The term "macrosociology" brings immediately to mind the idea of society, that social apparatus that has long been an ultimate point of reference in the organization of social life. Or so it seems. I begin this chapter with the observation that the idea of society is itself problematic—and is becoming more so all the time. I will close on the same note, and this will lead us naturally to the topic of the final chapter, the suprasocietal or global level.

The Central Place of the National Society in the Social Sciences

Virtually all of the social sciences, themselves children of the nineteenth- and twentieth-century domination of the nation-state, have, in one way or another, taken a version of that entity as the framing context for their respective intellectual enterprises. Consider the following:

•     For political science, the nation, the state, and the national government and its institutions have constituted the fundamental basis of study.

•     For economics, the basic macroeconomic unit has been the national economy. Writing only two decades ago, Simon Kuznets(1972: 1-2) stated that nation-states set the "institutional boundaries within which markets operate and within which human resources are relatively free to handle material capital assets and claims to them." Most analyses of international trade have dealt with interaction among national units.

•     For sociology, the corresponding unit has been the national society, the seat of social integration and social institutions. The notion of society, moreover, involves a confluence of self-sufficiency, political integrity, social solidarity, and cultural identity.

•     For cultural and social anthropology, the prime unit has been the "culture," stressing commonality of values, language, beliefs, and sense of identity but not necessarily having the attributes of a nation. This circumstance probably derives from the fact that many of the units studied by anthropologists have not been nations but rather tribal and other subnational groups. However, the concept of culture has proved easily translatable into the idea of a "national culture," as in references to German, Japanese, or American culture.

In this chapter and the next l will wonder about the continuing viability of this focus. But for the moment let us review some of the characteristics assigned to that favored unit.

The modern national society, or state, was consolidated in the intellectual and ideological work of writers like Thomas Paine and in the political and social work of the French Revolution. The composite view of the national state that emerged from that work was an identifiable social apparatus that fused a remarkable number of features of organized social life: geographic boundedness, political sovereignty, monopoly of force and violence by military and police forces, economic self-sufficiency, cultural integration or solidarity, a common language, anti the political identity of a citizenry.

To give several examples of this emphasis: Society as the basic organizing unit found expression in Durkheim's first work, which dealt with the division of labor in society ([1893] 1984). His primary conceptual unit was the society. That body, consistently regarded as a kind of organism, possesses an organic integrity. Intersocietal relations were not especially problematical for him, nor were subsocieties. In Durkheim's analysis of differentiation, he consistently spoke of segmentary and complex societies. Also, his treatments of the division of economic labor, the differentiation of political, administrative, and judicial functions (ibid., 1-2), and the differentiation of social institutions such as the family (ibid., xlv) were significant mainly at the societal level. Integration, too, is a societal phenomenon; mechanical solidarity is an attribute of undifferentiated societies, and organic solidarity an attribute of complex societies. With respect to the latter, Durkheim recognized the significance of subsocietal bases of solidarity but regarded them as on the wane. Thus:

[In peasant societies], . . . since economic activity has no repercussions outside the home, the family suffices to regulate it. . . . But this is no longer so when trades develop. . . . [I]f domestic society is no longer to play this [regulatory, integrative] role, another social organ must indeed replace it in order to exercise this most necessary function. (Ibid., xiv-vi)

If there is one truth that history has incontrovertibly settled, it is

that religion extends over an ever-diminishing area of social life. (Ibid., 119)

Gradually [local customs] merge into one another and unify, at the same time as dialects and patois dissolve into a single national language and regional administration loses its autonomy. (Ibid., 136)

The replacement for these declining functions was found in an assertion of the society itself; "the more we evolve, the more societies develop profound feeling of themselves and their unity" (ibid., 123). For Durkheim, the society frames all that is social.

This comprehensive view of the national society survived in the functionalist and other traditions of modern sociology. In his work on comparative sociology, Robert Marsh (1967: 10) defined a society as having the following characteristics: "(1) a definite territory; (2) recruitment in large part by sexual reproduction; (3) a comprehensive culture; that is, cultural patterns sufficiently diversified to enable the members of the society to fulfill all the requirements of social life; (4) 'political' independence; that is, a society is not a subsystem of any other system, except in a very partial sense." About the same time Parsons (1966: 9) defined a society as "a type of social system . . . which attains the highest level of self-sufficiency as a system in relation to its environments." These "environments" included the definition of ultimate reality, cultural systems, personality, behavioral organism, and the physical-organic environment—in relation to which the society was a self-sufficient, integrating, and coordinating agency.

This "strong" and "closed" notion of the national society was a product not only of the intellectual efforts of social theorists and social scientists. It also emerged from the more or less organized projects of modern national societies themselves, which, in their recent histories, have pursued policies of securing the monopoly of force and violence in the national state; cultural integration through schooling, language policies, and the media; and loyalty and identification by cultivating and appealing to nationalistic sentiments. In a word, national societies themselves have worked toward that fusion, or unity of national economy, polity, society, and culture—to make the "imagined communities" (Anderson 1983) of modern national societies into real communities.

In this chapter and the next I will take a double line of attack. On the one hand, I will recognize the continuing validity of the national society by discussing some of its own problematics; on the other, I will argue that the notion of the national society is coming into question, both as an empirical entity and as a core organizing construct in the social sciences.

Ongoing Problematics of Contemporary Societies

It is my impression that the concept of social structure—as well as the allied concepts of institution and role—has experienced a loss of status in sociological thinking in the past several decades (for a similar observation, see Eisenstadt 1995: 19-20)· If this impression is correct, three intellectual developments might be cited as partially responsible for the decline. The first was the assault on structural-functional analysis—to which both institution and role were central—in the 1960s and 1970s. The assault came mainly through the "microsociological revolution" of the period (which tended to treat those constructs as illegitimate reifications) and through the neo-Marxist and neo-Weberian ascendancy of the same period (which, however, retained social structure as a central organizing construct). The second was the subsequent assault on the Marxian perspective, stemming from both intellectual and political dissatisfactions with it. The third development was a revision of the notion of culture, earlier regarded mainly in its legitimizing role with respect to social structure but now increasingly conceived in its psychological significance ("identity"), its significance as "project" or "strategy," and its significance as an instrument of domination. In this chapter I try to right the balance and argue that a number of long-standing concerns with social structure do and should retain their traditional importance.

The Continuing Salience of Structural Differentiation

The notion of structural differentiation is a major thread of analysis in economics and sociology. That conception arises from the acknowledgment that the structured allocation of activities in society is variable and that a pivotal line of variation is the degree to which these activities are specialized, or differentiated from one another. Adam Smith ([1776] 1937) matte the division of labor—the economic version of structural differentiation—central to his analysis of the causes of increased economic productivity and the resulting wealth of nations, as well as the organizing concept for his theory of international trade. Karl Marx ([1867] 1949), too, recognized that an increased division of labor is a fundamental force in competitive capitalism. The idea of differentiation lay at the heart of Herbert Spencer's (1897) theory of evolution, and although that special theory was rejected by Durkheim ([1893] 1984), the idea of structural differentiation (the social division of labor) remained as the key structural element in the evolution from segmental to complex societies. Differentiation is central to the sociology of Georg Simmel as well, and formed the cornerstone of his concern with the development of modern society. However, Simmel stressed not only the economic and social efficiency of differentiation but also its capacity to create individualism and individual freedom (see Dahme 1990). Simmel's insight found expression in Parsons's (1966) subsequent observation that structural differentiation is the main lever for freeing individuals from their traditional ascriptive ties.

Structural differentiation also lies at the center of Parsons's (1961, 1966) general theory of social change, is a central theme in one strand of modernization theory (e.g., Smelser 1964), is a recurring theme in my own work (Smelser 1959, 1991), finds a significant place in the theoretical work of Luhmann (1982), and survives in "neofunctionalist" theory (Alexander and Colomy 1990). In most of these manifestations, differentiation appears as a description of and mechanism for the transition from traditional to modern social structure and, in that connection, carries an explicitly or implicitly adaptive—even evolutionary—connotation of the increasing rationalization anti efficiency of social life.

Even though we have presumably moved from the "modern" to a "postmodern" phase of civilization, differentiation remains a commanding feature of a contemporary society. The continuing proliferation of specialized occupations (especially in the service sector) and the continuing march of bureaucratic organizations give witness to the process, as does the dramatic increase in the international specialization of production. Nor is the phenomenon restricted to economic and administrative activities. The rise of the modern nuclear family involved a differentiation of economic activity away from the family by relocating work in factories and other formal organizations, leaving the family a more focused unit, "specializing" in socialization and intimacy. Also, the eclipse of arranged marriage and the rise of romantic love as the basis for marriage marked a differentiation of courtship both from kinship and from the transmission of property and status. In the contemporary world we witness a radical extension of that process. The increase in numbers and legitimacy of the nonconjugal household, the single-parent household, homosexual cohabitation, and communal living signifies, among other things, a differentiation and dispersion of the nuclear family's previous monopoly on intimacy to other kinds of relationships. Similarly, the establishment of nursery school, preschool, day care, play group, and other collective arrangements is a differentiation of socialization in the early years, with the family's previous near-monopoly once again dispersed. To choose a final example, one of the political aims of feminism has been to differentiate gender identification from occupational and status placement.

As indicated, the idea of structural differentiation has had an affinity with theories of progress and social efficiency, though that emphasis has weakened recently. While that dimension continues to be relevant, it constitutes only one aspect of the process. The following additional problematics are associated with the idea of differentiation.

•     More attention should be directed toward the inefficiencies and other costs associated with increased differentiation. Two traditions of research have this emphasis—first, the literature on diminished psychological gratification, increased alienation, and anomie associated with specialized roles; and second, the literature on inefficiencies (such as indecision, red tape, subversion of goals) associated with bureaucratization. Still other lines of inquiry are in order. To mention only one, while it is apparent that collective arrangements for the socialization of very young children are "efficient" in that they free parents for participation in the market and other activities, less is known about the psychic benefits and costs of placing so much socialization in the hands of professionals and other nonfamily agents.

•     The presumed causes of differentiation should be expanded beyond the more or less rational, often post facto assumption that social structures differentiate in order to augment social efficiency. I have in mind, as an example, the notion of structural differentiation as response to political conflict. Years ago Michel Crozier (1964) interpreted the proliferation of bureaucratic rules as an accumulation of responses to conflict situations so that similar conflicts either would not recur or, if they did, could be "handled" by the new machinery. Similarly, the proliferation of regulative and watchdog agencies to guard against conflicts of interest constitutes a differentiation of structural forms to deal with political and ethical problems and conflict. One may call this "efficiency" if one wishes, but that stretches the term and does not pinpoint the political process involved.

•     Because of the explicit or implicit linkage of differentiation with efficiency or progress, models of the process (e.g., Smelser 1959) tended to focus on successful differentiation, that is, sequences that actually produced more differentiated structures and more complex arrangements. Empirically, however, that process is not smooth, largely because differentiation involves the modification or even eradication of existing arrangements and often displaces incumbents of existing roles (as in technological unemployment). This means that efforts to change encounter corresponding counterpressures, usually in the form of vested interests. A frequent result is "blocked differentiation"—a kind of social paralysis as pressures to change build but yield chronic group conflict rather than structural change. In my study of the rise of state-supported education for the working classes in nineteenth-century Britain (Smelser 1991), I found it less profitable to regard the process as one of orderly differentiation than to treat it as a prolonged paralysis, with evident pressures to establish schooling for the working-class young (mainly concerns with pauperism and social order) being stalled for long periods by unresolvable conflicts among religious groups interested in promoting their kind of education.

•     In the first instance, differentiation produces changes in the social structure. Yet the results of differentiation also shape the structuring of groups, group interests, and group conflicts and in that way spill over directly into the political process. The story often unfolds in the following way. The differentiation of a new structure creates positions (or roles) that are occupied by new incumbents. Industrial development, for example, produces manual workers with various levels of skill, supervisors, engineers, sales personnel, and the like. An advanced medical system produces doctors, nurses, technicians, hospital administrators, and more. Incumbency in these roles, moreover, becomes the basis for common interests of incumbents, and for the formation of groups (mainly unions and associations) that may assume significance as conflict groups. Putting these ingredients together, we produce the following abstract model of process.

differentiation ® categorization ® social group ® consciousness of group ® political mobilization ® social change

     This kind of model informed Marx's ideas linking the economic and political processes. As the result of capitalist development (differentiation) a class of propertyless wage earners (category) is created; then, through mutual contact and communication, this category becomes a group with definite consciousness of its situation and on the basis of this consciousness becomes a politically active group that ultimately overthrows the system.

Marx regarded these transitions as more or less inevitable within capitalist development. But as subsequent history has demonstrated, the transitions from social categories to groups with consciousness to political action groups are problematical rather than inevitable. Some social categories (roles) become the basis for groups and others do not; moreover, category-based groups that do not have consciousness at one moment gain it at another—particularly when they arc threatened in some way. Furthermore, processes of differentiation can work to divide groups as well as unite them. More than one observer (Mills 1951; Dahrendorf 1959) has pointed out that the proliferation of multiple work roles, and especially service (white- and pinkcollar) roles has worked to subvert Marx's prediction that a propertyless proletariat as a whole would develop common class consciousness and become a directed conflict group.

One key agenda item for sociologists, then, is to link social structure (i.e., the kaleidoscope created by processes of structural differentiation) and group life in society by generating models and conducting empirical investigations that focus on the determinants of the contingent transitions among social structure, social categories, social groups, group consciousness, and group action.

The Increasing Salience of Diversity

The idea of differentiation concerns above all roles and institutions that have functional significance. It tells us a great deal, moreover, about groups and group conflicts precipitated from the panoply of structured roles. Crosscutting these functional roles, however, is another range of social categorizations, both ascribed anti self-assigned, that also constitute bases for assignment to functional roles, personal and group identification, prejudice anti discrimination, and the political process. Among ascribed categories are race, ethnic membership, native language, region or locality, age, gender, and religion (the latter a mixed category, because religion often involves a mixture of ascription and personal choice). Among nonascribed bases are membership in social movements, some based on the ascriptions mentioned—as in the case of feminism and regional political groups—but some issue based, as in the case of the peace movement, the environmental movement, the animal rights movement, and other groupings based on cultural choice, such as lifestyle and counterculture. Sometimes these categories overlap with functional structures—when women are assigned to certain occupations or to greater responsibility for child care in the family, or when occupations are segregated by race (slavery is the extreme case). Despite this overlap, a distinction can be made between functionally differentiated roles and these other social categories. The former describe the differentiation of society, the latter its diversity. Even this distinction is not a clean one, because part of the cultural diversity of modern societies arises from distinctive cultural groupings derived from functionally based groups (e.g., working-class culture, peasant culture, and yuppie lifestyle).

The social bases of diversity are historically variable. Long periods of Western history have been marked by the salience of religious diversity, though this has declined since the rise of industrialism and nationalism. (We cannot forget, however, the residual religious basis of some European political parties, and the continuing and extreme salience of religion in such areas as Northern Ireland, Lebanon, Syria, the former Yugoslavia and other Balkan areas, as well as fundamentalist movements everywhere.) Race as a biological category rose in salience during previous episodes of internationalization, especially enforced slavery and colonization. Before the onset of industrial-market and national bases of organization, locality and local culture served as the primary basis of social interaction and identification. This basis has withstood the institutional and sometimes-conscious political efforts of the market and nation-state to displace it, and in the past decades localism—expressed in terms of demands for autonomy, integrity, and recognition—has reasserted itself. In fact, the social bases of race ethnicity, language, gender, sexual preference, and to some extent age have come forward as salient bases of group identification and politics—so as to give life to labels such as "cultural diversification," "multiculturalism," "the new tribalism," and "identity politics." These developments have been as dramatic as they were difficult to predict. Whether or how long they will persist is uncertain and equally difficult to predict.

We do not understand the reasons for the resurgence of these kinds of diversity, but any ultimate explanation will have to take account of at least the following factors.

•     Certain categories have become more salient largely by virtue of realistic demographic and economic changes. For example, long-term demographic trends—mainly reduced fertility and mortality—have led to dramatic increases in the numbers (and therefore political significance) of the elderly in developed societies. The institutionalization of retirement has also given clearer visibility anti commonality of experience as a category removed from the active labor force. Furthermore, the mobilization of the elderly on their own behalf has raised the political consciousness of other age groups, especially in relation to taxation and welfare issues. In addition, the mobilization and political significance of the feminist movement cannot begin to be understood apart from the dramatically increased—but in many respects still disadvantaged—participation of women in the paid labor force since World War II, which created new interests and new consciousness among women.

•     In many respects cultural diversification has resulted from an actual diversification of populations in many nations through international, interregional, and intranational movement of peoples. This, in turn, has resulted from changes in demand for labor (e.g., guest workers), from wars and other political crises that have produced migrant populations, and from increased tourism. There seems to be no reason to believe that such movements will not increase.

•     The political dynamics of localities—both urban and nonurban—generate polarization between "newcomers" and "natives." These dynamics are overdetermined by several subprocesses—the tendencies for newcomers simultaneously to compete economically and to self-segregate culturally, both of which add to their visibility and their threat; the tendencies for natives to react defensively to preserve economic positions, political power, and ways of life; and the interaction of these two tendencies to produce cultural and political polarizations.

•     The most common verdict on the role of the media, especially television, is that they are culturally homogenizing, and their spread through the whole world is cited in support of this. The effects are, however, evidently more complex. Television brings cosmopolitan reality to localities, thus "diversifying" them, at least during that long and never-completed transition to cosmopolitanism. Similarly, the international presence of the media— to be discussed more in the final chapter—diversifies, and never completely conquers the developing world. Moreover, the media, particularly in the United States, tend to "tame" diversity by including it explicitly in programming and advertising, thus elevating issues such as race, gender, and sexual preference to greater salience for the general viewing public and imparting greater "diversity" of exposure to their audiences.

•     Presently I will lay out a number of reasons why it is difficult for polities—especially democratic ones—to deal with political groups that present their demands in cultural terms. In fact, there is evidence that polities often conspire in the unsuccessful attempt to downplay the political salience of categories such as race, ethnicity, gender, and sexual preference. However, the realities of politics sooner or later force them to recognize these groups as political entities in their own right, and when they do, they tend to heighten the political significance of those categories. The United States is a telling example. Largely as a result of the civil rights and feminist revolutions stretching from the late 1950s into the 1970s, federal and state governments came explicitly and officially to acknowledge the political presence of these groups, largely in the form of programs under the heading of "affirmative action." In doing so they made visible race and gender as political categories with a certain presumption to political entitlement, even as they denied that employment practices constituted favoritism and rejected the idea of quotas based on race and gender. And in doing that, they have conveyed the message that entitlement-like demands on the part of ethnic, sexual preference, physically disabled, and other groups were fair game in politics and, over time, have been greeted with similar political demands on the part of Native Americans, Asian Americans, Latino Americans, gay groups, and, not least, "white ethnics" and, to a lesser degree, white males in general. The recent efforts on the part of aspiring Republican politicians and others to diminish or abolish affirmative action can be understood as a response both to this backlash and to the difficulties created by ascriptive politics.

•     The combined effect of the internationalization of the economy (with a corresponding loss of control of nation-states over their economic fortunes) and the development of regional political alliances (such as the European Union and, to a lesser degree, North America) has no doubt given advantage to subnational regional, ethnic, and language movements in their programs to lay claim to political loyalty. And, again paradoxically, as these very movements gain momentum and legitimacy, they become active forces in the weakening of the nation-state as an object of loyalty and a focus of cultural identity.

•     It has been suggested that the tendencies to localization, including the dissolution of former empires and states, are, in fact, a protest against the growing scope of world markets and global politics, perhaps even some kind of reassertion of the limits on human bonding, which cannot extend itself indefinitely in scope, superficiality, anti diversity. This argument, while worth considering, is very difficult to demonstrate, and is perhaps beyond proof. However, it does make sense to interpret the reassertion of localism anti local autonomy as an effort on the part of human groups to gain control in a world that appears to be becoming increasingly uncontrollable.

These diverse factors constitute a helter-skelter list of plausibles; they do not provide anything like a full explanation. What is clear, however, is that the combination of accelerating differentiation and increasing diversification in contemporary societies has also changed the fundamental terms of two additional sociological dimensions: stratification and integration. To these phenomena we now turn.

The Changing Face of Stratification

There was a time in the recent history of sociology when two perspectives of social stratification held dominant positions. The first, the functional, proposed that a combination of occupational status and level of education operated as the prime determinants of social ranking in society—this ranking traceable, in turn, to the cultural values of industrial society. One subtradition of research, noting similarities of prestige rankings in most societies studied, held this kind of ranking to transcend both political systems and traditional cultural values (Treiman 1977). The second, the Marxist, tied social stratification to property relations in the capitalist system; this approach focused less on ranking than on class anti class conflict. The approaches resembled one another, however, in that each inextricably linked inequality with the dictates of modern industrial society, though the approaches differed in the particulars of diagnosis, explanation, and political flavor. In the 1990s both systems retain some relevance to the realities of social organization, but both seem increasingly out of date, for reasons I will now explore. A corollary of both perspectives—functionalist explicitly and Marxist implicitly—was that the unit of the stratification system was the nuclear family household and that the main agent in that household was the male job- or occupation-holder. For functional analysts in particular, the social ranking of the kinship unit depended on a mix of the occupational role (primary), education, and income of the husband-father. The neatness of that view of the stratification system has become more and more muddled in recent decades, in large part because of the following kinds of changes.

•     The universal basis of economic, occupational activity for social status has come in for repeated questioning and criticism in various quarters: formulations by some economists that people prefer leisure over work; the repeated assertion that the United States has moved historically from a production-oriented to a consumption-oriented society; debates in Germany about the "uncoupling" of work and social status; glimmerings of such debates in Japan; and the apparent nostalgia of postsocialist societies for the "welfare and security" aspects of the socialist era, while at the same time renouncing its politically repressive aspects and desiring some sort of market-based economy with its promise of greater prosperity and higher levels of consumption. This is not the place to evaluate the validity of these assessments; but insofar as they tend to dethrone the relationship between work and social status, they raise questions about the criteria to be invoked in assessing the ranking systems of societies.

•     The bases for assigning social rank have evolved to a new point of complexity and uncertainty. Increasing differentiation and numbers of occupations and jobs has yielded a less definite basis of ranking, if for no other reason than sheer multiplication. The simplicity of distinctions between—and translations into class terms of—manual and nonmanual labor, bourgeoisie and working class, and others, has become clouded in the light of the multiplication of occupations, especially in the service sector. Insofar as proletarianization has proceeded, further, it has not been as a form of manual labor but as a service proletariat, including low-wage clerical workers, fast-food workers, paid security personnel, and "temporary" workers of many descriptions. Interestingly, too, a new form of "duality" has appeared in labor markets; technological changes, foreign competition, and migration have created an unemployed surplus of low-skill workers. These workers, along with those who are hired on a periodic or parttime basis so as to avoid benefit payments (now as much as one-quarter of the American labor force, and growing), constitute an important segment of the lower-income population. Finally, the continuing embourgoisement of skilled workers, and their political alliances with some managers and owners on many issues dealing with free trade and protection, has blurred that classic division between labor and capital as well.

•     Insofar as there was validity in the claim that women's status was determined primarily by the occupational and educational status of their husbands, that claim has now been weakened. The main challenge is the increased representation of women in the labor force and their partial entry into high-status managerial and professional positions that endow them with the social status connected with those positions—if they are married, sometimes independently of their husband's status, sometimes mingled with it. However, the status of women derived from occupation and education still presents ambiguities, partly because of traditional values and prejudices that do not cede full equity of evaluation for women and partly because of traditional assumptions—held by women as well as men—that women should combine an occupational career with childbearing and child-rearing responsibilities, which remain proportionately greater than the corresponding responsibilities of men. In a word, the long-term revolution in labor force participation by women has yielded a more complicated and less certain basis for social ranking.

•     The traditional household itself that is, husband and wife with children—has also been thrown into question by changes in the kinship structure. The major changes are high divorce rates, increases in single-person and single-parent households, increases in nonconjugal living arrangements, and increases in homosexual cohabitation. To assume that an ideal-type traditional household is the unit of stratification becomes, as a result, increasingly problematic.

•     One of the infrequently recognized consequences of access of larger proportions of the population to higher education—evident in most developed societies—has also rendered education less certain as a determinant of status. General experience in higher education no longer constitutes a "ticket" to a high-status occupational position or a "credential" for social status. This is not to deny Bourdieu's emphasis on education as a source of cultural capital; rather, it is to agree with him that the opening of a previously elite avenue to status has become less valuable and less certain as a provider of that capital.

•     Insofar as ascribed anti quasi-ascribed bases of social organization assume greater salience, the more they are likely to be invoked in determining status and the more they cloud judgments about ranking and stratification. Put another way, diversification has become superimposed on differentiation as a basis for status, making both ranking and status identification more complex.

Most interpreters of the decline of class as an agency in the post-modern world—including both end-of-ideology theorists such as Daniel Bell and critical theorists such as Jürgen Habermas—have cited several factors: the increasing prosperity of the working classes, the politically calming effect of the institutionalized welfare state, and the incorporation of those classes into the political process via class-based political parties. Those diagnoses are true enough. I believe, however, that the points just enumerated lend an additional dimension of understanding. It has been not only a matter of incorporating a previously unincorporated political force into the polity; it has also been a matter of the progressive diffusion of class lines, so that the working class— or any other class, for that matter—has become less certainly identifiable, less conscious, and less mobilizable politically. These forces not only supplement the accounts given by postmodern theorists but also ramify the political process in other ways. They may, for instance, account not only for the relative weakening of class-based political parties; they may contribute to our understanding of the increasing salience of personality in political campaigns and our understanding of the increased reliance on media messages that are not specifically class messages. Furthermore, with the class and group structure of societies thus diffused, politicians themselves face a more ambiguous array of constituencies, mainly because familiar class lines of thinking match less well with social reality. Furthermore, the political salience of non-class-identified groups in the political process (ascribed groups and "new" social movements of various sorts) creates specific kinds of difficulties for the integration of society through the polity. To this last topic I now turn.

The Continuing Problematic of Societal Integration

In calling attention to social integration we must again begin with Durkheim, who more than any other scholar made that issue problematic. In doing so he was reacting in the first instance to the Spencerian notion, derived from Adam Smith's conception of the "invisible hand," that the individualistic pursuit of self-interest results in a collective or societal equilibrium that renders the issue of integration non-problematic. For Durkheim, the answer could not be so simple. A more active, positive regulation was required. Durkheim found this mainly in the generation of a legal system that served to regulate the interdependencies of differentiated structures and agents. In addition, he gave the state a distinct and expanding integrative role: "There is above all one organ in regard to which our state of dependence continues to grow: this is the state. The points we come into contact with it are multiplied, as well as the occasions when it is charged with reminding us of the sentiment of our common solidarity" (Durkheim [1893] 1984: 173). Despite this acknowledgment, Durkheim himself, in effect, fashioned his own version of an automatic solution to the problem of solidarity: organic solidarity is found in and arises from the division of labor itself His commitment to that notion lay behind his controversial proposition that anomie, class conflict, and other sources of instability are pathological and transitory.

Since Durkheim's time I believe that we, as sociologists, have revised his notions of solidarity in two fundamental ways. First, we have come to regard it as forever problematical and fragile, and forever requiring active efforts on the part of agents of integration to reproduce and sustain it in a national citizenry. Second, we have come to realize that there is not only one primary type of solidarity (organic solidarity) in complex societies but rather many types, and that these are related to anti overlapping with but not reducible to one another. By way of a nonexhaustive identification, I list the following:

•     Economic integration, or the interdependence of specialized economic agents via the market. This is the type of integration stressed by Adam Smith, which Durkheim criticized but at the same time acknowledged by placing differentiation so centrally in his own theory of integration.

•     Political-legal integration, involving the role of government in the maintenance of social order through the regulation of behavior and the resolution of conflict.

•     Cultural integration (including religion, common values, common ideology, and common language). This is the kind of integration associated with the writing of Talcott Parsons, who insisted, in perhaps his most controversial proposition, that all societies are characterized by a consensus on common, societywide values.

•     Integration through stratification-domination. Although this form bears a resemblance to political-legal domination, it is not the same. The premodern (and pre-nationtate) system of feudal "orders" is an example.

•     Kinship integration, which binds persons related by blood, marriage, and adoption to one another. In some historical situations kinship is fused with stratification domination, as in the case of hereditary monarchy.

•     Territorial integration, or the binding together of people by virtue of common residence and proximity.

As indicated at the outset, our sociological and political heritage has led us to expect that most of these aspects of integration are fused together in the modern nation-state—that is, the national economy, national territorial sovereignty, national monopolies over law, political regulation, and the means of violence, nationally based stratification systems, and national values or cultures.

The nub of the contemporary problem of both state and society, l would submit, is that this fusion at the societal level is by no means natural and that, in fact, we witness a growing disjunction, a systematic moving apart, of these bases of integration from one another and from the state and a corresponding weakening of the state as an integrative instrument. Let me only mention some salient evidence.

•     The increasing regionalization and internationalization of production, finance, markets, and trade have carried the economic differentiation and integration more and more beyond the capacity of the state (Cable 1995).

•     With the international movement of peoples, the augmentation of ascribed and semiascribed diversity within nations, and the survival of national minorities in newly founded states (conspicuously in the Balkans and in the former Soviet Union), the map of cultural solidarity coincides less and less with both territorial and national political integration (Brubaker 1995).

•     With the diffusion of the mass media, as well as the technological possibilities for instantaneous invisible communication (via fax and the Internet) and encryption, national boundaries tend to dissolve. Furthermore, whatever control over the flow of information (including market transactions) national states previously enjoyed, is correspondingly weakened. These technological possibilities also suggest the possibility of internationally based informal economies that escape the notice of national authorities even more than national informal economies do.

Put in concise form, the major contradiction is that the nationally based systems of political-legal and territorial integration are being increasingly besieged by economic and political developments mainly "from above" and by cultural developments mainly "from below" the nation-state level. Those developments present a special problem for the vitality of political democracy. One of the hallmarks of that system of governance, as it has evolved, is that political authorities at the state (and often local)level are elected by and ultimately accountable to national electorates. But by virtue of the erosion of certain aspects of the state's integrative capacities, democratic representatives of national peoples become progressively less able to govern and assure integration, because they lose control of many of the fundamental instruments of integration. In a word, they are, more and more, being held accountable for matters for which they cannot be accountable.

One final issue concerning the governability of democratic societies traces to the phenomena of cultural diversification. Democratic theory has come to mean many things since its formulations by Plato and Aristotle, but one of those meanings with special contemporary relevance is the notion of democracy as a set of representative governmental institutions in a pluralistic society with diverse and competing interests. The effectiveness of those institutions, moreover, is assessed according to their ability to hear those interests, negotiate with those leaders who speak for them ("prolocutors," to use Mayhew's [1990] term), and forge compromises that, with varying degrees of success, are aimed at settling current and forestalling future conflicts.

This version of the democratic process is built on several primary presuppositions: that demands made on the polity are in principle amenable to compromise; that prolocutors and their groups can envision compromise as an outcome; and that those in government can, in principle, fashion compromises. A type of conflict that readily fits with these presuppositions is industrial disputes, in which management and labor come into conflict over the adjustment of wages anti other conditions of work, and after a process of mediation or arbitration, some kind of mutually agreeable and binding, if not totally satisfactory, compromise position is put into place.

When claims on the polity do not meet these conditions, this creates difficulties for the democratic process. To choose another example from industrial relations, when conflicts between labor anti management concern principles of legitimacy, or the right of unions to exist and to be heard, they take on an either-or, nonnegotiable character and make incremental give-and-take and compromise more difficult. To state the matter more generally, the demands made by value-based or culture-based groups (often primordial in character) prove difficult for politicians to deal with precisely because they tend to assume an absolute, nonnegotiable character. The idea of primordialism implies above all that groups are usually rooted in sacred principles of membership or value-commitments. Defining themselves as sacred, primordial groups, they present positions and demands under the cloak of absolute principles that tend to have a noncompromising quality about them. To assert this is to argue neither that primordial groups do not have or express real interests nor that they do not engage in compromises. It is to argue, however, that primordial groups fuse interest claims with first principles, and this makes the process of compromise more difficult.

It follows that politicians and bureaucrats—the agents of the polity—tend to find the political demands of primordial groups uncongenial. The reason for this is that they do not easily lend themselves to compromise solutions that are the stock-in-trade of these agents. Consequently, when claims and conflicts of an absolute character arise, politicians in power tend to run for cover, to deny or otherwise minimize the primordial elements of those claims and conflicts, or to attempt to redefine them in ways that permit them to be dealt with as compromisable items. This is simply to assert that the increased salience of cultural diversification presents special challenges to democratic polities because they press against the edges of the tacit "rules of the game" of democratic governance.

To raise these points about the fragility of national boundaries and the capacity of nation-states to integrate and govern leads us logically to the concerns of international or global sociology. The developments that influence the permeability and fragility of national boundaries and the capacity of nations to govern are, as indicated, intricately tied to developments in world society itself. We will face that society directly in the final chapter, and note its characteristics and its capacity to penetrate national anti local bases of social organization.