PART IV—
CHANGE, DEVELOPMENT, REVOLUTION

Six—
The Idea of Political Development: From Dignity to Efficiency

Author's Note: The articles in this section deal with political change, usually the obverse of political stability. Obviously, the absence of conditions that make for the latter also may be assumed to make for change, but this is too facile. For one thing, we can say nothing about kinds of change by such reasoning. Furthermore, certain kinds of change (discussed in chapter 3) may involve the adaptation of the polity to changed conditions and thus denote in fact a kind of stability: the maintenance of a pattern, to whatever extent may be possible, if context changes.

One thing that must be known then if political change occurs is whether it results from a polity's malfunctioning or whether it happens in the nature of things: especially, whether it is "normal" developmental change—as germination, maturation, decay, and termination are normal organic changes.

During the 1960s and 1970s a large literature on political development grew, for reasons that are surely obvious. In the article that follows I tried to do three things: (1) to point out the considerable shortcomings of that literature; (2) to define the traits of proper "developmental theory," following the ideas of the aboriginal developmental theorists who wrote during the nineteenth and early twentieth centuries; and (3) to sketch a theory of political development that had those traits and also seemed plausible. The theory precipitated considerable feedback, most of it gratifying. It was also examined, again with gratifying results, in a conference on comparative law, in which the legal systems of societies at different levels of development were contrasted on the basis of the theory.

Originally published in World Politics 34 (1982): 451–486. Reprinted by permission of Princeton University Press. Copyright © 1982 by the Princeton University Press.

In chapter 1, I wrote that my German experience pervades my work. It may be well hidden in the theory, but it is there just the same: chiefly in the notion that totalitarianism, as what I here call "political society," may be regarded as a developmental stage, already latent in "elementary" political life, and not an aberration. It is all too easy and comfortable to think of totalitarianism as aberrant. The grotesque features of the Nazi version may well be so. But that does not mean that the general phenomenon is no more than a kind of madness. Nor is it to say that there may not be a "democratic" form of totalitarian government. That, however, is too large a subject to be dealt with here—except to remark that recent changes in "political societies" (Eastern Europe especially) may be dramatic, but far less vast than they seem: perhaps they are only changes from one to another kind of political system in its full (total) development.

From one point of view, the study of political development is a major area of achievement in recent political inquiry; from another, which matters more, it is a conspicuous failure.

What has been achieved is great and rapid growth. The study of political development, in contemporary form, started barely two decades ago.^[1] In short order, an extraordinary boom occurred in publications on the subject. By 1975, a standard overview listed over two hundred pertinent works. Accretion became especially rapid after 1964, though it seems to have "peaked out" (at a high level of production) in the early 1970s.^[2]

The negative side is that the result is mostly muddle. The study of political development has all the traits of too-rapid, jerry-built growth, and of its concomitant, "decay."^[3] The muddle is especially pronounced where it does the most harm: in regard to the very meaning of political development. Even scholars who were conspicuous in pushing the boom along are now viewing the matter of definition with dismay. Thus, Huntington and Dominguez start their review of the literature with remarks about loaded and wishful definitions—an "alarming proliferation" of them—and the consequent "superfluity" of much work on political development.^[4] It stands to reason that, if a concept is encumbered with many meanings, theories using it also will vary alarmingly because they are not about the same thing.

The study of political development thus is at a critical juncture. One can let it decay further or, not much different, choose to abandon itFrank Lloyd Wright's prescription for what to do about Pittsburgh (and, I think, Huntington's for political development). Or one can try a project in conceptual and, through it, theoretical renewal. In this essay, I develop a basis for renovation. Abandonment might, of course, be the wiser course. But the present conceptual muddle in studies of political development

seems to me due to avoidable causes; the early explorers of the subject were getting at something worth getting at—if it was attainable.

Critical Analysis of the Concept of Political Development

The reasons for the muddle about the meaning of political development start with a strange inversion of the proper and usual relations between concepts (labels), objects, and subjects. Normally, one begins with observations or ideas (or both). Concepts are used to make statements about them: "messages" that convey information. In making statements, difficulties may arise. More or less misinformation, or "noise," may be conveyed. Unless conventional language has been seriously abused, the fault can hardly lie in the words. Messages will be unclear to the extent that observations or ideas are crude or fuzzy. To achieve greater clarity, it is usually not to the point to revise the definitions; the obvious remedies are more exact observation and more lucid thought. The exceptions that call for abandonment are concepts that turn out to be vacuous because they label ignorance itself—like "phlogiston" or "having the vapors."

If, then, one asks what a concept means, the answer ought simply to be that it means what it is intended to mean. If disputes about the labels we ourselves have devised concern meaning as such, one can be sure that something went wrong at the first move. But, in point of fact, nothing is more discussed and debated among specialists in the study of political development than what the concept means. Huntington and Dominguez put their finger on the reason in pointing out that scholars first became concerned with a "thing" called political development, and that this concern "naturally" led them to try to define what the thing is.^[5] Trying to find a proper bottle to attach to a label had a predictable result: the "alarming proliferation" of meanings mentioned earlier. One sees this best in overviews of the literature on political development, which generally present lists of meanings—lists that tend to be lengthy as well as different from one another.^[6]

Why this topsy-turvy procedure of putting concepts before meanings? The explanation seems to me apparent in the actual unfolding of studies of development since the early fifties. At first, the term "development" became fashionable policy language for describing economic differences between Western (and other) societies and the assumed aspirations of Third World countries. In the economic sense, the concept was reasonably informative: it referred to levels of material abundance and processes of raising the levels. When political scientists appropriated the label, they also referred to assumed aspirations in the Third World and differences be-

tween it and the West; they then searched for a special content for the concept. Since "political abundance" has no immediately apparent meaning, the results were bound to be odd and eclectic.

The search for a political version of abundance is most evident in attempts to define political development in terms of capabilities—levels of them; "crises" regarding them; changes in them; various capabilities (as in Almond and Powell); or some particular kind (as in Organski). ^{[ 7]} This tack, however plausible, turned out, at best, to be problematic. Because capabilities are potential, it is intrinsically difficult to be precise about them. More important, the notion of capability only removes the problem of definition by one step. We now ask: Capability to do what, and how?

In most cases, the definition of the "thing" called political development was sought in a manner even more likely to lead to conceptual entropy. From the outset, development was, as stated, a norm-laden concept, distinguishing "us" from "them," Western achievements from non-Western aspirations—much like the earlier, less euphemistic word civilization . The difficulty here is not just parochial bias, but the tendency to conceive political development simplistically—as what exists in the West and had gone on in its history (especially its recent history), and what was sought to be replicated, or was bound to occur, in "backward" societies. Since a great many things have gone on in Western history, scholars were virtually compelled to choose for emphasis some aspect, or aspects, of Western history. They could—perhaps had to—choose by opinion. And so we get eclectic catalogues of meanings, which, taken altogether, simply spell Western history.

Thus, political development has been especially associated with increasing democratization; with growing bureaucratization; with the professionalization of politics (what Weber called living "off," not "for" politics);^[8] with the formalization of politics (actions based on explicitly prescribed legal rules); with the decline of ascription and the rise of achievement in political roles; with the growing clarification and resolution of political jurisdictions (the gradual vanishing of parallel or overlapping functions, like those of the Church, principalities, and other traditional corporations, and the clearcut rank-ordering of roles and structures, contrasted to the messy lack of hierarchy produced by infeudation).^[9] The very fact of the building of nation-states has also been equated with political development ("nation-building" once was the most common meaning), as has the increased penetration of societies by governmental authorities and the increase in redistributive policies. The list can easily be extended. History is enormously multifaceted.

Granted, development is not a sensible concept unless used in a historical sense. But if the concept remains closely tied to concrete history, there is no need for it. We already have a more intelligible concept for

development as history: namely, history. Development must surely stand for something more abstract and "ruleful" (theoretical)—something that is manifested in histories (plural).^[10]

Why not, then, simply abandon the idea of "political development"? In the first place, the present literature on political development simply does not represent "developmental" inquiry properly. In fact, little but the label itself (plus some other, mostly misused, concepts culled from developmental thinkers) links that literature to genuine developmental thought. The state of the literature, consequently, is irrelevant to assessing the utility of theories of development, properly constructed. (It might be added that the disjunction of the label from the theorizing it represents is ironic because developmental thought occurs at the very origin of modern social science in the nineteenth century and was devised to deal with its core issue: what it is to be "modern.") Second, the core issue that gave rise to developmental thought has hardly been transcended by later history. It remains obtrusive and urgent and it is still very much a puzzle. Finally, the potential of developmental thought for unraveling the puzzle it was meant to solve has hardly begun to be realized. The failure is the result of misunderstanding its intention. Very summarily, this intention was to understand our own societies (and, by extension, others) by finding their location in social time as a critical theoretical dimension.

The Nature of Developmental Thought

Social Time and the Puzzle of Modernity

Societies exist in, and "move" through, history—and social history differs from other sorts of history. Obviously, then, social theories must somehow come to grips with the nature and significance of "social time." One issue that is thus posed is how to think of the flow of history: whether to treat history annalistically, as a continuous thread, or as passing through distinct phases—as Geertz puts it, "a medium through which certain abstract processes move."^[11] A further issue arises if the latter tack is chosen: how to think of history as an abstract medium. For the most pan, that question has been dealt with via picturesque metaphors, above all that of organic growth and decay.^[12] Developmental thought is preeminent in trying to transcend mere imagery in the treatment of social time,^[13] as well as in the extent to which such time is treated as a puzzle and a key to understanding.

The context in which developmental thought appeared had much—perhaps everything—to do with its animus. Developmental theories began, simplistically, early in the nineteenth century and attained their apogee early in the twentieth. Roughly, the period runs from Comte, the first volume of whose Cours appeared in 1830. to Durkheim and Weber; this

period was, as we know, one of rapid, tumultuous, broad-scale, and above all almost wholly unprecedented changes, in scope and in kind, in the West. The changes need no elaboration. They include the spread of industry; the rapid growth of science and technology; the "disenchantment" of the world, as Weber called it, through rational perceptions and behavior; growing bureaucratization; democratization and its many concomitants (the appearance of political parties, changes in the composition of elites, and so on); nationalism and the emergence of new states; vastly increased social mobility; great cities; transformed networks of transport and communications; revolutions; reaction; and much else. Such extraordinary changes, especially while still novel, were bound to pose profound puzzles. They were also bound to unhinge established ideas about the nature of societies' movement through time. And they were, at the least, likely to make the proper understanding of historical process appear fundamental to all worthwhile social understanding. The belief in the critical significance of social time, and of the location of particular societies in it, emerges most categorically in Durkheim's sociology. For Durkheim, all was timebound and, therefore, culture-bound: not just "social facts," but even the epic issues of moral philosophy (because moral imperatives are senseless out of their historical context) and those of validity and truth (because no universal mind exists, only particular cultural consciences collectives ).^[14] Thus, the core problem of developmental thought was the puzzle of modernity. That puzzle had numerous related facets. Where were Western societies on the continuum of history? How did they get there? What forces had moved them from some long-past primal condition to the present? Where was historical process taking them? These questions differ from the way the problem of modernity is put now. In its earlier version, the issue was Western society:"we" rather than "they"; modernity rather than achieving modernization. Comparisons with premodern societies might, of course, help in finding solutions, but "we" were the problem. It follows that developmental theories could hardly have conceived of modernity in terms of the general condition, or selected facets, of "advanced" societies. Their first task was to diagnose that condition itself, in contrast to other conditions of societies.

In order to come to grips with the problem of modernity, and with the more general problem of passage through social time, the developmental theorists devised a mode of theory as unprecedented as the social conditions that they tried to understand—a change in thought perhaps as momentous as the appearance of Newtonian mechanics and cosmology in the physical sciences. To understand their mode of theorizing, it is particularly necessary to grasp two things at the outset. One is that the developmental theorists tried, in essence, to find patterns in pervasive novelty and seeming flux—to get bearings in a world devoid of all fixity and pre-

cedents. The second involves the nature of conceptions of social time available for them to find such patterns—conceptions they rejected as inadequate for their task. Discussion of these earlier conceptions should help to clarify the nature and novelty of developmental thought by contrast; and discussion of perceived shortcomings in the conceptions should clarify the animus of developmental theory—and thus also the traits such theory should possess.^[15]

Nondevelopmental Thought

Universal-abstract theory . The strict antithesis of developmental theory is a body of social thought not far removed from it in history, and still its chief competitor. For want of a conventional label, I will call it universalabstract thought. In such thought, there is no temporal dimension at all, or at least none that matters.

Although societies and polities exist in history, it is possible to theorize about them as if they did not. Many social and political issues are universal. In politics, the manifest universal issues are those of authority and subjection: the origin or basis of governmental authority, its proper domains, the obligations and rights of subjects and princes, the nature of justice, and so on. Since these issues arise wherever a polity exists, it may seem simple logic to seek their solutions independent of contexts, including temporal contexts. Before the nineteenth century, sociopolitical theory was addressed chiefly to such "universal" issues and aimed at such atemporal solutions.^[16]

If issues are not tied to contexts, it will seem plausible to seek solutions by methods that also ignore contexts. The systematic method of abstract thinking thus is deduction, and, in fact, the prototypical method of universal-abstract thought was social geometry. One invoked axiomatic truths and deduced theorems (mostly to serve as normative imperatives) from them. Both the axioms and deductions often were odd and hardly involved tight geometric reasoning. Sometimes, to be sure, premises were stated as explicit postulates, as in the opening of the Declaration of Independence. More often, they made apparently empirical assertions, especially about human nature: that human life is "nasty, poor, brutish, and short," or that people have a "natural identity of interests." Or they postulated some alleged primordial sociopolitical event—usually a social and/ or governmental "contract." At best, the links between the postulates and theorems also were quasi-mathematical; but the spirit of universal-abstract theories certainly was geometric.

Geometry, of course, is wholly abstract. It is a tool for imposing logical exactness on understandings of experience. If geometry as such is made into theory, the world that is theorized about will necessarily seem static,

timeless, ahistorical, and noncultural. The universal-abstract theories of Hobbes or Locke (among many others) do seem to pertain to a son of clockwork social universe. Time exists in it; but societies move in time like planets through their orbits.

Such a theoretical world can only be plausible if one's sense of social life is fundamentally one of fixity. Surely this at least partly explains the historical location of sociopolitical geometry. It postdates the upheavals of the Renaissance and the Reformation. It played a major role in the philosophic of the ancien régime . It was, in short, the characteristic thought of embryonic "modernity" not yet become mysterious and perceived as a sort of permanent, if still youthful, maturity. Therefore, developmental thought should be understood, above all, in counterpoint to universalabstract social and political theories. To repeat, Durkheim insisted on temporal answers even for the seemingly timeless issues of such theories.

Of course, there existed, alongside universal-abstract thought (indeed, long preceding it), modes of social thinking in which time does play a role, sometimes a considerable one. It would be strange if this were not the case. None of these modes of thought, however, could be suited to the goals of developmental theory—though one came close in time and traits and may be considered its flawed precursor.

Social time in classical and Christian thought . In Greek and Roman philosophies, as in the early myths that the classic philosophers rationalized,^[17] all time is cyclical: the myth of Demeter made into rational philosophy. The cycle manifestly is fundamental to political philosophy in Aristotle ("the same opinions appear among men infinitely often") and underlies history in Polybius. It was perhaps inevitable that early thought about social time should use the imagery of cycles. The obvious reason is that the perception of time as a cycle surely is the most immediate experience of our personal time as organisms, and of the organisms all about us, in the recurrence of germination, growth, withering, and rebirth.

The idea of organic growth is not incompatible with developmental thought. Perhaps it even is the prototype of all developmental thinking, as Nisbet argues.^[18] Growth and development have always been inseparable ideas. But this does not apply to the notion of organic cycles. The idea of cycles, after all, is close to being atemporal: it involves infinite repetitionchangeless change. The idea may be a useful deus ex machina for historical explanation: things happen, as they often do in Polybius, because the time is ripe for them; or they do not happen because their season is not yet. But the notion obviously will not do if the explanandum is considered unprecedented. In fact, when the Greeks and Romans thought of the emergence of their own civilization, they used a quite different imagery: that of a gradual trajectory, so to speak—from an age of childish ignorance

and contentment to philosophic wisdom and serenity.^[19] The inconsistency can be explained by the fact that the classical philosophers, quite unlike the developmentalists, knew their historic place: despite imperfections and corruption, they considered themselves to be at the end of "advancement"—certainly not on a novel journey to an unknown destination.^[20]

The conception of history as a trajectory, essential in developmental thought, is, of course, the basis of the Christian idea of social time. The Christian trajectory is a pilgrim's progress. This is especially explicit, before Milton, in St. Augustine: "The education of the human race . . . has advanced . . . through certain epochs or, as it were, ages, so that it might gradually rise from earthly to heavenly things."^[21] The pilgrim's advance, though, was gloomy, as befit the time of Alaric's sack of Rome; it was an advance through corruption toward apocalypse. Such a vision could hardly speak to thinkers in a hopeful age. But, more important, what makes the Christian idea of history irrelevant to developmental thought is, again, what is most basic in the latter: puzzlement over social time as a variable. The Christian (and especially the Augustinian) thinkers also knew their historic place: at the nadir of time, when an old world had died and a new one had not yet germinated. It is difficult to think of anything less appropriate to the mystifying ferment and burgeoning of the developmental theorists' world.

Progress theory . A considerable, and familiar, body of thought that lies much closer to developmental theory certainly does not share Christian gloom, nor does it envisage the apocalyptic transformation of a dying world: progress theory. The nature of theories of progress (from Fontenelle and Leibnitz, through the greater intricacies of Kant, to the detailed "sketch" of Condorcet, and beyond) surely needs no discussion here. Progress theory is one of the staples of academic political education.^[22] What does need discussion, and a fair amount of it, is why progress theories do not measure up to the developmentalists' task and therefore differ from developmental thought.

We must be especially careful to distinguish progress theories from developmental theories because it is common to think of progress theory as an early version of developmental thought. No doubt there are continuities and resemblances. They occur contiguously in time, and the line of division between late progress theory and early developmental thought (like Comte's) is blurred. But, seen whole, progress theory differs from developmental thought in crucial ways—especially in ways that are important for dealing with modernity as a puzzle.

Most significant, neither the future nor the past really were treated as a puzzle in progress theory in the first place; consequently, the present also was considered to be transparent to clear minds. Even if any specific

location on the path of history might be hard to fix exactly, the path itself was known. It was, typically, denned by the continuous growth of rational knowledge, not least through the waning of superstition and magic. Such a belief seemed plausible, even obvious, when only the lighter side of modern secular history was in evidence.^[23] Although in most progress theories, the future was a vague, poetic vision, the theories at least were of paradise glimpsed, if not yet gained. In any case, paradise was certain.

To be sure, there were, as always, thinkers of skeptical disposition (like Hume and Voltaire) who did not so much dissent as abstain; there were also the gloomy-minded. But it was not until the French Revolution and its aftermath that the simplistic, sunny theory of progress wobbled. In the context of developmental theory, pretty visions of constant progress toward utopia were as out of place as doom-speaking. The world was growing richer, more knowledgeable, busier, more secure. But much of what was novel also was patently ugly, and novelty coexisted with reaction.

Also important was the belief of progress theorists in a certain fixity in change itself. That is why some progress theorists fit well into the category of abstract-universal thought. Fixity existed especially in "human nature"; history was the social realization of that nature over time. In retrospect, we ourselves can discern a considerable temporal puzzle here. But that the theorists of progress did not do so is evident in their tendency to write what has been called "conjectural history," the prototype of which is the Discourse on the Origin of Inequality . It is still more evident in their searches for the essence of their own refined identity in rude or savage peoples. (Adam Ferguson's Essay on the History of Civil Society is the outstanding case in point).

Surely there is all the difference in the world between theories that profess to understand the present through the remote past and those in which the nature of the trajectory from past to present to future consists of deeply sensed mysteries. Granted that developmental thinkers, like theorists of progress, made much of rudimentary societies and did so in order to understand themselves and their own times. But developmental theory did not seek the sophisticate in the savage. Rather, like Melville, developmental theorists stressed contrast. Above all, where progress theorists wrote with certainty about the flow of social time, developmental thinkers puzzled and theorized.

Developmental Thought

We can now outline the essential traits of developmental thought (its form), in contrast to earlier treatments of social time.^[24] Each trait can be related to some facet of the puzzle of modernity. And, in each case, the point can

be substantiated that contemporary work on political development represents developmental thought inadequately.^[25]

Inherent change . The foremost trait of developmental thought is that it proceeds from the premise that change is inherent in society. Thus, social statics and dynamics coincide, and the dichotomy between order and change is false; order, as Comte said, is order-in-change.^[26] All other traits of developmental theories follow from the first in some fashion. The sheer scope and obviousness of change in the context in which developmental theories were formulated demanded that change be regarded as inherent, and thus ubiquitous, in social life.

What is most important theoretically about the premise of inherent change is that it involves an explicit choice at the fundamental dividing line between all theories, the primary branch point at which alternative theories occur. I have made this argument extensively elsewhere,^[27] so I will be sketchy about it here. In explanation, the matter to be explained can be considered inherent or contingent. If contingent, it occurs because of abnormal (literally aberrant) conditions that happen fortuitously. If inherent, it occurs ineluctably, unless impeded or diverted by chancy conditions. (Thus, our deaths are inherent, our illnesses contingent.) Everything in theory depends on the choice made at this branch point. As a familiar example, I would cite the fatefulness of the change from the Aristotelian conception of motion, which was contingent, to the Galilean, in which motion is considered to inhere in matter. Developmental theory was, potentially, just as revolutionary.

How does the premise of inherency set developmental thought apart from other modes of sociotemporal thinking? First, in early ideas of social time, change was inherent in society only as changeless change, as repetition; or else, necessary change was a "one-shot" apocalyptic transformation. Even in progress theories, the trajectory of social time always came, inherently, to rest—and rest was never far off.

In all thought before developmental theory, then, social motion was Aristotelian. It tended toward a telos: a literal end. Also (Thucydides is typical), the events of concrete history tended to be explained largely by manifestly contingent causes: by the particularities of acts, contexts, wise or unwise choices, and the like. Historical philosophy and explanation were bifurcated and inconsistent. In the former, cyclical necessity prevailed; in the latter, accidents. This bifurcation is still evident even (or especially) in Marx as a philosopher of history and interpreter of particular events.

The first task of developmental thought, then, is to formulate a really general theory of social time; the next is to apply it, as explanation, to particular cases at particular points in particular histories. "Really general"

means abstract theory regardless of time, place, and circumstance—theory that spans the whole of history, from primal origins to modernity. Clearly, the contemporary literature on political development does not come dose to discharging that task. In much of it, even our own development seems to have begun only in the nineteenth century, or not long before.^[28]

Dimensional change . It is necessary in developmental theory to construct what Simmel called "abstract grammars" with which to make sense of the immensely complex actual flow of social time and to locate concrete cases in such time.^[29] These grammars, in Kantian terms, are forms of history that fit the contents of particular histories. If social change is regarded as inherent, that flow must be along a continuous dimension—although aberrant conditions might slow down movement, or produce apparent rest, on the dimension.

To define such a dimension, developmental theorists had to characterize its poles: the nature of the primal and of the fully advanced conditions of society—the latter not really "final," but a vision of the future that could link with present and past. Only in this way could developmental theorists hope to find their own place in history, in relation to the remote past and the uncertain future. The polar conditions seem simplistic, and indeed they are so, intentionally. Maine, for instance, considered the poles of the continuum of history to be relations based on status and on contract; he explicitly called them formulas—abstract tools for constructing "the law of progress."^[30] The polar types of Toennies were community (Gemeinschaft ) and association (Gesellschaft ): abstract concepts constructed to correspond to "external" (viz., concrete) collectivities; to describe old and new; and to find underlying themes in familial, economic, political, religious, aesthetic, and scholarly life.^[31] Durkheim's orienting concepts referred to polar modes of social solidarity, the mechanical and organic;^[32] these concepts, again, were used as highly abstract forms to make sense of concrete cultures, by a theorist preeminently aware of their variability. Weber's poles were traditional and rational systems of action: pure types that were useful for describing and explaining impure (concrete) cases.^[33]

In regard to the construction of such continuous dimensions, the contemporary literature on political development again falls dismally short. Writers tend to take positions on modernity (as we saw, simplistically), but not explicitly on premodernity, and least of all on the primal conditions from which development flows. Huntington just tells us that traditional political systems vary.^[34] And Lerner, typically, says no more than that traditional systems lack modern traits.^[35]

Growth and stages . If social change is dimensional, all cases fall between minimal and maximal poles. Historic flow along a dimension thus involves

changes in degree, or quantitative growth. Throughout the corpus of developmental thought, change in scale in fact played a major role. In Toennies, for example, Gemeinschaft involved small-scale units (household, kinship, village, neighborhood, town); Gesellschaft was associated with large entities (city, nation, cosmopolitan life, markets, industries, scientific fields). In Durkheim, organic solidarity was associated with growth in the sheer "volume" of society and in "moral density": the number and variety of interactions in which people participate. "Lower" societies were spread out sparsely, lacked cities, had few and slow modes of communication and transport; "higher" ones had the opposite traits, and, above all, denser networks of interaction as a result.^[36] In Simmel, the quantitative element played a critical role even more explicitly. Virtually the whole of Simmel's thought was based on a distinction between two-person (dyadic) and threeperson (triadic) relations. This distinction was, for Simmel, a metaphor for small and large—a basis for showing that even a change merely from a relation of two to three would have enormous consequences.^[37]

An essential premise of developmental thought, nevertheless, is that not just change (singular) is inherent in societies, but also changes (plural). The developmental conception of social change is typological.^[38] Just as the premise of inherent changes responded to the obtrusiveness of change in the developmentalists' social world, so the positing of qualitative changes accommodated their sense of novelty—of new species being originated. Size and growth do matter, but chiefly as bases of generic changes. Durkheim's distinction between mechanical and organic solidarity clearly was typological; but the long process by which organic solidarity replaces the mechanical type results from quantitative change. At the outset, there is a growth of what Durkheim called "moral density"—of the volume of interactions (and hence of regulative mores). As this occurs, conflicts increase—if only through proximity, like conflicts among animals living off the same pans of a tree. The chief mechanism available for reducing conflicts caused by proximity is a kind of distancing: the division of labor separates people and, as a bonus to harmony, makes them more mutually dependent. In this way, a generically quite different kind of solidarity replaces that of early societies. In the latter, solidarity results from the sameness of their parts; in advanced societies, its source is just the opposite: differentiation.^[39]

If social change is both quantitative and typological, an obvious issue arises: how to reconcile the two. The developmentalists' solution (the only solution possible) was to think of the flow of social time as involving stages: critical thresholds at which growth in degree generates changes in kind. Comte posited three such stages: theological, metaphysical, and positive. Spencer considered social change to be more obviously continuous, but also to involve typological change: from homogeneous to heterogeneous

societies.^[40] Obviously such change is quantitative, but Spencer himself referred to the result as "transformations."

The idea of social stages, essential in developmental thought, sharply distinguishes such thought from progress theories.^[41] In theories of progress, change is the growth of some desirable aspect of society (typically, rational knowledge) and a concomitant zero-sum decline in its obverse (superstition or ignorance). Progress theorists thus could, and did, envisage a final resting place for society: in the elimination of its defects. If, on the other hand, the inherency of typological changes is posited, social time must be open-ended. A historical process not conceived in simple monotonic, zero-sum terms can hardly arrive at absolute completion. The resulting theory is gloomier, appropriate to a time of puzzlement about society.

Again, contemporary work on political development has virtually nothing to say about how quantities (and which quantities) flow through political time and become converted into generic transmutations at critical stages. The imagery of developmental thought is much used: for instance, "takeoff." So is terminology that seems typological—above all, the labels "traditional" and "modern." I have already commented on the vacuity of both. If the polar types are vacuous, it follows that "takeoff" and intermediary types (e.g., transitional society) also must be mere imagery. The essential task of developing a theory of political stages—linking changes in degree and kind—remains unfulfilled.

Ruleful, necessary change . One way to orient thought to a world inherently in transmutation, in which all is perceived as time-bound and culture-bound, is to treat all cultures as having "meaning" only in themselves. In that case, understanding consists merely of interpretation. It is no accident that Kulturwissenschafi and the method of Verstehen came into being when developmental thought also flourished. As universal-abstract thought best represents the perception of underlying social fixity and uniformity, so interpretative sociology—"thick description," or cultural aesthetics—is one major response to the perception of social flux and variability.^[42]

Cultural science, however, provides no orientation to the general flow of history and offers no solution to the puzzle of modernity. Changes (and differences) may be inherent in societies, but no account of the process of change is provided. The alternative—and the fundamental difference between social science and cultural interpretation—is to proceed from the premise that societies pass through the stages of social time in a ruleful manner. In that case, modernity—as a stage or passage to a stage—may become comprehensible, in comparison to other cultures, as a particular

location on a general and necessary sociotemporal continuum. Also, in this way, comfort can be provided in a still ugly, unfinished world. The tumult of early modernity may be regarded as the onset of a higher stage in which, as in any pure type, all will at last fit coherently. At the same time, the naive optimism of believers in finality can be avoided.

Thus, again, it is a critical task of developmental theory to construct a comprehensive theory of history—one that identifies distinct stages in a continuous flow of social time which links primal to modern society. Such abstract history must, of course, especially describe and explain how changes on the dimension of time have produced whatever is qualitatively distinct about modernity.

As we have seen, though, the puzzle of modernity did not just involve location in historical time. It also involved questions about how we arrived at our temporal location and others at theirs and about where history was taking us and them. A related task of developmental theory, then, was to identify, alongside uniform motion, the uniform forces pushing societies ineluctably along the continuum of history—especially forces inherent in societies. Only thus was it possible to see the necessity of what had occurred since primal times, and only thus can one have a sure sense of direction in looking ahead—orientation to present, past, and future.

All developmental theories posit such an underlying moving force (and special forces at different stages) that pushes societies through time. Specifying such a force—describing historical gravity, so to speak—is, needless to say, a very difficult task. What force could possibly do what was theoretically needed? Obviously, it had to be absolutely fundamental—something "essential" in the very nature of societies and thus always present as a dynamic force, through qualitative changes. The early evolutionists saw the problem, but they tended to circumvent it because they wanted a quick fix to make sense of change. Their circumventions involved truisms—resort to unspecified "properties of our species," as Comte put it^[43] —that is, to human nature. But as evolutionary thought itself evolved, the solutions became less vacuous. In Spencer, for instance, the driving force behind social change was the desire for social efficiency, which Spencer equated with complexity of structure. In Durkheim, it was something even more obviously essential: social existence as such—the need for solidarity among the elements of society.

Here, once more, contemporary theorists of political development fail dismally to live up to the form of developmental thought. Not only is their continuum of political time truncated, but the force supposedly driving underdeveloped societies is, generally, little more than an unexplained urge to be more developed. At bottom, this is a truism a la Comte; it involves an assumed intrinsic "property of our species," reinforced by the

extrinsic accelerator of cultural diffusion. What is absent, above all, is a theory of the fundamental force, or forces, that brought "us" to our political condition and continues to push us through political time.

Conclusion . It should be obvious that the mysteries of modernity, as I said above, still are very much with us. For a long time, in contemporary political inquiry, they were shifted to the Third World. But now, again, they arise in reference to ourselves: for example, in the concerns with the nature and future of postindustrial societies, and their governability. Developmental thought was itself developed to deal with these puzzles. Surely, it is uniquely suited to do so; thus, it is sensible to take such thought seriously—that is, to try to construct developmental theory properly. Hence this section, as groundwork for the next.

The quintessential developmental theorist, Durkheim, best summarized the spirit of the developmental mode of thought:

Every time we explain something human, taken at a given moment in history . . . it is necessary to go back to its primitive and simple form, to try to account for the characterization by which it was marked at that time, and then to show how it developed and became complicated little by little, and how it became that which it is at the moment in question.^[44]

I propose now to do this, in broad strokes, for the political aspect of human experience.

Sketch for a Revised Theory of Political Development

The passage from Durkheim succinctly describes what is needed to renovate the idea of political development. A more detailed agenda of questions to be dealt with follows from the summary of the traits of developmental thought:

1. What conception of continuous growth can plausibly describe the long passage from primal to highly advanced polities?

2. What is the essential nature of polity in its "primitive and simple" form?

3. What forces make the "advancement" of primal polities toward "higher" forms ineluctable (or at least highly probable)?

4. What distinctive stages lie along the trajectory of political time? In what ways do these stages involve both quantitative growth and change in kind?

5. What forces move polities from stage to stage?

6. What do the answers to these questions imply for polities that are at present less developed, and for "advanced," modern polities?

What Conception of Continuous Growth Describes the Passage from Simple to Highly Advanced Politics?

I have argued that contemporary theories of political development are historically myopic. Even in Georgian England—hardly remote history—the traits now most widely associated with political development were still embryonic. Democratization was certainly not far advanced. The suffrage was severely restricted; leaders (e.g., M. P. s) either were nobles and gentry or their handpicked clients, bound to serve their patrons' interests.^[45] In regard to bureaucratization, administrative and judicial roles remained entangled, nationally and locally; recruitment was highly ascriptive; specialization and formalization were elementary.^[46] Among the more familiar conceptions of political development, only the "clarification" of societal authority was mature, for the messiness of corporate jurisdictions had certainly been cleared up by the eighteenth century.

How, then, can one characterize a continuum of political time on which the Georgian polity itself belongs to a rather advanced period? Recall that such a continuum must involve quantitative growth and must be a "form" that can contain much variable content. Moreover, the dimension involved must be anchored in time by minimal and maximal poles, one corresponding substantially to rudimentary cases, the other a vision that links perceptions of modernity to its remote and nearer past and, still more important, to an approximated future.

I suggest that the most serviceable way to characterize such a continuum is also the simplest: what grows in political development is politics as such— the political domain of society. Through political history, political authority and competition for politically allocated values have continually increased. Using Durkheim's terminology, we might regard this as growth in "political density," perhaps as a special aspect of a growing "moral density." More and more political interactions occur, overall and in place of nonpolitical interactions.

To avoid confusion about what is being argued here, a conceptual distinction must be made. One can think of "the political" as any relations that involve, say, legitimate power, or conflict management, or the regulation of social conduct, and the like. In that case, "politics" may simply exist throughout society and not be located in any clearly denned social domain or institution. Or one can think of "politics" as the functions and activities of such a concrete domain: that of the heads of societies, the princes, chiefs, or kings (for, in its modern sense, politics is associated with government, and government and social headship are synonymous). What I argue is that, through political time, the "princely domain" has constantly grown—increasingly penetrating society. And, in conjunction, political activities and relations in the less concrete sense have also grown. Expro-

priation by "princes" and expansion of political activity occur in conjunction.

One pole of the dimension of political time thus might be called the social polity . In the social polity, as a pure type, there exists a "princely" domain: some institution of headship of society, chieftaincy, firstness. That domain, though, is little differentiated from others, in the sense of having separate organizations and administrative staffs; it is anything but a subsociety—neither a "machine" nor a "system" in itself. Above all, next to nothing is done by princes, at least as we understand political activity: there is almost no active princely management of society. The society is virtually all, and the polity virtually nothing. Relations of power exist; regulations of conduct and of conflicts occur; but they do so throughout society, not in special relation to chieftaincy.

At the other pole is political society . In political society as a pure type, "private" relations have been wholly preempted by the "public" domain of the chiefs. The institutions of that domain are highly differentiated and separately organized; governmental officers and staffs constitute a large subsociety. That subsociety is a complex system in itself, while at the same time it permeates social life.

The passage from social polity to political society can be described summarily. The domain of princes, who at the outset do virtually nothing, has great, indeed irresistible, potential for growth: power resources. Over a long period, these power resources are gradually realized. The chiefs of society convert headship into primacy, and primacy into actual control—at first very slowly, then with gathering, ultimately runaway, momentum. The momentum results from the fact that, as power resources are converted, they are not used up, but in fact increase. As this process unfolds, growth in degree corresponds, at specifiable periods, to transmutations of type. In our own modern period, we approach a condition in which the distinction between polity and society has again become blurred—not because the public realm is minimal, but because it has virtually eliminated all privacy. This, though, is not an end, but itself a stage in a continuing process. The political society generates its own dynamics; and we should at least be able to discern the forces likely to move it, even if not yet where it is destined to go.

This conception of political time has been anticipated by other theorists. It parallels Durkheim's view of more general social development. The minimal pole of the continuum is grounded in the anthropologists' notion of "stateless" societies.^[47] The conception of political development as expropriation is in Weber: the emergence of the modern state was, for Weber, a process of continuous expropriation by princes of "autonomous and 'private' bearers of executive power," resembling the expropriation by large capitalist enterprises of small, independent economic units.^[48] In his

publicist essays written shortly after the Russian Revolution, Weber envisaged the further, accelerating, and continuous expropriation by the political domain of economic life, and then also of the more intimate, and the scientific and cultural, spheres—a remarkable prevision. The idea of "total" politics now also is a recurrent theme in works on modern democratic states. Sharkansky, for example, refers to runaway governmental growth "in response to incessant demands for more services" and repeatedly alludes to the erosion of the "margins" of formal government as a consequence.^[49] The vision of political society informs especially the critiques of modern governments by perceptive (if also hotheaded) "libertarians": Hayek, Oakeshott, Ellul, Nisbet, and others.^[50]

What is Polity in its "Rudimentary" Form?

To sustain the thesis that what grows and changes in political development is the political domain per se, one must, first of all, characterize that domain in its "primitive and simple form," from which advancement proceeds. None of the many structural or functional notions that political scientists have used to define the essence of polity seem to make sense for its very early forms.^[51] What seems distinctive and universal to the princely realm in its simplest form is that its occupants and practices represent the very fact that society exists. Chiefs, khans, liegelords "embody" society. They are figures through whom societies personify themselves or sometimes (much the same) the ideal order of things imperfectly reflected in social order. They stand for the fact that a common, thus moral, life exists, and they celebrate the common life and make it compelling.

Surely that is fundamental in society, if anything is, because societies are nothing if not collective entities with which members identify—that is, define themselves. Thus, ceremony and symbolism—what Bagehot called the dignified parts of government—are not to be regarded as mere pretty trappings of power; nor are consummatory (expressive) and instrumental polities,^[52] or "sacred" and "secular" ones, distinctive types at all developmental stages. At the "simple" stage (thus, perhaps, always), symbolism is the very nature of the princely, not a guise. That is why, to us, the primal political domain seems empty. Primal "symbolic politics" does not stand for "real politics." It stands for society.

The evidence suggesting that primal politics is symbolic is considerable.^[53] For instance, in Schapera's study of sub-Saharan tribes the following points emerge:^[54] The chiefs, as heads of societies, do not do much at all; they are simply marked out from others (e.g., in costume), exalted (in special rituals), subjects of rejoicing and of eulogies.^[55] Tribes are often defined simply by identification with chiefs, not by territoriality or even kinship. Sometimes no abstract tribal name exists, only that of the chief. Often, tribal names are the inherited names of the ancestors of chiefs,

and at times chiefs are named by the tribal name. In some cases, any injury done to a member of a tribe is regarded as an injury to the chief (as we talk about crimes against society). In short, the collective and the personal are thoroughly joined in the chief's personage. Much the same comes out in Lucy Mair's studies of primitive governments and African kingdoms.^[56] Mair argues, indeed, that the substantive wielding of "power over the conduct of public affairs" generally is not so much the chief's or the court's function as that of lesser figures, for whom kings are mouthpieces.^[57] Lowie's work on North American Indian tribes makes a similar point.^[58]

More important from the developmental point of view, we find this to be true also in the primitive condition of a prototypical advanced society—English society. (England may be considered as a good concrete approximation of an idealized case of continuous development: something close to an experimentally contrived universe—free of uncontrolled, deceiving contingencies—which any theory of sociopolitical development should fit closely.)

Anglo-Saxon society approaches the extreme of what I have called social polity.^[59] If a "public sector" existed in that society, it could only have been that of king, folkmoot , and Witan . The king was principally a source of social identity, as were all lesser chiefs of the English tribes. His one significant activity was leadership in the common enterprise of making war, and practically no other common enterprise was engaged in. The folkmoot originally was not a council, but simply a local muster of warriors. By 900, local moots had pretty much been displaced by the Witan , a "national" council of "wise men." But the Witan's essential function simply was to advise the king on the nature of "unchanging custom." Here the primacy of society is especially evident: while the king embodied its consciousness of itself, the Witan kept him honest, as the guardian of its mores.

Much the most perceptive study of primal politics as I conceive it is Geertz's magnificent book on the nineteenth-century "theatre-state" in Bali, Negara .^[60] Geertz alone seems to have grasped fully the critical significance of political ceremony and ritual: of the "poetics" of power as against its "mechanics"—as Bagehot alone discerned that the dignified parts of English government were not mere vestigial histrionics, but essential to its "efficiency." Geertz does temporize between regarding theater as essential in polities as such and considering Bali an exotic alternative to politics as efficient power.^[61] But, at least in Bali, "power served pomp, not pomp power."^[62]

What Forces Make the Growth of Primal Polity Ineluctable?

Chieftaincy in primal polities is much indulged and rewarded, with awe and with goods. But that does not immunize chiefs (much less their retainers) against the appetite for mundane power; and, perhaps just because

the chiefs are symbolic figures—awesome rather than powerful—power struggles are pervasive in primal societies.^[63] For the purpose of developmental theory, it is necessary to show next that in such struggles the princely domain has overwhelming resources for subduing rivals and enlarging its effective control over society. What, then, are its power resources?

By itself, the representation of societies is an essential resource for power—perhaps the one seed that is capable of growing into political society. Societies are requisites of personal identity, safety, the satisfaction of material needs. But, though necessary, they are highly intangible. They are complex even when they are rudimentary. Seeing them as networks, or complexes of roles, or fields of interaction, or patterns of exchange—these are major feats even for modern professionals. Even if the task of abstract understanding were less difficult, such understanding would hardly move affections, which surely are needed for identification and legitimacy. So the personal symbols of society derive potential from the fact that they perform the most necessary of societal functions: making society appear "real."

It is true that there are other ways of making societies tangible. Primal societies, in fact, are always personified in their gods, through rites and magic. Thus, priests and magicians are the logical (and actual) main rivals of the chiefs for principal power. But the chiefs themselves are generally presumed to have special links to the supernatural, magical world—for instance, as rainmakers, healers, invokers of prosperity, possessors of sacred objects (fishing spears and the like), and as wielders of curses.^[64]

These links to the supernatural not only reinforce secular symbolism (or make it sacred) but also associate chiefliness and "potency," for the magical world is a world of fateful powers. Chiefs are also considered especially potent figures in the material sense of prowess. All societies have collective business of some sort—in primal societies, for instance, moving camp and herds.^[65] The function of making decisions about societal business naturally tends to be lodged in the locus of collectiveness. The one universal collective business of rudimentary societies is warfare: in defense against predatory others, for conquest (slaves, tribute, etc.), or, often, simply as a ritual.^[66] So chiefs, though they have rivals in heroes,^[67] generally are the main loci of potency as prowess. This accounts for the strange duty of chiefs in some tribal societies to be in good health, as well as for the use of wars of succession (in which the strongest survive) and for the frequent use of the phallus as a symbol of chieftaincy.

To exist, and to carry out collective enterprises, societies must, of course, be harmonious in some degree. Conflicts must be managed, quarrels mediated, crimes avenged. There is a universal social need for adjudication and, again, a "natural" tendency to associate that necessary function with

society's embodiments. The actual management of conflicts and deviance tends, in fact, to be decentralized and dispersed in primal societies—a matter of self-help in feuds, revenge, and exacting reparations. But the chief always has at least some vague special responsibility in regard to justice. For instance, we are told by Traill that a basic function of the Anglo-Saxon kings was to go about the kingdom putting down "evil customs." Traill's catalogue of judicial duties actually is a list of things kings could not do; and it seems evident that kings were little more than especially prestigious "oath-helpers."^[68] Still, justice and chieftaincy had special, even if largely hortatory, links.

The moral, surely, is evident. The primal princely domain is ages removed from the monopoly of legitimate power. But where could there be greater potential for eventual monopoly than in a domain standing for society itself; for potency, military and magical; and for justice? "Dignity" and "efficiency," granted, are obverse faces of politics—but also interchangeable resources.

What are the Stages of Political Time?

The fact remains that in primal polities, whatever the chief's potential, one can barely detect an active public core. Our own political world could hardly differ more. At "our" location in political time, as stated, it is difficult to find anything that is clearly private. I am not referring to "totalitarian" polities, or only to those, but (less categorically) to the other typically modern form of polity: popular democracies. (Modern democracies, in historical perspective, simply are the gentler twins of totalitarian rule, mitigated by open competition, free communications, and a sense of rights and liberties—which, compared to earlier times, no longer really divides the public from the private, but is a sense of political decency.)

I have described the extraordinary pervasiveness of political authority in contemporary British society elsewhere and need not dwell much on details.^[69] To convey the flavor of the matter, suffice it to say the following: (1) The national government (as in other modern democracies) now directly controls about half of GNP, and indirectly plans, guides, and channels most of the remainder. (2) Parliamentary sessions, once convened only occasionally, fill up the whole available legislative work-year, and even this at the cost of large omissions—uncontrolled "executive legislation" and a severe decline in the role of private members. (3) The Cabinet has virtually disappeared; as I wrote in 1958:

Cabinet functions have become dispersed to an almost unfathomably complex administrative and deliberative machinery. Decisions once made collectively in the Cabinet are now made by cabinet committees, by individual Ministers, bureaucrats, the Treasury, official committees, party machinery, and even private associations; and, most often, by interaction among all of

these bodies. If power is concentrated anywhere in the British machinery of government it is concentrated not in the Cabinet but in this complex framework of decision-making.^[70]

One can argue that what mainly mitigates the darker aspects of fully politicalized society is the very inability to control such a concentration of functions, due to sheer diversity and overload. The gentle myths of liberal rule surely help, but perhaps no more than the fact that monolithic authority itself is too large to manage. Privacy, in political society, is found in the interstices of authority; it is, perhaps, itself mainly a product of the structure of the public realm.

How did this transmutation to something close to "political society" come about? What lies beyond the primal polity's potential for growth? This is an enormous question, and we have not even the beginning of a plausible answer. As such a beginning, I suggest a six-stage process. The process is "logical" in that each stage manifestly is a condition for the next. The stages also make sense in the context of the English polity—our standard case for observing gradual, evolutionary "unfolding" (the literal meaning of développer ) in politics. For this reason I will use English history—in gross summary—to exemplify the stages.

The politics of primacy . I have already treated the first stage, primal polity, using Anglo-Saxon England to illustrate its nature. The second stage involves what might be called the struggle for, and achievement of, primacy. The forces that push polities to and through that stage (and later stages) will be discussed presently. Here, it must suffice to say that nothing in political development can possibly come before the clarification of a distinct public domain that, in regard to "efficient" functions, is minimally primus inter pares . Without this, there is nothing that may grow. One may suppose that the establishment of a realm of substantive primacy—one that involves more than symbolic headship—will not be a tranquil process but will involve stubborn conflicts over domination and autonomy. Aside from chiefs, there are others who have politically convertible resources: religious, economic, and military. But, as we have seen, the chiefs generally have much weightier resources for providing political goods—not least, safety, in a context of continuous struggle among social domains: Hobbes's good, and no doubt the fundamental value.

This general stage fits, in England, the period of feudal monarchy , say of the twelfth century. The feudal monarchy certainly was quite different from the Anglo-Saxon, despite the fact, generally agreed, that the Conquest caused no sharp break. The domain of the Angevin and early Plantagenet princes, to be sure, remained mainly on the level of symbol and pomp; its practical authoritative functions were sparse. What is most con-

spicuous about the period is struggle for "dominion" as such. The histories portray incessant turmoil. But the tumult was not about policy, in our sense. It involved competition about spheres of autonomy and subjection; and the fundamental source of that struggle was a lack of clarification and resolution of the functions of the great and small corporations of society—all authoritative in their own domains and constantly striving to expand or protect them.

Corporate boundaries now, though, mattered for more than symbolic reasons. They mattered because the princely domain had begun to acquire a critical function: material extraction—a condition of all effective action, and thus an obsession in feudal monarchy. The Treasury preceded all other political institutions in development. The classic account of twelfth-century royal "administration" is FitzNeal's Dialogue on the Exchequer , the exchequer being its one great administrative creation. It regularized the royal revenues, and the great pacification under Henry II was, at heart, a matter of reestablishing the central revenues in face of embezzlement by the barons. Extraction increased political "density," and the latter changed institutions.

Still, Henry's charter upon his coronation was little more than an assurance of liberties, grants, and customs. Petit-Dutaillis's study of feudal monarchy does tell us that the King's concilium attended to "all sorts of business";^[71] but, as to particulars, he lists only personal issues (e.g., marriages) and familiar matters of peace, war, loyalty, treason, and the administration of justice.

The last is important, however. Judicial activities now were much enlarged and wholly reorganized—coequal with pomp and war as the core of royal primacy. Indeed, aside from finance and war, the whole royal establishment now looked like a sort of national judiciary. The King's "prime minister" was the chief justiciar; the curia had become a "normal court" for the kingdom, not just an occasional tribunal; the judicial circuits, administering common law, had been established; and central justice had largely expropriated the seignorial jurisdictions, of which only a few islets remained.^[72]

The feudal monarchy thus achieved, gradually, a considerable legal and extractive permeation of society, as a material basis for primacy. Contestation persisted for a long time, but in an increasingly muted, one-sided way. The nascent monopoly over extraction, the increasing practical responsibility for the management of conflicts, and the emergence of specialized institutions to handle these functions, realized a potential already present in the primal polity; but, more important, all this added to the growth potential of the prince's domain.

The "prophylactic" polity . Substantive primacy, especially when added

to symbolic headship, is both gratifying in itself and a supremely valuable resource for acquiring additional resources. Once it is established, struggles for its possession inevitably occur. One of the fundamental tasks of politics is to institutionalize such struggles in order to defuse them—a basic function, for instance, of competition among political parties. But institutionalization is always gradual—a sort of subtheme of development. Early on, contestation for possession of the domain of primacy must involve—in greater or lesser degree—unregulated, brutal conflicts. Lacking institutionalization (or the transformation of real and deadly conflicts into ritualized competition) damage can be limited only by prevention: prophylaxis.

In the prophylactic polity, the overriding objective of the prince is to detect and disarm usurpation, while that of others is to seize or control principality. To protect principality, it is functional to place it in a tangible physical domain and to draw potential usurpers into that domain. Hence, the identification of primacy with the prince's court. It is there that the game of trying to get and keep primacy and its perquisites is played; politics turns inward.

To a degree, however, prophylactic politics must also reach out into society, further than before. Courtly politics cannot be wholly isolated because the discontents of society might play into the hands of usurpers. Therefore—rather than for altruistic reasons—the princely domain begins to furnish something else that is valuable to society: a degree of controlled social order, as prophylaxis in everyday life against society's misérables . The result is both a qualitative change in the nature of politics and the increased penetration of society by its political domain.

In England, the era of the Tudors illustrates the stage. The late medieval and Renaissance political struggles in England increasingly had a flavor different from those of feudalism. They were epitomized, and pretty much ended, by Tudor rule, for which "absolutism" is an egregious misnomer. Nothing really was absolute. Rather, the Tudors—especially Elizabeth—successfully coped with conspiracies within the realm of princely authority. If anything authoritative was absolute it was courtly absolutism, which transformed lords into mere courtiers. Concomitantly, political competition was courtly competition—scheming within the firm.

Nevertheless, one can discern a threshold in the permeation of society by authoritative policy. Outside of the royal palaces, authoritative regulation was still sparse; but before the Tudors (conflict management and extraction aside), authoritative space, outside its royal core, had been virtually empty. A good many histories refer to an abundance of "proclamations" by the Crown, and subservient parliaments and courts, in Tudor times.^[73] Elizabeth's parliaments did indeed pass 429 bills. The figure is

often mentioned to impress. Actually, it brings out only the limitations of policy making. Elizabeth's reign lasted forty-five years; nowadays, British legislative output runs to about a hundred bills a year. Much of Elizabethan "legislation" had to do with issues of diplomacy, foreign intrigues, war, and extraction. Some of the regime's authoritative activities, however, involved a novel extension of authority into society: the systematic maintenance of roads and bridges, the licensing of alehouses, controls over wages, the mobility of labor, entry into trades, dealings in commodities, interest rates, and—most familiar—a uniform law to care for the poor.

Growing political density surely is evident, especially since this reaching out into society supplemented unprecedented ceremonial activity (royal equipages and pageantry) and an even greater increase in foreign adventurism, war, and defense. The primacy of feudal monarchy clearly was now being put to use as a generalized resource. Perhaps this was a response to much-increased "social density": the manufacturing revolution in textiles, mining, iron making, and petty trades (perfumery, barbering, etc.)—a response, in general, to a busy society of promoters, speculators, patentees, dramatists, composers, astronomers, astrologers, physicians, surgeons, alchemists, sorcerers, explorers. What Black calls "the chaos of society,"^[74] however, did not engender policy as an attempt to impose any sort of rational order. Rather, the point of authoritative "outputs" seems to have been an extension of the defusing of courtly intrigues: the prevention of social discontents and marginality that were potentially threatening to the security and isolation of the courtly domain. The increased permeation of society under Tudor rule aimed, above all, at prophylaxis: controlling vagabonds, dealing with food riots, limiting speculators, usurers, and drunkards. The Poor Law and the relentless pursuit of religious recusants are all of a piece in this effort. A valuable resource was now being hoarded—though not yet much used for additional gain.

The polity of interests . When principality no longer needs to be preoccupied with usurpation, but has been institutionalized at least in accepted rules of succession, politics can turn outward for reasons other than prophylaxis. The primacy of a social domain above other domains and, even more, the "distancing" of courts from societies, inevitably lead to a conception of princely power and the social order (not "orderliness") as being somehow unrelated. The initial extroversion of the princely domain thus can hardly be concerned with such matters as engineering social harmony or just distribution. In introverted politics, these are matters for natural order or divine ordination. When politics turns outward from the court, then, the purpose initially is not so much to manage society as to exploit primacy as a resource: the gainful use of primacy by privilege. In the polity of interests, competition overshadows majesty. Though it in no sense in-

volves democratization, the arena of politics as competition becomes much enlarged and structurally altered. It still takes place in the court, but now also in institutions associated with the court (e.g., parliament) and, to a degree, in society. Through the "outputs" sought by patrons and their clients, the polity, as Durkheim would say, markedly "condenses." Royal administrative and judicial institutions become a rather complex "machinery" government.

In England, such acquisitive exploitation of established primacy—and through it the much enlarged penetration of society—is the essence of the Georgian period .^[75] One sees the scope of the eighteenth-century British polity best in the activities of its local officials. The justices of the peace were broadly charged with collecting and delivering revenues; assuring the proper practice and flow of trade; looking after the poor, the food supply, prices, and wages; licensing brewers and drinking houses; supervising gaols; establishing asylums and confining lunatics; seeing to the lighting of streets, their paving, policing, and cleaning. All this required at least an embryonic differentiation of political labor—though bureaucratization had hardly yet begun. There were now distinct judicial and administrative sessions, distinct highway and licensing councils, as well as individual specialists, like road surveyors and constables. Late in the century, new statutory authorities, with special dudes, appeared: for instance, turnpike trusts, corporations for administering relief to the poor, and, above all, a growing number and variety of improvement commissions.^[76]

This expansion of activities, and of organizations for performing them, was not intended to manage society. The overriding trait of the Georgian polity was that it was a marketplace of influence and spoils. The central level did not really manage society, yet there was extraordinary jockeying among parliamentarians and, as a result, ministerial instability. According to Namier, men went into parliament partly out of a sort of "predestination" (men of "political families"), but even more as clients looking after patrons' interests: as placemen and as purveyors and receivers of favors (there was, says Namier, a "universal . . . plaguing of Ministers on behalf of friends and relations");^[77] to advance themselves in the military and administrative services or reap rewards from service; to obtain contracts, jobs, subscriptions, loans, and remittances. The Enclosure Acts and what Beer calls "canal politics" epitomize this extraordinary politics of interests.

The politics of incorporation and of incumbency . When the domain of politics is used chiefly for acquisitive purposes by privileged groups, other groups will try to become incorporated into the game as players, rather than be excluded from it as passive victims. As the stakes grow (that is, the spoils increase) so, one may suppose, does the appetite for shares.

Certainly the pervasive theme of early modern (nineteenth-century) British politics is democratization. Tilly depicts the process as one in which excluded subjects first become "challengers," and then, through challenge, incorporated "members" of the polity:^[78] voters and those eligible to hold office. The transformation of challenge into membership occurs because the challengers have resources of their own that can be effectively mobilized—if only strikes, violence, and the like.

As the polity's membership expands and thus becomes more diverse in interest, the political penetration of society necessarily grows rapidly in scope; when "civic inclusion" is virtually total, so is the politcization of social life—but not just in the sense of universal citizenship. Two other processes occur that rapidly transform social into political space. One is familiar: as new members are incorporated, the volume of political demands grows, and with it, the volume of outputs; with outputs, the network of committees, agencies, departments, boards to define and deliver them; and, with such organizations, their own demands: "withinputs," as David Easton calls them.

Perhaps this chain reaction sufficiently explains the rapid development of political society out of acquisitive politics. I would suggest, though, that a second process supplements the demand-response relation and perhaps is more consequential. It bears at least a vague resemblance to the Tudor preoccupation with political prophylaxis. To put it starkly: political primacy in the modern polity clearly is more than ever worth possessing and keeping in possession; however great the resources of princes before, they were puny compared to the fully realized monopoly over legitimate power. The theater of political struggle, though, is no longer confined to the small stage of the court; it comprises society as such. Thus, the modern counterpart of coping with conspiracy in order to retain control over the princely domain is either mass suppression or the search for mass support (plus the special support of the more powerful, better organized interests). Mass support is elicited, at least in part, by going beyond responsiveness: by "redistributive" policies that make large public groups into clients—collective placemen. The unparalleled scale both of repression in authoritarian modern polities and of the political provision of all sorts of goods in welfare states serves the maintenance of incumbency. No doubt welfare policies and other distributions of benefits result from good intentions; but surely, they also provide benefits, in the form of political support, for their providers. At any rate, here is a parsimonious explanation of the substantial consensus on social policy in the contemporary British welfare state. The politics of incorporation leads logically to that of incumbency.

"Political density" during these stages grows rapidly toward its maximal pole. The vastness of the business done by the machine of government requires, as Durkheim realized, more and more internal complexity of

structure, in large part just for keeping things sorted and coordinated; it requires the development of a political "system,"^[79] which is not at all the same as a machinery of government. Structures of political competition also become highly organized and institutionalized networks of organizations. In gist, the pomp of primal chiefliness virtually disappears within the systems and networks of the polity.^[80]

Two important questions should be raised about the abstracted stages to determine whether they indeed constitute a general developmental sequence. First, do the stages occur, mutatis mutandis , in other longitudinal political processes, and do they furnish a good typology for the "cross-sectional" classification of polities in the present? If so, we can assert (in the manner of early exponents of the "comparative method"—Ferguson, Comte, Tylor, Morgan)^[81] that typological differences among polities are basically developmental: namely, that there is history, not just histories. Second, would a schematic treatment of political functions, goals, and structures by stages indeed show qualitative distinctions in each class, along with the quantitative growth of the political domain? These questions cannot be treated briefly; they are posed here as items on an agenda to follow up this essay.

What Forces Move Polities from Stage to Stage?

In the preceding section, I have tried to show sequential connections between stages of political development: how the earlier stages are preconditions for those that follow, and how these, in turn, are latent in preceding stages. (An important, familiar issue for praxis—too large to be tackled here—is raised by the question whether stages can be skipped, without the occurrence of pathologies and without regression.) This demonstration, though, says nothing about the forces that propel polities from stage to stage. We need at least a summary answer to complete our sketch for a theory of political development.

In developmental theory, one wants, ideally, to identify a general motive force that operates throughout developmental time (akin to physical inertia) and also special forces, generated in each earlier stage, which similarly lead to each later stage.

The general motive force at work in the sequence of stages I have described is surely the drive for the direct and indirect benefits of "efficient" primacy in and over society—the direct benefit of social elevation and indirect perquisites, such as material goods. That drive characterizes most directly the transformation of primal, ceremonial polity. The maintenance of primacy for getting other values follows in the polity of interests and leads to the challenges that incorporate excluded groups in the domain of primacy. The possession of higher positions—primacy in the domain of primacy—animates political motion in the most advanced stage.

Although primacy seeking is the essence of the initial developmental transformation of polities, it is clear that struggles for establishing an "efficient" principal domain are only resolved when an urgent societal need for such resolution arises. In the West, that need arose from the differentiation of society into distinct but overlapping "corporations" in virtually continuous collision. One may surmise, more generally, that an initial locus of efficient primacy will emerge when it is functionally critical to social integration that this occur—that is, when the integrative force of "mechanical solidarity" no longer works. The theatrical chiefs are destined to win struggles to perform the integrative function and to reap its benefits.

If there is such a thing as "pure" power politics, it occurs when struggles for primacy have been resolved. Pure power politics is about possessing primacy, not about establishing it. Once the domain of the prince itself is safe, a different propulsive force emerges; we might call it resource conversion.^[82]

The results of converting political into other goods now come to pose a quite different, but again functionally critical problem of integration: not of society but of the political domain itself, for the sake of its effective operation. The need for political integration has two facets. As new groups are incorporated into the polity, the plethora of interests and demands they generate must be coordinated: in Almond's terminology, a need exists to aggregate interests, so that demands may be effectively pressed and responded to. More important, as society is greatly politicalized through processes of civic incorporation, the machinery of government grows into a complex system; as a result, efficient management of the system itself must increasingly become a sine qua non of political goals, even exploitative ones. Without efficient political management, social life itself is imperiled, precisely because the polity pervades it; and, without such management, power itself is a chimera. In this way, we can see in political development a diminution, if not a metamorphosis, of pure power politics—and still avoid the "fault" of tendermindedness.

Thus, while struggles for primacy propel politics throughout developmental time, at each stage they take different forms and are reinforced by special forces: forces of greed and, more important, forces generated by collective functional needs. These themes of politics—primacy seeking, power seeking, greed, and integration—are familiar. What are not familiar are the special roles they play at different stages of political development.

The process of political development moved by these forces is monotonic in two senses. I have stressed one—the politicalization of society. The long trajectory from social polity to political society can also be considered a modulation from "dignity" to "efficiency" (the most fundamental qualitative social change conceivable), and each stage of the process can

be treated as a changing balance between the two. In parallel, polities change structurally from personage to court, to machine, to system.

Conclusion

The idea of political development, then, seems to me capable of renovation along the lines sketched. What I have tried to present is a design along proper "developmental" lines. The design is, and must continue to be, far from a completed theoretical structure. But if it proves to have merit, it helps to answer the final question raised above. It has important implications precisely for the issue that a developmental theory should illuminate: the puzzle of our own modernity. I will mention one such implication for a critical problem in modern political life.

We have lately heard much about a crisis of authority in highly advanced societies. The evidence is overwhelming that there is at least a malaise about authority. Strangely, that malaise seems to exist concurrently with the progressive growth of what people supposedly (and no doubt actually) want authority to be: decent, down to earth, participant, lenient, concordant, open to achievement. Might not the solution of this riddle lie in the "disenchantment" of theatrical politics (which moves affections), by rationally effective but too-drab systems? After all, society and polity remain intangible mysteries; the social sciences are devoted to their understanding. They have become all the more mystifying as they have grown in scale, density, and differentiation. At the same time, dignity has waned in relation to efficiency. More and more, our representative figures are capable but plain, managers not princes: Fords, Carters, Wilsons, Heaths; in our families, schools, and workplaces, authority increasingly also has derogated rank. We want this, and it seems good; but can we live with it?

Perhaps that is what Weber saw when he forecast a political "polar night of icy darkness and hardness." Perhaps, too, the tension between the needs for what Weber called matter-of-factness and devotion is the force propelling us into the future of political time.

Seven—
A Culturalist Theory of Political Change

Author's Note : I stated in chapter 1 that I chose the culturalist perspective as a basis for theorizing for reasons, but not as a closed dogma. One of the reasons—not the only one or even the most important—was that its principal alternative among other perspectives, rational-choice theory, seemed a highly inadequate basis for explaining what I wanted to illuminate for myself: my piece, so to speak, of the German experience. Ironically, the most serious criticism of culturalist theory became that it could not be a basis for explaining political change at all, least of all for very rapid change—Germany after the Second World War was the favored case in point—though other rapid changes in Germany's political history since 1871, including the change from Weimar to the Nazi regime, were always conveniently left out of discussion.

In this essay I deal with the criticism that the culturalist perspective cannot cogently cope with change, by constructing a culturalist theory that covers every possible kind of political change. Readers will note that the theory can be used to explain both the disruptive change in Germany to Nazism and the emergence of an apparently stable political order in the Bonn Federal Republic (especially in connection with chapter 5)—but via general theory, not idiographic interpretation. The explanation may not be valid; much work still must be done before one can make that claim. The point, as stated, is only that it is distinctly possible to construct pertinent explanations of political changes on a culturalist basis.

The chapter also may be read with the sections of chapter 4 that deal with testing theories. It has always seemed to me incumbent on theorists

Originally published in The American Political Science Review vol 82 (1988): 789–804. Reprinted by permission. Copyright © 1988 by the American Political Science Association.

to specify predictively, prior to observation, what data should and should not turn up if hypotheses are to be considered falsified or validated. In physical theories that method is generally evident to everyone, but often for technical reasons before the observations required are possible. One valid, logically deduced prediction of an unknown is worth a myriad post hoc explanations, however clever. Consequently, this chapter states both what sort of political changes the culturalist perspective can and cannot accommodate.

The political culture approach to building positive political theories and to political explanation has been with us since about 1960 and has been much described abstractly and much applied to concrete cases. The seminal works are Almond and Coleman's and Almond and Verba's.^[1] Applications of the approach are covered comprehensively in a retrospective on the influence of their work by Almond and Verba.^[2] Explications of it as a contender for paradigmatic status in political science, so to speak, occur in numerous works.^[3] My own use of the concept of culture, which I consider more precise than that of others, is discussed in the Appendix.

Political culture theory may plausibly be considered one of two still viable general approaches to political theory and explanation proposed since the early fifties to replace the long-dominant formal-legalism of the field—the other was political rational-choice theory. Indeed, determining which of the two modes of theorizing and explaining—the "culturalist" or the "rationalist"—is likely to give the better results may be the single most important item now on the agenda of political science.^[4]

Whether or not it is advisable to take the culturalist road to theory depends above all on the ability to produce a cogent culturalist theory of political change: a theory consistent with the assumptions (postulates) of the approach and confirmed by experience. Criticisms of culturalist political theories certainly have emphasized the occurrence of certain changes in political structures, attitudes, and behavior and culturalist accounts of their occurrence in order to impugn the approach. Rogowski,^[5] for example, has argued that political culturalists have been very offhand in dealing with change—that they have tended to improvise far too much in order to accommodate political changes into their framework. They have done so, he writes, to the point that they no longer have a convincing way to treat political change at all. His argument is directed at culturalist theory in general, but he singles out Almond's work with Powell as especially indicative of the sins that culturalists commit.^[6]

This argument—and others to similar effect—strikes me as cogent criticism of how culturalists have in fact dealt with political changes. Furthermore, difficulties accounting for change in general and for certain

kinds of change especially seem to me inherent in the assumptions on which the political culture approach is based.

Difficult, however, does not mean impossible or implausible. It is quite possible to deduce from these assumptions a logically cogent account of how political change, and every kind of such change, occurs. My purpose here is to provide such an account, as remedy for the "ad hocery" that Rogowski rightly criticizes.

The Postulates of Culturalist Theories and the Expectation of Continuity

The basic reason why a culturalist account of change is intrinsically difficult to construct (hence, why culturalists have in fact tended to waffle in explaining political change) is simple: the postulates of the approach all lead to the expectation of political continuity; they make political continuity the "normal" state.

The Postulates of Culturalism

To see why this is so we must first make explicit the fundamental assumptions from which culturalist theory proceeds—its "axiomatic" basis, so to speak. These assumptions unfortunately have been left implicit in culturalist writings. It is necessary to make them explicit if one is compellingly to specify what experiences are "normal" in a culturalist world and what conditions culturalist theory can and cannot accommodate.

The touchstone of culturalist theory is the postulate of oriented action : actors do not respond directly to "situations" but respond to them through mediating "orientations." All else either elaborates or follows from that postulate. What exactly, then, does the postulate assert?

"Orientations to action" are general dispositions of actors to act in certain ways in sets of situations. Such general dispositions pattern actions. If actors do not have them, or if orientations are ill formed or inconsistent, actions will be erratic: patternless, anomic. The idea of "orientations to action" follows a particular psychological stimulus-response model: not the simple "single-stage" behaviorist model in which nothing "subjective" intervenes between the experience of situations and responses to it (actions) but "mediational" models in which responses to stimuli (actions in situations) are considered results both of the experience of objective situations and actors' subjective processing of experience. "Orientations" do the processing. We may call them, as did Bentley, soul-stuff or mindstuff. The critical methodological task of studies based on such models is, of course, to penetrate reliably and with validity into the subjective.

Orientations are not "attitudes": the latter are specific, the former general, dispositions. Attitudes themselves derive from and express ori-

entations; though attitudes may, through their patterning, help us to find orientations. If orientations frequently occur in collectivities they may be called "culture themes," as by Mead and Metraux.^[7] Pye has distinguished four sets of such "themes" that he considers useful for making cultural comparisons on the societal level: trust-distrust, hierarchy-equality, liberty-coercion, parochial-national identifications. Putnam considers the theme of conflict or its counterpart, harmony, critical for cross-cultural analysis.^[8] These themes exemplify how orientations are general dispositions that pattern sets of actions and sets of specific attitudes. It is conventional to regard orientations as having three components: cognitive elements that, so to speak, decode experience (give it meaning); affective elements that invest cognition with feelings that move actors to act; and evaluative elements that provide goals toward which actors are moved to act.^[9]

The assumption of oriented actions would be vacuous without the addition of a second postulate, which we might call the postulate of orientational variability : orientations vary and are not mere subjective reflections of objective conditions. The significance of this postulate lies particularly in this: if the processing of experiences into actions were uniform—if it were fixed at the biological level or if it always involved "rationalist" cost-benefit calculation—then mediating mind-stuff could simply be left out of theory. In Hempel's terms, we would only need to know "initial conditions" (situations, structures) to explain actions, since we already know the universal covering law needed to complete an explanandum . No doubt ingenuity is required in relating conditions to actions via uniform orientations: the rational choice theories we have provide more than enough cases in point. But this does not alter the logic of the argument that without orientational variability we remain in a strictly behaviorist world. Similarly, if actions are merely "superstructural," we manifestly need only to know situations to explain actions. In that case, only the explanation of deviant cases (like false class consciousness) would require the use of mediating variables.

If orientations are not inherent in actors but variable, then something that is variable must form them. And if orientations are not simply subjective reflections of varying objective situations, then the variable conditions through which they are formed must themselves be cultural. Orientations are not acquired in some automatic way; they must be learned. Thus, a postulate of cultural socialization must hold if the first and second assumptions hold: orientations are learned through the agency of external "socializers." The repertoire of cognitions, feelings, and schemes of evaluation that process experience into action must be imparted by the socialized carriers of culture. The process can be direct, by "teachers" who are culturally variable actors; or it can occur indirectly simply through the experience of variable cultures.

"Rationalist" theorists do not, of course, reject the notion of political

socialization. That would be silly. What divides culturalist and rationalist theorists here involves the issue of late-in-life learning, or resocialization.

In regard to that matter, culturalists proceed from a postulate of "cumulative" socialization . This means two things. First, although learning is regarded as continuous throughout life (which is not likely to be questioned) early learning—all prior learning—is regarded as a sort of filter for later learning: early learning conditions later learning and is harder to undo. Second, a tendency is assumed toward making the bits and pieces of cognitive, affective, and evaluative learning form a coherent (consistent, consonant) whole.

The postulate of cumulative learning provides the culturalist account of how two fundamental needs of actors in societies are satisfied: the need for economy of action and the need for predictability in interaction. Life would hardly be bearable, even possible, if one had to think out every action, taking into account all pertinent information and lack of information. Orientational schemata thus save virtually all decision costs. Social life, similarly, would hardly be possible without reliable preknowledge of others' actions and of the effect of one's own actions on those of others. Without such preknowledge social life would tend to be entropic. As Crozier has cogently argued, "uncertainty" of action also begets power—arbitrary power.^[10]

Both economy of action and predictability in interaction are diminished to the extent that individual orientations are inconsistent and that early learning may readily be undone. These conditions have effects similar to a lack of orientations to actions and of socially shared orientations altogether. They lead to erratic, incoherent behavior by individuals and in social aggregates: anomie in the former; the absence of anything like a stable conscience collective in the latter.

It should be pointed out that the culturalist solution of the problems of economy of action and social predictability is not a unique solution, however plausible it may seem. Thus, in the rationalist perspective, economy of action is provided by "ideologies" or by the sensible delegation of decision-making powers.^[11] The fixity required for predictability in social life follows from the very fact that rational choice is considered a fixed disposition. If this is so, one can anticipate the actions of others and adjust one's own behavior to the anticipation. Social predictability may also be achieved through rationally formulated and enforced contractual arrangements or general legal rules. (It should be apparent that the two accounts of economy of action and social predictability provide a good basis for evaluating the relative power of culturalist and rationalist perspectives.)

To summarize: "Cultural" people process experience into action through general cognitive, affective, and evaluative predispositions; the patterns of such predispositions vary from society to society, from social

segment to social segment; they do not vary because objective social situations or structures vary but because of culturally determined learning; early learning conditions later learning and learning involves a process of seeking coherence in dispositions. And this is so in order to "economize" in decisions to act and to achieve predictability in social interactions.

The Expectation of Continuity

When the postulates of the political culture approach are made explicit, it should be evident why political culture theorists should have difficulties in accounting for political change. The assumptions of culturalist theory manifestly lead to an expectation of continuity, even in cases of changes in the objective contexts of political actions.

The expectation of continuity in aggregate (and individual) orientations follows most plainly from the assumption that orientations are not superstructural reflections of objective structures, but that they themselves invest structures and behavior with cognitive and normative meaning.

Cultural continuity also manifestly follows from the assumption that orientations are formed through processes of socialization. To the extent that socialization is direct (by precept), generational continuity must occur, the socializers being formed, "cultural" people. To the extent that socialization is indirect (by experience), generational continuity still follows; experience with authority occurs first in the family, then in schools, where unformed children encounter formed adults. In either case, what is true of one generation should continue substantially to be true in the next. This applies as much to cultural divisions in a society as to more general culture types and themes—if any exist in the first place. This, incidentally, makes the political culture perspective quite compatible with the finding that political regimes typically are short-lived.^[12]

The expectation of continuity in political cultures follows, most obviously, from the assumption of orientational cumulativeness, namely, that earlier learning conditions later learning and that actors tend to seek orientational consonance. The first allows some room for adult socialization and resocialization—but not much. The second makes unlikely the internalization of piecemeal orientational change that might increase dissonance.

But if change in culture patterns and themes are categorically excluded, political culture theory must immediately be thrown out as obvious nonsense: changes happen, including cultural changes. The saving grace of culturalist theory here is that continuity is, so to speak, an ideal-typical expectation—one that holds in an abstract , parsimonious cultural world. It is an expectation akin to that of inertia in the Galilean conception of motion. Physical inertia does not rule out changes of direction or rest, acceleration, and deceleration. It does make such phenomena depend on

contingent factors that may or may not impinge on objects in motion. Continuity is the inherent (lawful) expectation and so, therefore, is resistance to change of motion: exceptionally great forces are needed to induce great changes in direction or velocity. The notion of continuity as inertia in motivations (the psychological counterpart of physical motion) thus opens the door to culturalist accounts of change.

Through that door, however, the tendency toward improvised, post hoc accounts of political change may enter—may be bound to enter. If one's preferred theoretical approach implies a strong bias toward the continuity of culture or resistance to cultural change, then it is always tempting to extemporize theory-saving "special" conditions, or adjustments in concepts or theory to handle occurrences of change—especially major change. If, say, theoretical difficulties arise from emphasizing early socialization, then why not just relax that emphasis and assign more scope for late socialization or adult resocialization? If the assumption of a tendency toward orientational consonance makes it awkward to explain certain observations, then why not simply posit more toleration for dissonance? Or why not redefine consonance? In that way, however, one is likely to end with the term continuity meaning nothing more than "not completely (or instantaneously) changeable"—which drains the term of all reasonable meaning. This is exactly the point of Rogowski's criticism of how culturalists have in fact accounted for political change.

The remedy is to develop an explicit general culturalist theory of change, consistent with culturalist assumptions, in order to prevent ad hoc tinkering with culturalist postulates and their implications. Such a theory should state, prior to explanations of specific changes, the characteristics of change that the political culture approach can logically accommodate and those that do not fit its constraints. To formulate such a theory, I will consider two broad types of cultural changes: those arising "naturally" from changes in situations and structural conditions and those that result from "artifice"—deliberate attempts to transform political structures and behavior.

Situational Change

Pattern-Maintaining Change

Actors must often face novel situations with which their dispositional equipment is ill suited to deal. The world changes or presents us with experiences that are unfamiliar for other reasons (e.g., the penetration of peasant societies by market forces). The unfamiliar is encountered routinely in maturation, as one proceeds from family to school, from lower schools to higher ones, and from schools to participation in adult institutions. At the level of society and polity, novel situations arise from in-

ternal "development," however development may be conceived. Novel situations also arise from socially internal discontinuities (economic crises or political disruptions, like those caused by governmental instability or collapse, or from changes brought about by protest movements), or from externally imposed changes. Immigration brings actors into unfamiliar situations. So does internal migration and social mobility. The encounter of novel situations will, no doubt, occur much more frequently among individuals than on the macrolevel, but it also occurs in groups and societies.

Novel situations may be short-lived results of ephemeral upheavals. In that case no cultural adjustments are needed, nor are they likely to occur. What, however, should one expect if such situations persist?

If cultures exhibit inertia, then it should be expected that changes in culture patterns and themes will occur so as to maintain optimally such patterns and themes; that is to say, changes in culture are perfectly consistent with culturalist postulates if they occur as adaptations to altered structures and situations and if the function of change is to keep culture patterns in existence and consonant. "Pattern maintenance" (Parsons's concept) can take that form just as well as strict cultural continuity.

The French have a half-facetious adage for this sort of pattern maintenance: the more things change, the more they remain the same. The saying no doubt fits (used to fit?) France. The pragmatic masters at pattern-maintaining change, however, have been the British. Tory concessions to British working-class voters and interests are the usual case in point. Their function—sometimes "latent" but in the case of Disraeli's Tory democracy quite explicit—was to maintain Tory hegemony in the face of considerable sociopolitical change through the maintenance of as much as possible of what the Young England Circle considered the feudalistic virtues: the disposition to defer to one's betters and action by the betters on behalf of the lower orders. The point applies to reforms of the suffrage and also to the less well-known role of the Tories in the evolution of the British welfare state, which Tory governments not only have kept virtually intact but also much of which they pioneered.

An alternative to pattern-maintaining change is to subject unfamiliar experience to procrustean interpretation in order to obviate cognitive or normative change. "Perceptual distortion" has turned up frequently in experiments on how individual cognitive dissonance is handled. We know at least a little about the same way of dealing with the unfamiliar on the political macrolevel. To give just one example: party political elections in Northern Nigeria were initially regarded as a version of long-familiar elections to chieftaincy, in which the "candidates" were a small number of ascriptively defined eligibles.^[13] The extent to which perceptual distortion can be adaptive to unfamiliar experience no doubt is highly limited. How-

ever, where institutions like elections to chieftaincy exist in traditional cultures, the adaptation of dispositions to other kinds of elections should be easier than in other cases.

Change Toward Flexibility

Highly modern societies have traits that make it especially likely that actors and aggregates of actors will frequently confront novel situations. Social mobility, vertical and horizontal, is the most obvious cause. Because any changes in dispositions are costly (dysfunctional) in the culturalist perspective, one should expect, as a correlate to the expectation of pattern-maintaining cultural change, that the more modern societies are, the more the elements of their cultures will be general, thus flexible. No doubt there are considerable limits upon how general and flexible orientations can be and still perform their functions of making experience meaningful, actions economical, and interactions predictable. In more modern societies one should not expect culture to change as readily as situations and structures. Situational and structural change tend to occur with great frequency and rapidity in modern societies, and the assumption of orientational inertia postulates resistance to frequent, swift reorientation. Rather one should expect that the rigidity of cultural prescription will relax, so that culture can accommodate much social fluidity.

The tendency toward cultural flexibility can itself be regarded as a way to maintain cultural patterns and themes. As societies become more changeable, the elements of culture increasingly become "forms" that can subsume a variety of "contents." It is probably no coincidence that some sociologists early in the twentieth century (especially Simmel)^[14] adapted the Kantian distinction between form and content to social analysis. Durkheim argued much the same point directly. In early societies, he wrote, "the collective environment is essentially concrete . . . [and] the states of conscience then have the same character." ("Culture" is not a bad translation of his notion of a conscience collective .) As societies develop, the "common conscience" is obliged to rise above diversity and "consequently to become more abstract. . . . General ideas necessarily appear and become dominant."^[15]

I want to make three other points pertinent to the expectation that cultural abstractness and flexibility will grow with social development. First, the disposition to act "rationally" introduces just the kind of general and flexible culture trait that inherent social fluidity requires. (Durkheim already associated rational attitudes and behavior with the abstractness of thought necessary in highly developed societies.)^[16] The rationalization of modern life—which Weber considered to be its governing trait—thus may be an accommodation to structural conditions rather than, contra Weber, their underlying cause.

Second, the obviously difficult problem of finding a proper trade-off between two warring imperatives in modern societies, that of cultural flexibility and that of cultural fixity, is bound to be a practical difficulty, not just a theoretical one. Reconciling fixity with flexibility, abstractness, and formality may be a crucial element in what has widely been perceived as growing malaise in highly modern societies. Anomie will follow not only from lack of internal guides to action but from guidelines too general and loose to serve in the relentless particularity of experience. Highly modern society thus may be intrinsically acultural and, for that reason, transitory or susceptible to surrogates for culture—including cults and dogmas.

The expectation of cultural flexibility, finally, should apply to all highly modern societies. It thus pertains to polities initially based on rigid dogma (like communist societies) that have successfully pursued modernization. In such societies, the first expectation, that of cultural inertia, should hold. Old culture should resist new dogma. The expectation of pattern-maintaining change (or perceptual distortion) should hold as well. So one should expect also that as culture changes in such societies, it will change toward greater flexibility—and therefore to reinterpretations of dogma that make it increasingly pliable.

Cultural Discontinuity

Contextual changes can be so considerable or rapid or both that neither pattern-maintaining changes nor changes that gradually relax cultural rigidity to deal with social fluidity are possible. Rapid industrialization is the case in point usually cited. Changes resulting from war or from the formation of new polities also generally involve upheavals in social contexts. Such upheavals may result as well from economic traumas like the great inflation of 1923 in Germany (which led to far greater social disruption than the Great Depression—or possibly even the Black Death). And traumatic change sometimes strikes special segments of society rather than the whole.

We must deal, therefore, with social discontinuity, as well as "normal" change. Culturalists have tended either to avoid the matter or, worse, to treat cases of social trauma simply as "deviant cases" in which the theoretical constraints of their perspective are off—not least, the expectation of cultural inertia.

Obviously, traumatic social discontinuity will have cultural consequences different from contextual stability or less rapid, less pervasive change. Even in such cases, however, we may not simply improvise. If the assumptions of culturalists are correct, then traumatic social discontinuity should have logically expectable consequences, no less than other change.

The one consequence of social trauma absolutely precluded by culturalist assumptions is rapid reorientation. Social upheaval may overcome cultural inertia, but if so, actors should be plunged into a collective infancy in which cognitions that make experience intelligible and normative dispositions (affect, evaluative schemes) must be learned again, and learned cumulatively. No culturalist may expect, for instance, a democratic political culture to form, in a few short years, in a society like Germany after World War II, or "national" orientations to form rapidly in postcolonial tribal societies. Instead, changes in political cultures that occur in response to social discontinuity should initially exhibit considerable formlessness. For formlessness one may substitute other terms, like Durkheim's anomie or Merton's deinstitutionalization . The essence of the matter is that culture loses coherent structure. It becomes highly entropic.

The idea that rapid, large-scale contextual changes are personally disorienting and culturally disruptive is hardly new. lipset argued a generation ago that rapid economic development is associated with political extremism ("anomic protest movements" like anarchism and syndicalism),^[17] despite the fact that high levels of such development are related to political stability. Huntington later made much the same point,^[18] and Olson has probably developed it most cogently.^[19]

To say that formlessness under conditions of socioeconomic discontinuity should be "considerable" is not mere hedging. Cultural entropy can never be complete. If it were, no patterned action or interactions would be possible at all. In any case, social discontinuity never is total—intimate social units, like the family, survive the greatest upheavals (may, indeed, be strengthened by them, as refuges of predictable order); so too do structures that are supposedly merely instrumental—for example, bureaucracies. As well, if learning is cumulative, older people should exhibit a good deal of orientational inertia even when traumatic socioeconomic change occurs. We may surely suppose that the more ingrained orientations are and the more they are consonant systems, the less susceptible they are to "disorientation"—the more mechanisms like perceptive distortion will be used to invest experience with accustomed meaning.

Governmental authority will, of course, survive cultural discontinuity. In fact, it is likely to become more powerful to the extent that internalized dispositions cannot govern actions and interactions. How then do people act politically if political culture is highly formless?

We can get useful clues to answers from the growing literature on an analogous experience: how children adapt to novel situations that they enter in highly discontinuous ways: going to school, for instance, or going from one to another type or level of schooling. Much of the literature

on this subject, like studies by Wakeford and Woods,^[20] has been informed by Merton's path-breaking study of the bases of deviant behavior, which dealt in general terms with behavior under more or less "anomic" conditions.^[21]

Under conditions of cultural discontinuity, conformity with authority is still likely to occur, but it will tend to have certain characteristics. In Merton's terminology, it will tend to be ritualistic or else self-serving (opportunistic and of dubious morality, as general culture defines morality). Ritual conformity is compliance without commitment. One does what the rules or rulers prescribe, not for any discernible reason but (quoting from a lower-class British pupil interviewed by Woods) "because I behave meself . . . I just do what I'm told . . . [I] ain't got much choice." Conformity of this sort may be supposed to occur frequently in cases in which the former political cultures and subcultures prescribed high compliance ("subject cultures," as Almond and Verba called them). Self-serving, opportunistic conformity bends norms and rules for private advantage—including that of getting ahead in the competition for political power. Charles Dickens observed a lot of that sort of behavior in his travels in America as he reports them in his American Notes . Thus, in regard to a very successful businessman, "'He is a public nuisance, is he not?' 'Yes sir.' . . . 'And he is utterly dishonest, debased, and profligate?' 'Yes, sir.' 'In the name of wonder, then, what is his merit?' 'Well, sir, he is a smart man.'"^[22] I mention Dickens because one should especially expect "smart" conformity in immigrant societies or immigrant segments of society, where (as in schools) discontinuity occurs through movement into an unfamiliar but intact culture. Perhaps one should expect it even more in cultures greatly unsettled by upheaval. Thus, Burke presciently remarked (in 1790) that when cultural constraints are off, "the worst rise to the top."^[23]

More commonly than conformity, one should expect what Merton called retreatism under conditions of cultural discontinuity. Retreatism involves withdrawing from the "alien" larger society into the smaller, more familiar worlds of family, neighborhood, village, and the like. In Almond and Verba's scheme of concepts, it should show up as increased "parochialism." In the small worlds of schools, retreatism tends to involve self-imposed isolation—for instance, into remote places and daydreams or what Woods calls removal activities—"unserious pursuits which are sufficiently engrossing . . . [to make participants] oblivious for the time being of [their] actual situation"—or both.

Rebellion against, and intransigent resistance to, authority are also likely responses to the experience of cultural decay. A voluminous literature links social, economic, and political discontinuities to political vio-

lence—from Marx to Moore and Skocpol. Rebellion and intransigence, however, are always likely to be costly and call for much energy; retreatist behavior into parochial worlds or ritualistic conformity are thus more likely, especially where governing power—if not authority—is strong.

What should follow over time from contextual and cultural discontinuity? If economy of action and predictability indeed are imperatives in individual and collective life, one should expect new culture patterns and themes to emerge. But if dispositions are formed by cumulative learning, they should emerge only slowly (over generations) and, in the transitional period, at great costs resulting from raw power, withdrawal, and (because of withdrawal) forced mobilization and rebelliousness against it. Thus, the process of reformation of political cultures should be prolonged and socially costly. This is all the more likely to be the case if parochial units remain intact refuges from discontinuities in society, economy, or polity.

The expectation is logical also if older people, as is likely, cling to long-fixed dispositions even in the face of strong forces that might unsettle inertia. We might thus posit as a general expectation that in the process of cultural reformation considerable age-related differences should occur. In fact, age, in cases of pronounced discontinuity, might even be expected to be a major basis for subcultural differentiation. If indeed this were found to be so, the cultural perspective on theory would be enormously strengthened over alternatives. Empirical work pertinent to the expectation, however, is lacking; and as culturalists have built adult learning increasingly into their approach in order to accommodate illfitting facts, the incentive to inquire into age-related cultural differences, in both established and transitional contexts, has regrettably declined.

I want to make another point about the reformation of dispositions and culture patterns, more briefly. As the young should be more susceptible to reorientation than the old, so one should expect to find in social macrostructures particular segments that have traits especially conducive or susceptible to reorientation. By "conducive traits" I mean structural or dispositional traits readily accommodated to new culture patterns or, indeed, anticipations of them. In Western traditional societies, for instance, there always existed a large island of achievement in a sea of ascription—the celibate clergy, which hardly could be ascriptively recruited. The clergy, in fact, played a considerable role in the emergence of modern political institutions—despite their stake in the distribution of traditional privileges. Similarly, socially "marginal" groups—groups that occupy the fluid interstices of established cultures—should be highly susceptible to reorientation, thus "vanguards" in the reorienting of unsettled societies. There is a good deal of literature making the case that this is indeed so.^[24]

Political Transformation

By transformation I mean the use of political power and artifice to engineer radically changed social and political structures, thus culture patterns and themes: to set society and polity on new courses toward unprecedented objectives. Transformation, typically, is the objective of modern revolutions. It can also be the objective of military conquerors and of nation builders or other modernizers. Revolutions, however, provide the most unambiguous and dramatic cases. I will therefore confine my remarks to them—though what is said about them should also apply to transformation attempted in other ways.

Hannah Arendt undoubtedly was right in arguing that attempts at revolutionary transformation are distinctively modern^[25] —that revolutions as we think of them (not mere rebellious attacks on authorities or their actions) begin with the French and American revolutions. As long as political and social structures were considered divinely ordained, or natural, or simply the ways of a folk, the idea of their deliberate transformation hardly could occur. "History" then could only be endless repetition or an intrinsic progress toward a preordained end. Societies and polities could no more be transformed than the heavenly bodies set upon new orbits. One of the decisive traits of modern societies then is the belief that a "new beginning"—a felicitous and not redundant expression—could be made in political and social life.

Initially, making a new beginning did not seem to call for much artifice—no more, perhaps, than a proper constitution. Achieving liberty or equality throughout society simply called for setting polities and societies on their inherently right course—right, given human nature. For reasons not necessary to sketch in the age of the "God that failed," really making revolution—not seizing power but the accomplishment of transformation—came increasingly to be seen as a task, and a difficult task, for political artificers. Unfortunately, systematic studies of that process are few, although the exceptions often have been notable: for instance, Massell's study of Soviet attempts to bring Soviet Central Asia into modernity,^[26] and Kelley and Klein's study of the effects on inequality of the Bolivian Revolution of 1951.^[27] Inquirers into revolution still are hooked on the issue of their etiology.

Since revolutions are themselves major discontinuities and since they generally occur in periods of social or political upheavals, not least governmental breakdown,^[28] the expectations listed in the preceding section should apply to transformation. But I want to state here some expectations that follow from the culturalist perspective especially for processes of revolutionary transformation. Intrinsic interest and contemporary relevance

aside, these processes seem to me especially critical for evaluating culturalist theories and their bases. After all, transformative processes involve not only adjustment to necessity but also the deliberate engineering of great change, and they are typically backed by great power and control.

As a first expectation we may posit that revolutionary transformation is strictly impossible in the short run. Revolutions certainly bring upheaval. They may also be expected to bring about movement in the direction of their professed goals by readily accomplished actions—instituting wide suffrage, kicking out the landlords and redistributing land, ending feudal privileges and obligations, and the like. But if discontinuity begets "formlessness" of culture, then revolutionaries can hardly do much to reorient people in the short run (say, in a generation or so). Reorientation is, of course, the less likely the more intact is the prerevolutionary culture: the more it provides parochial refuges from transformative power or institutional centers of resistance to it. But even if revolution only reflects discontinuity instead of engendering it, the expectation stated still should hold.

If the conventional norms and practices of political life are disrupted by revolution, what can be put in their place? We may posit the answer that revolutionary transformation will initially be attempted by despotic or legalistic means. What, after all, could "order" societies and polities in place of conventional, internalized culture? Only brute power, or else the use of external legal prescriptions as a surrogate for internal orientational guides to behavior. "Revolutionary legalism" was in fact a device used early after the Bolshevik seizure of power, and it overlapped a good deal (even before Stalin) with attempts to "storm" society (especially its more backward parts) with head-on "administrative assault." Neither, according to Massell, accomplished much toward the realization of transformation; responses to it, he writes, included "avoidance," "selective participation," "evasion," "limited retribution," and "massive backlash."

"Legalism," it might be noted here, is likely to be a general response to massive cultural disruption, whether revolutionary or situational or both. Indeed, it can become, in highly unusual cases, a persistent surrogate for normative culture—indeed, a culture form. I have argued this elsewhere,^[29] defining "legalist" cultures as cultures in which legal rules are widely known, such rules are widely used (instead of justice or prudence) to justify political standpoints or decisions, legal actions are the normal mode of dealing with conflicts and disputes, and therefore laws deal in highly detailed—if possible, comprehensive—ways with social interaction and tend to be punctiliously adhered to. Durkheim argued the even more general, related proposition that in the course of development civil law (which regulates social interactions) constantly grows, while criminal, or

restitutive, law declines.^[30] His argument makes sense if indeed development loosens normative cultural prescription, as I argued, and lessens cultural similitude, as Durkheim argues.

The case I used to make this argument is contemporary West Germany. That, we should note, also is the case Rogowski mainly relies on to argue that reorientation can occur rapidly—the crucial point in his critique of culturalist theory.^[31] Rogowski seems to me to miss the real import of "deviant cases"—that through their very abnormal characteristics they can be used to shed light upon the factors that condition typical cases.

What of the long-run prospects of revolutionary transformation? I suggest the expectation that the long-run effects of attempted revolutionary transformation will diverge considerably from revolutionary intentions and resemble more the prerevolutionary condition of society. The expectation is not that little change in "content" will occur: in who holds power, gets privilege, and so on. No inevitable Thermidorean Reaction is posited. The argument is somewhat less categorical: reconstructed culture patterns and themes will diverge widely from revolutionary visions and will tend to diverge from them in the direction of the patterns of the old society and regime. The degree to which the expectation holds obviously depends on the extent to which the old culture was already in disarray.

Several points made earlier lead to this expectation. Culture must still be learned on a comprehensive scale, as in all societies; and although revolutionary teaching can no doubt play a considerable role in shaping the young, it can hardly replace socialization in small parochial units. Nor are teachers or role models likely to be, extensively, the sort of marginal individuals who are steeped in revolutionary dogma as a surrogate for convention—or people for whom the revolutionary vision has much meaning at all. Sheer cultural inertia will also play a role in the process of revolutionary decay; so will the tendency toward turning change into pattern maintenance—perhaps by a progressive transformation of revolutionary visions into mere revolutionary rhetoric; so—to the extent that the new rulers succeed in modernizing—will the tendency of modern cultures to be general, abstract, and (especially pertinent here) flexible; so will "retreatist" and "ritualist" responses to discontinuity; and so will the tendency of opportunistic conformists to get ahead, by scheming or approval, in unfamiliar contexts.

In fact, it may well be the case that the short-run effects of attempted transformation are greater than the longer-run effects. More can be done in upheaval than when life again acquires fixity. Kelley and Klein have argued precisely this point, on the basis of generalizing the case of the Bolivian Revolution of 1951.^[32]

Whether all this also entails the expectation that in the longer run incremental change will accomplish more than attempts at radical trans-

formation we can perhaps leave an open question here. But note that the rulers of the Soviet Union came increasingly to view the achievement of cultural change as a matter for what they called "systematic social engineering" for—as Massell describes it—"a pragmatic commitment to relatively patient and systematic social action, wherein at least as much time and effort would be devoted to the building of bridges to traditional society . . . as to actual and direct confrontation with the traditional system."

Conclusion

It may well be the case that the political culture approach has been used to explain political changes in the sort of ad hoc and post hoc manner that saves—and thus weakens—theories rather than testing and strengthening them. Culturalists hardly have a monopoly on such theoretical legerdemain—certainly not when compared to rational choice theorists—when discomfiting facts confront them. But I have tried to show here that culturalists must have a strong propensity toward improvised theory saving when dealing with political change, since their assumptions lead, necessarily, to an expectation of cultural continuity—at any rate in a "pure" (abstract, ideal-typical) cultural world, where all matters falling under "ceteris paribus" are in fact "equal."

Nevertheless, it should be evident that a cogent, potentially powerful theory of political change can be derived from culturalist premises. The theory sketched here specifies that changes in dispositions, in response to contextual changes, should be pattern-maintaining changes or—if the contextual changes involve modernization—changes toward normative generality and flexibility; that in response to abrupt social discontinuities cultural dispositions should, for a considerable period, be "formless"—incoherent in individuals and fragmented in aggregates; that in such cases retreating into intact parochial structures occurs, while conformity should become ritualistic or opportunistic; that revolutionary artifice cannot accomplish cultural transformation in the short run; that such transformation will be attempted by despotic power or (mainly hopeful) legal prescriptions; and that, in the longer run, attempts at revolutionary transformation will tend to be regressive or at least have quite unintended outcomes. Note, however, that nothing here rules out engineered change, so to speak—attempted structural reforms of politics. In the modern world, political tinkering, on small or grand scales, is endemic. The theory simply states what should result from such tinkering.

The problem of testing the theory against experience obviously remains, as do problems of operationalizing concepts for that purpose. But theory comes first.

If the power of a culturalist account of political change is to be com-

pared with that of different approaches to political theory and explanation, then general accounts of change, derived from noncultural postulates and similar to that presented here, are needed. Political-culture theories, admittedly, have not heretofore met the challenge of developing a general theory of change; but neither have other types of theories.

Appendix:
Culture

The term culture , unfortunately, has no precise, settled technical meaning in the social sciences, despite its centrality in them. The variable and ambiguous use of key concepts generates unprofitable arguments that are merely definitional. Hence I append a note that places my use of the term, as sketched in the first section, in its conceptual context.

My use of the term culture tries to make explicit, at the axiomatic level, what is implicit (occasionally almost explicit) in the works of Almond and his various collaborators. Their use of the concept seems to be based squarely on Talcott Parsons's "action frame of reference." Parsons first worked out that "frame of reference" as a way of synthesizing four apparently diverse, all highly influential, early modern social scientists: Mar-

Figure 7–1.
Interaction in the Action Frame of Reference

shall, Pareto, Durkheim, and Weber.^[33] He and collaborators developed action theory in a large series of works, the most useful of which probably is the multiauthored book. Toward a General Theory of Action .^[34]

The action frame of reference is based, at the microlevel, on Parsons's notion of an interaction, societies being complexes of interactions (some earlier sociologists called them acts of "sociation"). The notion is depicted on Figure 7–1. In brief translation, (1) ego (an actor) is in a "situation"—an objective context; (2) ego cognitively decodes that context and invests it with feeling (cathexis)—thus the context comes to have meaning for the actor; (3) the manner of investing situations with meaning is acquired through socialization, which consists mainly of early learning—this imparts the modes of understanding and valuing prevalent in societies or subsocieties or both, and in aggregate, these may be called a society's "culture"; (4) socialization leads to the internalization of cognitive and affective meanings (viz., the cultural becomes personal) and their institutionalization (the definition of expected behavior in social roles and of sanctions in case of deviation from expected behavior) makes smooth and regular patterns of interaction possible; (5) cognitions and affective responses to them define goals and ways to pursue them; (6) cognitions, feelings, and goals are communicated to alter (another actor) through the use of "signs" (symbolic expressions of culture that make ego's actions intelligible to alter)—but actions also depend on objective facilities that are part of any actor's situation and that independently affect the choice of goals; (7) alter responds, changing the situation in some respect, so that the process resumes.

Note especially that the action frame of reference emphasizes neither subjective nor objective factors but rather how the two are linked in interactions. Culturalists focus on the matters in the box on the right, but they should also bring that on the left into interpretation and theories. This I have tried to do throughout this essay, emphasizing how culture conditions change in varying contexts of objective change.

Alternatives to the notion of culture I use come chiefly from cultural anthropology. I use the plural intentionally because the meanings of culture vary a great deal in that field. One can probably subsume these meanings under four categories: (1) Culture is coterminous with society: it is the whole complex of the ways of a "folk," of human thought and action among particular people. Park comes close to that view.^[35] (2) Culture is social life in its subjective aspects: the knowledge, beliefs, morals, laws, customs, habits of a society. One finds this meaning (and these illustrative words) in the seminal work of Tyler and, later, Benedict and Kluckhohn.^[36] (3) Culture is what differentiates societies from one another, for the purpose of idiographic description but also for theorizing through compar-

and contrasts (agreements and differences). I take the seminal work here to be Malinowski's.^[37] (4) Culture is the distinctive, variable set of ways in which societies normatively regulate social behavior.^[38]

The fourth set of meanings comes closest to that used here. My use of the concept of culture here seems to be justified by usage in political science and, more important, by its suitability to testing theories through the catholic deduction of unknowns once its postulates are explicitly stated. Anyway, my version of the concept is that about which theoretical conflicts have thus far occurred in political inquiry.

Eight—
"Observing" Political Culture

Author's Note : This essay was completed just before this book was prepared and is published here for the first time. Because it deals with the second main criticism of "culturalist" theory, I have adjoined it to the chapter outlining a culturalist theory of political change. However, it deals with a major methodological issue and thus also belongs in the earlier section on "Political Science."

Introduction

The political-culture approach to theory and explanation has, in just one generation, passed from undoubted domination of macropolitical studies, to doubt or rejection, to what appears to be an early renaissance—even if in more or less modified forms.^[1]

Doubts about the approach no doubt arose from its too-quick popularity, its rapidly acquired faddishness.^[2] More important, they arose for two much more serious reasons: the apparent inability of culturalists to account for political change without cheating; and the fact that culturalists made fundamental something subjective, hence not immediately observable, which they variously called orientations to action, predispositions, or (dubiously) attitudes. In essence, they made central to the construction of "positive" political theory, akin to theories in the natural sciences, what had long been considered the principal barrier to doing so by the German Geistes - or Kulturwissenschaftler (cultural scientists, as conceived by Dilthey, Windelband, and Rickert) who influenced Max Weber's conception of verstehende Soziologie . This barrier was the meaning, or understanding, of experiences and actions to actors (cognitive, affective, evaluative, as usually

broken down), governed by "rules" that are always culturally defined,^[3] thus probably unique, and acquired by learning within cultures.^[4] One reaction to this emphasis on subjectivity (on highly variable "lifeworlds," as Husseri called them) was a sort of crusading antipositivism: the reduction of social science to the mere "interpretation" of cases. The rejection of social science on the ground of its subjectivity was pushed under more or less fancy labels: hermeneutics, phenomenology, semiotics, ethnomethodology, or "thick description."^[5] In able hands, like Geertz's, studies in this vein produced fascinating information and striking insights, but at extreme costs in reliability and validity (in their technical senses), not to mention parsimony and predictive power. Compared to earlier reactions to the philosophers of cultural science, especially Weber's, the later reactions argued, in effect, for throwing out both baby and bath water.

A second reaction also was extreme. In gist, it amounted to solving the problem by circumventing it. For the most part, this was done by rational-choice theorists, who assumed a universal disposition toward "efficient" action in cost-benefit terms. Culturally learned preferences would, at some point, have to be poured into utility functions, but not necessarily and certainly not initially.^[6] To a lesser extent, and less clearly, the difficulty also was evaded by the various forms of "structuralism," drawn from the diverse sources of Lévi-Strauss, Piaget, and, of course, structural linguistics. We might note, however, that structuralism, rather strangely, has been symbiotically related to the "phenomenological" reaction.^[7]

Now a substantial return to culturalist theory seems to be in process. But this is not due to the fact that culturalists have solved the decisive problems that caused earlier doubts about their approach. The renaissance of culturalist theory appears, in most cases, to be more a reaction against the earlier reactions. The revival of the approach occurred, in part, because "phenomenology," for all intents and purposes, abandoned positive theory, and also because clever formalistic exercises upon rational-choice assumptions are still a far cry from validity (and have been empirically impugned—e.g., in coalition-theory) as well as being based on dubious, or highly time-bound and, of all things, culture-bound assumptions. To justify the renaissance of culturalism, more is needed than the shortcomings of other modes of analysis. The problems that made culturalism doubtful must be plausibly solved, if we are not to go through this cycle of rejection and counterrejection all over again.

In an earlier essay I tried to deal with the issue of political change within the culturalist framework (see chapter 7). Here, I want to confront the second issue: how to deal with "meaning" for purposes of "positive theory."^[8]

The Subjectivity Problem

In the work that started political-culture theory, Almond and Verba, after granting that culture is a word overburdened with meanings in social study, define it succinctly as "psychological orientation toward social objects . . . the political system as internalized in the cognitions, feelings, and evaluations of its population,"^[9] instilled by socialization. Despite the large variety of meanings attached to the concept, before and since, "psychological" (that is, subjective) orientation has always been central in the idea of culture. To be sure, terminologies differ. For instance, Geertz writes about instilled "programs" that govern behavior, as against behavior per se,^[10] and Barnes identifies culture with "shared assumptions about how the world works and should work."^[11] Kaase used terms like Einstellungen (attitudes) and Prädispositionen as central to culture, as against Verhalten (behavior) or Handeln (acting).^[12]

Recent revised versions of political-culture theory retain the centrality of the subjective. The latest of these versions is "grid-group" culture theory, first developed in the work of Mary Douglas and in Douglas and Wildavsky on risk,^[13] and fully developed in a forthcoming work by a team headed by Wildavsky.^[14] "Grid-group" cultural theory à la Douglas and Wildavsky essentially does two things. It posits, first, a small number (five) of general clusters of "preferences" for ways of political life, to reduce (make "parsimonious") the vast variety of particular cultural traits and themes; Almond and Verba also did this, differently, in their work of 1963. Second, it proposes a theory of political viability: polities can survive if, and only if, social preferences and social relations are congruent. The preferences (viz., "ways of looking at the world") manifestly are general normative dispositions and thus subjective. It seems plausible to hold that the perception of congruence between preferences and actual relations also will largely be culturally defined, by cognitive dispositions—like those that led French revolutionary crowds to rally to the king, as the (perceived) people's guardian.

Political-culture theories, then, in all forms, focus inquiry on subjective matter. Anything subjective can, at best, only be indirectly observed in the phenomena we experience. This raises questions. Can subjective dispositions be observed at all? (Reichel,^[15] for example, has severe doubts about this; he holds that the decisive problem in political-culture research is "uncertainty": that is, problems of determining just what empirical observations are to be considered as manifestations of political culture.) If dispositions are to be considered observable, then how? And how rigorously can they be observed: how reliably and validly, in the technical methodological senses of these terms?

We may be tempted to dismiss these problems too easily, in the manner of Hartz on the role of "ideas" as data in political research.^[16] Very briefly, Hartz argues that positivist study is study of the "existential" (what exists); ideas exist; ergo, political ideas are as appropriate a subject for positivist study as any other: quod erat demonstrandum . This is simplistic. Hartz happened to be concerned with ideas that are the observable artifacts produced by political philosophers—with what the old-style behaviorist, Bentley,^[17] called "writing-activities." These undoubtedly "exist" and are directly observable, even if often hard to decipher correctly. But the matter is not so simple when one deals with silent and implicit orientations, dispositions, perceptions, especially on the macrolevel; they are not to be found in libraries or bookstores. There may be strongly suggestive evidence that orientations to action do "exist"; Hyman's seminal work on political socialization,^[18] among many others, provides heaps of it. But a version of Lord Hewart's dictum about justice applies here: facts, for positive study and theory, must not only exist; they must be seen to exist; they must be observable facts. The issue is not put correctly by Hartz. It is not whether ideas are facts, but whether they are observable and rigorously so.

In regard to cultural dispositions, the old behaviorists had a point not to be ignored. They held that one can observe only objective behavior and objects, and that ideas, attitudes, dispositions are, in Bentley's pungent words, nothing but "mind-stuff," "soul-stuff," or "spooks" imputed to behavior. Note that this was never meant to say that people have no subjective dispositions. It meant that if inquiry focused on the dispositions it must inevitably be infirm (unbestimmt , in Reichel's word), especially; to a considerable extent, arbitrary and intuitive. Not the least difficulty discerned was that subjective explanations would tend to be truistic and, in the philosophical sense, trivial (mere words, saying nothing). The danger of triviality in political-culture studies is in fact serious. It exists chiefly in the tendency of culturalists to state observed patterns of behavior in a cultural language and then to claim that doing so somehow explains the patterns. It has, for instance, been observed that some countries have multiparty systems. Why? A typical political-culture explanation is that the countries have fragmented political cultures. What is the evidence for this? Often, in large part, that they have multiparty systems. In the same vein, some countries experience frequent political violence, supposedly because they have "cultures of violence," for which the chief evidence is that they experience frequent political violence. The fallacy in this sort of "explanation" is evident. Political-culture theories tend to be circular: statements in which an explanandum (something to be explained) is identical with a proposed explanans (an explanation). Otherwise put, political-culture ex-

planations have tended to be translations into a special language, and no more.

The subjectivity problem that afflicts political-culture studies thus involves two issues: (1) Either political culture simply is a pattern of political behavior in a society or subsociety—in which case, why talk about political culture in the first place? Or (2) political culture is something else: the dispositions that underlie the activities and generally pattern them. In the latter case, back to square one: how can one observe the general dispositions with reasonable assurance of validity?

Political Culture and Survey Research

The standard solution of the subjectivity problem in political-culture research has been to use a special technique that supposedly taps orientations directly: survey research into political attitudes. The leading macrolevel example is by Almond and Verba,^[19] whose study of democratic cultures was largely based on an interview schedule of 131 items, many subdivided into numerous subitems: the schedule occupies twenty-three large pages of an appendix to The Civic Culture . Putnam's excellent book on the "beliefs" of politicians in Britain and Italy is another case in point, using both a large number of closed questions and the coding of replies to openended ones.^[20] A plethora of other examples could be given. The most intriguing, perhaps, is a sort of canned, standardized questionnaire proposed by Berg-Schlosser for determining the political cultures of any and all societies (which, if it made sense, would, of course, be very useful for comparative study).^[21]

The idea that underlies the survey-research solution of the subjectivity problem is familiar. One uses artfully devised questionnaires and interview schedules to obtain data considered to reveal unmistakably and directly people's general political orientations or more specific attitudes. Such research is considered particularly appropriate if one deals with large aggregates of respondents, so that psychological testing or even more arduous techniques for getting at perceptual maps are not feasible.

Much is said for the usefulness of survey research in numerous studies.^[22] As a solution to the subjectivity problem in political-culture research, however, the technique poses substantial difficulties, two of which, I think, are insuperable. I deal with most of them only briefly here because the matter is much discussed in the cookbook literature on social science methodology.

First, there is a practical difficulty: survey research is extremely costly. The resources required for it are unlikely to be available to most researchers. One should not dismiss this difficulty because it is only "practical," because the social sciences are, and are likely to remain, the poor relations of research funding. Sometimes it prevents research altogether; often it

forces researchers to make do with what can be done, not what they ought to do.

A second difficulty, practical but less readily soluble, involves access to respondents. Not all polities are open to survey research (especially by Americans), some by law (as in the case of Burma), more often because of obstacles by officials unable to distinguish between research and intelligence work, inquiry and snooping. In some cases, also, potential respondents are too fearful and suspicious to be good subjects for surveys.

A third difficulty, about which much has been written, is intrinsic in survey-research technique: there are nearly always doubts about the trustworthiness of responses to even the best designed survey instruments. Converse has discussed one especially serious reason for this,^[23] the expression of "nonattitudes": attitudes about matters regarding which subjects lack the knowledge to have attitudes in the first place. There are many others. Respondents, if suspicious, may give intentionally misleading responses. Or they may give answers they consider expected of them (a special kind of "nonattitude"); this is especially prevalent among school children who confuse questionnaires with tests. Or responses may vary with contextual factors; for example, there is the well-known phenomenon of deceptively high levels of political knowledge just before general elections. The wording of questions may contaminate responses, as may the presence or demeanor of researchers. And so on. Survey researchers have been ingenious at guarding against the difficulty posed by misleading responses but, of course, imperfectly so. The basic problem is not that distortions may occur, but that it is hard (indeed impossible) to know whether, and to what extent, they have occurred—unless one already knows what surveys are conducted to find out.

The absolutely insuperable difficulty posed by the survey-research solution is logical (and epistemological); that is what makes it insuperable. Responses to survey instruments still are "behavior" that only supposedly represents orientations. They undoubtedly provide more data than mere "raw observation" and conceivably more revealing data. But there is no getting around the fact that they are, in Bentleyan terms, "speakingactivities" or "writing-activities," not the subjective thing itself.

There is an ineluctable "epistemic gap" (as Northrop called it)^[24] between objective behavior and subjective dispositions. Certainly, survey research cannot bridge it; logically no observational technique can.

These arguments may seem glum, but they at least allow us to state the decisive problems in observing political culture. There are two. First, how can culturalists get more reliable, more trustworthy data than by survey research? Second, how can the apparently unbridgeable "epistemic gap" between behavior and orientations be narrowed, even if not eliminated, so that its existence makes little difference to theory?

A Proposed Solution of the Subjectivity Problem

Obtaining Data

If any single technique for getting data about a subject is likely to be flawed, it seems plain ordinary sense always to use a variety of such techniques. The results can then supplement one another; up to a point, they might cancel out errors that arise from each technique used singly; they provide checks on one another; and if they produce overlapping findings, one may be especially confident that something correct has been found. Campbell has called this procedure the use of multitrait-multimethod matrices.^[25] The method is analogous to geological "triangulation": measuring heights not otherwise measurable, or distances not readily traversable, by the use of a series of imperfect measures (of triangles from points on a baseline). That method is, in fact, sometimes used in preference to direct measuring, even where that is possible, because of its exactness.

In trying to get at culture, a considerable variety of different measures is, at least potentially, available for the similar purpose of observing the apparently unobservable. One of them, for all its imperfection, is survey research. There are many others. A large number of textbooks have been written about them. Without being comprehensive, I will discuss a few.

One method designed to get at the inherently unobservable is content analysis, a once widely used and quite sophisticated technique that seems now, unaccountably, to have lapsed into disfavor and even unfamiliarity in many cases. Its purpose was the quantitative analysis of communications, using explicity defined categories to capture their content; the use of the categories was binding in analysis to prevent the arbitrary selection of materials that might merely strike inquirers as intuitively "interesting"; and the measurement of the categories was carried out by appropriate, rigorous procedures. Newspaper editorials and speeches were the sort of materials studied, in all cases to find the "meanings" (mostly values) they conveyed. For instance, Lasswell, Leites, and their associates devised a "symbol analysis" that was used during the Second World War by several branches of the U.S. government to analyze the contents of newspapers, noting the frequency of the symbols they used and whether positive or negative meaning was attached to them.^[26] White similarly studied the prewar speeches of Hitler and Roosevelt, classifying every value statement (5,326 of them) under predefined categories.^[27] A good overview of the content-analysis technique, although it emphasizes attitudes in international relations, is provided by North et al.^[28]

Content analysis, of course, has many flaws of its own, including problems regarding the predefined variables used in carrying it out, problems in measuring them accurately, and problems of the "significance" of find-

ings because it is only too easy to get hooked on counting as such. Not least, it poses sampling problems; for instance, it seems heavily biased toward elite communications, such as speeches by conspicuous political leaders. However, its use, as stated, is recommended here only as a flawed way of deciphering meanings, among other such ways. And content analysis has undoubted advantages over survey research in getting at cultural meanings. One is that it is designed to get at implicit meanings of which subjects themselves might not be aware. Of course, details and nuances that idiographers usually catch get lost in the mechanical use of preselected variables and methods of measurement, but all techniques of generalizing, even about a particular society, pay that price—and the generalizers know it and consider it worthwhile.

A method that political science culturalists seem somehow to have missed, yet that seems tailor-made for deciphering meanings, is Osgood's "semantic differential" technique; its nature and rationale is discussed comprehensively in Osgood, Suci, and Tannenbaum's The Measurement of Meaning ,^[29] and the technique is applied, almost entirely by Osgood and his collaborators themselves, in a long-sustained series of works that covers some three decades.^[30] The technique is a type of projective technique. It is used to elicit responses to stimuli—in this case, cards that present words which are genuine opposites, like honest and dishonest, not short and tall—in regard to some object (a picture, for example) that may potentially reveal meanings with a minimum of deception by respondents (as in the case of nonattitudes) and contamination by the presence of observers.

Again, I refer to the semantic differential technique only as part of the arsenal of techniques available to get at meanings. However, that method seems to me to be especially notable because it has in fact produced enlightening findings; for instance, to cite just one, Israelis react more intensely to words denoting activity than to those that evoke strength, whereas Arab cultures seem to react intensely to cues that involve strength but not those that denote activity.^[31] On the face of it, this finding illuminates much experience. (After all, Nasser seemed genuinely surprised that his threats to drive the Israelis into the sea were taken at face value in 1967; he was, after all, only flexing muscles.)

Greenstein and Tarrow have used what they call a semiprojective technique that seems particularly suitable for studying children's dispositions because children are likely to confuse questionnaires and the like with tests, but their technique obviously has wider applicability.^[32] In gist, subjects are told the beginning of a story and then asked to continue it; their responses are then subjected to more or less systematic analysis. To illustrate, one story, designed to get at dispositions toward political authority began with this cue:"The [head of state] has to attend an important meeting. He [she] is late, so he [she] drives to the meeting over the speed

limit. A policeman stops him [her]." Here is part of how a twelve-year-old French girl continued, after considering that the president, then de Gaulle, might be issued a speeding ticket:

Girl : But they must let it pass, because it is the president. . . . He tells them he is late, that he has to go to a meeting. Perhaps the gendarmes let him go, telling him to go on less rapidly, because he could have an accident. Let's say that he pays attention to their orders, if he is afraid, or perhaps he does not pay attention to them saying to himself: "It is not right that a gendarme should give me orders." And maybe he will tell his chauffeur to speed up , or at any rate not to slow down.

Interviewer : What do you think? Would he say go slower or go faster?

Girl : Perhaps in order to be safe he slows down a little, for he says to himself: "If there is an accident, I will not be there any more to direct the country [la patrie ]."

Compare how a British working-class girl of the same age continued the story:

Girl : Well, really the police wouldn't really stop the queen driving the car because somebody would be escorting the car to the meeting and they would stop the rest of the cars so she wouldn't really need to be going fast. People have got time to wait for the queen. She wouldn't really be late.

Interviewer : Let's suppose for the purpose of this question that the queen was driving by herself in a car and she wanted to go out and get away from other people and the police stopped her for going too fast.

Girl : Well, the police might have stopped her, but as soon as they saw it was the queen they wouldn't take her in for speeding or anything like that.

Interviewer : What do you suppose the policeman would say?

Girl : I think he would be astounded, because really the queen, she doesn't drive the car on her own. And she knows Britain's speed limits, so she wouldn't have gone really fast.

Or take the way an American worker's son elaborated the scenario (not, to me, very reassuringly):

Boy : The policeman was giving him a ticket and everything. And once he saw it was the president he told him he could go and that. And the people that were standing around thought it was unfair and everything. So they started running after the car. They were pulling the bumper and that. They were wrecking it, and they forced the policeman to give him a ticket. He was late for his meeting.

Interviewer : Well, why did they think it was unfair that he didn't get a ticket?

Boy : Well, because the president is just a person and everything.

Interviewer : Well, what did the president think about all of this?

Boy : He thought it was all O.K. He thought it was fair that he should get one.

Interviewer : What does the policeman think when he sees that he has stopped the president's car?

Boy : He thinks he'd better let him go or else he'll get mad and that and start telling the chief and he'd fire him or something.

These responses surely tell one a lot, even on just an intuitive basis, and Greenstein and Tarrow do subject them to intensive analysis; but they also use rigorous quantitative techniques on the aggregates of responses.

Not to overlook the obvious, there also is "raw observation" (though we may feel better if we give it a label that sounds more dignified, like "participant observation") to provide clues to cultural meanings. A strange notion seems to be abroad in the social sciences: that such observation does not provide "data." When I used it in my own work, I was accused by a survey researcher of being merely "anecdotal"—not "scientific."^[33] Natural scientists, it seems to me, would regard this notion with disdain. Granted that they use vastly complex mechanisms and techniques of observation. They do not use them merely to observe, but to observe more and better: a necessity at a stage of development we hardly have reached. Moreover, raw observation can be carried out thoughtfully, even with considerable rigor (see the literature on participant observation). And it has several advantages over special techniques of observation designed to decipher the subjective. It is not artificial and is therefore unlikely to be either mistaken by subject or contaminated by observers; and, unlike survey research, one can do it almost anywhere.

In general, then, findings obtained by a variety of techniques always are worth more than those obtained by any single mode of observation, where all such modes are inadequate. Genuinely rigorous "triangulation," of course, requires one to know precisely the nature and extent of measurement errors in particular imperfect observations, as surveyors do. We know a little about this in regard to some of our methods; for instance, we know that, in community-power research, asking people questions like, Can you tell me who has political power here? will lead to findings that exaggerate considerably the concentration of power in the community. But we know how modes of inquiry distort observations in too few cases, and, where we know, we know too inexactly. However, the principle of the matter is clear: progress is possible; difficulty is not a good reason for not trying; and although it might be foolish to hope that we can attain the precision of geological triangulation, we can approximate it—but only if we try.

Narrowing the Epistemic Gap

The greater difficulty, the crux of the issue of whether political culture is "observable," is how to make negligible the epistemic gap between behavior and orientations (that is, meanings, dispositions, attitudes, mind-

stuff. The solution proposed here is, of course, applicable, mutatis mutandis, in all inquiries that pose the subjectivity problem.

To begin: One must never pretend that subjective orientations have been directly revealed by any technique of observation or by any set of such techniques. There simply is no way around the point that orientations are always imputed to behavior. This, however, is not (yet) a concession to the narrow behaviorists, for the simple reason that imputations may be valid. The issue we have come to can therefore be restated in a form clearly more conducive to solution: How can subjective orientations be validly imputed to behavior? In regard to this, we already have discussed the principal fallacy to be avoided if a plausible answer is to be found. It is the statement of circular explanations: the mere translation of objective observation into a cultural language. Such statements, as I have said, trivialize problems: the explanatory statements simply restate explananda ; they add nothing to them; and they allow us to deceive ourselves into thinking that we have said something when we have said nothing.

The trivialization of theories by translation and circularity is unavoidable if a particular orientation to action is imputed to any particular activity. We simply cannot say, for instance, that a person participates because he is participant. Political orientations must be regarded as dispositions that underlie, thus organize and pattern, sets of perceptions and activities: the larger and more diverse the sets the better for theory. This is inherent in the very conception of "orientations to action" anyhow. They are, as I wrote in an earlier paper, "general dispositions of actors to act in certain ways in sets of situations. [They] pattern actions. . . . Orientations are not 'attitudes': the latter are specific, the former general dispositions. Attitudes themselves derive from and express orientations; though attitudes may, through their patterning, help us find orientations."^[34] (For attitudes here one can, of course, substitute actions or behavior.) If, for example, we say that "X is tolerant," we say that X does not object (or, better, is oblivious) to sets of traits of people or of their behavior widely disapproved in a collectivity. The set might include Jews or Catholics, blacks or Asians, illegal immigrants or guest-workers, homosexuals or eccentrics, self-absorbed or stupid people, and the like; it is unlikely to cover all people or behaviors (say, child abusers: everyone draws a line somewhere), but the statement describes a general pattern of matters directly observable, or else it says nothing.

If orientations are so regarded, then statements about them or sets of them (cultures or subcultures) are statements of "regularity": they are lawlike statements or, more simply, theories. They are never descriptions, not even "thick" ones—though I do not mean to deride the clues to cultural theory that acute observation of a particular activity (say, a Balinese cockfight) may provide.

This, by the way, is rather an old view, dating to the origins of contemporary social science, although recent theorists of culture, and others who stress "meaning" in social inquiry, seem unaware of that fact. Weber, the apostle of verstehende Soziologie , for example, wrote, in his posthumous magnum opus:

Every interpretation aims at self-evidence or immediate plausibility. But an interpretation which makes the meaning of a piece of behavior as selfevidently obvious as you like cannot claim just on that account to be the causally valid interpretation as well. In itself it is nothing more than a particularly plausible hypothesis.^[35]

Weber's point was restated by Morris Ginsburg in 1956:

It appears to be a basic assumption of verstehende Soziologie  . . . that what we know within our minds is somehow more intelligible than what is outwardly observed. But this is to confuse the familiar with the intelligible. There is no inner sense establishing connexions between inner facts by direct intuition. Such connexions are in fact empirical generalizations , of no greater validity than the similar generalizations relating to outward facts.^[36]

This has a crucial implication, which Weber also pointed out. If one is to have confidence in the validity of statements about orientations, then they must be treated like any other hypothetical assertion. At the outset, statements about culture may simply be summaries of observations: verbal summaries or the scientific-looking ones we find in regression tables, frequency curves, histograms, smoothed ogives, and so on. Political culturists usually stop at that point—if they get to it. But theory, to be considered valid, must be tested; and statements about cultures, or anything else, cannot be tested by the data from which they were derived in the first place—even if the nature of a frequency distribution, or our intense immersion in a culture as participant observers, may justify the intuitive surmise that tests will validate or invalidate, so that they might not be worth the cost of being carried out.

If one cannot test theoretical statements by data used to formulate them, it follows that one must posit, by logically deduced prediction, as yet unobserved data that should turn up, given certain "initial (or determining) conditions,"^[37] if our assertions about orientations or cultural sets of them are to be considered valid. (Weber mistakenly thought that the validity of imputed meanings can only be determined by psychological tests—which are still a way of getting directly at the subjective.) Such predictive tests, positing unknowns, can be carried out in two ways.

First, one may predict behavior in situations either that have not yet occurred or, if they have, to which the responses are still uncertain. I will give a real-life example (that happened to turn out well) from a work of

mine that, as mentioned above, was criticized by an eminent authority for using "anecdotal material."^[38]

In my study of Norwegian authority-culture, I argued that Norwegians are consensualists in decision making. More specifically, it appeared that they follow this decision rule: agree widely (unanimously, if possible); if wide agreement is not possible, then appear to be in wide agreement; if that also is not possible, then drop the subject. This is a statement about subjective cognitions—"actions widely agreed upon are likely to be right actions"; about affect—Norwegians "feel" uncomfortable with small majorities; and about evaluation—Norwegians value wide agreement, and processes of forming consensus thereby become their decision processes. Thus, the statement contains all the elements into which Almond and Verba break down orientations.^[39]

The statement is a (low-level) theoretical generalization, based on a large and eclectic variety of mostly raw observations, from the behavior of small neighborhood and club committees to transactions of parliamentary business. Admittedly, the generalization was illustrated by what may accurately be described as "anecdotes." In an essay replying to this point, among others,^[40] I pleaded guilty to anecdotalism but also pointed out that the generalization was based on much else; that empirical generalizations should be based on the best obtainable data; but that despite their basis, what really counted in the end was how such generalizations stood up to appropriate testing by the accurate prediction of unknowns.

As it happened, a "natural" occasion for such prediction had arisen about the time that this debate was going on. In 1965, a general election brought to power what Norwegians called a bourgeois coalition, after thirty years of social-democratic (Labor party) rule. This coalition consisted of a very odd assortment of parties that had in common perhaps only opposition to the Labor party's long dominance. The coalition combined urban, and urbane, business interests with an agrarian party (in a country in which the rural-urban split looms large); nonreligious, if not downright antireligious, parties joined a Christian fundamentalist party; and a small party, chiefly of white-collar intellectual liberals, was added to the odd alignment. The common wisdom in political science is that such coalitions cannot last. They defy the now common rule that coalitions must minimize "preference disagreement," or "policy distance," merely to form.^[41] The common wisdom in Norway, academic and general, was the same. However, the common notions ignored the decision rule I have summarized. That rule led to a quite different expectation, stated in my paper of 1967:^[42] the coalition would endure by simply avoiding divisive issues (like religion or urban and rural interests likely to clash); it would work smoothly by concentrating on the least divisive issues (such as almost universally desired

tax reforms); it would do little or nothing to change the Labor party's welfare-state legislation (Labor still had about 40 percent of the parliamentary seats—hardly a negligible minority); and it would break down if absolutely forced by circumstances to deal with a deeply divisive issue (for example, an issue that involved what Norwegians call "cosmopolitanism": close connections with the continent, divisions about which originated, no doubt, in the long colonial domination of Norway by Denmark). As it happened, all these predictions turned out to be correct. They thus validated—failed to falsify, despite defying both plausible theory and sheer good sense—the cultural generalization. The coalition lasted, although it did little besides tax reform, until the question of joining the EEC came up as an unavoidable issue; it then disintegrated.

The case was, in one way, lucky. Something happened in Norway at just the right time to allow a "natural experiment." That probably is only rarely the case. Usually, one may surmise, similar tests do not yet exist; they must be awaited; and they may simply not turn up, anyway not for a long time—quite as in the natural sciences.

Because that is so, a second way of testing whether cultural "meaning" has been validly deciphered may be desired over the first. One may carry out such tests by predicting responses to stimuli controlled by inquirers—not least, by doing survey research into as-yet-unknown responses by logically predicting the findings that should turn up if one's conceptions about a culture are valid. If in fact one has conceptions of culture, however arrived at—multitrait-multimethod "triangulation" is still best—then it should not be difficult to devise questions or other stimuli to test the conceptions. This is the best use, in my view, to which survey research intended to get at orientations can be put, both epistemologically and practically. Models of political culture can be built in all sons of ways, less expensive in funds, time, personnel, and so on, than typical survey researches. Questions can be few if carefully devised for testing (especially if not used in attitudinal fishing expeditions), and samples may be small if thoughtfully selected for a carefully defined purpose. Unfortunately, neither survey research nor other modes of inquiry into culture have ever been so used. Their findings have been treated as definitive results, not as steps toward such results. But survey research, and other techniques, could be so used, at low cost and with potentially high payoffs.

Perhaps one other step that is desirable, even if not absolutely necessary, still is missing. What, in effect, we have, if all the above is done, is simply a tested hypothesis, essentially like any other. To narrow the epistemic gap between meaning and observation, it would still be useful to establish more firmly that a regularity exists because of subjective orientations rather than other factors. One can do this in at least three related ways: by multivariate

analysis intended to establish the relative strengths of direct contextual factors (e.g., SES) as against factors associated with processes of socialization; by comparing the strength of synchronic and diachronic factors in mature subjects (because orientations are supposedly instilled chiefly by experiences early in life and upon my argument are highly resistant to change);^[43] and by observations that might, or might not, establish that adult behavior is relatively stable despite changes in its context: something already done, by summarizing much psychological research, in Herbert Hyman's seminal work on political socialization.

Conclusion

All this is not meant to say that we can directly observe subjective matter by following the recipe here proposed. The epistemic gap between "inner" and "outward" facts, as stated, cannot be more than narrowed. But surely that is of no great consequence for the normal purposes of everyday academic inquiry: constructing, testing, and using theories.

It is especially worth adding that the solution of the subjectivity problem here proposed is exactly similar to how natural scientists treat unobservables, like time, electricity, or magnetism. No one in the "hard" sciences shies away from such concepts simply because they are not directly observable but experienced only through their effects—and in that sense are "spooky." They are, however, treated as theoretical constructs; their nature is imputed to a large number and variety of observations, using various research procedures; and conceptions of their nature have been subjected to careful tests. By now, they seem "real" in the objective sense; one would no doubt be held for an idiot if one were to contest paying one's electricity bills on the ground that one should not be charged for figmental "mindstuff." The construct works in all sorts of ways, and so we, and the utility companies, may safely consider it "real," for our, and their, purposes. Indeed, we must—or else dismiss much good theory.

I make this point to emphasize the assertion that nothing about positivist inquiry compels dealing only with direct observables. That Skinnerian view belongs to primitive and naive positivism; its resemblance to what goes on in the hardest sciences is not even coincidental. Nor does cultural meaning condemn us to hermeneutics or any of its relatives. In "observing" culture one must comply with only one imperative, though it is burdensome: one must be highly cautious, ingenious, rigorous, and, above all, skeptical about one's own ideas. But this is imperative in all researches.

Nine—
Explaining Collective Political Violence

Author's Note: Aristotle's Politics singled out three crucial problems of political theory, after long empirical and taxonomic discussions. Two are macrolevel problems: the causes of "revolutions" and political stability. One is a microlevel problem: civic education, comprising both what we would recognize as pedagogy and socialization.

Aristotle's emphasis on revolutions and their obverse had autobiographical foundations in the upheavals that occurred in Athens of the fourth century B.C.E. Inquiries into revolutions have occurred largely in reaction to immediate experience ever since. Following the headlines has been an affliction of political science, an obstacle to genuine accumulation; and inquiry into revolutions is a particularly good case in point. It has waxed and waned, in modern times, as revolutions themselves have been obtrusive or not. Much was written about the subject after the French Revolution and its aftershocks, after 1848, after the Russian Revolution and Stalin's "revolution from above," and, in my time, after Vietnam and other highly conspicuous insurgencies.

My own inquiries into the subject obviously also stem from elements of autobiography, as stated in chapter 1. Still, I have tried, in this case and others, to transcend immediate and personal experiences by inquiring into revolutions on a general level, recognizing that most significant events of current relevance pose universal problems and that particular events may best be illuminated by general theories. Yet much writing about the subject since the 1960s still seems tied to the headlines. Much of it seems

This article appeared in Handbook of Political Conflict: Theory and Research , ed. T. R. Gurr (New York: Free Press, 1980), 135–166. Reprinted by permission. Copyright © 1980 by the Free Press, a division of Macmillan, Inc.

to proceed from the premise that the phenomenon, as it concerns us, began only after the Second World War. And only a few inquirers (for example, Skocpol) have been led from current concerns back to earlier phenomena. (That we need history to understand recent events was the point of an article of mine recently published—coauthored by Raj Desai—"Insurgency: The Transformation of Peasant Rebellion," World Politics 42 (July 1990]: 441–465.)

I point this out, in part, to place my own involvement in studies of revolutions into perspective. It began at a luncheon with the associate director of Princeton's Center of International Studies, shortly after I went to that university in 1959. Distinguished work in the center (originally at Yale) had, by that time, been done in international relations and security studies, but the center had for some time also cultivated comparative political studies. Not long after joining it, at lunch in early 1960, the associate director asked me what, in my view, might be a good subject on which a center project might be constructed. I mentioned that revolution might be such a subject. It had a long pedigree in scholarship, but nothing significant of which I was, and am, aware had been done with it since Brinton's Anatomy of Revolution in 1938. I pointed out not only the classic nature of the subject but also that we were living in a revolutionary period; the Greek, Malayan, Algerian, and Indo-Chinese insurgencies (and some lesser ones) had already occurred, and the fact that they did not immediately involve the United States seemed to me beside the point. To show why it was beside the point, I argued that in a time of nuclear standoff, revolutionary conflict might reasonably be expected to be used, or at least be attempted to be used, as a surrogate for all-out war—not necessarily that the Great Powers would foment the conflicts, but that they would become involved in them, perhaps dangerously so. This is not hindsight after Vietnam; I made the same point in a publication that appeared in 1962.

The response to the suggestion was that, with the associate director's blessing, I should go forth into the world to try to raise money for the project.

I then began a Grand Tour of funding agencies, governmental and private. At first I was optimistic about obtaining funds; the events mentioned had happened, the arguments for renewed study seemed compelling, and we had started our own involvement, even if still barely perceptible, in Vietnam. But, only a short time before that involvement became vast, no one cared. John Gardner, then at the Carnegie Foundation, did, in the end, seem to see the point; he provided $ 90,000 for a three-year project. (Two or three years later, millions were being spent on research into insurgency—more correctly, counterinsurgency.)

The project started in 1961. We called it the Internal War Project. This

calls for explanation. I had long been uneasy about Brinton's choice of cases in Anatomy of Revolution , less because he included the American War of Independence, which bothered many people, than because he included the English Civil War. Oranges and apples should be distinguished—unless, of course, one studies fruit. The worst outcome of inquiry, surely, is discarding a perfectly decent theory because it does not fit a case it should not fit in the first place. More generally, "revolution" seemed to be a term used for a rotten pot of quite dissimilar events: coups, shortlived riots, rebellions to remedy grievances, long drawn-out and intense events to change all of society, anticolonial (and/or pro-Marxist) guerrilla wars, not to mention supposedly major peaceful changes, like the Industrial Revolution. In the project, we were concerned with violent internal conflicts broadly speaking, but, to avoid conceptual confusion, I decided to use the long-established French notion of guerre intérieure , simply because it had no conventional meaning in English. "Revolution" could be treated as a subtype of the genus, among others—a subject on which I wrote a long (unpublished) paper to orient research in the project. The tendency now is to label the genus differently: collective political violence. I have no quarrel with that, as long as "revolution" and other types of violent events remain a well-defined species under the genus so that one can theorize at both the more general and more specific levels. Perhaps it should be added that my unease about revolution as the general label was vague but became crystallized by Hannah Arendt's On Revolution (1963).

The project's first activity was convening a conference of outstanding scholars to help us get a grip on the subject. The conference was a gathering of glittering All-Stars in social science. For me, its outcome reinforced my conviction that the social sciences, as they then were, had little to offer on the subject. Considerable pressure was put on me to publish the conference papers and to write an introduction to them. I did not really want to do so, although, in view of the stature of the participants, it would have been easy to go into print. Pressure prevailed, and a book appeared (Internal War: Problems and Approaches [Westport, Conn.: Greenwood, 1964]). My introduction to the book was painful to write; it is a thinly veiled criticism of the book's contents, in the form of an agenda for required research on internal wars, which covers everything from "pretheoretical" conceptual problems to the issue of the outcomes of such wars: their shortterm and long-term effects on societies.

Subsequently, I wrote and published a more specific, orienting paper on the causes of internal wars, for the project ("On the Etiology of Internal Wars," History and Theory 4:2 [1965]: 133–163; not included in this book). In retrospect, both the book and article seem to me weak; but the book was used for an extensive period (including at the Army War College), and

the article was republished many times. I take this as an indication of the sudden hunger for academic studies of revolutionary violence, no doubt generated by the Vietnam war.

The Internal War Project was woefully underfunded. Still, it produced considerable results. Here is a sample. The project launched the work of perhaps the two most influential contemporary theorists on collective political violence, Ted Robert Gurr and Charles Tilly, whose divergent views are summarized here. Gurr was brought into the project because he had already written a dissertation on the subject, and Tilly was asked to join it because of his excellent book on the Vendée. In addition, the project spawned Paret and Shy's influential book on guerrilla warfare of 1962 (one year after Osanka's dreary compilation of pieces on the subject); Gil AlRoy's monograph on peasant involvement in internal wars; Janos's study of coups; and Victor Wolfenstein's psychoanalytic study of revolutionary personality, the first of its genre. Much, not all, of the voluminous work on collective political violence published in the 1960s and 1970s grew out of work begun in the project, directly or indirectly. I take pleasure in that. I would take more if, as well as directing the project, I had had resources to do research of my own in it.

In this chapter, I try to take stock of where studies of collective political violence had taken us by 1980. The chapter may be read as a jaundiced essay, but it was not intended to be gloomy; research rarely leads to the discovery of the son of critical juncture of alternatives discussed in the essay. That we cannot yet tell which alternative is likely to be the more fruitful is no disgrace.

Between problems and solutions in "positive" (if readers prefer, "scientific") inquiry there lies a crucial step. If this step is ignored, as is usual, cumulative theory building will almost certainly not occur. Instead, one ends up with an accumulation of alternative, untested, generally ambiguous hypotheses. The step I have in mind involves the most fateful aspect of what, in an earlier essay on political violence, I called "problemation":^[1]the discovery of the most fundamental problem requiring solution if a progressive development of theory about a subject is to occur . In this essay I try to show what that basic problem is in studying collective political violence.

Large numbers of social scientists have studied political violence since the early 1960s, when the subject, after one of those long hiatuses that characterize its study, was back in vogue. Production of work has been anything but scant: Zimmerman's magisterial review of the literature lists about 2,400 items,^[2] most of them published since 1960 (though he includes some stones perhaps better left unturned). It might seem odd, then, that an essay should now be devoted to a discussion—not even, as readers will

find at the end, a solution—of a "basic," a "primary" problem. But this is not at all odd. The discovery of primary problems usually culminates much work in positive study: it is a critical and difficult achievement. Before core problems can be defined with precision, there usually is much prior observation, speculation, debate, and, especially, diffuse dissatisfaction, a sense of growing mystery rather than of illumination. One gradually comes to see, through long groping, the basic puzzle that a subject presents: where to begin if a genuine unfolding of theory is to occur. Two decades seem a long time to get to that point, but it usually takes much longer (though afterward progress is swift).

The most basic problem in studying any subject is to choose a fruitful theoretical approach. Usually that choice lies among many candidates, in which case a still more basic problem is to narrow the choice down to viable options. This chapter deals with basic alternative approaches to theories about collective political violence. What then, first of all, is a theoretical approach?

In essence, an approach is, of course, a route or tactic one chooses to follow toward a new or difficult objective. All routes depend primarily on starting points. The starting point in theorizing is always axiomatic in nature, explicitly or implicitly: it involves assumptions deemed fruitful. Any such assumption is an initial commitment that is considered vital if one's goals in inquiry are to be achieved effectively. It is fairly common to call this a "perspective" of inquiry. A perspective is the core of a theoretical approach, which, when fully elaborated, also includes key descriptive concepts, key secondary problems, and the specification of appropriate methods—all matters that depend on the choice of a fundamental perspective and that are more subject to change through feedback from research. A perspective usually takes the form of specifying the basic nature of a subject matter. For instance, physical events are "shaped matter in motion," or markets are "patterns of interaction among rational valuemaximizers." Alternatively, a perspective may spotlight a key variable: one facet of a complex subject that is most important to grasp. Since all that constitutes a theoretical approach follows from its axiomatic basis, this essay asks: what are the crucial alternatives among basic perspectives that may be used to elaborate theories about collective political violence, and how can we choose among them?

The discussion cannot be brief, for the subject is enormously consequential, but space dictates that I omit much that ideally should be included. I ignore essentially descriptive case studies, still the main genre in the field, for my interest is in theory. Since my concern is with macrotheory, approaches to the explanation of individual aggression also are not considered here;^[3] because of the linkages between micro- and macrotheory, this is more regrettable. The chapter is concerned with process (especially

one aspect of it), and so omits ideologies and personality traits of revolutionary leaders.^[4] To make its coverage of the literature manageable I deal with contemporary studies, and only a sample of these.

More seriously, I skirt issues of definition (or, better, delimitation), thus ignoring my own early advice to students of the subject.^[5] But issues of definition and operanonalization have always been too intricate for brevity. Early on, the problem was a lack of considered definitions. Now we suffer from overabundance and too much diversity. In any case, the fate of competing definitions will turn in the end on the promise of opposing theoretical approaches to a tentatively conceptualized subject, rather than on abstract discussion: the issue of the "essential nature" of a subject must ultimately determine its conceptualization.

For my purpose here, I assume that there are no fundamental quarrels with the following definitional notions:

1. collective political violence involves destructive attacks by groups within a political community against its regime, authorities, or policies;^[6]

2. revolutions are the extreme cases of collective political violence, in regard to (a) their magnitude (scope, intensity), (b) targets (the political community or "regime"), (c) goals (degree and rapidity of change desired), and (d) the extent to which there is conflict between elites and counterelites.^[7]

Most important, I emphasize one theoretical problem, that of "etiology": why does collective political violence, in general or in particular forms, occur, and why does it occur at different levels of magnitude and intensity? That problem has certainly held center stage since about 1960, while the study of other phenomena (the "process" of revolution, issues of prudent action by authorities or rebels, determinants of outcomes, problems of postrevolutionary rule) have waxed and waned. Not least, the issue of etiology is the problem on which theoretical approaches now differ most, especially if we include in it the problem of why political violence takes different forms. And we may surmise that the solution of this problem has important repercussions for all others,^[8]

"Contingency" versus "Inherency"

In a very early essay on the etiology of collective political violence—the label then used was "internal war, "following French usage in the eighteenth and nineteenth centuries and the language of the Federalist Papers^[9] —I discussed a number of options in explaining its causes. Some of the choices to be made were between:

1. "preconditions" or "precipitants"—more remote or more proximate causes

2. "incumbents" or "insurgents"

3. "structural" or "behavioral" (cultural, attitudinal, psychological) factors

4. "specific occurrences" (say, economic depressions) or "general processes" (long-run patterns that may occur in numerous theoretically equivalent forms)

5. "obstacles" to collective political violence or "positive" factors that make for internal-war potential.

The result was a tentative model (not empirically grounded like that in Hibbs) in which internal wars are explained by complex balances of very different and logically heterogeneous factors.

The theme of such complex balances constantly recurs in the literature. Gurr's "simplified" model of the determinants of political violence lists seven factors that may act to enlarge or lessen its magnitude.^[10] Another version of his model list three proximate determinants, but also nineteen factors that determine the values of the more general determinants.^[11] In Hibbs's causal universe, positive and negative factors run amok. About thirty factors are directly or indirectly linked to coups, collective protests, and internal wars.^[12] The occurrence of these events involves the interplay of all the factors, facilitative and obstructive.

Undoubtedly, all concrete events, physical and social, result from the confluence of just such numerous positive and inhibiting factors. The business of positive theory is not to reproduce this complexity of the concrete, because that route takes one back to what makes experience mystifying in the first place. The task of positive theory is to cut through puzzling complexity to illuminating essentials: to find parsimony within intricacy. How can we do this after so much preliminary study on collective political violence?

The way to start, surely, is to find the most basic branch point for choice in theorizing. My thesis here can be put in a sentence: in studying collective political violence, the first and most fateful choice lies between regarding it as "contingent" or "inherent" in political life.

Nature of Contingency and Inherency

At the outset, I propose a broad thesis about basic branch points in building positive theories, regardless of subject. It seems historically true that primary branch points in theoretical inquiry are all alike: all involve a choice between contingency and inherency. Why so I will try to show momentarily. First, we must understand in a general way the nature of the two notions.

Something is contingent if its occurrence depends on the presence of

unusual (we might say aberrant) conditions that occur accidentally—conditions that involve a large component of chance. An auto accident clearly is contingent in this sense. Drivers may or may not make mistakes; cars may or may not fatefully malfunction. It is with reason then that we call such occurrences "accidents." Note immediately that contingency does not entail indeterminacy. We can specify that if a particular driver does something, an accident will probably occur. We can also determine general conditions that increase or decrease the probability of accidents. Contingencies thus are not random, and "may-may not" events can sometimes be controlled. Such events, though, do raise questions of explanation and theory in a special form: one wants to know what caused an accident where "normally" none was expected to occur. We are not mystified when a driver gets from here to there without malchance. We want to explain when the opposite occurs. Contingency implies unexpected, something out of the ordinary, something not understood without special explanation.

Per contra, something is inherent either if it always will happen (e.g., entropy) or if the potentiality for it always exists and actuality can only be obstructed. Just when the inevitable occurs or hindrances are removed is decided by contingencies: chance occurrences that hinder or facilitate. As contingency does not entail randomness or inability to control, so inherency does not imply fully predictable determinacy. The decay of an automobile surely is inherent, even without any accidents. When or how it will fall apart, though, is not fully predictable. But basic questions of explanation diner in cases of inherency: we usually want to know why the inherent did not occur sooner, what obstructed or delayed decay and "termination." In contingencies, then, the puzzle is "why"; in inherency it is "why not?"

There is a certain sense in which one could always say that contingencies are inherent in entities or events and thus score debating points. No car, no accident. If you drive, it is always possible that something untoward will happen. An accident could be considered one of a driver's inherent "repertoire" of possibilities. (My terminology comes from a theory of collective violence.) But that would be as fatuous and misleading as saying that walking into the path of a meteorite is included in my inherent repertoire of living. One could say, just as fatuously, that the final decay of cars generally is contingent, for something circumstantial usually does end them. Everything ends in a circumstance, but obviously it would be disastrous to throw out the second law of thermodynamics on that basis. The reinterpretation of contingency into inherency is verbal flimflam; the two are antithetical.

Nevertheless, the silly statements are of some help here. In the concrete world, contingency and inherency are almost always intertwined and hard to disentangle. What seems manifestly contingent to one observer may seem just as obviously inherent to another. This has been the case with

studies of collective political violence. Occurrences, at bottom, must always be regarded as the one or the other. Consequently the issue of contingency versus inherency always arises. Both always occur in a mix hard to disentangle, but they cannot, logically, both be equally basic. Any theory that supposes the contrary, or avoids the issue, must end by making experience illogical, hence unintelligible.

Note three matters before the argument is taken further. First, the distinction between contingency and inherency is often stated in different terminologies. The most common alternative language involves abnormality, irregularity, or aberration versus normality or regularity. In social science, the same distinction has been made as an antimony between "causal" and "purposive" behavior;^[13] Hempel's term for the latter is "dispositional."^[14] The distinction is between actions caused extrinsically (by contextual factors, and thus contingent) and actions chosen "intrinsically" (upon will, tactics, calculation, or other disposition in actors) from a repertoire of alternatives. Related to the notions of aberration and normality is the concept of "continuity." Thus we have been told that war may be treated as the extreme end of a single political dimension: routine politics continued by other means. War, of course, may also be regarded as lying in a realm separate from regular politics. As stated, these various differences are nominal. Those I employ here are rooted in long philosophic usage.

Second, precisely because contingent and inherent conditions are in most cases mixed in concrete occurrences, the issue usually is not whether something is the one or the other, but whether a subject is better regarded as basically inherent or contingent. The issue is not, literally, truth but explanatory "fruitfulness": the ability to explain cogently numerous matters through logically related propositions; to allow the subsumption of narrower theories under broader ones; perhaps, above all, to allow the deduction of good, previously unformulated theories. There have, to be sure, been decisive tests of the contingent or inherent nature of certain experiences. Usually, however, they occur long after the experience. (Between Copernicus's brainstorm and Galileo's tests, about sixty years elapsed; between Galileo and Newton, about a century.) In the meantime, one can only choose more or less sensibly—and try to devise projects that allow wiser theoretical choices at the crux.

Third, all kinds of occurrences (of collective violence or anything else) assumed to belong to a general "class" need not be contingent or inherent (see my remarks on this in the conclusion of this chapter). But if answers as to contingency or inherency differ by type of event, it is just about certain that the events do not belong to a single genus. Initially, though, let us assume that occurrences of collective political violence do constitute a broad type, in order not to sacrifice parsimony before we must.

Examples and Consequentiality of the Contingency-Inherency Distinction

The nature of our primary branch point can be illustrated by analogies to several very disparate fields.

Physical science furnishes the best-known example. At the divide between ancient and modern physics are divergent ways of regarding a critical property of objects: their motion. Nothing physical can be explained without a theory of motion, and thus also of rest. Motion can be regarded in Aristotle's manner: as the result of force exerted by (contingent) external movers, so that objects are "normally" at rest, or tend toward it. The alternative—inobvious, but correct—is that motion inheres in objects, so that rest is a condition resulting from a confluence of contingencies. External movers do affect motion, but they affect only its "accidents" (velocities, paths, rest), not motion in itself. It should be evident—as it was to the Inquisition—that the distinction is fateful in the extreme for physical theory, cosmology, and theology, even though particular motions are usually explained fully only by a combination of intrinsic and extrinsic factors. One thing that makes the distinction fateful, as suggested earlier, is that it poses different primary problems. For Aristotelians, the crucial problem involves the occurrence of motion: hence their labored teleological, to us absurd (but long believed) account of gravity. For Galileo the equivalent puzzle is rest, as itself a kind of motion. It would be nicely convenient to put Aristotelian and Galilean notions together in a single formulation. But since objects cannot be both "inertly" moving and unmoving, the result would be messy and illogical, in contrast to the power and elegance of modern physics.

In the life sciences, there exists an analogous core problem. Death (termination, entropy) is inexorable. But how should one regard illness and other disturbances of "normal" functioning? One possibility is the familiar bacterial explanation and its extensions: diseases result from the invasion of organisms by virulent microorganisms, and disabilities result from diseases, or from accidents, or from the consequences of other external matters (like dominant parents or competitive siblings). Diseases and disabilities are thus contingent—pathological. The alternative is to regard them as particular routine states of the living system: stress on systems is always present; the system defends and usually maintains itself through homeostatic devices, sometimes to avoid disturbance altogether, sometimes to recuperate; thus both illness and health essentially are intrinsic matters of the state of biological systems in interaction with their contexts. As one ages, of course, one becomes more vulnerable to stress, and homeostasis is more difficult to achieve. That does not resolve the fundamental problem: which version of disease and disability, the contingent or the inherent,

is the better for all or some pathologies as the base of theory? There is no single, agreed-upon view on this issue.

The current debate in structural linguistics between adherents of "deepstructure" theory and "empiricism" furnishes another example. Its consequentiality lies, of course, in that it raises the fundamental issue of the very nature of speech. In politics, as I have argued elsewhere, the analogous general branch point is between "culturalist" theories that explain political actions basically by (contingent) learned "orientations," and "rationalchoice" theories that postulate an inherent tendency to maximize influence. In studies of social stability and integration, which is close to our subject here, there is a basic confrontation between considering the "normal" state of society to be harmonious or conflictual—for instance, between functionalists and systems theorists, on the one hand, and class theorists, on the other.

Two Antithetical Explanation-Sketches for Collective Political Violence

How would a contingency theory about collective political violence compare with an inherency theory? At this point, we need broad "explanationsketches" to illustrate the opposing approaches; in the next section I summarize actual theories that should fit one or another of the sketches.

Explanation-sketches consist of an initial spelling out of laws and initial conditions—scientific explanantia —to be filled out and made into a fullfledged theory through research "for which the sketch suggests the direction."^[15] Explanation-sketches are more than the tentative commitments of theoretical approaches, but much less than final statements of theories.

Contingency theory should conform to the following sketch:

1. The fundamental disposition of individuals (or groups) in politics is toward "peace": the resolution or avoidance of violent conflicts. There would be no governments otherwise (see "contract" theorists). Satisfaction of political values is normally sought through pacific competition (electoral, through interest groups, by petitions, et cetera). Violent conflict is not in the normal "repertoire" of political competition.

2. The disposition toward pacific politics may be blocked and diverted under specifiable and "special" (aberrant) conditions. Given the disposition toward peace, the conditions should not readily occur, least of all in extreme forms of conflict. Collective violence thus involves the blockage of inherent tendencies by peculiar causes.

3. The critical problem in studying collective political violence thus is why it occurs as often as it does.

4. As to "peculiar causes," the pacific disposition may be blocked when

some other, discomfiting human disposition (which governments exist to suppress) is activated. This may be a disposition toward aggression or it may be a disposition toward comparing one's condition in life with that of others.

5. It follows that choices of collective political violence are highly "affective" rather than coolly calculated.

6. The tendency to act violently in politics may be increased by cultural patterns—learned modes of action (these are always variable and "contingent," of course). Violent action may be a learned response; and to the extent that this is so, pacific dispositions are more readily diverted.

7. Given the affectivity of collective political violence, two factors should play a rather minor role in its explanation (though they may play some role as "mediating" variables that reduce or increase probability). These are coercive balances between incumbents and their opponents and other factors that facilitate the successful use of violence.

Readers can construct an explanation-sketch for inherency theory by inverting the contingency-sketch, but it will save effort if an equivalent framework is constructed explicitly here:

1. The fundamental disposition of individuals (groups) in politics is to maximize influence, or power, over decisions. This disposition may flow through numerous channels, of which collective violent action is one: extreme but "normal."

2. Since there are alternative channels for seeking power, the choice of violence must be activated, but activation readily occurs—though, of course, not as readily at the extreme of revolution. Collective political violence is a normal response to commonplace conditions.

3. The critical problem for inherency theory, given the normality of violence, is why collective political violence does not occur more often than it does.

4. The activation of the choice of violent channels is a matter of tactical considerations (not arousal of virulent affect).

5. Tactical choice involves cost-benefit calculation. Thus, the violent mode of political competition is chosen if lower-cost channels of influence seeking are blocked, provided that violent means have a prospect of success that warrants their use. For extreme cases (revolutions) the ideal combination is blocked alternative channels, including those of lower-level violence; high valuation of goals; and perception of low capacity by opponents to inflict high costs.

6. Cultural patterns should play only a minor role; and to the extent

that learning plays a role, it should inhibit violence at least as much as promote it, by teaching people that it is a high-cost resource.

7. More objective factors, like coercive balances or facilitating factors, should play a major and primary role in explaining collective political violence.

It should be evident that the two sketches intersect at some points, so that the allure of unparsimonious combination is, as always, considerable. But it should also be evident that what is primary, important, necessary in one case is secondary, minor, chancy in the other. Most important, the sketches lead in quite different directions in research (in logic and, as we will see, in practice): most patently, toward conditions that arouse exceptional types and degrees of affect (especially anger) versus conditions that influence calculations of cost-benefit ratios in choosing modes of political goal-seeking (especially intrinsically high-cost channels). Most fundamental, as in all political theorizing since ancient times, are two antithetical conceptions of political man: as a creature in search of either peace or power.

Major Illustrations From Studies of Collective Political Violence

We proceed to theorists who illustrate the opposed approaches. The theories do not add much to the explanation-sketches. We do not yet have "finished" theories of collective political violence, but we do have "evolving" theories; unfortunately, they are becoming more and more complex and logically messy.

Contingency: the Relative Deprivation (RD) Family of Theories

Contingency theories of collective violence pivot on the notion of systemic breakdown where homeostatic devices normally provide negative entropy. It has been pointed out, correctly, that this implies sharp discontinuities between routine and nonroutine political activity, that the cause of violent action must be discontinuous (rapid, extensive) change in the context of politics and that collective and individual behavioral pathologies should significantly covary, where the former is a "version" of the latter.^[16] Almost all such theories are subsumable under the notion of relative-deprivation theory, of which Gurr has been the leading exponent.

"Why men rebel." Gurr's RD model can be summarized thus: (1) collective political violence is a form of aggression; (2) aggression results from anger, which is produced by frustration; (3) the fundamental cause of feeling frustration is an imbalance between what one gets and what one considers one's due: in Gurr's language, "discrepancy between men's value expectations and their value capabilities."^[17] Obviously, the propensity to

feel frustrated and its consequences are in a special sense "inherent." However, it is a dormant disposition until aroused by special extrinsic forces strong enough to overcome the tendency toward pacific acquiescence. The greater the scope and intensity of RD, of course, the more likely is violent behavior per se as well as at high "magnitudes."^[18]

The above is only a first step. Aggression is not yet rebellion. It must be politicized if it is to appear as collective political violence, and latency must become actuality. Here, mediating (secondary) variables that do not themselves involve the frustration-anger-aggression nexus come into play. They include "normative justifications" for political violence or the lack of them, from Sorelian glorifications of violence to Gandhi's doctrine of nonviolence at the other extreme.^[19] Such justifications are themselves, of course, contingent, and unlikely to have consequences without prior frustration.^[20] Also included are "utilitarian justifications," which are chiefly tactical considerations: estimations of the rational sense in collective violence. These involve calculations of numerous balances concerning the organizations of regimes and dissidents, their respective resources (actual and potential), and the availability of alternative channels of action. A third mediating variable decides whether politicized aggression surfaces as fully actualized collective violence, and involves something obviously tactical: the balance of coercion between regimes and dissidents.^[21] The relation of that balance to magnitude of violent political conflict is curvilinear: strife will be greatest if there is an even balance of coercion. At the extremes, where coercion is highly unequal, regimes collapse virtually without being pushed or dissidents lie low out of fear or are quickly put down. Gurr also has at times invoked still other factors, especially environmental conditions that facilitate strife:^[22] transportation networks, geographic traits, demographic characteristics—and, not least, the external support given dissidents.

In a contingency theory, such factors should themselves depend on rather fortuitous circumstances—as they do in Gurr. More important is an implication that must be read into relegating the factors to the inferior status of mere mediating variables. Causal-path analysis aside, the implication is that the role of tactical variables diminishes as the more fundamental factor of frustration grows: desperate, impassioned people will not act coolly or be much governed by tactical calculations, even about coercive balances. This is the only logical way to combine rationalistic with essentially nonrational motivation. Nonrationality also implies that a major role be assigned to culturally variable learning. This too occurs in Gurr's theory, the cultural variable being the extent to which a culture of violence, rooted in the past, exists.^[23]

Similar theories . Gurr's theory belongs to a large family. Tilly traces its ancestry to Emile Durkheim—though Tilly's treatment of Durkheim (whose

puzzle, after all, was solidarity in a differentiated society, not conflict) is debatable.^[24] More obvious precursors among the great sociologists are Gaetano Mosca and Vilfredo Pareto. Pareto traces the decline of elites to the contingency of insufficient cooptation of dangerous, competent members of the nonelite, those in whom the deep-structural residues of combination (organizational skill) and force (ability and will to use coercion) coalesce. The exclusion of such men produces in them a kind of political RD.^[25] Mosca propounded Pareto's theory of elite-circulation earlier, though less elaborately. He also argued that the resort to violence is often a reaction to the estrangement of elites from masses: an elite's adoption of foreign ways—a kind of cultural deprivation. Whether or not such estrangement occurs is, of course, no more intrinsic to elitism than is the exclusion of a dangerous counterelite. In Pareto, especially, the most obviously symptomatic indication of contingency, blockage of a normal process, is central.

Among contemporary writers, the most influential member of the family, next to Gurr, probably is Huntington.^[26] As in Pareto, the sense of deprivation in Huntington's theory is political, though less a matter of blocked channels than of their paucity or their insufficient capacity to handle "loads." In skeletal form: Huntington argues that revolutions and lesser forms of collective political violence are artifacts of rapid socioeconomic modernization. Such modernization "mobilizes" people and induces them to enter the arena of political conflict. No harm will be done if political channels can handle their demands and activities in pressing them. But if political development lags, blockage occurs, and aggressive modes of action are generated. Note the incidence of extreme political violence in conditions of socioeconomic development, especially in centralized monarchies, narrow-based military dictatorships, and in new nations.^[27]

The notions of overload and adaptation to stress belong to the world of systems theories. Since such theories are essentially concerned with negentropy as a normal state, any theory of collective political violence derived from the systems perspective belongs to the universe of contingency (though entropy is inevitable in the very, very long run). C. Johnson has been perhaps the leading systems theorist of revolution,^[28] at any rate if we do not look far beneath his language. The causal chain in Johnson is quite similar to Huntington's, ignoring nominal differences: rapid change leads (sometimes) to system disequilibrium (the overload of mechanisms of homeostasis), which produces individual pathologies as well as collective movements. The sense of deprivation (Johnson actually avoids psychological concepts and speaks of dysfunction) arises, of course, at the point of overload, or blockage. Similarly, Wolf's account of peasant rebellions,^[29] though making more of tactical considerations than Huntington

or Johnson, rests on aberration: peasants—not capitalistic "cultivators"—resist the encroachment of market economies; but when traditional peasant life cannot be maintained and alternative arrangements are too ill-developed or restrictive, tensions arise and peasants rebel.^[30] Here, aggression is unleashed by a combination of cultural and economic frustrations.^[31]

We can perhaps best divide the members of the family of contingency theories according to whether their normal model is essentially macrocosmic or microcosmic. In the first case, notions of systems and of their aberrations under conditions of extrinsically imposed strain are used to identify pathologies. In the latter case, apart from cultural learning, the microcondition is more manifestly and explicitly a sense of deprivation, relative to others or to more abstract conceptions of what is justly due. Thus Gurr's theory is microcosmic (individual), and Huntington's and Johnson's, macrocosmic (societal) in emphasis; but they converge at the explanandum, collective political violence.

Inherency:
the Collective Action (CA) Family of Theories

Inherency theories of collective political violence at present are less common than contingency theories. They seem more numerous than they actually are because of a proliferation of labels for the same thing: resource-mobilization theory, political process theory, theories of group dynamics, mobilization theory, strategic interaction models, and political contention theory. "Collective action theory" is used here because the postulate of the approach is that violent collective action is not aberrant but simply one of many alternative channels of group activity; like any other it is chosen by tactical calculation. Thus it belongs on a continuum or is part of a repertoire: different, sometimes extreme, but not off the normal scale.

"From mobilization to revolution." CA theory is chiefly the work of Charles Tilly and his associates; its summa is Tilly's recently published From Mobilization to Revolution .^[32] An earlier and more succinct overview appeared in 1975. Again, what I present here is a skeleton of a theory that, like Gurr's, has grown in complexity to accommodate data and objections, to the detriment of elegance.

Tilly begins with a simple conception of the polity. Polities have members, who have formal access to the political decision-making process, and challengers, who do not.^[33] All are contenders for power—with members, of course, enjoying privileges. Members use their resources in a game of continual jockeying to enhance their power; challengers try, as a condition to all else, to get into the game. To be allowed to play, there are entrance fees. The higher the fees, the greater the pressure needed to become

members; and at some point of cost-efficiency, violent action among contenders occurs, with revolution as the most extreme, but still normal, form of such action.

This is the barest précis. Some key points need to be added. Before any collective action (say, a strike, election, demonstration, charivari, riot) can occur, there must be a confluence of shared interests^[34] —though Tilly deliberately skirts the issue of how collective interests come to be perceived and pursued, perhaps wisely, given his purpose. The interests must possess organization: a combination of shared categoric traits and a pattern of frequent interaction, or network.^[35] Beyond this, organized interests must be mobilized; by this Tilly means the possession and use of resources that may help achieve goals.^[36] Even at this point, collective action will not occur unless there is sufficient opportunity for it.^[37] This is essentially a matter of power to repress (especially of credible threats) or, more generally, to make collective action costly. The obverse of repression is, of course, facilitation, not in the sense that Gurr usually employs the term, but with emphasis on political toleration of, or help to, the activities of groups in conflict.

There remains the question of what (if not contingent matters like anger or strain) activates violent collective action, particularly in the more extreme form of revolution rather than lower-cost actions. The answer is a process:^[38] (1) contenders (insiders or outsiders), organized around some specially motivated core group, make claims incompatible with a polity's survival in its existing form; (2) the claims gain increasing acceptance, usually under conditions of alienation resulting from governmental malfunctioning: failures to meet obligations (provide benefits) or unexpected demands for resources (usually taxes); note here the intrusion of a glaring contingency—but only (as in Galilean motion) as an accelerator or as something that channels activity into a special path; (3) threatened authorities either cannot, or will not, or will not efficiently, block the potential for extreme action by suppression; hence (4) a condition of multiple sovereignty comes to exist. That condition never occurs after a short-run breakdown; it is always the result of a long-run chain of events. Multiple sovereignty involves mutually exclusive claims to legitimate governmental control, accepted by many (on both sides); often it is manifest in the establishment of parallel governments; and a struggle for partners in coalition occurs. Upon the reintegration of sovereignty, the process ends.

Apart from challengers' egregious "claims to resources," mysterious in origin, it seems plain that the crucial force that channels collective actions to violent political actions, once the (obvious) conditions of any such actions exist, is governmental inefficiency, timidity, and weakness. Revolutions thus occur when obstacles to strong pressures are unblocked; they are not, as typical contingency theorists believe, the very result of blockage.

Hence the assertion in our explanation-sketches: for inherency theory, the pivotal problem is what prevents extreme conflict from taking place; in contingency theory, the issue is what causes it at all.

Relatives of CA theory . Tilly himself locates the ancestry of his theory in John Stuart Mill—with a nod also to Marx and Marxists, for whom inherency takes more the form of ineluctable historical process. Mill and the Utilitarians are aboriginal CA theorists in that they regard all action as based on the rational pursuit of self-interest (pleasure), in contrast to Durkheim's notion of aberrant phases in the unfolding of the division of labor, or Weber's notion of chiliastic traumas in the unfolding of a disenchanted world.

Among contemporary writers, we find versions of CA theory in numerous strategic interaction models of behavior.^[39] Such models treat forceful courses of action, such as strikes, not as releases for potent emotions but as moves in games—they involve bargains, coalitions, lying low or pouncing, to maximize one's take. A. O. Hirschman's elegant Exit, Voice, and Loyalty resembles Tilly's in that Hirschman accounts for "secession" from a social entity, protests (opposition) of various kinds within such entities, and acquiescence as a "repertoire" of responses to discontent;^[40] choice among them is considered a matter essentially of cost calculation. In an unfortunately discursive but fascinating work on violence by American blacks during the sixties, H. L. Nieburg argues at least implicitly in a similar vein.^[41] People prefer low-risk methods of resolving conflicts; the discovery of a method that offers a decent chance of success at low risk is a matter of trial and error (strategic interactions, in less plain words). In that process, violence may be used, usually under conditions of rapid social change, new group formations, and high levels of social uncertainty. In-groups and out-groups maneuver toward some new political balance, until a proper new low-risk mode of resolving conflicts is found; if not found, a life-and-death struggle for domination occurs. Violence here is, as in all CA theories, a "move" likely to be made if expected costs do not exceed expected benefits—more accurately, violence is used if it is the best available course of action. For manifest reasons, it often is the only viable choice for systematically disadvantaged groups. Nieburg's work represents a special form of rational choice: the choice of actions occurs not so much by hard calculation as by experience—trial and error among a set of (abstractly) equivalent actions.

Evaluation

We now come to the crux. How can one make a reasoned choice at the branch point?

Unfortunately, there is no simple, workable way. One might simply reflect on the actual incidence of collective political violence. But that leads nowhere. One reason is logical. One might, superficially, expect something "inherent" to occur more often than something "contingent," but after a bit more thought it is clear that this is not necessarily so: contingencies can frequently occur (like physical rest), and inherent tendencies may not generally be unblocked. (That is why people so often leap, like Aristotle, into obvious, but mistaken, positions. Concrete nature is masterful at deceit.) In actuality, violent actions occur very often; but they do not occur as often as alternatives. Sorokin found, over two millennia, about one year of violent disturbance out of every four.^[42] From this it may follow that it does not take much to make violence occur contingently or that it does not take much to block the tendency toward it or make other actions more attractive.

It also goes without saying that studies by the theorists themselves fit whatever tack they choose—though sometimes in an eyebrow-raising way. Gurr, for instance, consistently does get good statistical results. But so does Tilly, when he confronts his models with data (which he does less well than Gurr). This is hardly surprising, since their models must come together at some point of explanation of concrete events, which, as stated, do nearly always have both contingent and inherent causes. At the same time, difficulties of research findings, even if manifest, are often glossed over or interpreted in a dubious way. To my knowledge, Tilly and collaborators have never succeeded in solving a crucial problem early recognized: finding "reliable procedures" for enumerating contenders, measuring mobilization, and specifying the relationship of groups to existing structures of power. Operationally, the theory is in limbo at all crucial points. Gurr, however, has been much criticized for his choice of "indicators" of RD. Deprivation is a "state of mind" that Cantril studies psychologically,^[43] but that is inferred in Gurr from objective (economic and political) indices. That begs many questions. One also wonders about: (1) the fact that Gurr and Duvall account for 75 percent of the variance in "civil conflict" across eighty-six countries in the early 1960s with five "causes" and eleven variables, in a simultaneous equation models^[44] —which could be worse, but hardly is conclusive; (2) the fact that mediating (secondary) variables in work by Gurr always account for a good deal of the variance in magnitude of civil strife, with social-structural facilitation always a significant variable—a result that CA theorists surely can turn to their own account.

At the very outset, then, we confront ambiguity. We should try to reduce it by inspecting available data bearing logically on one theory or the other. I will do so by discussing a number of selected issues, potentially helpful in choosing at the branch point.

Alternative Channels

If CA theory is the fruitful tack, a clear relationship should show up between the incidence of collective political violence and the availability of alternative channels for making and realizing "claims." We may thus posit that democracies ("open" polities) will rank low on the dependent variable. By extension, political violence should at least decline discernibly in cases of regular electoral competition. But since the less advantaged do not have equal access even in open polities one should find them playing a specially important role in political violence—following the cliché that violence is the resort of the weak: everyone's equal capacity, in Hobbes's state of nature.

The matter of alternative channels would seem immediately vital for anyone who considers all collective actions a set, or repertoire, of equivalent events. However, astonishingly little has been done with the subject by Tilly: some secondary analyses of lower-class actions such as strikes and food riots, and a study of the connection between elections, organized associations, and the occurrence of "demonstrations."^[45] We do have more direct evidence—though the subject cries for far more investigation by all inquirers into our subject.

In general, the data run counter to CA theory. Hibbs found virtually no statistical relationship between democratic polities and magnitude of political violence, either of the milder protest variety or with more virulent internal wars.^[46] To avoid the possibility that findings were distorted by including "ill-developed" democracies, a relationship was sought between levels of political violence and democratic development, in the manner of Neubauer.^[47] Again, no significant relationship emerged with protest, and a weak negative association with internal war was convincingly explained away as spurious. Worse, a positive association turns up between elections and political violence,^[48] suggesting, perhaps, that electoral processes activate emotions appropriate also to other outlets. Hibbs also finds that effective exclusion from valued political positions due to ethnic, religious, or linguistic traits usually leads only to mild forms of protest—a finding that also turns up in Gurr.^[49]

In the most recent, and most persuasive, study of the subject by Gurr,^[50] the results are more complex, but still of scant comfort to CA theorists. One critical finding is that democracies typically had more extensive "civil conflict" (a broad notion that ranges from demonstrations to guerrilla wars) than autocracies. (For simplicity's sake I ignore a third type that Guncalls "elitist.") Democratic civil conflicts, however, were much less deadly. The first finding clearly impugns CA theory. However, the second provides CA theorists with a measure of comfort, since it must be due to a toleration in democracies of protests that, in repressive regimes, never surface or else are forced to take virulent forms. RD theorists can rejoin that griev-

ances in democracies are generally less serious (most obviously because of lesser political deprivation per se). They can also sensibly hold that the greater deadliness of civil conflicts in autocracies may be a result of the actions of regimes, not dissidents; this depends, obviously, on who is killed and under what circumstances.

Two other relevant points: the old saw that violence is the political means of the impoverished—the basis of the McCone Report about the Los Angeles riots in the mid-1960s—simply does not stand up to close examination.^[51] There is also the much documented fact that revolutionary leaders do not much differ socioeconomically from other salient political figures, and that they differ more in regard to social marginality than in regard to resources at their disposal.

It seems evident—logically and empirically—that CA theorists must deflate many quantitative findings about the effects of alternative channels by stressing facilitation. Open polities do not block propensities to act of many, or any, kinds as much as closed polities. So they produce more collective political actions of all sorts. And the more advantaged and powerful have more means for deadly violence, no less than other actions: particularly military elites, who are more likely to act politically in closed polities. Our first test thus is one-sided in statistical results, but not hard to argue away.

Facilitation

It seems necessary then to look hard at the CA theorist's chief route of escape: facilitation—that is, how difficult or how easy pertinent circumstances make it to use collective violence in politics. Here, contingency theorists sometimes seem to turn the tables on themselves. Consider Gurr. In his early (now much modified) report,^[52] certain highly contextual factors presumed related to the possibility of violence (e.g., transportation networks, density of population, other geographic characteristics, and "external support of insurgents") account for more variance in magnitude of civil strife than anything else (twice as much as persistent deprivation!). Similarly, such facilitating factors as the distribution of value-stocks, complexity and cohesion of organization, number and scope of values (resources) are likely to decrease strife if regimes are better equipped and to increase it to the extent that dissidents possess them. In Gurr's later work, facilitative matters still seem to have crucial effects on whether dissidence is peaceful or violent.

A picture begins to emerge. Having less costly channels available does seem to affect the choice of violent means: so, score a point for inherency. But the cost of violence does not much reduce deadly conflict in democracies, and still less in autocracies, where presumably it is likely to cost more: so, score a point for the other side. All would seem to depend then

on what to regard as fundamental. I know no way to decide that issue yet, since CA theory incorporates facilitation (viz., opportunity), as does RD theory. The most we can say is that violence is more likely if easier to engage in and perceived to be more likely to succeed. Tactics play a role—but perhaps only for people afflicted by high RD. It seems necessary, then, to look at other issues that might break what, so far, appears to be a tie.

The Balance of Coercion

The most manifest "facility" for collective political violence is the ability and willingness of regimes to repress, relative to that of dissidents to destroy. We should, of course, expect the balance of coercion to make a difference both in a contingency-sketch of political violence and in its opposite. For contingency theory, though, the tactical consideration involved is of lesser import: as stated earlier, very angry men are not likely to act coolly, even in the face of what Tilly calls "threat." For CA theories, per contra, little would seem more important. What does the evidence suggest?

To begin with, we are handicapped by a flaw in method: the almost universal tendency to use the coercive capacities of regimes as measures of the coercive balance between authorities and dissidents. One exception is an article by Gurr.^[53] Gurr uses measures of loyalty and dissidence by military forces and weighs familiar measures of the coercive potential of regimes against foreign support of dissidents and aspects of dissident groups, such as their size and organization, that may be assumed to have a bearing on coercive capacity. The results Gurr obtains actually seem to offer some, but not much, support to CA theorists. The coercive capacity of dissidents does "enhance the prospects for rebellion," the most extreme category of dissidence, whereas that of authorities reduces it. This is what CA theorists would expect. However, (1) regime coercion has "very little effect" on other forms of protest; (2) a different balance, concerning "institutional support" (support by "dense" and pervasive networks of organizations) consistently explains more; (3) throughout, the combined factor of institutional/coercive balance fares only a little better or worse in regard to different kinds of collective violence than a quite different explanatory variable, "justification," which includes both tactically relevant factors (e.g., success of past strife) and nontactical ones (legitimacy), but the latter yields better results than the former; (4) dissident coerciveness enhances the likelihood of rebellion much more than regime coercion inhibits it; (5) the latter has virtually no effect on lesser forms of strife. What follows? Perhaps nothing, for operational reasons: Leaving aside questions of data sources and scaling, Gurr's evidence comes from twentyone Western countries over a mere five-year span (1961–1965). However, the clear and strong result that CA theory would seem to call for manifestly

fails to turn up. Also, CA theorists must make much of the capacity to inflict high costs (as pointed out in the explanation-sketch for inherency theory above); one therefore should not find that regime coercion is, as it seems to be, the least weighty explanatory factor in all types of strife. After an initial leaning toward CA theory, one thus is led to a contrary conclusion—though that, again, is offset by Gurr's conclusion that "variations in deprivation are not an important direct determinant of total 3strife."^[54] Deprivation exists in the remote background, waiting to be "converted."

CA theory would be best supported by finding a strongly curvilinear relation between conflict and coercion. If one or another side greatly outweighs the other's capacity for coercion, rebellions should not start or regimes ought quickly to collapse, at low cost. The most intense conflicts ought to occur where coercive capacities are closely matched. Granted difficulties in the available data, the most recent study by Gurr shows only slight curvilinearity. In addition, some other results that are very odd from the standpoint of CA theory turn up: in general, a positive relationship seems to exist between governmental coercion and conflict. This includes the finding that the cumulative application of sanctions increases conflict, even at the extreme of sanctions, and that the use of sanctions has no discernible time-lagged effect on conflict. These findings are contrary to Gurr's earlier position or that of the Feierabends.^[55] However, they are strongly supported by Hibbs.^[56] And they are more consistent with contingency theory than the earlier position.

The unsatisfactory state of the available evidence does provide an escape hatch to CA theorists, at least for now. The tendency, though, has been for coercive balance to be of less importance as studies have improved. Most damaging, perhaps, has been the tendency of CA theorists themselves to argue away inconvenient findings by making fuzzy what should be especially clear in their theories. The problem is illustrated by Tilly's magnum opus,^[57] in which repression is treated with unusual convolution—and by aphorism: for instance, "governments which repress also facilitate." Ultimately, Tilly resorts to a promising line of argument, one that involves historical patterns of repression. The special variable involves abrupt changes in such patterns. Unfortunately, the hypothetical relationship is either to "encourage" or "discourage" types of collective action (again, excepting only the obvious exception: very high levels of successful repression). That is simply not permissible—not, anyway, without a lot of added theory.

On the whole, contingency theory emerges healthier than inherency theory from our third test. But we cannot escape the problem of adequate data. For this reason, CA theorists need not throw in their towel yet. We need far more and better light on the issue. We want it especially from

CA theorists themselves; the matter is critical for theories in which "causation" and "channeling" (or blockage) are the same thing. This is useful to know, but leaves our issue still undecided.

When Men Rebel

I referred above to Tilly's use of the historical pattern of repression as an explanatory factor. His doing so may be subsumed under a more general, widely followed line of assessment: to find out why men rebel we may be helped by studying when they do so. As we shall see, this involves a set of diverse tests that could also be invoked separately. Each, though, fits a general deduction from our two explanation-sketches:

1. For contingency theory: Collective political violence, or such violence in its more extreme forms, should occur when, as the result of some temporal pattern, the specified contingency, such as RD, is or may be expected to be particularly great.

2. For inherency theory: Collective political violence, or its more extreme forms, should occur when, as a result of a temporal pattern, (a) the costs of violent collective action are expected to be especially low, or (b) nonviolent actions in pursuit of highly valued goals have been shown to be unproductive.

There is much historical precedent for seeking explanations of political violence in the nature of historical moments of such violence. The classic source is Tocqueville. His basic argument in The Ancient Regime is familiar:

Revolution does not always come when things are going from bad to worse. It occurs most often when a nation that has accepted, and indeed has given no sign of even having noticed the most crushing laws, rejects them at the very moment when their load is being lightened. . . . Usually the most dangerous time for a bad government is when it attempts to reform itself.

This argument is the theme for numerous variations. We will consider the most important.

Rapid change . There is a large family of theories that attributes extreme and destructive political behavior to an obviously "contingent" condition: unusually rapid, hence unusually unsettling, socioeconomic conditions. Frustrations are likely to arise under such conditions because of disorientation (anomie) and because of the familiar occurrence of an excessive rise in expectations. At a minimum, the conditions of any contingency theory are more likely to be satisfied when change is rapid (and abrupt) than under more settled circumstances.

Olson's influential article on the consequences of rapid economic growth is prototypical of this view.^[58] A political, and otherwise mod-

ified version of Olson's argument is Huntington's mobilization-institutionalization hypothesis. Both arguments are backed by persuasive reasoning, as well as the usual selectively chosen, post hoc illustrations. However, the evidence once again is surprisingly inconclusive; in general, only illustrations are used, and illustrations can be found of almost anything that is not wholly absurd.

Tilly and Johnson both have pointed out that contingency theorists should expect a high correlation, in cases of rapid change, between individual pathology (crime) and the collective variety (political violence). One can see why this should be so, the causes of individual and collective aberration (such as frustration) being presumably the same. But again, though the point seems crucial, evidence is strangely meager, and, for us, confusing. Gurr's work of 1970 cites a study that reports a decline in aggressive crimes (by blacks and against blacks) during periods of civil rights demonstrations.^[59] That fact, though, can be interpreted almost any way one chooses—though, superficially, it runs against the expectations of contingency theory. By way of compensation, Tilly et al. do find a correlation of the two,^[60] but they emphasize that it is low. Perhaps, though, this is the correct expectation—anyway for them—since CA theorists regard different responses to similar conditions as alternatives in a repertoire of actions. If so, individual and group violence should be associated, but not very closely, for some people will choose the one response and others its alternative. Obviously, we need here both better data and better reasoning.

In any case, an impressive number of studies suggests that there is no simple, direct relationship between rates of socioeconomic change and political violence. The relationship, again, seems unusually ambiguous. Tanter and Midlarsky found a negative relationship between economic growth and, as they use the term, "revolutions" in Latin America, but a positive relationship in the Middle East and Asia.^[61] Bwy confirms the Latin American result but argues that a different relationship among the variables holds for less developed countries.^[62] Flanagan and Fogelman find a negative relationship;^[63] Alker and Russett find the same.^[64] On the other hand, the Feierabends and Nesvold report a high association between rate of modernization and "instability."^[65] Hibbs finds different relations between rapid change and different types of political violence.^[66] And so it goes, in a very extensive set of works.

With enough ingenuity, one might perhaps trace this extreme confusion to different uses of concepts and measures. But if the relationship between the variables was very strong, mere differences in preferred indicators ought not to produce such wildly divergent findings.

Adaptation to change . The virulent potential effects of most stresses may, of course, be offset by the proper adaptation of "systems." If socio-

economic change is similar, differences in "adaptation" to it should be matched by differences in its consequences, including collective political violence. An example is the apparent relation (to be sure, small) between historical bourgeois radicalism and the relative lack of opportunities to become ennobled—^[67] a nice example of Pareto's hypothesis.

More to the point here is Barrington Moore's thesis that political outcomes (in Moore's case, the nature of regimes, but, by implication, also political processes more generally) depend chiefly on the adaptation of the traditional landed upper classes to "bourgeoisification"—economic modernization, as most of us think of it.^[68] That famous thesis fits, post hoc, eight widely assorted cases. But it has run into trouble when extended to other cases (or when examined more closely): by Tilton (Sweden) and Rokkan (the smaller countries in general); note also the critique by Skocpol.^[69] Moore's thesis can also be logically subsumed under either contingency theory, if change is emphasized, or inherency theory, if his argument is interpreted as hinging on choices of coalitions. Note, too, that despite the long vogue of "systems" theories in political science, operationally rigorous work on stress due to change, and adaptations to such stress, is virtually nonexistent.

Structural imbalance . As implied earlier, Huntington's thesis rests on the notion of a "balance" of structures rather than on rapid mobilizing as such. This, in a sense, combines the variables of change and adaptation to change. Huntington's theory, of course, places the most inflammatory point in polities where the divergence between mobilization and institutionalization is greatest. As the space between the two narrows, collective political violence should decline. Like Moore, Huntington illustrates his argument. The evidence of other studies, however, runs strongly against him.

Schneider and Schneider accept Huntington's basic argument on the basis of a cross-national study,^[70] but their evidence leads them to reject the corollary that "mobilization" should be slow if it is to be balanced by "institutional" adaptations; this is damaging evidence, for one can easily show that the corollary is inherent in the postulate. Sanders concocts a resounding empirical refutation and also presents a strong critique of Huntington's conceptualization (which, of course, weakens the empirical refutation).^[71] Other important empirical rebuttals may be found in Yough and Sigelman, and Duvall and Welfling.^[72] Huntington himself distinguishes among Western and Eastern types of polities, to which his thesis (presumably) applies differently, and adds some very ill-fitting variables that confuse the nontactical character of his theory: for instance, the effects of foreign wars and interventions.^[73]

Unfortunately, again, there seem to be no other theories of structural

"imbalance" that have sufficient empirical support to allow a more definite verdict about this mode of theorizing; nor are other possibilities worked out as fully as is Huntington's theory (really, itself an explanation-sketch).

The J-curve . One major hope remains: that we can infer the conditions of collective political violence from some less simply wrought theory of change—some theory that finds the point of explosion at a particular point, or range, on a curve of change. The leading exposition of such a view is Davies's J-curve theory.^[74]

The theory is that revolution is likely when periods of prolonged improvement, the historical pattern most likely to raise expectations, are interrupted by abrupt reversals; then frustrations due to unrequited expectations become intolerable. The J-curve theory, needless to say, involves contingency in its most pristine form, especially considering that it has never been diluted with logically confusing factors suggesting tactical choice—all honor to Davies for theoretical courage. What is the state of the evidence regarding the theory?

The supporting evidence should be very strong, for all contingency theory implies that the condition described by Davies frustrates and angers very deeply. In Davies's original formulation four cases were invoked—once again, as illustrations. Later, four other cases were added.^[75] In an impressive independent check, Grofman and Muller provide a clear measure of support by using data on individuals.^[76] But they also find strong evidence for a "relative gratification" theory of conflict behavior (droprise, or V-curve, theory). Still, their study manifestly remains within the realm of contingency theory and rests on reasoning analogous to, though not wholly the same as, that of Gurr and Davies.

The chief problem with J-curve theory is the abundance of countercases. Consider, for example, the many countries in which the Great Depression of the 1930s did not increase political violence. Surely, the effects of sudden depression, following the orgiastic recovery of the 1920s, were crucial—and no more in Germany than in all the countercases. Moreover, Tilly provides, of all things, a tactical interpretation of the meaning of the J-curve patterns of change.^[77] Unfortunately, it seems labored: members should "break commitments where least dangerous." This begs crucial questions: What commitments, and to whom? Moreover, the interpretation accepts the importance of contingent frustration and simply counsels increased coercion in the right place.

Group dynamics . The question of when men rebel obviously has led, again, to puzzlement, if viewed from the standpoint of contingency theory. Does inherency theory, then, fare better with the question of timing? For RD theorists and their kin, the flashpoint of collective political violence

should occur at a point of social process when rage or despair are most acute—when expectations and capabilities are most distant. For CA theorists, in contrast, violent action, being relatively high in cost in most circumstances, should generally be chosen after lower-cost channels are perceived as ineffectual, provided only that the balance of coercion does not manifestly rule out successful violence.

A tactical scenario that makes the resort to violence especially likely can readily be constructed. A group of contenders makes political claims cheaply: say, by petition. No response is made by authorities. Pressure is stepped up, perhaps through a stoppage of work; but still no response. A more dangerous organized demonstration is used next. The authorities remain intransigent and call out some squads of police to indicate determination. A more intense demonstration occurs; perhaps now some undisciplined elements or provocateurs throw rocks or do some looting. The authorities call on police and militia; heads are broken. At this point it will be clear that only violent collective action has any prospect of success at all. And, given some chance of success and intensity of claims, the highcost method will be used. The moral is that negative sanctions precede collective political violence. On this point, CA theorists have differed from RD theorists almost from the beginning, for manifest reasons.

Tilly's argument along the lines just sketched is predated by the report of the National Advisory Commission on Civil Disorders. The report held that intervention by the police "almost invariably" preceded larger-scale violence—though, of course, the police might simply be acting in learned anticipation of what might occur anyway. A more convincing finding occurs in Hibbs:^[78] milder forms of aggression do generally escalate if regimes use coercion; in fact, that is the strongest of all factors leading to "internal war" in Hibbs's eclectic model. Hibbs, though, can hardly be conclusive on the point at issue here. For one thing, it is not peaceful action that is converted by force, but aggressive action into more aggressive action: the optional channels start with violent behavior. In addition, Hibbs's data are cross-sectional; they do not follow the same conflicts over time. And, again, there is empirical support also for the other side. Tilly and partners themselves have presented contrary evidence:^[79] repression, they argue, works—though, needless to say, they use the point to support the tactical-choice argument. This is a fine example of how one may read the same evidence in contrary ways if one starts from antithetical bases—hence the bases of theory matter a great deal. Repression is used and violent action stops; a routine channel is blocked, or aberrant anger is suppressed. The evidence fits both conclusions equally—and we remain ambiguously suspended between antitheses.

War . A different argument for inherency involves the removal or weak-

ening of coercive blockages, rather than gradual escalation as a forceful response to repression. A tendency may be present and emerge when resistance to it weakens, or actions may occur to overcome resistance when otherwise behavior would be more moderate.

When is coercion, as an obstacle to turbulent political action, likely to be unusually low? The most obvious answer is that the coercive potential of regimes will be exceptionally low when military forces have disintegrated, leaving the field free to less potent groups. This condition is likely after defeat in war, at least if occupation forces do not step in to help incumbent authorities.

Examples of revolutions after lost wars abound: France in 1871; Russia in 1905 and 1917; Turkey in 1918; China after World War II. Unfortunately, countercases can be invoked just as readily: Japan or Italy after the last great war, for instance. Perhaps these cases only show that losing a war is not a sufficient condition for revolution; but then, neither is it a necessary condition, or even a "normal" occurrence (e.g., France, 1789; Mexico, 1910; Cuba, 1959; etc.). Anyway, the argument surely is more pertinent to the outcome of revolutionary conflicts than to their inception. This is Hammond's point about communist take-overs and, more broadly, that of D. E. H. Russell's study of twenty-eight "mass rebellions" since 1906.^[80] Consider also Trotsky's familiar argument that the armed forces usually reflect popular conditions. If so, they would always facilitate violent action, if there is sufficient popular disposition to use it.

Learning:
The "Culture of Violence"

Contingency theories belong, in social analysis, to the same family as culturalist theories—theories based on learned orientations to action. Inherency theories are related to rationalist theories—based on the notion that actions are chosen by calculations of cost-efficiency.

There is much writing about the role of learning and culture in relation to political violence. Bandura is the leading exponent of the view that using violence individually is learned, and he makes a good case.^[81] On the macrolevel, though, we should expect ambiguous findings about the relation of present to past violence. The use of violent action might, logically, become a learned response, but it might also teach people that its costs tend to be disproportionately large, even if successful. The problem here (as with violence by authorities: repression) is that any curve, or none, can support the cultural interpretation. Much, in other words, is made in the literature of "cultures" of violence, but nothing definitive for the debate between contingency and inherency theories is likely to emerge from studies linking past to present violence.

This point fits findings. At best, a moderate relationship tends to turn up, as in Gurr's work.^[82] Still, Gurr later strongly argues the "culture-of-

violence hypothesis on "varied evidence and arguments"—which, as I read them, support equally well utilitarian explanations of violence.^[83] Another recurrent theme in studies of the culture of violence, as we should expect, is that "it depends." For instance, statistical relations seem to vary, for some reason, with geographic areas. In Africa, past rebellions are associated with reduced "turmoil";^[84] but Latin America is different, and Western Europe different again. Welfling (with Duvall) squares the circle on the basis of type of collective violence: turmoil, it is argued, feeds on itself whereas "elitist" kinds of violence reduce the likelihood of later violence. But they do not do so through learning; rather, the cause seems to be the tactical factor of suppression.^[85] Hibbs also reports quite contrary findings for "protest" and "internal war";^[86] in his study, conflict akin to the "elitist" type in Welfling seems to be increased by the existence of a "culture" of violence.

As I argued, this contrariness can be accommodated to the premises of contingency theory. But to accommodate the evidence, we will have to specify when violence teaches violence or the imperative of peace. This will be difficult, at best. At present, the evidence may be regarded as typical of the too often ambiguous findings in the literature.

Conclusion

We began with an explication of a recurrent antithesis at the basic branch points of theorizing about any subject. We went on to a version of that branch point especially tailored to explaining collective political violence. We found that theories in the field do divide along the lines indicated. And we reviewed many empirical studies that bear on expectations deduced from the postulates of the antithetical approaches. No clear result emerged. Granted, conditions such as inequalities, unsatisfied demands, discrimination, and societal cleavages are related consistently to degrees of conflict. What remains mysterious, in the end, is the basic nature of the link between causes and effects—not least, therefore, understanding why similar conditions so often have dissimilar results.

The findings do not point strongly toward the superior fertility of a particular species of theorizing. "Fertility" in this connection is, first, the capacity of an approach to provide consistently superior explanations of many aspects of an independent variable—explanations that are themselves logically consistent or entailed by a higher-order theory. Second, fertility is the ability of such higher-order theory deductively to yield new, and good, lower-order theoretical relations. On the whole, contingency theory probably has fared better than inherency theory. That may only be a byproduct of its more frequent use or it may result from a lesser empirical bent among inherency theorists. More important, we consistently found

ambiguities in data bearing on the contingency perspective, no less than its antithesis—problems of interpreting correctly their implications—and often findings that were offset or made doubtful by other findings. Also, relationships that should have been clear and strong frequently were opaque and weak.

Something, then, must be amiss—at least if the premise of this essay concerning basic theoretical choice in studying collective violence is correct. I will conclude with some speculations on what the problem might be.

Surely, the difficulty is not any lack of studies or data. We are inundated with both, and the data cover a vast range of variables, history, geography, and types of polities.

Somewhat more likely (though I doubt it) is that we arrived at a confusing result because the empirical studies cited were an inadequate sample. They certainly do not exhaust the literature and are a mix of methodologically good, bad, and indifferent studies. This implies a challenge to specialists in the field: before embarking on new empirical studies, identify the high-quality works in both approaches and analyze their procedures and results from the perspectives sketched here.

Another strong possibility is that confusion has emerged because of a taxonomic error. We have treated cases of collective political violence as a single class—as Tilly treats a still much larger universe: all collective actions. Even if this is typologically plausible, doing so may be theoretically confusing: as suggested early in this essay, certain kinds of political violence might better be treated as "contingent," others as "inherent." We did frequently find such results in the literature, especially between lesser and greater types of strife. For reasons of parsimony, we ought perhaps to resist using this possibility, but evidence might compel resorting to it. In any case, here is another challenge for analysis.

A third likely difficulty—one common in the social sciences, and one that may well condemn to futility exercises like those just suggested—is method: an overdose of induction. One doubts that underlying determinants can ever be mechanically teased out of the complexity of highly disparate events, no matter how sophisticated the quantitative methods used. A related problem is deliberate eclecticism. Choosing eclecticism at the outset guarantees confusion in the end and thus seems self-defeating—even perverse.

Although we have not found a hero, we have surely found villains: prehedged models and models ground out mechanically from motley data. The resulting "modeled" world appears about as complex (and thus mystifying) as the concrete world it models. In Hibbs,^[87] we find no illuminating simplification: we find only a complex world of variables, vertiginous with arrows and proportions. The artfulness of nature wins out over that of

theory because, contra Bacon, nature is not "put to the test." One should not be too censorious about this, since Hibbs's deliberate aim is eclecticism. Gurr's model—more surprisingly, because based on an elegantly parsimonious theory—is not much less complex. Is not the reason fear of potential disconfirmation, even at the cost of mixing up a highly plausible line of analysis with its antithesis?

The result always is a realized intention: a model that "fits" the data. But if such a model is irrefutable in principle, it will fit only without really illuminating experience. That, I suggest, is inadvertently the case in Gurr's work, however tenacious and ingenious the work has been. Try to refute the following, which is essential RD theory, à la Gurr:

1. no political violence can occur without politicized discontent (satisfied people do not rebel);

2. no discontent will exist unless somebody feels deprived;

3. politicization involves both normative and utilitarian (tactical) considerations;

4. even so, little or nothing will happen when facultative and coercive resources available to dissidents and authorities are distributed one-sidedly.

Admittedly, this is a great simplification. But it surely shows, by omitting nonessentials, how RD theory has been insulated against tactical accounts of collective political violence by incorporating them.

Tilly, too, has increasingly complicated his theoretical model, as well as swallowed up inconvenient data by reinterpretations. A still greater difficulty that emerges in his work, and also disarms invalidation, is a kind of clever triviality (in the philosophic sense). Something seems to be said to explain a mystifying set of events. But except for labels (members, challengers, etc.) we know it already and so remain mystified. No substantial violent action will occur unless:

1. some group wants something it does not have;

2. a fair number of people agree that their claim is justified;

3. the group is not successfully suppressed to begin with;

4. the group controls some suitable resources (and wants to control more).

And who does not know that extreme rebels lay claim to nothing less than sovereign authority, against the claims of incumbent authorities? I do not mean to be sardonic. The point of the argument is that the "interesting" issues (those that really need explaining) always are a step removed from those Tilly faces: why do "outgroups" actually (not latently) come to want "in," where before they acquiesced? Under what conditions do people

come to perceive the illegitimacy of a pattern of governmental authority, rather than continue to acquiesce?

A conclusion seems to emerge. The literature, even if differently surveyed, will probably be inconclusive for us because a well-defined choice among theoretical approaches has not been faced at all by the many scholars of collective political violence. Perhaps this is owing to failure to recognize that such a choice exists. If so, this essay, though it has no "result," should help. More immediate reasons are the understandable desire to be "right," which, in a messy world, is easier to achieve with messy theory than with parsimonious theory. If not tautological, our explanations of collective political violence thus far have been too close to descriptions: too close, that is, to being depictions of the concrete, in the jargon of either contingency or inherency theory. On that basis, explanation hardly can fail; but it also cannot succeed in getting to, or even near, essentials. The remedy is to regard theory as what it is: a tool of explanation, not something that models all facets of the concrete. We want deliberate one-sidedness that may fall in clean, competitive tests, not deliberate, or face-saving, eclecticism. If the choice of a theoretical approach is a choice among conceptions of the essential nature of a set of phenomena, studies that circumnavigate that issue can only defeat us—unless I totally misunderstand the powers of multivariate methods.

We can choose between the two most basic models of the political individual: peace-seeking versus power-seeking. But in my view we cannot do it effectively by incorporating both in our initial models. If we do that, results pointing to both assuredly will turn up; both are at work—in physical motion, language, or the malfunctioning of organisms. Nor can we do it by largely arbitrary commitments. Political theorists have always dealt with basic branch points in one way or the other. Thus they have come to no resolution at the foundations of theory; rather, they simply have divided "members" and "challengers" into disciplinary factions, always for suspect reasons—the worst of which is not facing up thoughtfully, with open minds, to the problem of primary theoretical choice.

One way to do so would be to work along a given line to see how far it leads. All considered, I would at present choose contingency theory in the RD version—but in a much more simplified form than Gurr and his associates have used, recognizing that the goal is to construct good "theory," which is an abstract tool of understanding, not to reconstruct concrete reality in all its nuances and complexities. Once the essentials are known, the nagging complexities will (on past scientific evidence) fall into place more persuasively.