundation 1996a Global villagers: the rise of transnational communities, American Prospect, no. 25, pp. 747 1996b Transnational communities: their emergence and signi cance in the contemporary world-system, pp. 15168 in Roberto Patricio Korzeniewicz and William C. Smith (eds), Latin America in the World-Economy, Westport, CT: Greenwood Press 1998 Divergent destinies: immigration, the second generation, and the rise of transnational communities, pp. 3357 in Peter H. Schuck and Rainer Munz (eds), Paths to Inclusion: The Integration of Migrants in the United States and Germany, New York: Berghahn Books 1999a Conclusion: toward a new world the origins and effects of transnational activities, Ethnic and Racial Studies, vol. 22, no. 2, pp. 46377 1999b Immigration theory for a new century: some problems and opportunities, pp. 2133 in Charles Hirschman, Philip Kasinitz, and Josh DeWind (eds), The Handbook of International Migration: The American Experience, New York: Russell Sage Foundation PORTES, ALEJANDRO, GUARNIZO, LUIS E. and LANDOLT, PATRICIA 1999 The study of transnationalism: pitfalls and promise of an emergent research eld. Ethnic and Racial Studies, vol. 22, no. 2, pp. 21737 PORTES, ALEJANDRO and RUMBAUT, RUBN G. 1990 Immigrant America: A Portrait, Berkeley, CA: University of California Press PORTES, ALEJANDRO and ZHOU, MIN 1999 Entrepreneurship and economic progress in the 1990s: a comparative analysis of immigrants and African Americans, pp. 14371 in Frank D. Bean and Stephanie Bell-Rose (eds), Immigration and Opportunity: Race, Ethnicity, and Employment in the United States, New York: Russell Sage Foundation ROBERTS, BRYAN 1995 Socially expected durations and the economic adjustment of immigrants, pp. 4286 in Alejandro Portes (ed.), The Economic Sociology of Immigration, New York: Russell Sage Foundation ROBERTSON, ROLAND 1992 Globalization: Social Theory and Global Culture, London: Sage ROSENBAUM, GERALD 1973 Immigrant Workers: Their Impact on American Radicalism, New York: Basic Books RUMBAUT, RUBN 1999 Assimilation and its discontents: ironies and paradoxes, pp. 17295 in Charles Hirschman, Philip Kasinitz, and Josh DeWind (eds), Handbook of International Migration: The American Experience, New York: Russell Sage Foundation SASSEN, SASKIA 1996 Losing Control? Sovereignty in an Age of Globalization, New York: Columbia University Press SKLAIR, LESLIE 2001 The Transnational Capitalist Class, Oxford: Blackwell Publishers SMITH, MICHAEL P. and GUARNIZO, LUIS E. (eds) 1998, Transnationalism from Below, New Brunswick, NJ: Transaction Publishers SOYSAL, YASEMIN N. 1994 Limits of Citizenship: Migrants and Postnational Membership in Europe, Chicago, IL: University of Chicago Press

Theorizing transnational immigration 577

2000 Citizenship and identity: living in diasporas in post-war Europe? Ethnic and Racial Studies, vol. 23, no. 1, pp. 115 TAKAKI, RONALD 1989 Strangers from a Different Shore: A History of Asian Americans, Boston, MA: Little, Brown and Company 1993 A Different Mirror: A History of Multicultural America, Boston, MA: Little, Brown and Company URRY, JOHN 2000 Sociology Beyond Societies: Mobilities for the Twenty-First Century, London: Routledge VERTOVEC, STEVEN 1999 Conceiving and researching transnationalism, Ethnic and Racial Studies, vol. 22, no. 2, pp. 44762 WATERS, MALCOLM 1995 Globalization, London: Routledge

PETER KIVISTO is Professor of Sociology at Augustana College, Rock Island, IL. ADDRESS: Department of Sociology, Augustana College, Rock Island, IL 61201, USA. email: sokivisto@augustana.edu