Ethnic boundaries and identity in plural societies.

INTRODUCTION

Ethnic boundaries are patterns of social interaction that give rise to, and subsequently reinforce, in-group members' self-identification and outsiders' confirmation of group distinctions. Ethnic boundaries are therefore better understood as social mediums through which association transpires rather than as territorial demarcations. Various ways of defining ethnic groups are found in the literature, but most emphasize cultural and geographical elements. The first of these elements is usually viewed as a social construction involving insiders and outsiders mutually acknowledging group differences in cultural beliefs and practices. Insiders and outsiders do not necessarily agree over the details of the acknowledged cultural division, nor do groups necessarily develop similar interpretations of the relative merits of the various distinguishing cultural attributes. The second basic element used to define an ethnic group pertains to geographical origins, and therefore social origins, that are foreign to the host s ociety. While this element usually has an objective basis, it is also partly subjective. The native-born generations of an ethnic group sometimes continue to be identified by outsiders, and in-group members may self-identify, in terms of their foreign origin. The ways in which insiders and outsiders go about characterizing a group, and thereby positioning it and its members in the larger society, are responsive to the social and historical context within which intergroup interactions take place (Nagel 1994, Waters 1990). Consequently, ethnic identities are fluid across time and social contexts, sometimes even to the point of "ethnic switching" (Alba 1990, Nagel 1995). The public presentation of ethnic identity is also situational, which reveals the plural or hybrid character of modem ethnicity (Espiritu 1992, Lessinger 1995). Howard's (2000) recent review of the social psychology of identity includes a helpful section on ethnic identity.

The situational and subjective aspects of ethnic identity mean that researchers who wish to understand how ethnicity emerges as an important factor in a range of social processes must do more than identify key cultural and behavioral components of groups. Researchers must also investigate patterns of interaction that link groups. The locations of cross-group interactions are usually better understood in terms of social space than as physical places. The social spaces wherein cross-group interactions take place are the effective social boundaries between groups. In this sense, it is "the ethnic boundary that defines the group, not the cultural stuff that it encloses" (Barth 1969, pp. 15). According to Barth, studying ethnic groups only in terms of their cultural traits and institutional forms leads researchers to confound the effects of cultural tradition with how ecological circumstances lead to changes in patterns of belief and behavior. The cultural traits of an ethnic group respond to ecological circumstan ces; therefore, forms of institutionalized behavior emerge that represent reactions to the environment as much as they reflect a cultural orientation. For example, the set of relationships that have traditionally defined a Vietnamese family has been modified as a response to the dissolution of refugee families. This adaptive redefinition of the family has been accompanied by transformations in traditional gender- and age-based hierarchies of power (Gold 1992, Kibria 1993). Similar transformations have emerged for groups as diverse as the Sudanese (Holtzman 2000), Koreans (Min 1998), Dominicans (Pessar 1995), and Southeast Asians (Camino & Krulfeld 1994).

Ethnic distinctions sometimes coincide with territorial segregation in the host society and with social constructions of racial identity. Widely acknowledged racial differences can sharpen in-group members' self-identification and out-group acknowledgment of intergroup distinctions. Similarly, when interaction between groups is limited and otherwise conditioned by territorial segregation, intergroup differences gain emphasis. Constraints on cross-group interaction contribute to the respective groups ignorance of one another. This, in turn, encourages stereotyping. Race and the segregating tendencies of territorial concentrations are not necessarily components of ethnic boundaries, but when one or both of these elements of social organization obtain, they can play important roles in the maintenance of ethnic boundaries.

Emphasis on ethnic boundaries at the cost of giving less attention to the cultural content within those boundaries is an example of a good idea pushed too far, according to Cornell (1996). He critiques two lines of research for their failure to adequately address the cultural content of what defines an ethnic group. One set of studies concentrates on the situational contexts that "construct and give significance to ethnic boundaries, and thereby give logic to ethnic group formation and persistence" (Cornell 1996, pp. 266). This approach, in Cornell's (1996, pp. 267) view, "grants to ethnic identity itself little meaning ... and makes it difficult to account for the tenacity with which some groups cling to identities." He goes on to argue, "While circumstances construct identities, identities, via the actions they set in motion, are also capable of reconstructing circumstances." The second line of research critiqued by Cornell concentrates on the responses of groups to the situational context. Although these s tudies pay attention to the cultural content of ethnic groups, and how in-group members construct and reconstruct their own identity, ethnicity is treated as though driven by factors external to the group. Cornell accepts the importance of ethnic boundaries in defining ethnic groups, but he cautions the field against giving short shrift to what goes on within those boundaries--the shared cultural content of ethnicity.

Plan and Scope of the Review

The literature reviewed below reveals that scholars give a great deal of attention to ethnic boundaries, and to the cultural content within those boundaries, as they study ethnicity in plural societies. In so doing, scholars place emphasis on ethnic networks and social capital derived from such networks. I briefly examine studies from the mid twentieth century, but the bulk of the review samples more recent contributions to the literature. Due to the foreign origin dimension of ethnicity, most of the literature reviewed considers immigrants and their offspring. The main lines of inquiry found in contemporary studies, the economic action of adults and youth socialization and education, receive the greatest attention in this review. However, I also attend to other important areas of study including ethnic churches, population concentrations, and transnationalism. I distinguish race from ethnicity, but racial boundaries and identity sometimes overlap with ethnic boundaries and identity, and this is reflected in the review. Although the literature has a strong international component, the lion's share of the work focuses on the United States. The review is restricted to work available in English.

SOCIAL NETWORKS AND SOCIAL CAPITAL

Social networks and the social capital derived from them are central to the study of ethnicity in plural societies. The importance of these social forces is documented by studies of transnational networks that encourage labor migration (Palloni et al. 2001). These networks are shaped by characteristics of the immigrant stream and by structural conditions in the host society (Grieco 1998, Massey & Espinosa 1997). Adaptive social networks that emerge in the host society exert strong influences on the labor-market experiences of adults and on parental and community efforts to facilitate the success of the next generation in the host society (Waldinger 1996, Zhou & Bankston 1998).

Ethnic networks, like other forms of social organization, must protect themselves from internal corruption and incompetence. The more closed a network, the greater the chance that effective forms of monitoring and sanctioning will develop (Coleman 1990). Much of the sociological study of closure originates from Weber ([1922] 1978), but it is contemporary scholars (Murphy 1988, Parkin 1974) who have advanced the study of closure to include nonsubordination processes (e.g., a minority group is able to monopolize opportunities within a limited part of the labor market). What Parkin calls "solidaristic closure" involves social relations with underpinnings of ethnic solidarity that generate and channel opportunities to a cross-section of the group. Such networks are effective in providing opportunities to group members, but a high degree of closure may limit their capacity to broadly serve the ethnic community. Powerful ethnic networks are effective in expanding the employment opportunities of group members throug hout a metropolitan labor market (Light et al. 1994, Nee & Sanders 200 la). Therefore, network closure and power are not the same thing. Given the limited ability of ethnic entrepreneurs to provide employment opportunities (Light & Gold 2000), the economic fortunes of immigrant groups must include participation in the larger metropolitan economy. Consequently, immigrants are dispersed throughout metropolitan labor markets (Logan et al. 2000). Nee & Sanders (2001b) have developed a multiple-forms-of capital conceptualization of the immigrant incorporation process in an effort to explain this diversity of employment. Their approach is sensitive to class origins and transferred (or transferable) assets, human and cultural capital acquired in the sending and receiving societies, and social capital derived from social networks. Differences in the labor-force experiences of men and women are an important source of this diversity (Greenlees & Saenz 1999, Zhou 1992). Gender differences occur at the high end (Choy 200 0) and the low end (Kaufman 2000) of the labor market.

Sociologists have found the concept of social capital to be useful in explicating how ethnic-based forms of social organization and collective action are embedded in interpersonal networks, and how these forms of organization and action generate and distribute resources. In a recent review piece, Portes (1998) traces the rise of the terminology of social capital. He points out that the most fundamental social capital thesis--taking part in the organizational forms and collective actions of a group can lead to both positive and negative consequences for the individual participants and for the group as a whole--is found in the work of nineteenth century scholars who provided the intellectual foundations upon which sociology is based. Portes & Sensenbrenner (1993), covering much of the ground included in Portes (1998), made a major contribution in identifying two sources or mechanisms of social capital that are especially relevant to the issues of ethnic boundaries and identity. One of the concepts refers to an in-group oriented, principled form of behavior termed "bounded solidarity." This mechanism for generating social capital involves a sense of group solidarity that is most likely to manifest as a reaction to real or perceived threats to the group. The other mechanism, "enforceable trust," refers to the monitoring and sanctioning capacity ofa group. As Portes & Sensenbrenner (1993, pp. 1332) point out with the oxymoron "trust exists in economic transactions precisely because it is enforceable by means that transcend the individuals involved," enforceable trust has little to do with trust in any altruistic sense, but it has a lot to do with the capacity to enforce group norms. Mechanisms that reduce the free-rider problem (Olson 1965) are essential for effective and sustained collective action.

The tight connection between social capital and social networks makes another recent review article relevant. Lin (1999) reviews the social-resources literature, which usually applies the methodologies of network analysis. Like social-capital theory, social-resources theory focuses on resources derived from, and disseminated through, social networks. Lin observes that the concept of social capital is usually used to frame theoretical arguments, whereas social resources are used to operationally define the "goods" that social networks provide.

THE EXPERIENCES OF THE EUROPEAN FOREIGN STOCK: MID-TWENTIETH CENTURY CONCLUSIONS

Approximately four decades ago, on the eve of a profound shift in the immigration policies of the United States, the repercussions of which scholars did not anticipate (GIazer 2000), the literature was concentrating on the cultural, social, and economic circumstances of groups populated by two or more generations born and raised in the host society. Since most of this research focused on the United States, moreover, the quota system of immigration restrictions in place between the mid 1920s and mid 1960s meant that the foreign-born contingent of most ethnic groups was not being renewed. The literature concluded that important factors slowing the economic advancement and fuller social acceptance of ethnic groups included persisting cultural pluralism despite an intergenerational tendency toward greater acculturation, institutionalized restrictions that had stratifying consequences, and the informal exercise of everyday normative practices that discouraged the entry of ethnic minorities into the primary social circles of the dominant group.

While slow acquisition of social skills that come with acculturation was seen as one of the serious problems faced by immigrants, and sometimes by their children, the main problems encountered by later generations were seen as having little to do with acculturation because for them, the process was largely complete. The primary problem facing acculturated generations was seen as lagging assimilation. This lack of assimilation was usually viewed as more a function of the ethnic group being denied full acceptance by the dominant group than as a reluctance of ethnic groups to become assimilated.

By the mid 1960s, several important sociological works dealing with ethnic groups and their social boundaries and identity had emerged. Much of this literature is reviewed by Hirschman (1983) and Yinger (1985), and therefore I limit my treatment of the earlier literature. Yet a brief look at a few of the important works is helpful for conveying the state of the literature at that time.

Scholars had generally come to the conclusion that acculturation (or adaptation) tended to obtain in two or three generations, whereas assimilation lagged and thus ethnic pluralism persisted. At most, the literature offered only qualified support for the evolutionary natural history assimilation arguments forwarded by Park (1926, pp. 196), which are often summarized in the immigrant-adaptation and race-relations cycle of "contacts, competition, accommodation, and eventual assimilation." The qualifier of "eventual" assimilation has to be taken seriously, as the earlier generation of scholars documented. Perhaps it is this long-term dimension of assimilation theory, in conjunction with its ethnocentric tone, that engenders dissatisfaction from scholars. Certainly, it can be frustrating to work in an area wherein falsifying one of the major theoretical positions is difficult because of the temporal nature of the argument. Indeed, Lyman (1968) argues that the lack of specificity in matters of timing renders assim ilation theory untestable.

In the 1950s, Otis Dudley Duncan and his colleagues (Duncan & Duncan 1955, Duncan & Lieberson 1959) confirmed an …