The Infrastructure of Modernity:
Indirect Social Relationships, Information Technology, and Social Integration

Craig Calhoun

Social Relationships and Social Integration

During the last decade the reemergence of human geography as a vital field of social science has brought renewed attention to the spatial organization of social relationships (see, e.g., Gregory and Urry 1985). Over a slightly longer time span network studies have brought new power and sophistication to the analysis of concrete patterns of social relationships.^[1] These advances draw attention to a weak spot in contemporary social theory. The study of structures of social relationships, that is, the concrete connections among social actors, has not been used to improve our understanding of social integration. Those focusing on structural analysis have failed to show how patterns of relationships constitute social life and hold social institutions and populations together; they have thrown out the problematic of social integration like a baby in dirtily functionalist bathwater. Functionalists have turned to cultural understandings of social integration, but these analyses at best

[1] The advances in network studies have been primarily methodological and only to a lesser extent empirical; it is tendentious to speak of network "theory." Contemporary network studies in sociology are often conducted in atheoretical, positivistic terms, but they draw at least implicitly on a range of structuralist and economic individualist or utilitarian theories. The modern network approach had its origins, and some of its most substantively fruitful work, in structural-functionalist anthropology. See Barnes 1954, Mitchell 1969, and Nadel 1957. Nadel, in particular, was both theoretically oriented and sophisticated. The main modern attempt to trace the theoretical foundations of network studies is that of Burt (1982).

omit and at worst obscure attention to the concrete patterns of social relationships.^[2]

This failure, or even refusal, to approach social integration on the basis of patterns of concrete relationships is common to work in a wide range of otherwise divergent theoretical perspectives. Relational structures are too narrowly sociological a concern for many cultural theorists. For others the idea of social integration as a variable is too reminiscent of crude contrast of Gemeinschaft and Gesellschaft . It seems too normative to many theorists (including some who treat utilitarianism as value free) and too vague and unmeasurable to others. In what is perhaps the foremost contemporary effort develop a theory of social integration, Habermas (1984) focuses on a distinction between social integration/life-world and system integration/system in which concrete social relationships are seen as the stuff of the former, and the latter is conceived in cybernetic rather than social-relational terms.^[3] In functionalist accounts integration is usually conceptualized as a system state that is partially dependent on interaction patterns but distinct from these patterns. Clearly, in Parsons's sense (e.g., 1951), the extent of integration of a social system cannot be reduced to relational structures. However, in this chapter I contend that such relational structures have been neglected compared to other aspects of integration and that they have been conceptualized in ways that focus on face-to-face interaction and obscure the fact that mediated relationships are still social relationships.

Before we can explain social change satisfactorily we need a clearer conception of the relational dimension of social integration and the beginnings of a descriptive account of variation in concrete social relationships. The contribution of social relationships to social integration may be taken loosely as the complex variable measuring the extent to which the action of each person in a population implies, depends on, or predicts that of the others.^[4] Of course, the extent of social integration is not the only product

[2] I refer not only to Parsonian functionalism but also to Lévi-Straussian, or linguistic, structuralism insofar as work in either vein proposes to find in culture an autonomous level of logical and/or functional integration. This point is true even of Sahlins (1978) despite his intention to break fully from the functionalist problematic.

[3] "Thus there is a competition not between the types of action oriented to understanding and to success, but between principles of societal integration …. The rationalization of the lifeworld makes possible a kind of systematic integration that enters into competition with the integrating principle of reaching understanding and, under certain conditions, has a disintegrative effect on the lifeworld" (Habermas 1984, 342–43, emphasis in original).

[4] Note that this definition in no way requires sentiments of affection or solidarity as to common purposes or values. These are perhaps likely where social integration is great (both as results and as supports) but so are dissensus, anger, and resentment. The key is stable mutual determination, not pleasure in shared company. Note also that this definition examines the contribution of social relationships to social integration. Although social

integration is a broader concept than social relationships, I am arguing that more of social integration can be understood through relational analysis than is conventionally assumed. What Parsons termed the "media of interchange" (Parsons and Platt 1973, 23–25) help to constitute an infrastructure of social integration. Even very generalized media (like money) can be understood in concretely social relational rather than cybernetic terms. Where Parsons's concept of integration thematizes the issues of internal control and self-regulation of a social system, the present argument about social relations addresses a crucial source of capacity for control and regulation without treating these cybernetically as system properties. The control and regulation may often be intentional and sometimes arbitrary exercises of power. Using the same infrastructural capacity, they may as readily be destructive of harmonious relations among actors (e.g., in Habermas's sense of the destruction of life-world integration based on mutual understanding [see n. 3]) or among subsystems of social action (in Parson's sense) as constructive of them.

of variation in social relationships. These relationships may vary qualitatively in kind, quantitatively in density, and both qualitatively and quantitatively in pattern (including relative boundedness). The key is that we neither ignore concrete relationships nor privilege them as exclusively communal, and that we not leave large-scale organizations to representations in reified, actionless terms.^[5] I argue that, by paying attention to patterns of social relationships, we can provide a strictly sociological dimension to complement accounts of the contemporary age in terms of cultural and/or economic tendencies, 'modernity' and/or capitalism.

The first part of this chapter returns to the classical conceptions of modernity, which group cultural, economic, and social structural dimensions together more or less indiscriminately. I suggest a conceptual distinction between direct and indirect relationships that can help to illuminate many of the changes evocatively suggested by Gemeinschaft vs. Gesellschaft and similar oppositions while retaining a much clearer potential for empirical specification. Taking large-scale markets and especially corporations as examples, I show the utility of this simple conceptualization by focusing on the distinctive features of key modern social institutions. At least one of these institutions, the corporation, has been given surprisingly short shift in sociological theory. One of the constitutive features of the modern age, I argue, is the ever-increasing prevalence of indirect social relationships, that is, those relationships constituted through the mediation

[5] Part of Habermas's reason for maintaining his strong distinction between life-world and system is to leave a grounding for hope that the communicative constitution of relationships in the life-world can embody the potential for social transformation or (at least) resistance against the encroachments of the system. This is problematic (a) because it tends to accept a reified view of the system rather than to theorize it in a way that aims simultaneously to reveal and explain this reification, and (b) because it tends to overestimate the extent to which actually existing concrete social relationships can be understood as manifesting the ideal of communicative action. In this connection Fraser (1985) has suggested how Habermas's conception leaves little room for recognizing the mutual constitution of life-world and system, as for example in the way in which gender relations and identities are reproduced.

of large-scale markets, administrative organizations, and/or information technology. More and more these relationships are coming to be the basis on which society "at large" is constituted. This does not mean, however, that direct relationships are disappearing or losing their emotional potency for individuals, only that they are becoming compartmentalized and therefore altered in sociological significance.

The second section of the chapter links this analysis of the changing patterns of social relationships to changes in infrastructural—especially transportation and communication—technologies. Sociological and economic accounts of new technologies tend to focus disproportionately on production technologies and their effects on the labor force. I suggest that infrastructural technologies are at least as important and that the infrastructural uses of such new technologies as computers hold at least as much potential for social change. Such change, however, at present (and for the plausibly foreseeable future) seems to lie primarily in the extension of the trends of the last two hundred or more years, including the increasing importance of indirect social relationships, not in a reversal of these trends. In other words, modernity continues; we are not undergoing an epochal transformation comparable to that ushered in by industrial capitalism.^[6]

The third sections follows directly from this point and examines why terms like "postindustrial society" are exaggerations and points out a key sociological weakness of the theories on which they rest: failure of these theories to develop an account of what constitutes society in a postindustrial (or any other) age. In other words, lacking an account of social relationships, theories such as Bell's (1973, 1979) describe features of society but not society itself. In this failure, perhaps ironically, these theories fail to make use of openings to social-relational analysis and the problematic of social integration provided by the very functionalist theories with which they are often lumped by critics and on other parts of which they (like their progenitors in theories of industrial society) depend.

The last section of the paper takes up the Marxist account of capitalism. I try to show that however strong the Marxist theory of capitalism, it must remain a theory of part but not all of social life. Marxism lacks a theory of concrete social relationships (even though it offers a powerful theory of abstract relationships such as those mediated through the commodity

[6] This conclusion is true, I would suggest, in terms of both cultural and economic dimensions as well as social-relational dimensions (although there is no intrinsic reason why the three must always covary). Cultural accounts of postmodernism tend (a) to exaggerate the contemporary novelty of antimodernist movements, which in fact have a history as old as modernity, and (b) to neglect the continuing importance of such cultural tendencies as the privileging of quantity over quality or of accounts of change over those of continuity (the latter tendency being one that these accounts themselves often exhibit).

form). It offers an account of the dynamic tendencies that capitalism imposes on modern social actors, but not an account of social integration.^[7]

From Kinship and Community to Markets and Corporations

Contrast between country and city were a staple of nineteenth-and-early-twentieth-century social commentary (Williams 1973). Nearly everyone preferred the country. The country was clean, while the chimneys of city factories belched black smoke; the country was morally pure, while cities were dens of iniquity; perhaps most important, country dwellers enjoyed true community and social order, while cities were chaotic, unregulated, and anonymous. Early social theorists believed that cities somehow embodied the core features of a new kind of society, and this new society contrasted sharply with the previous, more communally solidaristic social order. Today, however, Tönnies's (1887) Gemeinschaft-Gesellschaft contrast, Wirth's (1938) and Redfield's (1941) folk-urban continuum, and other contrasts of tradition and modernity are as familiar as objects of abuse as they are as mandatory bits of vocabulary in introductory sociology textbooks. Those who attack this approach generally focus on the community or tradition side of the dichotomy (e.g., Gusfield 1967, 1975). They argue that the depictions by Tönnies and others of traditional community life are nostalgic and unrealistic; they also note that the portrayal of most of the Third World as traditional rather than modern is both patronizing and predisposed to neglect the extent to which contemporary Third World patterns are produced by modern capitalism and international relations.^[8]

Surprisingly, the sociological inadequacies of Tönnies's (and others') conception of Gesellschaft have not received comparable comment; the same goes for most of the other well-known binary oppositions.^[9] The

[7] Marxism's lack of an account of social integration is not simply the result of the fact that Marxist class theory suggest fundamental social contradictions whereas Parsons and other functionalists consider the stratification system to be primarily integrative. The problematic of social integration is not developed in Marx's work or most Marxism. Indeed, the concrete relational dimension to internal class solidarity is not much developed by Marx; when it has been studied by other Marxists, the theoretical or conceptual bases have been drawn from outside Marxism and often have been left in unclear relationship to Marx's more central categories theory of capitalism.

[8] This voluminous literature has been reviewed recently by Worsley (1985). The arguments of Wallerstein (1974, 1979) and Frank (1969, 1978) are perhaps the most prominent. Calhoun 1978, 1980, 1983) ties to salvage something of the notions of community and tradition from these critical dismissals and from the genuine nostalgia, paternalism, and error found in earlier formulations.

[9] The one real exception to this contention is the Marxist critique that most of these conceptions neglect the centrality of capitalism to "modernity" or Gesellschaft .

impact of the transition from Gemeinschaft to Gesellschaft in Tönnies's conceptualization was largely the loss of a felt sense of belonging together in favor of an exaggerated individualism and a focus on instrumental relations. Tönnies had little social-structural foundation for his notion of changed personal experience, which accordingly remained unsatisfactorily impressionistic.^[10] Simmel's analysis of "the metropolis and mental life" made a good deal more of the change in concrete social relationships that accompanied the emerging social psychology of urban life ([1903] 1971]). His attention, however, was focused on the larger issue of the development of individuality in the modern West. His characterization of cities, along with most of the other famous typologies of tradition and modernity, offered only a very general view, one that was lacking in historical specificity—or rather, one that failed to recognize the historical specificity implicit in its apparently general account (Abu-Lughod 1969). To be more precise about the experience—let alone the sociological significance—of modern urban life, we need to go beyond such broad statements about sociopsychological differences to a specific analysis of change in the structure and the kind of social relationships.

Almost all major premodern forms of social organization depended primarily on direct interpersonal relationships. Kinship, community life, and even the most stable, recurrent relationships of economic exchange all took place within the conscious awareness, and usually the face-to-face copresence, of human individuals. Such relationships might be more or less systematic and complex: for example, webs of kinship can link hundreds of thousands of members of traditional African societies. However, the actualization of each relationship, as opposed to its latent potential, was normally directly interpersonal. Although state apparatuses certainly predate the modern era (and occurred historically throughout the world), Giddens is surely right to argue that few if any were able to "govern" in the modern sense of the word; their capacity for regularized administration of a territory and its residents was very limited.^[11] This limitation was largely the result of the fact that power relations could not be extended

[10] It was partly to avoid this sort of impressionistic account that human ecology went to the opposite extreme in borrowing models from biology and developing a highly "objective" account that purported to treat cities as wholes without reference to their constituent interpersonal relationships. Here the key variables in human ecology became population, organization, ecology, and technology. See Duncan 1959 and Hawley 1950, 1981. Haines (1985) has offered a cogent critique of this biological emergentism and the human ecologists' corresponding neglect of alternative "relational" approaches to their subject matter.

[11] Giddens 1985, 63. Such "administrative power can only become established if the coding of information is actually applied in a direct way to the supervision of human activities, so as to detach them in some part from their involvement with tradition and with local community life" (Giddens 1985b, 47).

effectively over large distances.^[12] Although their cultural variation was enormous and their variation in specific patterns of social organization was considerable, premodern peoples were only rarely able to produce the physical infrastructure and administrative practices that are necessary to build large-scale social organization of much intensity.

The direct relationships that prevailed included both "primary" and "secondary" ones, to use Cooley's language ([1909] 1962).^[13] Useful though it may be for some purposes, Cooley's conceptual distinction does not differentiate the modern age adequately from its predecessors. Modernity is not constituted by the presence of secondary relationships or the absence of primary relationships; both sorts exist in a wide range of modern and nonmodern societies. Rather modernity is distinguished by the increasing frequency, scale, and importance of indirect social relationships. Large-scale markets, closely administered organizations, and information technologies have produced vastly more opportunities for such relationships than existed in any premodern society. This trend does not mean that direct relationships have been reduced in number or that they are less meaningful or attractive to individuals. Rather it means that direct relationships tend to be compartmentalized. They persist as part of the immediate life-world of individuals, both as the nexus of certain kinds of instrumental activities (e.g., the many personal relationships that smooth the way for or make possible business transactions [see Granovetter 1985]) and, especially, as the realm of private life (family, friends, and neighbors). However, direct interpersonal relationships organize less and less of public life, that is, fewer and fewer of the crucially determinant institutions controlling material resources and exercising social power. Indirect relationships do not eliminate direct ones, but they

[12] This point was recognized some time ago by Innis ([1950] 1972) in his arguments about the centrality of space-transcending communications media to the building of empires.

[13] Cooley's opposition was between relationships that linked people merely as the enactors of specific social roles and those that involved whole persons, linked to each other in ways that transcended the fragmentation of life into different spheres. See also the discussion in Nisbet and Perrin 1977. Cooley's version of the Gemeinschaft-Gesellschaft dichotomy implied a Rousseauian critique of the inauthenticity of secondary relationships. This and other criticisms that secondary relationships are less fulfilling, less meaningful, and weaker than primary relationships are central to the concepts of primary and secondary relationships as they have commonly been used. Cooley's conceptualization exists to describe an expansion in the number of relationships that is accompanied by a deterioration in their meaningful content and social strength. There may be something to this critique, but it would be better to make the issue one of the relative importance of each sort of relationship in organizing various kinds of activities, to recognize the secondary relationships are central to public life and democratic political participation, and to keep in mind the distinctive difference between both sorts of directly interpersonal relationships, on the one hand, and indirect relationships, on the other hand.

change both their meaning and their sociological significance.^[14] Although they are as sociopsychologically and culturally powerful as ever, direct relationships are no longer constitutive of society at its widest reaches.^[15]

The growing importance of indirect relationships was recognized by both Marx and Weber. For Marx these relationships characterized above

[14] For a general discussion of some of the distinctive features of face-to-face relationships in contrast to mediated ones see Meyrowitz 1985. For example, nonmediated relationships are more easily clustered, he suggests, so that individuals can communicate in different styles and contents to different groups:

The combination of many different audiences is a rare occurrence in face-to-face interaction, and even when it occurs (at a wedding, for example) people can usually expect the speedy resumption of private isolated interactions. Electronic media, however, have rearranged many social forums so that most people now find themselves in contact with others in new ways…. And the merger of different audiences and situations through radio and television has made it difficult for national politicians to say very specific things to particular constituencies or to behave differently in different social situations. (Meyrowitz 1985, 5)

This observation seems quite true, but Meyrowitz's account is biased toward broadcast media (rather than computers and other electronic media that use numbers, text, and other more abstract codes). Although broadcast may indeed maintain a kind of surveillance over public figures that shapes their behavior and eliminates certain "privacies," two qualifications seem in order. First, earlier modes of social control may have been at least as powerful in forcing individuals to adhere to constant standards of rectitude in their behavior. Television cameras may never have invaded Victorian gentlemen's clubs or brothels to embarrass their patrons, but standards of public propriety in dress, speech, and the like allowed less flexibility in many regards even though enforcement was only through direct observation. (Giddens's [1985b] Foucault-inspired account of the growing capacity for surveillance is similarly not complemented by attention to either more informal means of social control or changing opportunities for political participation.) Second, computerized communications allow a great deal of tailoring of messages to specific audiences, as any recipient of direct-mail political advertisements knows. Where candidates speaking on television must appeal to a certain common denominator of "the general public," those targeting various population categories for funding can shape each appeal in a distinctive, perhaps quite contradictory way. An elaborate variety of statistical and other consulting services help candidates know just which issues to stress with mailing lists of churchgoers, veterans, schoolteachers, people of high incomes, parents of school-age children, or any of a hundred other segments into which the population of potential donors may be categorized. Although it is a bit dated (particularly with regard to the computerization of direct mail), Sabato 1981 is probably the best general account of this phenomenon.

[15] This situation is one source of modern populist politics—the politics of local communities and traditional cultural values. It is a potent kind of politics, and it offers potentially radical and important visions of alternative modes of social organization. Many of its variants, however, are based on some combination of (1) systematic misrecognition of the opportunities for local autonomy available in a world structured largely by large-scale organizations of indirect social relationships, and (2) systematically biased analogies between the world of direct personal relationships and that of large-scale organizations of indirect relationships (e.g., "balancing the U.S budget is just like balancing your family checkbook").

all the system of commodity production and capital accumulation. For Weber the commodity form was also key, but, characteristically, market rather than production relations were central; the "indirect exchange of money" was prototypical:

Within the market community every act of exchange, especially monetary exchange, is not directed, in isolation, by the action of the individual partner to the particular transaction, but the more rationally it is considered, the more it is directed by the actions of all parties potentially interested in the exchange. The market community as such is the most impersonal form of practical life into which humans can enter with one another. This is not due to that potentiality of struggle among the interested parties which is inherent in the market relationship. Any human relationship, even the most intimate, and even though it be marked by the most unqualified personal devotion, is in some sense relative and may involve a struggle with the partner…. The reason for the impersonality of the market is its matter-of-factness, its orientation to the commodity and only to that. When the market is allowed to follow its own autonomous tendencies, its participants do not look toward the persons of each other but only toward the commodity; there are no obligations of brotherliness of reverence, and none of those spontaneous human relations that are sustained by personal unions. (Weber [1922] 1978, 636)

Weber's ideal-typical market does not correspond to any actuality, of course, any more than Marx's pure model of capitalism does. But each expresses a distinctly modern tendency.

Weber's analysis of bureaucracy suggests another such tendency: the creation of social apparatuses for rational administration. Weber tended to assume that bureaucracies would be sociogeographically concentrated; he associated them with cities and treated their development as a specification and enhancement of the role of cities as centers for the exercise of power. In fact Weber wrote at about the point in Western history when cities began to lose some of their distinctive centrality to systems of power and administration.^[16] In ancient empires, dispersed city-states, and late-feudal Europe alike, cities had been at the heart not only (obviously) of civilization but also of both power relations and trade. Cities were the nodes that could anchor structures of indirect relationships in an age of minimal information technology. They could provide for mediation among participants in far-flung markets, and they were

[16] Thus Weber was one of the many classical sociologists who together placed urban studies at the heart of sociology without recognizing the historical specificity of the centrality of cities, especially to the period of the transition to modern capitalism and nation-states. See Saunders 1985 for an argument that this error is at least partially repeated (with less excuse) by the geographers and sociologists who would make the spatial analysis of social relationships a field of its own.

the focal points for political and military control. As a result, they created networks of power and exchange stretching well beyond their boundaries. Moreover, they provided(and to a considerable extent still provide) for the direct relationships that make systems of indirect relationships work (the personal relationships that connect banking houses, for example, and the direct communications among central government officials).^[17] Cities also provided for public life, which is composed of direct—although not necessarily intimate—relationships among strangers (Sennett 1977; Calhoun 1986). But the development of modern transportation and communications technologies, on the one hand, and the growing administrative organization and power of the state, on the other, meant that both economic and political activity could begin to bypass the cities.^[18]

In short, state power could grow because the new forms of organization and the improved transportation and communications infrastructures (based partly on new technologies but, at first, more on heavy investments in the extension of old methods) enabled the spread of increasingly effective administration throughout the various territories of a country. This is the story Giddens (1985b) offers as the centerpiece of his critique of historical materialism, and it is a necessary complement to Marx's analysis of capitalism. It is a crucial complement, but it is not sufficient.

A full account needs to recognize, first, that the growth of the state, like the capitalist economy, developed infrastructures that could be used by ordinary people to develop connections with each other. Roads, trains, telegraphs, and telephone furthered the social integration of dispersed populations, promoted their common participation in capitalist production and exchange, and made possible their common subjection to state surveillance and administration. Class struggle itself, in the sense of the mobilization of workers organized on the same sociogeographic

[17] Ancient and early-modern cities relied much more on these direct relationships because they lacked the material and organizational infrastructures to do otherwise. Written communications were the only means of transcending the spatial and temporal limits of copresence—and literacy was not widespread, especially outside the cities. In countries of limited infrastructural technology and organization cities still appear to be largely aggregations of smaller populations linked almost exclusively by direct interpersonal relationships. Cities remain central (and often swell beyond their ready supporting capacity or their leaders' desires) because the lack of communications, transportation, and organizational infrastructures makes it all but impossible to create economic (and political) opportunities at a distance.

[18] "The growing obsolescence of the city, in its traditional form, in political, economic, and military terms, is one of the most fundamental transitions initiated—although certainly not completed—as part of the emergence of the absolutist state" (Giddens 1985b, 97).

scale as capital accumulation, had to wait for a communications infrastructure that was adequate to the formation of large-scale trade unions and political parties (Calhoun 1988).

Second, a full account also needs to recognize that modern states are in fact special (and critically important) instances of a more general phenomenon: corporations. As Giddens notes, the absolutist kings were distinct from other traditional rulers in the crucial sense that they not only sat at the pinnacle of state power but also incorporated the state symbolically within their own sovereign persons: "The religious symbolism of 'divine right' should actually be seen as a traditional accoutrement to something very new—the development of 'government' in the modern sense, the figure of the ruler being a personalized expression of a secularized administrative entity" (1985b, 93–94). This notion is part of what Kantorowicz (1956) meant in his brilliant portrayal of the late-medieval doctrine of the "king's two bodies," one personal and the other public. The king had begun to assume the status (still common to Roman Catholic bishops and other ecclesiastical nobles) of a "corporation sole" (see also Gierke 1934; Maitland [1900] 1958). Eventually, a doctrine of corporate personality developed that freed the corporation from any legal need for embodiment. On this and other bases corporations (starting at least as much with the state and various monastic bodies as with the urban corporations from which the lineage is usually traced) were eventually able to command routine public, jural, and even (rather unanalytical) sociological acknowledgment as unitary actors.

Oddly, the corporate form of social organization has received very little attention in sociological theory even though it is central to modern institutional arrangements.^[19] A brief discussion of this remarkable form of organization is therefore in order before considering more recent information technology and the question of whether we have left, or are about to leave, modernity behind us.

The corporation is a remarkable cultural artifact. One of the most extraordinary things about business corporations—as well as the other types of corporations from religious and charitable institutions to governments and quasi-governmental organizations of various sorts—is that we so routinely reify or anthropomorphize them.^[20] With minor variations and qualifications corporations throughout the Western world may own

[19] Coleman 1982 is one of a handful of modern exceptions to this stricture; see also Selznick 1969. One of the best sociological treatments comes not from a sociologist but from the legal theorist Dan-Cohen (1986).

[20] In Social and Cultural Dynamics, Sorokin (1957, chapter 18) found that the modern West exhibited a "reemergence of singularism," with continued rapid growth into the late-nineteenth and early-twentieth centuries. Sorokins's analysis links singularism—the claim of ontological or nominalist social reality for collectivities—with a "triumphant individualism."

property, litigate, and make contracts in the same way as "natural persons." They may, indeed, enter into a variety of relationships—usually highly asymmetrical—with natural persons.^[21]

Such relationships are quintessentially indirect. Although real human beings are linked by them, this linkage is almost invisible. Indeed, social relationships seem to disappear in the operation of an apparently self-moving technical and social system.^[22] With even a minimal communications technology the relationships constituting an organization can be rendered indirect, that is, distanced both in time and in space (e.g., by the storage, retrieval, or transmission of the written word) and socially mediated (by transmission through the official functions of other corporate agents). Thus a corporation is in one sense merely an aggregate or structure of social relationships, most (but not all) of which are indirect. In another sense, however, it is a social creature at a different level, a whole unto itself.^[23] Our Western—especially American—culture grants the corporation the status of an autonomous actor, one that is capable of "responsibility," thus offering its members limited liability for their actions.^[24]

Both corporations and large-scale markets depend on the flow of information through indirect social relationships, and both are accordingly

[21] "Asymmetrical" is Colemen's (1982) apt term for relationships between corporations and "natural persons." The relationships that spring to mind are, usually, contractual ones such as employment, the sale of goods, or credit; the ownership of shares in a corporation is perhaps a special case of contractual relationship. But corporations also enter into other sorts of relationships with natural persons, as for example when they produce or distribute toxic substances that kill them.

[22] Studies have long explored the impact of mechanical analogies on our understanding of human nature. the idea of automation—self-movement—came early on to influence conceptions of social organizations. Hobbes ([1651] 1962, 81), for example, describes Leviathan as an artificial man, or automaton. The image of an artificial man suggests much of our ordinary understanding of the independence of such social automata from human action. Sociotechnical systems subject to automation not only industrial production and office work but also the control and coordination of social relations. In doing so, they create indirect relationships that are particularly conducive to reification.

[23] A corporation exhibits, for example, the characteristics Durkheim ([1895] 1966) thought defining of social facts: it endures, at least potentially, longer than any of its individual members of agents; if is external to any individual (although whether it is external to all is perhaps better treated as a question of theoretical presupposition rather than one of empirical fact; see Alexander 1982); and it is capable of coercion over individuals, both conscious and intentional, and unintended and/or unrecognized on either side.

[24] See the discussions in Dan-Cohen 1986, French 1984, Orhnial 1984, and Stone 1975. Attempts to apply criminal law sanctions to corporations raise particularly difficult questions about their "responsibility" and ontological status. It is not entirely clear, for example, what is meant by the notion of "punishing" a firm, as distinct from the individuals who act as its agents, own it, or otherwise create and compose it (see Coffee 1981).

subject to routine reification.^[25] Economists predict, and nearly everyone discusses, the economy as though it were as natural and objective a phenomenon as the weather. This tendency reflects a culture that is at once pervasively individualistic—and thus underrecognizes the social dimension in the creation of both markets and corporations—and at the same time supports a maximally "disembodied" ontology that allows people to accord some manner of unitary individual existence to bodiless social creatures. Markets differ from corporations, however, in that they lack administration. They are the aggregate of individual actions, and sometimes collective action, but they are not collective actors.

Because of this difference, corporations tend to be not only reified but also anthropomorphized. As note earlier, we no longer find it necessary to embody states in the persons of their rulers; er attribute individuality to the disembodied state itself (see also Manning 1962; Giddens 1995b, chapter 4). Similarly, corporations are readily understood to exist, and in some sense to act, independently of their chief executives. However attractive Chrysler Corporation may find it to promote Lee Iacocca as its symbolic image or however much Ronald Reagan may appeal to Americans as a symbol of their country, no once confuses the person with the corporation. As Justice Marshall wrote nearly a hundred and seventy years ago, a corporation is "an artificial being, invisible, intangible, and existing only in contemplation of law."^[26] The confusion comes in treating the corporation as a person.

[25] Although administrative flows of information are the most obvious aspect of this dependence in corporations (including states), we should not forget that the maintenance and operation of such organizations as collective actors depends on a whole variety of "informal," that is, not specifically administered, information flows. Indeed, one of the questions that some ask about the new information technology is whether it does not subject many of these informal lines of communication to increasing surveillance or formalized administration and thus in one way weaken the organizations it more generally facilitates.

Both money and commodities can be understood as the basis of information flows in markets. Parsons (1963), Luhmann (1979, chapter 3), and Giddens (1985b, chapter 5) all offer understandings of monetary transactions as communications. Simmel ([1907] 1978, 284–85, 297–302) stresses the role of money in making possible impersonal relations between people and thus promoting individualism; he does not, however, quite develop an account of money as a medium of communications. Marx's notion of commodity fetishism has been developed in this direction (Taussig 1978; Lukács [1922] 1971). Both Luká and Simmel draw connections between commodification and what Simmel calls "the calculating character of modern times" ([1907] 1978, 443).

[26] Dartmouth v. Woodward, 17 U.S. (4 Wheat.) 518 (1819). See also Simmel's stress on the significance of the personal unity of the owner, including the socially created corporate owner:

Every sum of money has a different qualitative significance if it belongs to a number of people rather than to one person. The unit of the personality is thus the correlate or the pre-condition for all qualitative differences of possessions and their importance; here the assets of legal persons are, in terms of their function, on the same level because of the uniformity of their administration. Similarly, we may speak of a nation's

wealth only if we conceive of the nation as a unified possessing subject. That is to say, we have to conceive the assets owned by the individual citizens as being unified by their interaction within the national economy, in the same way as the fortune of one individual comes together as a practical unity through such interactions—for example, distribution, the relation of individual expenditures to the total, balance between income and expenditure, etc. ([1907] 1978, 271).

For Weber ([1922] 1978, 48–52), social relationships constitute an organized or corporate group (Verband ) only insofar as a set of specific individuals (usually together with an administrative staff) regularly enforces its boundaries. Weber refuses, however, to recognize the corporate whole as distinct from the individuals in authority. See also the discussion in Sorokin (1957, chapters 18–19)

Information Technology and the Expansion of Indirect Relationships

The past two decades have seen a near passion for labeling new ages and new kinds of society: postindustrial, technetronic, Third Wave, etc.^[27] These visions of a new and different age derive substantially from the anticipated effects of new technology, most prominently computers and related information technologies. Although this technology is indeed powerful, such accounts of a qualitative break with the previous two hundred years of modernity are misleading. New technologies have extended the most basic trends in social integration more often than they have countered them, and this pattern will probably continue unless substantial social effort is invested to the contrary.

Corporations, large-scale markets, and other organizations of indirect social relationships have grown in size and importance throughout the modern era. Advances in information technology have repeatedly facilitated their extension. Computers and new telecommunications technologies continue this pattern. They not only offer a large quantitative increase in indirect relationships but also contribute to a shift in balance between two qualitative kinds of indirect relationships. By extension from Cooley's notions of primary and secondary relationships, we might conceptualize two kinds of indirect relationships: tertiary and quaternary.

Tertiary relationships need involve no physical copresence; they may be mediated entirely by machines, correspondence, or other persons, but the parties are well aware of the relationship. A tertiary relationship may be created, for example, by writing to a more or less anonymous functionary of a large bank to complain about an error in one's statement. Most ordinary citizens have only tertiary relationships with their national political representatives, relationships that are mediated by broadcast and print media, voting in elections, and, occasionally, correspondence. In a large

[27] Among many, see Bell 1973, 1979. Touraine 1971, Brzezinski 1977, Toffler 1980, and Naisbitt 1982; see also the review by Badham (1985) and the sampling in Forester 1986.

corporation most employees have this sort of relationship with top managers. Such relationships are more or less fully contained by their explicit purposes and systemic roles. Because they are not characterized by physical copresence, they are not as open to redefinition and expansion as are secondary relationships. The various media through which the relationship is carried out help in varying degrees to seal role performance off from the other attributes of individuals. But the relationships retain a degree of mutual recognition and intentionality; each party can (at least in principle) identify the other and the relationship itself is manifest.

Quaternary relationships, by contrast, occur outside of the attention and, generally, the awareness of at least one of the parties to them. They are the products of surveillance and exist wherever a sociotechnical system allows the monitoring of people's actions and turns these actions into communication, regardless of the actors' intentions. Quaternary relationships are created by the tapping of telephones, the theft of computerized data, or even the analysis of stored data for purpose other than those for which they were initially provided by people other than those to whom they were initially provided. Each person who uses a credit card, travels on an airplane, pays income taxes, applies for a visa, or completes employment applications—that is, nearly everyone in a modern society—provides data that can be subjected to reanalysis. Such reanalyses can be used to trace behavior of particular individuals as groups. The purpose of such surveillance may be as benign as providing marketing information or as threatening as discovering the members of minority ethnic groups for purposes of control or persecution.^[28] Modern markets and governmental apparatuses could not function in their present manners without substantial use of such data. This use nonetheless constitutes surveillance and creates very indirect, nearly invisible, but potent quaternary relationships.

As we saw in the case of markets, not all quaternary relationships depend on the administration of information. The flow of money in successive transactions is itself a form of communication; monetarization laid the basis for an extension of markets that created extremely indirect, almost invisible quaternary relationships of their own. However, the distinction between monetary communication in large markets and administered information flows is diminishing as money comes increasingly to be one version of electronically encoded information. In addition to markets a variety of noneconomic relationships are facilitated by new technologies. For example, hobbyists may use computer networks to keep in touch, a sort of semiadministered use. However, new technologies have

[28] For example, the U.S. government provided supposedly confidential census data to police and legal authorities for use in finding Japanese-Americans as part of the unconstitutional program of mass detention during Worlds War II (Burnham 1983, 20–26).

their most dramatic impacts on various kinds of more fully administered information flows.^[29]

The use of writing marked the first great historical breakthrough in creating the capacity for indirect social relationships. Our present capacities still depend more on literacy than on any other invention or skill. But information technology has advanced enormously throughout the modern era, from the invention of printing presses through telegraphs, telephones, radio, television, communications satellites, cable and microwave transmission, and now computers.^[30] Improvements in transportation facilities have also been vast and for centuries constituted the basis for most long-distance communications. It is worth reemphasizing how recent and how enormous transportation improvements have been. As recently as the mid-1750s, it took ten to twelve days to travel from London to Edinburgh; by 1836 less than two days were needed for the trip (Bagwell 1971); the train (which on that route is nowhere near "state of the art") now takes four hours, the plane one hour, and electronically mediated communication is virtually instantaneous. German immigrants to America after 1848 could not count a reliable post (the International Postal Union dates from 1874) and could not expect ever again to see the family member they left behind. Yet, their great-grandchildren not only travel and phone between the two countries with ease but also own shares in corporations doing business simultaneously in both countries and depend on military coordination in which computers and satellites link commanders and staffers thousands of miles apart and monitor the entire face of the earth.

Through most of history wars of conquest and the migrations that they precipitated were among the few major vehicles of long -distance and cross-cultural communication. At considerably greater intervals wars and migrations were supplemented by waves of religious conversion. More frequently, religious pilgrimages and commercial expeditions were mounted. Trading routes provided a flow of gossip, but the everyday scale of activities was much narrower. Communications capacities grew out of the political needs of managing empires and the logistic needs of armies, drawing on the resources of clerical literacy (hence, "clerks"). In the early-modern period commerce began to complete with

[29] Administration may refer to both the monitoring of information flows and the organized creation of such flows. In other words, it may refer to both phone tapping and television broadcasting as well as to the combination of both aspects in electronic credit checks.

[30] Media theorists focusing on telephone, broadcast, and other electronic technologies have frequently emphasized the renewed prominence of oral communications in shaping modern culture (McLuhan 1964; Ong 1982; Meyrowitz 1985, 16–23). The role of computers in the latest wave of electronic communications technologies presages a partial reversal of this trend and a revival in the centrality of literacy.

and then surpass warfare as the occasion for international communications. It multiplied the demands for literacy and improved communications media. The growth of strong national states was tied closely to both this growth of commerce (which simultaneously provided the wealth with which to run the states and a need for the protection of trades and international boundaries) and the communications media themselves. National integration was further by these communications media not only through enhanced means of administration (as Giddens 1985b emphasizes) but also through the growth of cohesive national cultures and shared "consciousness" or ideologies. Linguistic standardization and codification was a key step in nation building (even through the histories of the German-speaking and the Romance-language countries differ somewhat in this regard) and in turn made long-distance communications still easier.^[31]

Like literacy itself, new information technologies enable the transcendence of not only space but also time; fewer relationship or transactions require the copresence.^[32] Although it is a mundane sort of time machine and disappointingly inert, each gray metal file cabinet enables communication to take place between those who put information in it and those who take it out. Computers are able to do this, of course, on a much larger scale, with much more sophisticated procedures for matching the stored bits of information to the inquiries of future searchers. One of the most distinctive features of modern corporations is their ability to combine a high level of continuity on their patterns (in the face of both environmental complexity and fluctuation and internal personnel changes) with a capacity for organized change in response to managerial decisions. Computers can be used to monitor activities in connection with very long-term plans or simply to maintain conformity with preset rhythms and routines. Communication is increasingly separated from transportation, surveillance from direct observation.

New information technologies may be used in the following ways: to organize more of social life through indirect relationships, to extend the power of various corporate actors, to coordinate social action on a larger scale, or to intensify control within specific relationships. This broad set of potential can indeed be socially transformative; the technologies are

[31] Today television is one of the central means of promoting a shred national consciousness, especially for large countries. Many Third World countries have invested have invested heavily in it for just this reason, but few can rival China's purchase of several entire color television factories from Japanese manufacturers in order to meet its goal of one television per household by the end of the century.

[32] In Innis's ([1951] 1964) sense, new information technologies are biased, like their forebears, toward the transcendence of space more than of time, toward reach and flexibility rather than toward permanence

powerful and pliable. But will this transformation be a break with the trend of modernity toward an increasing reliance on indirect social relationship to organize large-scale social integration?

An excessive focus on the question of the extent to which new technologies supplant human labor in the production of material goods has obscured the deeper import of these new technologies for social integration. It is indeed true that the proportionate contribution of "artificial" (not directly to human) labor to the production process has grown and is likely to grow much further.^[33] This trend is important, and the potential employment impacts of computerization are not insignificant (see Jones 1982; Gill 1985). Similarly, computerization offers major advances in productivity (not only for labor, but also for capital facilities). Focusing only on these issues, however, obscures other very significant changes, including changes in the material production process itself.

The greatest changes in most production facilities are not in the numbers of people employed or even in their skill levels, but in "throughput" process (Gunn 1982). Computerization enables not just the automation of a variety of different specific production processes (welding, say, or painting) but also the automation of the flow of goods, materials, and information through the factory.^[34] At an automobile assembly plant, for example, each chassis is given a computerized identification card at the start. It corresponds to a specific car ordered by a specific dealer. The computer indicates to each worker (or robot) what parts to add to the basic chassis, what color to paint it, and what trim or finishing to give it. The computer also orders all necessary parts and materials as needed, thus cutting down on both the clerical work force and the necessary inventory. As with assembly line and indeed factories themselves, the change here is in the organization of the production process. Computerization's most profound industrial impact comes in increasing the scale of technically

[33] This trend may shift large parts of the work forces of the advanced economies out of material production, but two qualifications needed to be suggested to Bell's characterization of this shift as a move to postindustrial society (1973, 1979). First, most of the jobs created for former industrial workers (or their children or their future counterparts) are at least as mundane and often at least as manual. What could be more exemplary manual labor than typing? Yet typing is "information-sector employment." Much the same goes for such service-sector jobs as making and selling hamburgers at McDonalds. Second, proportionate declines in employment in material production industries are not mirrored by proportionate declines in investment in capital goods or value added in economic production.

[34] One merit Beniger's (1986) account of the "control revolution" is to grasp this feature of information technology, which is overlooked by many accounts. Beniger's cybernetic model for understanding this, however, tends to obscure many of the issues addressed in this chapter, as well as the matter of struggle over the course of technological change. Beniger prefers to present control as an attribute of reified "systems" rather than chosen or created eventualities.

coordinated activity at the same time that it establishes flexibility for small-batch production. This description of the production process (based on observations in wayne, Michigan) is part of a computerized integration of design, production, and marketing facilities in seven countries on four continents (Ford Motor Company's World Car project).

What is changed, in short, is the social integration of the production process. Just as the factory and the division of labor transformed the production of goods in the classical revolution, so computerization today transforms not just individual task but whole organizational forms. Although automation displaces some manual workers by having machines do their jobs, it changes society by replacing the human component in many organizational links.^[35] Social organizations itself is in some sense automated as computers and related information technologies help to create an artificial locus of self-environment. We call a machine automated if it can drive itself or perform autonomously. So, too, we could call the process of creating factories or even more far-flung but autonomously working organizations one of automation. We might even consider that corporations are a kind of social automaton. They are made possible by indirect relationships in which human functionaries serve as intermediaries, but they are greatly expanded on the basis of new information technologies for the mediation of relationships.

Social Integration and the Idea of Postindustrial Society

In attempting to revitalize and reformulate the problematic of social integration, I am following most closely in the path of Durkheim but also in varying degrees in those of Tönnies, Weber, and Simmuel. Durkheim made social solidarity and social integration more distinctively his concern than any other classical sociologist, but he did not for the most part approach these issues concretely through the study of patterns of relationships. Rather he concentrated on the sociopsychological sense of mutuality, the cognitive implications of life in society, and the functional analysis of cohesion among institutions.^[36] Functional analysis of this sort

[35] It is common to stress the "deskilling" of such manual workers as machinists and their partial replacement by technical specialists, who are more easily controlled (Noble 1984). Whatever the merits of this argument as to the motives behind corporate policy, it should be noted that the transformation of production organization also affects some managers. At Ford's Wayne Assembly Plant none of the managers of the facility had the ability to reprogram their robots or their throughput controls. They too had lost control to a centralized organization working through more specialized agents.

[36] Durkhiem's failure to develop an approach to concrete social relationships is centrally responsible for one of the most important weaknesses of his theory of social solidarity (or integration). Concentrating on differences in form between mechanical and organic

solidarity, Durkheim ([1893] 1964) reduces transformations in scale to a minor independent variable, not a major substantive feature of social change. As interesting a variable as dynamic density is, it does not substitute for an account of how dramatically different numbers of people come to be organized in a common web of social relationships or what this means for people's experiences, actions, or social structure.

is abstract even though it is not abstract in the sense or to the extent that Marx's categorial analysis of capitalism is.^[37] It is from Weber, above all, that we derive the concrete analysis of social relationships, which he understood as the probability that the actions on one person will influence the course of action of another.^[38] Not all social relationships are direct, of course, many are mediated. What renders the Weberian approach concrete is its focus on relationships from the point of view of the actors (thus necessarily recognizing qualitative distinctions) rather than on the categorial nature of meditation.^[39]

Ironically, although much of the structural-functionalism of the 1950s retained a focus on social integration and even on concrete social relationships (the latter perhaps more visible in the social anthropological variants), the "industrial society" theories that developed on Durkheimian and Weberian foundations as alternatives to Marxism exhibit the same neglect of social integration that characterizes Marxism. Industry is no more definitive of society than capitalism; if anything, it is less so (Kumar 1978; Giddens 1985b; Badham 1985). Industry as a way of organizing material production is clearly a feature of social life and is somewhat influential, but it is neither dynamic in itself nor the source of the basic web of relationships linking people to each other. The same problem is carried forward in Bell's theory of postindustrial society. The question of how value is produced is mistaken for the question of what society is. Regardless of the merits or demerits of Bell's notion that information

[37] Functional analysis is abstract in the sense that it does not study concrete interpersonal relationships or social relationships (in Weber's sense) but rather the influence of socially constructed institutional phenomena on each other. In still another sense statistical analysis is also abstract: it focuses on distributions rather that relationships. The language here is often fuzzy; statistical "relationships" are discussed where one set of distributional indicators is held to predict another. This analysis is different from the analysis (statistical or otherwise) of concrete relationships among people.

[38] Simmel's work adds greatly to Weber's in this respect. Gidden's (1985a, 1985b) development of a Weberian complement to Marxist analysis almost reduces Weber to a one-dimensional theorist of power. At points Giddens focuses so predominantly on the state that social life appears as little besides economic relations with the nature (analyzed in a manner primarily indebted to Marx) and power relations.

[39] It is in the sense that Marxist often hold that Weber does not have a theory but only a highly systematic, formalized account empirical of history. In this view only the sort of abstract analysis I describe below for Marx's theory of capitalism counts as theory.

replaces labor as the source of value, this contention cannot be an account of the achievement of social integration.^[40] Planning, one of Bell's central new institutions, seems to be charged with the maintenance of social coherence. Its failures are traced to the "cultural contradictions of capitalism" (Bell 1976), not to an analysis of social relationships or their integration as such.

It is particularly unfortunate that Bell's account should exhibit this lack of attention to social integration. This absence vitiates much of the value of what is the most serious sociological attempt to come to terms with the significance of information technology. Moreover, as I have tried to show, the notion of a fundamental discontinuity between post-industrial society (or any of its myriad synonyms) and its putative precursor is misleading. Although I have described much that is new, and a lengthier treatment of technologies and social change could described a great deal more that is new, the set of conditions—especially social-structural conditions—we vaguely term modernity continues and appears likely to continue for some time to come (barring a material cataclysm of one form or another). There has been no basic shift in the form of social integration such that a new sort of society might reasonably be declared to exist. The changes that have occurred and are occurring are more or less of a piece with the changes that have occurred throughout the modern age. Indeed, a high rate of change and an expectation of change are among the defining features of modernity. Capitalism's relentless pushing is a major source of this continuous social (as well as technological, economic, and cultural) change. But it is not the only source, and at points it is resisted, so it should not be thought to contain the whole explanation.

[40] Bell (1973, xiv; 1979, 168) shares with Habermas (1970, 104), Touraine (1971), and several others the notion of an "information theory of value" that replaces the formerly applicable labor theory of value. In each case it is remarkable that no attempt is made to revise the labor theory of value so that intellectual labor can be handled better. Habermas (1976, 1984) apparently wishes to retain a narrow sense of labor so that it can be assimilated readily into his distinction between instrumental and communicative action. The desirability of a sharp distinction for Touraine may have to do with the appeal of a "new class" analysis of workers' politics in the France of 1968 and immediately thereafter. Unlike Bell, however, Touraine (e.g., 1977) and Habermas (1984) attempt to develop an account of both social relationships as such and social integration. I conceive of the present effort to be at least partially complementary to that embodied in Habermas's recent work, which distinguishes between social integration and system integration. I wish, however, to avoid "the seducements of systems theory" (McCarthy 1985) by specifying the concrete relational basis on which the tendencies toward reification and the privileging of techne over praxis arise. Rather than ceding the "world" of formal organization and large-scale social integration to systems theory, we need to examine why it is that immediate, communicative relationships cannot account for society's integration.

Marx: Abstract Totality and Social Relationships

The issue of concrete social relationships and the integration of social groupings is almost entirely suppressed in the works of Marx and most Marxists (Calhoun 1982; Alexander 1983). This suppression is partly because of an overemphasis on one of Marx's most fundamental insights: the totalizing drive of capitalist commodity production and capital accumulation.^[41] Marx recognized in a profound way that capitalism was not established on the basis of direct interpersonal relationships. It existed only through the mediation of commodities that were produced and exchanged in the pursuit of capital accumulation. Moreover, as Giddens (1985b, chapter 5) has recently reminded us, a central feature of Marx's theory is missing from competing accounts of industrial society. The missing feature is the dynamism of capitalist production and commodification, and its ceaseless expansion into new lines of production, new areas of life, and new parts of the world.^[42]

Capitalism, according to Marx, must by its nature increase its extension in the world and the intensity or completeness of its domination wherever it organizes economic activity. Capitalism drives the creation of new technology (for both production and distribution), new products, and new markets. As Giddens stresses (1985b), this analysis neglects the coeval rise of the state, which was crucial to the creation and maintenance of a distinct economic sphere. But this does not go far enough. We must both recognize the accuracy of Marx's argument for the dynamic by which capitalism

[41] This is the theme of Volume 1 of Capital insofar as it shows how abstract forms dominate concrete relationships in modern capitalism. (For Marx, one must remember, the concrete is not the simple and the obvious but the complex and in some sense arbitrary sum of many different determinations.) The commodity from encapsulates a new kind of social mediation in which people are knit together in the most determinative ways of their production and exchange of things rather than by direct interpersonal relationships; qualitative particularity is suppressed in favor of quantitative generality. Lukács, more than any other thinker, held fast to this theme—developed in his analysis of reification—as the center of Marxism ([1922] 1971, esp. the chapter titled "Reification and the Consciousness of the Proletariat"). Although he expands the notion of totality to include many other variants of Marxist holism, Jay's (1984) analysis is very helpful on this point. For Lukács, totality is not just a tendency but an essential category that is recognizable from the abstract standpoint of the proletariat as self-moving subject-object. One of the most powerful recent attempts to read Marx's theory in this way and at the level of basic categories is that of Postone (1983). In this work totalization is seen as a historically specific tendency of capitalism.

[42] "The bourgeoisie cannot exist without constantly revolutionising the instruments of production, and thereby the relations of production, and with them the whole relations of society. Conservation of the old modes of production in unaltered form was, on the contrary, the first condition of existence for all earlier industrial classes" (Marx and Engels [1848] 1976, 487).

pushes toward totality and complement it with an analysis of the concrete social relations with which capitalism (like the notion-state) coexists but which cannot be reduced to it. Capitalism, on other words, is not society. It exists in some part precisely in opposition to direct interpersonal relationships. As Marx remarked: "Their own exchange and their own production confront individuals as an objective relation which is independent of them. In the case of the world market, the connection of the individual with all, but at the same time also the independence of this connection from the individual, have developed to such a high level that the formation of the world market already at the same time contains the conditions for going beyond it" ([1939] 1973, 161, emphasis in original).^[43] Commodities confront human beings as objectifications of human activities (in relation to nature, self, and others). Commodities mediate human relationships to create the abstract totality that is capitalism. At the same time, the commodity form reifies human activity and relationships, obscuring the fact that capitalism is the product of human labor and making it appear as an independent object. Both Marx and, especially, Engels were fond of borrowing Carlyle's phrase that capitalism left no other nexus between man and man than "callous cash payment" (e.g., Marx and Engels [1848] 1976, 487; Engels [1880] 1972, 608). Just as capitalism must disregard or even attack the irreducibly qualitative nature of commodities, so it must disregard or attack the qualitative content of human relationships (Marx [1867] 1974, chapter 1; Lukáacs [1922] 1971, 83–148). But capitalism can go only so far in this attack, even in Marx's theory. Commodities tend to the purely quantitative, but they remain physical things and thus have qualities. Similarly, capitalism cannot wholly dominate or eradicate quantitative cultural contents, interpersonal relationships, or purely personal thoughts and affects.

Indeed, it is central to at least one reading of Marx that this should be

[43] "Comparison," Marx went on, takes the "place of real communality and generality: It has been said and may be said that this is precisely the beauty and the greatness of it: this spontaneous interconnection, this material and mental metabolism which is independent of the knowing and willing of individuals, and which presupposes their reciprocal independence and indifference. And, certainly, this objective connection is preferable to the lack of any connection, or to a merely local connection resting on blood ties, or on primeval, natural or master-servant relations" ([1939] 1973, 161). See the similar discussion by Engels ([1880] 1972, 627–28) and in the Manifesto (Marx and Engels [1848] 1976, 486–87). It is, however, above all in Capital, especially in the relationship between Volumes 1 and 3, that we see Marx creating a theory of a mode of totalization that makes social life appear systematically as other than it is, that is, makes capital seem the cause and not the product of human action. If we can identify capitalism with the system-world, it does not just "colonize" the life-world, as Habermas would have it, but constitutes the very severance of each from the other, the compartmentalization of the life-world and the reification of mediated action. (Whether there are other forms of mediation comparably conducive to reification remains an open question.)

so. In this reading one cannot explain the revolutionary transformation or supersession of capitalism solely on the basis of dialectical negation. That is, there must be an alternative, qualitatively separate mode or dimension of existence on the basis of which opposition to capitalism can build. Such a basis may remain outside the domination of capitalism, as in the notion, arguably suggested by Gramsci, of a counterhegemonic culture (Gramsci 1982; Boggs 1984). Or such a basis could be created by capitalism itself. Marx, for example, considered that the very concentration of workers in cities and factories (and the social organization of the factories) might provide the basis for radical mobilization (Calhoun 1983). But there is a tension between this line of reasoning in which Marx expects the coalescence of the working class as a collective actor, subjectively unified on the basis of direct relationships among workers, and Marx's more predominant analysis of how the indirect, abstract relationships of capitalism dominate and destroy direct ones.^[44] In this latter line of reasoning Marx focuses on the purely categorical position of the proletariat as the negation of capitalism; the proletariat is unified by common place in the formal relations of production rather than by qualitative relationships to each other.^[45]

The other side of capitalist totality in Marx's categorial analysis turns out, ironically, to be a kind of individualism. On the one hand, this is the universal individualism of utilitarianism, and Marx critiques aspects of it.

[44] This tension is one that does not much trouble Engels. Where Marx (in Capital ) has seen the anarchy of everyday capitalist economic activity as the superficial inverse of a relentless push toward totality, Engels uses similar language to contrast two levels of concrete social phenomena:

The old bonds were loosened, the old exclusive limits broken through, the producers were more and more turned into independent, isolated producers of commodities. It became apparent that the production of society at large was ruled by absence of plan, by accident, by anarchy; and this anarchy grew to greater and greater height. But the chief means of aid by which the capitalist mode of production intensified this anarchy of socialized production was the exact opposite of anarchy. It was the increasing organization of production, upon a social basis, in every individual productive establishment By this the old, peaceful, stable condition of things was ended. (Engels [1880] 1972, 97)

As a result, Engels can provide a more transition to an evolutionary socialist expectation that socialized production will lead naturally to proletarian solidarity and socialist society. It is this sort of Engelsian account that is revised, domesticated, and appropriated by Bell (1973) and that Galbraith (1978) and others believe they have answered by showing that modern capitalist production is not anarchic but planned, not the reckless pursuit of profit but the carefully modulated pursuit of organizational growth.

[45] In different ways both Lukács and the Althusserian structuralists extend this argument from the pure categories of capitalism even further than Marx did. See Poulantzas: "Social classes are not empirical groups of individuals, social groups that are 'composed' by simple addition; the relations of these agents among themselves are thus not inter-personal relations" (1975, 17).

But, on the other hand, Marx seems to accept this "implicit" individualism as at least a partially accurate description of reality under capitalism: concrete qualitative examples of proletarian social solidarity (such as direct interpersonal relationships as opposed to political commonality or organization) are taken by Marx as nothing other than the residues of the old order. Capitalism is purely formal, impersonal, and quantitative; the working class is unified by the commonalities of a category of individuals.^[46] If any relationships are held to be defining or productive of solidarity, they are the relationships of opposition to the bourgeoisie, the ruling class, not the relationships among workers. Nowhere does Marx endeavor to show that individuals in capitalist society (including capitalists as well as workers) are anything other than quantitatively interchangeable, except in potential.

It is important, however, to keep the issue of human social potential in mind. To the extent that capitalism is the object of analysis, direct interpersonal relationships are of minimal significance. In the pre-Capital writings where Marx envisages a communist future, however, he does not contrast quantitatively interchangeable individuals with an abstract totality. Rather he takes pains to stress that "above all we must avoid postulating 'society' again as an abstraction vis-a-vis the individual. The individual is the social being " (Marx [1844] 1975, 299, emphasis in original). But such a condition is a possible future to be created, not a timeless feature of human nature (other than in potential): "Universally developed individuals, whose social relations, as their own communal [gemeinschaftlich ] relations, are hence also subordinated to their own communal control, are no product of nature, but of history" (Marx [1939] 1973, 162). Natural law and social contract theorists, Marx says at the same point in Grundrisse, focus their attention on "merely objective" bonds among people and mistake them for the spontaneous relationships that are not possible in the existing state of society. So long as the abstract relationships of capitalism remain determinant, the analysis of concrete relationships will be the analysis of more or less arbitrary epiphenomena. When capitalism and the human self-estrangement of private property are transcended, there will still be a difference between activities that are carried out in direct communality with others and those (e.g., science) that depend less on the immediate copresence of the group but that are nonetheless self-consciously social. But each of these activities will be self-determining in a way impossible under the domination of capitalism:

Social activity social enjoyment exist by no means only in the form of some directly communal activity and directly communal enjoyment, although

[46] This approach is continued in the work of many contemporary "analytic Marxists"; see, e.g., Wright 1985.

communal activity and communal enjoyment—i.e., activity and enjoyment which are manifested and affirmed in actual direct association with other men—will occur wherever such a direct expression of sociability stems from the true character of the activity's content and is appropriate to the nature of the enjoyment. ([1848] 1976, 298, emphasis in original)

In terms of Marx's own political interests, that is, his theory of revolution, there is a tension between his account of the tendency of capitalism to eradicate all interpersonal relationships not created by capitalist commodity production and exchange and the need for a basis of social solidarity in the struggle against capitalism. Marx's few comments on interpersonal relationships other than those constituted by capitalism itself indicate that real communality would have to be postponed to a postcapitalist world. Thus Marx has no substantial theory of social integration under capitalism (as opposed to system integration, in Habermas's sense, or the integration of the capitalist totality itself). Although Marx's account of capitalism is powerful, it is an account not of the experiential or observed world of social relationships—that is, society—but of (1) a factor pushing continuously for certain directions of transformation in those relationships and (2) a form of mediation producing systematic misrecognition of those relationships. Marx's theory of capitalism is a more local or specific theory and a less universal one than is frequently claimed. It cannot be the basis for all sociology. Indeed, in the most literal sense it is not a sociology at all.^[47] Insofar as they are composed of concrete social relationships, even some of the most characteristic institutions of modern capitalism—business corporations, for example—must be explained by factors other than capitalism.

The some extent the aim of this chapter is to explore these other factors. I have offered a preliminary conceptualization of the structures of the indirect relationships that are distinctive to the modern world. Capitalism has in part helped to produce these relationships, but, at least equally, it depends on them. Corporations and large-scale markets are crucial examples. Indirect relationships have been and continue to be furthered by developments in information technology. Each relationship is also subject to a tendency toward reification, which sets it apart from the social institutions that are formed primarily on the basis of directly interpersonal relationships.

The reader should not think that only Marxism suffers from the lack

[47] Of course, taking Marx's theory of capitalism in this narrow sense excludes a good deal of Marx's own work, some of which is more directly sociological. The point is that this sociological part is not based directly on, and is still less deducible from, Marx's categorial theory of capitalism.

of a good account of the role of concrete relationships in social integration, that a good account of social integration is to be found in the whole cloth in some other theoretical corpus, or that my argument is entirely in opposition to Marxist theory. Taking the last point first, my aim is to delimit the application of the most central part of Marxist theory, treating it strictly as a formal of capitalism. Marxism is a theory of capitalism as a form of mediation among human actors (again, considered abstractly as procedures, consumers, and owners), not as a form of social or economic action (as it is for Weber). In this sense capitalism is dynamic and pushes toward totality.^[48] Although it may need improvement, with these limits Marx's theory is powerful and still offers an almost unique insight. However, I insist that capitalism is not a form of society. Marx's theory of capitalism should not be privileged as a theory of all social life. We may grant the claim that it is the tendency of capitalism as a form mediation to increase the extent to which a theory of capitalism will explain other aspects of social life, but many of these aspects nonetheless remain "other." Central among these other aspects for the purposes of this chapter are the various relationships through which members of populations are knit to each other and enable the coordination of social action on a very scale. Much the same could be said for the contents of culture.

Conclusion

I have argued that a dominant sociological trend of the modern era is the extension of social integration to an ever-larger scale, albeit with greater internal intensity, through reliance on indirect social relationships. I have suggested that new information technologies do not mark a break with this long-term trend. As material productivity continues to increase, so, too, do our capacities for organization through indirect social relationships. At the same time, social system are extended further beyond the bounds of locality, and the capacity of those empowered by them to reach into the daily lives of ordinary people is extended. Issues of information technology and control are thus central for modern sociology, but this situation does not imply any qualitative break in the kind of social processes

[48] Of course capitalism never achieves complete totality. On the basis of historical experience it does not seem to come even as close Marx imagined. We can envisage its totality by contemplating it as a purely abstract system (in the manner of Althusser), but my concern here is with concrete historical analysis. Even though the limits on the extension and intensification of capitalism have proved to be considerable, I still think it marks a qualitative break with all previous historical experience. Human organization may have seen varying and systematicity, but only capitalism has created the kind of abstract mediation that in capital accumulation not just a human desire or also an apparently independence compulsion.

at work at the most fundamental level. Modernity, if that is what we wish to to call our age, continues.^[49]

Neither Marx's theory of capitalism nor any theory of industrial society (or postindustrial society) offers an adequate account of society itself, that is, social integration. I have offered a conceptual contribution toward this end by trying to specify the distinctive nature and modern role of indirect social relationships. Of course this discussion raises other issues that I have left untouched. New information technologies may facilitate the reversal of an ancient trend toward population concentration (as least in the rich countries of the world). Will this reversal take place? With what effects? Simultaneously, the same technologies offer an increased capacity for centralized surveillance and control. Will this be checked or balanced by new means of democratic participation? What are the meanings and potentials of direct interpersonal relationships in an age in which so much of social life is constituted through indirect relationship?

Adequate answers to these and related questions depend on our ability to analyze the varying forms and extent of social integration. This analysis in turn calls for the study of concrete social relationships. Questions of social integration cannot be addressed by purely cultural analysis or through an atomistic utilitarian individualism. Nor does the Marxist theory of capitalism suffice. Despite the centrality of its insight into the dynamic pressures for change in social integration, it remains focused on the abstract, totalizing mediation of qualitative human labor (and the qualitative activity of living itself insofar as it is "consumption" or "enjoyment" of use-values) through capitalist production and the exchange of commodities. From the more structural variants of "structural-functional" thought—especially from Weber and Simmel—we may derive an approach to the study of concrete social relationships. This approach is essential to tackling the issue of social integration, a necessary, it recently neglected, counterpart to cultural accounts of modernity and Marxist accounts of capitalism. We must not, however, limit the study of concrete relationships to the direct ones that constitute the life-world while ceding all analysis of large-scale social organization to systems theory. Rather we must extend the analysis of concrete relationships into the realm of indirect relationships, showing that large-scale organizations are still part of social integration even if they are based on relationships over which participants have little control, of which they may not even be aware, and the results of which they may tend to reify.

[49] I suspect that the modern age is best characterized vaguely rather than defined precisely. In my loose-knit view certain dispositions (including both compulsions to novelty and its counterpart, nostalgia) join with the capitalist push for productivity and capital accumulation and the prevalence of indirect social relationships.

References

Abu-Lughod, J. 1969. The city is dead—long live the city: Some thoughts on urbanity . Center for Planning and Development Research Monograph no. 12. Berkeley: Center for Planning and Development.

Alexander, J. 1982. Theoretical logic in sociology . Vol. 1, Positivism, presuppositions, and current controversies . Berkeley: University of California Press.

Alexander, J. 1983. Theoretical logic in sociology . Vol. 2, The antinomies of classical social thought: Marx and Durkheim . Berkeley: University of California Press.

Badham, R. 1985. The sociology of industrial and post-industrial societies. Contemporary Sociology 32, no. 1:1–36.

Bagwell, R. 1971. The transportation revolution from 1770 . London: Batsford.

Barnes, J. A. 1954. Class and committee in a Norwegian parish. Human Relations 7, no. 1:39–58.

Bell, D. 1973. The coming of post-industrial society . New York: Basic Books.

Bell, D. 1976. The cultural contradictions of capitalism . New York: Basic Books.

Bell, D. 1979. The social framework of the information society. In The computer age: A twenty-year view, ed. M. Dertouzos and J. Moses, 163–211. Cambridge: MIT Press.

Beniger, J. 1986. The control revolution . Cambridge: Harvard University Press.

Boggs, C. 1984. The two revolutions; Gramsci and the dilemmas of Western Marxism . Boston: South End Press.

Brzezinski, Z. 1977. Between two ages: America's role in the technetronic era . Harmondsworth: Penguin.

Burnham, D. 1983. The rise of the computer state . New York: Random House.

Burt, R. 1982. Toward a structural theory of action . New York: Academic Press.

Calhoun, C. J. 1978. History, anthropology and the study of communities: Some problems in MacFarlane's proposal. Social History 3, no. 3:363–73.

Calhoun, C. J. 1980. Community: Toward a variable conceptualization for comparative research. Social History 5, no. 1:105–29.

Calhoun, C. J. 1982. The question of class struggle: Social foundations of popular protest in industrializing England . Chicago: University of Chicago Press.

Calhoun, C. J. 1983. The radicalism of tradition: Community strength or venerable disguise and borrowed language? American Journal of Sociology 88, no. 5:886–914.

Calhoun, C. J. 1986. Computer technology, large scale social integration and the local community. Urban Affairs Quarterly 22, no. 2:329–49.

Calhoun, C. J. 1988. Class, place and industrial revolution. In Class and space: The making of urban society, ed. P. Williams and N. Thrift, 51–72. London: Routledge and Kegan Paul.

Coffee, J. C., Jr. 1981. "No soul to damn: No body to kick": An unscandalized inquiry into the problem of corporate punishment. Michigan Law Review 79 (January):386–459.

Coleman, J. S. 1982. The asymmetric society . Syracuse, N.Y.: Syracuse University Press.

Cooley, C. H. [1909] 1962. Social organization . New York: Shocken.

Dan-Cohen, M. 1986. Rights, persons, and organizations: A legal theory for bureaucratic society . Berkeley: University of California Press.

Duncan, O. D. 1959. Cultural, behavioral, and ecological perspectives in the study of social organization. American Journal of Sociology 65, no. 2:132–53.

Durkheim, E. [1893] 1964. The division of labor in society . New York: Free Press.

Durkheim, E. [1895] 1966. The rules of sociological method . New York: Free Press.

Engels, F. [1880] 1972. Socialism: Utopian and scientific. In The Marx-Engels reader, ed. R. Tucker, 605–39. New York: Norton.

Forester, T., ed. 1986. The information technology revolution . Cambridge: MIT Press.

Frank, A. G. 1969. Capitalism and underdevelopment in Latin America . Rev. ed. New York: Monthly Review Press.

Frank, A. G. 1978. Dependent accumulation and underdevelopment . New York: Monthly Review Press.

Fraser, N. 1985. What's critical about critical theory? The case of Habermas and gender. New German Critique 35:97–132.

French, P. A. 1984. Collective and corporate responsibility . New York: Columbia University Press.

Galbraith, J. K. 1978. The new industrial state . 3d ed. New York: Houghton-Mifflin.

Giddens. A. 1985a. The constitution of society . Berkeley: University of California Press.

Giddens, A. 1985b. A contemporary critique of historical materialism . Vol. 2, The nation-state and violence . Berkeley: University of California Press.

Gierke, Otto von. 1934. Natural law and the theory of society . Cambridge: Cambridge University Press.

Gill, C. 1985. Work, unemployment and the new technology . Cambridge, Eng.: Polity Press.

Gramsci, A. 1982. Selections from the prison notebooks . Ed. and trans. Q. Hoare and G. N. Smith. London: Lawrence and Wishart.

Granovetter, M. 1985. Economic action and the problem of embeddedness. American Journal of Sociology 91, no. 3:481–510.

Gregory, D., and J. Urry, eds. 1985. Spatial relations and social structures . New York: St. Martin's Press.

Gunn, T. 1982. The mechanization of design and manufacturing. Scientific American 247, no. 3:115–30.

Gusfield, J. 1967. Tradition and modernity: Misplaced polarities in the study of social change. American Journal of Sociology 72, no. 3:351–62.

Gusfield, J. 1975. Community: A critical response . New York: Harper and Row.

Habermas, J. 1970. Towards a rational society . Boston: Beacon Press.

Habermas, J. 1984. The theory of communicative action . Vol. 1, Reason and the rationalization of society . Boston: Beacon Press.

Haines, V. 1985. Organicism, ecology, and human ecology. Sociological Theory 3, no. 1:65–74.

Hawley, A. 1950. Human ecology: A theory of community structure . New York: Ronald Press.

Hawley, A. 1981. Urban society . 2d ed. New York: Wiley.

Hobbes, T. [165] 1962. Leviathan . Harmondsworth: Penguin.

Innis, H. A. [1950] 1972. Empire and communication . Toronto: University of Toronto Press.

Innis, H. A. [1951] 1964. The bias of communication . Toronto: University of Toronto Press.

Jay, M. 1984. Marxism and totality: The adventures of a concept from Lukacs to Habermas . Berkeley: University of California Press.

Jones, B. 1982. Sleepers wake! Technology and the future of work . New York: Oxford University Press.

Kantorowicz. E. 1956. The king's two bodies . Princeton: Princeton University Press.

Kumar, K. 1978. Prophecy and progress: The sociology of industrial and post-industrial societies . Harmondsworth: Penguin.

Luhmann, N. 1979. Trust and power . Chichester: Wiley.

Lukács, G. [1922] 1971. History and class consciousness . Cambridge: MIT Press.

McCarthy, T. 1985. Complexity and democracy, or the seducements of systems theory. New German Critique 35:27–53.

McLuhan, M. 1964. Understanding media: The extensions of man . New York: McGraw-Hill.

Maitland, F. W. [1900] 1958. Introduction. In Political theory of the Middle Ages, by O. von Gierke, vii–xiv. Boston: Beacon Press.

Manning, C. A. W. 1962. The nature of international society . London: Bell.

Marx, K. [1844] 1975. The economic and philosophical manuscripts of 1844 . In Collected works, by K. Marx and F. Engels, 3:229–348. London: Lawrence and Wishart.

Marx, K. [1867] 1974. Capital . Vol. 1. London: Lawrence and Wishart.

Marx, K. [1939] 1973. Grundrisse . Harmondsworth: Pelican.

Marx, K., and F. Engels. [1848] 1976. Manifesto of the Communist party . In Collected works, by K. Marx and F. Engels, 6:477–519. London: Lawrence and Wishart.

Meyrowitz, J. 1985. No sense of place: The impact of electronic media on social behavior . New York: Oxford.

Mitchell, J. C., ed. 1969. Social networks in urban situations . Manchester: Manchester University Press.

Nadel, S. 1957. The theory of social structure . London: Cohen and West.

Naisbitt, J. 1982. Megatrends . New York: Warner Books.

Nisbet, R. A., and R. G. Perrin. 1977. The social bond . Rev. ed. New York: Knopf.

Noble, D. 1984. Forces of production: A social history of industrial automation . New York: Knopf.

Ong, W. 1982. Orality and literacy: The technologizing of the word . New York: Methuen.

Orhnial, T., ed. 1984. Limited liability and the corporation . London: Croom Helm.

Parsons, T. 1951. The social system . New York: Free Press.

Parsons, T. 1963. On the concept of political power. Proceedings of the American Philosophical Society 107:232–62.

Parsons, T., and G. Platt. 1973. The American university . Cambridge; Harvard University Press.

Postone, M. 1983. The present as necessity: Towards a reinterpretation of the Marxian critique of labor and time . Inauguraldissertation, Goethe-Universität, Frankfurt am Main.

Poulantzas, N. 1975. Political power and social classes . London: New Left Books.

Redfield, R. 1941. Folk culture of the Yucatan . Chicago: University of Chicago Press.

Sabato, L. 1981. The rise of the new political consultants . New York: Basic Books.

Sahlins, M. 1978. Culture and practical reason . Chicago: University of Chicago Press.

Saunders, P. 1985. Space, the city and urban sociology. In Social relations and spatial structures . ed. D. Gregory and J. Urry, 67–89. New York: St. Martin's Press.

Selznick, P. 1969. Law, society and industrial justice . New Brunswick, N.J.: Transaction Books.

Sennett, R. 1977. The fall of public man . New York: Knopf.

Simmel, G. [1903] 1971. The metropolis and mental life. In Georg Simmel on individuality and social forms, ed. D. N. Levine, 324–39. Chicago: University of Chicago Press.

Simmel, G. [1907] 1978. The philosophy of money . London: Routledge and Kegan Paul.

Sorokin, P. A. 1957. Social and cultural dynamics . Abridged ed. Boston: Porter Sargent.

Stone, L. 1975. Where the law ends . New York: Harper and Row.

Taussig, M. 1978. The devil and commodity fetishism . Chapel Hill: University of North Carolina Press.

Tönnies, F. 1887. Community and association (Gemeinschaft und Gesellschaft) . London: Routledge and Kegan Paul.

Toffler, A. 1980. The third wave . New York: Bantam.

Touraine, A. 1971. Post-industrial society . London: Wildwood House.

Touraine, A. 1977. The self-production of society . Chicago: University of Chicago Press.

Wallerstein, I. 1974. The modern world-system . New York: Academic Press.

Wallerstein, I. 1979. The capitalist world economy . New York: Academic Press.

Weber, M. [1922] 1978. Economy and society . 2 vols. Ed. Guenther Roth and Claus Wittich. Berkeley: University of California Press.

Williams, R. 1973. The country and the city . St. Albans, Eng.: Paladin.

Wirth, L. 1938. Urbanism as a way of life. American Journal of Sociology 44, no. 1:1–24.

Worsley, P. 1985. The three worlds . Chicago: University of Chicago Press.

Wright, E. O. 1985. Classes . London: New Left Books.