State-Centered Approaches to Social Revolutions: Strengths and Limitations of a Theoretical Tradition

Jeff Goodwin

Center for Studies of Social Change

August, 1995

I argue in this essay that state-centered theoretical approaches are the most powerful analytic tools that are currently available to analysts of social revolutions. By contrast, fashionable poststructuralist conceptions of power simply beg too many fundamental questions. Certain types of cultural analysis, as we shall see, as well as the recent turn to "civil society" are somewhat more useful. But state-centered approaches are even more helpful for resolving the key puzzles that are distinctive to the study of social revolutions. ₁ (Throughout, I refer to state-centered approaches in the plural, because -- as I detail below --there is no single statist perspective or argument, but several overlapping ones.) Of course, state-centered approaches, like any theoretical tradition, have their blindspots and limitations, which I will also address. Fortunately, these limitations point the way toward a more powerful synthetic theory of revolutions and collective action.

What is the statist theoretical tradition all about? All of the state-centered approaches that I shall review emphasize or "center" a particular set of causal mechanisms -- namely, those processes whereby states (foreign as well as domestic) shape, enable, or constrain economic, associational, cultural, and even social-psychological phenomena. ₂ State-centered theorists argue that these mechanisms are, for certain purposes, more powerful or causally important than (or at least complementary to) a range of alternative causal processes -- for example, those emphasizing social class, civil s9ciety, culture, or social psychology. Statist perspectives, then, are intentionally one-sided.

And yet partly because of this one-sidedness, state-centered approaches are exceptionally valuable for understanding social revolutions. This follows, at least in part, from the fact that revolutions themselves are unusually state-centered phenomena. As Charles Tilly notes, "whatever else they involve, revolutions include forcible transfers of power over states, and therefore any useful account of revolutions must concern, among other things, how states and uses of force vary in time, space and social setting." ₃

I should note at the outset that I do not write as an unbiased observer. My own empirical investigations into insurgencies and social revolutions have been resolutely state-centered. ₄ I must obviously believe, then, that statist approaches to social revolutions are especially powerful! At the same time, I will try to clarify the various limitations of this perspective. ₅ After discussing the considerable strengths of state-centered approaches to social revolutions, accordingly, I will review the main weaknesses of statist analysis and suggest some of the theoretical resources that are available for redressing them. ₆ I also examine how certain strengths and limitations of state-centered approaches are exemplified in a recent case study of the Cuban Revolution.

Before discussing the analytic strengths of state-centered approaches to social revolutions, let me begin by distinguishing the distinctive forms of state-centered analysis. Understanding the variety of statist perspectives is important for appreciating both the strengths and limitations of this theoretical tradition.

Four types of state-centered analysis

A good deal of confusion has resulted from the failure of proponents and critics alike to distinguish among -- or even to note the existence of -- four distinctive versions of statist analysis, namely, the state-autonomy, state-capacity, political-opportunity, and state-constructionist approaches. Because individual states exist within an international state system, furthermore, each of these approaches has geopolitical as well as domestic dimensions.

The state-autonomy perspective, with which the others are most often conflated, emphasizes the variable autonomy of state officials or "state managers" from the dominant social class, civil society more generally, or other states. ₇ According to this perspective -- which derives from Max Weber's political sociology ₈ -- politicians, bureaucrats, and military officers may develop identities, interests, ideologies, and (ultimately) lines of action that are very different from those of organized groups in civil society or the officials of other states; they may not be usefully conceptualized, accordingly, as representatives of powerful capitalists, interest groups, the "popular will," or foreign potentates. In fact, the interests of state officials in accumulating resources (through taxes, for example) and mobilizing the population (for war against other states, for example) may sometimes conflict with the interests of powerful social groups (including the dominant class), not to mention external actors. Overt conflicts between state officials, on the one hand, and economic elites or mobilized groups, on the other, are typically adduced as evidence for this perspective.

A second statist approach -- which may also be traced to Weber emphasizes the actual material and organizational capacity (or lack thereof) of state officials to implement successfully their political agenda, even in the face of opposition from powerful actors in civil society or from other states. This perspective focuses on variations in states' fiscal resources, military power, and organizational reach (or "penetration") into civil society what Michael Mann has termed the "infrastructural power" of states. ₉ Key determinants of such variations include the organizational or bureaucratic rationality of state institutions as well as the extent to which states confront threats from other states that require war preparation; some states also receive large infusions of resources from other states. (A state's position in the international state system, in other words, may strongly shape its capacities. ₁₀ ) While this second, state-capacity approach is typically utilized along with the state-autonomy perspective, the two are analytically distinct; state officials, after all, may have very different aims than economic elites or other states and yet lack the capacity to actually implement their preferred policies. State autonomy, in other words, does not necessarily imply state capacity, or vice versa.

A third state-centered approach emphasizes how the apparent responsiveness or permeability of states or "polities" influences the ability of mobilized social groups to act collectively or substantively influence state policies. ₁₁ More specifically, "political opportunities" (which mobilized groups themselves usually attempt to create or manipulate) are deemed necessary -- in addition to (for example) grievances and organization -- for people to act collectively or to shape the agenda of state officials. ₁₂ At the very least, according to this political-opportunity perspective, the state must be either lack the means (infrastructurally speaking) or simply be unwilling to violently suppress such groups; it also helps if these groups can find powerful allies within a divided state or polity. ₁₃ And geopolitics is again important here. Some social groups, for example, may form alliances with, and receive significant resources from, foreign states; and international wars and imperial over extension have often produced political crises that have created unprecedented opportunities for political mobilization. ₁₄

There exists, finally, what Skocpol calls a "Tocquevillian" approach, which emphasizes how states shape the very identities, ideas, and organization of actors in civil society.' ₁₅ To my mind, this is perhaps the most interesting statist approach of all, yet it is often elided in discussions of state-centered theory or else conflated with the political-opportunity perspective. I propose that we label this approach the state-constructionist perspective, ₁₆ because it examines the ways in which states help to construct or constitute various aspects of civil society that are (falsely) conceptualized as wholly external to states. ₁₇ In other words, the focus here -- as against a political-opportunity approach -- is not so much on whether a state or polity provides incentives or opportunities to act for already existing networks of like-minded people; rather, state constructionism emphasizes (among other things) how the actions of foreign as well as domestic states help to make cognitively plausible and morally justifiable certain sorts of collective grievances, identities, ideologies, and associational activities (but not others) in the first place. ₁₈

Analytic strengths of state-centered approaches to revolutions

A. The Importance of State Power and State Breakdowns

How are these various theoretical approaches useful for understanding social revolutions in particular? In what follows, I emphasize how statist approaches help to resolve a series of key problems that are distinctive to the study of social revolutions.

To begin with, consider this puzzle: Why is social revolution. unlike many other forms of social conflict a peculiarly "modern" phenomenon? Why, in other words, have social revolutions occurred with considerable frequency during the nineteenth and twentieth centuries, yet don't seem to have occurred at all before the seventeenth? This puzzle concerns the "conditions of existence" of revolutions -- that is, the background conditions (which have only widely existed, evidently, for the past century or two) that are necessary for social revolutions to occur. A state-centered perspective offers a compelling solution to this puzzle: the international state system itself. In other words, no states, no revolutions. This basic proposition, frequently reiterated by Charles Tilly, is usually overlooked by analysts of revolutions; it is taken for granted by virtually all recent students of revolutions, whether by Marxists, culturalists, or state-centered analysts themselves.

From a state-centered approach, it is much more than a convention or a mere matter of convenience that scholars write books and articles about, for example, the "French," "Russian," and "Cuban" revolutions. In fact (as a state-capacity approach would suggest), prior to the emergence of consolidated national states, ₁₉ social revolutions as we now understand them -- whether as radically transformative processes, a distinctive political repertoire, or a moral ideal -- were simply impossible. Until the modern era, that is, there existed no institution with sufficient infrastructural power to remake extensive social arrangements in fundamental ways; the consolidated national state, however, made it possible to do -- and to think of doing -- just that. (Revolutionaries themselves, in fact, have often consolidated states precisely in order to remake societies.) Thus, while social conflict may be old as humanity itself, the reality and ideal of radically transforming a "society," "nation," or "people" -- the economic, political, and cultural arrangements of a large population -- is coeval with the modern state system as it originated in Europe and was then transported and emulated around the globe. ₂₀

This analysis immediately suggests a solution to another puzzle: Why are revolutionary movements, unlike other social movements concerned with "seizing" or "smashing" state power? If the preceding analysis is correct, those who would radically transform modern societies must obviously concern themselves with the state. (If they don't, the state will certainly concern itself with them!) In other words, because the state enforces (through violence if necessary) the most fundamental "rules" of a society (whether these are codified as laws or not) by virtue of its control of the principal means of coercion, any radical recasting of these rules requires access to, and indeed a fundamental reorganization of, state power itself. Because of their actual and potential infrastructural power, in other words, states are necessarily the target (although not the only target) of revolutionary movements.

This view of revolutions, I should note, is shared by state-centered and Marxist analysts alike, even though the latter are otherwise keen to emphasize how class struggles are supposedly the driving force behind them. "The basic question of every revolution," wrote Lenin, "is that of state power." ₂₁ The task of revolutionaries, in his view, was not simply to change laws or to replace government officials, but rather to change the structural characteristics of the state -- to create "an entirely different kind of power" ₂₂ with which society as a whole could then be recast. Perry Anderson similarly notes that

one of the basic axioms of historical materialism [is] that secular struggle between classes is ultimately resolved at the political -- not at the economic or cultural -- level of society. In other words, it is the construction and destruction of States which seal the basic shifts in the relations of production, so long as classes subsist. ₂₃

It follows that successful revolutionary movements must, at the very least, secure or "seize" state power. And this implies, by definition, that the old state must collapse; for if it persists in the face of a revolutionary challenge, then the revolutionaries have obviously failed to attain the sort of power that they need in order to change society as a whole in a more or less radical fashion. ₂₄

Note that we now possess the solution to yet another conundrum: Why must the state break down. collapse, or capitulate for a revolution, unlike many other forms of social protest, to succeed? That state breakdowns create the sort of political opportunities necessary for full-fledged revolutionary change is perhaps the best-known idea to emerge from statist analyses of revolution; it is a point that is central, for example, to Theda Skocpol's influential state-centered study, States and Social Revolutions. ₂₅ In fact, Skocpol not only utilizes political-opportunity approach in order to explain why transformative, classbased revolts from below could occur in France, Russia, and China; she also employs a state-autonomy perspective in order to explain the political crises that created such opportunities in the first place. Indeed, one of the more interesting claims of Skocpol's study is that the political crises that made revolutions possible in France, Russia, and China were not brought about by revolutionaries; rather, conflicts between dominant classes and autonomous state officials -- conflicts, Skocpol emphasizes, that were produced or exacerbated by geopolitical competition --directly or indirectly brought about such crises, thereby opening up opportunities that rebellious lower classes and self-conscious revolutionaries seized, sometimes years later.

By illuminating the origins of, and the political opportunities created by, these sorts of state crises and breakdowns, state-centered approaches help to resolve yet another classic puzzle: Why do social revolutions occur when and where they do? It has become virtually obligatory for scholars to note that people are not often rebellious in the poorest of societies or during the hardest of times; and even where and when people are rebellious, and strong revolutionary movements form, they are not always able to seize state power -- unless, as we have seen, they are able to exploit the political opportunities opened up by state breakdowns.

The limited utility of poststructuralist conceptions of power, ₂₆ at least for the analysis of revolutions, should now be apparent. In fact, any view of power as "decentered," largely nonviolent, local, mobile, and ubiquitous fails to grasp the crucial difference that centralized state power (and its breakdown) makes for a variety of social processes, including social revolutions Furthermore, the notion that the state itself is simply the "institutional crystallization" or "institutional integration of power relationships" that are fundamentally local in nature ₂₇ fails to grasp the potential autonomy and distinctive capacities of states; it also underestimates the role of state power in constructing, or reconstructing, localized power relationships in the first place.

B. The Formation of Revolutionary Movements

And yet, of course, state power and its breakdown cannot alone explain (or predict) social revolutions; analysts also need to explain why and how specifically revolutionary movements are able to take advantage of these crises and actually seize power. ₂₈ After all, an organized revolutionary movement simply may not exist or possess the sufficient leverage or "hegemony" within civil society that is necessary to take advantage of extant political opportunities. In such cases, state power will be reconsolidated -- if it is reconsolidated at all -- by surviving factions of the old regime or by political moderates who eschew any radical transformation of the state or society.

Here again, I would propose, a state-centered perspective provides us with some indispensable analytic tools. For although statist approaches (as we shall see) do not adequately theorize collective action as such (radical or otherwise), they. are particularly helpful in resolving the following puzzle: Why are groups with a specifically revolutionary agenda or ideology. as well as a radical or high-risk strategic repertoire. sometimes able to attract broad popular support? ₂₉ A state-centered perspective suggests that at least five distinctive state practices or characteristics help to engender hegemonic revolutionary movements. These traits are causally "cumulative," in the sense that a hegemonic revolutionary movement is more likely to develop the more such traits characterize a given state:

1. State sponsorship or protection of unpopular economic and social arrangements In certain societies, economic and social arrangements -- particularly those involving people's work or livelihood -- may be widely viewed as unjust (that is, as not simply unfortunate or inevitable). Yet unless state officials are seen to sponsor or protect those arrangements, specifically revolutionary collective action is unlikely. People may blame their particular bosses or superiors for their plight, for example, or even whole classes of bosses, yet the state itself may not be challenged -- even when the aggrieved are well organized and the political context is opportune -- unless there exists a palpable symbiotic relationship between the state and these elites. (Of course, the fact that the state must actively protect certain institutions will itself serve, in many instances, to delegitimate them.)

In this sense, "ruling classes" that do not directly rule may be safer than those that do; other things being equal, in other words, some measure of state autonomy from the dominant economic class may act as a bulwark against revolution. In such contexts, contentious, anti-elite actions may be chronic, in such forms as pilfering, malingering, sabotage, riots, strikes, and demonstrations; yet such actions are unlikely to escalate beyond a local or, at most, regional level in a way that would seriously and directly threaten a strong state. ₃₀ And yet rebels are not revolutionaries, we have seen, unless they seriously contend for state power.

In sum, grievances may only become "politicized" (that is, framed as resolvable only at the level of the state), and thereby a basis for specifically revolutionary collective action, when the state sponsors or protects economic and social conditions that are viewed as grievous. Note that this is a state-constructionist argument: State practices, in this case, help to construct a distinctive political orientation among aggrieved groups in civil society.

2. Exclusion of mobilized groups from state power or resources. Even if aggrieved groups direct their claims at the state, they are unlikely to seek its overthrow (or radical reorganization) if they manage to attain some significant share --7 or believe they can attain such a share -- of state power or influence. Indeed, even if such groups view their political influence as unfairly limited, their access to state resources or inclusion in policy-making deliberations -- unless palpably cosmetic -- will likely prevent any radicalization of their strategic repertoire or guiding ideology. In fact, the political "incorporation" of mobilized groups -- including the putatively "revolutionary" proletariat -- has typically served to deradicalize them. ₃₁ For such groups often view this sort of inclusion as the first step in the accumulation of greater influence and resources; in any event, they are unlikely to jeopardize their relatively low-cost access to the state -- unless that state itself is in deep crisis -- by engaging in "disloyal" or even illegal activities.

Political inclusion also discourages the sense that the state is unreformable or an instrument of a narrow class or clique and (accordingly) needs to be fundamentally overhauled. Tocqueville emphasized how the exclusionary nature of French absolutism bred a political culture among intellectuals characterized by a utopian longing for total revolution -- even though French social conditions were comparatively benign by European standards of the time, ₃₂

Accordingly, neither open, democratic polities nor authoritarian yet inclusionary (for example, "populist") regimes have generally been challenged by powerful revolutionary movements. By contrast, chronic exclusion of mobilized groups from access to state power is likely to push them toward a specifically revolutionary strategy or repertoire -- that is, militant, extralegal, and even armed struggle aimed at overthrowing the state. Such exclusion, after all, serves as an object lesson in the futility of legalistic or constitutional politics (i.e., "playing by the rules"). Exclusionary authoritarian regimes, accordingly, tend to "incubate" radical collective-action: Those who specialize in it tend to prosper, because they come to be viewed by many people as more realistic and potentially effective than political moderates, who themselves come to be viewed as hopelessly ineffectual. ₃₃ Partly for this reason, virtually all of the powerful revolutionary movements of the present century developed under exclusionary regimes, including the Bolsheviks in Russia, the Communists in China and Vietnam, Castro's July 26th Movement in Cuba, the broad coalition that opposed the shah in Iran, and the guerrilla movements of Central America.

Note that this point has both political-opportunity and state-constructionist aspects. In the former sense, this argument emphasizes how the lack of routine opportunities to influence state policy tends to push certain groups and individuals towards radical politics; in the latter sense, emphasis falls on the ways in which exclusionary state practices reinforce the plausibility and justifiability of a radical political orientation or identity.

3. Indiscriminate but not overwhelming, state violence against mobilized groups and oppositional political figures. Like political exclusion, indiscriminate state violence against mobilized groups and oppositional figures is likely to reinforce the plausibility, justifiability, and (hence) diffusion of the idea that the state needs to be violently "smashed" and radically reorganized. For reasons of simple self-defense, in fact, people who are literally targeted by the state may arm themselves or join groups that have access to arms. Unless state violence is simply overwhelming, then (see below), indiscriminate coercion tends to backfire, producing an ever-growing popular mobilization by armed movements and an even larger body of sympathizers. ₃₄ Revolutionary groups may thus prosper not so much because of their ideology per se, but simply because they can offer people some sort of protection from certain sorts of states. Many studies of particular revolutions emphasize that groups have only turned to militant strategies or armed struggle after their previous efforts to secure change through legal means were violently repressed. ₃₅

Like political exclusion, indiscriminate state violence also reinforces the plausibility and diffusion of specifically revolutionary ideologies, that is, ideologies that envisage a radical reorganization not only of the state, but of society as well. After all, a society in which aggrieved people are routinely denied an opportunity to redress perceived injustices, and even murdered on the mere suspicion of political disloyalty, is unlikely to be viewed as requiring a few minor reforms; those people are more likely to view such a society as in need of a fundamental reorganization. In other words, violent, exclusionary regimes often unintentionally foster the hegemony of their most radical social critics -- religious zealots, virtuous ascetics, socialist militants, and radical nationalists, for example, who view society as more or less totally corrupted, incapable of reform, and thus requiring a thorough and necessarily violent reconstruction. ₃₆

4. Weak policing capacities and infrastructural power. Of course, as the political-opportunity approach emphasizes, no matter how iniquitous or authoritarian a state may be -- or the society which it rules -- it can always retain power so long as it is capable of ruthlessly repressing its enemies. Such a state may in fact have many enemies -- including revolutionary foes -- yet they will prove quite ineffective so long as the state's coercive might remains overwhelming.

Long before a state breakdown, however, revolutionaries may become numerous and well organized if the state's policing capacities and infrastructural power more generally are chronically weak or geographically uneven. Guerrilla movements, for example, have typically prospered in peripheral and especially mountainous areas where state control is weak or nonexistent: The Communist movement in China grew strong in the northwest periphery, Castro's movement in Cuba's Sierra Madre, and El Salvador's guerrilla armies in that country's mountainous northern departments. ₃₇ And revolutionaries are doubly fortunate if they confront states and armies that are ineffectual due to corruption or bureaucratic incoherence, which are often purposively fostered by ruling cliques or autocrats who fear palace coups. ₃₈ In such situations, revolutionaries themselves may bring about or accelerate; state breakdowns not only through direct military pressure, but also by exacerbating conflicts between states (especially personalistic dictatorships) and dominant classes, and between states and their foreign supporters. These sorts of conflicts, in addition to the general insecurity that revolutionary situations engender, may also accelerate state breakdowns by creating economic downturns that bring on fiscal crises for states. ₃₉

5. Corrupt and arbitrary personalistic rule that alienates. weakens or divides civilian and military elites. As these last remarks suggest, autocratic and so-called neopatrimonial dictatorships are especially vulnerable to revolution. In fact, such regimes not only tend to facilitate the formation of hegemonic revolutionary movements, but cannot easily defeat such movements once they have formed; examples include the absolutist monarchies in France, Russia, and Iran, and the dictatorships of Diaz in Mexico, Chiang in China, Batista in Cuba, Somoza in Nicaragua, and Ceaupescu in Romania. ₄₀ As especially narrow and autonomous regimes, such dictatorships not only have few fervid supporters, but also the discretionary power that may alienate certain state officials and military officers as well as vast sectors of society, including middle strata and elites in addition to lower classes. In fact, because dictators often view economic and military elites as their chief foes, they may attempt to weaken and divide them in various ways, even though such groups share with dictators a counter-revolutionary outlook. By weakening counterrevolutionary elites, however, dictators may unwittingly play into the hands of revolutionaries, since such elites thereby become too weak either to effectively oppose the revolutionaries or to oust the dictator and reform the regime, thereby preempting revolution.

Of course, not all dictators are equally adept at controlling their armed forces and rival elites; their incompetence or incapacity in this regard does not bode well for them personally, but it may prove decisive in preempting social revolution. For if civilian and military elites can remove corrupt and repressive dictators, and perhaps institute democratic reforms, they thereby undermine much of the appeal of revolutionaries; in fact, this is precisely what happened in the Philippines with the ouster of Ferdinand Marcos. ₄₁

In sum, certain sorts of states are not only liable to break down and thereby to create the sort of political opportunities that strong revolutionary movements can exploit; certain states also unintentionally foster the very formation and indeed hegemony of radical movements by politicizing popular grievances, foreclosing possibilities for peaceful reform, compelling people to take up arms in order to defend themselves, making radical ideologies and identities plausible, providing the minimal political space that revolutionaries require to organize disgruntled people, and weakening counterrevolutionary elites, including their own officer corps. By thus illuminating both state breakdowns and processes of revolutionary mobilization, I hope to have shown, state-centered approaches comprise our most powerful tools for explaining social revolutions.

Some common criticisms of state-centered approaches

Of course, like any theoretical tradition, the statist perspective has its share of critics. The various complaints that they have directed against this tradition, however, are very uneven in their persuasiveness. Before turning to some of the more potent criticisms of statist analysis, I want to examine several that either rest upon unfounded assumptions or are simply unconvincing. Four such criticisms merit a brief response:

1. "Societies affect states as much as, or possibly more than states affect societies." ₄₂ This thesis challenges one combination of the state-autonomy and state-capacity approaches, namely, the view that all states are autonomous from civil society and actually have the capacity to implement their preferred policies. ₄₃ This is certainly a view worth challenging, but it is not clear that many state-centered theorists would defend it. In fact, state-centered theorists have generally emphasized that state autonomy and capacities are potential and variable rather than "given" a priori. ₄₄ As we have seen, moreover, statist analysts have emphasized precisely how state breakdowns, as well as infrastructurally weak states, have encouraged important social processes, including the formation of revolutionary movements.

This criticism also seems to confuse state-centered analysis with a sort of sweeping political determinism that robs "society" of any analytic autonomy whatsoever. ₄₅ But a state-centered perspective hardly implies that states are the only institutions that matter or that states themselves are not potentially shaped and constrained by a variety of social factors. In fact, it is possible and sometimes desirable to combine or complement a state-centered analysis with, for example, class analysis. ₄₆

2. State officials are usually not autonomous actors; instead they typically respond to the demands of the dominant class or (occasionally) of militant lower classes This criticism of state-centered perspectives -- the principal one expressed, of course, by Marxists ₄₇ -- is a narrower version of the preceding one, emphasizing how specifically class-based demands determine state policies. Like the previous criticism, therefore, this one only challenges one extreme version of the state-autonomy approach, namely, the idea that all states are autonomous from the demands of social classes (and, accordingly, are never influenced by such classes). Again, this is a claim that few if any statists would wish to make; it seems more reasonable, in fact, to assume that the relationship between states and classes is in fact quite variable over space and time.

Two other points about this criticism also need to be made. First, it has usually been raised in the context of complex, detailed debates about the relative importance of class and state actors in formulating specific state policies. ₄₈ These debates, whichever side one finds most convincing, hinge upon the marshalling and interpretation of particular facts and sequences of events. Neither side, including those who emphasize the importance of class actors, has suggested that their opponents must be wrong a priori, irrespective of the actual historical record. The theoretical grounds for believing that states may be autonomous from class forces, in other words, have not been seriously challenged in these debates; what is disputed is the relative autonomy of particular state actors in specific times and places (e.g., Democratic politicians in the U.S. Congress during the 1930s).

Second, even in those cases in which the class-based character of state policies has been convincingly established, it would be quite unfortunate to dismiss or ignore state-centered perspectives on that account. In fact, state autonomy may very well explain why such policies were adopted in the first place. ₄₉ The state-capacity approach, furthermore, may be helpful for understanding which, if any, class-based policies can actually be implemented. The political-opportunity perspective, furthermore, may be helpful for understanding whether other classes or groups can successfully mobilize against such policies. (In this regard, it may make a great deal of difference whether individual capitalists are simply acting in similar ways or the state is enforcing -- with violence, if necessary -- certain laws or policies at their behest.) And a state-constructionist analysis may be helpful for understanding why specifically class-based actors are politically organized and influential.

3. As a type of "structuralism," state-centered analysis necessarily neglects the purposive (including strategic) and cultural dimensions of social action. The conflation of state-centered analysis with the sort of "structuralism" that denies the importance of purposive human agency would seem to rest upon an elementary confusion. ₅₀ In fact, statist analysis may emphasize the actions and policies of state actors just as much as the impersonal "structural" characteristics of states (and both are undoubtedly important). For example, rationally calculating (and acting) state officials are the analytic pivot in some types of state-centered studies. ₅₁

The criticism that state-centered analysis fails to treat culture seriously is only partially correct. ₅₂ (I discuss the sense in which it is accurate in the next section.) While most statist analyses have in fact been "structuralist" or "instrumentalist" in the sense of neglecting the shared beliefs of politicians and state officials, this quality seems fortuitous rather than inevitable. ₅₃ So, for example, one important state-centered study, James M. Jasper's Nuclear Politics: Energy and the State in the United States. Sweden. and France, ₅₄ emphasizes precisely the ways in which the ideologies and "policy styles" of state officials shape state policies. Jasper's study is no less state-centered for treating such officials as cultural actors rather than as rational calculators or as puppets of external forces. As Jasper's study emphasizes, moreover, state practices are "always already" cultural practices. ₅₅

It is also possible, as Robert Wuthnow has convincingly shown, to explain the diffusion and institutionalization of ideologies from a state-centered perspective. ₅₆ As we have seen, in fact, a state-constructionist approach is indispensable for understanding how radical ideologies and strategic repertoires sometimes resonate with and diffuse among broad masses of people.

4. Because they interpenetrate one another the very distinction between "states" and "societies" is untenable and should be scrapped. This criticism, which is perhaps the most radical that has been raised against statist approaches, has been elaborated most fully in a much discussed article by Timothy Mitchell. ₅₇ Mitchell notes that "the edges of the state are uncertain; social elements seem to penetrate it on all sides, and the resulting boundary between state and society is difficult to determine." ₅₈ Mitchell terms this the "boundary problem." He points out, for example, that upper classes have sometimes controlled certain state institutions, making it difficult if not impossible to distinguish state power from the class or economic power of such groups. Mitchell concludes that "The state should not be taken as . . . an agent, instrument, organization or structure, located apart from and opposed to another entity called society." ₅₉ He also seems to question the analytic utility of the conceptual distinction between states and societies.

This is a problematic argument. To begin with, upper-class control of certain state institutions does not destroy the statist character of those institutions -- the fact, that is, that they are backed (unlike other institutions) by a monopoly of legitimate violence. Indeed, this situation would seem to be one in which state institutions are an "instrument" of the upper classes; yet Mitchell concludes that such a formulation is somehow impermissible.

In my view, Mitchell's argument exemplifies what Margaret Archer has termed, in a different context, the fallacy of "central conflation." ₆₀ Archer uses the term to characterize studies that, striving mightily to avoid either cultural or "structural" determinism, posit that ideas and social structure are so closely connected that "there is no way of 'untying' the constitutive elements. The intimacy of their interconnection denies even relative autonomy to the components involved." ₆₁ Mitchell, analogously, seems to assume that because states and societies are so closely bound together, "it becomes impossible to examine their interplay." ₆₂

The "boundary problem" that Mitchell discusses is real enough, and social analysts do often reify the concepts of state and society in problematic ways. ₆₃ Yet it seems more helpful (and interesting) to recognize that concrete institutions may sometimes share certain analytic characteristics of both states and societies rather than to jettison these concepts completely. Throughout his article, in fact, Mitchell himself refers quite un-self-consciously to such things as "the French state," "state practices," and "state-society relations." His own language, in other words, would seem to testify to the unavoidable importance of the conceptual distinction between states and societies. ₆₄

Limitations of state-centered approaches

Although these general criticisms of state-centered perspectives are ultimately unhelpful, statist approaches do have their limitations. In this section, I examine some of the more serious theoretical gaps in state-centered analysis and point to some theoretical resources than can help to bridge them. A proper recognition of these gaps not only reveals the limits of what state-centered analyses can reasonably hope to explain, but also helps to highlight more clearly what statist approaches to social revolutions can explain.

For analysts of revolutionary movements (or collective action in any of its forms), the fundamental problem with statist analysis is that it does not theorize the nonstate or nonpolitical sources -- or the independent explanatory weight -- of associational networks (i.e., "civil society"), material resources, or collective beliefs and discourses (including grievances and identities) ₆₅ Needless to say, this is a significant problem indeed given the potentially crucial connection between social organization, resources, and culture, on the one hand, and collective action (revolutionary or otherwise), on the other. Fortunately, there are some powerful theoretical resources at hand that can help to make that connection.

For example, the role of social organization and interpersonal ties in mobilization processes has been powerfully addressed in recent years by so-called social-network analysts. ₆₆ These analysts emphasize the crucial role of networks of social ties in recruiting people into, and then sustaining their collective identification with and commitment to, social movements and perhaps even larger political communities (thereby obviating a need, in some cases, for substantial material resources). Network analysts also stress how such social ties, sometimes in the shape of formal organizations, provide the relational infrastructure of actual collective actions. These insights have been underscored by those who emphasize the importance of "civil society" -- that is, voluntary associational activities -- as a mechanism for democratic dialogue and a bulwark against state oppression; these insights may also be found in the work of Marxists who emphasize the importance of class-based collective action in particular. ₆₇

From all these perspectives, in fact, individuals with a strong inclination to pursue reformist or even revolutionary change, and who also find themselves in a political context that allows or even encourages such pursuits, will still be unable to act effectively unless they are connected to a sufficiently large social network of like-minded people. Seemingly "appropriate" political opportunity structures, in other words, will not give rise to collective action if such networks do not exist.

Of course, state-centered analysts can justly counter that these associational networks are often politicized and radicalized, and even built up in the first place, as a result of specific state structures and policies. Social networks, after all, do not simply fall from the sky. Network analysts, proponents of "civil society," and Marxists, unfortunately, often neglect the ways in which state actions shape the formation, or prevent the formation, of voluntary organizations and revolutionary associations in particular. Still, these associations are also typically rooted in class or ethnic relations, extended kinship networks, religious communities, urban neighborhoods, or rural villages -- still other social networks, that is, which do not derive wholly or even in part from state practices. And associational networks and practices have their own dynamics and emergent properties that need to be taken seriously and analyzed in their own right. Revolutionaries themselves, for example, may act in ways that expand or corrode their ties to other people. For these reasons, a state-centered perspective on civil society is inherently limited.

The potentially autonomous influence of material resources on collective action, for its part, has been most carefully theorized by resource-mobilization and political-process theorists, ₆₈ as well as by certain rational-choice theorists. ₆₉ All of these analysts point out (albeit in somewhat different ways) that even tightly knit groups may not be able to act collectively -- at least not for long or with much effectiveness -- if they do not have steady access to the money and various sorts of infrastructure and technology -- means of communication and transportation, weapons, safehouses, etc. -- that are necessary to sustain their activities and (perhaps) motivate people to contribute to their cause. Even tightly knit groups that would seem to have the opportunity as well as an interest in acting collectively may not be able to do so effectively without substantial material resources. So again, collective action (whether revolutionary or not) may depend on much more than an encouraging political context.

Of course, a group's access to material resources generally depends on how it is inserted into specific social networks and institutions; the class composition of such groups is of particular importance in this regard. Nonetheless, access to specifically state resources may also be quite important for political mobilization -- even for would-be revolutionaries who are violently excluded from the state. Defectors from the state's armed forces, for example, often bring along their guns. Guerrilla armies, furthermore, usually build up their arsenals through raids on peripheral army garrisons or ambushes of government troops. And some revolutionary groups, of course, have had access to the resources of foreign states -- which is one of the ways in which the international state system (and geopolitical competition in particular) matters for revolutionary conflicts. While the extent of external aid to revolutionaries has often been exaggerated by their opponents, ₇₀ there is little doubt that such aid figured prominently (which is not to say decisively) in the revolutionary conflicts. in (for example) Mexico, Vietnam, Algeria, and Afghanistan. ₇₁

Finally, the potentially independent role of beliefs and identities in collective action has been powerfully underscored recently by theorists of so-called "framing processes," especially David Snow and Robert Benford and their collaborators. ₇₂ These theorists, drawing on Goffman's important study, ₇₃ argue that "objective" reality is only recognized (or indeed recognizable) as unjust and alterable when it is interpreted or "framed" as such by means of specific cultural systems or discursive formations. When extant collective frames do not allow such an interpretation --even of a reality that an external observer might find both unconscionable and easily rectified -- then collective action aimed at altering that reality is obviously impossible. In fact, even resourceful groups that would seem to have the opportunity as well as a rational interest in changing their predicament will not (indeed, cannot) do so in the absence of an appropriate cognitive frame.

Of course, as we have seen, a state-centered perspective would emphasize that the plausibility, justifiability, and diffusion of a militant collective-action repertoire or a specifically revolutionary ideology may be strongly shaped by specific state practices. Revolutionary "frames," ideologies, and cultures, that is, no more drop from the sky than do social networks or material resources. Unfortunately, framing theory and other forms of cultural analysis often overlook the specifically political processes by which collective beliefs and discourses are formulated pand broadly diffused.

Still, like associational networks, revolutionary ideologies are also rooted in a variety 9f social relations and cultural systems that may not derive wholly or much at all from state practices. And such ideologies have their own substantive properties that demand to be taken seriously and analyzed in their own right. Revolutionary Marxism and Islamic "fundamentalism," for example, envisage the radical reconstruction of societies in very different and distinctive ways. For these reasons, a state-centered perspective on culture and ideology -- like that on networks and resources -- is inherently limited.

The case of the Cuban Revolution

The strengths and limitations of a statist perspective are exemplified in several recent comparative studies of Latin American social revolutions, including works by Robert Dix, Timothy Wickham-Crowley, and John Booth and Thomas Walker. ₇₄ All of these studies explicitly engage and often endorse different strands of state-centered theory (as well as other theoretical approaches) in their attempt to explain why radical movements in the region have only seized power in Cuba and Nicaragua in recent decades. More interestingly, I think, the strengths and limitations of statism are also evident in a recent case study of the Cuban Revolution --a nonpolemical analysis that does not in any way explicitly draw upon or attempt to criticize a state-centered (or any other) theoretical perspective.

Marifeli Perez-Stable's The Cuban Revolution: Origins, Course and Legacy ₇₅ develops a generally persuasive multicausal account of the revolution, albeit one that "highlights the importance of social classes in the breakdown of the old Cuba and the making of the revolution." ₇₆ (Perez-Stable notes that while she has been influenced by both Theda Skocpol and Charles Tilly, she has "refrained from engaging the literature on revolutions." ₇₇ ) In fact, two of the factors that, according to Perez-Stable, "interacted to render Cuba susceptible to radical revolution" were the weakness of Cuba's clases economicas (i.e., the bourgeoisie) and the relative strength of the clases populares or popular sectors, influenced in part by the ideology of radical nationalism. ₇₈ However, Perez-Stable also draws attention to two other causal factors that refer, at least in part, to characteristics 6f the Cuban state, namely, what she terms "mediated sovereignty" (i.e., the Cuban state's lack of autonomy from the U.S. government and U.S. corporations) and a near-chronic "crisis of political authority" that deepened with the dictatorship of Fulgencio Batista of the 1950s. ₇₉

Perez-Stable's account would thus seem to demonstrate both the necessity and the insufficiency of treating the prerevolutionary Cuban state as an independent causal factor in the revolution. She implies that the geopolitical subservience and weakness of that state, as well as the serious legitimation crisis that developed following Batista's coup of 1952, created a political opportunity for some sort of revolution; at the same time, she suggests that analysts also need to take into account the strength (and ideology) of Cuba's social classes in order to understand why a radical revolution actually occurred. Indeed, Perez-Stable strongly suggests that the political crisis of the 1950s would not have resulted in social revolution were it not for the weakness of the Cuban bourgeoisie and the strength of the radicalized popular sectors.

And yet the story is even more interesting than this. For Perez-Stable's account also suggests that the very weakness of conservative and moderate political forces in Cuba on the eve of the revolution, as well as the gradual attachment of the popular sectors to Fidel, the July 26th Movement, and the Rebel Army, were themselves primarily a result of actions taken by the Batista dictatorship. In other words, without fully realizing it, Perez-Stable makes a number of state-constructionist arguments in her account of the fidelistas' rise to power.

Perez-Stable repeatedly suggests, for example, that "Batista's resistance to calling elections undermined the moderate opposition and bolstered the July 26th Movement" and, more generally, "bolstered those who argued that armed struggle was the only way to challenge his rule." ₈₀ Indeed, both the moderate opposition and the Communist Party -- which at first viewed Fidel as a "putschist" and "adventurist" -- positively "endorsed armed rebellion when other avenues of struggle against Batista had all but disappeared." ₈₁ Perez-Stable further notes how broad sectors of the popular classes and even members of the clases economicas were disgusted by the harsh repression and undisguised corruption of the batistato. She notes that many wealthy Cubans supported the insurrection, contributing 5 to 10 million pesos to the rebels; indeed, Perez-Stable suggests that "virtually all Cubans" backed Fidel and the Rebel Army by January 1, 1959: "the clases economicas . . . joined in celebrating the revolution." ₈₂

At the same time, Perez-Stable emphasizes that Batista might have preempted social revolution had he simply been less intransigent: "The general might have consented to free and honest elections and ushered in a provisional government in late 1955 when Cosme de la Torriente led the civic dialogue movement or early in 1958 when the Catholic church revived it." ₈₃ Unfortunately, "Batista became more intransigent as momentum gathered against his rule." ₈₄

Revolution might also have been averted had the Cuban military replaced Batista with some sort of provisional government -- as the United States came to hope and scheme ₈₅ -- or had the armed forces simply contained the guerrillas in the Sierra Madre. But the corruption and politicization of the Cuban military under Batista divided and fatally weakened that institution. Perez-Stable notes the unsuccessful coup attempt led by Colonel Ram6n Barquin and the much more serious naval uprising against Batista at Cienfuegos. ₈₆ She also refers to the failed government offensive against the rebels during the summer of 1958, although she might have said more about the reasons for this crucial failure. This offensive, after all, clearly demonstrated that the Cuban armed forces as a whole had neither the will nor the capacity to fight an effective counterinsurgency war, thereby sealing Batista's fate. Perez-Stable might have noted, for example, that the commanding officer in northern Oriente province -- a political appointee whose promotion rankled many professional officers -- simply refused to engage the rebels. ₈₇ "By the end of the summer," Luis Perez has noted,

The army simply ceased to fight. Desertions and defections reached epidemic proportions. Retreating units became easy prey for advancing guerrilla columns. . . Local military commands surrendered, often without firing a shot. Some defected and joined the opposition. ₈₈

In sum, the failure to either contain the rebels or preempt popular support for a radical revolution was itself primarily a consequence of the character and decisions of the Batista dictatorship.

Perez-Stable's analysis thus clearly demonstrates the utility of a state-centered perspective for understanding the Cuban Revolution. The Batista regime, she shows, not only created a political opportunity for some sort of revolution in Cuba; it also positively weakened the civilian and military enemies of a radical revolution and unwittingly enhanced the popular appeal of the fidelistas. The alignment (and ideology) of class forces in Cuba that Perez-Stable highlights, in other words, was itself very strongly shaped by the nature of the batistato.

On the other hand, Perez-Stable's account of the causes of the Cuban Revolution also points to some of the limitations of a purely statist perspective. The weakness of the Cuban bourgeoisie, for example, was not simply a result of state policies, but was also rooted in (among other factors) the economic division of interests between nascent industrialists and the sugar industry. ₈₉ The oppositional hegemony of the fidelistas, moreover, while certainly bolstered by the character of the batistato, was also a result of the astute political maneuvering of the rebels themselves and of their unswerving commitment to armed struggle and Cuban self-determination. ₉₀ Fidel himself first "captured the popular imagination," as Perez-Stable puts it, with his "integrity, compassion, and dignity, and . . . a political program of nationalist reform." ₉₁ Radical nationalism itself, in fact, appealed to many Cubans not simply because their state was historically subservient to the United States, but also because the Cuban economy and class relations -- which were strongly but of course not wholly shaped by that state -- were widely viewed as exploitative and unjust. ₉₂ Perez-Stable's study suggests, in sum, that an adequate explanation of the Cuban Revolution requires an examination not only of the Cuban state and its effects on civil society; it also demands an analysis of the independent role of class relations, popular culture, and the nature and actions of the revolutionaries themselves as they built a vast network of active supporters and sympathizers.

Conclusion

Due to its various theoretical shortcomings, a state-centered perspective alone will not completely explain (nor accurately predict) the emergence or character of collective action, including revolutionary movements. ₉₃ These very shortcomings, however, point the way toward a more powerful theory of social revolutions and collective action. Clearly, such a theory will necessarily highlight the role of social networks, resource mobilization, and framing processes in addition to state structures and practices. Of course, networks, resources, and culture cannot be simply "tacked on" to a state-centered analysis in the guise of "independent variables." For, as the state-constructionist approach in particular emphasizes, all these factors are themselves more or less strongly influenced by state-centered processes.

In my view, we still await the formulation of the sort of synthetic theory of revolutions and collective action that we clearly need. ₉₄ Until that theory materializes, however, state-centered approaches will remain our single most powerful theoretical perspective on social revolutions, and any superior perspective will need to incorporate the insights of this theoretical tradition. Indeed, this tradition's insights into both state breakdowns and revolutionary mobilization tell us much, if not everything, that we need to know about social revolutions, and they help to resolve the key puzzles that revolutions have raised for social analysts

Footnotes

Note 1: By social revolution, I mean a relatively rapid and fundamental change not only of state institutions, but also of the economic, cultural, and associational arrangements among the population governed by those institutions. A revolutionary movement, as I use the term, is a social movement with self-consciously social-revolutionary goals. Back.

Note 2: Following Weber, I define a state as those institutions that monopolize the legitimate use of violence, and control the principal. means of coercion, within a bounded territory. Back.

Note 3: Charles Tilly, European Revolutions. 1492-1992 (Oxford: Blackwell, 1993), 5. Back.

Note 4: See, e. g. , my "Old Regimes and Revolutions in the Second and Third Worlds: A Comparative Perspective," pp. 575-604 in Social Science History 18 (1994); "Why Guerrilla Insurgencies Persist, or the Perversity of Indiscriminate State Violence," paper presented at the 1993 annual meetings of the American Sociological Association, Miami Beach; "Colonialism and Revolution in Southeast Asia: A Comparative Analysis," pp. 59-78 in Revolution in the World-System, edited by Terry Boswell (Westport, CT: Greenwood, 1989); State and Revolution in the Periphery: The Cold War Era and Beyond (Berkeley and Los Angeles: University of California Press, forthcoming); Jeff Goodwin and Theda Skocpol, "Explaining Revolutions in the Contemporary Third World," pp. 489-507 in Politics and Society 17 (1989); and John Foran and Jeff Goodwin, "Revolutionary Outcomes in Iran and Nicaragua: Coalition Fragmentation, War, and the Limits of Social Transformation," pp. 209-247in Theory and Society 22 (1993). Back.

Note 5: In this regard, see my "Toward a New Theory of Revolutions," pp. 731-766 in Theory and Society 23 (1994), and Mustafa Emirbayer and Jeff Goodwin, "Network Analysis, Culture, and the Problem of Agency," pp. 1411-1454 in American Journal of Sociology 99 (1994). Back.

Note 6: This essay only discusses the relevance of state-centered analysis for understanding the origins or causes of social revolutions, including the formation of strong revolutionary movements. I should note, however, that statist perspectives have also been employed to explain the long-term outcomes of revolutions. See, e. g. , Theda Skocpol, States and Social Revolutions (Cambridge: Cambridge University Press, 1979), part 2; and Foran and Goodwin, "Revolutionary Outcomes in Iran and Nicaragua. " Back.

Note 7: See, e. g. , Peter Evans, Embedded Autonomy: States and Industrial Transformation (Princeton, NJ: Princeton University Press, 1995); Michael Mann, The Sources of Social Power. Volume II: The Rise of Classes and Nation-States, 1760-1914(Cambridge: Cambridge University Press, 1993); Skocpol, States and Social Revolutions; Ellen Kay Trimberger, Revolution from Above: Military Bureaucrats and Development in Japan. Turkey. Egypt. and Peru(New Brunswick, NJ: Transaction Books, 1978); Otto Hintze, The Historical Essays of Otto Hintze, edited by Felix Gilbert (New York: Oxford University Press); and Katharine Chorley, Armies and the Art of Revolution (Boston: Beacon Press, 1973 [1943]). Back.

Note 8: See Theda Skocpol, "Bringing the State Back In: Strategies of Analysis in Current Research," pp. 3-37 in Bringing the State Back In, edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1985). Back.

Note 9: Mann, The Sources of Social Power, Volume II. Infrastructural power refers, more specifically, to "the institutional capacity of a central state, despotic or not, to penetrate its territories and logistically implement decisions. " Ibid. , 59. See also Evans, Embedded Autonomy; Joel S. Migdal, Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World (Princeton, NJ: Princeton University Press, 1988); and Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities (Cambridge: Cambridge University Press, 1982). Back.

Note 10: See, e. g. , Randall Collins, "Prediction in Macrosociology: The Case of the Soviet Collapse," pp. 1552-93 in American Journal of Sociology 100 (1995); Bruce D. Porter, War and the Rise of the State:The Military Foundations of Modern Politics (New York: Free Press, 1994); Charles Tilly, Coercion. Capital. and European States. AD 990-1992, revised edition (Cambridge, MA: Blackwell, 1992); and Hintze, Historical Essays. Back.

Note 11: This approach rests upon two important distinctions made by Charles Tilly: the distinction between "states," on the one hand (i. e. , organizations that control the principal means of coercion within a' bounded population) and "polities," on the other (i. e. , the state plus those "member" groups with routine access to it), and the distinction between the capacity to act collectively, which Tilly terms "mobilization" (i. e. , the quantity of resources, including labor and skills, collectively controlled by a group) and actual collective action. See Tilly, From Mobilization to Revolution (Reading, MA: Addison-Wesley, 1978), chapter 3. Back.

Note 12: Sidney Tarrow defines "political opportunity structure" as those "dimensions of the political environment that provide incentives for people to undertake collective action by affecting their expectations for success or failure. " Power in Numbers: Social Movements. Collective Action and Politics(Cambridge: Cambridge University Press, 1994), 85. The notion of "structure" here seems problematic; after all, short-term or even ephemeral state actions can be as consequential as the long-term "structural" characteristics of a state. Back.

Note 13: See, e. g. , Tarrow, Power in Movement; David S. Meyer, "Institutionalizing Dissent: The United States Structure of Political Opportunity and the End of the Nuclear Freeze Movement," pp. 157-179 in Sociological Forum 8 (1993); Edwin Amenta and Yvonne Zylan, "It Happened Here: Political Opportunity, the New Institutionalism, and the Townsend Movement," pp. 250-265 in American Sociological Review 56 (1991); and Herbert P. Kitschelt, "Political Opportunity Structures and Political Protest: Anti-Nuclear Movements in Four Democracies," pp. 57-85 in British Journal of Political Science 16 (1986). Back.

Note 14: See, e. g. , Collins, "Prediction in Macrosociology"; Tilly, European Revolutions; Paul Kennedy, The Rise and Fall of the Great Powers: Economic Change and Military Conflict from 1500 to 2000 (New York: Random House, 1987); Skocpol, States and Social Revolutions; and Chorley, Armies and the Art of Revolution. Back.

Note 15: This approach is so named, of course, because of Alexis de Tocqueville's masterful use of it in The Old Regime and the French Revolution, translated by Stuart Gilbert (New York: Doubleday, 1955), and in Democracy in America (New York: Modern Library, 1981). See Skocpol, "Bringing the State Back In," 21. Back.

Note 16: This label is modeled on the well known idea of "social constructionism," that is, the notion that certain social phenomena are recognized, defined, or even produced (in whole or in part) through cultural and discursive symbols (e. g, political grievances, social problems, and collective identities). I do not limit the idea of state constructionism, however, to the cultural or discursive effects of states; as I suggest below, the organization and practices of states -- which are only partially discursive in nature -- are equally if not more consequential. Back.

Note 17: A "private" corporation, for example, cannot logically or temporally exist outside of a state-enforced legal order; the corporate form itself is legally defined and enforced, as are the property rights that attach to it. Back.

Note 18: See, e. g. , Meyer, "Institutionalizing Dissent"; Pierre Bimbaum, States and Collective Action: The European Experience (Cambridge: Cambridge University Press, 1988); Ira Katznelson, "Working-Class Formation and the State: Nineteenth-Century England in American Perspective," pp. 257-284 in Bringing the State Back In , edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1985); Robert Wuthnow, "State Structures and Ideological Outcomes," pp. 799-821 in American Sociological Review 50 (1985); and Kitschelt, "Political Opportunity Structures and Political Protest. " Back.

Note 19: Tilly differentiates "consolidated" national states ("large, differentiated, [and] ruling heterogeneous territories directly, claiming to impose a unitary fiscal, monetary, judicial, legislative, military and cultural system on its citizens") from "segmented" states (for example, "a city-based bishopric and its immediate hinterland, or . . . a composite of different sorts of unit, each enjoying considerable distinctness and autonomy"). See Tilly, European Revolutions, 35, 31. Note that "national" states in this sense are not necessarily "nation-states," which rule peoples who share a homogeneous ethnic or religious identity (and which are, in fact, quite rare). See Tilly, Coercion. Capital. and European States, 2-3. Back.

Note 20: Sidney Tarrow has made the same point about the modern social movement in Power in Movement, chapter 4. Note, moreover, an important implication: If the state system disintegrates (as many now believe it is doing, due to "globalization"), then social revolutions will disappear with it. Back.

Note 21: Lenin, "The Dual Power" [1917), in The Lenin Anthology, edited by Robert C. Tucker (New York: Norton, 1975), 301. Back.

Note 22: Ibid. , 301-302. Back.

Note 23: Anderson, Lineages of the Absolutist State (London: Verso, 1974), 11; emphasis in original. Back.

Note 24: This does not rule out the possibility, on the other hand, that revolutionaries may institute radical changes in those parts of a country that they effectively control, even if the central government has not been toppled. Back.

Note 25: See also Skocpol, Social Revolutions in the Modern World (Cambridge: Cambridge University Press, 1994). State breakdowns are also emphasized in Jack Goldstone's Revolution and Rebellion in the Early Modern World (Berkeley and Los Angeles: University of California Press, 1991). Although Goldstone presents an explanation of these breakdowns that is very different from Skocpol's (one that emphasizes demographic pressures), he shares her view that revolts from below cannot succeed so long as states remain fiscally and militarily strong. See Randall Collins, "Maturation of the State-Centered Theory of Revolution and Ideology," pp. 117-128 in Sociological Theory 11 (1993); see also Chorley's classic study, Armies and the Art of Revolution. Back.

Note 26: See especially Michel Foucault, Discipline and Punish: The Birth of the Prison (New York: Vintage, 1979), and The History of Sexuality Volume 1: An Introduction (New York: Vintage, 1990), and Timothy Mitchell, "The Limits of the State: Beyond Statist Approaches and Their Critics," pp. 77-96 in American Political Science Review 85 (1991). Back.

Note 27: Foucault, History of Sexuality. Vol. 1, 93, 96. Back.

Note 28: I thus disagree with Randall Collins to the extent that his writings sometimes seem to imply that state breakdowns themselves automatically induce revolutionary movements or popular mobilizations. See, e. g. , Collins, "Prediction in Macrosociology," 1561; "Maturation of the State-Centered Theory," 119; and "The Romanticism of Agency/Structure Versus the Analysis of Micro/Macro," Current Sociology 40 (1992), 82. Back.

Note 29: I borrow the notion of a collective-action repertoire from Tilly, From Mobilization to Revolution, chapter 5. The concept of high-risk activism is borrowed from Doug McAdam, "Recruitment to High-Risk Activism: The Case of Freedom Summer," pp. 64-90 in American Journal of Sociology 92 (1986). Back.

Note 30: As James C. Scott has emphasized, class struggles "from below" only very rarely break out of their localistic and necessarily disguised forms, even when inequalities, class identities, and oppositional subcultures are quite salient. See Scott, Domination and the Arts of Resistance: Hidden Transcripts (New Haven: Yale University Press, 1990). Back.

Note 31: See, e. g. , Mann, The Sources of Social Power. Volume II, chapter 18; Reinhard Bendix, Nation-Building and Citizenship (Berkeley and Los Angeles: University of California Press, 1977); Alan Dawley, Class and Community: The Industrial Revolution in Lvnn (Cambridge, MA: Harvard University Press, 1976); and Selig Perlman, A Theory of the Labor Movement (Philadelphia: Porcupine Press, 1979 [1928]). Back.

Note 32: Tocqueville, The Old Regime and the French Revolution, esp. part 3, chapter 1. I have elsewhere argued that Tocqueville sheds considerable light on the gradual rejection by Eastern European dissidents of a reformed socialism or "socialism with a human face"; by 1991 these dissidents generally rejected Communism in toto and were proponents of a Western-style, democratic capitalism. See my "Old Regimes and Revolutions in the Second and Third Worlds. " Back.

Note 33: See, e. g. , Goodwin and Skocpol, "Explaining Revolutions in the Contemporary Third World"; Seymour Martin Lipset, "Radicalism or Reformism: The Sources of Working-Class Politics," pp. 1-18 in American Political Science Review 77 (1983); Dawley, Class and Community; and Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968). Back.

Note 34: See, e. g. , Goodwin, "Why Guerrilla Insurgencies Persist"; T. David Mason and Dale A. Krane, "The Political Economy of Death Squads: Toward a Theory of the Impact of State-Sanctioned Terror," pp. 175-98 in International Studies Quarterly 33 (1989); and Ted Robert Gurr, "Persisting Patterns of Repression and Rebellion: Foundations for a General Theory of Political Coercion," pp. 149-68 in Persistent Patterns and Emergent Structures in a Waning Century, edited by Margaret P. Karns (New York: Praeger, 1986). Back.

Note 35: See, e. g. , Marifeli Perez-Stable, The Cuban Revolution: Origins, Course, and Legacy (New York: Oxford University Press, 1993); John A. Booth and Thomas W. Walker, Understanding Central America, second ed. (Boulder, CO: Westyjew Press, 1993); John Walton,Reluctant Rebels: Comparative Studies of Revolution and Underdevelopment (New York: Columbia University Press, 1984); William G. Rosenberg and Marilyn B. Young, Transforming Russia and China: Revolutionary Struggle in the Twentieth Century (New York: Oxford University Press, 1982); and Benedict J. Kerkyliet, The Huk Rebellion: A Study of Peasant Revolt in the Philippines (Berkeley and Los Angeles: University of California Press, 1977). Back.

Note 36: This argument is nicely developed in Tim McDaniel, Autocracy. Modernization. and Revolution in Russia and Iran (Princeton, NJ: Princeton University Press, 1991), chapter 7. Back.

Note 37: See, e. g. , Eric R. Wolf, Peasant Wars of the Twentieth Century (New York: Harper and Row, 1969), and Jenny Pearce, Promised Land: Peasant Rebellion in Chalatenango El Salvador (London: Latin America Bureau, 1985). Back.

Note 38: See, e. g. , McDaniel, Autocracy, Modernization. and Revolution, chapter 2; and Timothy P. Wickham-Crowley, Guerrillas and Revolutions in Latin America: A Comparative Study of Insurgents and Regimes Since 1956 (Princeton, NJ: Princeton University Press, 1992), chapters 8 and 11. Back.

Note 39: See John Foran's essay in this volume on the comparative-historical sociology of Third World revolutions. Back.

Note 40: See, e. g. , Goodwin, "Old Regimes and Revolutions in the Second and Third Worlds"; John Foran, "A Theory of Third World Social Revolutions: Iran, Nicaragua, and El Salvador Compared," pp. 3-27 in Critical Sociology 19 (1992); Wickham-Crowley, Guerrillas and Revolution; McDaniel, Autocracy, Modernization. and Revolution; Farideh Farhi, States and Urban-Based Revolutions (Urbana and Chicago: University of Illinois Press, 1990); Goodwin and Skocpol, "Explaining Revolutions in the Contemporary Third World"; Jack A. Goldstone, "Revolutions and Superpowers," pp. 34-48 in Superpowers and Revolutions, edited by Jonathan R. Adelman (New York: Praeger, 1986); and Robert H. Dix, "Why Revolutions Succeed and Fail," pp. 423-46in Polity 41 (1984). Back.

Note 41: See, e. g. , Richard Snyder, "Explaining Transitions from Neopatrimonial Dictatorships," pp. 379-99 in Comparative Politics 24 (1992). Back.

Note 42: Joel S. Migdal, Atul Kohli, and Vivienne Shue, "Introduction: Developing a State-in-Society Perspective," in State Power and Social Forces: Domination and Transformation in the Third World edited by Joel S. Migdal, Atul Kohli, and Vivienne Shue (Cambridge: Cambridge University Press, 1994), 2. Back.

Note 43: Migdal, for example, emphasizes how state-centered theories "encounter . . . difficulties when they assume that the state organization is powerful and cohesive enough to drive society. " This assumption, he notes, is especially problematic for students of African societies, such as Senegal, which has a conspicuously "weak" state. Migdal, "The State in Society: An Approach to Struggles for Domination," in State Power and Social Forces, 20. Back.

Note 44: See, e. g. , Skocpol, "Bringing the State Back In. " Back.

Note 45: See, e. g. , Kohli and Shue, "State Power and Social Forces: On Political Contention and Accommodation in the Third World," in State Power and Social Forces, 303. To be sure, a few state-centered theorists (e. g. , Bimbaum and Kitschelt) sometimes lapse into a sort of political determinism, but this is hardly a logical requirement of statist analysis as such! Back.

Note 46: See, e. g. , Skocpol, States and Social Revolutions, and Wickham-Crowley, Guerrillas and Revolution. Back.

Note 47: See, e. g. , Paul Cammack, "Bringing the State Back In?" pp. 261-290in British Journal of Political Science 19 (1989). Back.

Note 48: See, e. g. , Michael Goldfield, "Worker Insurgency, Radical Organization, and New Deal Labor Legislation," pp. 1257-1282 in American Political Science Review 83 (1989), and Theda Skocpol and Kenneth Finegold's response, "Explaining New Deal Labor Legislation," pp. 1297-1304 in American Political Science Review 84 (1990). Back.

Note 49: This was Nicos Poulantzas's position in his famous debate with Ralph Miliband. See the discussion in Clyde W. Barrow, Critical Theories of the State: Marxist. Neo-Marxist. Post-Marxist(Madison, WI: University of Wisconsin Press, 1993), chapters 1-2. Back.

Note 50: See, e. g. , Youssef Cohen, Radicals. Reformers. and Reactionaries: The Prisoner's Dilemma and the Collapse of Democracy in Latin America (Chicago: University of Chicago Press, 1994), chapters 2-3. The confusion probably derives from Skocpol's polemic against "voluntaristic" accounts of revolutionary political crises. See her States and Social Revolutions, chapter 1. But this polemic was clearly directed against the view that such crises, as they arose in France, Russia, and China, were caused by the actions of self-conscious revolutionaries and/or revolts from below; for Skocpol, that argument (as we have seen) stood the actual historical record on its head. Nowhere, in any event, did she question the potential importance of human agency as such. Back.

Note 51: See, e. g. , Margaret Levi, Of Rule and Revenue (Berkeley and Los Angeles: University of California Press, 1988). Back.

Note 52: See, e. g. , Roger Friedland and Robert R. Alford, "Bringing Society Back In: Symbols, Practices, and Institutional Contradictions," in The New Institutionalism in Organizational Analysis, edited by Walter W. Powell and Paul J. DiMaggio (Chicago: University of Chicago Press, 1991), esp. 2. 35-238. Back.

Note 53: See my discussion in "Toward a New Sociology of Revolutions. " Back.

Note 54: Princeton, NJ: Princeton University Press, 1990. Back.

Note 55: See also Peter A. Hall, Governing the Economy: The Politics of State Intervention in Britain and France (New York: Oxford University Press, 1986), and Frank Dobbin, Forging Industrial Policy: The United States, Britain. and France in the Railway Age (New York: Cambridge University Press, 1994). The idea that culture stands opposed to structure can also be faulted for a too narrow understanding of "structuralism. " Cultures, after all, can themselves be treated as persistent, supraindividual "systems" or "structures" that enable and constrain social behavior as much as "social" structures. See, e. g. , Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1973), and Emirbayer and Goodwin, "Network Analysis, Culture, and the Problem of Agency. " Back.

Note 56: Wuthnow, "State Structures and Ideological Outcomes," and Communities of Discourse: Ideology and Social Structure in the Reformation, the Enlightenment. and European Socialism(Cambridge, MA: Harvard University Press, 1989). Back.

Note 57: Mitchell, "The Limits of the State. " Back.

Note 58: Ibid. , 88. Back.

Note 59: Ibid. , 95. Back.

Note 60: Ardher, Culture and Agency (Cambridge: Cambridge University Press, 1988). Back.

Note 61: Ibid. , 80. Back.

Note 62: Ibid. ; emphasis in original. Back.

Note 63: See my own comments on this problem in "Toward a New Theory of Revolutions. " Back.

Note 64: Mitchell would apparently have us focus on "disciplinary power," which he argues has produced the state-society distinction as a "metaphysical effect. " "The Limits of the State," 94. Yet this would simply recreate the "boundary problem" in a new form, since it is often difficult to distinguish disciplinary from nondisciplinary practices. Here again, it seems to me, he is talking about an analytic distinction that is often blurred in the real world. Back.

Note 65: State-centered approaches also neglect autonomous social-psychological processes, including the role of collective emotions. In this respect, however, they are no different than most other theoretical traditions in sociology, including the dominant perspectives in social-movement research. See Mustafa Emirbayer and Jeff Goodwin, "Symbols, Positions, Objects: Rethinking Network Analysis," paper presented at the International Conference on Social Networks, London (1995). Back.

Note 66: See, e. g. , John F. Padgett and Christopher K. Ansell, "Robust Action and the Rise of the Medici, 1400-1434," pp. 1259-1319 in American Journal of Sociology 98 (1993); Peter Bearman, Relations into Rhetoric: Local Elite Social Structure in Norfolk England 1540-1640(New Brunswick, NJ: Rutgers University Press, 1993); Roger V. Gould, "Multiple Networks and Mobilization in the Paris Commune, 1871," pp. 716-729 in American Sociological Review 56 (1991); Doug McAdam, "Recruitment to High-Risk Activism"; and David A. Snow, Louis A. Zurcher, Jr. , and Sheldon Ekland-Olson, "Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment," pp. 787-801 in American Sociological Review 45 (1980). Back.

Note 67: Recent works on civil society include Ernest Geliner, Conditions of Liberty: Civil Society and Its Rivals (New York: Allen Lane, 1994); Robert D. Putnam, Making Democracy Work: Civic Traditions in Modern Italy (Princeton, NJ: Princeton University Press, 1993); Jean L. Cohen and Andrew Arato, Civil Society and Political Theory (Cambridge, MA: MIT Press, 1992); and John Keane, Civil Society and the State (London: Verso, 1988). The Marxist and class-analytic literature on revolutions and collective action is of course vast. Among the more influential recent studies are Jeffery M. Paige, Agrarian Revolution: Social Movements and Export Agriculture in the Underdeveloped World (New York: Free Press, 1975); Wolf, Peasant Wars of the Twentieth Century; and Barrington Moore, Jr. , Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World (Boston: Beacon Press, 1966). Back.

Note 68: See, e. g. , Tarrow, Power in Movement; Doug McAdam, Political Process and the Development of Black Insurgency 1930-1970 (Chicago: University of Chicago Press, 1982); Tilly, From Mobilization to Revolution; and John D. McCarthy and Meyer N. Zald, "Resource Mobilization and Social Movements: A Partial Theory," pp. 1212-1241in American Journal of Sociology 82 (1977) Back.

Note 69: See, e. g. , Mancur Olson, The Logic of Collective Action (Cambridge, MA: Harvard University Press, 1965), and Samuel L. Popkin,The Rational Peasant: The Political Economy of Rural Society in Vietnam (Berkeley and Los Angeles: University of California Press, 1979). Back.

Note 70: See the discussion in Wickham-Crowley, Guerrillas and Revolution, chapter 5. Back.

Note 71: See, e. g. , the case studies in Revolutions of the Late Twentieth Century, edited by Jack A. Goldstone, Ted Robert Gurr, and Farrokh Moshiri (Boulder, CO: Westview Press, 1991); Friedrich Katz, The Secret War in Mexico: Europe. the United States. and the Mexican Revolution (Chicago: University of Chicago Press, 1981); and Wdlf, Peasant Wars of the Twentieth Century. Back.

Note 72: David A. Snow and Robert D. Benford, "Master Frames and Cycles of Protest," pp. 133-155 in Frontiers in Social Movement Theory, edited by Aldon D. Morris and Carol McClurg Mueller (New Haven: Yale University Press, 1992), and David A. Snow, E. Burke Rochford, Jr. , Steven K. Worden, and Robert D. Benford, "Frame Alignment Processes, Micromobilization, and Movement Participation," pp. 464-481 in American Sociological Review 51 (1986). For different approaches to integrating ideologies and cultures into explanations of revolutions, see also Forrest D. Colburn, The Voaue of Revolution in Poor Countries (Princeton, NJ: Princeton University Press, 1994); Eric Selbin, Modern Latin American Revolutions (Boulder, CO: Westview Press, 1993); Goldstone, Revolution and Rebellion in the Early Modern World, chapter 4; William H. Sewell, Jr. , "Ideologies and Social Revolutions: Reflections on the French Case," pp. 57-85 in Journal of Modern History 57 (1985); and John Foran's chapter on culture and revolution in this volume. Back.

Note 73: Erving Goffman, Frame Analysis: An Essay on the Organization of Experience (New York: Harper and Row, 1974). Back.

Note 74: Dix, "Why Revolutions Succeed and Fail"; Wickham-Crowley, Guerrillas and Revolution; and Booth and Walker, Understanding Central America. Back.

Note 75: New York: Oxford University Press, 1993. Back.

Note 76: Ibid. , 8. Back.

Note 77: Ibid. , 184-85, fnl6. Back.

Note 78: Ibid. , 7. Although it reappears throughout her text, and she might have treated it as an analytically independent factor, Perez-30 Stable does not actually list radical nationalism as an independent cause of the revolution. Back.

Note 79: Ibid. Perez-Stable also includes two other factors in her list of the causes of the revolution, namely, "sugar-centered development" and "uneven development. " These particular factors, however, seem only indirectly related to the revolution. They powerfully influenced both class and state formation in Cuba, to be sure, but since they characterized the island since the 19th century (at least), they do not tell us all that much about why a social revolution occurred there in 1959. "Uneven development," furthermore, is a characteristic of virtually every country in the so-called Third World, including many that have never had anything remotely resembling a social revolution. Back.

Note 80: Ibid. , 9, 56. See also 57. Back.

Note 81: Ibid. , 69. Back.

Note 82: Ibid. , 62, 63. Back.

Note 83: Ibid. , 58. Back.

Note 84: Ibid. , 57. Back.

Note 85: See, e. g. , Jules R. Benjamin, The United States and the Origins of the Cuban Revolution: An Empire of Liberty in an Age of National Liberation (Princeton, NJ: Princeton University Press, 1990), chapter 6. Back.

Note 86: Ibid. , 56. See also Ram6n L. Bonachea and Marta San Martin, The Cuban Insurrection, 1952-1959 (New Brunswick, NJ: Transaction Books, 1974), 63-64, 147-152. The Cienfuegos revolt was led by "naval officers [who] felt frustrated at Batista's appointments of men who had not graduated from the Mariel Naval Academy to the highest ranks in the service. " Ibid. , 147. Back.

Note 87: Ibid. , 231, 262. Back.

Note 88: Luis A. Perez, Jr. , Cuba: Between Reform and Revolution (New York: Oxford University Press, 1988), 309. Back.

Note 89: Perez-Stable, The Cuban Revolution, chapter 1. Back.

Note 90: Ibid. , 58-59. Back.

Note 91: Ibid. , 53. Back.

Note 92: Ibid. , 3-5. Back.

Note 93: This point has been made recently by Sidney Tarrow, who is otherwise concerned to emphasize the importance of "political opportunity structures" for collective action. Tarrow, Power in Movement, 90-92. Back.

Note 94: Tarrow's Power in Movement certainly approaches such a synthesis, although I believe that it says too little about the social-psychological dynamics of collective action. For a rather different sketch of what such a theory might look like -- one that tries to reincorporate social psychology -- see Emirbayer and Goodwin, "Symbols, Positions, Objects. " Back.