Durkheim's Problem and Differentiation Theory Today

Jeffrey C. Alexander

Differentiation comes closer than any other contemporary conception to identifying the overall contours of civilizational change and the texture, immanent dangers, and real promises of modern life. As a general process, differentiation is fairly well understood, and it provides a backdrop for making sense of everyday life today. Institutions gradually become more specialized. Familial control over social organization decreases. Political processes become less directed by the obligations and rewards of patriarchy, and the division of labor is organized more according to economic criteria than by reference simply to age and sex. Community membership can reach beyond ethnicity to territorial and political criteria. Religion becomes more generalized and abstract, more institutionally separated from and in tension with other spheres. Eventually, cultural generalization breaks the bonds of religion altogether. Natural laws are recognized in the moral and physical worlds and, in the process, religion surrenders not only its hierarchical control over cultural life but also its institutional prominence.

It is in terms of these general contours of world history, and the intuitive representation of modernity they provide, that the immanent dangers and promises of modernity can be understood. Thus the need to develop flexible and independent control over social complexity leads to the emergence of large-scale bureaucratic and impersonal organizations (Eisenstadt 1963). Such centralization—political, economic, informational—provides an ever-present resource for the exercise of organized cruelty and domination. Yet precisely because it is impersonal

and bureaucratic rather than primordial and diffuse, that is, because it is differentiated, this centralization is experienced, even in totalitarian societies, in important new ways. Rarely is it experienced as an all-powerful and archetypical reality; more typically it is experienced as a development that challenges the existence of deeply entrenched institutions of private and public life (see, for example, Touraine et al. 1983).

The countercenters that mark private and public life are not confined to the primary groups, or life-worlds, that Habermas (1984) presents as the last bastion against colonization by rational systems. Uneven differentiation, not one-dimensional colonization, characterizes the modern world. Indeed, as Walzer (1983) has shown, it is the very existence of social and culture differentiation—not colonization—that allows social critics who are dedicated to justice in modern societies to demand ever greater autonomy and self-control for the spheres of public and private life.

But it is not enough to know the outlines of differentiation and its problems and possibilities in general terms. If the perspective of differentiation is going to produce a theory of social change, it must be brought down to earth. Obviously, not all societies, and institutions differentiate. Sometimes they stagnate. Often they become brittle and reactionary, concentrated and inflexible. Why do these responses happen? Why, by contrast, is differentiation sometimes able to proceed?

Merely to describe differentiation as a general process, moreover, makes it appear to be automatic, an equilibrating mechanism that occurs whenever adjustments must be made to conflict and strain. This is not the case. The social processes that produce differentiation must be described in specific, concrete terms. When they are, the contingent nature of differentiation will be more clearly understood, as will the fact that differentiation takes different forms in different historical settings. Is a certain orienting ideology necessary for differentiation to occur? Are particular kinds of interest group formations necessary? If so, in what societies and historical conjunctures are such requirements likely to occur?

Finally, what is the relation between differentiation and historical formations that are the traditional objects of classical theories of social change? Do feudalism, fascism, capitalism, and socialism represent a continuum of differentiation, or do they represent amalgamations of institutions that are differentiated in varying degrees? Does thinking of change as differentiation allow us to conceptualize the strains and conflicts in these formations more effectively than traditional theories do?

These questions mark the frontier of differentiation theory. They arise not just from scientific curiosity but out of theoretical competition (Wagner and Berger 1984). They are the questions that other theories

put to theories who think they see differentiation in social change. If the theory is to be maintained, it must be improved, and these questions must be answered.

In the chapter follows I begin to formulate what some answers might be. I do this, in part, by suggesting that in the theoretical community today there is already an upsurge of investigation (for example, Alexander 1985; Alexander and Colomy 1989) directed precisely to these ends. In larger part, However, I try to provide some answers to these questions myself, or at least to produce a framework within which such answers can be more readily conceived. I begin by suggesting that the questions I have enumerated can be viewed not simply as the parochial preoccupations of recent neofunctionalist work but as issues that go back to the classical foundations of sociology itself. Indeed, I argue that, properly understood, they are generic questions that must be faced by every effort that seeks to understand social change in a serious way. I show how these questions define the achievements and limitations of Durkheim's change theory. By examining Parson's later theorizing in these terms, I argue, we gain a new handle not only on the criticisms of the functionalist theory of change but on the efforts that have been made to improve it as well. These considerations inform my suggestions, offered in the conclusion, about what future efforts at understanding differentiation might be.

1. Durkheim's Problem

Although the notion that society changes through a process of institutional specialization can be traced back to ancient times, the modern theory of social change as differentiation began with Durkheim.^[1] In The Division of Labor in Society Durkheim ([1893] 1933) put Spencer's earlier theory in a new form and started a research program that extends to the present day. Although Durkheim's first great work has, of course, become one of the classics of Western social science, the association with differentiation theory has not usually been made. In the context of the present discussion, therefore, Division is of particular interest. Although each of the problems I find in this classical work have been noted before, they have never been understood in reference to differentiation theory.

Because they have not, their theoretical interrelation has been impossible to see.

Durkheim's first great work serves as an exemplar of differentiation theory in several different ways. It can be considered the first and still one of the most powerful applications of the theory itself. It can also be seen as embodying some of this traditions's most typical and debilitating weaknesses. In other words, Durkheim's early work presents in a nutshell both the achievements of differentiation theory and the difficulties it often creates.

In book 1 of Division, titled "The Function of the Division of Labor," Durkheim outlines a general portrait of social as differentiation. Societies were once mechanically organized. They had repressive laws and were dominated by a particularistic and omnipresent collective conscience. Gradually, they have moved toward organic solidarity, where laws are restitutive and collective is generalized and abstract. In terms of institutional references Durkheim focuses on economic change on the one hand and the separation of religion from political and legal functions on the other hand. There is also a brief but important discussion of cultural generalization as indicating the increasingly person-centered character of the collective conscience.

This initial discussion, however, is of a particularly sweeping kind. Although this sweep confers power and scope, it makes it difficult to incorporate any real discussion of particulars, that is, the specific historical phases through which differentiation proceeds, the particular institutions and sectors on which distinct periods of differentiation depend, and historically specific social problems that differentiation systematically might generate. Durkheim's argument in Book 1 is evolutionary rather than development in the sense that there are no phase-specific strains outlined. It is functional in the sense that there is no theory of how particular structures are involved. It is ideal-typical in the sense that there is no account of the processes of change by which an episode of social differentiation actually occurs.^[2]

What is fascinating about this work, however, and what makes it so paradigmatic of differentiation theory as such, is that Durkheim goes on to try to supply these missing particulars in Books 2 and 3. Book 3, titled "Causes and Conditions," is his effort to supply a theory of social process. Durkheim argues that population growth leads to greater density and that greater specialization is a quasi-Malthusian response to the need for a more adaptive and efficient distribution of resources. Durkheim's Book 3, "Abnormal Forms," is an effort to discuss a particular historical

phase of differentiation and the problems it typically engenders. He suggests that because industrial society is not yet fully differentiated, the division of labor is coercive and disruptive. When birth is further separated from wealth, and political from economic organization, industrial relations will be mature and society less conflictual.

The fatal weakness of Division is that its three books cannot be related to one another in a systematic way. That demographic pressure is the principal process through which differentiation proceeds, as Durkheim asserts in Book 2, is in itself open to doubt. More significant from a theoretical point of view is that this emphasis seems to directly contradict the notion, which Durkheim argued in Book 1, that differentiation involves cultural and political phenomena. And what either demographics or systemic differentiation more generally understood have to do with the forced division of labor—Durkheim's topic in Book 3—is problematic as well. For if indeed the division of labor is anomic and coercive in 1890, there is nothing in Durkheim's general theory, or in his specific account of social process, to supply explanation for it. What is necessary is a more phase-specific model of general differentiation and of social process alike. Only with such a theory would it be possible to stipulate the criteria for predicting the "normal" and the "pathological" outcomes of a particular social formation.

To establish links between the three parts of Durkheim's work, in other words, requires a detailed account of structures and processes and a systematic effort to link these theories to the general theory of differentiation. I argue that this is precisely the goal for which contemporary differentiation theory must strive.

2. Social Change Theory and Durkheim's Problem

In order to relate this agenda for a particular research program to issues about social change more generally, one must recognize that "Durkheim's problem" was not unique to him. He used differentiation theory to grope with issues that are generic to the study of social changes as such. Each of Division's three parts represents one important way in which social change has been conceptualized: through the construction of general models, through developing accounts of social process, and through historically specific analyses of tensions and strains. Durkheim's problem, in other words, is an enduring one with which every perspective on change must come to grips.

In these terms I now briefly examine the principal classical theories of change with which Durkheim's must compete. Although Weber certainly defines a general theme, "rationalization," he does not emphasize the general level of his analysis in a way comparable to Durkheim. Weber's

only effort to produce a general account of rationalization is the "Author's Preface" (Weber [1920] 1958, 13–31) to his collected essays in the sociology of religion, which was written only at the end of his career and which was much more an afterthought than the basis for his theoretical program. The minimalist character of the rationalization theme also can be seen from the fact that a debate is still raging about the simple definition of rationalization itself.^[3] I am not suggesting that this general conception was not important for guiding Weber's thinking, for most certainly it was. But to conceptualize and elaborate it was not something with which Weber was centrally concerned.

The heart of Weber's work is his theorizing about processes of change, the role of institutions and groups in these processes, and the historically specific strains that are involved. The Protestant ethic creates capitalism in the West, patrimonialism overwhelms autonomous urban centers in the East, charismatic leadership becomes routinized and bureaucratic, priests and later legal notables have an interest in producing formally rational law. These are the middle-range propositions with which Weber is concerned. How and why these are connected to historical rationalization is implicit but never clearly spelled out. One result is that the relationship between Weber's various middle-range theories of change is never easy to see. Bendix (1961) devoted one very ambitious book to spelling out these connections, and Schluchter (1981) has recently devoted another to this same subject. But while presented as commentaries on Weber's theories, these works must actually be seen as theoretical constructions that try to fill this gap. Another result of this disarticulation of Weber's specific theories from one another and from his general perspective is that the relevance of these historical accounts for explaining other episodes of change, and for thinking about the future course of change, is far from clear.

Moreover, although Weber's historical explanations of traditional society often involve phase-specific accounts of conflict and strain—his theory of the patrimonialism-feudalism dilemma must be seen as a prototype in this regard—this genetic, or developmental, quality disappears from his treatment of the capitalist and modern periods. Again, this disarticulation between the strands of Weber's change theory leaves fundamental questions unanswered. Will bureaucratization dominate party

politics in the modern era, or will it be continuously challenged by charismatic politicians? Will formal law reign indefinitely, or will there be challenges to such formulations from different kinds of social groups, whose demands can be formulated in a substantive and historically specific way? Does the otherworldly character of Puritanism lead eventually to cultural universalism or to secularism in a purely political sense?

At the back of these problems are Weber's historicist difficulties with the concept of capitalism. Does late capitalism vitiate the processes that Weber has identified with its earlier creation? What can distinctively define late capitalism, if indeed a new postcapitalist historical phase will have to be introduced? Will this phase differ at all from the socialist form of industrial society, which at one point Weber ([1918] 1971) suggested must be seen merely as capitalism in another form, or from communist industrialism, which at another point (Beetham 1974, 46–48, 82–87) Weber believed to differ fatefully from capitalism not only in economic but in political and moral terms? Once again, my point is not that Weber has nothing to say about these issues; obviously he does. My point rather is that the failure to articulate the different levels or forms of his theorizing makes his contributions in these regards fragmentary and ad hoc. To suggest that there are paradoxes created by the rationalization of culture (Schluchter 1979) is suggestive but does not go nearly far enough. Nor is it sufficient to translate Weberian political theory into a story of the production of citizenship (Bendix 1964), even though such an effort is centainly valuable in its own right. Weber's theory remains the most perceptive theory of institutional change ever written, and it continues to inspire the most searching writing on the processes of change today (see, for example, Collins 1986b). Even for Weberian theory, however, Durkheim's problem remains.

Marxists, of course, have pointed most forcefully to these weaknesses in Weber's change theory, and when we look at Marx's approach to change, by contrast, we cannot help but admire its beauty and theoretical power. Marx united the different kinds of theorizing about social change in a coherent and compelling way. His general theme describes a dialectical movement—thesis, antithesis, synthesis—which occurs within each historical period and over the course of human history as a whole. His institutional theorizing neatly translates this dialectic by defining thesis as class domination in the service of economic production, antithesis as the struggle by classes who are exploited in production, and synthesis as the revolutionized social formation that ensues. Phase-specific strains are handled in an equally elegant and interconnected way, at least for the capitalist period: production processes rest on the forces of production; classes are established by property rights that define their relations to production; as the relations of production begin to strangle

the forces of production, class conflict begins; and equilibrium can be restored only if the revolutionary transformation of property relationships is achieved.

Because Marx seems, at least in part, to provide the solution to Durkheim's problem, his theory of change has had wide appeal. In times of great conflict and anxiety, it supplies a coherent interpretation of events. It has also clearly identified some of the most specific and obvious features of contemporary social life. That there is capitalism and class conflict cannot be denied. It is also clear that the redistribution of property continues to preoccupy capitalist welfare states, and that the twentieth century has been transformed by a series of communist revolutions Despite its intellectual power, however, Marxist change theory has, in my view, been refuted time and time again, indeed first and still most powerfully by Max Weber himself. Only when domination is experienced as intensive and relatively monolithic do Marxist theories become plausible. Insofar as social life returns to its more typically fragmented and pluralized shape, Marxism loses its attraction. We are living in such a period today. The social convulsions of the 1960s produced a renewal of Marxism but in the contemporary period Marxism is in definite decline. the centrality to change of relatively autonomous noneconomic institutions has come to be emphasized once again (see, for example, Sewell 1980; Evans, Rueschemeyer, and Skocpol 1985) and, against sweeping dialectical theories, temporal and spatial specificity has been emphasized (Giddens 1981, 1986).

As this consideration of Marxism indicates, there is more to the development of social change theories than Durkheim's problem alone. In every mode of theorizing the theorist must make specific commitments, describe empirical processes, predict conflicts, and prescribe moral possibilities. Indeed, the more explicit a theory becomes at each of the different levels of theoretical work, and the more tightly knit the interrelation it can propose, the more contestable its substantive empirical and moral commitments become. It should not be surprising that, as an advocate of more pluralistic theorizing. I find Marxism's substantive formulations implausible, even while I admire its theoretical scope. It is one thing to solve Durkheim's problem; it is quite another to solve it in an empirically and morally reasonable way. It seems to me that Weber's change theory is much closer to empirical reality than Marx's, and the moral possibilities Weber implies, although flawed in many ways, are more liberal and emancipating as well.

The challenge is to solve Durkheim's problem without giving up Weber's institutional work, which is to suggest that differentiation theory must be pushed in a Weberian direction. This was Parsons's intention. Let us see the kinds of advances he made over Durkheim's earlier theorizing

before I insist, once again, that he did not really solve Durkheim's problem at all.

3. Parsons's Change Theory and Durkheim's Problem

Parsons is generally considered, both by himself (Parsons [1960] 1967, Parsons 1971, 74, 78) and by others (Smith 1973), to have taken up differentiation theory where Durkheim left off. It is worth noting, however, that Parsons saw himself as carrying out Weber's perspective on social change as well. Although I argue that Parsons's theory is Durkheimian in its most fundamental thrust, in a certain sense Parsons's self-perception must be credited. The substantive formulations in Parsons's evolutionary writings cannibalize Weber's change theory in an extraordinary way. No one has ever taken Weber's institutional theorizing as seriously; on one has pursued the implications as strenuously or tried as hard to find a model within which they could be interrelated and explained. It is here that the paradox of Parsons's differentiation theory lies. For although Parsons finds his critical evidence and illustrations in Weber's institutional work, he never theorizes from within the institutional and processual level as such.^[4] Weber's work is grist for the mill of Parsons's improved differentiation theory, but it never threatens to displace Durkheim's approach as such.

It is good grist, to be sure. Parsons's account of change is vastly superior to Durkheim's because it can be couched in the terms that Weber provides. In Durkheim's there is sketchy generalization and, even in the most historical of his works (for example, Durkheim [1938] 1977), shifts from one historical phase to another are described in schematic terms. In Parsons's theory (1966, 1971), by contrast, differentiation is mapped in terms of actors, groups, institutions, social movements, civilizations, and states. As a result, Parsons is able to provide a much more intuitively compelling reconstruction of the modern world than Durkheim was able to provide himself. He can succeed in demonstrating what Durkheim merely suggested, namely, the extraordinary distance that has been traveled from band societies to the societies of the present day. In doing so, Parsons succeeds in legitimating the meaningful foundations of modern life.

It is impossible here to communicate the nuance and complexity of this Parsonian account, but I can make some indication of the scope and coherence of his generalized scheme. For Parsons historical evolution involves what might be called the defamilialization of the world. In band societies kinship ties define important social, cultural, and even psychological activity. Totemism is a good example. As an animal or vegetable symbol of ethnic identity and "religion." it fuses the band's existence with the natural world and with human kinship as well. It is no wonder, according to Parsons, that prohibitions like the incest taboo play such a socially decisive role, for the intermixing of kinship and social criteria makes behavior diffuse, particularistic, affective and, above all, prescriptive and ascribed. If societies are to become more flexible and individualized, they must make such "blood-related" qualities a much smaller part of social life. In order to do so, the significance of kinship must be drastically reduced.

This fused situation changes when one of the two lineages that usually form a band society seeks to improve its status. The equality of marriage exchange is altered; restricted intralineage marriage emerges and other resources are controlled as well. On the one hand, it is here that stratification and inequality arise. On the other hand, because power has itself become the basis for defining the extension of kinship ties, it marks the beginning of the possibility for more powerful and adaptive forms of social direction and control. Property comes into being, and kinship begins to be strategically subordinated to it. States are developed to protect the surplus wealth of the dominant lineage but, Parsons emphasizes, this is differentiation too, for from this point on the institutional structure of politics cannot be deduced from the nature of kinship itself.

These economic and political developments, moreover, cannot be sustained for any length of time without a religion that is far more elaborate and independent of kinship than totemism. This new religion must stretch over nonmarrying lineages and must explain and justify the social hierarchy and inequality. It does so not only by formulating a broader and more differentiated conception of the supernatural realm but also by developing a more generalized conception of "the people." Another result of the initial creation of stratification is the emergence, for the first time, of a nonfamilial conception of the societal community. There emerges a territorial referent for the human community that strongly emphasizes group as distinguished from lineage boundaries.

These processes continue in archaic and historical societies. Religion becomes more formalized and abstract. Cults emerge, as do other groups with specifically religious ambitions. Eventually churches, institutions with highly specialized religious personnel, develop. Politics continues to differentiate as well. It becomes more impersonal and bureaucratic, both

in order to gain control for the privileged class—which involves placating lower-class groups by developing primitive welfare functions—and in order to ensure the safety of larger territories and continuous productivity of economic life. Economic life becomes more functionally divided, and stratification increases. Within the now established range of "national" solidarity, heterogeneous groupings develop. They are arranged in horizontal as well as in vertically segmented ways. Although these developments ensure a more flexible and productive social organization, they also ensure new levels of hierarchy and inequality. Aristocracies represent the continued linkage of function to kinship, and new forms of domination emerge, like kingship and church, that fuse the control of various goods.

In the early modern and modern periods, primarily in the West, these intermediate levels of social development are pushed much further still. The Reformation moves religion toward a more abstract and less institutionally fused position. The emergence of parliaments and common law makes government more independent of social groups and economic position. With the advent of citizenship, social solidarity eventually becomes more independent of actual position in various spheres. The advance of universal education makes culture still more generalized and accessible, regardless of one's particular and origin. Competence rather than traditional connection or personal charisma becomes the arbiter of authority. The organization of technical knowledge through professional authority provides a systematic counterbalance to the hierarchical power derived from bureaucracies and the money power derived from markets.

Because he has one hand resting on Weber's shoulders, Parsons is able to describe the stages of differentiation with much more precision and concreteness than Durkheim himself. Even so, Durkheim's problem remains. Parsons has taken his general bearings from Durkheim, primarily from The Division of Labor, Book 1. Like Durkheim's before him, Parson's general theory does not provide an account of how change occurs. To suggest that, because a differentiated institution is more effective and flexible, it will eventually develop to cope with problems posed by other spheres says little about the actual processes by which that new and more differentiated institution actually comes about. Parsons acknowledges the imbalance. He is concerned with "the structural ordering of social data," he argues (1966, 112), not in the first instance with "the analysis of process and change." He does not seem aware, however, of the intellectual difficulties that such a position presents. His insistence (1966, 111) that "structural analysis must take a certain priority over the analysis of process and change" recalls his dogged assertion in The Social System (Parsons 1951) that the analysis of stability must precede the analysis of

change. The manifest inadequacy of this earlier claim, however, is what moved Parsons to the differentiation theory I have just described. The problem now seems to have reappeared in a new form. Even when he is committed to a theory of social change, it is the morphology of change, not its dynamics, that must come first.

But whatever Parsons's personal inclinations, this separation is impossible to make. Book 1 of Division was followed by Book 2, even though Durkheim could never connect them in an intelligible way. There is no second book for Parsons, but there is in fact an implicit strain of theorizing about what some of the actual processes of change might be. Unfortunately, the tone of this unwritten second book is Darwinian in a rather vulgar sense. Parsons himself has a more sophisticated parallel to Darwin in mind. He suggests that, like Darwin, he is justified in setting out a structural morphology of evolution without an explanation of just how evolution occurs. This was certainly true of Darwin's work. Because he did not have access to Mendel's theory of genetic mutation, he could only outline the macroconstraints within which species changed. But surely this situation does not apply to Parsons. Darwin could not set out a theory of evolutionary process; the knowledge simply was not there. When Parsons is writing, by contrast, a great deal of knowledge about the processes of social evolution already exists. Parsons chooses not to discuss it. The real parallel between Parsons and Darwin is less sophisticated. In Parsons's implicit theorizing about social change processes he tries to make do with Darwin's theory of macroconstraints alone. He takes over Darwin's theory of species competition and adaptation, which Spencer called the survival of the fittest. Even while eschewing an institutional understanding of process, therefore, there are suggestions in Parsons's work of how and why transitions from one form to another take place. This latent perspective, we will discover, allows Parsons to overlook knowledge about change processes that he prefers not to see.

For Parsons the world is an evolutionary field. Societies are species. They may die out, but innovations—breakthroughs to more differentiated phases—eventually occur. As a general theory of evolutionary change, there is nothing to fault this. The problem is that Parsons implies that it is a specific theory as well, that it is in order to adapt to an environment that breakthroughs in evolution actually occur. In presenting institutions and societies as problem solvers, Parsons's implicit second book takes a dangerous turn. In the long run adaptation may be the result of a given institutional innovation, but it is rarely its efficient cause (see Alexander and Colomy 1985). Because Parsons incorporated so many of Weber's specific and antiteleological explanations, this confusion could often be avoided. However, it cannot be denied that one

implication of his work is that adaptation and problem solving are everywhere at work (see Smelser 1985).

The results for his change theory are often disastrous. Thinking that adaptation is both cause and result provides an ideological patina for thinking about the moral implications of rationalizing change. It also hides from Parsons's understanding the full theoretical implications of his decision to ignore real processes. These ideological and theoretical difficulties come together in Parsons's sotto voce dialogue with war.

At several critical points in his evolutionary work Parsons seems to acknowledge that the transitions between phases of differentiation can be carried out by war. In his discussion of early societies, I suggested earlier, Parsons emphasizes that upper-class lineages typically depended on religious legitimation to maintain their domination, using this fact to explain the beginnings of religious generalization. He acknowledges, however, that an exception to this dependence on legitimation exists in cases "in which a group subordinates another group by military conquest" (Parsons 1966, 44). He tries to mitigate this fact in a revealing way. Although domination through conquest may have "played an important part in processes of social change," he insists, military conquest cannot be considered "differentiation in the present sense" (emphases mine). The conquerors in such situations are "a foreign group, not a structural segment of the original society." Moreover, it is "a rare, limiting case when such a group altogether eschews claims to religious legitimation and operates in terms of its naked self-interest alone." But Parsons's efforts to avoid the implications of his insight into the significance of war are beside the point. Of course domination through conquest is not differentiation; of course these conquerors are not part of domestic society but a foreign group; of course this conquering group will at some point need religious legitimation itself. None of this, however, denies the crucial fact that the transition toward a more complex society is often the result of war.

What if we know that the transition from band to stratified societies often involves political repression and ferocious violence? This does nothing to negate the fact that as the result of this transition more differentiation and flexibility occur. Nonetheless, this knowledge certainly changes our understanding of the meaning and implications of differentiation itself.

By underplaying process in his change theory, Parsons is able to deny the centrality of war in human history (see, for example, MacNeill 1982). Military conquest, of course, is not practiced only by conquering bands. Differentiated societies have experienced dark ages and the massive destruction of their civilizations as well. No matter what the innovations of a group, its survival is not assured. Even if a society is significantly more

differentiated than those around it, one of its neighbors may be developed in a direction that is, at that historical moment, much more strategically significant in military terms.

Parsons cannot see this because he confuses differentiation with adaptive success. When he cannot avoid historical disasters, he becomes whiggish in a truly embarrassing way, discussing them from the viewpoint of the comfortable present. He writes (1966, 130), for example, that "the Nazi movement, even with its immense mobilization of power, seems to have been an acute sociopolitical disturbance, but not a source of major future structural patterns." But what does "seems to have been" mean? If a repressive system is defeated on the field of battle, this does not mean that its features were less adaptive in any short-run sense. If certain contingencies had turned out differently, historians of World War II suggest, the Nazis could well have emerged victorious. Their vicious and reactionary structures would, then, certainly have established the dominant social patterns throughout Europe for an uncertain period of time.

Because he ignores processes like war, Parsons's differentiation theory cannot understand the fundamental role of backwardness and structural fusion in creating the history of the modern world. Sandwiched between his elegiac accounts of the Renaissance and Reformation, on the one hand, and his laudatory analysis of the industrial and democratic revolutions, on the other hand, one finds scarcely four pages in Parsons's book (1971, 50–54) about the Counter-Reformation and its enormous repercussions of social and cultural life. Indeed, after his analysis of the democratic revolution in France Parsons moves directly to his analysis of how the high degree of social and cultural differentiation has stabilized American and Western European nations in our time. The clear implication is that steady progress was made, that "problems" like the Counter-Reformation came up and that they were solved by cultural and institutional adaptation.

It might well be argued, however, that quite the reverse is true. It took hundreds of years to destroy the effects of the Counter-Reformation, which was itself a response to differentiation in the early modern period. Divisions were created throughout Europe, murderous and long-lived conflicts broke out between nations, and basic patterns of cultural particularism and social authoritarianism emerged. The massive wars of the twentieth century must be seen in this context. It was not adaptation through differentiation that ended the authoritarian systems whose roots lay in the reaction to the Renaissance and Reformation, it was more or less continuous war and revolution (Maier 1975). In the twentieth century war has created not just the restabilized democratic systems Parsons extols but totalitarian and repressive states as well.

By ignoring process and war, however, Parsons does not simply commit the sin of sanguinity. He also fails to generate a powerful and coherent theory of social change. In the conclusion to his work on modern societies he acknowledges (1971, 140–41) that "there has, of course, been a great deal of conflict, 'frontier' primitivism, and lag in some of the older parts of the system relative to the more progressive parts." He even allows that "certainly the history of modern systems has been one of frequent, if not continual, warfare." The conclusion that follows has about it a stunning incongruity. "The striking point," Parsons writes, "is that the same system of societies within which the evolutionary process that we have traced has occurred has been subject to a high incidence of violence, most conspicuously in war but also internally, including revolutions" (emphasis in original). As I have just suggested, of course, this striking point is exactly what Parsons's history of the modern world has not explained.

I have spent a great deal of time on the unwritten second book of Parsons's change theory. One reason is that it spells out so clearly the problems with Parsons's unwritten third. In his own third book Durkheim developed a compelling, if theoretically contradictory, account of the strains that threatened the social and moral equipoise of his time. Because Parsons emphasizes adaptation through differentiation, however, he can do nothing of the kind.

It is worth noting, I think, that this was not always the case. In what I have called the "middle period" of Parsons's work, which extended from the late 1930s to the late 1940s and resulted in a series of essays on modern society (Parsons 1954b), Parsons's writing about social change had a sharply critical edge (see Alexander 1981b; Alexander 1983, 61–71). He did not write about differentiation as such, but in the light of his later work it was clearly differentiation that he had in mind. The tensions between home and office, the discontinuous and sex-linked socialization processes this separation implied, the abstraction and rationalization of modern culture, the discipline and market-orientation of labor—these were institutional developments, which Parsons would later call differentiation, that he viewed as creating enormous problems for the modern world. They led to the distortion of gender identities and relationships, to alienation and interpersonal aggression, to harsh ethnic and racial conflicts, and indeed also to war (see, for example, Parsons 1954a). In the midst of the period from the Great Depression to fascism and world war, Parsons saw differentiation as a cause of social problems and upheavals. In the period of postwar equipoise, however, he saw differentiation as a problem-solving solution.

Because the tensions of the past are underplayed in Parsons's differentiation theory, the strains of the present cannot be displayed. In

Parsons's later account the anxiety and pathos that continue to mark the twentieth century simply fail to exist. It is not that, like Durkheim, Parsons recognizes these problems but fails to integrate his account with his general theory. It is that Parsons cannot write Durkheim's third book at all. His theory lacks a developmental notion of historically specific strains and conflicts. Thus while he plausibly argues against the feudalism-capitalism-socialism trichotomy of Marx, he does not distinguish coherent phase of his own. Parsons refers to "coming phases" of modernization and to "major changes … in process" (1971, 141–43), but aside form vague reminders about the dangers of excessive rationality and impersonality he never tells us what these phases and changes might be.

In regard to the contemporary period it appears that Parsons is not as interested in explaining changes as in changing explanations. In the closing pages of his studies on evolution he attacks the "widespread" pessimism over the survival of modern societies … especially among intellectuals," and suggests that the goal of his work should be understood in those terms, "To establish sufficient doubt of the validity of such views." Once again, there is a furtive backward glance at the tabooed subject of war. Parsons acknowledges "the undeniable possibility of overwhelming destruction." But the possibility of war in the future will not be pursued any more than its reality was pursued in the past.

Parsons sees the twentieth century as a period of opportunity and achievement, not a period of massive destruction and total war. In the last phase of his life he has become the "can-do" American pragmatist, the irrepressible evangelical utterly confident that the future will be shaped in a humane way. "Our view is relatively optimistic," Parsons concludes. The problem is that he cannot identify exactly the historical period he is optimistic about. His general theory certainly established the meaningful validity of "modern society." His inability to explain institutional process and to engage in more fully historical forms of explanation has made it impossible, however, to know whether this meaningful social framework will be able to survive.

In the midst of the Great Depression, classical economist predicted that Say's law remained valid. In the long run, they continued to maintain, demand would come back into equilibrium with supply and the slumping capitalist economies would revive. Keynes responded that in the long run we are all dead; the problem was in short run. In our own lifetimes, Keynes demonstrated, there is only partial equilibrium and Say's law does not always apply. Without confronting pathologies in the short run, even the most meaningful civilizations may not survive. Durkheim's second and third books must be written, and they must be systematically integrated with the first.

4. Theoretical Revision and Durkheim's Problem

In the polarized political climate of the 1960s and 1970s Parsons's version of differentiation theory became increasingly hard to sustain. It was challenged in the name of more historically specific and processual theorizing (for example, Nisbet 1969; Smith 1973). Theorists wanted to speak of specific events like the French Revolution (Tilly 1967) and of precise variations in national outcomes (Moore 1966). They wanted to explain specific phases and uneven development, for example, the emergence of the world capitalist system in early modern Europe (Wallerstein 1974) and the monopoly phase of capitalism (Baran and Sweezy 1966). They wanted to be able to talk about how modernization creates systemic conflicts and strains (Gusfield 1963; Gouldner 1979). Interactionists and resource mobilization theorists (Turner 1964, Gamson 1968) made claims for the centrality of social movements, and on this basis they developed explanations about the scope of change that went far beyond anything in Parson's work. Conflict theorists (Collins 1975, Skocpol 1979) developed theories of state-building and revolution that were much more historically specific and comparatively precise.

There was, moreover, a pervasive shift in ideological tone. Theories became more critical and sober about the possibility that change would take a satisfactory course. These challenges insisted that Durkheim's second and third books must be written. Eventually, Marxism drew up many of these particular theories and challenged Parson's first book as well. As I suggested above, Marxism is remarkably successful in interrelating general and specific theories of change. In the 1960s and early 1970s, a period of turbulent movements for social liberation, the elegance of Marxist theory seemed empirically compelling as well.

Empiricist philosophy of science, which continues to legitimate most social science today, holds that theories live and die through falsification. As Kuhn (1969), Lakatos (1969), and other postpositivist philosophers and historians of science have shown, however, falsification cannot—or at least in practice usually does not—disprove a general theory, even in the natural sciences. Lakatos has developed the most plausible account of how the resistance to falsification occurs (see Wagner and Berger 1985). Theoreticians differentiate between a theory's core notions, which are positions considered essential to the theory's identity, and other commitments that are more peripheral. When faced with studies that throw some of their important commitments into doubt, theorists sustain the viability of their general theories by discarding peripherals and defending the core notions. They seek to incorporate challenges by reworking and elaborating these new peripheral points. Of course this kind of defense is no more than a possibility. Whether an effective shoring-up

process actually occurs depends on the empirical actors and the social and intellectual conditions at a particular time.

When differentiation theory first encountered the challenges to its predictions and its mode of explanation, it seemed as if no successful defense would be made. Parsons himself was never able to throw the weaker points of his general change theory overboard or to expand it in an ambitious way. Faced with the choice of abandoning the theory or changing it, many functionalists simply left it behind. A theory can be abandoned even if it is not refuted, and the effect on the course of scientific development is much the same.

Some of the most important early works of Parsons's students can be seen as attempts to set the theory right by writing what should have been Parsons's second book. Smelser (1959) and Eisenstadt (1963) discussed differentiation in terms of distinctive historical events and elaborated specific processes of change; Bellah (1969) and Smelser (1963) tackled the problems of specific institutional spheres. Although these studies were important examinations of change in their own right, as theoretical revisions they did not go far enough. Eventually Smelser and Eisenstadt separated the core from the periphery of differentiation theory in a much more radical way. Eisenstadt (1964) insisted that theorizing about general differentiation was impossible in isolation from concepts of specific social processes. He showed (see Alexander and Colomy 1985) that there are particular carrier groups for particular kinds of differentiation and that the interest structures and ideological visions of these groups determine the actual course differentiation will take. He insisted on the historical and comparative specificity of differentiation and gave to civilizational factors such as culture a permanently arbitrating role (see Eisenstadt 1982, 1986). Smelser also initiated a fundamental critique from within. In his work on higher education in California he insisted that differentiation might be seen as a self-limiting process. He insisted on the resistances to differentiation and outlined a theory of the symbiotic relationship between differentiation and self-interested elites. Eventually, Smelser (1985) attacked the very problem-solving framework of Parsons's differentiation theory itself.

These revisions were intellectually powerful but they did not, at least at first, have a significant impact on debate in the field of social change. By the late 1970s this situation began to change. Several factors were involved:

1.     The glow began to fade from the more institutional and phase-oriented theories that had initiated the response to Parsons's work. Neo-Marxist theories of the world capitalist system, for example, were challenged by rising economic growth in some Third World

     nations and by the fact that the threat of imminent world economic crisis began to recede. For their part conflict perspectives appeared to have underestimated the resilience of capitalist and democratic institutions. Weber's approach to institutional process and social strain began to seem plausible once again.

2.     These developments created strains between Marxism's general theory and its more specific predictions and explanations. Ideological events, moreover, lessened the political attractiveness of not only Marxism's more sweeping conclusion but also its phase-specific theory of strains.

3.     A new generation of theorists emerged who had not personally been involved in the revolt against differentiation and, more generally, modernization theory; they did not, therefore, have a personal stake in continuing the controversy.^[5]

By the late 1970s and early 1980s the revision of differentiation theory, which have been signaled by Smeler's and Eisenstadt's work, became both more pronounced and more widespread.^[6] This work emerged in both Germany (Schluchter 1979; Luhmann 1981, 1987; Münch 1982, 1989) and the United States (Rueschemeyer 1977; Robertson 1978; Alexander 1978). These revisions proceed from the common assumption that differentiation does indeed provide an intuitively meaningful framework for understanding the nature of the modern world. But it is efforts to interrelate this general model to institutions, processes, and phase-specific strains that preoccupy most differentiation theorists in the present day.^[7]

One group of efforts has been particularly directed to the issue of phase-specific conflicts and strains. Indeed, Gould (1985) first formulated

the distinctiveness of this theoretical task in his prolegomena to a theory of social crisis, and he has concretized it in a study of the capitalist and patrimonial origins of the English Revolution (Gould 1987). Lechner (1985, 1989) has used differentiation theory to find indicators for contemporary fundamentalist movements and for structural reactions against modernity more generally. Mayhew (1984, 1989) has developed a notion of the differentiated public as corresponding to the early modern origins of capitalist society.

Other developments have been directed more to Durkheim's second book, to linking differentiation to specific theories of institutional behavior and processes of change. Champagne (1989) has formulated a complex model for explaining the failure and success of differentiation in particular American Indian societies, and Rhoades (1989) has explained why the differentiation of higher education systems has been blocked by the nationally specific organization of professional and governmental spheres. Colomy (1982, 1985, 1989a) has developed the most ambitious program in this regard. Elaborating a theory of "uneven structural differentiation," he has explained the actual paths differentiation has taken in terms of the "institutional projects" developed by strategic social groups. He explains the forces that form these projects in a systematic way and distinguishes between institutional entrepreneurs, conservatives, and accommodationists.

There still remains too large a gap between this new wave of differentiation theory and the actual strains and conflicts that characterize change in the contemporary world. Obviously, social science must separate itself from the direct preoccupations of everyday life. However, a clear and identifiable linkage must be made, especially in theorizing about social change. Only this connection anchors theorizing in the effervescence of everyday life, and only this value-relevance makes such theorizing compelling as well as true. I close with some illustrations of the linkages I have in mind.

Even in relatively developed countries, the autonomy of the societal community—its differentiation from religious, primordial, political, and economic spheres—is tentative. In liberal capitalist nations, for example, the media of mass communication are often still partly fused with political, economic, and ethnic groupings (Alexander 1980). Even when a certain autonomy is achieved, moreover, social stability may not be the result. Similarly, even in societal communities that are relatively differentiated, particularistically defined core groups continue to occupy privileged positions (Alexander 1980). Because exclusion from this core on religious, ethnic, and social class grounds remains, struggles for inclusion are not bounded episodes but are permanent and inescapable features of modern life.

It is possible to argue, in other words, that in contemporary "modern" societies differentiation still has a very long way to go. Contemporary activities in virtually every social sphere can be understood in this way. Thus, although there is no doubt that kinship and blood have vastly receded in civilization terms, the significance of gender in almost every area of modern society demonstrates that much fusion remains. Feminist movements can be seen, in these terms, as efforts to differentiate kinship and biology from evaluations of competence and hence from the distribution of economic, political, and cultural goods (see Walzer 1983, 227–42). Current struggles for workers' control and participation can be seen in much the same way. Although public governments in democratic societies have gained some independence from economic control, private governments—for example, the organization of power in factories and organizations—remain dominated by market criteria in corporate economic life (Walzer 1983, 281–312). How sharply private government can be differentiated remains to be seen, but an autonomous political and participatory sphere can certainly be extended (Siriani 1981).

To recognize that differentiation is a process that is carried by contemporary movements of social change suggests that differentiation theory needs to elaborate a conception of social polarization. Differentiation is demanded by coalitions of elites and masses, and it is opposed by other coalitions that benefit from less differentiated structural and cultural arrangements. In the course of this polarization, crises emerge (Alexander 1984). Depending on the structural setting, revolution, reform, or reaction will be the result (Alexander 1981b).

The refusal to identify differentiation with the Western status quo, and the access to a more systematic understanding of conflict that this refusal opens up, is demonstrated in the most dramatic manner when attention is shifted from the domestic to the international plane. As I have intimidated above, the emergence of more powerful and adaptive social systems not only has been stimulated by war-making but has in turn laid the basis for much more continuous, widely diffused, and deadly warfare (see Collins 1986a). Not only can the intranational causes of war become an object of differentiation theory; the international social control of war can as well. The world system is not only an economic order but also a social one. Differentiation theory suggests that social systems can control conflict only through the creation of relatively autonomous regulatory mechanisms. From this perspective the contemporary world system remains in a primitive and archaic form. Primordial solidarities are dominant and the possibilities for intrasystemic regulation are only regionally conceived. The relationship between this deficient regulatory system and war constitutes a vital but virtually unexplored topic for differentiation theory. War will be eliminated only to the

degree that the world system replicates the processes of differentiation—incomplete as they are—that have transformed the framework of national societies.

Contemporary struggles and strains need not be conceived only in terms of the structural and cultural fusions that remain. The achievement of differentiation does not do away with social problems, but rather shifts them to a different plane. Even when news media are independent, for example, they are subject to dramatic fluctuations in their trustworthiness (Alexander 1981a), and they can magnify and distort contemporary information as a result. The competition that ensues between autonomous media and other powerful institutions, moreover, generates manifold possibilities for corruption. Similar strains affect the relationship between autonomous universities and their host societies. Once the university has become committed to defending the autonomy of scientific or cognitive rationality, conflicts about the university's moral obligations to society can take on new and extraordinarily vexing forms (Alexander 1986).

In social science general theories are never disproved. Rather, like the proverbial soldiers of old, they simply fade away. For quite a few years it looked as if this would be the fate of differentiation theory. In this chapter I have argued that this not the case. I have suggested that the difficulties it has faced are the same as those encountered by every ambitious theory of social change, and after examining Durkheim's classic work, The Division of Labor in Society, I have called these difficulties "Durkheim's problem." Parsons's revisions of Durkheim's original contribution went beyond the substance of Durkheim's theorizing in many ways, but they did not overcome Durkheim's problem in a more generic sense. Indeed, in critical respects Parsons did not face this problem nearly as well. Weberian ideas have addressed this problem in important ways but have neglected other aspects at the same time. Marxism addresses Durkheim's problem most successfully of all, but its empirical implausibility, I have suggested, undermines its considerable theoretical power. In response to these difficulties, and to internally generated revisions as well, a new round of differentiation theory has begun. That it addresses Durkheim's problem more effectively is certain. Whether it can solve his problem and retain its verisimilitude remains to be seen.

References

Alexander, Jeffrey C. 1978. Formal and substantive voluntarism in the work of Talcott Parsons. American Sociological Review 43: 177–98.

Alexander, Jeffrey C. 1978. Core solidarity, ethnic outgroup and structural differentiation: Toward a multidimensional model of inclusion in modern societies. In National

and ethnic movements, ed. Jacques Dofny and Akinsola Akiwowo, 5–28. Beverly Hills: Sage.

Alexander, Jeffrey C. 1981a. The mass news media in systemic, historical, and comparative perspective. In Mass media and social change, ed. Elihu Katz and Thomas Szecsko, 17–52. Beverly Hills: Sage.

Alexander, Jeffrey C. 1981b. Revolution, reaction, and reform: The change theory of Parsons's middle period. Sociological Inquiry 5:267–80.

Alexander, Jeffrey C. 1983. Theoretical logic in sociology . Vol.4, The modern reconstruction of classical thought: Talcott Parsons . Berkeley: University of California Press.

Alexander, Jeffrey C. 1984. Three models of culture/society relations: Toward an analysis of the Watergate crisis. Sociological Theory 2:290–314.

Alexander, Jeffrey C., ed. 1985. Neofunctionalism . Beverly Hills: Sage.

Alexander, Jeffrey C. 1986. The university and morality: A revised approach to university autonomy and its limits. Journal of Higher Education 57, no. 5:463–76.

Alexander, Jeffrey C. 1987. The dialectic of individuation and domination: Max Weber's rationalization theory and beyond. In Max Weber, rationality and modernity, ed. Sam Whimster and Scott Lash. London: Allen and Unwin.

Alexander, Jeffrey C. 1989. Introduction. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 1–15. New York: Columbia University Press.

Alexander, Jeffrey C., and Paul Colomy. 1985. Towards neofunctionalism: Eisenstadt's change theory and symbolic interactionism. Sociological Theory 3:11–23.

Alexander, Jeffrey C., and Paul Colomy., eds. 1989. Differentiation theory and social change . New York: Columbia University Press.

Baran, Paul, and Paul Sweezy. 1966. Monopoly capital . New York: Monthly Review.

Beetham, David. 1974. Max Weber and the theory of modern politics . London: Allen and Unwin.

Bellah, Robert N. 1969. Beyond belief . New York: Random House.

Bendix, Reinhard. 1961. Max Weber: An intellectual portrait . New York: Doubleday Anchor.

Bendix, Reinhard. 1964. Nation-building and citizenship . New York: Doubleday Anchor.

Brubaker, Rogers. 1984. The limits of rationality: An essay on the social and moral thought of Max Weber . London: Allen and Unwin.

Champagne, Dwayne. 1989. Culture, differentiation, and environment: Social change in Tlingit society. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 52–87. New York: Columbia University Press.

Collins, Randall. 1975. Conflict sociology . New York: Academic Press.

Collins, Randall. 1986a. Sociological theory, disaster research, and war . Paper delivered at symposium, Social Structure and Disaster, 15–16 May, at College of William and Marry, Williamsburg, Va.

Collins, Randall. 1986b. Weberian sociological theory . Cambridge University Press.

Colomy, Paul. 1982. Stunted differentiation: A sociological examination of political elites in Virginia, 1720–1850. Ph.D. diss., Department of Sociology, University of California at Los Angeles.

Colomy, Paul. 1985. Uneven differentiation. In Neofunctionalism, ed. Jeffrey C. Alexander, 131–56. Beverly Hills: Sage.

Colomy, Paul. 1989a. Revision and progress in differentiation theory. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 465–95. New York: Columbia University Press.

Colomy, Paul. 1989b. Strategic groups and political differentiation in the antebellum United States. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 222–64. New York: Columbia University Press.

Durkheim, Emile. [1893] 1933. The division of labor in society . New York: Free Press.

Durkheim, Emile. [1938] 1977. The evolution of educational thought . London: Routledge and Kegan Paul.

Eisenstadt, S. N. 1963. The political system of empires . New York: Free Press.

Eisenstadt, S. N. 1964. Institutionalization and social change. American Sociological Review 29:235–47.

Eisenstadt, S. N. 1982. The axial age: The emergence of transcendental visions and the rise of clerics. European Journal of Sociology 23:294–314.

Eisenstadt, S. N. 1986. A sociological approach to comparative civilizations: The development and directions of a research program . Jerusalem: Hebrew University.

Evans, Peter B., Dietrich Rueschemeyer, and Theda Skocpol, eds. 1985. Bringing the state back in . Cambridge: Cambridge University Press.

Gamson, William. 1968. Power and discontent . Homewood, Ill.: Dorsey Press.

Giddens, Anthony, 1981. A contemporary critique of historical materialism . Vol. 1, Power, property and the state . Berkeley: University of California Press.

Giddens, Anthony. 1986. A contemporary critique of historical materialism . Vol. 2, The nation-state and violence . London: Macmillan.

Gould, Mark. 1985. Prolegomena to future theories of societal crisis. In Neofunctionalism, ed. Jeffrey C. Alexander, 57–71. Beverly Hills: Sage.

Gould, Mark. 1987. Revolution in the development of capitalism . Berkeley: University of California Press.

Gouldner, Alvin W. 1979. The new class . New York: Seabury.

Gusfield, Joseph. 1963. Symbolic crusade . Champaign-Urbana: University of Illinois Press.

Habermas, Jurgen. 1984. The theory of communicative action . Vol. 1. Boston: Beacon Press.

Kuhn, Thomas. 1969. The structure of scientific revolutions . Chicago: University of Chicago Press.

Lakatos, Imre. 1969. Criticism and the methodology of scientific research programmes. Proceedings of the Aristotelian Society 69:149–86.

Lechner, Frank. 1985. Modernity and its discontents. In Neofunctionalism, ed. Jeffrey C. Alexander, 157–76. Beverly Hills: Sage.

Lechner, Frank. 1989. Fundamentalism and sociocultural revitalization: On the logic of dedifferentiation. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 88–118. New York: Columbia University Press.

Luhmann, Niklas. 1981. The differentiation of society . New York: Columbia University Press.

Luhmann, Niklas. 1987. The evolutionary differentiation between society and interaction. In The micro-macro link, ed. J. Alexander, B. Giesen, R. Münch, and N. Smelser, 112–31. Berkeley: University of California Press.

MacNeill, William H. 1982. The pursuit of power: Technology, armed force, and society since A.D. 1000 . Chicago: University of Chicago Press.

Maier, Charles. 1975. Recasting bourgeois Europe . Princeton: Princeton University Press.

Mayhew, Leon. 1984. In defense of modernity: Talcott Parsons and the utilitarian tradition. American Journal of Sociology 89: 1273–1305.

Mayhew, Leon. 1989. The differentiation of the solidary public. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 294–322. New York: Columbia University Press.

Moore, Barrington. 1966. The social origins of dictatorship and democracy . Boston: Beacon Press.

Münch, Richard. 1982. Talcott Parsons and the theory of action. Part 2. American Journal of Sociology 87:771–826.

Münch, Richard. 1989. Differentiation, rationalization, and interpenetration: Three basic features of the emergence of modern societies. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 441–64. New York: Columbia University Press.

Nisbet, Robert. 1969. Social change and history . London: Oxford University Press.

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

Parsons, Talcott. 1954a. Certain primary sources and patterns of aggression in the social structure of the Western world. In Essays in sociological theory, by Talcott Parsons, 298–322. New York: Free Press.

Parsons, Talcott. 1954b. Essays in sociological theory . New York: Free Press.

Parsons, Talcott. [1960] 1967. Durkheim's contribution to the theory of integration of social systems. In Sociological theory and modern society, by Talcott Parsons, 3–34. New York: Free Press.

Parsons, Talcott. 1966. Societies: Evolutionary and comparative perspectives . Englewood Cliffs, N.J.: Prentice-Hall.

Parsons, Talcott. 1971. The system of modern societies . Englewood Cliffs, N.J.: Prentice-Hall.

Rhoades, Gary. 1989. Political competition and differentiation in higher education. In Differentiation theory and social change, ed. Jeffrey C. Alexander and Paul Colomy, 187–221. New York: Columbia University Press.

Robertson, Roland. 1978. Meaning and change: Explorations in the cultural sociology of modern societies . New York: Oxford University Press.

Rueschemeyer, Dietrich. 1977. Structural differentiation, efficiency, and power. American Journal of Sociology 83: 1–25.

Schluchter, Wolfgang. 1979. The paradoxes of rationalization. In Max Weber's vision of history, ed. Guenther Roth and Wolfgang Schluchter, 11–64. Berkeley: University of California Press.

Schluchter, Wolfgang. 1981. The rise of Western rationalism: Max Weber's development history . Berkeley: University of California Press.

Sewell, William H., Jr. 1980. Work and revolution in France: The language of labor from the Old Regime to 1848 . Cambridge: Cambridge University Press.

Siriani, Carmen. 1981. Production and power in a classless society: A critical analysis of the utopian dimensions of Marxist theory. Socialist Review 59:33–82.

Skocpol, Theda. 1979. States and social revolutions . New York: Cambridge University Press.

Smelser, Neil J. 1959. Social change in the industrial revolution . Chicago: University of Chicago Press.

Smelser, Neil J. 1963. The sociology of economic life . Englewood Cliffs, N.J.: Prentice-Hall.

Smelser, Neil J. 1973. Epilogue: Social structural dimensions of higher education. In The American University, ed. Talcott Parsons and Gerald Platt, 389–422. Cambridge: Harvard University Press.

Smelser, Neil J. 1974. Growth, structural change, and conflict in California higher education, 1950–1970. In Public higher education in California, ed. Neil J. Smelser and Gabriel Almond, 9–141. Berkeley: University of California Press.

Smelser, Neil J. 1985. Evaluating the model of structural differentiation in relation to educational change in the nineteenth century. In Neofunctionalism, ed. Jeffrey C. Alexander, 113–30. Beverly Hills: Sage.

Smith, Anthony. 1973. The concept of social change: A critique of the functionalist theory of change . London: Routledge and Kegan Paul.

Tilly, Charles. 1967. The Vendée . New York: Wiley.

Touraine, Alain, François Dubet, Michel Wieviorka, and Jan Strzelecki. 1983. Solidarity: The analysis of a social movement: Poland, 1980–1981 . Cambridge: Cambridge University Press and Paris: Editions de la Maison des Sciences de l'Homme.

Turner, Ralph H. 1964. Collective behavior and conflict. Sociological Quarterly 5:122–32.

Wagner, David G., and Joseph Berger. 1985. Do sociological theories grow? American Journal of Sociology 90:697–728.

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

Walzer, Michael. 1983. Spheres of justice . New York: Basic

Weber, Max. [1918] 1978. Socialism. In Max Weber: Selections in translation, ed. W. G. Runciman, 251–62. London: Cambridge University Press.

Weber, Max. [1920] 1958. Author's introduction. In The Protestant ethic and the spirit of capitalism, 13–31. New York: Scribners.