Theda Skocpol

In this part of the course we are undertaking such an analysis, which is based on Tocqueville's two classic works: Democracy in America and The Old Regime and the French Revolution. The analysis is based on Neil Smelser's book Comparative Methods in The Social Sciences. Smelser treats the Tocqueville's two books as a single study in order to analyze Tocqueville's comparative strategies.

Introduction

Chapter 1: Explaining Social Revolutions: Alternatives to Existing Theories

A Structural Perspective

International and World-historical Contexts

The Potential Autonomy of the State

A Comparative Historical Method

Why France, Russia, and China

Part I: Causes of Social Revolutions in France, Russia, and China

Chapter 2: Old-Regime States in Crisis

Old-Regime France: The Contradictions of Bourbon Absolutism

Manchu China: From the Celestial Empire to the Fall of the Imperial System

Imperial Russia: An Underdeveloped Great Power

Japan and Prussia as Contrasts

Chapter 3: Agrarian Structures and Peasant Insurrections

Peasants Against Seigneurs in the French Revolution

The Revolution of the Obshchinas: Peasant Radicalism in Russia

Two counterpoints: The Absence of Peasant Revolts in the English and German Revolutions

Peasant Incapacity and Gentry Vulnerability in China

Part II: Outcomes of Social Revolutions in France, Russia, and China

Chapter 4: What Changed and How: A Focus on State Building

Political Leadership

The Role of Revolutionary Ideologies

Chapter 5: The Birth of a "Modern State Edifice" in France

A Bourgeois Revolution?

The Effects of the Social-Revolutionary Crisis of 1789

War, the Jacobins, and Napoleon

The New Regime

Chapter 6: The Emergence of a Dictatorial Party-State in Russia

The Effects of the Social-Revolutionary Crisis of 1917

The Bolshevik Struggle to Rule

The Stanilist "Revolution from Above"

The New Regime

Chapter 7: The Rise of a Mass-Mobilizing Party-state in China

The Social-Revolutionary Situation after 1911

The Rise and Decline of the Urban-Based Kuomintang

The Communists and the Peasants

The New Regime

Conclusion

SOME BOOKS PRESENT fresh evidence; others make arguments that urge the reader to see old problems in a new light. This work is decidedly of the latter sort. It offers a frame of reference for analyzing social-revolutionary transformations in modern world history. And it uses comparative history to work out an explanation of the causes and outcomes of the French Revolution of 1787-1800, the Russian Revolution of 1917-1921, and the Chinese Revolution of 1911-1949. Developed through critical reflection on assumptions and types of explanation common to most received theories of revolution, the principles of analysis sketched in the first chapter of the book are meant to reorient our sense of what is characteristic of - and problematic about - revolutions as they actually have occurred historically. Then the remainder of the book attempts to make the program of Chapter I, calling for new kinds of explanatory arguments, come alive in application. In Part I, the roots of revolutionary crises and conflicts in France, Russia, and China are traced through analyses of the state and class structures and the international situations of the Bourbon, Tsarist, and Imperial Old Regimes. Particular emphasis is placed upon the ways in which the old-regime states came into crisis, and upon the emergence of peasant insurrections during the revolutionary interregnums. Then, in Part II, the Revolutions themselves are traced from the original outbreaks through to the consolidation of relatively stable and distinctively structured New Regimes: the Napoleonic in France, the Stalinist in Russia, and the characteristically Sino-Communist (after the mid-1950s) in China. Here special attention is paid to the state-building efforts of revolutionary leaderships, and to the structures and activities of new state organizations within the revolutionized societies.

In their broad sweep from Old to New Regimes, the French, Russian, and Chinese Revolutions are treated as three comparable instances of a single, coherent social-revolutionary pattern. As a result, both the similarities and the individual features of these Revolutions are highlighted and explained in ways somewhat different from previous theoretical or historical discussions.

Books grow in unique ways out of the experiences of their authors, and this one is no exception. The ideas for it germinated during my time as a graduate student at Harvard University in the early 1970s. This was - however faint the echoes now - a vivid period of political engagement for many students, myself included. The United States was brutally at war against the Vietnamese Revolution, while at home movements calling for racial justice and for an immediate end to the foreign military involvement challenged the capacities for good and evil of our national political system. The times certainly stimulated my interest in understanding revolutionary change. And it was during these years that my commitment to democratic-socialist ideals matured. Yet it would be a mistake to imply that States and Social Revolutions sprang immediately from day-to-day political preoccupations. It didn't. Instead it developed in the relative "ivory tower" quiet of the library and the study. As a graduate student, I pursued studies in macro-sociological theory and in comparative social and political history. Puzzles kept emerging at the interface of these sets of studies. My attempts to formulate answers to problematic issues, and then to follow answers through to their conclusions, led me, through many stages of formulation, to the arguments and analyses now embodied here.

There was, for one thing, my early intellectual confrontation with the case of South Africa. The history of that unhappy land struck me as an obvious refutation of Parsonian structure-functionalist explanations of societal order and change, and as an insuperable challenge to commonplace and comforting predictions that mass discontent would lead to revolution against the blatantly oppressive apartheid regime. Liberal justice, it seemed, did not inevitably triumph. Marxist class analysis impressed me as much more useful than structure-functionalism or relative deprivation theory for understanding the situation of the nonwhites in South Africa and deciphering the long-term tendencies of socioeconomic change. But, working strictly in terms of class analysis, it was difficult to conceptualize, let alone adequately explain, the structure of the South African state and the political role of the Afrikaners. Yet these seemed to be the keys to why no social revolution had occurred - or likely soon would - in South Africa.

Another formative experience was a lengthy, in-depth exploration of the historical origins of the Chinese Revolution. To structure my program of study, I compared and sought to explain the relative successes and failures of the Taiping Rebellion, the Kuomintang Nationalist movement, and the Chinese Communist Party, looking at all three movements in the historically changing overall context of Chinese society. Deeply fascinated by late Imperial and modem China, I came away from this research profoundly skeptical about the applicability (to China, and perhaps to other agrarian states as well) of received social-scientific categorizations such as "traditional" or '-feudal." I also became convinced that the causes of revolutions could only be understood by looking at the specific interrelations of class and state structures and the complex interplay over time of domestic and international developments.

If most other students of comparative revolutions have moved, so to speak, from the West to the East-interpreting the Russian Revolution in terms of the French, or the Chinese in terms of the Russian - my intellectual journey has been the other way around the globe. After first investigating China, I next learned about France as part of a general program of studies on the comparative political development of Western Europe. Although I realized that France was "supposed" to be like England, her absolutist Old Regime seemed in many ways similar to Imperial China. I also deciphered basic similarities in the French and Chinese revolutionary processes, both of which were launched by landed upper class revolts against absolutist monarchs, and both of which involved peasant revolts and culminated in more centralized and bureaucratic New Regimes. Finally, I came to interpret old-regime and revolutionary Russia in the same analytic terms that I had worked out for China and France. And the emphases on agrarian structures and state building seemed a fruitful way to understand the fate of this "proletarian" revolution after 1917, through, 1921 and the early 1930s.

There was yet another peculiarity worth noting about my induction into systematic research on revolutions. Unlike most sociologists who work this area, I learned a good deal about the histories of actual revolutions before I read very extensively in the social-scientific literature that purports to explain revolutions theoretically. When I did survey this literature, I quickly became frustrated with it. The revolutionary process itself was envisaged in ways that corresponded very poorly to the histories I knew. And the causal explanations offered seemed either irrelevant or just plain wrong, given what I had learned about the similarities and differences of countries that had, versus those that had not, experienced revolutions. Before long, I decided (to my own satisfaction, at least) what the fund mental trouble was: Social-scientific theories derived their explanations of revolution from models of how political protest and change were ideally supposed to occur in liberal-democratic or capitalist societies. Thus non-Marxist theories tended to envisage revolutions as particularly radical and ideological variants of the typical social (reform movement, and Marxists saw them as class actions spearheaded by the bourgeoisie or the proletariat. No wonder, I said to myself, that these theories offer so little insight into the causes and accomplishments of revolutions in predominantly agrarian countries with absolutist-monarchical states and peasant-based social orders.

From this melange of intellectual experiences, a possible project, destined to culminate in this book, presented itself to me: Use comparisons among the French, Russian, and Chinese Revolutions, and some contrasts of these cases to other countries, to clarify my critique of the inadequacies of existing theories of revolution, and to develop an alternative theoretical approach and explanatory hypotheses. Although I rejected the assumptions and substantive arguments of the theories of revolution I knew, I still had the urge to clarify the general logic that I sensed was at work across the diversely situated major revolutions I had studied. Comparative historical analysis seemed an ideal way to proceed.

To my good fortune, the three Revolutions that I wanted to include in my comparative analysis had been extensively researched by historians and area specialists. A large existing literature may be a bane for the specialist who hopes to make a new contribution based upon previously undiscovered or under-exploited primary evidence. But for the comparative sociologist this is the ideal situation. Inevitably, broadly conceived comparative historical projects draw their evidence almost entirely from "secondary sources"- that is, from research monographs and syntheses already published in book or journal-article form by the relevant historical or culture-area specialists.

The comparative historian's task - and potential distinctive scholarly contribution - lies not in revealing new data about particular aspects of the large time periods and diverse places surveyed in the comparative study, but rather in establishing the interest and prima facie validity of an overall argument about causal regularities across the various historical cases. The comparativist has neither the time nor (all of) the appropriate skills to do the primary research that necessarily constitutes, in large amounts, the foundation upon which comparative studies are built. Instead, the comparativist must concentrate upon searching out and systematically surveying specialists' publications that deal with the issues defined as important by theoretical considerations and by the logic of comparative analysis. If, as is often the case, the points debated by specialists about a particular historical epoch or event are not exactly the ones that seem most important from a comparative perspective, then the comparative analyst must be prepared to adapt the evidence presented in the works of the specialists to analytic purposes somewhat tangential to those they originally envisaged. And the comparativist must be as systematic as possible in searching out information on the same topics from case to case, even though the specialists are likely to emphasize varying topics in their research and polemics from Cn6 country to the next. Plainly, the work of the comparativist only becomes possible after a large primary literature has been built up by specialists. Only then can the comparativist hope to find at least some material relevant to each topic that must be investigated according to the dictates of the comparative, explanatory argument that he or she is attempting to develop.

As the Bibliography for this book is meant to indicate, I have been able to draw extensively upon rich literatures about France, Russia, and China. Each literature has great depth and scope, and each includes many books and articles originally published in (or translated into) English and French, the two languages that I read most easily. With occasional exceptions attributable to the thinness of interest about particular topics in one historical literature or another, the challenges I have faced have not been due to difficulties of finding basic information. Rather they have been challenges of surveying huge historical literatures and appropriately weighing and using the contributions of specialists, in order to develop a coherent comparative historical argument. How well I have met these challenges is for readers (including historians and area specialists) to judge for themselves. For myself, I shall be satisfied if this book serves in some small measure to provoke debate and inspire further investigations, both among people interested in one particular revolution or another and among people concerned to understand modern revolutions in general, their past causes and accomplishments and their future prospects. Comparative history grows out of the interplay of theory and history, and it should in turn contribute to the further enrichment of each.

Working and reworking the argument of this book over the last few years has often felt like an unending lonely struggle with a giant jig-saw puzzle. But, in actuality, many people have lent a hand, helping me to see better the overall design and pointing out where particular pieces fit, or do not. My most fundamental scholarly debt is to Barrington Moore, Jr. If was my reading of his Social Origins of Dictatorship and Democracy while was still an undergraduate at Michigan State University that introduced me to the magnificent scope of comparative history and taught me that agrarian structures and conflicts offer important keys to the patterns of modern politics. Moreover, the graduate seminars I took from Moore at Harvard were the crucibles within which my capacities to do comparative analysis were forged, even as I was allowed the space to develop my own interpretations. Moore set rigorous tasks and reacted with telling criticisms. And student fellowship in the seminars provided a supportive and intellectually lively atmosphere. In fact, two friends among fellow students in Moore's seminars, Mounira Charrad and John Mollenkopf, have given me encouragement and advice through all stages of this project on comparative revolutions.

From pp. 33-40 of the Book

"Social revolutions" as defined at the beginning of this work-rapid, basic transformations of a society's state and class structures, accompanied and in part carried through by class-based revolts from below- have been relatively rare occurrences in modern world history. Each such revolution, furthermore, has occurred in a particular way in a unique set of social-structural and international circumstances. How, then can a sociologist hope to develop historically valid explanations of social revolution as such?.

The study of social revolutions in their own right has been avoided in recent American social science because scholars believe that only phenomena of which there are a large number of cases can be studied in a truly scientific manner. There has been a self-conscious reaction against the "natural history" approach to revolutions favored by an earlier generation of American social scientists. The "natural historians," chiefly Lyford Edwards, Crane Brinton, and George Pettee, examined handfuls of cases in an attempt to develop generalizations about the typical process of revolution. Spurring this approach as too "historical," later students of revolution sought, instead, to theorize only about large numbers of cases. Thus, in the introduction to a 1964 book entitled Internal War, Harry Eckstein defines "a theoretical subject" as "a set of phenomena about which one can develop informative, testable generalizations that hold for all instances of the subject, and some of which apply to those instances alone," and he goes on to assert that whereas "a statement about two or three cases is certainly a generalization in the dictionary sense, a generalization in the methodological sense must usually be based on more; it ought to cover a number of cases large enough for certain rigorous testing procedures like statistical analysis to be used." Many other contemporary students of revolution agree with Eckstein. Consequently, the favored strategies for explaining revolutions have been premised upon subsuming them within much broader categories. These include structure-functionalist social-system categories (e.g., Chalmers Johnson) and categories such as "political violence" (e.g., Ted Gurr) or "collective action" (e.g., Charles Tilly) that refer to aspects shared by many types of political events.

It is not that contemporary analysts of revolution-subsuming phenomena see their theories as irrelevant to social revolutions. They believe, of course, that their general theories should be "applied" to instances of revolution by historians or by social scientists who do analyses of single cases. In a sense, theories such as those of Johnson, Gurr, and Tilly certainly are applicable to individual cases of social revolution: One can find relative deprivation, multiple sovereignty, and system disequilibria and value-oriented ideological movements in any and all instances of social revolution. Historians or case analysts thus could, in principle, use any or all of these ideas in a discussion of a given revolution. Indeed, because the contemporary social-scientific theories are framed in such general conceptual terms, it is very difficult to tell if they ever do not apply to a given case. What society, for example, lacks widespread relative deprivation of one sort or another? And how do we tell a synchronized social system when we see one? Ironically, theoretical approaches that set out to avoid the pitfalls of a too-historical approach to revolutions can end up providing little more than pointers toward various factors that case analysts might want to take into account, with no valid way to favor certain explanations over others.

Marxist theory works with less general, more historically grounded categories than the recent social-scientific theories, and it offers a more elegant and complete explanation of social-revolutionary transformations as such (rather than, say, political violence in general). It is thus no accident that Marxism has been the social-scientific theory most consistently and fruitfully used by historians to elucidate various particular revolutions. Yet the interaction between Marxist theory and history is incomplete because historical cases have not been used to test and modify the explanations offered by the theory. Marxist analysts have devoted themselves to high-lighting the class conflicts and changes in class relations that certainly do occur during revolutions. But they have not devised ways to test whether these factors really distinguish between revolutions and other kinds of transformations or between successful and abortive revolutionary outbreaks. Perhaps especially because the factors that they consider are indeed an important part of the story, Marxists have failed to notice a crucial point: Causal variables referring to the strength and structure of old-regime states and the relations of state organizations to class structures may discriminate between cases of successful revolution and cases of failure or nonoccurrence far better than do variables referring to class relations and patterns of economic development alone. Similarly, in their explanations of the outcomes of revolutions, Marxist-oriented scholars emphasize changes in class structures and even very long-run economic developments. But they virtually ignore the often much more striking and immediate transformations that occur in the structure and functions of state organizations such as armies and administrations, and in the relations between the state and social classes. Again, this has meant that they have missed identifying the distinctive political-institutional changes that set revolutions apart from non-revolutionary patterns of national development.

A gap of one sort or another between theory and history thus plagues both Marxist scholarship and recent academic social-science theories about revolution. Historians, especially, note the existence of this gap from time to time. Some of them complain about the vagueness of recent social-scientific theories of revolution. Others polemically assert the inappropriateness of Marxist concepts or explanations for whatever case they are concerned to analyze. Unfortunately, disillusioned historians sometimes conclude that their discipline should avoid social-scientific theories altogether. They advocate instead analyzing revolutions case by case, each in its own analytic terms, or else each in terms of the language of the actors at that time and place. In practice, no such relativist approaches are really possible, for historians must always draw, at least implicitly, upon theoretical ideas and comparative points of reference. But a hiatus of communication between historians and area specialists, on the one hand, and social theorists, on the other, is always possible. To the extent that such a hiatus exists, as it always does to some degree, it only encourages, simultaneously, the proliferation of putatively general theories of (or about) revolution that do not actually illuminate historical revolutions and an increase of specialists' accounts of particular cases that are not self-consciously informed by more general principles of analysis and explanation. The way to counter such a split, however, is not to deplore it from a vantage point above the fray. Rather, the only effective antidote is the actual development of explanations of revolutions that illuminate truly general patterns of causes and outcomes, without either ignoring or totally abstracting away from the aspects particular to each revolution and its context.

Fortunately, a method is available to aid in the development of such explanations of revolutions, at once generalizable across cases and historically sensitive. Social revolutions as such can be treated as a theoretical subject; there is no inescapable requirement to formulate explanatory hypotheses only about categories with large numbers of cases. Nor need theorists content themselves only with applying general concepts to particular cases. To generalize about social revolutions, to develop explanations of their causes and outcomes, one can employ comparative historical analysis with selected slices of national historical trajectories as the units of comparison. "Comparative history" is commonly used rather loosely to refer to any and all studies in which two or more historical trajectories nation-states, institutional complexes, or civilizations are juxtaposed. In this very broad sense, the term refers to studies with very different kinds of purposes. Some comparative histories, such as The Rebellious Century 1830-1930 by Charles, Louise, and Richard Tilly, are meant to show that a particular general sociological model holds across different national contexts. Other studies, such as Reinhard Bendix's Nation-building and Citizenship and Perry Anderson's Lineages of the Absolutist State, use comparisons primarily to bring out contrasts among nations or civilizations taken as synthetic wholes. But there is still a third version of comparative history-which I am here labeling the method of comparative historical analysis - in which the overriding intent is to develop, test, and refine causal, explanatory hypotheses about events or structures integral to macro-units such as nation-states.

Comparative historical analysis has a long and distinguished pedigree in social science. Its logic was explicitly laid out by John Stuart Mill in his A System of Logic. The method was applied to powerful effect by such classical social and historical analysts as Alexis de Tocqueville and Marc Bloch. And it continues to be elaborated and applied by contemporary scholars, including (perhaps most notably) Barrington Moore, Jr., in Social Origins of Dictatorship and Democracy. Comparative historical analysis is distinctively appropriate for developing explanations of macro-historical phenomena of which there are inherently only a few cases. This is in contrast to more plentiful and manipulable kinds of phenomena suitable for experimental investigations, and in contrast to other phenomena where there are the large numbers of cases required for statistical analyses. Comparative historical analysis is, in fact, the mode of multivariate analysis to which one resorts when there are too many variables and not enough cases.

Logically speaking, how does comparative historical analysis work? Basically one tries to establish valid associations of potential causes with the given phenomenon one is trying to explain. There are two main ways to proceed. First, one can try to establish that several cases having in common the phenomenon one is trying to explain also have in common a set of causal factors, although they vary in other ways that might have seemed causally relevant. This approach is what Mill called the "Method of Agreement." Second, one can contrast the cases in which the phenomenon to be explained and the hypothesized causes are present to other cases in which the phenomenon and the causes are both absent, but which are otherwise as similar as possible to the positive cases. This procedure Mill labeled the "Method of Difference." Taken alone, it is a more powerful method than the Method of Agreement alone for establishing valid causal associations (provided that one can find suitable negative cases for the required contrasts). In practice) though, it is often possible, and certainly desirable, to combine these two comparative logics. This is done by using at once several positive cases along with suitable negative cases as contrasts.

That will be the approach of this book. France, Russia, and China will serve as three positive cases of successful social revolution, and I shall argue that these cases reveal similar causal patterns despite their many other differences. In addition, I shall invoke negative cases for the purpose of validating various particular parts of the causal argument. In so doing, I shall always construct contrasts that maximize the similarities of the negative case(s) to the positive case(s) in every apparently relevant respect except the causal sequence that the contrast is supposed to validate. Thus, for example, the abortive Russian Revolution of 1905 will be contrasted to the successful Revolution of 1917 in order to validate arguments about the crucial contribution to social-revolutionary success in Russia of war-related processes that led to the breakdown of state repressive capacities. Moreover, selected aspects of English, Japanese, and German history will be used in various places to strengthen arguments about the causes of revolutionary political crises and peasant revolts in France, Russia, and China. These cases are suitable as contrasts because they were comparable countries that underwent non-social-revolutionary political crises and transformations in broadly similar times and circumstances to France, Russia, and China.

At first glance, comparative historical analysis may not seem so very different from the approach of the "natural historians" Lyford Edwards, Crane Brinton, and George Pettee. They, too, analyzed and compared a few historical cases in depth. Actually, however, comparative-historical and natural-history approaches to revolutions differ both in objective and in method of analysis. Whereas the goal of comparative historical analysis is to establish causes of revolutions, the natural historians sought to describe the characteristic cycle, or sequence of stages, that should typically occur in the processes of revolutions. As Robert Park put it in his introduction to Lyford Edwards's The Natural History of Revolutions,

Every social change that is capable of description in conceptual terms will have . . . its characteristic cycle. This is one of the presuppositions upon which this study is based. As a matter of scientific method, this description of the cycle seems to be the first step in the analysis of social change everywhere.

Methodologically, the natural historians analyzed revolutions by trying to fit either parts of various cases (e.g., Edwards) or a few entire cases (e.g., Brinton) to metaphors that seemed to best describe their shared stages of development, hence the sequence putatively "natural" to revolutions. Brinton, for example, explicitly employed a metaphor of disease that had also been used implicitly by Edwards:

We shall regard revolutions as a kind of fever . . . In the society during the generation or so before the outbreak of revolution . . . there will be found signs of the coming disturbance. .. They are.. [well] described as prodormal signs, indications to the very keen diagnostician that a disease is on its way, but not yet sufficiently developed to be the disease. Then comes a time when the full symptoms disclose themselves, and when we can say the fever of revolution has begun. This works up, not regularly but with advances and retreats, to a crisis, frequently accompanied by delerium, the rule of the most violent revolutionists, the Reign of Terror. After the crisis comes a period of convalescence, usually marked by a relapse or two. Finally the fever is over, and the patient is himself again, perhaps in some respects actually strengthened by the experience, immunized at least for a while from a similar attack, but certainly not wholly made over..

To be sure, the natural historians also offered, at least implicitly, some theoretical hypotheses about the causes of revolution. These were primarily social-psychological, and-the significant point for our purposes-little attempt was made to use comparisons of historical cases to validate them. Instead, the theoretical hypotheses were simply applied to the analysis as a whole, and the historical materials used primarily to illustrate the metaphorical stage sequence. The resulting natural-history analyses were certainly not without value - indeed, they offer many insights into revolutionary processes and can still be read with profit today-but they were very different from a comparative historical analysis. Such an analysis uses comparisons among positive cases, and between positive and negative cases, to identify and validate causes, rather than descriptions, of revolutions. Moreover, a comparative historical analysis does not in any way assume or attempt to argue that revolutionary processes should appear descriptively similar in their concrete trajectories from case to case. For analytically similar sets of causes can be operative across cases even if the nature and timing of conflicts during the revolutions are different, and even if, for example, 'one case culminates in a conservative reaction, whereas another does not (at all or in the same way). In a comparative historical analysis, such differences are not obstacles to the identification of similar causes across cases of revolution. At the same time, they represent variations that can themselves be explained by comparisons of the positive historical cases among themselves.

Of course, comparative history is not without its difficulties and limitations, and several especially relevant ones deserve brief discussion. There are, in the first place, inevitable difficulties in applying the method according to its given logic. Often it is impossible to find exactly the historical cases that one needs for the logic of a certain comparison. And even when the cases are roughly appropriate, perfect controls for all potentially relevant variables can never be achieved. Thus, strategic guesses have to be made about what causes are actually likely to be operative - that is, which ones could, or could not actually affect the object of study. The upshot is that there always are unexamined contextual features of the historical cases that interact with the causes being explicitly examined in ways the comparative historical analysis either does not reveal, or must simply assume to be irrelevant.

Another set of problems stems from the fact that comparative historical analysis necessarily assumes (like any multivariate logic) that the units being compared are independent of one another. But actually, this assumption is rarely if ever fully valid for macro-phenomena such as revolutions. For, as we have already noted, these phenomena occur in unique world-historical contexts that change over time, and they happen within international structures that tie societies to one another. For much of any given comparative analysis the fiction of independent units can often be maintained. Thus, for example, I am willing to treat old-regime France, Russia, and China as basically similar and unrelated agrarian states for the purposes of exploring the causes of the French, Russian, and Chinese Revolutions. But, sooner or later in most macro-analyses, one must make allowance for the unique effects of the world setting and timing, and for interrelations among the units. Thus, I shall work into my analysis the effects of the unique world-historical contexts of the eighteenth-century French versus the twentieth-century Russian and Chinese Revolutions, and I shall take into account the fact that Russian revolutionaries actually played a role in the Chinese Revolution through the transmission of Communist party models and policies via the Comintern.

Finally, it needs to be stressed that comparative historical analysis is no substitute for theory. Indeed, it can be applied only with the indispensable aid of theoretical concepts and hypotheses. For the comparative method alone cannot define the phenomenon to be studied. It cannot select appropriate units of analysis or say which historical cases should be studied. Nor can it provide the causal hypotheses to be explored. All of these must come from the macro-sociological imagination, informed by the theoretical debates of the day, and sensitive to the patterns of evidence for sets of historical cases.

Still, comparative historical analysis does provide a valuable check, or anchor, for theoretical speculation. It encourages one to spell out the actual causal arguments suggested by grand theoretical perspectives, and to combine diverse arguments if necessary in order to remain faithful to the ultimate objective-which is, of course, the actual illumination of causal regularities across sets of historical cases. Whatever the source(s) of theoretical inspiration, comparative history succeeds only if it convincingly fulfills this goal. And when it is successfully employed, comparative historical analysis serves as an ideal strategy for mediating between theory and history. Provided that it is not mechanically applied, it can prompt both theoretical extensions and re-formulations, on the one hand, and new ways of looking at concrete historical cases, on the other.

Skocpol's Case Selection

From pp. 40-42 of the book

WHY FRANCE, RUSSIA, AND CHINA?

The preceding parts of this chapter have sketched a theoretical frame of reference and introduced a method of analysis, both of which are in principle applicable to the investigation of many possible sets of social revolutions. This book does not, of course, analyze in depth all available historical cases of social revolution. Nor does it analyze a "random" sample from the entire universe of possible cases. In fact, comparative historical analysis works best when applied to a set of a few cases that share certain basic features. Cases need to be carefully selected and the criteria for grouping them together made explicit. In the following chapters, the French, Russian, and Chinese Revolutions are to be treated together as basically similar examples of successful social-revolutionary transformations. At this point, therefore, some words are in order to justify this selection of cases.

There are some important practical reasons why these social revolutions rather than others were chosen for analysis. All of them, for one thing, happened in countries whose state and class structures had not been recently created or basically altered under colonial domination. This consideration eliminates many complexities that would need to be systematically included in any analysis of revolutions in postcolonial or neocolonial settings. Furthermore, the French, Russian, and Chinese Revolutions all broke out and-after more or less protracted processes of class and political struggle-culminated in the consolidation of revolutionary state power, long-ago enough in the past to allow a study and comparison to be made of all three as entire revolutionary transformations. It is possible, in other words, to trace each Revolution from the demise of the old regime through to the emergence of a distinctively structured new regime. For comparative history, Hegel's maxim indubitably holds: The owl of Minerva flies at dusk.

Stronger reasons than these, however, are needed to explain not only why France, Russia, and China have each been selected for intense study, but also why all three have been grouped together as fundamentally similar cases of social revolution. For, according to most existing ways of defining and grouping revolutions for comparative study, France, Russia, and China simply do not belong together- certainly not all of them in one set" France was a pre-twentieth-century European revolution, typically understood as bourgeois-capitalist or liberal-democratic in nature. Depending upon one's category scheme, Russia was either an anti-absolutist revolution, or a statist-developmental revolution, or a proletarian-communist revolution. Some analysts might be willing to group it with France, others with China, but none would agree that it belongs together with both."" For China, especially, is not considered legitimately classifiable with France, either because the French Revolution was "bourgeois" or "liberal" and the Chinese obviously neither, or else because China should be grouped with Third World national-liberation revolutions and not with European revolutions of any sort.

But it is the premise of this work that France, Russia, and China exhibited important similarities in their Old Regimes and revolutionary processes and outcomes-similarities more than sufficient to warrant their treatment together as one pattern calling for a coherent causal explanation. All three Revolutions occurred in wealthy and politically ambitious agrarian states, none of which was ever colonially subjugated. These Old Regimes were proto-bureaucratic autocracies that suddenly had to confront more economically developed military competitors. In all three Revolutions, the externally mediated crises combined with internal structural conditions and trends to produce a conjuncture of: (1) the incapacitation of the central state machineries of the Old Regimes; (2) wide-spread rebellions by the lower classes, most crucially peasants; and (3) attempts by mass-mobilizing political leaderships to consolidate revolutionary state power. The revolutionary outcome in each instance was a centralized, bureaucratic, and mass-incorporating nation-state with enhanced great-power potential in the international arena. Obstacles to national social change associated with the pre-revolutionary positions of the landed upper class were removed (or greatly curtailed), and new potentials for development were created by the greater state centralization and mass political incorporation of the New Regimes.

Whatever other category systems may assume, the French and Chinese Revolutions-the two "polar" cases of my trio-were not so different from one another, nor so similar (respectively) to early European, liberal revolutions and to Third World, nation-building revolutions, as their contrasting spatio-temporal and cultural settings might suggest. The French Revolution actually was in important respects strikingly different from the English Revolution of the seventeenth century, and rather similar to the Chinese and Russian Revolutions. Peasant revolts played a key role in the process of the French Revolution, and the political result was a more centralized and bureaucratic state, not a liberal-parliamentary regime. As for the Chinese Revolution, it seems remarkably shortsighted in historical terms to regard it as a new-nation-building revolution of the mid-twentieth century. China had an imperial Old Regime with a cultural and political history stretching back many hundreds of years. And the Chinese Revolution as an entire process was launched in 1911 by an upper-class revolt against an absolute monarchical state, not unlike the aristocratic revolt that started the French Revolution.99 Furthermore, the Chinese Revolution eventually gave rise to a developmentally oriented Communist regime that is certainly as much or more similar to the post-revolutionary Soviet regime as to contemporary, noncommunist Third World governments.

Given that there are, indeed, sufficient similarities to allow these three Revolutions to be grouped together for comparative historical analysis, much is to be gained by actually doing so. The similar sociopolitical features of the French, Russian, and Chinese Revolutions can be highlighted and explained in ways that would necessarily be missed by analysts determined to keep them segregated in separate type categories. Above all, there is much to be learned from the juxtaposition of these Revolutions about the causes and results of peasant participation in social revolutions. There is also much to be learned about the dynamics of the breakdown and reconstruction of state administrative and coercive organizations from old to new regimes. It is not incidental that these aspects of revolutions tend either to be played down or assumed away by many other comparative analyses. This happens because most of the alternative category schemes serve to highlight instead either bourgeois proletarian class configurations or patterns of legitimate political authority and the ideological self-conceptions of old and new regimes.

But we shall not only emphasize the common patterns shared by the French, Russian, and Chinese Revolutions. Given the flexibility and the historical sensitivity of the comparative method, attention can also be paid to the particular features of each of the three Revolutions. There will be no need to deny that the French Revolution had bourgeois and liberal features, that the Russian Revolution was extremely statist in its outcome, or that the Chinese Revolution had in its process elements of a national-liberation struggle. For even as we primarily look for and attempt to explain patterns common to France, Russia, and China, we can also attend to the variations that characterize pairs of cases or single cases. These can then be explained as due in part to variations on the shared causal patterns, in part to contrasts among the social structures of France, Russia, and China, and in part to differences in the world-historical timing and succession of the three great Revolutions. As a result, exactly those distinctive characteristics of the Revolutions and their world-historical setting that have prompted other scholars to segregate them into separate type categories will be cast in a new explanatory light as they are studied against the background of the patterns shared by all three Revolutions.