Ethnicity without Groups
ROGERS BRUBAKER
HARVARD UNIVERSITY PRESS Cambridge, Massachusetts, and London, England
Introduction
The essays collected in this volume address a cluster of closely related themes: ethnicity, race, nationalism, ethnic violence, identity, collective memory, migration, assimilation, and the nation-state. These are issues that, in one form or another, have preoccupied me for nearly two decades. Yet the present essays—written in analytical counterpoint to sustained ethnographic research, and in critical engagement with contemporary theoretical debates—mark a new direction in my work.
My earlier work on immigration, citizenship, and the nation-state in France and Germany (1992) and on the interplay between nationalizing states, national minorities, and external national homelands in eastern Europe (1996) involved what Charles Tilly (1984) has called "big structures, large processes, [and] huge comparisons." The latter project, for example, embraced three sprawling multinational empires— Habsburg, Ottoman, and Romanov—and their successor states, two of which—the Soviet Union and Yugoslavia—were themselves multinational (and on some accounts imperial) states, whose disintegration yielded another score of successor states. In the course of this work, I began to experience a tension between my emerging theoretical commitments, which pointed increasingly in a microanalytic direction, and the scale and scope of my empirical work.
In response to this tension, I began in the summer of 1995 to conduct fleldwork in the ethnically mixed Transylvanian town of Cluj,
2 ■ Introduction
Introduction • 3
whose flamboyantly nationalist Romanian mayor had earned considerable notoriety for his anti-Hungarian pronouncements and his unflagging crusade to nationalize the town's public space. Together with Jon Fox, Margit Feischmidt, and Liana Grancea, I have been studying the meanings, workings, and variable salience of ethnicity (or "nationality," as it is called in the region) in everyday life, and the ways in which such everyday ethnicity is both affected by and insulated from nationalist politics on local, statewide, and interstate levels (Brubaker et al. 2004).
Working in Cluj, I have come to appreciate the force of Eric Hobs-bawm's (1990: 10) dictum that nationhood and nationalism, while constructed from above, "cannot be understood unless also analysed from below, that is in terms of the assumptions, hopes, needs, longings and interests of ordinary people, which are not necessarily national and still less nationalist." Studying the everyday preoccupations of ordinary Clujeni—to which ethnicity is indeed largely irrelevant—helped make sense of certain puzzles: in particular the lack of popular mobilization in response to, and the considerable popular indifference in the face of, intense and intractable elite-level nationalist conflict.
Yet this nonresponsiveness to the appeals of ethnonational entrepreneurs does not mean that ethnicity is experientially insignificant in Cluj. Social life is pervasively, though unevenly, structured along ethnic lines, and ethnicity "happens" in a variety of everyday settings. Ethnicity is embodied and expressed not only in political projects and nationalist rhetoric but in everyday encounters, practical categories, commonsense knowledge, cultural idioms, cognitive schemas, interactional cues, discursive frames, organizational routines, social networks, and institutional forms. Such everyday ethnicity—like what Michael Billig (1995) has called "banal nationalism"—may be invisible to the student of collective action or ethnic violence, but it merits study in its own right.
Although they do not, with one exception, directly engage my work in Cluj,1 the essays in this volume were prompted in part by this shift to a smaller scale of empirical research and the concomitant concern with everyday ethnicity. They were also occasioned by dissatisfaction with prevailing analytical idioms. As critical contributions, the essays share two main targets. The first is what I call "groupism": the tendency to take bounded groups as fundamental units of analysis (and basic constituents of the social world). Grounded in what Pierre Bour-dieu called "our primary inclination to think the social world in a
substantialist manner" (Bourdieu and Wacquant 1992: 228), this tendency has proved surprisingly robust. It has managed to withstand a quarter century of constructivist theorizing in the social sciences, a sustained critique of reification in anthropology and other disciplines, the influential and destabilizing contributions of feminist, post-structuralist, post-modernist, and other theories, and even the widespread acknowledgment, in principle, that "cultures," "communities," "tribes," "races," "nations," and "ethnic groups" are not bounded wholes. Despite these and other developments, ethnic and other groups continue to be conceived as entities and cast as actors.
The second target is complacent and cliched constructivism. Social construction has been a fertile metaphor in recent decades, inspiring a large body of work that has enriched and transformed our understanding of ethnicity (and of many other phenomena). Yet by virtue of its very success, the constructivist idiom has grown "weary, stale, flat, and unprofitable." Once an insurgent undertaking, a bracing challenge to entrenched ways of seeing, constructivism has become the epitome of academic respectability, even orthodoxy. It is not that the notion of social construction is wrong; it is rather that it is today too obviously right, too familiar, too readily taken for granted, to generate the friction, force, and freshness needed to push arguments further and generate new insights. One symptom of this intellectual slackness is that one often finds constructivist and groupist language casually conjoined.
The essays seek to develop ways of analyzing ethnicity without invoking bounded groups, and to do so in a manner that gives the constructivist project renewed analytical purchase. They share a commitment to disaggregated modes of analysis, but this does not entail an ontological or methodological individualism.2 The alternative to the substantialist idiom of bounded groups is not an idiom of individual choice, but rather (as Bourdieu never tired of emphasizing) a relational, processual, and dynamic analytical language.
The eponymous Chapter 1 sets the agenda for the book by developing the critique of groupism and proposing ways of studying "ethnicity without groups." The title should not be taken too literally. The book does not seek to banish "groups" from the study of ethnicity; it seeks, rather, to open up that study to other ways in which ethnicity "works." Bounded and solidary groups are one modality of ethnicity (and of social organization more generally). But they are only one modality.
4 ' Introduction
Introduction • 5
"Groupness" is a variable, not a constant; it cannot be presupposed. It varies not only across putative groups, but within them; it may wax and wane over time, peaking during exceptional—but unsustainable— moments of collective effervescence. Ethnicity does not require such groupness. It works not only, or even especially, in and through bounded groups, but in and through categories, schemas, encounters, identifications, languages, stories, institutions, organizations, networks, and events. The study of ethnicity—even the study of ethnic conflict— should not, in short, be reduced to, or even centered on, the study of ethnic groups. Notwithstanding the title of Horowitz's magisterial (and still indispensable) book, ethnic conflict is not always a matter of "ethnic groups in conflict."
Chapter 2, written with Frederick Cooper, takes on that central, indeed inescapable term in the social sciences and humanities: identity. This is a term that has been asked to do a great deal of analytical work, and much of that work is legitimate and important. But it is also a deeply ambiguous term, divided between "hard" and "soft" meanings, between groupist assumptions and constructivist qualifiers, between connotations of unity and multiplicity, sameness and difference, permanence and change. Understood in a strong sense—as implying a singular, abiding, foundational sameness—"identify" tends to mean too much; understood in a weak sense—as multiple, fluid, fragmented, negotiated, and so on—it tends to mean too little. This essay argues that the work done by "identity" might better be done by several clusters of less congested terms: identification and categorization, self-understanding and social location, commonality and connectedness.
One of the most significant intellectual developments of the latter part of the twentieth century in the human sciences was the "cognitive turn," which revolutionized psychology, recast debates in linguistics, created a new subdiscipline in anthropology, and founded entirely new fields such as artificial intelligence and cognitive science. In sociology, too, cognitive perspectives have opened up new lines of analysis. In the study of ethnicity, however, the cognitive turn has remained incipient and largely implicit. Chapter 3, written with Mara Loveman and Peter Stamatov, suggests ways of consolidating and extending this incipient cognitive turn by drawing on cognitive research in social psychology and anthropology. Cognitive perspectives provide resources for conceptualizing ethnicity, race, and nation in a non-groupist manner, as perspectives on the world rather than entities in the world, while at
the same time helping to explain the tenacious hold of groupist ways of thinking in practice.
Ethnic and nationalist violence has become a major focus of public and scholarly concern since the end of the Cold War. Yet what is often rather too casually called "ethnic violence" is not a unitary phenomenon; and any attempt to construct a unitary theory of ethnic violence would be vitiated by its lack of a meaningful explanandum. Chapter 4, written with David Laitin, reviews the burgeoning literature on the subject and proposes to take account of the composite and causally heterogeneous texture of the phenomenon through a strategy of analytical disaggregation.
The last third of the twentieth century witnessed a massive differen-tialist turn in ways of conceptualizing and responding politically to cultural heterogeneity in Western democracies. This movement of social thought and public policy resulted in greatly increased sensitivity to, appreciation for, and encouragement of "difference" in a variety of domains. Chapter 5 suggests that this movement may have reached its peak, and that, in the domain of immigration at least, prevailing understandings of a linear move "beyond assimilation" require rethinking. Examining public discourse in France, public policy in Germany, and scholarly research in the United States, I find evidence of a modest "return of assimilation" in recent years. Yet what has "returned," I emphasize, is not the old, analytically discredited and politically disreputable "assimilationist" understanding of assimilation, but a more analytically complex and normatively defensible understanding.
For more than a century, scholars and public figures have distinguished "civic" and "ethnic," western and eastern, liberal and illiberal forms of nationalism. These and similar distinctions have provided a way of coming to terms with the empirically unruly and morally and politically Janus-faced phenomena of nationhood and nationalism. But nationalism resists easy parsing into types with clearly contrasting and neatly correlated empirical and moral profiles. The distinction between civic and ethnic nationalism, I argue in Chapter 6, is conceptually ambiguous, empirically misleading, and normatively problematic.
Chapter 7 examines the politically charged questions of ethnicity, migration, and statehood in post-Cold War Europe. In Western Europe, ethnic heterogeneity generated by immigration is often seen as one of many expressions of a fundamental move "beyond the nation-state" or, alternatively, as a catalyst of reactive, exclusionary
6 • Introduction
nationalism. In Eastern Europe, ethnicity, migration, and statehood are seen as linked in a more ominous way, through violent conflict and ethnic cleansing. This chapter seeks to provide a more nuanced account of these closely intertwined issues, sensitive to persisting regional differences yet avoiding an oversimplified East-West contrast.
The volume concludes with a comparative study of the sesquicenten-nial commemoration of the revolutions of 1848 in Hungary, Romania, and Slovakia, written with Margit Feischmidt. Constructivist studies of collective memory emphasize the malleability and manipulability of the past at the hands of contemporary cultural and political entrepreneurs. This essay too finds ample evidence of the ways in which the politics of the present shape the representation of the past. Yet at the same time, the differential resonance of official commemorative efforts in the three countries underscores the point that the past is also refrac tory to presentist reconstruction.
CHAPTER ONE
Ethnicity without Groups
Commonsense Groupism
Few social science concepts would seem as basic, even indispensable, as that of group. In disciplinary terms, \"group" would appear to be a core concept for sociology, political science, anthropology, demography, and social psychology. In substantive terms, it would seem to be fundamental to the study of political mobilization, cultural identity, economic interests, social class, status groups, collective action, kinship, gender, religion, ethnicity, race, multiculturalism, and minorities of every kind.
Yet despite this seeming centrality, the concept "group" has remained curiously unscrutinized in recent years. There is, to be sure, a substantial social psychological literature addressing the concept (Hamilton et al. 1998; McGrath 1984), but this has had little resonance outside that subdiscipline. Elsewhere in the social sciences, the recent literature addressing the concept "group" is sparse, especially by comparison with the immense literature on such concepts as class, identity, gender, ethnicity, or multiculturalism—topics in which the concept "group" is implicated, yet seldom analyzed on its own terms.1 "Group" functions as a seemingly unproblematic, taken-for-granted concept, apparently in no need of particular scrutiny or explication. As a result, we tend to take for granted not only the concept "group," but also "groups"—the putative things-in-the-world to which the concept refers.
My aim here is not to enter into conceptual or definitional casuistry. It is rather to address one problematic consequence of the tendency to
8 • Ethnicity without Groups
Ethnicity without Groups • 9
take groups for granted in the study of ethnicity, race, and nationhood, and in the study of ethnic, racial, and national conflict in particular. This is what I will call "groupism," by which I mean the tendency to take discrete, bounded groups as basic constituents of social life, chief protagonists of social conflicts, and fundamental units of social analysis.21 mean the tendency to treat ethnic groups, nations, and races as substantial entities to which interests and agency can be attributed. I mean the tendency to reify such groups, speaking of Serbs, Croats, Muslims, and Albanians in the former Yugoslavia, of Catholics and Protestants in Northern Ireland, of Jews and Palestinians in Israel and the occupied territories, of Turks and Kurds in Turkey, or of Blacks, Whites, Asians, Hispanics, and Native Americans in the United States as if they were internally homogeneous, externally bounded groups, even unitary collective actors with common purposes. I mean the tendency to represent the social and cultural world as a multichrome mosaic of monochrome ethnic, racial, or cultural blocs.
From the perspective of broader developments in social theory, the persisting strength of such groupism is surprising. After all, several distinct traditions of social analysis have challenged the treatment of groups as real, substantial things-in-the-world. These include not only individualistic approaches such as rational choice, game theory, and agent-based modeling, but also network theory, cognitive theory, feminist theory, and densely relational micro-interactionist approaches such as ethnomethodology and conversation analysis. More generally, many constructivist stances treat groups as constructed, contingent, and fluctuating, while a diffuse post-modernist sensibility emphasizes the fragmentary, the ephemeral, and the erosion of fixed forms and clear boundaries. These developments are disparate, even contradictory in analytical style, methodological orientation, and epistemological commitments. Network theory, with its methodological (and sometimes ontological) relationalism (Emirbayer and Goodwin 1994; Wellman 1988) is opposed to rational choice theory, with its methodological (and sometimes ontological) individualism; both are sharply and similarly opposed, in analytical style and epistemological commitments, to post-modernist approaches. Yet these and other developments have converged in problematizing groupness and undermining axioms of stable group being.
Challenges to "groupism," however, have been uneven. They have been striking—to take just one example—in the study of class,
especially in the study of the working class, a term that is hard to use today without quotation marks or some other distancing device. Yet ethnic groups continue to be understood as entities and cast as actors. To be sure, constructivist approaches of one kind or another are now dominant in academic discussions of ethnicity. Yet everyday talk, policy analysis, media reports, and even much ostensibly constructivist academic writing routinely frame accounts of ethnic, racial, and national conflict in groupist terms as the struggles "of" ethnic groups, races, and nations.3 Somehow, when we talk about ethnicity, and even more when we talk about ethnic conflict, we almost automatically find ourselves talking about ethnic groups.
Now it might be asked: "What's wrong with this?" After all, it seems to be mere common sense to treat ethnic struggles as the struggles of ethnic groups, and ethnic conflict as conflict between such groups. I agree that this is the—or at least a—commonsense view of the matter. But we cannot rely on common sense here. Ethnic common sense—the tendency to partition the social world into puta-tively deeply constituted, quasi-natural intrinsic kinds (Hirschfeld 1996)—is a key part of what we want to explain, not what we want to explain things with; it belongs to our empirical data, not to our analytical toolkit.4 Cognitive anthropologists and social psychologists have accumulated a good deal of evidence about commonsense ways of carving up the social world—about what Lawrence Hirschfeld (1996) has called "folk sociologies." The evidence suggests that some commonsense social categories—and notably commonsense ethnic and racial categories—tend to be essentializing and naturalizing (Rothbart and Taylor 1992; Hirschfeld 1996; Gil-White 1999). They are the vehicles of what has been called a "participants' primor-dialism" (Smith 1998: 158) or a "psychological essentialism" (Medin 1989). We obviously cannot ignore such commonsense primor-dialism. But that does not mean we should simply replicate it in our scholarly analyses or policy assessments. As "analysts of naturalizes," we need not be "analytic naturalizers" (Gil-White 1999: 803).
Instead, we need to break with vernacular categories and common-sense understandings. We need to break, for example, with the seemingly obvious and uncontroversial point that ethnic conflict involves conflict between ethnic groups. I want to suggest that ethnic conflict— or what might better be called ethnicized or ethnically framed conflict—need not, and should not, be understood as conflict between
10 ■ Ethnicity without Groups
Ethnicity without Groups ■ 11
ethnic groups, just as racial or racially framed conflict need not be understood as conflict between races, or nationally framed conflict as conflict between nations.
Participants, of course, regularly do represent ethnic, racial, and national conflict in such groupist, even primordialist terms. They often cast ethnic groups, races, or nations as the protagonists—the heroes and martyrs—of such struggles. This is entirely understandable, and doing so can provide an important resource in social and political struggles. But this does not mean analysts should do the same. We must, of course, take vernacular categories and participants' understandings seriously, for they are partly constitutive of our objects of study. But we should not uncritically adopt categories of ethnopoliti-cal practice as our categories of social analysis. Apart from the general unreliability of ethnic common sense as a guide for social analysis, we should remember that participants' accounts—especially those of specialists in ethnicity such as ethnopolitical entrepreneurs, who, unlike nonspecialists, may live "off" as well as "for" ethnicity—often have what Pierre Bourdieu has called a performative character. By invoking groups, they seek to evoke them, summon them, call them into being. Their categories are for doing—designed to stir, summon, justify, mobilize, kindle, and energize. By reifying groups, by treating them as substantial things-in-the-world, ethnopolitical entrepreneurs can, as Bourdieu notes, "contribute to producing what they apparently describe or designate" (1991c: 220).5
Reiflcation is a social process, not simply an intellectual bad habit.6 As a social process, it is central to the practice of politicized ethnicity. And appropriately so. To criticize ethnopolitical entrepreneurs for reifying ethnic groups would be a kind of category mistake. Reifying groups is precisely what ethnopolitical entrepreneurs are in the business of doing. When they are successful, the political fiction of the unified group can be momentarily yet powerfully realized in practice. As analysts, we should certainly try to account for the ways in which—and conditions under which—this practice of reiflcation, this powerful crystallization of group feeling, can work. But we should avoid unintentionally doubling or reinforcing the reiflcation of ethnic groups in ethnopolitical practice with a reiflcation of such groups in social analysis.7
Beyond Groupism
How, then, are we to understand ethnic conflict, if not in commonsense terms as conflict between ethnic groups? And how can we go beyond groupism? Here I sketch eight basic points and then, in the next section, draw out some of their implications. In the final section, I illustrate the argument by considering one empirical case.
Rethinking Ethnicity. We need to rethink not only ethnic conflict, but also what we mean by ethnicity itself. This is not a matter of seeking agreement on a definition. The intricate and ever-recommencing definitional casuistry in studies of ethnicity, race, and nationalism has done little to advance the discussion, and indeed can be viewed as a symptom of the noncumulative nature of research in the field. It is rather a matter of critically scrutinizing our conceptual tools. Ethnicity, race, and nation should be conceptualized not as substances or things or entities or organisms or collective individuals—as the imagery of discrete, concrete, tangible, bounded, and enduring "groups" encourages us to do—but rather in relational, processual, dynamic, eventful, and disaggregated terms. This means thinking of ethnicity, race, and nation not in terms of substantial groups or entities but in terms of practical categories, situated actions, cultural idioms, cognitive schemas, discursive frames, organizational routines, institutional forms, political projects, and contingent events. It means thinking of ethnicization, racialization, and nationalization as political, social, cultural, and psychological processes. And it means taking as a basic analytical category not the "group" as an entity but groupness as a contextually fluctuating conceptual variable. Stated baldly in this fashion, these are of course mere slogans; I will try to develop them somewhat more fully in what follows.
The Reality of Ethnicity. To rethink ethnicity, race, and nationhood along these lines is in no way to dispute their reality, minimize their power, or discount their significance; it is to construe their reality, power, and significance in a different way. Understanding the reality of race, for example, does not require us to posit the existence of races. Racial idioms, ideologies, narratives, categories, and systems of classification, and racialized ways of seeing, thinking, talking, and framing claims, are real and consequential, especially when they are embedded in powerful organizations. But the reality of race—and even its overwhelming coercive power in some settings—does not depend on the existence of "races." Similarly, the reality of ethnicity and
12 ■ Ethnicity without Groups
nationhood—and the overriding power of ethnic and national identifications in some settings—does not depend on the existence of ethnic groups or nations as substantial groups or entities.
Groupness as Event. Shifting attention from groups to groupness, and treating groupness as variable and contingent rather than fixed and given,8 allows us to take account of—and, potentially, to account for— phases of extraordinary cohesion and moments of intensely felt collective solidarity, without implicitly treating high levels of groupness as constant, enduring, or definitionally present. It allows us to treat groupness as an event, as something that "happens," as E. P. Thompson (1963: 9) famously said about class. At the same time, it keeps us alert to the possibility that groupness may not happen, that high levels of groupness may fail to crystallize, despite the group-making efforts of ethnopolitical entrepreneurs, and even in situations of intense elite-level ethnopolitical conflict. Being analytically attuned to "negative" instances in this way enlarges the domain of relevant cases, and helps correct for the bias in the literature toward the study of striking instances of high groupness, successful mobilization, or conspicuous violence—a bias that can engender an "overethnicized" view of the social world, a distorted representation of whole world regions as "seething cauldrons" of ethnic tension {Brubaker 1998b), and an over-estimation of the incidence of ethnic violence (Fearon and Laitin 1996; this volume, Chapter 4). Sensitivity to such negative instances can also direct potentially fruitful analytical attention toward the problem of explaining failed efforts at ethnopolitical mobilization.
Groups and Categories. Much talk about ethnic, racial, or national groups is obscured by the failure to distinguish between groups and categories. If by "group" we mean a mutually interacting, mutually recognizing, mutually oriented, effectively communicating, bounded collectivity with a sense of solidarity, corporate identity, and capacity for concerted action, or even if we adopt a less exigent understanding of "group," it should be clear that a category is not a group.9 It is at best a potential basis for group-formation or "groupness."10
By distinguishing consistently between categories and groups, we can problematize—rather than presume—the relation between them. We can ask about the degree of groupness associated with a particular category in a particular setting, and about the political, social, cultural, and psychological processes through which categories get invested with groupness (Petersen 1987). We can ask how people—and
Ethnicity without Groups • 13
organizations—do things with categories. This includes limiting access to scarce resources or particular domains of activity by excluding categorically distinguished outsiders," but it also includes more mundane actions such as identifying or characterizing oneself or others (Levine 1999; Brubaker et al. 2004) or simply "doing being ethnic" in an ethnomethodological sense (Moerman 1974). We can analyze the organizational and discursive careers of categories—the processes through which they become institutionalized and entrenched in administrative routines (Tilly 1998) and embedded in culturally powerful and symbolically resonant myths, memories, and narratives (Armstrong 1982; Smith 1986). We can study the politics of categories, both from above and from below. From above, we can focus on the ways in which categories are proposed, propagated, imposed, institutionalized, discursively articulated, organizationally entrenched, and generally embedded in multifarious forms of "governmentality."12 From below, we can study the "micropolitics" of categories, the ways in which the categorized appropriate, internalize, subvert, evade, or transform the categories that are imposed on them (Dominguez 1986). And drawing on advances in cognitive research, ethnomethod-ology, and conversation analysis,13 we can study the sociocognitive and interactional processes through which categories are used by individuals to make sense of the social world, linked to stereotypical beliefs and expectations about category members,14 invested with emotional associations and evaluative judgments, deployed as resources in specific interactional contexts, and activated by situational triggers or cues. A focus on categories, in short, can illuminate the multifarious ways in which ethnicity, race, and nationhood can exist and "work" without the existence of ethnic groups as substantial entities. It can help us envision ethnicity without groups.
Group-Making as Project. If we treat groupness as a variable and distinguish between groups and categories, we can attend to the dynamics of group-making as a social, cultural, and political project, aimed at transforming categories into groups or increasing levels of groupness (Bourdieu 1991c, 1991d). Sometimes this is done in quite a cynical fashion. Ethnic and other insurgencies, for example, often adopt what is called in French a politique du pire, a politics of seeking the worst outcome in the short run so as to bolster their legitimacy or improve their prospects in the longer run. When the small, ill-equipped, ragtag Kosovo Liberation Army (KLA) stepped up its attacks
14 • Ethnicity without Groups
Ethnicity without Groups * 15
on Serb policemen and other targets in early 1998, for example, this was done as a deliberate—and successful—strategy of provoking massive regime reprisals. As in many such situations, the brunt of the reprisals was borne by civilians. The cycle of attacks and counterattacks sharply increased groupness among both Kosovo Albanians and Kosovo Serbs, generated greater support for the KLA among both Kosovo and diaspora Albanians, and bolstered KLA recruitment and funding. This enabled the KLA to mount a more serious challenge to the regime, which in turn generated more brutal regime reprisals, and so on. In this sense, group crystallization and polarization were the result of violence, not the cause (Brubaker 1999). The same can be said, mutatis mutandis, about the dynamics of the second intifada in Israel and the occupied territories.
Of course, the KLA was not starting from scratch in the late 1990s. It began already with relatively high levels of groupness, a legacy of earlier phases of conflict. The propitious "raw materials" the KLA had to work with no doubt help explain the success of its strategy. Not all group-making projects succeed, and those that do succeed (more or less) do so in part as a result of the cultural and psychological materials they have to work with. These materials include not only, or especially, "deep," longue-duree cultural structures such as the mythomoteurs highlighted by Armstrong (1982) and Smith (1986), but also the moderately durable ways of thinking and feeling that represent "middle-range" legacies of historical experience and political action. Yet while such raw materials—themselves the product and precipitate of past struggles and predicaments—constrain and condition the possibilities for group-making in the present, there remains considerable scope for deliberate group-making strategies. Certain dramatic events, in particular, can galvanize group feeling, and ratchet up pre-existing levels of groupness (Laitin 1995b). This is why deliberate violence, undertaken as a strategy of provocation, often by a very small number of persons, can sometimes be an exceptionally effective strategy of group-making.
Groups and Organizations. Although participants' rhetoric and commonsense accounts treat ethnic groups as the protagonists of ethnic conflict, in fact the chief protagonists of most ethnic conflict— and a fortiori of most ethnic violence—are not ethnic groups as such but various kinds of organizations, broadly understood, and their empowered and authorized incumbents. These include states (or more
broadly autonomous polities) and their organizational components such as particular ministries, offices, law enforcement agencies, and armed forces units; they include terrorist groups, paramilitary organizations, armed bands, and loosely structured gangs; and they include political parties, ethnic associations, social movement organizations, churches, newspapers, radio and television stations, and so on. Some of these organizations may represent themselves, or may be seen by others, as organizations of and for particular ethnic groups.15 But even when this is the case, organizations cannot be equated with ethnic groups. It is because and insofar as they are organizations, and possess certain material and organizational resources, that they (or more precisely their incumbents) are capable of organized action, and thereby of acting as more or less coherent protagonists in ethnic conflict.16 Although common sense and participants' rhetoric attribute discrete existence, boundedness, coherence, identity, interest, and agency to ethnic groups, these attributes are in fact characteristic of organizations. The IRA, KLA, and Kurdistan Workers' Party (PKK) claim to speak and act in the name of the (Catholic) Irish, the Kosovo Albanians, and the Kurds of Turkey respectively; but surely analysts must differentiate between such organizations and the putatively homogeneous and bounded groups in whose name they claim to act. The point applies not only to military, paramilitary, and terrorist organizations, of course, but to all organizations that claim to speak and act in the name of ethnic, racial, or national groups—or indeed in the name of any other kind of group (Heisler 1990).
A fuller and more rounded treatment of this theme, to be sure, would require several qualifications that I can only gesture at here. Conflict and violence vary in the degree to which, as well as the manner in which, organizations are involved. What Donald Horowitz (2001) has called the deadly ethnic riot, for example, differs sharply from organized ethnic insurgencies or terrorist campaigns. Although organizations (sometimes ephemeral ones) may play an important role in preparing, provoking, and permitting such riots, much of the actual violence is committed by broader sets of participants acting in relatively spontaneous fashion, and in starkly polarized situations characterized by high levels of groupness. Moreover, even where organizations are the core protagonists, they may depend on a penumbra of ancillary or supportive action on the part of sympathetic nonmembers. The "representativeness" of organizations—the degree
16 • Ethnicity without Groups
to which an organization can justifiably claim to represent the will, express the interests, and enjoy the active or passive support of its constituents—is enormously variable, not only among organizations, but also over time and across domains. In addition, while organizations are ordinarily the protagonists of conflict and violence, they are not always the objects or targets of conflict and violence. Entire population categories—or putative groups—can be the objects of organized action, even if they cannot easily be the subjects or undertakers of such action.17 Finally, even apart from situations of violence, ethnic conflict may be at least partly amorphous, carried out not by organizations as such but spontaneously by individuals through such everyday actions as shunning, insults, demands for deference or conformity, or withholdings of routine interactional tokens of acknowledgment or respect (Bailey 1997). Still, despite these qualifications, it is clear that organizations, not ethnic groups as such, are the chief protagonists of ethnic conflict and ethnic violence, and that the relationship between organizations and the groups they claim to represent is often deeply ambiguous.
Framing and Coding. If the protagonists of ethnic conflict cannot, in general, be considered ethnic groups, then what makes such conflict count as ethnic conflict? And what makes violence count as ethnic violence? The answer cannot be found in the intrinsic properties of behavior. Violence becomes "ethnic" (or "racial" or "nationalist") through the meanings attributed to it by perpetrators, victims, politicians, officials, journalists, researchers, relief workers, and others. Such acts of framing and narrative encoding do not simply interpret the violence; they constitute it as ethnic.™
When an ethnic frame is established, we "see" conflict and violence not only in ethnic, but in groupist terms. Although such perceived groupness does not necessarily reflect what is felt and experienced by participants in an event, a compelling ex post framing can exercise a powerful feedback effect, shaping subsequent experience and increasing levels of groupness. A great deal is at stake, then, in struggles over the interpretive framing and narrative encoding of conflict and violence.
Interpretive framing, of course, is often contested. Violence—and more generally, conflict—regularly occasions social struggles to label, interpret, and explain it. Such "metaconflicts" or "conflict[s] over the nature of the conflict," as Donald Horowitz has called them (1991a: 2), do not simply shadow conflicts from the outside, but are integral parts
Ethnicity without Groups ■ 17
of them. To impose a label or prevailing interpretive frame—to cause an event to be seen as a "pogrom" or a "riot" or a "rebellion"—is no mere matter of external interpretation, but a constitutive and often consequential act of social definition (Brass 1996b). Interpretive struggles over the naming and framing of violence therefore merit study in their own right (Brass 1996a, 1997; Abelmann and Lie 1995).
How conflict and violence are seen, interpreted, and represented depends significantly on prevailing interpretive frames. Today, ethnic and national frames are readily accessible, powerfully resonant, and widely understood as legitimate. This encourages actors and analysts alike to interpret conflict and violence in ethnic rather than other terms. Analysts are thereby prone to overestimate the incidence of ethnic conflict and violence by "coding" as ethnic instances of conflict or violence that might have been coded in other terms (Bowen 1996; this volume, Chapter 4). Actors, in turn, can take advantage of this coding bias, and of the generalized legitimacy of ethnic and national frames, by strategically using ethnic framing to mask the pursuit of clan, clique, or class interests. The point here is not to suggest that clans, cliques, or classes are somehow more real then ethnic groups, but simply to note the existence of structural and cultural incentives for strategic framing.
Ethnicity as Cognition. These observations about the constitutive significance oi coding and framing suggest a final point about the cognitive dimension of ethnicity. Ethnicity, race, and nationhood are fundamentally ways of perceiving, interpreting, and representing the social world. They are not things in the world, but perspectives on the world.19 These include ethnicized ways of seeing (and ignoring), of construing (and misconstruing), of inferring (and misinferring), of remembering (and forgetting). They include ethnically oriented frames, schemas, and narratives, and the situational cues—not least those provided by the media—that activate them. They include systems of classification, categorization, and identi6cation, formal and informal. And they include the tacit, taken-for-granted background knowledge, embodied in persons and embedded in institutionalized routines and practices, through which people recognize and experience objects, places, persons, actions, or situations as ethnically, racially, or nationally marked or meaningful.
Cognitive perspectives, broadly understood,21' can help advance con-structivist research on ethnicity, race, and nationhood, which has stalled in recent years as it has grown complacent with success. Instead
18 ' Ethnicity without Groups
of simply asserting that ethnicity, race, and nationhood are constructed, they can help specify how they are constructed. They can help specify how—and when—people identify themselves, perceive others, experience the world, and interpret their predicaments in racial, ethnic, or national rather than other terms. They can help specify how "group-ness" can "crystallize" in some situations while remaining latent and merely potential in others. And they can help link macrolevel outcomes with microlevel processes (Hirschfeld 1996).
Implications
At this point a critic might interject: "What is the point of all this? Even if we can study 'ethnicity without groups,' why should we? Concepts invariably simplify the world; that the concept of discrete and bounded ethnic groups does so, suggesting something more substantial and clear-cut than really exists, cannot be held against it. The concept of ethnic group may be a blunt instrument, but it's good enough as a first approximation. This talk about groupness and framing and practical categories and cognitive schemas is all well and good, but meanwhile the killing goes on. Does the critique matter in the real world, or—if at all—only in the ivory tower? What practical difference does it make?"
I believe the critique of groupism does have implications, albeit rather general ones, for the ways in which researchers, journalists, policymakers, NGOs, and others come to terms, analytically and practically, with what we ordinarily—though perhaps too readily—call ethnic conflict and ethnic violence. Here I would like to enumerate five of these, before proceeding in the final section to discuss an empirical case.
First, sensitivity to framing dynamics, to the generalized coding bias in favor of ethnicity, and to the sometimes strategic or even cynical use of ethnic framing to mask the pursuit of clan, clique, or class interests can alert us to the risk of overethnicized or overly group ist interpretations of (and interventions in) situations of conflict and violence (Bowen 1996). One need not subscribe to a reductionist "elite manipulation" view of politicized ethnicity (Brubaker 1998b) to acknowledge that the "spin" put on conflicts by participants may conceal as much as it reveals, and that the representation of conflicts as conflicts between ethnic or national groups may obscure the interests at stake and the dynamics involved. What is represented as ethnic conflict or
Ethnicity without Groups • 19
ethnic war—such as the violence in the former Yugoslavia—may have as much or more to do with thuggery, warlordship, opportunistic looting, and black-market profiteering than with ethnicity (Mueller 2000; cf. Kaldor 1999; Collier 2000).
Second, recognition of the centrality of organizations in ethnic conflict and ethnic violence, of the often equivocal character of their leaders' claims to speak and act in the name of ethnic groups, and of the performative nature of ethnopolitical rhetoric, enlisted in the service of group-making projects, can remind us not to mistake groupist rhetoric for real groupness, the putative groups of ethnopolitical rhetoric for substantial things-in-the-world.
Third, awareness of the interest that ethnic and nationalist leaders may have in living off politics, as well as for politics (to borrow the classic distinction of Max Weber [1946: 84]), and awareness of the possible divergence between the interests of leaders and those of their putative constituents, can keep us from accepting at face value leaders' claims about the beliefs, desires, and interests of their constituents.
Fourth, sensitivity to the variable and contingent, waxing and waning nature of groupness, and to the fact that high levels of groupness may be more the result of conflict (especially violent conflict) than its underlying cause, can focus our analytical attention and policy interventions on the processes through which groupness tends to develop and crystallize, and those through which it may subside. Some attention has been given recently to the former, including tipping and cascade mechanisms (Laitin 1995b; Kuran 1998b; this volume, Chapter 4: 107) and mechanisms governing the activation and diffusion of schemas and the "epidemiology of representations" (Sperber 1985; this volume, Chapter 3). But declining curves of groupness have not been studied systematically, although they are just as important, theoretically and practically. Once ratcheted up to a high level, groupness does not remain there out of inertia. If not sustained at high levels through specific social and cognitive mechanisms, it will tend to decline, as everyday interests reassert themselves, through a process of what Weber (in a different but apposite context [1968 (1922):246-54]) called "routmization" (Ver-alltäglichung, literally "towards everydayness").
Lastly, a disaggregating, non-groupist approach can bring into analytical and policy focus the critical importance of intra-ethnic mechanisms in generating and sustaining putatively interethnic conflict (this volume, Chapter 4: 98-101). These include in-group "policing," monitoring, or
20 • Ethnicity without Groups
Ethnicity without Croups • 21
sanctioning processes (Laitin 1995b); the "ethnic outbidding" through which electoral competition can foster extreme ethnicization {Rothschild 1981; Horowitz 1985); the calculated instigation or provocation of conflict with outsiders by vulnerable incumbents seeking to deflect in-group challenges to their positions; and in-group processes bearing on the dynamics of recruitment into gangs, militias, terrorist groups, or guerrilla armies, including honoring, shaming, and shunning practices, rituals of manhood, intergenerational tensions, and the promising and provision of materia! and symbolic rewards for martyrs.
Ethnicity at Work in a Transylvanian Town
At this point, I would like to add some flesh to the bare-bones analytical argument sketched above. It is tempting to comment on the United States. It would be easy to score rhetorical points by emphasizing that the "groups" taken to constitute the canonical "ethnoracial pentagon" (Hollinger 1995)—African Americans, Asian Americans, Whites, Native Americans, and Latinos—are (with the partial exception of African Americans) not groups at all but categories, backed by political entrepreneurs and entrenched in governmental and other organizational routines of social counting and accounting (Office of Management and Budget 1994). It would be easy to highlight the enormous cultural heterogeneity within these and other putative "groups," and the minimal degree of groupness associated with many ethnic categories in the United States (Gans 1979; Heisler 1990).21
But rather than take this tack, I will try to address a harder case, drawn from a region that, for a century and a half, has been the locus classicus of ethnic and nationalist conflict. 1 want to consider briefly how ethnicity works in an East Central European town characterized by continuous and often intense elite-level ethnonational conflict since the fall of communism (and, of course, by a much longer history of ethnonational tension). Here too, I want to suggest, we can fruitfully analyze ethnicity without groups.22
The setting is the city of Cluj, the main administrative, economic, and cultural center of the Transylvanian region of Romania. Of the approximately 320,000 residents, a substantial minority—just under 20 percent, according to the 2002 Census—identify themselves as Hungarian by ethnocultural nationality.23 The city has been the site of protracted and seemingly intractable ethnonational conflict since the
collapse of the Ceausescu regime in December 1989. But this is not, I will argue, best understood as a conflict between ethnic or national groups. To think of it as a conflict between groups is to conflate categories ("Hungarian" and "Romanian") with groups ("the Hungarians," "the Romanians"); to obscure the generally low, though fluctuating, degree of groupness in this setting; to mistake the putative groups invoked by ethnonational rhetoric for substantial things-in-the-world; to accept, at least tacitly, that nationalist organizations speak for the "groups" they claim to represent; and to neglect the everyday contexts in which ethnic and national categories take on meaning and the processes through which ethnicity actually "works" in everyday life.
Here, as elsewhere, the protagonists of the conflict have been organizations, not groups. The conflict has pitted the town's three-term mayor—the flamboyant Romanian nationalist Gheorghe Funar—and the statewide Romanian nationalist parties against the Cluj-based Democratic Alliance of Hungarians of Romania (DAHR), at once a statewide political party with its electoral base in Transylvania and an organization claiming to represent and further the interests of the Hungarian minority in Romania. Rhetoric has been heated on both sides. Mayor Funar has accused Hungary of harboring irredentist designs on Transylvania;24 he has called the DAHR a "terrorist organization"; and he has accused Transylvanian Hungarians of secretly collecting weapons, forming paramilitary detachments, and planning an attack on Romanians. Funar has ordered bilingual signs removed from the few buildings that had them; banned proposed celebrations of the Hungarian national holiday; called for the suspending of Hungarian-language broadcasts on Romanian state television; called for punishment of citizens for displaying the Hungarian flag or singing the Hungarian anthem; and proposed to rename after Romanian personages the few Cluj streets that bear the names of Hungarians.
The DAHR, for its part, is committed to a number of goals that outrage Romanian nationalists.25 It characterizes Hungarians in Romania as an "indigenous community" entitled to an equal partnership with the Romanian nation as a constituent element of the Romanian state—thereby directly challenging the prevailing (and constitutionally enshrined) Romanian understanding of the state as a unitary nation-state such as France. At the same time, it characterizes Transylvanian Hungarians as an "organic part of the Hungarian nation," and
22 • Ethnicity without Groups
as such claims the right to cultivate relations with the "mother country" across the border, which leads Romanian nationalists to call into question their loyalty to the Romanian state. It demands collective rights for Hungarians as a national minority, and it demands autonomy, including territorial autonomy, for areas in which Hungarians live as a local majority, thereby raising the specter of separatism in the minds of Romanian nationalists. It demands that Hungarians have their own institutional system in the domain of education and culture—yet that this institutional system should be financed by the Romanian state. It demands the right to public, state-funded education in Hungarian at every level and in every branch of the educational system. It demands the right to take entrance exams to every school and university in Hungarian, even if the school or department to which the student is applying carries out instruction in Romanian. And it demands the reestablishment of an independent Hungarian university in Cluj.
Like ethnic and nationalist organizations everywhere, the DAHR claims to speak for the Hungarian minority in Romania, often characterizing it as a singular entity, "the Hungariandom of Romania" (a romdniai magyarsdg). But no such entity exists.26 The many Cluj residents who self-identify as Hungarian are often sharply critical of the DAHR, and there is no evidence that the demands of the DAHR are the demands of "the Hungarians." On the question of a Hungarian university—the most contentious political issue of the last few years—a survey conducted by a Hungarian sociologist found that a plurality of Hungarian university students in Cluj preferred an autonomous system of Hungarian-language education within the existing university to the DAHR goal of reestablishment of a separate Hungarian university (Magyari-Nandor and Peter 1997). Most Hungarians, like most Romanians, are largely indifferent to politics, and preoccupied with problems of everyday life—problems that are not interpreted in ethnic terms. Although survey data and election results indicate that they vote en bloc for the DAHR, most Hungarians are familiar only in a vague way with the DAHR program. Similarly, there is no evidence that Mayor Funar's anti-Hungarian views are widely shared by the town's Romanian residents. When Funar is praised, it is typically as a "good housekeeper" (bun gospodar); he is given credit for sprucing up the town's appearance and for providing comparatively good municipal services. Almost everyone—Romanian and
Ethnicity without Groups • 23
Hungarian alike—talks about ethnic conflict as something that "comes from above" and is stirred up by politicians pursuing their own interests. The near-universal refrain is that ethnicity is "not a problem." To be sure, a similar idiom—or perhaps ideology—of everyday interethnic harmony can be found in many other settings, including some deeply divided, violence-plagued ones. So the idiom cannot be taken as evidence of the irrelevance of ethnicity. The point here is simply to underscore the gap between nationalist organizations and the putative "groups" in whose names they claim to speak.
Despite the continuous elite-level ethnopolitical conflict in Cluj since the fall of Ceausescu, levels of "groupness" have remained low. At no time did Hungarians and Romanians crystallize as distinct, solidary, bounded groups. The contrast with Targu Mures, a few hours' drive to the east, is instructive. In Targu Mures, ethnically framed conflict over the control of a high school and over the control of local government in the immediate aftermath of the fall of Ceausescu intensified and broadened into a generalized conflict over the "ownership" and control of the ethnodemographically evenly divided city. The conflict culminated in mass assemblies and two days of street fighting that left at least six dead and 200 injured. In the days leading up to the violent denouement, categories had become palpable, sharply bounded groups, united by intensely felt collective solidarity and animated by a single overriding distinction between "us" and "them." The violence itself reinforced this sense of groupness, which then subsided gradually as life returned to normal, and no further Hungarian-Romanian violence occurred, here or elsewhere in Transylvania.
No such crystallization occurred in Cluj. There were, to be sure, a few moments of moderately heightened groupness. One such moment—among Hungarians—occurred when Mayor Funar ordered a new plaque installed on the base of a monumental equestrian statue of Matthias Corvinus, celebrated king of Hungary during the late fifteenth century, in the town's main square. The statue, erected at the turn of the last century at a moment of, and as a monument to, triumphant Hungarian nationalism, is perceived by many Hungarians as "their own," and the new plaque deliberately affronted Hungarian national sensibilities by emphasizing the (partly) Romanian origin of Matthias Corvinus and representing him—contrary to the triumphalist image projected by the statue—as having been defeated in battle by "his own nation," Moldavia (Feischmidt 2001). Another moment
24 ■ Ethnicity without Groups
occurred when archeological excavations were begun in front of the statue, again in a manner calculated to affront Hungarian national sensibilities by highlighting the earlier Roman—and by extension, Romanian—presence on the site. A third moment occurred in March 1998, when Mayor Funar tried to bar Hungarians from carrying out their annual 15 March celebration commemorating the revolution of 1848, this year's celebration, in the sesqukentennial year, having special significance.27 A final moment occurred in June 1999 at the time of a much-hyped soccer match in Bucharest between the national teams of Romania and Hungary. In Cluj, the match was televised on a huge outdoor screen in the main square; some fans chanted "Afam, afara, cu Ungurii din tarar (out, out, Hungarians out of the country!) and vandalized cars with Hungarian license plates.28
In each of these cases, groupness—especially among Hungarians, though in the final case among Romanians as well—was heightened, but only to a modest degree, and only for a passing moment. The first event occasioned a substantial but isolated Hungarian protest, the second a smaller protest, the third some concern that the commemoration might be broken up (in the event it proceeded without serious incident), and the last some moments of concern for those who happened to be in the town center during and immediately after the soccer match. But even at these maximally grouplike moments, there was no overriding sense of bounded and solidary groupness for those not immediately involved in the events.29 What is striking about Cluj in the 1990s, in short, is that groupness failed to happen.
To note the relatively low degree of groupness in Cluj, and the gap between organizations and the putative groups they claim to represent, is not to suggest that ethnicity is somehow not "real" in this setting, or that it is purely an elite phenomenon. Yet to understand how ethnicity works, it may help to begin not with "the Romanians" and "the Hungarians" as groups, but with "Romanian" and "Hungarian" as categories. Doing so suggests a different set of questions than those that come to mind when we begin with "groups," Starting with groups, one is led to ask what groups want, demand, or aspire towards; how they think of themselves and others; and how they act in relation to other groups. One is led almost automatically by the substantialist language to attribute identity, agency, interests, and will to groups. Starting with categories, by contrast, invites us to focus on processes and relations rather than substances. It invites us to specify how people
Ethnicity without Groups • 25
and organizations do things with, and to, ethnic and national categories; how such categories are used to channel and organize processes and relations; and how categories get institutionalized, and with what consequences. It invites us to ask how, why, and in what contexts ethnic categories are used—or not used—to make sense of problems and predicaments, to articulate affinities and affiliations, to identify commonalities and connections, to frame stories and self-understandings.
Consider here just two of the many ways of pursuing a category centered rather than a group-centered approach to ethnicity in Cluj. First, a good deal of commonsense cultural knowledge about the social world and one's place in it, here as in other settings, is organized around ethnonational categories.30 This includes knowledge of one's own and others' ethnocultural nationality, and the ability to assign unknown others to ethnonational categories on the basis of cues such as language, accent, name, and sometimes appearance. It includes knowledge of what incumbents of such categories are like,11 how they typically behave, and how ethnonational category membership matters in various spheres of life. Such commonsense category-based knowledge shapes everyday interaction, figures in stories people tell about themselves and others, and provides ready-made explanations for certain events or states of affairs. For Hungarians, for example, categorizing an unknown person as Hungarian or Romanian may govern how one interacts with him or her, determining not only the language but also the manner in which one will speak, a more personal and confidential (bizalmas) style often being employed with fellow Hungarians. Or for Romanians, categorizing two persons speaking Hungarian in a mixed-language setting as Hungarian (rather than, for example, as friends who happen to be speaking Hungarian) provides a ready-made explanation for their conduct, it being commonsense knowledge about Hungarians that they will form a bisericuta (clique, literally: small church) with others of their kind, excluding co-present Romanians, whenever they have the chance. Or again for Hungarians, categorically organized commonsense knowledge provides a ready-made framework for perceiving differential educational and economic opportunities as structured along ethnic lines, explaining such differentials in terms of what they know about the bearing of ethnic nationality on grading, admissions, hiring, promotion, and firing decisions, and justifying the commonly voiced opinion that "we [Hungarians] have to work twice as hard" to
26 • Ethnicity without Groups
Ethnicity without Groups • 27
get ahead. These and many other examples suggest that ethnicity is, in important part, a cognitive phenomenon, a way of seeing and interpreting the world, and that, as such, it works in and through categories and category-based commonsense knowledge.
Ethnic categories shape institutional as well as informal cognition and recognition. They not only structure perception and interpretation in the ebb and flow of everyday interaction but channel conduct through official classifications and organizational routines. Thus ethnic (and other) categories may be used to allocate rights, regulate actions, distribute benefits and burdens, construct category-specific institutions, identify particular persons as bearers of categorical attributes, "cultivate" populations, or, at the extreme, "eradicate" unwanted "elements."32
In Cluj—as in Romania generally—ethnic categories are not institutionalized in dramatic ways. Yet there is one important set of institutions built, in part, around ethnic categories. This is the school, system.33 In Cluj, as in other Transylvanian cities, there is a separate Hungarian-language school system paralleling the mainstream system, and running from preschool through high school. These are not private schools, but part of the state school system. Not all persons identifying themselves as Hungarian attend Hungarian schools, but most do (85 to 90 percent in grades 1 through 4, smaller proportions, though still substantial majorities, in later grades).34 In Cluj, moreover, there are also parallel tracks at the university level in many fields of study.
Categories need ecological niches in which to survive and flourish; the parallel school system provides such a niche for "Hungarian" as an erhnonational category. It is a strategically positioned niche. Hungarian schools provide a legitimate institutional home and a protected public space for the category; they also generate the social structural foundations for a small Hungarian world within the larger Romanian one. Since the schools shape opportunity structures and contact probabilities, and thereby influence friendship patterns (and, at the high school and university level, marriage patterns as well), this world is to a considerable extent self-reproducing. Note that the (partial) reproduction of this social world—an interlocking set of social relationships linking school, friendship circles, and family—does not require strong nationalist commitments or group loyalties. Ethnic networks can be reproduced without high degrees of groupness, largely through the logic of contact probabilities and opportunity structures and the resulting moderately high degrees of ethnic endogamy.35
This brief case study has sought to suggest that even in a setting of intense elite-level ethnic conflict and (by comparison to the United States) deeply rooted and stable ethnic identifications, one can analyze the workings of ethnicity without employing the language of bounded groups.
Conclusion
What are we studying when we study ethnicity and ethnic conflict? I have suggested that we need not frame our analyses in terms of ethnic groups, and that it may be more productive to focus on practical categories, situated actions, cultural idioms, cognitive schemas, common-sense knowledge, organizational routines and resources, discursive frames, institutionalized forms, political projects, contingent events, and variable groupness. It should be noted in closing, however, that by framing our inquiry in this way, and by bringing to bear a set of analytical perspectives not ordinarily associated with the study of ethnicity—cognitive theory, ethnomethodology, conversation analysis, network analysis, organizational analysis, and institutional theory, for example—we may end up not studying ethnicity at all. It may be that "ethnicity" is simply a convenient—though in certain respects misleading—rubric under which to group phenomena that, on the one hand, are highly disparate, and, on the other, have a great deal in common with phenomena that are not ordinarily subsumed under the rubric of ethnicity.36 In other words, by raising questions about the unit of analysis—the ethnic group—we may end up questioning the domain of analysis: ethnicity itself. But that is an argument for another occasion.
<^>  CHAPTER TWO
Beyond "Identity"
"The worst thing one can do with words," wrote George Orwell (1953: 169-70) a half a century ago, "is to surrender to them." If language is to be "an instrument for expressing and not for concealing or preventing thought," he continued, one must "let the meaning choose the word, and not the other way about." The argument of this essay is that the social sciences and humanities have surrendered to the word "identity"; that this has both intellectual and political costs; and that we can do better. "Identity," we will argue, tends to mean too much (when understood in a strong sense), too little (when understood in a weak sense), or nothing at all (because of its sheer ambiguity). We take stock of the conceptual and theoretical work "identity" is supposed to do, and suggest that this work might be done by other terms, less ambiguous, and unencumbered by the reifying connotations of "identity."
We argue that the prevailing constructivist stance on identity—the attempt to "soften" the term, to acquit it of the charge of "essentialism" by stipulating that identities are constructed, fluid, and multiple—leaves us without a rationale for talking about "identities" at all and ill equipped to examine the "hard" dynamics and essentialist claims of contemporary identity politics. "Soft" constructivism allows putative "identities" to proliferate. But as they proliferate, the term loses its
This chapter was coauthored with Frederick Cooper.
Beyond "Identity" • 29
r analytical purchase. If identity is everywhere, it is nowhere. If it is fluid, how can we understand the ways in which self-understandings may harden, congeal, and crystallize? If it is constructed, how can we understand the sometimes coercive force of external identifications? If it is multiple, how do we understand the terrible singularity that is often striven for—and sometimes realized—by politicians seeking to transform mere categories into unitary and exclusive groups? How can we understand the power and pathos of identity politics?
"Identity" is a key term in the vernacular idiom of contemporary politics, and social analysis must take account of this fact. But this does not require us to use "identity" as a category of analysis or to conceptualize "identities" as something that all people have, seek, construct, or negotiate. Conceptualizing all affinities and affiliations, all forms of belonging, all experiences of commonality, connectedness, and cohesion, all self-understandings and self-identifications in the idiom of "identity" saddles us with a blunt, flat, undifferentiated vocabulary.
We do not aim here to contribute to the ongoing debate on identity politics.1 We focus instead on identity as an analytical category. This is not a "merely semantic" or terminological issue. The use and abuse of "identity," we suggest, affects not only the language of social analysis but also—inseparably—its substance. Social analysis—including the analysis of identity politics—requires relatively unambiguous analytical categories. Whatever its suggestiveness, whatever its indispensability in certain practical contexts, "identity" is too ambiguous, too torn between "hard" and "soft" meanings, essentialist connotations and constructivist qualifiers, to serve well the demands of social analysis.
The "Identity" Crisis in the Social Sciences
"Identity" and cognate terms in other languages have a long history as technical terms in Western philosophy, from the ancient Greeks through contemporary analytical philosophy. They have been used to address the perennial philosophical problems of permanence amidst manifest change and of unity amidst manifest diversity (Stroll 1967: 121 ).2 Widespread vernacular and social-analytical use of "identity" and its cognates, however, is of much more recent vintage and more localized provenance.3
The introduction of "identity" into social analysis and its initial diffusion in the social sciences and public discourse occurred in the
30 * Beyond "Identity"
United States in the 1960s (with some anticipations in the second half of the 1950s).4 The most important and best-known trajectory involved the appropriation and popularization of the work of Jarjk Enkson (who was responsible, among other things, for coining the term "identity crisis").5 But there were other paths of diffusion as well. The notion of identification was pried from its original, specifically psychoanalytic context (where the term had been initially introduced by Freud) and linked to ethnicity on the one hand (through Gordon Allport's influential 1954 book The Nature of Prejudice) and to sociological role theory and reference group theory on the other (through figures such as Nelson Foote and Robert Merton). Symbolic interactionist sociology, concerned from the outset with "the self," came increasingly to speak of "identity," in part through the influence of Anselm Strauss (1959). More influential in popularizing the notion of identity, however, were Erving Goffman (1963), working on the periphery of the symbolic interactionist tradition, and Peter Berger (Berger and Luckmann 1967; Berger et al. 1973; Berger 1974), working in social constructionist and phenomenolog-ical traditions.
For a variety of reasons, the term identity proved highly resonant in the 1960s,ri diffusing quickly across disciplinary and national boundaries, establishing itself in the journalistic as well as the academic lexicon, and permeating the language of social and political practice as well as that of social and political analysis. In the American context, the prevalent individualist ethos and idiom gave a particular salience and resonance to "identity" concerns, particularly in the contexts of the 1950s thematization of the "mass society" problem and the 1960s generational rebellions. And from the late 1960s on, with the rise of the Black Power movement, and subsequently other ethnic movements for which it served as a template, concerns with and assertions of individual identity, already linked by Erikson to "communal culture,"7 were readily, if facilely, transposed to the group level. The proliferation of identitarian claim-making was facilitated by the comparative institutional weakness of leftist politics in the United States and by the concomitant weakness of class-based idioms of social and political analysis. As numerous analysts (e.g., Calhoun 1993b) have observed, class can itself be understood as an identity. Our point here is simply that the weakness of class politics in the United States (vis-a-vis Western Europe) helps explain the profusion of identity claims.
Beyond "Identity" • 31
Already in the mid-1970s, W.J. M. Mackenzie could characterize identity as a word "driven out of its wits by overuse," and Robert Coles could remark that the notions of identity and identity crisis had become "the purest of cliches."8 But that was only the beginning. In the 1980s, with the rise of race, class, and gender as the "holy trinity" of literary criticism and cultural studies (Appiah and Gates 1995: 1), the humanities joined the fray in full force. And "identity talk"—inside and outside academia—continued to proliferate in the 1990s.9 The "identity" crisis—a crisis of overproduction and consequent devaluation of meaning—shows no sign of abating.10
Qualitative as well as quantitative indicators signal the centrality— indeed the inescapability—of "identity" as a topos. Two new interdisciplinary journals devoted to the subject, complete with star-studded editorial boards, were launched in the mid-1990s." And quite apart from the pervasive concern with "identity" in work on gender, sexuality, race, religion, ethnicity, nationalism, immigration, new social movements, culture, and "identity politics," even those whose work has not been concerned primarily with these topics have felt obliged to address the question of identity.12
Categories of Practice and Categories of Analysis
Many key terms in the interpretative social sciences and history— "race," "nation," "ethnicity," "citizenship," "democracy," "class," "community," and "tradition," for example—are at once categories of
social and.....political practice and categories of social and political
analysis. By "categories of practice," following Bourdieu, we mean something akin to what others have called "native" or "folk" or "lay" categories. These are categories of everyday social experience, developed and deployed by ordinary social actors, as distinguished from the experience-distant categories used by social analysts.13 We prefer the expression; "category of practice" to the alternatives, for while the latter imply a relatively sharp distinction between "native" or "folk" or "lay" categories on the one hand and "scientific" categories on the other, such concepts as "race," "ethnicity," or "nation" are marked by close reap rocal connection and mutual influence between their practical and analytical uses.14
"Identity," too, is both a category of practice and a category of analysis. As a category of practice, it is used by "lay" actors in some
32 • Beyond "Identity"
(not all!) everyday settings to make sense of themselves, of their activities, of what they share with, and how they differ from, others. It is also used by political entrepreneurs to persuade people to understand themselves, their interests, and their predicaments in a certain way, to persuade certain people that they are (for certain purposes) "identical" with one another and at the same time different from others, and to organize and justify collective action along certain lines.15 In these ways the term "identity" is implicated both in everyday life and in "identity politics" in its various forms.
Everyday "identity talk" and "identity politics" are real and important phenomena. But the contemporary salience of "identity" as a category of practice does not require its use as a category of analysis. Consider an analogy. "Nation" is a widely used category of social and political practice. Appeals and claims made in the name of putative "nations"—for example, claims to self-determination—have been central to politics for 150 years. But one does not have to use "nation" as an analytical category designating an entity in the world in order to understand and analyze such appeals and claims. One does not have to take a category inherent in the practice of nationalism—the realist, reifying conception of nations as real communities—and make this category central to the theory of nationalism.16 Nor does one have to use "race" as a category of analysis—which risks taking for granted that "races" exist—in order to understand and analyze social and political practices oriented to the presumed existence of putative "races" (Loveman 1999).17 Just as one can analyze "nation-talk" and nationalist politics without positing the existence of "nations," or "race-talk" and "race"-oriented politics without positing the existence of "races," so one can analyze "identity-talk" and identity politics without, as analysts, positing the existence of "identities."
The mere use of a term as a category of practice, to be sure, does not disqualify it as a category of analysis.18 If it did, the vocabulary of social analysis would be a great deal poorer, and more artificial, than it is. What is problematic is not that a particular term is used, but how it is used. The problem, as Wacquant (1997: 222) has argued with respect to "race," lies in the "uncontrolled conflation of social and sociological . . . [or] folk and analytic understandings."19 The problem is that "nation," "race," and "identity" are used analytically a good deal of the time more or less as they are used in practice, in an implicitly or explicitly reifying manner, in a manner that implies or asserts that "nations," "races," and "identities" "exist" as
Beyond "Identity" • 33
substantial entities and that people "have" a "nationality," a "race," an "identity."
It may be objected that this overlooks recent efforts to avoid reifying "identity" by theorizing identities as multiple, fragmented, and fluid.20 "Essentialism" has indeed been vigorously criticized, and constructivist gestures now accompany most discussions of "identity."21 Yet we often find an uneasy amalgam of constructivist language and essentialist argumentation.22 This is not a matter of intellectual sloppiness. Rather, it reflects the dual orientation of many academic identitarians as both analysts and protagonists of identity politics. It reflects the tension between the constructivist language that is required by academic correctness and the foundationalist or essentialist message that may be required if appeals to "identity" are to be effective in practice.25 Nor is the solution to be found in a more consistent constructivism: for it is not clear why that which is routinely characterized as "multiple, fragmented, and fluid" should be conceptualized as "identity" at all.
The Uses of "Identity"
What do scholars mean when they talk about "identity"?24 What conceptual and explanatory work is the term supposed to do? This depends on the context of its use and the theoretical tradition from which the use in question derives. The term is richly—indeed for an analytical concept, hopelessly—ambiguous. But one can identify a few key uses:
1. Understood as a ground or basis of social or political action, "identity" is often opposed to "interest" in an effort to highlight and conceptualize noninstrumental modes of social and political action.25 With a slightly different analytical emphasis, it is used to underscore the manner in which action— individual or collective—may be governed by particularistic self-understandings rather than by putatively universal self-interest (Somers 1994). This is probably the most general use of the term; it is frequently found in combination with other uses. It involves three related but distinct contrasts in ways of conceptualizing and explaining action. The first is between self-understanding and (narrowly understood) self-interest.26 The second is between particularity and (putative) universality. The third is between two ways of construing social location. Many (though not all) strands of identitarian
Beyond "Identity"
Beyond "Identity" ■ 35
theorizing see social and political action as powerfully shaped by position in social space.27 In this they agree with many (though not all) strands of universalist, instrumentalist theorizing. But "social location" means something quite different in the two cases. For identitarian theorizing, it means position in a multidimensional space defined by particularistic categorical attributes (race, ethnicity, gender, sexual orientation). For instrumentalist theorizing, it means position in a universalistically conceived social structure (for example, position in the market, the occupational structure, or the mode of production).28
Understood as a specifically collective phenomenon, "identity" denotes a fundamental and consequential "'sameness" among members of a group or category. This may be understood objectively (as a sameness "in itself") or subjectively (as an experienced, felt, or perceived sameness). This sameness is expected to manifest itself in solidarity, in shared dispositions or con-sciousness, or in collective action. This usage is found especially in the literature on social movements (Melucci 1995); on gender;29 and on race, ethnicity, and nationalism (e.g., Isaacs 1975; Connor 1994). In this usage, the line between "identity" as a category of analysis and as a category of practice is often blurred.
Understood as a core aspect of (individual or collective) selfhood or as a fundamental condition of social being, "identity" is invoked to point to something allegedly deep, basic, abiding, or foundational. This is distinguished from more superficial, accidental, fleeting, or contingent aspects or attributes of the self, and is understood as something to be valued, cultivated, supported, recognized, and preserved.30 This usage is characteristic of certain strands of the psychological (or psychologizing) literature, especially as influenced by Erikson,31 though it also appears in the literature on race, ethnicity, and nationalism. Here too the practical and analytical uses of "identity" are frequently conflated.
Understood as a product of social or political action, "identity" is invoked to highlight the processual, interactive development of the kind of collective self-understanding, solidarity, or "groupness" that can make collective action possible. In this
usage, found in certain strands of the "new social movement" literature, "identity" is understood both as a contingent product of social or political action and as a ground or basis of further action (e.g., Calhoun 1991; Melucci 1995; Gould 1995).
5. Understood as the evanescent product of multiple and competing discourses, "identity" is invoked to highlight the unstable, multiple, fluctuating, and fragmented nature of the contemporary self. This usage is found especially in the literature influenced by Foucault, post-structuralism, and post-modernism (e.g., Hall 1996). In somewhat different form, without the post-structuralist trappings, it is also found in certain strands of the literature on ethnicity—notably in "situationalist" or "contextu-alist" accounts of ethnicity (e.g., Werbner 1996).
Clearly, the term "identity" is made to do a great deal of work. It is used to highlight noninstrumental modes of action; to focus on self-understanding rather than self-interest; to designate sameness across persons or sameness over time; to capture allegedly core, foundational aspects of selfhood; to deny that such core, foundational aspects exist; to highlight the processual, interactive development of solidarity and collective self-understanding; and to stress the fragmented quality of the contemporary experience of self, a self unstably patched together through shards of discourse that are contingently activated in differing contexts.
These usages are not simply heterogeneous; they point in sharply differing directions. To be sure, there are affinities between certain of them, notably between the second and third, and between the fourth and fifth. And the first usage is general enough to be compatible with all of the others. But there are strong tensions as well. The second and third uses both highlight fundamental sameness—across persons and over time—while the fourth and fifth uses both reject notions of fundamental or abiding sameness.32
"Identity," then, bears a multivalent, even contradictory theoretical burden. Do we really need this heavily burdened, deeply ambiguous term? The overwhelming weight of scholarly opinion suggests that we do.33 Even the most sophisticated theorists, while readily acknowledging the elusive and problematic nature of "identity," have argued that it remains indispensable. Critical discussion of "identity" has thus sought not to jettison but to save the term by reformulating it so as to
36 • Beyond "Identity"
make it immune from certain objections, especially from the charge of "essentialism." Thus Stuart Hall (1996: 2) characterizes identity as "an idea which cannot be thought in the old way, but without which certain key questions cannot be thought at all." What these key questions are, and why they cannot be addressed without "identity," remain obscure in Hall's sophisticated but opaque discussion.34 Hall's comment echoes an earlier formulation of Claude Levi-Strauss (1977: 332), characterizing identity as "a sort of virtual center (foyer virtuel) to which we must refer to explain certain things, but without it ever having a real existence." Lawrence Grossberg (1996: 87-88), concerned by the narrowing preocuppation of cultural studies with the "theory and politics of identity," nonetheless repeatedly assures the reader that he does "not mean to reject the concept of identity or its political importance in certain struggles" and that his "project is not to escape the discourse of identity but to relocate it, to rearticulate it." Alberto Melucci (1995: 46), a leading exponent of identity-oriented analyses of social movements, acknowledges that "the word identity ... is semanticaliy inseparable from the idea of permanence and is perhaps, for this very reason, ill-suited to the processual analysis for which I am arguing." Ill suited or not, "identity" continues to find a central place in Melucci's writing.
We are not persuaded that "identity" is indispensable. We will sketch below some alternative analytical idioms that can do the necessary work without the attendant confusion. Suffice it to say for the moment that if one wants to argue that particularistic self-understandings shape social and political action in a noninstrumental manner, one can simply say so. If one wants to trace the processes through which persons sharing some categorical attribute come to share definitions of their predicament, understandings of their interest, and a readiness to undertake collective action, it is best to do so in a manner that highlights the contingent and variable relationship between mere categories and bounded, solidary groups. If one wants to examine the meanings and significance people give to constructs such as "race," "ethnicity," and "nationality," one already has to thread through conceptual thickets, and it is not clear what one gains by subsuming them under the flattening rubric of identity. And if one wants to convey the late modern sense of a self that is constructed and continuously reconstructed out of a variety of competing discourses, while remaining fragile, fluctuating and fragmented, it is not obvious how the word "identity" can help.
Beyond "Identity" • 37
"Strong" and "Weak" Understandings of "Identity"
We suggested at the outset that "identity" tends to mean either too much or too little. This point can now be elaborated. Our inventory of the uses of "identity" has revealed not only great heterogeneity but a strong antithesis between positions that highlight fundamental or abiding sameness and stances that expressly reject notions of basic sameness. The former can be called strong or hard conceptions of identity, the latter weak or soft conceptions.
Strong conceptions of "identity" preserve the commonsense meaning of the term—the emphasis on sameness over time or across persons. And they accord well with the way the term is used in most forms of identity politics. But precisely because they adopt for analytical purposes a category of everyday experience and political practice, they entail a series of deeply problematic assumptions:
1. Identity is something all people have, or ought to have, or are searching for.
2. Identity is something all groups (at least groups of a certain kind—e.g., ethnic, racial, or national) have, or ought to have.
3. Identity is something people (and groups) can have without being aware of it. In this perspective, identity is something to be discovered, and something about which one can be mistaken. The strong conception of identity thus replicates the Marxian epistemology of class.
4. Strong notions of collective identity imply strong notions of group boundedness and homogeneity. They imply high degrees of groupness, an "identity" or sameness between group members, a clear boundary between inside and outside.35
Given the powerful challenges from many quarters to substantialist understandings of groups and essentialist understandings of identity, one might think we have sketched a "straw man" here. Yet in fact strong conceptions of "identity" continue to inform important strands of the literature on gender, race, ethnicity, and nationalism (e.g., Isaacs 1975; Connor 1994).
Weak understandings of "identity," by contrast, break consciously with the everyday meaning of the term. It is such weak or "soft"
38 ■ Beyond "Identity"
conceptions that have been heavily favored in theoretical discussions of "identity" in recent years, as theorists have become increasingly aware of and uncomfortable with the strong or "hard" implications of everyday meanings of "identity." Yet this new theoretical "common sense" has problems of its own. We sketch three of these.
The first is what we call "cliched constructivism." Weak or soft conceptions of identity are routinely packaged with standard qualifiers indicating that identity is multiple, unstable, in flux, contingent, fragmented, constructed, negotiated, and so on. These qualifiers have become so familiar—indeed obligatory—in recent years that one reads (and writes) them virtually automatically. They risk becoming mere placeholders, gestures signaling a stance rather than words conveying a meaning.36
Second, it is not clear why weak conceptions of "identity" are conceptions of identity. The everyday sense of "identity" strongly suggests at least some self-sameness over time, some persistence, something that remains identical, the same, while other things are changing. What is the point of using the term "identity" if this core meaning is expressly repudiated?
Third, and most important, weak conceptions of identity may be too weak to do useful theoretical work. In their concern to cleanse the term of its theoretically disreputable "hard" connotations, in their insistence that identities are multiple, malleable, fluid, and so on, soft identitarians leave us with a term so infinitely elastic as to be incapable of performing serious analytical work.
We are not claiming that the strong and weak versions sketched here jointly exhaust the possible meanings and uses of "identity." Nor are we claiming that sophisticated constructivist theorists have not done interesting and important work using "soft" understandings of identity. We will argue, however, that what is interesting and important in this work often does not depend on the use of "identity" as an analytical category. Consider three examples.
Margaret Somers (1994), criticizing scholarly discussions of identity for focusing on categorical commonality rather than on historically variable relational embeddedness, proposes to "reconfigure] the study of identity formation through the concept of narrative" (605), to "incorporate into the core conception of identity the categorically destabilizing dimensions of time, space, and relationality" (606). Somers makes a compelling case for the importance of narrative to social life and social analysis, and argues persuasively for situating social
Beyond "Identity" • 39
narratives in historically specific relational settings. She focuses on the ontological dimension of narratives, on the way in which narratives not only represent but, in an important sense, constitute social actors and the social world in which they act. What remains unclear from her account is why—and in what sense—it is identities that are constituted through narratives and formed in particular relational settings. Social life is indeed pervasively "storied" (614); but it is not clear why this "storiedness" should be axiomatically linked to identity. People everywhere and always tell stories about themselves and others, and locate themselves within culturally available repertoires of stories. But in what sense does it follow that "narrative location endows social actors with identities—however multiple, ambiguous, ephemeral, or conflicting they may be" (618)? What does this soft, flexible notion of identity add to the argument about narrativity? The major analytical work in Somers's article is done by the concept of narrativity, supplemented by that of relational setting; the work done by the concept of identity is much less clear.37
Introducing a collection on Citizenship, Identity, and Social History, Charles Tilly (1996: 7) characterizes identity as a "blurred but indispensable" concept and defines it as "an actor's experience of a category, tie, role, network, group or organization, coupled with a public representation of that experience; the public representation often takes the form of a shared story, a narrative." But what is the relationship between this encompassing, open-ended definition and the work Tilly wants the concept to do? What is gained, analytically, by labeling any experience and public representation of any tie, role, network, etc., as an identity! When it comes to examples, Tilly rounds up the usual suspects: race, gender, class, job, religious affiliation, national origin. But it is not clear what analytical leverage on these phenomena can be provided by the exceptionally capacious, flexible concept of identity he proposes. Highlighting "identity" in the title of the volume signals an openness to the cultural turn in the social history and historical sociology of citizenship; beyond this, it is not clear what work the concept does. Justly well known for fashioning sharply focused, "hardworking" concepts, Tilly here faces the difficulty that confronts most social scientists writing about identity today: that of devising a concept "soft" and flexible enough to satisfy the requirements of relational, constructivist social theory, yet robust enough to have purchase on the phenomena that cry out for explanation, some of which are quite "hard."
40 • Beyond "Identity"
Craig Calhoun (1991) uses the Chinese student movement of 1989 as a vehicle for a subtle and illuminating discussion of the concepts of identity, interest, and collective action. Calhoun explains students' readiness to "knowingly risk death" (53) in Tiananmen Square on the night of 3 June 1989, in terms of an honor-bound identity or sense of self, forged in the course of the movement, to which students became increasingly and, in the end, irrevocably committed. His account of the shifts in the students' lived sense of self during the weeks of their protest—as they were drawn, in and through the dynamics of their struggle, from an originally "positional" (67), class-based self-understanding as students and intellectuals to a broader, emotionally charged identification with national and even universal ideals—is a compelling one. Here too, however, the crucial analytical work appears to be done by a concept other than identity—in this case, that of honor. Honor, Calhoun observes, "is imperative in a way interest is not" (64). But it is also imperative in a way identity, in the weak sense, is not. Calhoun subsumes honor under the rubric of identity, and presents his argument as a general one about the "constitution and transformation of identity." Yet his fundamental argument in this paper, it would seem, is not about identity in general, but about the way in which a compelling sense of honor can, in extraordinary circumstances, lead people to undertake extraordinary actions, lest their core sense of self be radically undermined.
Identity in this exceptionally strong sense—as a sense of self that can imperatively require interest-threatening or even life-threatening action—has little to do with identity in the weak or soft sense. Calhoun himself underscores the incommensurability between "ordinary identity—self-conceptions, the way people reconcile interests in everyday life" and the imperative, honor-driven sense of self that can enable or even require "bravery to the point of apparent foolishness" (Calhoun 1991: 68, 51). Calhoun provides a powerful characterization of the latter; but it is not clear what analytical work is done by the former, more general conception of identity.
Introducing his edited volume on Social Theory and the Politics of Identity, Calhoun works with this more general understanding of identity. "Concerns with individual and collective identity," he observes, "are ubiquitous." It is certainly true that "we know of no people without names, no languages or cultures in which some manner of distinctions between self and other, we and they are not made" (Calhoun
Beyond "Identity" • 41
1994: 9). But it is not clear why this implies the ubiquity of identity, unless we dilute "identity" to the point of designating all practices involving naming and self-other distinctions. Calhoun—like Somers and Tilly—goes on to make illuminating arguments on a range of issues concerning claims of commonality and difference in contemporary social movements. Yet while such claims are indeed often framed today in an idiom of "identity," it is not clear that adopting that idiom for analytical purposes is necessary or even helpful.
In Other Words
What alternative terms might stand in for "identity," doing the theoretical work "identity" is supposed to do without its confusing, contradictory connotations? Given the great range and heterogeneity of the work done by "identity," it would be fruitless to look for a single substitute, for such a term would be as overburdened as "identity" itself. Our strategy has been rather to unbundle the thick tangle of meanings that have accumulated around the term "identity," and to parcel out the work to a number of less congested terms. We sketch three clusters of terms here.
Identification and Categorization. As a processual, active term, derived from a verb, "identification" lacks thevreifying'connotations of " identity. "3S It invites us to specify the agents that do the identifying. And it does not presuppose that such identifying (even by powerful agents, such as the state) will necessarily result in the internal sameness, the distinctiveness, the bounded groupness that political entrepreneurs may seek to achieve. Identification—of oneself and of others—is intrinsic to social life; "identity" in the strong sense is not.
One may be called upon to identify oneself—to characterize oneself, to locate oneself vis-a-vis known others, to situate oneself in a narrative, to place oneself in a category—in any number of different contexts. In modern settings, which multiply interactions with others not personally known, such occasions for identification are particularly abundant. They include innumerable situations of everyday life as well as more formal and official contexts. How one identifies oneself—and how one is identified by others—may vary greatly from context to context; self-and other-identification are fundamentally situational and contextual.
One key distinction is between relational and categorical modes of identification. One may identify oneself (or another person) by position
42 • Beyond "Identity"
in a relational web (a web of kinship, for example, or of friendship, patron-client ties, or teacher-student relations). On the other hand, one may identify oneself (or another person) by membership in a class of persons sharing some categorical attribute (such as race, ethnicity, language, nationality, citizenship, gender, sexual orientation, etc.). Calhoun (1997: 36ff.) has argued that, while relational modes of identification remain important in many contexts, categorical identification has assumed ever greater importance in modern settings.
Another basic distinction is between self-identification and the identification and categorization of oneself by others.39 Self-identification takes place in dialectical interplay with external identification, and the two need not converge.40 External identification is itself a varied process. In the ordinary ebb and flow of social life, people identify and categorize others, just as they identify and categorize themselves. But there is another key type of external identification that has no counterpart in the domain of self-identification: the formalized, codified, objectified systems of categorization developed by powerful, authoritative institutions.
The modern state has been one of the most important agents of identification and categorization in this latter sense. In culturalist extensions of the Weberian sociology of the state, notably those influenced by Bourdieu and Foucault, the state monopolizes, or seeks to monopolize, not only legitimate physical force but also legitimate symbolic force, as Bourdieu puts it. This includes the power to name, to identify, to categorize, to state what is what and who is who. There is a burgeoning sociological and historical literature on such subjects. Some scholars have looked at "identification" quite literally: as the attachment of definitive markers to an individual via passport, fingerprint, photograph, and signature, and the amassing of such identifying documents in state repositories (Noiriel 1991, 1993,1998; Fraenkel 1992; Torpey 2000; Caplan and Torpey 2001). Other scholars emphasize the modern state's efforts to inscribe its subjects onto a classificatory grid (Scott 1998: 76-83), to identify and categorize people in relation to gender, religion, occupation, property ownership, ethnicity, literacy, criminality, health, or sanity. Censuses apportion people across these categories,41 and institutions—from schools to prisons—sort out individuals in relation to them. To Foucauldians in particular, these individualizing and aggregating modes of identification and classification are at the core of what defines "governmentality" in a modern state (Foucault 1991 ).42
Beyond "Identity" • 43
The state is thus a powerful "identifier," not because it can create "identities" in the strong sense—in general, it cannot—but because it has the material and symbolic resources to impose the categories, classificatory schemes, and modes of social counting and accounting with which bureaucrats, judges, teachers, and doctors must work and to which nonstate actors must refer.43 But the state is not the only "identifier" that matters. As Tilly (1998) has shown, categorization does crucial "organizational work" in all kinds of social settings, including families, firms, schools, social movements, and bureaucracies of all kinds. Even the most powerful state does not monopolize the production and diffusion of identifications and categories; and those that it does produce may be contested. The literature on social movements— "old" as well as "new"—is rich in evidence on how movement leaders challenge official identifications and propose alternative ones. It highlights leaders' efforts to get members of putative constituencies to identify themselves in a certain way, to see themselves—for a certain range of purposes—as "identical" with one another, to identify emotionally as well as cognitively with one another (e.g., Melucci 1995; Martin 1995).
The social movement literature has valuably emphasized the interactive, discursively mediated processes through which collective solidarities and self-understandings develop. Our reservations concern the move from discussing the work of identification—the efforts to build a collective self-understanding—to positing "identity" as their necessary result. By considering authoritative, institutionalized modes of identification together with alternative modes involved in the practices of everyday life and the projects of social movements, one can emphasize the hard work and long struggles over identification as well as the uncertain outcomes of such struggles. However, if the outcome is always presumed to be an "identity"—however provisional, fragmented, multiple, contested, and fluid—one loses the capacity to make key distinctions.
"Identification," we noted above, invites specification of the agents that do the identifying. Yet identification does not require a specifiable "identifier"; it can be pervasive and influential without being accomplished by discrete, specified persons or institutions. Identification can be carried more or less anonymously by discourses or public narratives (Hall 1996; Somers 1994). Although close analysis of such discourses or narratives might well focus on their instantiations in particular
44 • Beyond "Identity"
discursive or narrative utterances, their force may depend not on any particular instantiation but on their anonymous, unnoticed permeation of our ways of thinking and talking and making sense of the social world.
There is one further meaning of "identification," alluded to above, that is largely independent of the cognitive, characterizing, classificatory meanings discussed so far. This is the psychodynamic meaning, derived originally from Freud.44 While the classificatory meanings involve identifying oneself (or someone else) as someone who fits a certain description or belongs to a certain category, the psychodynamic meaning involves identifying oneself emotionally with another person, category, or collectivity. Here again, "identification" calls attention to complex (and often ambivalent) processes, while the term "identity," designating a condition rather than a process, implies too easy a fit between the individual and the social.
Self-Understanding and Social Location. "Identification" and "categorization" are active, processual terms, derived from verbs, and calling to mind particular acts of identification and categorization performed by particular identifiers and categorizers. But we need other kinds of terms as well to do the varied work done by "identity." Recall that one key use of "identity" is to conceptualize and explain action in a noninstrumental, nonmechanical manner. In this sense, the term suggests ways in which individual and collective action can be governed by particularistic understandings of self and social location rather than by putatively universal, structurally determined interests. "Self-understanding" is therefore the second term we would propose as an alternative to "identity." It is a dispositional term that designates what might be called "situated subjectivity": one's sense of who one is, of one's social location, and of how (given the first two) one is prepared to act. As a dispositional term, it belongs to the realm of what Pierre Bourdieu (1990a) has called sens pratique, the practical sense— at once cognitive and emotional—that persons have of themselves and their social world.
The term "self-understanding," it is important to emphasize, does not imply a distinctively modern or Western understanding of the "self" as a homogeneous, bounded, unitary entity. A sense of who one is can take many forms. The social processes through which persons understand and locate themselves may in some instances involve the psychoanalyst's couch and in others participation in spirit possession
Beyond "Identity" • 45
cults.45 In some settings, people may understand and experience themselves in terms of a grid of intersecting categories; in others, in terms of a web of connections of differential proximity and intensity. Hence the importance of seeing self-understanding and social locatedness in relation to each other, and of emphasizing that both the bounded self and the bounded group are culturally specific rather than universal forms.
Like the term "identification," "self-understanding" lacks the reifying connotations of "identity." Yet it is not restricted to situations of flux and instability. Self-understandings may be variable across time and across persons, but they may be stable. Semantically, "identity" implies sameness across time or persons; hence the awkwardness of continuing to speak of "identity" while repudiating the implication of sameness. "Self-understanding," by contrast, has no privileged semantic connection with sameness or difference.
Two closely related terms are "self-representation" and "self-identification." Having discussed "identification" above, we simply observe here that, while the distinction is not sharp, "self-understandings" may be tacit; even when they are formed, as they ordinarily are, in and through prevailing discourses, they may exist, and inform action, without themselves being discursively articulated. "Self-representation" and "self-identification," on the other hand, suggest at least some degree of explicit discursive articulation.
"Self-understanding" cannot, of course, do all the work done by "identity." We note here three limitations of the term. First, it is a subjective, autoreferential term. As such, it designates one's own understanding of who one is. It cannot capture others' understandings, even though external categorizations, identifications, and representations may be decisive in determining how one is regarded and treated by others, indeed in shaping one's own understanding of oneself. At the limit, self-understandings may be overridden by overwhelmingly coercive external categorizations.46
Second, "self-understanding" would seem to privilege cognitive awareness. As a result, it would seem not to capture—or at least not to highlight—the affective or cathectic processes suggested by some uses of "identity." Yet self-understanding is never purely cognitive; it is always affectively tinged or charged, and the term can certainly accommodate this affective dimension. However, it is true that the emotional dynamics are better captured by the term "identification" (in its psychodynamic meaning).
46 ■ Beyond "Identity"
Finally, as a term that emphasizes situated subjectivity, "self-understanding" does not capture the objectivity claimed by strong understandings of identity. Strong, objectivist conceptions of identity permit one to distinguish "true" identity (characterized as deep, abiding, and objective) from "mere" self-understanding (superficial, fluctuating, and subjective). If identity is something to be discovered, and something about which one can be mistaken, then one's momentary self-understanding may not correspond to one's abiding, underlying identity. However analytically problematic these notions of depth, constancy, and objectivity may be, they do at least provide a reason for using the language of identity rather than that of self-understanding.
Weak conceptions of identity provide no such reason. It is clear from the constructivist literature why weak understandings of identity are weak; but it is not clear why they are conceptions of identity. In this literature, it is the various soft predicates of identity—constructedness, contestedness, contingency, instability, multiplicity, fluidity—that are emphasized and elaborated, while that which they are predicated of— identity itself—is taken for granted and seldom explicated. When identity itself is elucidated, it is often represented as a sense of who one is (Berger 1974: 162), or a self-conception (Calhoun 1991: 68), that is, as something that can be captured in a straightforward way by "self-understanding." This term lacks the theoretical pretensions of "identity," but this should count as an asset, not a liability.
Commonality, Connectedness, Groupness. One particular form of affectively charged self-understanding that is often designated by "identity"—especially in discussions of race, religion, ethnicity, nationalism, gender, sexuality, social movements, and other phenomena conceptualized as involving collective identities—deserves separate mention here. This is the emotionally laden sense of belonging to a distinctive, bounded group, involving both a felt solidarity or oneness with fellow group members and a felt difference from or even antipathy to specified outsiders.
The problem is that "identity" is used to designate both such strongly groupist, exclusive, affectively charged self-understandings and much looser, more open self-understandings, involving some sense of affinity or affiliation, commonality or connectedness to particular others, but lacking a sense of overriding oneness vis-á-vis some constitutive "other."47 Both the tightly groupist and the more loosely affiliative forms of self-understanding—as well as the transitional
Beyond "Identity" ■ 47
forms between these polar types—are important, but they shape personal experience and condition social and political action in sharply differing ways.
Rather than stirring all self-understandings based on race, religion, ethnicity, and so on into the great conceptual melting pot of "identity," we would do better to use a more differentiated analytical language. Terms such as commonality, connectedness, and groupness could be usefully employed here in place of the all-purpose "identity." This is the third cluster of terms we propose. "Commonality" denotes the sharing of some common attribute, "connectedness" the relational ties that link people. Neither commonality nor connectedness alone engenders "groupness"—the sense of belonging to a distinctive, bounded, solidary group. But commonality and connectedness together may indeed do so. This was the argument Tilly (1978: 62 ff.) put forward some time ago, building on Harrison White's idea of the "catnet," a set of persons comprising both a category, sharing some common attribute, and a network. Tilly's suggestion that groupness is a joint product of the "catness" and "netness"—categorical commonality and relational connectedness—is suggestive. But we would propose two emendations.
First, categorical commonality and relational connectedness need to be supplemented by a third element, what Max Weber called a Zusammengehörigkeitsgefühl, a feeling of belonging together. Such a feeling may indeed depend in part on the degrees and forms of commonality and connectedness, but it will also depend on other factors such as particular events, their encoding in compelling public narratives, prevailing discursive frames, and so on. Second, relational connectedness, or what Tilly calls "netness," while crucial in facilitating the sort of collective action Tilly was interested in, is not always necessary for "groupness." A strongly bounded sense of groupness may rest on categorical commonality and an associated feeling of belonging together with minimal or no relational connectedness. This is typically the case for large-scale collectivities such as "nations": when a diffuse self-understanding as a member of a particular nation crystallizes into a strongly bounded sense of groupness, this is likely to depend not on relational connectedness, but rather on a powerfully imagined and strongly felt commonality.48
The point is not, as some partisans of network theory have suggested, to turn from commonality to connectedness, from categories to networks, from shared attributes to social relations.49 Nor is it to
48 • Beyond "Identity"
celebrate fluidity and hybridity over belonging and solidarity. The point in suggesting this last set of terms is rather to develop an analytical idiom sensitive to the multiple forms and degrees of commonality and connectedness, and to the widely varying ways in which actors (and the cultural idioms, public narratives, and prevailing discourses on which they draw) attribute meaning and significance to them. This will enable us to distinguish instances of strongly binding, vehemently felt groupness from more loosely structured, weakly constraining forms of affinity and affiliation.
Three Cases: "Identity" and Its Alternatives in Context
Having surveyed the work done by "identity," indicated some limitations and liabilities of the term, and suggested a range of alternatives, we seek now to illustrate our argument—both the critical claims about "identity" and the constructive suggestions regarding alternative idioms—through a consideration of three cases. In each case, we suggest, the identitarian focus on bounded groupness limits the sociological—and the political—imagination, while alternative analytical idioms can help open up both.
A Case from Africanist Anthropology: "The" Nuer
Identitarian thinking in African studies is most extreme, and most problematic, in journalistic accounts that see primordial "tribal identities" as the main cause of Africa's woes. Africanist scholars have long been troubled by this reductive vision and, influenced by Barth (1969), developed a constructivist alternative well before such an approach had a name (Cohen 1969; Lonsdale 1977).50 The argument that ethnic groups are not primordial but the products of history—including the reifying of cultural difference through imposed colonial identifications—became a staple of African studies. Even so, scholars have tended to emphasize boundary formation rather than boundary crossing, the constitution of groups rather than the development of networks. And while Africanists have been critical of the concepts of "tribe," "race," and "ethnicity," they often still use "identity" in an unexamined way (e.g., Dubow et al. 1994). Acknowledgment that identity is multiple is rarely followed by explanation of why that
Beyond "Identity" • 49
which is multiple should be considered identity.51 In this context, it is worth going back to a classic of African ethnology: E. E. Evans-pritchard's (1940) book The Nuer.
Based on research in Northeast Africa in the 1930s, The Nuer describes a distinctively relational mode of identification, self-understanding, and social location, one that construes the social world in terms of the degree and quality of connection among people rather than in terms of categories, groups, or boundaries. Social location is defined in the first instance in terms of lineage, consisting of the descendants of one ancestor reckoned through a socially conventional line: patrilineal, via males in the case of the Nuer, via females or more rarely via double descent systems in some other parts of Africa. Children belong to the lineage of their fathers, and while relationships with the mother's kin are not ignored, they are not part of the descent system. A segmentary lineage can be diagrammed as shown in Figure 1.
Everybody in this diagram is related to everybody else, but in different ways and to different degrees. One might be tempted to say that the people marked in circle A constitute a group, with an "identity" of A. as distinct from those in circle B, with an "identity" of B. Yet the very move which distinguishes A and B also shows their relatedness, as one moves back two generations and finds a common ancestor. If someone in
rr"n n   r~n \
l Mais > Female Marriage
Figure \. A segmentary patrilineage. Lines represent descent; marriage partners come from another lineage; children of daughters belong to the lineage of the husband and are not shown; children of sons belong to this lineage and are represented here.
SO • Beyond "Identity"
set A gets into a conflict with someone in set B, such a person may well try to invoke the commonality of "A-ness" to mobilize people against B. But someone genealogically older than these parties can invoke the linking ancestors to cool things off. This practice—and the ever-present possibility of construing relatedness on different levels—fosters relational rather than categorical understandings of social location.
One could argue that this patrilineage as a whole constitutes an identity, distinct from other lineages. But Evans-Pritchard's point is that segmentation characterizes an entire social order, and that lineages themselves are related to one another as male and female lineage members are to each other. Virtually all segmentary societies insist on exogamy; in evolutionary perspective, this may reflect the advantages of cross-lineage connectedness. The male-centered lineage diagram presumes another set of relationships, through women who are born into the lineage of their fathers but whose sons and daughters belong to the lineage they married into.
One could then argue that all the lineages connected through intermarriage constitute the "Nuer" as an identity distinct from "Dinka" or any of the other groups in the region. But recent work in African history offers a more nuanced perspective. The genealogical construction of re-lationality offers possibilities for extension that are obscured by the contemporary scholar's tendency to look for a neat boundary between inside and outside. Marriage relations could be extended beyond the Nuer (both via reciprocal arrangements and by forcing captive women into marriage). Strangers—encountered via trade, migration, or other form of movement—could be incorporated as fictive kin or more loosely linked to a patrilineage via blood brotherhood. The people of northeastern Africa migrated extensively, as they tried to find better ecological niches or as lineage segments moved in and out of relations with each other. Traders stretched their kinship relations over space, formed a variety of relationships at the interfaces with agricultural communities, and sometimes developed a lingua franca to foster communication across extended networks.52 In many parts of Africa, one finds certain organizations—religious shrines, initiation societies—that cross linguistic and cultural boundaries, offering what Paul Richards (1996) calls a "common 'grammar' " of social experience within regions, for all the cultural variation and political differentiation that they contain.
The problem with subsuming these forms of relational connectedness under the "social construction of identity" is that linking and separating
Beyond "Identity" ■ 51
get called by the same name, making it harder to grasp the processes, causes, and consequences of differing patterns of crystallizing difference and forging connections. Africa was far from a paradise of sociability, but both war and peace involved flexible patterns of affiliation as well as differentiation.
Sliding scales of genealogical connection are not unique to small-scale "tribal" society. Kinship networks structure larger-scale political organizations as well, with authoritative rulers and elaborate hierarchies of command. African kings asserted their authority by developing patrimonial relations with people from different lineages, creating a core of support that cut across lineage affiliations, but they also used lineage principles to consolidate their own power, contracting marriage alliances and expanding the royal lineage (Lonsdale 1981). In almost all societies, kinship concepts serve as symbolic and ideological resources, yet while they shape norms, self-understandings and perceptions of affinity, they do not necessarily produce kinship "groups" (Guyer 1981; Amselle 1990).
To a greater extent than earlier forms of domination, colonial rule sought to map people with putatively common characteristics onto territories. These imposed identifications could be powerful, but their effects depended on the actual relationships and symbolic systems that colonial officials—and indigenous cultural entrepreneurs—had to work with, and on the countervailing efforts of others to develop, articulate, and maintain different sorts of affinities and self-understandings. The colonial era did indeed witness complex struggles over identification, but it flattens our understanding of these struggles to see them as producing "identities." People could live with shadings—and continued to do so in everyday undertakings even when political lines were sharply drawn.
Sharon Hutchinson's (1995) remarkable reanalysis of Evans-Pritchard's "tribe" takes such an argument into a contemporary, conflict-ridden situation. Her aim is "to call into question the very idea of 'the Nuer' as a unified ethnic identity" (29). She points to the fuzzi-ness of the boundaries of people now called Nuer: culture and history do not follow such lines. And she suggests that Evans-Pritchard's segmentary schema gives excessive attention to the dominant male elders of the 1930s, and not enough to women, men in less powerful lineages, or younger men and women. In this analysis, it not only becomes difficult to see Nuerness as an identity, but imperative to examine with precision how people tried both to extend and to consolidate connections.
52 • Beyond "Identity"
Bringing the story up to the era of civil war in the southern Sudan in the 1990s, Hutchinson refuses to reduce the conflict to one of cultural or religious difference between the warring parties and insists instead on a deep analysis of political relationships, struggles for economic resources, and spatial connections.
hi much of modern Africa, indeed, some of the most bitter conflicts have taken place within collectivities that are relatively uniform culturally and linguistically (Rwanda, Somalia) and between loose economic and social networks based more on patron-client relations than ethnic affiliation (Angola, Sierra Leone), as well as in situations where cultural distinction has been made into a political weapon (Kwa Zulu in South Africa).53 To explain present or past conflict in terms of how people construct and fight for their "identities" risks providing a prefabricated, presentist, teleological explanation that diverts attention from questions such as those addressed by Hutchinson.54
East European Nationalism
We have argued that the language of identity, with its connotations of boundedness, groupness, and sameness, is conspicuously ill suited to the analysis of segmentary lineage societies—or of present-day conflicts in Africa. One might accept this point yet argue that identitarian language is well suited to the analysis of other social settings, including our own, where public and private "identity talk" is widely current. But we are not arguing only that the concept of identity does not "travel" well, that it cannot be universally applied to all social settings. We want to make a stronger argument: that "identity" is neither necessary nor helpful as a category of analysis even where it is widely used as a category of practice. To this end, we briefly consider East European nationalism and identity politics in the United States.
Historical and social scientific writing on nationalism in Eastern Europe—to a much greater extent than writing on social movements or ethnicity in North America—has been characterized by relatively strong or hard understandings of group identity. Many commentators have seen the postcommunist resurgence of ethnic nationalism in the region as springing from robust and deeply rooted national identities—from identities strong and resilient enough to have survived decades of repression by ruthlessly antinational communist regimes. But this "return-of-the-repressed" view is problematic.55
Consider the former Soviet Union. To see national conflicts as
Beyond "Identity" ■ 53
struggles to validate and express identities that had somehow survived the regime's attempts to crush them is unwarranted. Although anti-nationalist, and of course brutally repressive in all kinds of ways, the Soviet regime was anything but antinational.56 Far from ruthlessly suppressing nationhood, the regime went to unprecedented lengths in institutionalizing and codifying it. It carved up Soviet territory into more than fifty putatively autonomous national "homelands," each "belonging" to a particular ethnonational group; and it assigned each citizen an ethnic "nationality," which was ascribed at birth on the basis of descent, registered in personal identity documents, recorded in bureaucratic encounters, and used to control access to higher education and employment. In doing so, the regime was not simply recognizing or ratifying a preexisting state of affairs; it was newly constituting both persons and places as national.57 In this context, strong understandings of national identity as deeply rooted in the precommunist history of the region, frozen or repressed by a ruthlessly antinational regime, and returning with the collapse of communism are at best anachronistic, at worst simply scholarly rationalizations of nationalist rhetoric.
What about weak, constructivist understandings of identity? Con-structivists might concede the importance of the Soviet system of institutionalized multinationality, and interpret this as the institutional means through which national identities were constructed. But why should we assume it is "identity" that is constructed in this fashion? To assume that it is risks conflating a system of identification or categorization with its presumed result, identity. Categorical group denominations—however authoritative, however pervasively institutionalized—cannot serve as indicators of real "groups" or robust "identities."
Consider for example the case of "Russians" in Ukraine. At the time of the 1989 census, some 11.4 million residents of Ukraine identified their "nationality" as Russian. But the precision suggested by this census data, even when rounded to the nearest hundred thousand, is entirely spurious. The very categories "Russian" and "Ukrainian," as designators of putatively distinct ethnocultural nationalities, or distinct "identities," are deeply problematic in the Ukrainian context, where rates of intermarriage have been high, and where millions of nominal Ukrainians speak only or primarily Russian. One should be skeptical of the illusion of "identity" or bounded groupness created by the census, with its exhaustive and mutually exclusive categories. One can imagine circumstances in which "groupness" might emerge
54 • Beyond "Identity"
among nominal Russians in Ukraine, but such groupness cannot be taken as given.5S
The formal institutionalization and codification of ethnic and national categories implies nothing about the depth, resonance^ or power of such categories in the lived experience of the persons so categorized. A strongly institutionalized ethnonational classificatory system makes certain categories readily and legitimately available for the representation of social reality, the framing of political claims, and the organization of political action. This is itself a fact of great significance, and the breakup of the Soviet Union cannot be understood without reference to it. But it does not entail that these categories will have a significant role in framing perception, orienting action, or shaping self-understanding in everyday life—a role that is implied by even constructivist accounts of "identity."
The extent to which official categorizations shape self-understandings, and the extent to which the population categories constituted by states or political entrepreneurs approximate real "groups," are open questions that can only be addressed empirically. The language of "identity" is more likely to hinder than to help the posing of such questions, for it blurs what needs to be kept distinct: external categorization and self-understanding, objective commonality and subjective groupness.
Consider one final, non-Soviet example. The boundary between Hungarians and Romanians in Transylvania is certainly sharper than that between Russians and Ukrainians in Ukraine. Here too, however, group boundaries are considerably more porous and ambiguous than is widely assumed. The language of both politics and everyday life, to be sure, is rigorously categorical, dividing the population into mutually exclusive ethnonational categories, and making no allowance for mixed or ambiguous forms. But this categorical code, important though it is as a constituent element of social relations, should not be taken for a faithful description of them. Reinforced by identitarian entrepreneurs on both sides, the categorical code obscures as much as it reveals about self-understandings, masking the fluidity and ambiguity that arise from mixed marriages, from bilingualism, from migration, from Hungarian children attending Romanian-language schools, from intergenerational assimilation, and—perhaps most important—from sheer indifference to the claims of ethnocul-tural nationality.59
Even in its constructivist guise, the language of "identity" disposes
Beyond "Identity" • 55
us to think in terms of bounded groupness. It does so because even constructivist thinking on identity takes the existence of identity as axiomatic. Identity is always already "there," as something that individuals and groups "have," even if the content of particular identities, and the boundaries that mark groups off from one another, are conceptualized as always in flux. Even constructivist language tends therefore to objectify "identity," to treat it as a "thing," albeit a malleable one, that people "have," "forge," and "construct." This tendency to objectify "identity" deprives us of analytical leverage and constricts political possibilities. It makes it more difficult for us to treat "groupness" and "boundedness" as emergent properties of particular structural or conjunctural settings rather than as always already there in some form.
Identity Claims and the Enduring Dilemmas of "Race" in the United States
The language of identity has been particularly prominent in the United States in recent decades. It has served both as an idiom of analysis in the social sciences and humanities and as an idiom in which to articulate experience, mobilize loyalty, and advance claims in everyday social and political practice.
The pathos and resonance of identity claims in the contemporary United States have many sources, but one of the most profound is that central problem of American history—the importation of enslaved Africans, the persistence of racial oppression, and the range of African-American responses to it. The African-American experience of "race" as both imposed categorization and self-identification has been important not only in its own terms, but also—from the late 1960s on—as a template for other identity claims, including those based on gender and sexual orientation as well as those formulated in terms of ethnicity or race (Gitlin 1995: 134).
In response to the cascading identitarian claims of the last three decades, public discourse, political argument, and scholarship in nearly every field of the social sciences and humanities have been transformed. There is much that is valuable in this process. History textbooks and prevailing public narratives tell a much richer and more inclusive story than those of a generation ago. Specious forms of uni-versaiism—the Marxist category of "worker" who always appears in the guise of a male, the liberal category of "citizen" who turns out to
56 • Beyond "Identity"
be white—have been powerfully exposed. "First-generation" identi-tarian claims themselves—and scholarly literatures informed by them— have been criticized for their blindness to cross-cutting particularities: African-American movements for acting as if African-American women did not have gender-specific concerns, feminists for focusing on white, middle-class women.
Constructivist arguments have had a particular influence in Americanist circles, allowing scholars to stress the contemporary importance of imposed identifications and the self-understandings that have evolved in dialectical interplay with them, while emphasizing that such self- and other-identified "groups" are not primordial but historically produced. The treatment of race in the historiography of the United States is an excellent example.60 Well before "social construction" became a fashionable term, scholars were showing that far from being a given dimension of America's past, race as a political category originated in the same moment as America's republican and populist impulses. Edmund Morgan (1975) argued that in early eighteenth-century Virginia, white indentured servants and black slaves shared a subordination that was not sharply differentiated; they sometimes acted together. It was when Virginian planter elites started to mobilize against the British that they needed to draw a sharp boundary between the politically included and the excluded, and the fact that black slaves were more numerous and replaceable as laborers and less plausible as political supporters led to a marking of distinction, which poor whites could in turn use to make claims.61 Subsequent historical work has identified key moments of redefinition of racial boundaries in the United States, as well as moments in which other sorts of ties and affiliation became salient. Whiteness (Roediger 3991) and blackness were both historically created and variably salient categories. Comparative historians, meanwhile, have shown that the construction of race can take still more varied forms, and have highlighted the peculiarity of the American system of racial classification, based on the "one-drop" rule.62
American history thus reveals the power of imposed identifications, but it also reveals the complexity of the self-understandings of people defined by circumstances they did not control. Pre-Civil War collective self-definitions situated black Americans in particular ways in regard to Africa—often seeing an African (or an "Ethiopian") origin as placing them close to the heartlands of Christian civilization. Yet early back-to-Africa movements often treated Africa as a cultural tabula rasa or as a
Beyond "Identity" ■ 57
fallen civilization to be redeemed by African-American Christians.63 Self-identification as a diasporic "people" did not necessarily imply claiming cultural commonality. One can write the history of African-American self-understanding as the development of a black nationality, or one can explore the interplay of such a sense of collectivity with the efforts of African-American activists to articulate different kinds of political ideologies and to develop connections with other radicals. The important point is to keep in mind the range of possibilities and the seriousness with which they were debated.
It is not the historical analysis of social construction as such that is problematic, but the presumptions about what it is that is constructed. Scholars have been more inclined to focus on the construction of racial (or other) "identities" than on that of other, looser forms of affinity and commonality. Setting out to write about "identifications" as they emerge, crystallize, and fade away in particular social and political circumstances may well inspire a rather different history than setting out to write of an "identity," which links past, present, and future in a single word.
Cosmopolitan interpretations of American history have been criticized for taking the pain out of the distinctive ways in which that history has been experienced by African Americans: above all the pain of enslavement and discrimination, and of struggle against them. This has indeed been distinctive to African Americans (Lott 1996). Calls to understand the particularity of experience therefore resonate powerfully. Yet there are risks of flattening a complex history through a focus on a singular "identity," though there may be gains as well as losses in such a focus, as thoughtful participants in debates over the politics of race have made clear.64
Yet to subsume further under the generic category of "identity" the historical experiences and allegedly common cultures of other "groups" as disparate as women and the elderly, Native Americans and gay men, poor people and the disabled is not in any obvious way more respectful of the pain of particular histories than are the uni-versalist rhetorics of justice or human rights. And the assignment of individuals to such "identities" leaves many people—who have experienced the uneven trajectories of ancestry and the variety of innovations and adaptations that constitute culture—caught between a hard identity that does not quite fit and a soft rhetoric of hybridity, multiplicity, and fluidity that offers neither understanding nor solace.
58 ■ Beyond "Identity"
This conceptually impoverished identitarian sociology, in which the "intersection" of race, class, gender, sexual orientation, and perhaps one or two other categories generates a set of all-purpose conceptual boxes, has become powerfully entrenched in American academia in the 1990s—not only in the social sciences, cultural studies, and ethnic studies, but also in literature and political philosophy. In the remainder of this section, we shift our angle of vision and consider the implications of the use of this identitarian sociology in the latter domain.
"A moral philosophy," wrote Alisdair Maclntyre (1981: 22), "presupposes a sociology"; the same holds a fortiori of political theory. A weakness of much contemporary political theory is that it is built on a dubious sociology—indeed precisely on the reductively groupist representation of the social world just mentioned. We are not taking the side of "universality" against "particularity" here. Rather, we are suggesting that the identitarian language and groupist social ontology that informs much contemporary political theory occludes the problematic nature of "groupness" itself and forecloses other ways of conceptualizing particular affiliations and affinities.
There is a considerable literature now that is critical of the idea of universal citizenship. Iris Marion Young, one of the most influential of such critics, proposes instead an ideal of group-differentiated citizenship, built on group representation and group rights. The notion of an "impartial general perspective," she argues, is a myth, since "different social groups have different needs, cultures, histories, experiences, and perceptions of social relations." Citizenship should not seek to transcend such differences, but should recognize and acknowledge them as "irreducible" (Young 1989: 257, 258; 1990).
What sorts of differences should be ratified with special representation and rights? The differences in question are those associated with "social groups," defined as "comprehensive identities and ways of life," and distinguished from mere aggregates on the one hand—arbitrary classifications of persons according to some attribute—and from voluntary associations on the other. Special rights and representation would be accorded not to all social groups, but to those who suffer from at least one of five forms of oppression. In practice, this means "women, blacks, Native Americans, Chicanos, Puerto Ricans and other Spanish-speaking Americans, Asian Americans, gay men, lesbians, working-class people, poor people, old people, and mentally and physically disabled people" (Young 1989: 267, 261).
J:
Beyond "Identity" • 59
What constitutes the "groupness" of these "groups"? What makes them grouPs father than categories around which self- and other-identifications may, but need not necessarily, crystallize? This is not addressed by Young. She assumes that distinctive histories, experiences, and social locations endow these "groups" with different "capacities, needs, culture, and cognitive styles" and with "distinctive understandings of all aspects of the society and unique perspectives on social issues" (Young 1989: 267, 268). Social and cultural heterogeneity is construed here as a juxtaposition of internally homogeneous, externally bounded blocs. The "principles of unity" that Young repudiates at the level of the polity as a whole—because they "hide difference"—are reintroduced, and continue to hide difference, at the level of the constituent "groups."
At stake in arguments about group-differentiated or "multicultural" citizenship are important issues that have been long debated outside as well as inside the academy, all having to do in one way or another with the relative weight and merits of universalist and particularist claims.65 Sociological analysis cannot and should not seek to resolve this robust debate, but it can seek to shore up its often shaky sociological foundations. It can offer a richer vocabulary for conceptualizing social and cultural heterogeneity and particularity. Moving beyond identitarian language opens up possibilities for specifying other kinds of connectedness, other idioms of identification, other styles of self-understanding, other ways of reckoning social location. To paraphrase what Adam Przeworski (1977) said long ago about class, cultural struggle is a struggle about culture, not a struggle between cultures. Activists of identity politics deploy the language of bounded groupness not because it reflects social reality, but precisely because groupness is ambiguous and contested. Their groupist rhetoric has a performative, constitutive dimension, contributing, when it is successful, to the making of the groups it invokes (Bourdieu 1991b, 1991c).
Here there is a gap between normative arguments and activist idioms that take bounded groupness as axiomatic and historical and sociological analyses that emphasize contingency, fluidity, and variability. At one level there is a real-life dilemma: preserving cultural distinctiveness depends at least in part on maintaining bounded groupness and hence on policing the "exit option," and accusations of "passing" and of betraying one's roots serve as modes of discipline (Laitin 1995a). Critics of such policing, however, would argue that a
60 • Beyond "Identity" ■
liberal polity should protect individuals from the oppressiveness of social groups as well as that of the state. At the level of social analysis, though, the dilemma is not a necessary one. We are not faced with a stark choice between a universalist, individualist analytical idiom and one that is identitarian and groupist. Framing the options in this way misses the variety of forms (other than bounded groups) which affinity, commonality, and connectedness can take—hence our emphasis on the need for a more supple vocabulary. We are not arguing for any specific stance on the politics of cultural distinction and individual choice, but rather for a vocabulary of social analysis that helps open up and illuminate the range of options. The politics of group "coalition" that is celebrated by Young and others, for example, certainly has its place, but the groupist sociology that underlies this particular form of coalition politics—with its assumption that bounded groups are the basic building blocks of political alliances—is unduly constraining.66
None of this belies the importance of current debates over "univer-salistic" and "particularistic" conceptions of social justice. Our point is that the identitarian focus on bounded groupness does not help in posing these questions. We need not in fact choose between an American history flattened into the experiences and "cultures" of bounded groups and one equally flattened into a single "national" story. Reducing the complex and dynamic heterogeneity of American society and history to a formulaic pluralism of identity groups hinders rather than helps the work of understanding the past and pursuing social justice in the present.
Conclusion: Particularity and the Politics of "Identity"
We have not made an argument about identity politics. Nonetheless, the argument does have political as well as intellectual implications. Some will think these regressive, and will worry that the argument undermines the basis for making particularistic claims. That is neither our intention nor a valid inference from what we have written.
To persuade people that they are one; that they comprise a bounded, distinctive, solidary group; that their internal differences do not matter, at least for the purpose at hand—this is a normal and necessary part of politics, and not only of what is ordinarily characterized as "identity
Beyond "Identity" • 61
politics." It is not all of politics; and we do indeed have reservations about the way in which the routine recourse to identitarian framing may foreclose other equally important ways of framing political claims. But we do not seek to deprive anyone of "identity" as a political tool, or to undermine the legitimacy of making political appeals in identitarian terms.
Our argument has focused, rather, on the use of "identity" as an analytical concept. Throughout the essay, we have asked what work the concept is supposed to do, and how well it does it. We have argued that the concept is deployed to do a great deal of analytical work— much of it legitimate and important. "Identity," however, is ill suited to perform this work, for it is riddled with ambiguity, riven with contradictory meanings, and encumbered by reifying connotations. Qualifying the noun with strings of adjectives—specifying that identity is multiple, fluid, constantly renegotiated, and so on—does not solve the problem. It yields little more than a suggestive oxymoron—a multiple singularity, a fluid crystallization—and begs the question of why one should use the same term to designate all this and more. Alternative analytical idioms, we have argued, can do the necessary work without the attendant confusion.
At issue here is not the legitimacy or importance of particularistic claims, but how best to conceptualize them. People everywhere and always have particular ties, self-understandings, stories, trajectories, histories, predicaments. And these inform the sorts of claims they make. To subsume such pervasive particularity under the fiat, undifferentiated rubric of "identity," however, does nearly as much violence to its unruly and multifarious forms as would an attempt to subsume it under "universalist" categories such as "interest."
Construing particularity in identitarian terms, moreover, constricts the political as well as the analytical imagination. It points away from a range of possibilities for political action other than those rooted in putatively shared identity—and not only those that are praised or damned as "universalist." Identitarian political advocates, for example, construe political cooperation in terms of the building of coalitions between bounded identity groups. This is one mode of political cooperation, but not the only one.
Margaret Keck and Kathryn Sikkink (1998), for example, have drawn attention to the importance of "transnational issue networks," from the antislavery movement of the early nineteenth century to
62 • Beyond "Identity"
Beyond "Identity" • 63
international campaigns about human rights, ecology, and women's rights in recent years. Such networks necessarily cross cultural as well as state boundaries and link particular places and particularistic claims to wider concerns. To take one instance, the antiapartheid movement brought together South African political organizations that were themselves far from united—some sharing "universalist" ideologies, some calling themselves "Africanist," some asserting a quite local, culturally defined "identity"—with international church groups, labor unions, pan-African movements for racial solidarity, human rights groups, and so on. Particular groups moved in and out of cooperative arrangements within an overall network; conflict among opponents of the apartheid state was sometimes bitter, even deadly. As the actors in the network shifted, the issues at stake were reframed. At certain moments, for example, issues amenable to international mobilization were highlighted, while others—of great concern to some would-be participants—were marginalized (Klotz 1995).fi7
Our point is not to celebrate such networks over identitarian social movements or group-based claims. Networks are no more intrinsically virtuous than identiarian movements and groups are intrinsically suspect. Politics—in southern Africa or elsewhere—is hardly a confrontation of good universalists or good networks versus bad tribal-ists. Much havoc has been done by flexible networks built on clientage and focused on pillage and smuggling; such networks have sometimes been linked to "principled" political organizations; and they have often been connected to arms and illegal merchandise brokers in Europe, Asia, and North America. Multifarious particularities are in play, and one needs to distinguish between situations where they cohere around particular cultural symbols and situations where they are flexible, pragmatic, readily extendable. It does not contribute to precision of analysis to use the same words for the extremes of reifi-cation and fluidity, and everything in between.
To criticize the use of "identity" in social analysis is not to blind ourselves to particularity. It is rather to conceive of the claims and possibilities that arise from particular affinities and affiliations, from particular commonalities and connections, from particular stories and self-understandings, from particular problems and predicaments in a more differentiated manner. Social analysis has become massively, and
durably, sensitized to particularity in recent decades; and the literature on identity has contributed valuably to this enterprise. It is time now to go beyond "identity"—not in the name of an imagined univer-salism, but in the name of the conceptual clarity required for social analysis and political understanding alike.
Ethnicity as Cognition • 65
CHAPTER THREE
Ethnicity as Cognition
In recent years, categorization has emerged as a major focus of research in the study of ethnicity, as it has in many other domains. As long as ethnic groups were conceived as substantial, objectively definable entities, there was no reason to focus on categorization or classification. As constructivist stances have gained ground in the last quarter century, however, objectivist understandings of ethnicity (a term we use broadly here to include race and nationhood as well)' have been displaced by subjectivist approaches. The latter define ethnicity not in terms of objective commonalities but in terms of participants' beliefs, perceptions, understandings, and identifications. One consequence of this shift has been an increasing concern with categorization and classification.
We see the emergent concern with categorization as an incipient, and still implicit, cognitive turn in the study of ethnicity.2 We argue that the understanding of ethnicity can be enriched by making explicit this heretofore implicit cognitive reorientation, and by engaging research in cognitive psychology and cognitive anthropology.3 Doing so, we suggest, has far-reaching implications for how ethnicity should be conceived as both object and field of study. Cognitive perspectives provide resources for avoiding analytical "groupism"—the tendency to treat ethnic groups as substantial entities to which interests and
This chapter was coauthored with Mara Loveman and Peter Stamatov.
agency can be attributed—while helping to explain the tenacious hold of groupism in practice. They suggest strong reasons for treating race, ethnicity, and nationalism together rather than as separate subfields. And they afford new purchase on the old debate between primor-dialist and circumstantialist approaches to ethnicity.
We begin by reviewing historical, political, institutional, ethnographic, and microinteractional work on classification and categorization in the study of ethnicity, and by suggesting why cognitive perspectives have remained implicit in such work. We next consider expressly cognitive work on stereotypes, social categorization, and schemas, and we suggest ways in which the latter concept, in particular—designating more complex knowledge structures than categories— might be used in research on ethnicity. Finally, we consider the broader implications of cognitive perspectives, which suggest that ethnicity is fundamentally not a thing in the world, but a perspective on the world. Our aim is not to advance specific hypotheses, but to sensitize students of ethnicity to cognitive dimensions of the phenomenon, and to point to ways in which attention to these dimensions can fruitfully inform research in the field.
Categories and Categorization: An Incipient Cognitive Turn
Anthropology has a long-standing interest in classification and categorization (Dürkheim and Mauss, 1963 [1903]; Levi-Strauss, 1966; Needham 1979), so it is not surprising that anthropologists took the lead in highlighting the centrality of classification and categorization to ethnicity. The key work here is that of Norwegian anthropologist Fredrik Barth (1969). Ethnicity, Barth argued, is not a matter of shared traits or cultural commonalities, but rather of practices of classification and categorization, including both self-classification and the classification of (and by) others. Richard Jenkins (1997) and others have developed this idea further, emphasizing the interplay between self-identification and external categorization, and drawing attention to the various levels (individual, interactional, and institutional) and contexts (informal and formal) in which categorization occurs.4
While Barth formulated his argument with respect to ethnicity, it applies, mutatis mutandis, to race and nation as well. As its biological underpinning came to seem increasingly dubious, race came to be
66 • Ethnicity as Cognition M
-'3
reconceptualized as "a manner of dividing and ranking h uman beings J by reference to selected embodied properties (real or imputed) so as to subordinate, exclude and exploit them . . ." (Wacquant 1997: 229). The introduction to a recent anthology on Race and Racism (Boxill 2001:1) begins as follows: "Racial classification today is commonplace; people routinely catalogue each other as members of this or that race, and seem to assume that everyone can be thus classified." The American Anthropological Association has issued an official "Statement on 'Race'" that refers to race as "a mode of classification," a "worldview," and an "ideology" that employs socially exclusive categories to naturalize status differences.5 In sociology, too, the ascendancy of social constructivist perspectives has led analysts to emphasize "the absence of any essential racial characteristics" and "the historical flexibility of racial meanings and categories" (Omi and Winant 1994: 4).6
A general retreat from objectivism has been apparent in the study of nationhood as well: a shift from definitions of nationhood in terms of common language, culture, territory, history, economic life, political arrangements, and so on to definitions that emphasize the subjective sense of or claim to nationhood, as in Hugh Seton-Watson's interestingly circular suggestion that "a nation exists when a significant number of people in a community consider themselves to form a nation, or behave as if they formed one" (1977: 5). Like ethnicity and race, nation too has been expressly conceptualized as "a basic operator in a widespread system of social classification" (Verdery 1993: 37) and as a "practical category" (Brubaker 1996: Chapter 1).
Empirical work influenced by this new understanding of the cen-trality—indeed the constitutive significance—of categorization and classification for ethnicity, race, and nation clusters in two broad areas.7 One cluster comprises historical, political, and institutional studies of official, codified, formalized categorization practices employed by powerful and authoritative institutions—above all, the state. Foucault's notion of governmentality has been an important point of reference here (Burchell et al. 1991), as has Bourdieu's (1994) account of symbolic power as the power to state what is what and who is who, and thereby to impose legitimate principles of vision and division of the social world. The second, smaller, cluster comprises ethnographic and microinteractionist studies of the unofficial, informal, "everyday" classification and categorization practices of ordinary people.
Ethnicity as Cognition * 67
Research on official practices of ethnic, racial, and national categorization began with studies of colonial and postcolonial societies. Without dwelling on categorization per se, several now classic works pointed out how colonial rule transformed antecedent patterns of social identification and shaped patterns of ethnic mobilization through the identification, labeling, and differential treatment of ethnic groups (e.g., Young 1976; Geertz 1963; Horowitz 1985). More recent studies, giving more sustained attention to systems of classification and practices of categorization themselves, have shown how rulers' practices of naming, counting, and classifying affected the self-understandings, social organization, and political claims of indigenous populations (Anderson 1991, Chapter 10; Appadurai 1996; Dirks 1992; Hirschman 1986; Jackson 1999; Jackson and Maddox 1993).
A growing literature addresses official categorization practices in noncolonial settings as well. Much of this literature has focused on censuses. Drawing inspiration from Bourdieu's work on the symbolic power of modern states, recent works have examined how censuses inculcate the idea that national societies are bounded wholes, composed of discrete, mutually exclusive ethnic, racial, or cultural groups (Patriarca 1996; Kertzer and Arel 2002: 5-6; Nobles 2000; Loveman 2001, forthcoming). Even when census categories are initially remote from prevailing self-understandings, they may be taken up by cultural and political entrepreneurs and eventually reshape lines of identification (Starr 1987; Nagel 1995; Petersen 1987, 1997). Especially when they are linked through public policy to tangible benefits, official census categories can have the effect of "making up people" (Hacking 1986), or "nominating into existence" (Goldberg 1997: 29-30) new kinds of persons for individuals to be. Such categories, Goldberg argues from a Foucauldian perspective, are central to the state's exercise of "racial governmentality": censuses have comprised a "formative governmental technology in the service of the state to fashion racialized knowledge— to articulate categories, to gather data, and to put them to work" (Goldberg 1997: 30).
Censuses classify people anonymously and fleetingly; they do not permanently assign individuals to categories, or attach enduring, legally consequential identities to specific persons. Other forms of state categorization, however, do just this, imposing ethnic or racial categories on persons, inscribing them in documents, and attaching consequences—sometimes fateful ones—to these official identities
68 • Ethnicity as Cognition
Ethnicity as Cognition • 69
(Jenkins 1997: 69). The most notorious cases are the official schemes of racial classification and identification employed by Nazi Germany (Burleigh and Wippermann 1991) and South Africa (Bowker &c Star ; 1999, Chapter 6). More recently, attention has been called to the uses made of official ethnic identities, specified in forma! identity documents, in the Rwandan genocide (Fussell 2001; Longman 2001 ).8 In the Soviet Union, too, ethnic nationality was not only a statistical category, a fundamental unit of social counting and accounting, but a legal category that was inscribed in personal documents, transmitted by descent, recorded in bureaucratic encounters and official transac- : tions, and used in certain contexts to govern admission to higher education and access to certain types of jobs (Vujacic and Zaslavsky 1991; Roeder 1991; Slezkine 1994; Brubaker 1994; Martin 2001).
Studies of official categorization practices generally argue or imply that the ways in which states and other organizations count, classify, and identify their subjects, citizens, and clients have profound consequences for the self-understandings of the classified. This is no doubt often the case, but the connection between official categories and popular self-understandings is seldom demonstrated in detail. And the literature on classification and categorization in everyday life shows that the categories used by ordinary people in everyday interaction often differ substantially from official categories. The categorized are themselves chronic categorizers; the categories they deploy to make sense of themselves and others need not match those employed by states, no matter how powerful.
Research on the production and reproduction of racial, ethnic, and national distinctions and boundaries in everyday life demonstrates great complexity and variability in the categories actually used. An extreme example is the very large number of race and color categories used in Brazil (Harris 1970; Sanjek 1971), but complex and variable categorization practices have been documented in many other settings (see e.g., Sanjek 1981; Leach 1954; Kunstadter 1979; Moerman 1965). A common thread in studies of everyday classification is the recognition that ordinary actors usually have considerable room for maneuver in the ways in which they use even highly institutionalized and powerfully sanctioned categories (Baumann 1996; Sokefeld 1999; Alexander 1977; Levine 1987; Berreman 1972; Dominguez 1986; Kay 1978; Sanjek 1981; Starr 1978). They are often able to deploy such categories strategically, bending them to their own purposes; or they
may adhere nominally to official classificatory schemes while infusing official categories with alternative, unofficial meanings.9
Although most work on everyday categorization is ethnographic, a few works derive inspiration from ethnomethodology and conversation analysis, and notably from the pioneering work of Harvey Sacks. These works treat ethnicity as a skilled practical accomplishment, as something that "happens" when ethnic categories are made relevant to participants in the course of a particular interactional trajectory (Moerman 1974; Day 1998; Schegioff 2002; Brubaker et al. 2004). Such research sees ethnic and other category memberships as "ascribed (and rejected), avowed (and disavowed), displayed (and ignored) in local places and at certain times ... as part of the interactional work that constitutes people's lives" (Antaki and Widdicombe 1998: 2).10
In its concern with the social organization and interactional deployment of knowledge, the literature on official and everyday categorization represents an incipient cognitive turn in the study of ethnicity. The scope of this cognitive turn, however, has been limited by the lack of engagement with expressly cognitive research in psychology and cognitive anthropology. Indeed most discussions of categorization and classification proceed without any explicit reference to cognition.11
Two reasons for the reluctance to engage expressly cognitive research can be identified. First, to extend a point DiMaggio (1997: 264-66) made about the sociology of culture, the humanistic, interpretive, holistic, and antireductionist commitments that inform most sociological, anthropological, and historical work on ethnicity clash with the posi-tivist, experimentalist, individualist, and reductionist commitments of cognitive science. Yet as DiMaggio goes on to argue, there has been a certain rapprochement in recent years. On the one hand, holistic understandings of culture—and, one might add, of ethnicity—have come to seem increasingly problematic; on the other, cognitive research has paid increasing attention to more complex, culturally and historically specific mental structures and processes—to the "sociomentai" domain, as Zerubavel calls it (1997: 5).
Second, advocates of ethnographic and especially interactionaliy oriented research have drawn a sharp distinction between cognitive and discursive approaches. The cognitive approach takes "discourse as a realization of. . . underlying processes and structures of knowledge," and "culture itself ... as a kind of socially shared cognitive
70 • Ethnicity as Cognition
organization" (Edwards 1991: 517). The discursive approach, in con-, trast,
treats talk and texts ... as forms of social action. Categorization is something we do, in talk, in order to accomplish social actions (persuasion, blaming, denials, refutations, accusations, etc.). From this perspective, we would expect language's "resources" not to come ready-made from a process in which people are trying their best to understand the world [as in the cognitive approach] ... but rather, or at least additionally, to be shaped for their functions in talk, for the business of doing situated social actions (ibid.).12
This is a valid—and important—criticism of some strands of cognitive research. Yet it overstates the opposition by relying on a narrow understanding of cognitive research as premised on an "individualistic, mentalistic, computational, and culture-minimal" notion of mind and as seeking to reduce "all of psychological life, including discourse and social interaction, to the workings of cognitive, or even computational, mental processes" (Edwards 1997: 32, 19). As DiMaggio (1997) has pointed out, and as Edwards and Potter (1992: 14-15, 21, 23) themselves acknowledge, there is much recent cognitive research that cannot be characterized in this way.
The incipient cognitive turn in the study of ethnicity could be extended in fruitful ways by drawing on the empirical findings and analytical tools of cognitive research. Strong cognitive assumptions— though generally unacknowledged and therefore unanalyzed ones-inform almost all accounts of the ways in which race, ethnicity, and nation "work" in practice. When we characterize an act of violence as racial, ethnic, or nationalist; when we analyze the workings of racially, ethnically, or nationally charged symbols; when we characterize police practices as involving "racial profiling"; when we explain voting patterns in terms of racial or ethnic loyalties; when we impute identities or interests to racial, ethnic, or national groups; when we analyze nationalist collective action; when we characterize an action as meaningfully oriented to the race, ethnicity or nationality of another person; when we identify an expression as an ethnic slur—in these and innumerable other situations, we make cognitive assumptions about the ways in which people parse, frame, and interpret their experience. At a minimum, we assume that they are identifying persons, actions, threats, problems, opportunities, obligations, loyalties, interests, and so on in racial, ethnic,
Ethnicity as Cognition • 71
or national terms rather than in terms of some other interpretive scheme. Engaging cognitive anthropology and cognitive psychology would help specify—rather than simply presuppose—the cognitive mechanisms and processes involved in the workings of ethnicity, and would strengthen the microfoundations of macroanalytic work in the field. Towards this end, the next section reviews cognitive work on stereotyping, social categorization, and schemas.
Cognitive Perspectives: From Categories to Schemas
We have considered categorization as a political project and as an everyday social practice. But categorization is also a fundamental and ubiquitous mental process. As George Lakoff put it, "There is nothing more basic than categorization to our thought, perception, action, and speech." We employ categories whenever we "see something as a kind of thing . .. [or] reason about kinds of things"; we do so equally, it should be emphasized, whenever we—persons, organizations, or states—talk about kinds of things, or treat something as a kind of thing (or as a kind of person, a kind of action, or a kind of situation). Categories are utterly central to seeing and thinking, but they are equally central to talking and acting. "Without the ability to categorize, we could not function at all, either in the physical world or in our social and intellectual lives" (Lakoff 1987: 5-6).
Categories structure and order the world for us. We use categories to parse the flow of experience into discriminable and interpretable objects, attributes, and events. Categories permit—indeed entail— massive cognitive, social, and political simplification. Following a principle of "cognitive economy," they "provide maximum information with the least cognitive effort" (Rosch 1978: 28). They allow us to see different things—and treat different cases—as the same. They focus our attention and channel our limited energies, leaving us—individuals and organizations alike—free to disattend to "irrelevant" stimuli. They thereby make the natural and social worlds intelligible, interpretable, communicable, and transformable. Without categories, the world would be a "blooming, buzzing confusion"; experience and action as we know them would be impossible. Thus categories underlie not only seeing and thinking but the most basic forms of "doing" as well, including both everyday action and more complex, institutionalized patterns of action.
72 ■ Ethnicity as Cognition
Ethnicity as Cognition ■ 73
When we make sense of our experience by seeing objects, persons, actions, or situations as instances of categories, this always involves more than mere sorting. It always carries with it expectations and "knowledge"—sometimes rather elaborate knowledge (Medin 1989)— about how members of those categories characteristically behave. Such beliefs and expectations are embodied in persons, encoded in myths, memories, narratives, and discourses, and embedded in institutions and organizational routines. Even when we are not consciously aware of them, they can subtly (or not so subtly) influence our judgments, and even our very perceptions, of the objects or persons so categorized, and thereby the way we behave toward them. This holds true not only in laboratory settings, but also in everyday interactional contexts and in the workings of organizations and institutions.
Stereotypes
Recent work on stereotypes emphasizes the continuities between stereotypical thinking and categorical thinking in general.13 Stereotypes are no longer defined in terms of cognitive deficiencies—in terms of false or exaggerated or unwarranted belief—but more neutrally as cognitive structures that contain knowledge, beliefs, and expectations about social groups (Hamilton and Sherman 1994: 2-3). Nor are stereotypes seen as the distinctive and pathological propensity of particular kinds of personalities (the "authoritarian personality" or "high-prejudice" individual, for example), but rather as rooted in normal and ubiquitous cognitive processes. There is no need to postulate special "needs"—for example the alleged need to feel superior to others—to explain stereotypes; they are more parsimoniously explained as an outgrowth of ordinary cognitive processes.
On this understanding, which has antecedents in the work of Gordon Allport (1954), stereotypes are simply categories of social groups, and their structure and workings mirror those of categories in general. Like other categories, stereotypes are represented in the mind through some combination of prototypical features, concrete exemplars, behavioral expectations, and theory-like causal knowledge. Like other categories, stereotypes obey the principle of cognitive economy, generating inferences and expectations that go "beyond the information given" (Bruner 1973 [1957]) with minimal cognitive processing. Like other categories, stereotypes work largely automatically. They can be primed or cued subliminally, and can influence subjects'
judgments without their awareness. This does not mean that stereotypes are wholly beyond conscious control, but it does mean that stereotyping is deeply rooted in ordinary cognitive processes and that countering or correcting stereotypes is effortful and costly (Devine 1989).
The content of stereotypes—and therefore their substantive social significance and in particular their perniciousness—is of course highly variable across cultural settings, over time, and across target groups. Clearly, cognitive research cannot explain such variations in content. But it can help explain the universality of stereotyping, based as it is in categorical thinking in general; the resistance of stereotypes to discon-flrming information; the dynamics of activation of stereotypes; the ways in which stereotypes, once activated, can subtly influence subsequent perception and judgment without any awareness on the part of the perceiver; and the extent to which and manner in which deliberate and controlled processes may be able to override the automatic and largely unconscious processes through which stereotypes are activated.
Because they are not the products of individual pathology but of cognitive regularities and shared culture, stereotypes—like social categories more generally—are not individual attitudinal predilections, but deeply embedded, shared mental representations of social objects. As a consequence, macro- and mesolevel research cannot dismiss research on stereotypes as "individualistic" or "psychologically reductionist." Research on stereotypes clarifies the relationship between the individual and the social in the production and operation of standardized templates for making sense of social objects. Among these templates are those that frame social objects and social experience in racial, ethnic, or national terms and are activated by particular, culturally specific cues. Cognitive research on stereotypes can thus illuminate the sociocognitive underpinnings of the variable resonance and salience of racial, ethnic, and national ways of seeing, interpreting and reacting to social experience.
Social Categorization
Stereotyping is of course one key aspect of social categorization, but it is by no means the only one. Other aspects have been explored by the largely European tradition of research known as "social identity theory" (or in some later variants as "self-categorization theory") that grew out of the work of social psychologist Henri Tajfel. Arguing
74 ■ Ethnicity as Cognition
against the paradigm of "realistic group conflict theory," according to which intergroup conflicts are grounded in accurate perceptions of underlying conflicts of interest, Taj fel demonstrated the autonomous significance of categorization. His "minimal group" experiments revealed a robust tendency toward in-group bias—the tendency to favor members of one's own category—even in the absence of any intergroup conflict or hostility, indeed even when the "groups" or categories were constructed along purely arbitrary lines (for example, through random experimental assignment of subjects to artificial categories of "reds" and "blues"). In other words, "the mere perception of belonging to two distinct groups—that is, social categorization per se—is sufficient to trigger intergroup discrimination favoring the in-group" (Tajfel and Turner 1986: 13).14
A second aspect of social categorization (indeed of categorization in general) documented by Tajfel and associates is the tendency of categorization to produce "accentuation effects." People tend to exaggerate both the similarity of objects within a category and the differences between objects in different categories (Hogg and Abrams 1988: 19). When the categories at hand are categories of "human kinds," the over-estimation of intercategory differences and of intracategory (especially out-group) homogeneity15 facilitates the reification of groups. Ethnic classification depersonalizes individuals by transforming them "from unique persons to exemplars of named groups" (Levine 1999:169). Together with more recent research on the causes and consequences of perceptions of the "entitativity"—that is, the unity and coherence—of social categories or groups, these findings can help explain the resilience of "groupist" representations of the social world.16
Schemas
Schemas (and related concepts such as scripts and cultural models) became a central focus of research in cognitive psychology and cognitive anthropology in the 1970s as researchers developed more complex models of cognition than had characterized earlier phases of cognitive research (for overviews see Rumelhart 1980; Casson 1983; Markus and Zajonc 1985; D'Andrade 1995: Chapter 6; Strauss and Quinn 1997: Chapter 3). Recent sociological theory has also invoked the notion of schema (Bourdieu 1977; Giddens 1984; Sewell 1992; DiMaggio and Powell 1991; DiMaggio 1997), while the related concept of frame, originally given sociological formulation by Goffman (1974), has been
Ethnicity as Cognition • 75
adapted by the cognitively oriented literature on social movements (Gamson 1992; Johnston 1995).
Schemas are mental structures in which knowledge is represented. They range from the universal to the idiosyncratic (Casson 1983: 440). Most sociologically interesting schemas, however, are neither universal nor idiosyncratic but "culturally [more or less widely] shared mental constructs" (D'Andrade 1995: 132; cf. D'Andrade 1981; Zerubavel 1997). As mental structures, schemas are of course not directly observable. Rather, they are posited to account for evidence—experimental, observational, and historical—about how people perceive and interpret the world and how knowledge is acquired, stored, recalled, activated, and extended to new domains.
Schemas are not simply representations but also "processors" of information (Rumelhart 1980: 39; Casson 1983: 438; D'Andrade 1995: 122, 136). They guide perception and recall, interpret experience, generate inferences and expectations, and organize action. In this way they function as "a kind of mental recognition 'device' which creates a complex interpretation from minimal inputs; [they are] not just a 'picture' in the mind" (D'Andrade 1995: 136). In contrast to piecemeal processing, which "relies only^ on the information given and combines the available features without reference to an overall organizing structure," schematic processing treats each "new person, event, or issue as an instance of an already familiar category or schema" (Fiske 1986). As processors, schemas function automatically, outside of conscious awareness. They process knowledge in an "implicit, un-verbalized, rapid, and automatic" manner, unlike modes of controlled cognition, which process knowledge in an "explicit, verbalized, slow, and deliberate" manner (D'Andrade 1995: 180). In this respect they are congruent with, and indeed the means of specifying further, sociological constructs such as Bourdieu's notion of sens pratique, the "regulated improvisation" of practical action governed by the habitus (Bourdieu 1977, 1990a; Wimmer 1995: 62 ff.; Strauss and Quinn 1997: 44-47).17
Schemas are organized hierarchically. The top levels, representing core, invariant aspects of concepts, are fixed, but lower levels have "slots" that need to be filled in by contextual cues, by information revealed in the course of interaction, or by "default values" (Casson 1983: 431-32; D'Andrade 1995: 123, 136, 139ff.). In this respect the concept resonates with the core ethnomethodological idea that all
76 ■ Ethnicity as Cognition
Ethnicity as Cognition • 77
mundane interaction requires participants to "fill in" unspecified information from their stocks of tacit background knowledge.
Schemas must be activated by some stimulus or cue. Activation depends on proximate, situationally specific cues and triggers, not directly on large-scale structural or cultural contexts, though structural and cultural changes can affect the distribution of such proximate cues and thereby the probabilities of activation of schemas. An important limitation of existing research is that activation of schemas, as of stereotypes, has been studied chiefly in experimental settings that cannot capture the enormous complexity of the actual interactional contexts in which schemas are activated. As DiMaggio points out, a central challenge for cognitively minded sociologists is to understand the interaction between the distribution of schemas across persons and the distribution of the "external cultural primers" that evoke them (DiMaggio 1997: 274).18 To the extent that progress is made in this respect, the schema concept has the potential to bridge private and public, mental and social, the individual mind and the supraindi-vidual world of public representations.
Surprisingly, given its application in many other social and cultural domains (for reviews see Casson 1983; D'Andrade 1995; Quinn and Holland 1987), the schema concept has not been used systematically in the study of ethnicity.19 There has of course been a great deal of work on ethnic and racial (and to a lesser extent national) categories. And there is certainly some overlap between the notion of categories and that of schemas. Both concern the organization and representation of knowledge and the ways in which knowledge structures permit us to go beyond immediately given information, make inferences, and interpret the world. Yet the schema concept allows consideration of more complex knowledge structures. The recent literature on categories, to be sure, stresses the complexity of category-based knowledge. It suggests, for example, that categories are "theory-like," in that causal knowledge—not simply prototypical attributes or characteristic exemplars—is built into categories themselves (Medin 1989). Nonetheless, the issue of categories and categorization has been interpreted relatively narrowly in studies of ethnicity.
When we think of categorization in connection with ethnicity, we tend to think of categories of people. We don't think of categories of situations, events, actions, stories, theories, and so on. Yet as Lakoff observes, most categories "are not categories of things; they are
categories of abstract entities. We categorize events, actions, emotions, spatial relationships, social relationships, and abstract entities of an enormous range" (1987: 6).
A cognitive perspective focuses our analytical lens on how people see the world, parse their experience, and interpret events. This raises a different and broader set of questions about racial, ethnic, and national categorization. The relevant questions are not only about how people get classified, but about how gestures, utterances, situations, events, states of affairs, actions, and sequences of actions get classified (and thereby interpreted and experienced). The questions, in short, are about seeing the social world and interpreting social experience, not simply about classifying social actors, in ethnic terms. The schema concept can help elucidate and concretize this notion of ethnic "ways of seeing."
Consider for example schemas for events and for standardized sequences of events. In the cognitive literature, these are sometimes called scripts (Schank and Abelson 1977). A standard example is the "restaurant" schema or script for the stereotypical sequence of events involved in ordering, being served, eating, and paying for food at a restaurant. Much knowledge (in the broadest sense) that is relevant to—indeed partly constitutive of—race, ethnicity, and nationhood is embedded in such event schemas. For example, a significant part of the knowledge that many African Americans have about race may be contained in schemas for recurrent events or stereotypical sequences of events. These might include the "being stopped by the police for DWB ['driving while black']" schema or the " being-watched-in-the-store-as-if-one-were-considered a-potential-shoplifter" schema. Like all schemas, event schemas such as these can be activated and generate interpretations with minimal or ambiguous inputs. There is no doubt—there is indeed abundant evidence—that conscious and unconscious "racial profiling" exists; but it may also be that event schemas such as these can generate the interpretation and experience of racial profiling even in marginal or ambiguous situations, thereby further "racializing" social experience.
Or consider social interpretation schemas—a loose and heterogeneous class of schemas that includes all kinds of templates for making sense of the social world. Ethnicity can be slotted into many of these so as to generate ethnic variants or subtypes of the schemas. Consider for example a generic social competition schema, an abstract
78 • Ethnicity as Cognition
Ethnicity as Cognition ■ 79
representation of two or more parties competing over some scarce good or resource. In the generic schema, there is no restriction on the object of competition (which might be money, prestige, love, market share, power, etc.) or on the parties (which might be persons, families, cliques, factions, teams, coalitions, firms, occupational groups, organizations, states, and so on). In addition to this generic social competition schema, however, there may be a variety of more specific social competition schemas, defined by specific sorts of objects or by specific sorts of parties. One of these might be an ethnic competition schema, perhaps informed by a strong normative "sense of group position" (Blumer 1958: 3-7; Bobo 1999), in which the parties would be ethnic (or racial or national) groups. If this ethnic competition schema is easily activated, people may be more prone to see and experience competition in ethnic rather than other terms. This is part of what is meant by ethnicization. Given the pervasive ambiguity of the social world, there is always a great deal of room for interpretation, and schemas are the mechanisms through which interpretation is constructed. One key aspect of processes of ethnicization is that ethnic schemas can become hyperaccessible and in effect crowd out other interpretive schemas.
Broader Implications
Apart from their direct applications to the study of ethnicity, the cognitive perspectives we have reviewed challenge us to revisit foundational issues and recast certain fundamental debates in the field. In this final section, we consider the implications of cognitive perspectives for (1) the conceptualization of the domain of study; (2) the question whether race, ethnicity, and nation require separate or integrated analytical treatment; and (3) the perennial debate between "primor-dialist" and "circumstantialist" approaches.
Conceptualizing the Domain: From Things in the World to Ways of Seeing
Despite the constructivist stance that has come to prevail among sophisticated analysts, the study of ethnicity remains informed by "groupism": by the tendency to treat ethnic groups, nations, and even races as things-in-the-world, as real, substantial entities with their own cultures, identities, and interests. In accordance with what
Hollinger (1995) has called pluralist rather than cosmopolitan understandings of diversity and multiculturalism, the social and cultural world is represented as an assemblage of self-enclosed, homogeneous racial, ethnic, or cultural groups.20
Cognitively oriented work offers resources for avoiding such groupism, while at the same time helping account for its tenacious hold on our social imagination. Cognitive perspectives suggest treating racial, ethnic, and national groups not as substantial entities but as collective cultural representations, as widely shared ways of seeing, thinking, parsing social experience, and interpreting the social world. Instead of conceptualizing the social world in substantia list terms as a composite of racial, ethnic, and national groups—instead, that is, of uncritically adopting the folk sociological ontology that is central to racial, ethnic, and national movements—cognitive perspectives address the social and mental processes that sustain the vision and division of the social world in racial, ethnic, or national terms. Rather than take "groups" as basic units of analysis, cognitive perspectives shift analytical attention to "group-making" and "grouping" activities such as classification, categorization, and identification. By its very nature, categorization creates "groups" and assigns members to them; but the groups thus created do not exist independently of the myriad acts of categorization, public and private, through which they are sustained from day to day. Race, ethnicity, and nationality exist only in and through our perceptions, interpretations, representations, classifications, categorizations, and identifications. They are not things in the world, but perspectives on the world—not ontoiogical but epistemolog-ical realities.21
To say this is not to espouse a radical subjectivism or psycholo-gism.22 It is not to privilege what goes on in people's heads over what goes on in public. The promise of cognitive approaches is precisely that they may help connect our analyses of what goes on in people's heads with our analyses of what goes on in public. Dan Sperber (1985), for example, has proposed an "epidemiological" perspective on the distribution and diffusion of representations within a population. Representations, according to Sperber, are of two kinds: public representations23 (embodied in texts, talk, monuments, etc.) and mental representations. Representations of either kind may be idiosyncratic, or they may be more or less widely shared. Some representations are "easier to think" than others. Lawrence Hirschfeld (1996)
80 - Ethnicity as Cognition
and Francisco Gil White (2001) have argued that representations of the social world in terms of putative intrinsic kinds (including ethnic "kinds") may be easy to think because of our cognitive architecture. Representations that are easy to think will be more easily communicated, transmitted, and remembered, and as a result more widely shared, than others. When more or less similar versions of a representation are widely (but not universally) shared, we may speak of a cultural (rather than an idiosyncratic personal) representation. If Hirschfeld, Gil-White, and others are right about racial, ethnic, and national categories being easy to think—easier to think than, say, class—this would help explain in part why they tend to be widely shared and powerfully entrenched cultural representations.24
If racial, ethnic, and national categories are easy to think, this does not, of course, mean that they, or the various schemas in which they may be embedded, are universally active or salient. Indeed a concern with the diffusion, distribution, accessibility, and salience of schemas can help us avoid taking for granted the centrality and salience of race, ethnicity, and nation. Instead of speaking routinely of racial, ethnic, or national "groups," for example, which carries with it the usual implications of boundedness and homogeneity, and biases the discussion by presuming the relevance of a racial, ethnic, or national frame or self-understanding, a cognitive perspective suggests speaking of groupness as a variable.25 Here cognitive perspectives complement other attempts to think relationally rather than substantially and to problematize groupness rather than taking it for granted (Tilly 1978: 62 ff.). In its cognitive dimensions, groupness can be understood as depending not simply on the content of representations (i.e., on the extent to which the representations highlight the "entitativity,"26 the internal homogeneity and external boundedness of the "group") but on the distribution of such representations within a population,27 on their accessibility or ease of activation, on their relative salience once activated, and—not least—on the relative ease with which they "slot" into or "interlock" with other key cultural representations. This last might be understood as the cognitive counterpart to the notion of "resonance," central to the social movement literature on framing and frame alignment. Changes in groupness—short-term fluctuations as well as long-term developments—are cognitively mediated, depending on changes in the distribution or propagation of groupist representations, or on changes in their accessibility, activation, salience,
Ethnicity as Cognition • 81
or resonance. Clearly, social structural, cultural, and situational factors will be key determinants of such changes; but we will understand them better when we understand the cognitive micromechanisms through which such macrolevel determinants are mediated (cf. DiMaggio 1997: 280).
What cognitive perspectives suggest, in short, is that race, ethnicity, and nation are not entities in the world but ways of seeing the world. They are ways of understanding and identifying oneself, making sense of one's problems and predicaments, identifying one's interests, and orienting one's action. They are ways of recognizing, identifying, and classifying other people, of construing sameness and difference, and of "coding" and making sense of their actions. They are templates for representing and organizing social knowledge, frames for articulating social comparisons and explanations, and filters that shape what is noticed or unnoticed, relevant or irrelevant, remembered or forgotten.
One Domain or Several?
Race, ethnicity, and nationalism were long considered separate analytical domains, with largely nonoverlapping literatures. In the last two decades, as the literature has become more comparative and less parochial, the boundaries have blurred.28 The wider spectrum of cases has undermined neat distinctions that might have worked in some limited settings—for example, in the United States, between "race" (cori ceptualized in strictly black-white terms mirroring the one-drop rule), ethnicity (seen as generated by immigration), and nationalism (understood as something that happens elsewhere, and as definitionally linked to state formation).
Still, much ink continues to be spilled in an effort to draw analytical distinctions between race, ethnicity, and nation. In our view, this conceptual casuistry—sometimes informed by political concerns—is misplaced. It is not that we wish to treat race, ethnicity, and nation as one undifferentiated domain. Clearly, the domain is highly differentiated. But it does not parse into three clearly bounded subdomains. Rather, there are many dimensions of differentiation, none of them coinciding precisely with conventional definitions of domain. An abbreviated list of these would include:
• criteria and indicia of membership
• transmission: manner in which membership is acquired
82 • Ethnicity as Cognition
Ethnicity as Cognition • 83
• fixedness versus fluidity of membership
• degree and form of naturalization, that is, degree and form of appeal to natural grounding for community
• degree and form of embodiment; importance attributed to phenotypic and other visible markers
. importance attributed to distinctive language, religion, customs, and other elements of culture
• degree and nature of territorialization; importance of territorial organization and symbolism
• nature of claims, if any, to autonomy and self-sufficiency
These multiple dimensions of differentiation do not map neatly onto any conventional distinction between race, ethnicity, and nation.
Cognitive perspectives suggest further reasons for treating race, ethnicity, and nation together, as one integrated domain rather than several distinct domains of study. As we suggested above, race, ethnicity, and nation are fundamentally ways of seeing. The cognitive processes and mechanisms underlying these ways of seeing are identical throughout the larger domain. If nation, for example, is famously treated as an "imagined community" (Anderson 1991 [1983]) or a "conceived order" (Lepsius 1985), this is no less true of ethnicity or race. If race, according to Hirschfeld, involves folk sociologies that divide people into intrinsic, putatively natural human kinds, this is no less true for ethnicity and nation. If ethnic boundaries, as Barth says, are sustained by processes of categorical self- and other-ascription, this is no less true for racial and national boundaries. The processes of classification and categorization, formal and informal, that divide "us" from "them"; the forms of social closure that depend on categorizing and excluding certain potential competitors as "outsiders"; the categories and frames in terms of which social comparison and social explanation are organized; the schemas, scripts, and cultural models that allow one to perceive, experience, or interpret situations and sequences of action in standardized racial, ethnic, or national terms; the cognitive biases in the retrieval and processing of information that lead us to evaluate evidence in selective ways that tend to confirm prior expectations and strengthen stereotypes—all of these and many more cognitive and sociocognitive mechanisms and processes are involved in essentially similar forms in phenomena conventionally coded as belonging to distinct domains of race, ethnicity,
and nationalism. Of course there are great variations in the content of patterns of classification and closure, social comparison and explanation, schemas and cultural models, but these cut across conventional distinctions of domain.
Primordialism and Circumstantialism
Cognitive research also invites us to revisit and reframe the classic, though too often hackneyed, debate between primordialist and cir-cumstantialist or instrumentalist approaches.29 This debate pits an understanding of ethnicity as rooted in deep-seated or "primordial" attachments and sentiments10 against an understanding of it as an instrumental adaptation to shifting economic and political circumstances. Cognitive perspectives allow us to recast both positions and to see them as complementary rather than mutually exclusive.
With the ascent of the social constructionist paradigm, serious engagement of primordialist positions has given way to dismissive references to "naturalizing" and "essentializing" perspectives. But primordialism is more subtle and interesting than this. In the oft-cited but seldom closely analyzed formulation of Clifford Geertz (1963: 109), primordial attachments stem "from the 'givens'—or, more precisely, as culture is inevitably involved in such matters, the assumed 'givens'—of social existence," including blood ties, religion, shared language, and customs. In most discussions, this crucial distinction between perceived "givens" and actual "givens" is elided. Primordialists are depicted as "analytical naturalizers" rather than "analysts o/" naturalizes" (Gil-White 1999: 803). In fact, on the primordialist account, it is participants, not the analysts, who are the real primordialists, treating ethnicity as naturally given and immutable.
Thus clarified, the primordialist position cannot be so easily dismissed. And cognitive research can give it a stronger empirical foundation, by specifying the natural foundations of the often observed tendency to naturalize ethnicity. Research on "psychological essen-tialism" (Medin 1989: 1476-77) suggests that "people act as if things . . . have essences or underlying natures that make them the things they are," and that even if this is "bad metaphysics," it may in many circumstances serve as "good epistemology." Even young children, traditionally understood to attend primarily to external, visible features of things, in fact have a firm grasp of notions of "insides" and essences (Gelman and Wellman 1991). Social categories, in particular,
84 • Ethnicity as Cognition
are often (incorrectly) perceived as if they were natural kinds; as a result, people often infer "deep essential qualities on the basis of surface appearance" and "imbue even arbitrary categorizations with deep meaning" (Rothbart and Taylor 1992: 12).
Hirschfeld (1996) and Gil-White (2001) extend this line of analysis to race and ethnicity, positing a deep-seated cognitive disposition to perceive human beings as members of "natural kinds" with inherited and immutable "essences." Drawing on experiments with three- and four-year-olds, Hirschfeld (1996) argues that humans have a special-purpose cognitive device31 for partitioning the social world into what he calls "intrinsic kinds" based on "shared essences."32 This provides the cognitive foundations for what Hirschfeld calls "folk sociology," by which he means the "commonsense partitive logic or social ontology that picks out the 'natural' kinds of people that exist in the world" (1996: 20). Hirschfeld emphasizes the presence worldwide of a similar deep classificatory logic—one that naturalizes social difference by dividing the social world into putatively deeply constituted groups seen as based on some shared intrinsic essence—underlying what seem at first glance to be strikingly different systems of racial, ethnic, and national classification. Gil-White (2001) argues that essentialist reasoning about ethnicity is derived by analogical transfer from reasoning about biological species. He speculates that this occurs through the adaptation of an existing special-purpose cognitive module—a "living-kinds" module evolutionarily tailored to perception of and reasoning about species—to perception of and reasoning about ethnic groups.
Although Hirschfeld and Gil-White disagree about the particular nature of the cognitive mechanism at work, both suggest that the extremely widespread tendency to "naturalize" and "essentialize" racial, ethnic, and national categories may be grounded in the human cognitive apparatus. Cognitive perspectives enable us to analyze "participants' primordialism" (Smith 1998: 158) without endorsing analytical primordialism. And rather than attribute the naturalization of social differences to vaguely conceived emotional commitments (Connor 1994), to an irreducible sense of "identity,"33 or to "a certain ineffable significance . . . attributed to the tie of blood" (Shils 1957: 142), cognitive perspectives provide potentially powerful explanations for this tendency.
Cognitive perspectives can help respecify and strengthen the
Ethnicity as Cognition • 85
circumstantialist position as well.34 Circumstantialists have characterized ethnicity as situationally malleable and context-dependent. But how does this work? Accounts have been implicitly cognitive. Oka-niura (1981: 454), for example, has suggested that ethnic identities are activated depending on "the actor's subjective perception of the situation in which he finds himself " and "the salience he attributes to ethnicity as a relevant factor in that situation." But what governs the perception of the situation and the perceived salience of ethnicity? Most accounts are rather narrowly instrumentalist at this point, suggesting that individuals strategically manipulate, deploy, mobilize, or downplay ethnicity to suit their interests. Such deliberate and calculated manipulation of ethnicity certainly occurs, but circumstantialist perspectives would be strengthened by a less restrictive account of the micromechanisms that enable and prompt situational shifts in identification.
As we observed above, cognitive research indicates that much cognition (and schema-governed cognition in particular) is unselfcon-scious and quasi-automatic rather than deliberate and controlled. This suggests that the explicit, deliberate, and calculated deployment of an ethnic frame of reference in pursuit of instrumental advantage may be less important, in explaining the situational variability of ethnicity, than the ways in which ethnic—and nonethnic—ways of seeing, interpreting, and experiencing social relations are unselfconsciously "triggered" or activated by proximate situational cues.35 Attention to framing processes, too, can help explain the variable salience of ethnicity and variable resonance of ethnicized discourse.36 By illuminating the cognitive processes that underlie ethnic ways of seeing and talking, cognitive perspectives can provide a firmer microfoundation for accounts of "situational ethnicity."
Once each position is respecified in cognitive terms, it becomes apparent that primordialist and circumstantialist accounts need not be mutually exclusive. The former can help explain the seemingly universal tendency to naturalize and essentialize real or imputed human differences, while the latter can help explain how ethnicity becomes relevant or salient in particular contexts. Rather than contradicting one another, they can be seen as directed largely to different questions: on the one hand, how groups are conceived, and folk sociologies constructed and sustained; on the other hand, how ethnicity works in interactional practice.
86 • Ethnicity as Cognition Conclusion
Cognitive perspectives, we have argued, suggest new ways of conceptualizing ethnicity as a domain of study. By treating ethnicity as a way of understanding, interpreting, and framing experience, these perspectives provide an alternative to substantialist or groupist ontologies. They afford strong reasons for treating ethnicity, race, and nationalism as one domain rather than several. And they suggest a fresh and fruitful way of recasting the perennial debate between primordialist and circumstantialist accounts of ethnicity. In addition, the empirical findings and conceptual tools of cognitive research can help illuminate the mechanisms that link the microdynamics of race, ethnicity, and nationalism to macrolevel structures and processes.
The skeptic may counter that attending seriously to cognitive research risks abandoning the social constructionist agenda for a psychologistic and individualistic approach. We thus conclude with a reminder that there is nothing intrinsically individualistic about the study of cognition. The domain of the "mental" is not identical with the domain of the individual. Indeed the kind of knowledge in which we are interested—the schemes of perception and interpretation through which the social world is experienced in racial, ethnic, or national terms—is social in a double sense: it is socially shared knowledge of social objects. A cognitive approach to the study of ethnicity directs our attention not to individual psychology but to "sociomental" (Zerubavel 1997) phenomena that link culture and cognition, macro- and mi-crolevel concerns (DiMaggio 1997, Straus and Quinn 1997). Cognitive construction, in short, is social construction. It is only in and through cognitive processes and mechanisms that the social construction of race, ethnicity, and nation can plausibly be understood to occur.
Cognitive perspectives can also advance the constructivist agenda by correcting for the elite bias of much constructivist research. By this we mean the tendency to focus on conspicuously visible constructions, such as those of political entrepreneurs, high-level state bureaucrats, or public intellectuals, to the neglect of the less visible (but no less "constructive") activities of common people in their everyday lives. In his "insider's critique" of the framing perspective in social movement literature, Benford (1997) points to the need for studies of "rank-and-file" framing. Similarly, social constructivism needs studies of the "rank-and-file" construction of racial, ethnic and national "realities."
Ethnicity as Cognition • 87
Cognitive research provides the conceptual vocabulary and analytical tools for such an enterprise.
Finally, cognitive perspectives can help realize the constructivist aspiration to capture the relational and dynamic nature of race, ethnicity, and nation by treating them as products of reiterative and cumulative processes of categorizing, coding, framing, and interpreting. Instead of asking "what is race?", "what is an ethnic group?", "what is a nation?", a cognitive approach encourages us to ask how, when, and why people interpret social experience in racial, ethnic, or national terms.
The phenomena we call race, ethnicity, and nation surely count among the most significant social and cultural structures—and among the most significant social and political movements—of modern times. Yet they continue to exist only by virtue of being reproduced daily in and through the quotidian ways of thinking, talking, and acting of countless anonymous individuals. Although this is widely recognized in principle, the mechanisms of this daily reproduction remain little known. The promise of a cognitive perspective is that it can help us understand the ways in which these great principles of vision and division of the social world work in the world at large by specifying the way they work in ordinary minds and seemingly insignificant everyday practices.
The Return of Assimilation? • 117
CHAPTER FIVE
The Return of Assimilation?
The Differentialist Turn
"The point about the melting pot," wrote Nathan Glazer and Daniel Patrick Moynihan in the preface to their influential Beyond the Melting Pot, "is that it did not happen." This "failure to melt" thesis was iconoclastic when the book was published in 1963. But it had become widely accepted already by the end of the decade—well before the post-1965 revival of mass immigration began to transform the American urban landscape. By the 1980s, when the effects of the "new 'new immigration' " had become unmistakable, earlier conceptions of assimilation seemed to many to have lost all relevance. When Glazer published We Are All Multiculturatists Now in 1997, he was writing as eminence grise, not as iconoclastic intellectual.1 Pluralistic understandings of persisting diversity, once a challenge to the conventional wisdom, had become the conventional wisdom, not only in the United States and other classic countries of immigration such as Canada and Australia, but also in much of northern and western Europe.
There is obviously a good deal of truth to this conventional wisdom. Public discourse and public policies bearing on the integration of immigrants are indeed vastly more "differentialist"—vastly more sensitive to and supportive of "difference"—today than they were, say, in the period between the two world wars in France or the United States, or in the early postwar decades in the United States. The 1980s and 1990s witnessed an unprecedented efflorescence of differentialist
discourse—and differentialist integration policies—in all Western countries of immigration.
This differentialist turn has not been restricted to, or even centered on, immigration. Especially in the United States, but in a more limited sense in Western Europe as well, it has been a much broader and more general movement of thought and opinion. It has found expression in movements to preserve or strengthen regional languages and cultures in Europe (Keating 1996); in demands for, and greater recognition of, the autonomy of indigenous peoples in the United States, Canada, Australia, Russia, Latin America, and elsewhere (Brested et al. 1985; Kymlicka 1995); in Black Power, Afrocentrist, and other antiassimila-tionist movements involving African Americans (Howe 1998); in the shift from an individualist, opportunity-oriented, and color-blind to a collectivism results-oriented, and color-conscious interpretation of civil rights legislation in the United States (Glazer 1978); in multicul-turalist revisions of school and university curricula (Nash et al. 1997; Glazer 1997); in gynocentric or "difference" feminism (Irigaray 1993); in gay pride and other movements based on the public affirmation of alternative sexualities (Johnston 1973); in claims by other putative cultural communities—including for example the deaf (Lane 1992)—for autonomy; in generalized opposition to the homogenizing, centralizing claims of the modern nation-state; in antifounda-tionalist understandings of the production of knowledge in historically and socially situated epistemic communities (Hollinger 1997); in other poststructuralist and postmodernist critiques of the universalist premises of Enlightenment thought; and in the shift from an understanding of politics emphasizing the pursuit of putatively universal interests to one emphasizing the recognition of avowedly particularist identities (Young 1990).
Today, however, this massive differentialist turn in social thought, public discourse, and public policy shows signs of having exhausted itself. Differentialist stances have long been a lightning rod for criticism from cultural conservatives (D'Souza 1991) and from the economistic, resolutely anti-identitarian left. More recently, criticism has come increasingly "from within," that is from the "cultural left" itself, from persons sympathetic to the claims of cultural difference, yet uncomfortable with their absolutization and with the pervasive "culturaliza-tion" of political rhetoric.2 Opposition to the relativistic and ultimately solipsistic implications of epistemological insiderism; concern over the
118 ■ The Return of Assimilation?
fragmenting and in certain respects disabling consequences of identity politics; resurgent interest in forms of civic commonality; rethinking of the modalities of and rationale for affirmative action, not only on the part of its longstanding critics on the right, but on the part of its longstanding defenders on the left—these and other developments suggest that, in some respects at least, the maximally differentialist moment may have passed.3
In the domain of immigration, too, there are signs that the differentialist tide may have begun to ebb. Instead of a definitive, unidirectional shift from assimilation to multiculturalism, there is evidence of an incipient shift in the opposite direction. To call this the "return of assimilation" is undoubtedly too grand a label for the relatively modest and uneven shift I will describe; hence the question mark in my title. But it may usefully caution us against overhastily consigning assimilation to the dustbin of history.
Two Meanings of "Assimilation"
By the "return of assimilation," I do not mean a return to the normative expectations, analytical models, public policies, or informal practices associated with the ideal of Anglo-conformity or the increasingly nativist Americanization movement after the First World War (Gleason 1980); or to those associated with the schoolteachers of the French Third Republic, notorious for shaming and humiliating those who spoke languages or dialects other than standard French (Weber 1976: 313); or to those associated with the harsh Imperial German effort to "Germanize" its largely Polish-speaking eastern borderlands (Broszat 1972: 129-72);4 or to any of the many other lamentable instances of harshly homogenizing state projects.
This should go without saying, but assimilation has acquired such a bad name among American differentialists that it has come to be associated almost automatically with narrow Anglo-conformity or aggressive Americanization. In Germany, the word "assimilation" has been even more strongly "contaminated" and disqualified by its association with forcible Germanization. In France, by contrast, the word itself was never so thoroughly discredited. But in France, too, it was tainted by association with the sometimes brutally homogenizing aspirations and practices of Jacobin Republicanism.
So what are we talking about when we talk about "assimilation"?
The Return of Assimilation? -119
What is it that is "returning," if it is not these normatively and analytically discredited models? To address this question, we must distinguish between two basic meanings of "assimilation." One is general and abstract; the other is specific and organic. The two meanings are related, but they differ sharply in their affective overtones, moral and political connotations, and intellectual respectability.
In the general and abstract sense, the core meaning is increasing similarity or likeness. Not identity, but similarity. To assimilate means to become similar (when the word is used intransitively) or to make similar or treat as similar (when it is used transitively). Assimilation is thus the process of becoming similar, or of making similar or treating as similar.
In the specific and organic sense, the root meaning is transitive. To assimilate something is to "convert [it] into a substance of its own nature, as the bodily organs convert food into blood, and thence into animal tissue ... to absorb into the system, [to] incorporate" {Oxford English Dictionary). Assimilation in this sense implies complete absorption.
In the general, abstract sense, the accent is on the process, not on some final state, and assimilation is a matter of degree. Assimilation designates a direction of change, not a particular degree of similarity. In the specific, organic sense, by contrast, the accent is on the end state, and assimilation is an either-or matter, not a matter of degree.
It is the connotations of this organic meaning, with its biological metaphor of incorporation, that have discredited the term, making it seem normatively retrograde (given our contemporary appreciation of difference and diversity), analytically disreputable (given its superannuated organismic understanding of society), and empirically wrong (with its implication of complete absorption).
In addition, one aspect of the general, abstract meaning has stood out as normatively and analytically problematic. This is the transitive use of "assimilate" to mean "make similar," which suggests state policies and programs of "forced assimilation," or at least policies and programs that seek to assimilate people against their will. Such policies and programs have rightly come to be seen as morally and politically repugnant. Abundant historical and comparative evidence, moreover, suggests that they rarely work, and that they are indeed more likely to strengthen than to erode differences, by provoking a reactive mobilization against such assimilatory pressures. Analytically, we may have
120 • The Return of Assimilation?
good reason to speak of assimilationist policies; but such policies need not have assimilationist outcomes.5
Yet when used intransitively in the genera), abstract sense of becoming similar—becoming similar in certain respects, that obviously have to be specified—assimilation does not seem to be morally objectionable, analytically useless, or empirically wrong as a conceptual instrument for studying populations of immigrant origin. Indeed the use of some such notion—if only to pose certain questions about patterns of "integration," "adaptation," or "incorporation," terms that have been preferred to "assimilation" in many recent discussions6—would seem to be analytically indispensable. I return to this point in the conclusion. Here I simply wish to underscore that it is this intransitive understanding of "assimilation," this normative and analytical concern with the nature and extent of emerging similarities in particular do- j mains between populations of immigrant origin and "host" populations, that I see "returning" in recent years.
Three Cases
I sketch in the following sections three illustrative vignettes, drawn from different countries and from different domains. I discuss the return of assimilation in public discourse in France, in public policy in Germany, and in scholarly research in the United States. It might be argued that whatever "return of assimilation" I find is largely an artifact of the cases I have chosen: the United States is historically the paradigmatic country of immigrant assimilation, while France is the European country with the longest, strongest, and most ideologically elaborated tradition of assimilation. There is something to this: had I chosen different cases—for example the U.K., Sweden, the Netherlands, and Germany—the trend would have been less clear-cut.
Yet the trend does not simply reflect the cases chosen. In the first place, the return of assimilation in France and the United States involves a marked return, not simply the persistence of something always present. I want to stress this reactive moment of return, and to situate it in the context of a preceding "differentialist" turn in both France and the United States. Moreover, there has been a modest assimilationist turn in Germany as well, and in the Netherlands (Koopmans and Statham 2000; Thranhardt 2000) and Sweden (Soininen 1999: 6S9-91), two other countries with relatively "differentialist" incorporation regimes. -i
The Return of Assimilation? ■ 121
Over the longer term, as a third generation of immigrant origin emerges, it is likely that a concern with at least some dimensions of assimilation will become increasingly salient throughout Europe.7
France: From droit a la difference to droit a la resemblance
One might think there was not much of a story to be told about France. Why talk of a return of assimilation in a country that has a long tradition of assimilation, transforming peasants—and immigrants—into Frenchmen, in what Gerard Noiriel has called le creuset frangais, the French melting pot (Weber 1976; Noiriel 1988)? But to frame the issue in this way—to focus only on the Jacobin-Republican assimilationist tradition, or myth—is to forget the strong differentialist turn that occurred in French public discussion of immigration and other issues in the 1970s and early 1980s, precisely in reaction against the Jacobin and assimilationist tradition. Indeed differentialist discourse received one of its sharpest and most lapidary, if ambiguous, formulations in the characteristic French slogan of those years: the droit a la difference. True, the differentialist turn was much stronger in rhetoric than in reality: differentialism remained largely symbolic (Schnapper 1992: 119) and was embedded only relatively weakly in policies and institutionalized practices—for example in the program in which foreign instructors, selected and paid by foreign governments as a result of bilateral agreements concluded with the French state, were recruited to offer instruction in so-called "languages and cultures of origin" in French public schools (Boyzon-Fradet 1992: 155 ff .}.8 But at the level of public discussion, differentialism was clearly ascendant in the early 1980s, during the early years of the Socialist government.9
It is important to note that differentialism was gaining ground on the French right as well as the left. The historian and philosopher Pierre-Andre Taguieff has analyzed the rise of a differentialist—one could even say multiculturalist—"new right" in France in the 1970s and 1980s, clustered around the enigmatic figure of Alain de Benoist. No longer xenophobic but formally "heterophile," antiracist, and egalitarian, the new differentialists of the right emphasizxd, indeed absolutized, cultural difference, seeking to "preserve at any price collective identities, and thus differences between communities, haunted by the danger of their destruction through mixing, physical and cultural" (Taguieff 1994: 66-67).
122 • The Return of Assimilation?
What happened to the ascendant differentialism? In two words: Le Pen. Although Le Pen and the intellectuals associated with him actually belonged to a different segment of the right than the small circle of principled differentialists analyzed by Taguieff, they too adopted a dif-ferentialist idiom, adroitly turning it to their own purposes. Droit a la difference? Mais out, bien sur, chez vous. But here, in France—so went the argument—it's we, the "real" French, who have our own right to be different, our own right to preserve our own "identity" from unwanted admixture. As a result, the moral and political ambiguity—and the exclusionary potential—of culturalist differentialism were brought into sharp focus.
It was this political and ideological conjuncture that set the stage for the return of assimilation. The much-vaunted droit a la difference ceased to be invoked; by the late 1980s, one was more likely to hear of the droit a la resemblance, to which Harlem Desir himself appealed in a widely watched TV appearance in 1987—or of the droit a ('indifference, in effect the right to be treated like everyone else. In the wake of the differentialist collapse, there was a resurgence of neore-publican, neouniversalist, and at least hesitatingly neoassimilationist discourse, elaborated by such public intellectuals as Alain Finkiefkraut (1987), Taguieff (1996), and especially Emmanuel Todd (1994). Their views do not go unchallenged, of course; there are sophisticated analysts such as Michel Wieviorka (1996) who continue to defend a moderately differentialist position. Yet the sudden collapse of simplistic, sloganeering differentialism and the equally sudden resurgence of universalist, assimilationist discourse about immigration is striking. Certainly, no equally sharp shift in the center of gravity of public discourse has occurred elsewhere.
Germany: Rethinking Institutionalized Separateness
While my French vignette concerned public discourse, my German story is about public policy. German policy vis-a-vis immigrants and their descendants has been strongly differentialist—much more so than French policy even during the years of ascendant differentialist rhetoric in France.10
Consider three indicators of differentialist policy in Germany. First, instruction in languages and cultures of origin has been much more widespread in Germany than in France, and indeed has been part of
The Return of Assimilation? ■ 123
the obligatory curriculum in some Länder (Castles et al. 1984: 175). (Since education is the responsibility of the individual Lander, this has varied a good deal from state to state; Bavaria, in particular, was long notorious for educating foreigners in segregated, homeland-oriented classes.)
Second, there is the peculiar German system of social service provision to populations of immigrant origin. Responsibility for such provision was farmed out by the state to the three major nonstate charitable organizations—one affiliated with the Catholic Church, a second with the Evangelical Church, the third with the Social Democratic Party. Jurisdiction was apportioned in such a way that foreigners were allocated to a particular charitable organization on the basis of their national origin, so that all Turks were the responsibility of one organization, all Italians of a second, and so on. As critics have observed (Puskeppeleit and Thränhardt 1990; cf. Älund & Schierup 1991 on the somewhat similar Swedish case), this system not only treats immigrants as passive clients of the charitable organizations, but also tends to reinforce and perpetuate national origin distinctions.
The third policy I want to discuss is citizenship. Until its recent liberalization, German citizenship law was well known for its restrictive-ness vis-ä-vis non-German immigrants. What was and remains less well known is that except for political rights, long-settled noncitizen immigrants have possessed rights virtually identical to those of German citizens. Of course, as immigrant populations became more settled, and as a second and an incipient third generation population developed, the lack of political rights became increasingly anomalous. What was distinctive about the response to this anomaly, and indicative of deep-rooted German differentialism, was that the solution was long seen on the left not in terms of incorporating immigrants and their descendant as full citizens, but rather in terms of extending even political rights—along with social, civil, and economic rights—to resident foreigners. Until the early 1990s, there was little interest in the anomalous formal citizenship status of immigrants, but considerable interest in extending voting rights to foreigners in local elections, and a large literature addressing this possibility. This was seen as the "progressive" solution—one that would extend the substantive rights of citizenship to immigrants without questioning their "differentness," their foreignness, their otherness.11
124 • The Return of Assimilation?
These policies, and the idioms in which they were rationalized and justified, were indicative of a kind of benevolent, paternalistic, and egalitarian (or pseudoegalitarian) "apartheid" or institutionalized separatc-ness. As suggested in the oxymoronic phrase "unsere ausländische Mitbürger'" ("our foreign fellow citizens"), this has been a leitmotif of well-meaning public discussions of what continues to be called Ausländerpolitik—politics or policies regarding foreigners—in Germany. Left differentialists of course criticized existing policies on various counts; but they too endorsed this separate-but-equal logic.
It is against this background of deep differentialism that signs of a modest assimilationist turn can be discerned in the manner in which citizenship has been legally redefined and politically reconceived in recent years (Joppke 1999: 202-208). Naturalization rules were substantially eased in the early 1990s, and naturalization rates of Turks— extraordinarily low until the late 1980s—have soared. In 1999, naturalization rules were further liberalized. More importantly, the rules for the attribution of citizenship at birth were changed as well, supplementing the previously exclusively descent-based law, founded on the principle of jus sanguinis, with the territorial principle of jus soli.n Henceforth, citizenship will be attributed at birth to children born in Germany to foreign parents, one of whom has resided legally in Germany at least eight years. (This citizenship will, however, be provisional; in most cases, the child will have to choose either the German or the foreign citizenship at maturity, and renounce the other.)
The legal changes, increasing naturalization rates, and new ways ; of thinking and talking about citizenship on the part of Germans and ■ foreigners alike are indicative of a limited but significant assimilationist turn. Not in the sense that full assimilation is required as a prerequisite for citizenship. To the contrary: the liberalization of naturalization law broke expressly with this principle, previously enshrined in the regulations governing naturalization. The new practices, policies, and discourses surrounding citizenship are assimilationist, rather, in the sense of politically recognizing, legally constituting, and symbolically emphasizing commonality rather than difference. Assimilation, it is worth remembering, means becoming similar, or treating as similar, and i this new inflection in the policies and practices of citizenship in the 1990s has involved a modest but significant assimilationist turn in ! both senses.
The Return of Assimilation? • 125
The United States: Assimilation without "Assimilationism"
Having discussed public discourse in France and public policy in Germany, I turn to a third domain in which one can discern a return of assimilation in recent years: scholarly research. Here I will focus on the United States, though I should note that in France, too, researchers have shown a renewed interest in assimilation (Tribalat 1996; Todd 1994). Jn Germany, by contrast, most scholarly research on immigrant integration continues scrupulously to avoid at least the term assimilation, even when it addresses questions that could be seen as falling under this rubric (exceptions include Esser 1980 and Nauck et al. 1997).
In the United States, research on immigrant integration was dominated from its beginnings in the 1920s through the mid-1960s by as-simifationist perspectives of one kind or another. Then, from about 1965 to 1985, largely under the impact of external events, the historical and sociological literature—at least the more theoretically ambitious strands of that literature—was characterized mainly by pluralist perspectives, emphasizing and documenting ethnic persistence in a variety of ways.13 Since about 1985, however, one can discern a renewed theoretical concern with assimilation in the scholarly literature (see for example Gans 1992, Glazer 1993, Portes and Zhou 1993, Morawska 1994, Kazal 1995, Barkan 1995, Alba and Nee 1997 and 2003, Rumbaut 1997, Alba 1999).
The ethnic persistence literature has made and continues to make valuable contributions. But "a way of seeing," as Kenneth Burke observed, "is also a way of not seeing" (Burke 1954: 40). Focusing on ethnic communities, on ethnically marked places or ethnic organizations rather than on persons or wider social processes, this literature has missed those who moved out of such ethnically marked places, who "disappeared," as Ewa Morawska put it (1994: 83). With its "unexamined assumptions that cultural maintenance is always a good thing, that immigrants as a rule tried to preserve as much of their traditional culture as possible, [and] that ethnocentric Anglo-America equally reflexively resisted both cultural transplantations and assimilation," it has
tended to take ethnic communities—places—as opposed to individually experienced adaptation—immigrant lifecourses—as its object of inquiry,
126 • The Return of Assimilation?
The Return of Assimilation? • 127
and . .. has tended to focus precisely on those kinds of places—areas of concentrated first-generation settlement—where the odds of finding evi-dence for ethnic maintenance ate greatest. It has sought to restore agency to the immigrant actor, but has not always followed that agency into all the varied paths that it could take. In particular,... by confining its focus to ethnic maintenance it neglects to extend its concern for the immigrant as historical actor to the assessment of the immigrant's impact upon society as a whole. We have constructed an oppositional history of virtuous, autonomous, ethnic outsiders interacting minimally with others except in the workplace, outsiders who thereby bear, to be sure, little moral responsibility for the sins of the broader nation, but also, by implication, little significance in its broader history (Conzen 1996:21).
Inwardly focused, the ethnic persistence literature has neglected wider social and cultural processes such as the formation of transethnic (but often racially closed) working-class communities in the early part of the century (Kazal 1995); the spatial dispersion that has accompanied post-World War II suburbanization, in which even recent immigrants have been participating (Alba and Nee 1997: 836-37, 857-62); increasing rates of ethnic intermarriage (Spickard 1989; Alba 1999; Qian 1997); and the dynamic renegotiation of ethnic and racial categories and identifications (Roediger 1991; Ignatiev 1995; Perlmann and Waldinger 1997). All of these processes have led to the blurring or shifting of some ethnic boundaries (Zol-berg and Long 1999) in ways that undermine stable ethnic enclosures (Hollinger 1999).
The new theorists of assimilation do not simply replicate the old, pre-1965 approaches. The older work—even work as sophisticated as Gordon's—was analytically and normatively Anglo-conformist. It posited, endorsed, and expected assimilation towards an unproblemati-cally conceived white Protestant "core culture." Recent work on assimilation, by contrast, is agnostic about its directions, degrees, and modalities, and ambivalent about its desirability. There is nothing today comparable to the complacent empirical and normative expectancies of midcentury. Of course, this is partly because the notion of a universally acknowledged "core culture" has lost all its plausibility since the late 1960s. This, in turn, has raised the question of the reference population toward which assimilation is said to occur. Characteristic of the newer literature on assimilation is its willingness to consider multiple reference
populations and correspondingly segmented forms of assimilation (Fortes and Zhou 1993; Zhou 1997; Waters 1994; Neckerman et al. 1999).14 It is no longer true that assimilation (or integration, a term that often, especially in the European context, refers to much the same thing) is "inevitably" conceptualized as occurring "into one, single, indivisible (national) 'state,' and one, simple, unitary (national) 'society'" (Favell 2000).
Today, concern with assimilation is not necessarily "assimila-tionist." It implies no global belief in the inevitability or desirability of assimilation. This does not mean that the newer literature on assimilation has no normative thrust. Normative concerns about civic commonality do underlie and inform much work on assimilation today (Alba 1999). But they do not entail any blanket endorsement of assimilation. Some forms of assimilation are indeed widely thought to be desirable. One aspect of linguistic assimilation, for example—the intergenerational acquisition of English at levels sufficient to permit success in schooling, occupational mobility, and full participation in public life—is clearly desirable. But note that this in no way entails the desirability of what Portes and Rumbaut (1990: 209-21) call "subtractive" linguistic assimilation—the intergenerational loss of competence in the language of origin.
Some aspects of socioeconomic assimilation are also clearly desirable (Hirschman 1983: 403 ff.; Alba and Nee 1997). Consider, for ex ample, a population with mean income and education levels well below the respective means for the population at large. Surely, assimilation in these domains—in the sense of a shift in the direction of convergence with the income and educational distributions of the wider society—would be desirable for this population, and it is important to know whether and to what extent this is occurring. But the desirability of assimilation in these respects does not imply its desirability in other respects. It does not imply the desirability of complete acculturation, for example; or of full "identiflcational assimilation" (the "development of a sense of peoplehood based exclusively on the host society" [Gordon 1964: 71, emphasis added]); or of spatial assimilation through suburbanization and the concomitant decline of ethnic neighborhoods; or of full occupational assimilation and the concomitant decline of ethnic niches, enclaves, and professional specializations; or of the erosion of group boundaries through high rates of intermarriage or what Gordon called structural assimilation (participation in
128 • The Return of Assimilation?
the "social cliques, clubs, and institutions of the core society at the primary group level" [Gordon 1964: 80]). My point is not that assimilation in these respects is necessarily undesirable, though evidence— for example, about better health outcomes of infants born to immigrants than to United States-born mothers, even after controlling for ethnicity and a variety of socioeconomic factors (Rumbaut 1997)—suggests that certain forms of assimilation may indeed be undesirable. This point is forcefully developed in the segmented assimilation literature, which argues that socioeconomic success, for second generation immigrants in predominantly minority inner-city neighborhoods, may depend on resisting assimilation to the surrounding youth milieu, with its adversarial stance toward mainstream culture (Portes and Zhou 1993; Zhou 1997). The broader point is that one can study assimilation in its various domains and directions without being an "assimilationist"; one may be agnostic about its destinations and ambivalent or even skeptical about its desirability.
Assimilation is not a single process of the sort envisioned by "straight-line" accounts. Already by Gordon's time, a picture had emerged of assimilation as a complex and only partially interlocking set of processes (see also Yinger 1981). Some of these (notably structural assimilation on Gordon's account and spatial assimilation on some recent accounts [Massey and Denton 1993: 149 ff.]) bear significantly on processes in other domains by shaping opportunity structures and contact probabilities. But other domains are at most loosely coupled with one another. Recent accounts are sensitive to the possibilities of different rhythms and trajectories of assimilation—or dissimilation—in different domains (Banton 1983: 144-46). On current understandings, assimilation is always domain-specific and relative to a particular reference population; and the normative stance one takes toward it will also depend on the particular domain and reference population.
Conclusion: A Concept Transformed
In all three countries, what I have called "the return of assimilation" has involved a subtle but significant change in perspective. Analytically, this has involved a shift from an overwhelming focus on persisting difference—and on the mechanisms through which such cultural maintenance occurs—to a broader focus that encompasses
The Return of Assimilation? ■ 129
emerging commonalities as well. Normatively, it has involved a shift from the automatic valorization of cultural differences to a renewed concern with civic integration.
This shift in analytical and normative emphasis does not presage a radical reversal. It does not amount to a return to the bad old days of arrogant assimiiationism. For while the term "assimilation" has returned, the concept has been transformed. I sketch in conclusion the main elements of this transformation:
1. A shift from organic understandings of assimilation, focusing on an end state of complete absorption, to abstract understandings, focusing on a process of becoming similar (in some respect, to some reference population).
2. A shift from transitive to intransitive understandings of assimilation. The former see populations of immigrant origin as moldable, meltable objects; the latter see persons comprising such populations as active subjects. As such, to be sure, they are not busy consciously "assimilating." Assimilating can,
of course, be a deliberate, self-conscious activity, and the poignant—and sometimes tragic—ambiguities and ambivalences bound up with it have been movingly explored by novelists, memoirists, essayists, historians, and even a few sociologists (Bauman 1988; Laitin 1995a). Yet for most historians and social scientists, assimilation is an emergent property of social processes at an aggregate level, rather than something that happens (consciously or unconsciously) at the level of individual persons. As an emergent tendency at the aggregate level, assimilation is largely unintended and often invisible; and when made visible, it may be lamented. Yet even when it is lamented, the processual tendency we call "assimilation" is not something done to persons, but rather something accomplished by them, not intentionally, but as an unintended consequence of myriad individual actions and choices in particular social, cultural, economic, and political contexts (cf. Alba 1995: 4).
3. The unit within which change occurs—the unit that undergoes assimilation—is not the person but a multigenerational population. Population-level assimilation can occur without any individual-level assimilation. Linguistic assimilation at the
The Return of Assimilation?
The Return of Assimilation? • 131
population level, for example, can occur without any adult learning a new language, simply through the acquisition of the language of the reference population by children. Of course this is not what ordinarily happens; we do observe some language shift at the level of individual persons. But key changes (in language and in other domains) occur intergenerationally; they occur not within persons but within abstractly constructed multigenerational populations, as new (genealogical) "members" of the population turn out to be different—that is, they dissimilate—from other, older members of the source population, in ways that make them more similar to members of some reference population.
4. A shift from thinking in terms of homogeneous units to thinking in terms of heterogeneous units. Assimilation does not involve a shift from one homogeneous unit to another. It involves, rather, a shift from one mode of heterogeneity—one distribution of properties—to another mode of heterogeneity, that is, to a distribution of properties more similar to the distribution prevailing in some reference population.
5. A shift in the focus of normative concern informing research on assimilation from cultural to socioeconomic matters.
A general openness to cultural diversity, coupled with confidence among specialists—if not always among the wider public—in the continuing robustness of processes of linguistic acculturation (Portes and Schauffler 1994) has alleviated anxieties about cultural dimensions of assimilation. Yet the bifurcation of recent immigrants into high-skill and low-skill segments—at a moment when macroeconomic changes associated with the "hourglass economy" have decreased the rewards to low-skill, uneducated labor—has generated concerns about long-term structural marginalization (Gans 1992; Portes and Zhou 1993; somewhat more optimistic: Waldinger 1996; Perlmann and Waldinger 1997). As a normatively charged concept, assimilation, in this sense, is opposed not to difference 1 but to segregation, ghettoization, and marginalization.
6. A shift from a holistic approach that conceptualized assimilation towards a taken-for-granted reference population—the j "core culture" or "national society" as a whole—to a
disaggregated approach that discards the notion of assimilation as a single process, considers multiple reference populations, and envisions distinct processes occurring in different domains. This has entailed a shift from the monodimensional question, "how much assimilation?" to the multidimensional question, "assimilation in what respect, over what period of time, and to what reference population?" It has also entailed a shift from an assimilationist understanding of assimilation—a global empirical expectation and normative endorsement of assimilation—to an agnostic stance, varying by domain and reference population, concerning both the likelihood and the desirability of assimilation.
Reformulated in this manner, and divested of its "assimilationist" connotations, the concept of assimilation—if not the term itself—seems not only useful but indispensable. It enables us to ask questions about the domains and degrees of emergent similarities, and persisting differences between multigenerational populations of immigrant origin and particular reference populations. There are good reasons for us to want to ask such questions, regardless of whether we applaud or lament such emerging similarities. Naturally, to pose such questions is only a beginning. Assimilation is not a theory; it is simply a concept. But it is a concept we can ill do without.
CHAPTER SIX
'Civic" and "Ethnic" Nationalism
From its late nineteenth-century beginnings to the present, the study of nationhood and nationalism has been marked by deep ambivalence and intractable ambiguity. On the one side, nationalism has been associated with militarism, war, irrationalism, chauvinism, intolerance, homo-genization, forced assimilation, authoritarianism, parochialism, xenophobia, ethnocentrism, ethnic cleansing, even genocide; it has been characterized as the "starkest political shame of the twentieth century" (Dunn 1979: 55). On the other side, nationhood and nationalism have been linked to democracy, self-determination, political legitimacy, social integration, civil religion, solidarity, dignity, identity, cultural survival, citizenship, patriotism, and liberation from alien rule.
One reason for the ambivalence, of course, is that "nation" and "nationalism" designate a whole world of different things. To a great extent, the ambivalence reflects not so much competing understandings and evaluations of the same thing, as alternative uses of the same : term. Much of the ambivalence, that is, has been rooted in ambiguity. How people have evaluated nationalism has depended on what they have understood it to be.
Recognition of the protean quality of "nation" and "nationalism"— and of the normative ambivalence and conceptual ambiguity surrounding the subject—has engendered innumerable attempts at classification. Some typologies have been elaborate. In his early book Theories of Nationalism, for example, Anthony Smith classified national
"Civic" and "Ethnic" Nationalism • 133
movements by the "formal" criteria of "intensity" and "achievement" and by the "substantive" criteria of "independence" and "distinctiveness." The former yielded six types, the latter 12; cross-classifying them, with some simplification, yielded no fewer than 39 types for which Smith found corresponding historical or contemporary instances (Smith 1983 [1971]: 211-29). Most classifications, however, have been quite simple, often founded on a single dichotomous distinction. And such distinctions have often been intended to do both normative and analytical work.
The most well known distinctions—between voluntaristic and organic, political and cultural, subjective and objective, liberal and illiberal, and civic and ethnic forms of nationalism—overlap to a great extent. They have an illustrious pedigree, going back to Friedrich Mei-necke's (1919 [1907]) distinction between Staatsnation and Kulturna-tion at the beginning of the century and, more immediately, to Hans Kohn's (1944) influential midcentury work, usually glossed as distinguishing between "Western" and "Eastern" forms of nationalism.1
Of these overlapping distinctions, the one with the greatest resonance today, especially outside the narrow circle of researchers working primarily on nationalism, is the distinction between civic and ethnic understandings of nationhood and forms of nationalism. This has been used to suggest that there are, fundamentally, only two kinds of nationalism: civic nationalism, characterized as liberal, voluntarist, universalist, and inclusive; and ethnic nationalism, glossed as illiberal, ascriptive, particularist, and exclusive. These are seen as resting on two corresponding understandings of nationhood, based on common citizenship in the first case, common ethnicity in the second.
Sometimes, as in Kohn's work, this distinction is projected in space, and used to contrast the civic nationalism of Western Europe, or of "the West" in general, with the ethnic nationalism of Eastern Europe or other world regions. Such grand contrasts of world regions easily acquire a neo-orientalist flavor and lend themselves to the invocation of a dubious series of linked oppositions—between universalism and particularism, inclusion and exclusion, civility and violence, reason and passion, modern tolerance and ancient hatreds, transnational integration and nationalist disintegration, civic nationhood and ethnic nationalism.2
But this is not the prevalent use of the distinction today. The triumphalist—or, at best, complacent—account of Western civic
134 ■ "Civic" and "Ethnic" Nationalism
nationalism is too obviously problematic for this view to be seriously entertained. The unexpected (and partly nationalist) resistance to the Maastricht treaty, the longstanding violent conflicts in northern Ireland and the Basque country, the intensifying ethnopolitical conflict in Belgium, and the electoral successes of xenophobic parties in many countries—all these have made it impossible to hold such an uncritical view of the essentially "civic" quality of West European nationalism.
More common is the use of the civic-ethnic opposition to make distinctions between states—or between national movements—rather than between whole world regions. This is often done in an ideological mode, to distinguish one's own good, legitimate civic nationalism from the illegitimate ethnic nationalism found elsewhere. The leaders of postindependence Ukraine and Kazakhstan, for example, have selfconsciously used the language of civic nationhood to present their states to domestic and especially international audiences as paragons of civic inclusiveness and tolerance, as states of and for all their citizens, rather than as states of and for a single ethnocuttural group. They—and scholars sympathetic to their cause—have pointed to inclusive citizenship legislation, liberal language laws, and rhetoric of civic inclusiveness to mark a contrast with Estonia and Latvia, with their restrictive citizenship legislation, tough language laws, and rhetorical emphasis on ethnocultural survival.
Many separatist movements, too, use this self-legitimating language of civic nationalism. The general election manifesto of the Welsh nationalist party Plaid Cymru, for example, proclaims its commitment to a "civic nationalism [that] welcomes all those living in Wales to join us in finding the solutions to [social and environmental] challenges and in restoring the equilibrium of social justice and environmental sustainablity in Wales and Europe."^ Scottish National Party (SNP) leaders emphasize even more strongly the party's civic nationalism, especially its inclusive, residentially based definition of Scottishness. So pronounced is this emphasis that a fringe nationalist group opposed to the SNP's rhetoric of civic nationalism has caustically criticized the "hogwash about being Scottish just because you happen to live in Scotland ... it is to be hoped that Scottishness will, through means of education and restored ethnic consciousness, cease to be the sad joke which in many cases it has become."4
Scottish nationalist leaders generally like to align themselves with the Catalan, Quebecois and other regional nationalisms. Yet they are
"Civic" and "Ethnic" Nationalism • 135
will'11? to distance themselves from these movements to underscore their own commitment to civic nationalism. For example, after the narrow defeat of the Quebec sovereignty referendum in 1995, notoriously blamed by Quebecois separatist leader Jacques Parizeau on the "ethnic vote," SNP leader Alex Salmond said that "Quebec is not Scotland and Scotland is not Quebec. . . . The linguistic and ethnic basis of their nationalism is a two-edged sword. ... we follow the path of civic nationalism."5 For their part, Quebec nationalists have sought in recent years to project a more "modern," unifying image of civic nationalism. But Parizeau's gaffe, together with a remark a few weeks earlier by separatist leader Lucien Bouchard about the low birth rate of Quebecois, allowed critics of Quebecois nationalism to turn the civic-ethnic distinction back against their opponents. To cite but one of many examples, the Toronto Globe and Mail, Canada's leading Anglophone newspaper, characterized Quebecois separatism as "rooted in ethnic rather than civic nationalism. Blood is more important than citizenship."6
Paralleling this frankly political use of the civic-ethnic distinction to legitimize or discredit particular state policies or nationalist movements is its use in a scholarly mode to draw distinctions between different instances of nationalism and different modes of national self-understanding. Often this scholarly accounting of nationalism— bestowing the imprimatur of the civic on some states or movements, denying it to others—itself belongs to the sphere of nationalist politics in a broad sense. There is nothing new about this; for a century and a half, scholars have been participants in, as well as observers of, nationalist politics. But the work done by the notion "civic," with its normative prestige, in such accounts may be more political than analytical: it may speak more to the putative international respectability and legitimacy of the state or movement in question than to its empirical characteristics.
In recent years, many scholars of nationalism have grown uncomfortable with the unequivocal sorting of cases into "civic" and "ethnic" categories. From a detached, analytical point of view, as numerous commentators have pointed out, it is often impossible, or at best problematic, to characterize an entire state, or an entire national movement, simply as civic or ethnic. As a result, efforts have been made to use the distinction in a more abstract manner. Instead of being used to characterize concrete cases, it is now most often used to characterize
136 • "Civic" and "Ethnic" Nationalism
opposed analytical "elements" or tendencies and to show how they are mixed in different manners and proportions in concrete cases. Indeed so prevalent in the literature is this notion that individual states or national movements display a mixture of civic and ethnic elements or tendencies that it can be said to constitute a kind of theoretical "common sense."
In the hands of sophisticated observers such as Anthony Smith, whose Ethnic Origins of Nations was particularly influential in promoting it, this use of the civic-ethnic distinction to designate analytical elements that are found in concrete cases "in varying proportions at particular moments of their history" (Smith 1986: 149) is certainly an improvement over the unequivocal sorting of states and nationalist movements as a whole—to say nothing of entire regions—into "civic" or "ethnic" categories.7 Yet even in this more abstract and analytical mode, I want to argue, the civic-ethnic distinction remains both analytically and nor-matively problematic.8
Analytical Ambiguities
Let me begin with what I see as the analytical weaknesses of the civic-ethnic distinction. Both terms are deeply ambiguous. Their ambiguity can be highlighted by asking how culture fits in to the civic-ethnic scheme. There are in fact two very different ways of mapping culture onto the civic-ethnic distinction, but I will argue that neither is fully satisfactory.
What is "ethnic" about ethnic nationalism? Advocates of the civic-ethnic distinction have a ready answer: nation-membership is understood to be based on ethnicity. But this simply pushes the question one step back. What is "ethnicity"? As analysts going back to Max Weber have emphasized, "ethnicity" is an exceedingly ambiguous notion.9 Consider here just one aspect of that ambiguity, involving the relation between "ethnicity" and culture.
On the one hand, ethnic nationalism may be interpreted narrowly, as involving an emphasis on descent, and, ultimately, on biology. "Strictly speaking," as Anthony Smith noted in his first book on nationalism, "ethnicity refers to common descent" (Smith 1983 [1971]: 180). Yet construing ethnicity narrowly in this manner severely constricts the domain of ethnic nationalism. For as Smith himself went on to observe, many "commonly accepted 'nations1 ... do not invoke a common
i-
"Civic" and "Ethnic" Nationalism • 137
ancestor," and even when nationalist argumentation does involve "imputed common descent," this is "usually a minor claim" (ibid., 180-81).10 On the strict understanding of ethnicity, nationalist rhetoric that emphasizes common culture, but not common descent,11 has to be coded as a kind of civic nationalism.12 But then the Category of civic nationalism becomes too heterogeneous to be useful, while that of ethnic nationalism is severely underpopulated.
On the other hand, "ethnic" may be construed broadly, as ethno-cultural. This is the path Smith chose in Theories of Nationalism, treating "' ethnic' [as] identical with the term 'cultural,' without further specification" (1983 [1971]: 180). In this case, the problem is just the opposite: virtually all nationalisms would have to be coded as ethnic. Thus for Eric Hobsbawm, "Every separatist movement in Europe . . . bases itself on 'ethnicity,' linguistic or not, that is to say on the assumption that 'we'—the Basques, Catalans, Scots, Croats, or Georgians are a different people from the Spaniards, the English, the Serbs or the Russians" (Hobsbawm 1996b: 256). By defining "ethnicity" so expansively that it is coextensive with a sense of separate "peoplehood," however that sense of peoplehood is grounded, Hobsbawm codes as "ethnic" what others often classify as "civic"—Catalan and Scottish nationalism, for example. Civic nationalism is thereby reduced to an empty set or, as on Hobsbawm's account, relegated to an earlier phase of historical development.
Nor is ambiguity limited to the term "ethnic." The category "civic" is equally ambiguous. On the one hand, civic nationalism may be interpreted strictly, as involving an acultural, ahistorical, univer-salist, voluntarist, rationalist understanding of nationhood. "The nation" is then construed as a voluntary association of culturally unmarked individuals. Nation-membership is understood as chosen rather than given, as a "daily plebiscite," in Renan's celebrated metaphor.
Yet construing civic nationalism strictly in this fashion risks defining the phenomenon out of existence. Even the cases most often cited as paradigmatic of civic nationalism—France and America—involve a crucial cultural component or, in Hobsbawm's terms, a strong sense of separate peoplehood,13 A purely acultural understanding of nationhood has never been widely held. It is a model of nationhood that has never been instantiated, existing only as a conceptual ideal type. Even as an ideal type, it is problematic. Although Ernest Renan is often cited
138 • "Civic" and "Ethnic" Nationalism
as the locus classicus for this model, this reflects a one-sided reading ol :B his famous lecture. His characterization of the existence of a nation as a "daily plebiscite"—a self-conscious rhetorical flourish which Renah S prefaced by asking his audience to "pardon the metaphor"—does indeed underscore the importance, for Renan, of subjective self-understanding in constituting nationhood (Renan 1996 [1882]: 53) But Renan's understanding of nationhood is far from acultural oi purely voluntaristic. It is a "thick," not a "thin" understanding. Renan f stresses the constitutive significance of the "possession in common of s 0 rich legacy of memories"; he characterizes the nation as "the culmina- 3 tion of a long past of endeavors, sacrifice, and devotion" (ibid., 52), In -If this sense, the nation is "given" as well as "chosen."14
On the other hand, civic nationalism may be defined broadly. The definition offered by Michael Keating, a sympathetic yet sophisticated "S analyst of Scottish, Catalan, and Quebecois nationalisms, is worth quoting at length. Keating defines civic nationalism as a collective en-terprise
rooted in individual assent rather than ascriptive identity. It is based on common values and institutions, and patterns of social interaction. The bearers of national identity are institutions, customs, historical memories and rational secular values. Anyone can join the nation irrespective of birth or ethnic origins, though the cost of adaptation varies. There is no myth of common ancestry . . . [Nationhood is] based on territorially defined community, not upon a social boundary among groups within a territory. This is not to say that any piece of real estate can form the basis for a nationalism. There needs to be a structured set of political and social interactions guided by common values and a sense of common identity (Keating 1996: 5-6). Keating wants to have it both ways. He retains the rationalist, univer-salist emphasis on choice characteristic of "thin" understandings of civic nationalism. At the same time his more sociologically realistic understanding of nationhood pushes him to acknowledge the importance of "common values," "customs," "historical memories," and a "sense of common identity." Yet these are just the sort of particularism thick, given factors highlighted by broad, culturalist understandings of ethnicity. The factors highlighted by Keating are not all that dif-ferent, for example, from the quartet of "myths, memories, values, and symbols" emphasized by Anthony Smith in The Ethnic Origins of Nations.
"Civic" and "Ethnic" Nationalism • 139
To sum up the argument so far: A narrow understanding of ethnicity severely constricts the domain of ethnic nationalism and leaves the residually defined civic category too large and heterogeneous to be useful. Conversely, a narrow understanding of "civic" severely constricts the domain of civic nationalism and leaves the residually defined ethnic category too large and heterogeneous to be useful. If one combines a strict understanding of civic and a strict understanding of ethnic nationalism, then one is left with few instances of either one and a large middle ground that counts as neither, and one can no longer think of the civic-ethnic distinction as an exhaustive way of classifying types or manifestations of nationalism. If one combines, finally, a broad understanding of civic and a broad understanding of ethnic nationalism, one confronts a large middle ground that could be classified either way, and one can no longer think of the civic-ethnic distinction as mutually exclusive.
Advocates of the civic-ethnic distinction would argue that this large middle group consists of cases that combine civic and ethnic elements in varying ways. But the problem is not that it is difficult to know, on balance, how to classify a "case." The problem is rather that the deep ambiguity of the terms "civic" and "ethnic," and in particular the uncertain place of culture in the civic-ethnic scheme, calls into question the usefulness of the distinction itself. It can be just as difficult to classify an "element" as it is to classify an entire "case."
How, for example, are we to classify policies designed to promote a particular language at the state or provincial level? From the point of view lyrically articulated by Benedict Anderson, for whom nations are "conceived in language, not in blood," and are therefore "joinable in time" (Anderson 1991: 145), there can be nothing "ethnic" about such policies, even if they might be judged restrictive, illiberal, or even chauvinistic. Indeed, from another point of view one could go further and characterize such policies as positively civic, that is, as indispensable for the promotion of republican citizenship. The assimilationist language policies of the French Revolution were justified in just such a civic idiom in Abbé Grégoire's report "On the necessity and means of abolishing the patois and universalizing the use of the French language." Only when all citizens speak the same language, the report argued,  can   all  citizens   "communicate  their  thoughts without hindrance" and enjoy equal access to state offices (de Certeau et al. 1975: 302).i5 And as John Stuart Mill put it in Considerations on
140 • "Civic" and "Ethnic" Nationalism
Representative Government, "Among a people without fellow-ltelin^, especially if they speak different languages, the united public opinion necessary to the working of representative government, cannot exist" (Mill 1975 [1861]: 382).
From another point of view, however, linguistic nationalism js simply a particular expression of ethnic nationalism. When "ethnic'' is understood broadly as ethnocultural, or simply as cultural without qualification, then conceptualizing the nation as a community of language, demanding autonomy or independence in the name of such a community, limiting access to citizenship to persons knowing the Ian guage, and promoting or requiring teaching, publishing, broadcasting, administering, or advertising in that language must be considered central, indeed paradigmatic manifestations of ethnic nationalism.
Normative Ambiguities
The distinction between civic and ethnic understandings of nationhood and forms of nationalism is not only, or even primarily, an analytical distinction. It is also, at the same time, a normative one. This fusion of analytical and normative criteria was characteristic already of Hans Kohn's work. Kohn's portrayal of pioneering Western nationalisms joined neutral analytical observations about their "predominantly political" character, reflecting the fact that national consciousness developed within the framework of existing states, to a normative celebration of the spirit of "individual liberty and rational cosmopolitanism" that he saw as informing such nationalisms. Similarly, his portrayal of the later nationalisms of Germany and central and Eastern Europe joined neutral analytical observations about their initially cultural character, reflecting the fact that national consciousness developed outside of and in opposition to the framework of existing states, to a normatively charged evocation of the illiberal tendencies that he saw as inherent in those nationalisms (Kohn 1944: 329-31).
Even as the distinction has been stripped, in most uses, of the concrete spatial reference given to it by Kohn, it has retained the same normative valence. Civic nationalism is generally glossed as liberal, voluntarist, universalist, and inclusive, ethnic nationalism as illiberal, ascriptive, particularism and exclusive. Except for the opposition between universalism and particularism, which finds contemporary
"Civic" and "Ethnic" Nationalism • 141
partisans on both sides, it is hard to imagine a more normatively loaded, one-sided characterization. Who could have a good word for a form of nationalism routinely glossed as illiberal, ascriptive, and exclusive? How could one criticize a form of nationalism understood to (jt. liberal, voluntarist, and inclusive? When civic and ethnic nationalism are paired, the former is invariably a term of praise, the latter of abuse.
Yet although the normative opposition seems unambiguous, matters are in fact more complicated. Take for example the characterization of civic nationalism as inclusive and of ethnic nationalism as exclusive. 16 In fact all understandings of nationhood and all forms of nationalism are simultaneously inclusive and exclusive. What varies is not the fact or even the degree of inclusiveness or exclusiveness, but the bases or criteria of inclusion and exclusion.17
Civic understandings of nationhood are glossed as inclusive for one of two reasons. The most common is that the civic nation is based on citizenship, and therefore includes all citizens, regardless of their particularistic traits. But citizenship itself, by its very nature, is an exclusive as well as an inclusive status. On a global scale, citizenship is an immensely powerful instrument of social closure (Brubaker 1992). It shields prosperous and peaceful states from the great majority of those who—in a world without borders and exclusive citizenries—would seek to flee war, civil strife, famine, joblessness, or environmental degradation, or who would move in the hope of securing greater opportunities for their children. Access to citizenship is everywhere limited; and even if it is open, in principle, to persons regardless of ethnicity, this is small consolation to those excluded from citizenship, and even from the possibility of applying for citizenship, by being excluded from the territory of the state. This "civic" mode of exclusion is exceptionally powerful. On a global scale, it is probably far more important, in shaping life chances and sustaining massive and morally arbitrary inequalities, than is any kind of exclusion based on putative ethnicity. But it is largely invisible, because we take it for granted. Only among philosophers and political theorists, in recent years, has there been some attention to issues such as open borders, or some moves to recast Rawlsian accounts of justice on a global scale.18 In wider spheres of public debate, this kind of closure and exclusion is simply never questioned.
Civic understandings of nationhood have also been characterized as
142 • "Civic" and "Ethnic" Nationalism
inclusive because they comprise "all those—regardless of race, u>lor creed, gender, language, or ethnicity—who subscribe to the nation's political creed" (Ignatieff 1993: 6). The emphasis on a constltntive political creed echoes an older literature on American nationalism, .it-cording to which American national identity was essentially ideological and therefore uniquely open.1s That view has been much criticized in the last two decades, notably by Rogers Smith, who sees American understandings of nationhood as pervasively informed, for much of the country's history, by an ethnocultural or "inegalitarian ascriptive" strand of thinking as well as by liberal and republican strands (Smith; 1997: 2ff., 14 ff.). But even apart from its historical accuracy in the American context, the creedal model of membership has its own logic of exclusion. The French Revolution provides the paradigmatic examples of such exclusions—of emigres, refractory priests, noblemen, rebels, and other presumed political opponents. At the opposite end of the political spectrum, McCarthyism provides the paradigmatic example in the American context. But it is worth remembering that even in Germany—often treated as the key exemplar of ethnic nationalism— Catholics and Social Democrats were excluded from the moral community of the nation and characterized as internal "enemies of the Reich" in Bismarck's time not by virtue of ethnicity, but by virtue of their imputed lack of loyalty to the national state.
Understandings of nationhood as based on citizenship or political creed, then, are not more inclusive than those that emphasize cultural community or common descent; they are differently inclusive (and exclusive). And not only are the exclusions on which they are premised normatively problematic, but so too, in certain contexts, is their very inclusiveness. Transylvanian Hungarians, for example, resent and resist the putatively inclusive, citizenship-based rhetoric of nationhood which construes them as members of the Romanian nation. On their own self-understanding, they are citizens of the Romanian state, but members of a Hungarian cultural nation that cuts across the boundaries of state and citizenship.
In the early 1980s—to take another example—some second generation Algerian immigrants protested against the French nationality that had been attributed to them automatically at birth. For reasons having to do with a technicality of French citizenship law, they had been unaware of this attribution until, upon reaching age 16 and applying for residence permits as foreigners, they were stupefied to be told by
"Civic" and "Ethnic" Nationalism • 143
officials that they were French. While some welcomed this news— S r French nationality, after all, would protect them against expulsion— o[jlt.rs "experienced the attribution of French nationality as a violation of their personality, their familial attachments, and their membership of a newly emancipated [Algerian] nation" (GISTI 1983: 6), and several thousand formally requested—in vain—to be released from the nationality that had been attributed to them without their knowledge, against their will, and in violation of their self-understanding as Algerians. The Algerian government too objected to the unilateral imposition of citi-- P ■ zenship on "its" emigrants; after "the years of murderous conflict aimed precisely at giving them their own nationality," this was regarded as a neocolonial affront to Algerian sovereignty (Mangin 1981: 23).
The conventional gloss of civic and ethnic understandings of nationhood as voluntaristic and ascriptive, respectively, is also problematic. In the first place, it is greatly overdrawn. Only on implausibly acultural and ahistorical construals of civic nationalism can nation-membership be understood as entirely voluntary; on richer and more realistic accounts, including Renan's own account, as we have already seen, the nation is understood as given as well as chosen. On the other hand, choice is far from irrelevant in settings where nationhood is understood to be based on ethnocultural commonality, such as Central and Eastern Europe, usually considered the locus classicus of ethnic -:1       nationalism. As Hobsbawm observed, commenting on the "paradoxes *:       of primordial ethnicity," "early twentieth century Europe was full of " ,        men and women who, as their very names indicate, had chosen to be Germans or Magyars or French or Finns" (Hobsbawm 1996b: 260, •        259; emphasis in the original).
| Moreover, the normative valence of the opposition between chosen-
-[ ness and givenness is more complex than the loaded contrast between \nluntary and ascriptive suggests. Liberal moral and political theory ] have indeed celebrated voluntary engagements, commitments, and affiliations over ascribed statuses. But the communitarian critique of liberalism (Sandel 1982) and the development of a variant of liberalism more sensitive to the cultural contexts of choice (Kymlicka 1989) have led to an enhanced appreciation of the ways in which choices are meaningful only against the horizon of unchosen cultural contexts. And this in turn has led to a tempering and relativization of the opposition between chosenness and givenness. I have mentioned Kymlicka in connection with newly "culturalist"
144 • "Civic" and "Ethnic" Nationalism
accounts of liberalism. But he has also, of course, been a central figure in recent discussions of multiculturalism (Kymlicka 1995). These discussions, too, have problematized the normative opposition between civic and ethnic nationalism. By valorizing particular cultural attachments and identities—including ethnic or ethnocultural ones—and by seeing the public recognition of such particularistic attachments as central to and supportive of rather than antithetical to citizenship (even to liberal citizenship, on Kymlicka's account), multiculturalism destabilizes and relativizes the normative contrast between civic and ethnic nationalism.
A Modest Alternative
From an analytical point of view, a less ambiguous distinction than that between civic and ethnic nationalism can be drawn between state-framed and counter-state understandings of nationhood and forms of nationalism. In the former, "nation" is conceived as congruent with the state, and as institutionally and territorially framed by it. In the latter, "nation" is imagined as distinct from, and often in opposition to, the territorial and institutional frame of an existing state or states. The former is equivalent to Meinecke's notion of the Staatsnation; the latter, however, is a wider category than Meinecke's Kulturnation.
There is not necessarily anything "civic"—in the normatively robust sense of that term—about state-framed nationhood or nationalism. It is the state—not citizenship—that is the cardinal point of reference; and the state that frames the nation need not be democratic, let alone robustly so.20 The sense of "nation" that developed gradually in ancien regime France was framed by the state from the beginning, but it became linked to ideas of citizenship only during the Revolution. To take another example, when Prussian reformers sought to transform Prussia into a "nation" in the early nineteenth century, to "do from above what the French had done from below," as one of the leading reformers put it, the "nation" they envisaged—Prussian, not German!—was conceived as framed by the state, yet one could not characterize it as based on citizenship. The same is true of the nationalisms of many authoritarian contemporary states.
Moreover, the notion of state-framed nationhood or nationalism enables us to talk about the way in which linguistic, cultural, and even
"Civic" and "Ethnic" Nationalism • 145
(narrowly) ethnic aspects of nationhood and nationalism may be framed, mediated, and shaped by the state. For while there is a definitional antithesis between civic nationhood and ethnicity—and in some interpretations between civic nationhood and culture—there is no such antithesis between state-framed nationhood and ethnicity or culture. State-framed nationalisms are often imbued with a strong cultural content.21 France, for example, is a paradigmatic instance of state-framed nationhood. At the same time, culture is understood as constitutive of French nationhood.22 There is no contradiction here. The culture that is understood to be constitutive of nationhood is a pervasively state-framed, and, in modern times, state-propagated one; it is not conceived as prior to and independent of the territorial and institutional frame of the state.
Counter-state nationalisms, on the other hand, need not be specifically ethnic; nationhood conceived as distinct from or in opposition to an existing state need not be conceived in ethnic terms, or even, more loosely, in ethnocultural terms. Quite apart from the difference, discussed above, between narrowly ethnic and broadly ethnocultural understandings of nationhood, counter-state definitions of nation may be based on territory, on historic provincial privileges, or on the possession of a distinct political history prior to incorporation into a larger state. Early anti-Habsburg Hungarian nationalism, for example, was couched in the idiom of historic constitutional privileges until the end of the eighteenth century, when increasing emphasis began to be placed on protecting and developing the Magyar language. An intriguing contemporary example is furnished by Northern Italian regional nationalism, in which "Padania" (the term refers to the Po river valley) is conceptualized not simply as a "region" but as a north Italian "nation" entitled to national self-determination.
Moreover, even when the nation in question is defined in cultural or ethnic terms, counter-state nationalisms may partake of "civic" qualities. This is most evident in cases such as Catalonia, Scotland, or Quebec where there is an institutionally defined sphere within which a substantial degree of self-government is possible (Keating 1996). But even counter-state nationalist movements without a formally secured sphere of institutionalized autonomy within the larger state can provide settings for the cultivation and exercise of "civic" virtues—for example by organizing and running schools, credit associations, cooperative enterprises, and welfare organizations.
1.46 • "Civic" and "Ethnic" Nationalism Conclusion
The civic-ethnic distinction addresses important analytical and normative issues, but it does not do so in a satisfactory fashion. It can be seen as a routinization and codification of the various efforts scholars have made to come to terms with the normative ambivalence and empirical ambiguity surrounding the protean phenomena grouped under the umbrella term "nationalism." It represents an effort to domesticate these normatively and empirically unruly phenomena, to impose conceptual and moral order on them, to subsume them under a convenient formula, to render them suitable grist for academic mills.
Yet nationalism resists neat parsing into types with clearly contrasting empirical and moral profiles. Distinctions are of course unavoidable in analytical and normative inquiry alike, but we should not expect too much of a single distinction. The civic-ethnic distinction is overburdened; it is expected to do too much work. We would do better to disentangle the work of analytical ordering from that of normative appraisal. The distinction between state-framed and counter-state understandings of nationhood is offered as one modest way of doing some of the analytical work done by the civic-ethnic distinction without the attendant confusion. The inexhaustible moral and political ambiguities and dilemmas generated by nationalism can then be addressed on their own terms.
I i
o&>  CHAPTER SEVEN
Ethnicity, Migration, and Statehood in Post-Cold War Europe
Among the most salient and politically charged issues of the last two decades in Europe have been questions of ethnicity, migration, and statehood. These closely interlinked issues have figured centrally in political, cultural, and social transformations throughout the continent. In eastern Europe, they are often understood to be linked in a vicious circle. States founded on ethnicity—and understood as the states of and for particular ethnocultural nations—are seen as engendering violent conflict and forced migration. Ethnic cleansing has come to epitomize this diabolical intertwining of ethnicity, migration, and statehood. In western Europe, by contrast, some observers have seen a more benign intertwining. The postnational erosion of sovereign statehood, on this view, has produced a continent-wide space for free migration, and has allowed previously suppressed ethnoregional cultures—and even autonomous ethnonational polities like Catalonia and Scotland—to flourish. Ethnicity, on this account, has been uncoupled from statehood. Darker accounts, to be sure, stress migration from outside Europe, which is seen as generating unwanted ethnic pluralism, newly ethnicized or re-ethnicized understandings of nationhood, and pressures for a renationalization of the state (or for a statelike— and perhaps nationlike—"fortress Europe" that would keep outsiders at bay). While none of these accounts is particularly nuanced, each points to the importance of the intertwined themes of ethnicity, migration, and statehood, and together they suggest that these issues can
148 • Ethnicity, Migration, and Statehood
be configured in quite different ways. In this essay, I seek to specify persisting differences in the way these questions are posed in different parts of Europe, yet to avoid the often caricaturaljy oversimplified east-west contrasts that inform many accounts of contemporary Europe.
Ethnicity
Almost all European societies, like almost all societies worldwide, arc ethnically heterogeneous, but that heterogeneity takes sharply differing forms. In order to highlight crucial differences in the configuration-— the genesis, form, and political consequences—of ethnic heterogeneity in Europe, I distinguish two ways in which ethnic heterogeneity can be socially organized and politically expressed. The first I call "immigrant ethnicity," and the second, "territorial nationality."1
On the first model, characteristic mainly of western Europe, ethnic groups arise through migration and are generally territorially dispersed.2 On the second model, characteristic of east central and eastern Europe, ethnic groups are indigenous (or at least make claims to be so); they are in many cases generated by the movement of borders across people, rather than that of people across borders; and they are generally territorially concentrated. Their members are ordinarily citizens of the country in which they reside, yet they often identify culturally and sometimes politically with a neighboring "kin" or "homeland" state, to which they see themselves as "belonging" by shared ethnicity or culture, though not by legal citizenship {Brubaker 1996). Lastly, and crucially, they define themselves in national terms. They see themselves as belonging not simply to a distinct ethnic group, but to a distinct nation or nationality that differs from the nation or nationality of their fellow-citizens. In this second model, then, ethnicity takes the form of nationality, and ethnic heterogeneity is coded as national heterogeneity. This territorial ethnicity-as-national ity is very different from immigration-engendered polyethnicity. Using the same term—"ethnicity" or "ethnic minorities"—to designate both can be misleading.
The political claims that can be made in the name of ethnicity differ sharply in the two cases. Immigrant ethnicity evokes a politics of antidiscrimination, civic inclusion, and "soft multiculturalism" {claims to recognition, resources, and sometimes immunities and exemptions).
Ethnicity, Migration, and Statehood • 149
I v -       Territorial nationality involves claims for national self-determination, for symbolic recognition as a state-bearing nation rather than as a mere ,. "minority," for extensive language rights, for territorial autonomy or
IE; even full independence, and sometimes for rapprochement with a a neighboring "kin" or "homeland" state.
i Clearly, the claims of territorial nationality can threaten the basic na-
| ture of the state in a way that the claims of immigrant ethnicity gener-
al ally do not. When ethnic claims become national claims, based on
putative territorial nationhood and nationality, they become more fun-1 damental, and potentially more threatening, precisely because they raise
\ what Linz and Stepan (1996) have called the "stateness" problem—the
• problem of the integrity and boundaries of the state.1
In east central Europe, ethnicity speaks this potentially explosive 2, language of nationality. Nationality or nationhood, in turn, is under-
4 ;       stood as based on ethnicity (language, culture, a vague sense of shared
i descent, and so on), rather than on citizenship or state frontiers. One
might say that ethnicity is nationalized, while nationality and nation-
; hood are ethnicized. In western Europe, in contrast, after decades of heavy labor migration and subsequent family reunification, public attention has focused on immigrant ethnicity, while ethnic claims have not generally been framed as national claims.
1 There are, of course, important exceptions to this pattern on both
sides. In much of east central Europe, there are fundamental issues as-
'■        sociated with the large, socially stigmatized, spatially segregated, and
,        in large part economically marginalized Gypsy or Roma population
a|        (Barany 2002). These issues are sui generis and cannot be neatly subsumed under our usual conceptual rubrics. Depending on how Roma
I       are represented by others, and how they represent themselves, they
]       can be conceived as an ethnic group, a national group, a caste, or a so-
.]       rial underclass (Vermeersch 2003 ).4
In western Europe, on the other hand, ethnicity sometimes involves claims to territorial nationality or nationhood, and the politics of eth-
f       nicity then becomes a politics of national autonomy and self-
• determination. This is true above all in Spain, Belgium, and Britain, all \       of them multinational (and not simply multiethnic) polities. There is
also the interestingly ambiguous case of Italy, where the Northern J       League sometimes claims that northern Italy, or Padania, is a distinct nation. Only in the case of Northern Ireland—the western European
• case most similar to the classic national conflicts of central and eastern
150 ■ Ethnicity, Migration, and Statehood
Europe—is a cross-border "kin" state or ethnic homeland involved m any significant way. As a result—and notwithstanding the political v olence associated with Irish, Basque, and Corsican nationalist movements—this type of ethnonationalist politics is less threatening to states than the characteristic eastern European configuration.
A further crossover, blurring the sharp outlines of the east-west 11 tinction, is that just as ethnicity is nationalized—understood as nationality—in some western European as well as in most east central European cases, so too nationality and nationhood may be ethnicized in western as well as in eastern Europe. And this is true not only for ethnoregional nationalisms. In response to growing Muslim and non-European immigrant populations, national self-understandings have also been ethnicized, to some degree, even in the so-called stagnations of northern and western Europe, in countries with traditionally state-framed understandings of nationhood.
Ethnicity in east central Europe, I have suggested, often takes a specifically national—and nationalist—form. Yet despite this potentially explosive configuration, and despite the resurgence of nationalism that accompanied the collapse of communist regimes, ethnic violence has been less widespread, ethnic mobilization less strong, and ethnic identity less pervasively significant than is ordinarily assumed. Having made a good part of my professional living recently off ethnicity and nationalism in eastern Europe, I have no interest in minimizing their significance. In general, however, I think that discussions of the region are overly ethnicized and that an exaggerated fociis on ethnicity and nationalism risks crowding out other, often more important theoretical and practical perspectives.
Of the ghastly violence in Yugoslavia and parts of the former Soviet Union since the end of the Cold War we need no reminder. But as Tom Nairn (1995: 91-92) put it, even though one would certainly not want-to make light of these terrible conflicts, one should also beware of "making dark" of them. Ethnonationalist violence has been limited to a relatively small part of eastern Europe and the former Soviet Union—overwhelmingly concentrated in Yugoslavia, Transcaucasia, and the North Caucasus; elsewhere, what is striking is the absence of violence, and the relatively peaceful character of the disintegration of the Soviet Union. Consider, for example, the 25 million Russians stranded as minorities in nationalizing successor states by the breakup of the Soviet Union. Many analysts—myself included, in
Ethnicity, Migration, and Statehood • 151
the early 1990s—thought that at least some of these Russians might well be the flashpoints of ethnonational conflict and violence. Yet outside the self-proclaimed "Dniester Republic" in Moldova, successor state Russians have been neither the objects nor the perpetrators of nationalist violence (Laitin 1998: Chapter 12; Melvin 1998; Braun 2000).
What about ethnic and nationalist mobilization? Here too there is a case-selection bias at work. We pay attention to the spectacular moments of high mobilization—the human chain across the Baltic republics in 1989, the great crowds that filled the main squares of Yerevan, Tbilisi, Berlin, Prague, and other cities in 1988-1990. But these have been the exception, not the rule. Moments of high mobilization have been few and ephemeral. Even where "nation" was a galvanizing category at one moment, it was not at the next. On the whole, especially since 1990, people have remained in their homes, not taken to the streets. In conspicuous contrast to east central Europe in the interwar period, demobilization and political passivity, rather than fevered mobilization, have prevailed. Much has been written on the strength of nationalist movements in the former Soviet Union, not enough on their comparative weakness.5
There is, moreover, a kind of optical illusion involved in the view from afar. From a distance, one risks taking at face value the claims of ethnonational entrepreneurs, and forgetting that people do not necessarily respond particularly energetically or warmly to the nationalist utterances of politicians who claim to speak in their name.
In the Transylvanian town of Cluj, where I conducted field work in the second half of the 1990s, a bitterly nationalist local politics pits majority Romanian against minority Hungarian claims.6 Yet there has been virtually no nationalist mobilization by ordinary people, and most remain indifferent to the endless cycles of nationalist talk. This has made palpable for me the loose coupling, or lack of congruence, between nationalist politics—which seems to run in a sphere of its own, unmoored from its putative constituencies—and everyday life. And there are many parallels elsewhere in the region. The general political passivity of Russians in Soviet successor states, for example, has been striking, despite various attempts to mobilize them.
Forty years ago, sociologist Dennis Wrong (1961) criticized Parson ian functionalism for its "oversocialized conception of man." Much social analysis today is informed by what might be called an
152 • Ethnicity, Migration, and Statehood
overethnicized conception of history, politics, and social interaction; The ethnic categories deployed by political and cultural entrepreneurs are often uncritically adopted by social analysts. As a result, the salience of ethnicity tends to be assumed rather than demonstrated; ethnic identities are ascribed to persons who may define themselves in other terms. Ethnicity and nationalism need to be understood as particular ways of talking about and experiencing the social world and as particular ways of framing political claims, not as real boundaries inscribed in the nature of things.7 At some places and times, these ways of talking about the social world and of making political claims have deep resonance and powerfully shape how people think and talk and act in everyday life, as well as how they understand and act on their political interests. At other times and places, the language of ethnicity and nationalism deployed by political entrepreneurs falls on deaf or simply indifferent ears.
Migration
Like ethnicity—and in part, of course, in connection with ethnicity— migration too has become a central issue throughout Europe. But just as patterns and perceptions of ethnicity differ, so too do patterns and perceptions of migration. First, and most obviously, the problematics of migration in western Europe have focused on immigration, especially from outside the region,8 seen both as a problem (in political terms) and as a solution (in economic and, increasingly, demographic perspectives). In eastern Europe, questions of migration have been, in the first instance, about emigration—seen again both as a problem and as a solution. Emigration is seen as a problem insofar as it involves the disproportionate outmigration of highly educated or skilled younger people, or even a declining overall population (the population of Romania declined by nearly 5 percent in the 1990s, in considerable part because of emigration). But emigration is also seen as a solution: by ordinary citizens, insofar as getting out offers a means of getting by or getting ahead; by the state, insofar as it generates remittances; and by some nationalists, insofar as it removes or weakens "unwanted elements."
As a corollary of this basic difference, migration has been experi-entially marginal in western Europe. After long years of invisibility, migrants—and their distinctive cultural practices—have become
Ethnicity, Migration, and Statehood * 153
conspicuously visible and central to everyday experience in many western European cities and towns. But migration itself—even in former countries of emigration such as Greece, Italy, Portugal, and Spain—is something that others do. In eastern Europe, by contrast, migration has become experientially central, figuring pervasively in the way ordinary people think and talk about their plans, strategies, and aspirations.
Within the European Union (EU)—and within the Schengen zone in particular—migration has of course become more free through the creation and enlargement of a space of borderless free movement. In much of eastern Europe, migration has become less free, in certain respects, as political space has contracted; as borders, visas, and new citizenships have proliferated; and as the initially open door with which Western countries welcomed migrants fleeing collapsing communist regimes quickly closed. In other respects, to be sure, migration possibilities in eastern Europe have expanded. For several years now, citizens of most east central European countries have not required visas to travel to E.U. countries. This does not, of course, grant them the right to work, and even after the eastward enlargement of the E.U. in 2004, existing member states will be permitted to limit labor migration from new member states for a transitional period of up to seven years. But the ability to travel without the hurdles and indignities of having to seek a visa nonetheless marks a significant improvement for citizens of these countries (and also, of course, makes it easier to work without documents).
In western Europe—to highlight a final stark dimension of difference—migration involves mixing, and generates new forms and degrees of ethnic, racial, linguistic, and religious heterogeneity, together with the new challenges to national self-understandings and new forms of politicized ethnicity sketched above. In eastern Europe, much migration—not only in the last fifteen years, but over the last century—has involved unmixing, reducing rather than increasing heterogeneity (Brubaker 1995). This is notoriously the case, of course, for the infamous instances of forced migration—starting with the Balkan Wars at the beginning of the twentieth century, via the massive displacements during and after the Second World War, to the "Balkan Wars" at the century's close—that have come to be known as "ethnic cleansing" (Naimark 2001; Mann 2004). But it is also the case for quieter, less dramatic forms of ethnic unmixing, involving, for example,
154 • Ethnicity, Migration, and Statehood
the migration of Germans from Poland, Russia, and the former Soviet Union to Germany; of Hungarians from Romania, Yugoslavia, Ukraine, and Slovakia to Hungary; of Russians from various Soviet successor states to Russia; and of Jews from the former Soviet Union to Israel (Brubaker 1998a; Joppke 2004).9
Of course, patterns of migration are a great deal more complicated than this. "Western Europe" and "Eastern Europe" are not single places but differentiated series of places, differently positioned—for economic, political, and geographic reasons—with respect to migration flows. Consider just one example. In the more prosperous east central European countries—especially Poland, the Czech Republic, Hungary, and Slovenia10—emigration pressures are weaker, while labor migration from points further east, and requests for political asylum from Asian and African as well as eastern European countries, have emerged as significant issues. In this respect, these countries seem to be following in the path of Spain, Portugal, Italy, and Greece, which made the transition from emigration to immigration countries during the last quarter century.
A decade and a half after the end of the Cold War, it is worth keeping in mind the migration that has not occurred from—and within— eastern Europe. In 1990, experts warned of an "exodus," a "human deluge,"11 an "invasion" of "hungry hordes," a "mass migration on a scale unseen since World War II,"12 a "flood of desperate people," amounting to a modern-day Völkerwanderung akin, in the words of Peter Jankowitsch, chair of the Foreign Relations Committee of the Austrian parliament, to that in which "the Germanic peopiefs] moved west and destroyed the Roman Empire." "How many Poles will stay in Poland?" Jankowitsch asked rhetorically. "How many Romanians will stay in Romania?"13 Plenty, it turned out. Sizeable though westward migration has been in the experience and—even more so—in the social imagination of ordinary citizens of eastern Europe, its magnitude, for western countries, has remained modest. In the "frontline" states of Germany and Austria, such migration has been much more significant, but even there its rhythms have been measured, not cataclysmic.
Around the same time, haunted by the Yugoslav refugee crisis, analysts envisioned convulsive episodes of forced or politically induced migration, pointing with special concern, in this context too, to the 25 million Russians outside Russia. Yet while many Russians have left Central Asia and Kazakhstan, the migration has been comparatively
Ethnicity, Migration, and Statehood • 155
orderly, and the large majority of Kazakhstani Russians have chosen so far to remain in Kazakhstan.
Yet while vast east-west migrations have not occurred, this is not for the reasons suggested by the widespread imagery of Fortress Europe. Migration policy has indeed been subordinated to security concerns in certain ways, and this trend is likely to intensify after the Madrid bombings. Yet as Favell and Hansen (2002) have argued, the imagery of Fortress Europe overstates the restrictiveness of European migration policies. Even in the domain of asylum, where policies are indeed very restrictive, only a small minority of those whose claims to asylum have been rejected are ever deported. And the Fortress Europe model neglects the market-driven dynamics that have generated new labor migration flows even in ostensibly zero-immigration countries like Germany and Britain. It also neglects the fact that the eastward enlargement of the EU, like the earlier deepening of European integration through the creation of the Schengen zone, involves both inclusion and exclusion, both liberalizing and restrictive aspects. What, then, does EU enlargement mean for migration patterns? It : already has entailed the development of stricter controls along the ex-
ternal frontiers of the enlarged union (and along those anticipated ■ after the accession of Romania and Bulgaria in 2007); this has dis-
\ rupted cross-border flows in the border regions (ibid). Will the east-
* ward enlargement generate a substantial increase in east—west
migration within the EU? Not necessarily. Restrictions on labor migration can remain in place for up to seven years. And even after free movement is introduced, there may be no huge increase in east-west migration. To be sure, many citizens of central European countries, I Poland especially, have already been working in EU member states, i legally and illegally, and their numbers may well increase in response
1 to economic and demographic pressures. But just as the southern en-| largement of what was then the European Community to include I Greece, Spain, and Portugal in the 1980s did not lead to dramatically J new migration from those countries, so the eastern enlargement may j        not generate dramatically new east-west migration.
1 Statehood
\
\ My final cluster of themes concerns the state. The restructuring of the
J state has been a central issue throughout Europe. But in this domain,
156 • Ethnicity, Migration, and Statehood
too, questions have been posed in very different ways in different parts of Europe.
The most striking difference would seem to be this: while the reorganization of political space in western Europe has pointed—at least in anticipation—beyond the nation-state, the spectacular post-Cold War reconfiguration of central and eastern Europe has involved a move back to the nation-state. Apart from unified Germany, nineteen of the twenty-two successor states to the multinational Soviet Union and Yugoslavia and binational Czechoslovakia are generally understood as nation-states, that is, as the states of and for the particular nations whose names they bear (and the three exceptions—the Russian Federation, rump Yugoslavia, and Bosnia-Herzegovina—are themselves closely linked to particular nations). If western Europe is entering a postnational age, the political context for much of eastern Europe might be more aptly described as post-multinational. Just as the great Habsburg, Romanov, and Ottoman empires crumbled at the beginning of the "short twentieth century," leaving an array of nationally defined successor states in their stead, so too, at the close of the century, multinational states have again fragmented into sets of soi-disant nation-states.
Yet this view requires qualification, and not only because the massive eastward enlargement of the EU in May 2004 blurs the east-west distinction. More fundamentally, the EU does not represent a linear or unambiguous move "beyond the nation-state" to a supranational form of political authority. As Milward (1992) argues, the initially limited moves towards supranational authority worked—and were intended—to restore and strengthen the authority of the nation-state. What has been occurring is a complex unbundling and redistribution— upwards, downwards, and in various oblique directions—of previously tightly bundled powers and competencies. The resultant "multilevel" or even "neomedieval" polity does not look much like a supranational superstate: an oft-quoted remark describes the EU as an "economic giant, a political dwarf, and a military worm."14 Events of the last decade, notwithstanding the Treaty of Maastricht and the announced formation of a common security and defense policy, have done little to undermine that view.15
Yet while there has been no clear move beyond the nation-state, the classical model of unitary, centralized, sovereign statehood, in which all authority derives from a single central point, no longer describes
Ethnicity, Migration, and Statehood • 157
political reality. Authority has been reconfigured, and competencies unbundled and redistributed, not only to the EU (itself a set of institutions and authorities, not a single entity) but also to other international organizations, and to subnational polities and jurisdictions. This raises fundamental questions about the changing nature of statehood and political authority.
Granted that the EU is not very statelike at present, how might it become more statelike in the future? What attributes historically associ ated with statehood might it come to acquire? What does its development imply about the statehood—or, following J. P. Nettl (1968), the "stateness"—of existing states? Are they becoming less statelike as they give up conventional sovereign powers, such as control over borders and over monetary and fiscal policy?
Once we revise our understanding of statehood to allow for the unbundling and sharing of powers and competencies previously monopolized by a single sovereign center, then questions of stateness also arise for lower-level polities emerging within federalizing or otherwise decentralizing states. To what extent do more or less autonomous but nonsovereign polities such as Catalonia, Flanders, and Scotland take on attributes of stateness as they gain new and often quite considerable powers and competencies,16 even while remaining parts of larger, more embracing states? This is a familiar issue in the literature on federalism, but that literature has been quite separate from the historical and political sociological literature on the development of the modern state. The latter has defined the modern state as centralized and sovereign—as monopolizing the means of coercion within a particular territory, in Weber's classic formulation—and has cast the story of its development in teleological form, involving the progressive appropriation of previously dispersed powers by a single center. This perspective has marginalized the experience of federal states. Their very existence is something of an anomaly; they are by definition not very stateiike.
The complex unbundling and redistribution of powers and competencies, in short, are forcing a fundamental rethinking of the very notion of "the state." The notion may prove too heavily encumbered by the political theory of sovereignty and its monist, unitarist connotations to be of much analytical use in conceptualizing the complex multilevel polity that is emerging.
In eastern Europe, questions of statehood and stateness are posed in
158 • Ethnicity, Migration, and Statehood
quite different terms. There is, in the first place, the sheer proliferation of new states. Almost all of them, as noted above, have defined and constituted themselves as sovereign nation-states, drawing on highly institutionalized—if outdated—rhetorics and models of sovereignty and nationhood (Meyer 1987). These institutionalized "performances" of sovereign nation-statehood do not represent an unambiguous move "back to the nation-state." Almost all the new states are involved, in one way or another, in processes of regional integration, notably as members or candidate members of the North Atlantic Treaty Organization (NATO) and/or the EU on the one hand and the Commonwealth of Independent States on the other. Yet the invocations of sovereignty and nationhood are not mere rhetoric. There is a real tension between the model of sovereign nation-statehood and that of supranational integration; the latter does not automatically trump the former. The model of sovereign nation-statehood remains normatively more robust in eastern than in western Europe and has its attractions not only for newly constituted states but also for those newly freed from the Soviet economic and security embrace.
Second, there are the special "stateness" problems—in the Linz and Stepan's sense, not Nettl's—posed by politicized ethnicity in eastern Europe. As I indicated above, the ethnically framed challenges—or perceived challenges17—to the territorial integrity and boundaries of existing states are particularly delicate in eastern Europe because they often involve cross-border links connecting ethnonational claimants within particular states and a patron state abroad that represents the same ethnocultural nationality.
Third, and most important, although the initially prevailing understanding of postcommunist "transition" posited the need to liberate economy and society from the grip of an overly strong state, more recent analyses have made almost the opposite argument.18 The post-Cold War moment of triumphant anti-statism has passed. As Stephen Holmes and others have argued with respect to Russia—although the point has broader relevance for the region—it is not the strength of the state, but its weakness, that threatens the basic rights and well-being of citizens.19 The "withering away of the state" that occurred in Russia and elsewhere in the 1990s destroyed the capacity to provide the most elementary public goods and services. Neoliberals increasingly concede what paleoliberals knew all along: a strong, even powerful state is a precondition for everything that they hold
Ethnicity, Migration, and Statehood '159
dear, including the orderly workings of markets, the protection of citizens against violence, and the enforcement of human rights. Hence the calls to strengthen and build up the state, to liberate what are in theory the distinctively public powers of the state from the clutches of those who have expropriated and in effect privatized them.
The force of renewed calls for a "strong" or "powerful" state depends of course on how we understand these terms. Here Michael Mann's (1993: 59-60) distinction between "despotic" and "infra-structural" power is helpful, the former denoting arbitrary power over civil society, the latter the power of state institutions to coordinate and regulate social life by penetrating and working through civil society. Despotically "strong" states may be infrastructurally "weak," and vice versa. What is urgently needed in much of eastern Europe—and throughout the Third World—is an infrastructurally strong state, one that can keep the peace, punish force and fraud, enforce contracts, collect taxes, provide basic services, protect public health, implement legislation, and prevent wholesale plundering by criminal and quasi-criminal networks.
State-building, then, is still very much on the agenda in eastern Europe. While western and parts of east central Europe move towards the unbundling and redistribution of previously concentrated powers, in much of eastern Europe we see (or at least hear about the need for) moves in the opposite direction, toward the rebundling and reconcen-tration of previously dispersed—and in considerable part privately appropriated—powers.20 Whether such changes will succeed—whether an effective, infrastructurally strong state can be built—is by no means certain. Over the long sweep of European history in the last millennium, sustained military competition eventually led to the weeding out of the most blatant forms of patrimonial administration.21 Today, however, pressures to reform conspicuously corrupt, grossly inefficient state administrations are much weaker. States (and non-state actors) continue to make war, but war no longer makes states the way it used to.22 The worldwide club of states includes a large and perhaps increasing number of "quasi-states" (Jackson 1990)—organizations that are officially recognized and certified internationally as "states" yet fail to do the most elementary things that states are supposed to do, such as maintaining order throughout a given territory. Today, thanks to the reification and sacralization of existing state borders in prevailing international discourse and practice,23 such quasi-states can continue to
160 ■ Ethnicity, Migration, and Statehood
exist, irrespective of their abysmal performance, with little threat that they will go out of business. Eastern Europe may not harbor the worst specimens of this lamentable genre, and of course there are great differences within the region. In much of the region, however, the making of the modern state, far from being a completed chapter of history, re^ mains a matter of great contemporary urgency.
v. t
ov  CHAPTER EIGHT
1848 in 1998: The Politics
of Commemoration in Hungary,
Romania, and Slovakia
The year 1.998 marked the 150th anniversary of the cascading wave of revolution that swept across Europe in the spring of 1848. Like all great upheavals (indeed like all great events, personalities, or works
of art), the revolutions of 1848 do not contain their own meaning. Powerful cultural objects—whether events, persons, or cultural cre-J--1        ations—are always ambiguous: indeed that ambiguity, according to \        Griswold (1987a), is a key part of what constitutes their power. Such objects always offer rich and varied, though not unlimited, interpretive possibilities. It is now widely agreed that the meanings of such i        cultural objects are not fixed, given, or uniquely ascertainable, but in--1        stead are created and recreated in different times, places, and settings 'jj        through a series of "interactions" or "negotiations" between the ob-i        jects and their socially situated, culturally equipped, and often politi-]        cally engaged interpreters (Hall 1980; Griswold 1987a, 1987b;
liebesandKatz 1996). j In the last fifteen years, commemorations—and social memory gener-
i ally—have emerged as a fruitful site for studying this interactive pro-:\ duction of meaning. That the past is constructed and reconstructed to | suit the needs and purposes of each succeeding generation; that even I personal memory is a thoroughly social and cultural construct; that I      collective or social memory is not only constructed but chronically
This chapter was coauthored with Margit Feischmidt.