In attacking the problem of revolution, as most others of major significance in history, we historians should think twice before we spurn the help offered by our colleagues in the social sciences, who have, as it happens, been particularly active in the last few years in theorizing about the typology, causes, and evolutionary patterns of this particular phenomenon. The purpose of this chapter is not to advance any new hypothesis, but to provide a summary view and critical examination of the work that has been going on.
The first necessity in any inquiry is a careful definition of terms: what is, and what is not, a revolution? According to one view, it is change effected by the use of violence in government, and/or regime, and/or society.1 By society is meant the consciousness and the mechanics of communal solidarity, which may be tribal, peasant, kinship, national, and so on; by regime is meant the constitutional structure – democracy, oligarchy, monarchy; and by government is meant specific political and administrative institutions. Violence, it should be noted, is not the same as force; it is force used with unnecessary intensity, unpredictably, and usually destructively.2 This definition of revolution is a very broad one, and two historians of the French Revolution, Crane Brinton and Louis Gottschalk, would prefer to restrict the use of the word to the major political and social upheavals with which they are familiar, the ‘Great Revolutions’, as George S. Pettee calls them.3
Even the wider definition allows the historian to distinguish between, on the one hand, the seizure of power that leads to a major restructuring of government or society, the establishment of a new set of values for distributive justice, and the replacement of the former élite by a new one, and on the other hand, the coup d’état involving no more than a change of ruling personnel by violence or threat of violence. This latter is the norm in Latin America, where it occurred thirty-one times in the ten years 1945–55. Merle Kling has arrived at a suggestive explanation of this Latin American phenomenon of chronic political instability, limited but frequent use of violence, and almost complete lack of social or institutional change. He argues that ownership of the principal economic resources, both agricultural and mineral, is concentrated in the hands of a tiny, very stable, élite of enormously wealthy monoculture landlords and mining capitalists. This élite is all-powerful and cannot be attacked by opposition groups within the country; externally, however, it is dependent on foreign interests for its markets and its capital. In this colonial situation of a foreign-supported, closed plutocracy, the main avenue of rapid upward social mobility for non-members of the élite leads, via the army, to the capture of the government machine, which is the only accessible source of wealth and power. This political instability is permitted by the élite on the condition that its own interests are undisturbed. Instability, limited violence, and the absence of social or institutional change are therefore all the product of the contradiction between the realities of a colonial economy run by a plutocracy and the façade of political sovereignty – between the real, stable power of the economic élite and the nominal, unstable control of politicians and generals.4
The looser definition of revolution thus suits both historians of major social change and historians of the palace coup. It does, however, raise certain difficulties. First, there is a wide range of changes of government by violence which are neither a mere substitution of personalities in positions of power nor a prelude to the restructuring of society; second, conservative counter-revolutions become almost impossible to fit into the model; and last, it remains hard to distinguish between colonial wars, civil wars, and social revolution.
To avoid these difficulties, an alternative formulation has recently been put forward by a group of social scientists working mainly at Princeton. They have dropped the word ‘revolution’ altogether and put ‘internal war’ in its place.5 This is defined as any attempt to alter state policy, rulers, or institutions by the use of violence in societies where violent competition is not the norm and where well-defined institutional patterns exist.6 This concept seems to be a logical consequence of the preoccupation of sociologists in recent years with a model of society in a stable, self-regulating state of perpetual equipoise. In this utopian world of universal harmony all forms of violent conflict are anomalies, to be treated alike as pathological disorders of a similar species. This is a model which, although it has its uses for analytical purposes, bears little relation to the reality familiar to the historian. It looks to a society without change, with universal consensus on values, with complete social harmony, and isolated from external threats; no approximation to such a society has ever been seen.
The crude opposite model is based on pure interest theory and postulates that the social order rests on physical coercion of the majority by a minority in order to distribute material rewards and power in an inequitable way. The state claims a monopoly of violence, and the consequent suppression of internal disorder is its raison d’être. Societies become unstable only because the relation between authority and force becomes unstable. It is obvious that this model bears no more relation to reality than its opposite, since society is in fact both a moral community held together by shared values, which give the state legitimacy, and also a system of control, employing in the last resort the force necessary to prevent deviance and disorder. A more reasonable model is one which accepts that all societies are in a condition of uneasy equilibrium, whose stability is always threatened by a host of political conflicts, but which is usually held in balance partly by social norms and ideological beliefs and partly by physical sanctions.7 Instability may arise from material conflicts over the distribution of scarce economic resources or political power; or from a breakdown of values due to inadequate socialization of the young, exacerbated conflict between mutually incompatible roles, or group dissensus over norms. Such conflicts are usually settled by suitable adjustment mechanisms, and it is only in rare moments that a society is sufficiently shaken to undertake major structural alterations.
The first objection to the all-embracing formula of internal war is that, by covering all forms of physical conflict from strikes and terrorism to civil war, it isolates the use of violence from the normal process of societal adjustment. Though some of the users of the term express their awareness that the use of violence for political ends is a fairly common occurrence, the definition they have established in fact excludes all times and places where it is common. It thus cuts out most societies the world has ever known, including Western Europe in the middle ages and Latin America today. Second, it isolates one particular means, physical violence, from the political ends that it is designed to serve. Clausewitz’s famous definition of external war is equally applicable to internal war, civil war, or revolution: ‘War is not only a political act, but a real political instrument; a continuation of political transactions, an accomplishment of them by different means. That which remains peculiar to war relates only to the peculiar nature of its means.’8
It is perfectly true that any means by which society exercises pressure or control, whether it is administrative organization, constitutional law, economic interest or physical force, can be a fruitful field of study in its own right, so long as its students remain aware that they are looking at only one part of a larger whole. It is also true that there is something peculiar about violence, if only because of man’s highly ambivalent attitude towards the killing of his own species. Somehow he regards physical force as different in kind from, say, economic exploitation or psychological manipulation as a means of exercising power over others. But this distinction is not one of much concern to the historian of revolution, in which violence is a normal and natural occurrence. The concept of internal war is too broad in its comprehension of all types of violence from civil wars to strikes, too narrow in its restriction to normally non-violent societies, too limited in its concern with one of many means, too arbitrary in its separation of this means from the ends in view, and too little concerned with the complex roots of social unrest to be of much practical value to him.
The most fruitful typology of revolution is that of Chalmers Johnson, set out in a pamphlet that deserves to be widely read.9 He sees six types, identified by the targets selected for attack, whether the government personnel, the political regime, or the community as a social unit; by the nature of the carriers of revolution, whether a mass or an élite; and particularly by the goals and the ideologies, whether reformist, eschatological, nostalgic, nation-forming, élitist or nationalist. The first type, the jacquerie, is a spontaneous mass peasant rising, usually carried out in the name of the traditional authorities, Church and King, and with the limited aims of purging the local or national élites. Examples are the English Peasant Revolt of 1381, Kett’s Rebellion in Norfolk of 1549, and the Pugachev Rebellion in Russia in 1773–5. The second type, the millenarian rebellion, is similar to the first but with the added feature of a utopian dream, inspired by a living messiah. This type can be found at all times, in all parts of the world, from the Florentine Revolution led by Savonarola in 1494, to the Anabaptist rebellion in Munster led by John Mathijs and John Beukels in 1533–1535, to the Sioux Ghost-Dance Rebellion inspired by the Paiute prophet, Wovoka, in 1890. It has attracted a good deal of attention from historians in recent years, partly because the career of Hitler offered overwhelming proof of the enormous historical significance of a charismatic leader, and partly because of a growing interest in the ideas of Max Weber.10 The third type is the anarchistic rebellion, the nostalgic reaction to progressive change, involving a romantic idealization of the old order; the Pilgrimage of Grace and the Vendée are examples of this.
The fourth is that very rare phenomenon, the Jacobin Communist revolution. This type of revolution can occur only in a highly centralized state with good communications and a large capital city, and its target is government, regime, and society – the lot. The result is likely to be the creation of a new national consciousness under centralized, military authority, a re-distribution of property and authority, and the erection of a more rational, and hence more efficient, social and bureaucratic order on the ruins of the old ramshackle structure of privilege, nepotism, and corruption.
The fifth type is the conspiratorial coup d’ état, the planned work of a tiny élite fired by an oligarchic, sectarian ideology. This qualifies as a revolutionary type only if it in fact anticipates a mass movement and inaugurates social change – for example, the Nasser Revolution in Egypt or the Castro Revolution in Cuba; it is thus clearly distinguished from the palace revolt, assassination, dynastic succession-conflict, strike, banditry, and other forms of violence, which are all subsumed under the ‘internal war’ rubric.
Finally, there is the militarized mass insurrection, a new phenomenon of the twentieth century in that it is a deliberately planned mass revolutionary war, guided by a dedicated élite. The outcome of guerilla warfare is determined by political attitudes, not military strategy or material, for the rebels are wholly dependent on broad popular support. In all cases on record the ideology that attracts the mass following has been a combination of xenophobic nationalism and Marxism, with by far the greater stress on the former. This type of struggle has occurred in Yugoslavia, China, Algeria, and Vietnam.
Although, like any schematization of the historical process, this six-fold typology is concerned with ideal types, and although in practice individual revolutions may sometimes display characteristics of several different types, the fact remains that this is much the most satisfactory classification we have so far; it is one that working historians can recognize and use with profit. The one obvious criticism is semantic, an objection to the use of the phrase ‘Jacobin Communist revolution’.11 Some of Johnson’s examples are Communist, such as the Russian or Chinese Revolutions; others are Jacobin but not Communist, such as the French Revolution or the Turkish Revolution of 1908–22. It would be better to revert to Pettee’s category of ‘Great Revolutions’, and treat Communist revolutions as a sub-category, one type, but not the only type, of the modernizing revolutionary process.
Given this classification and definition of revolution, what are its root causes? Here everyone is agreed in making a sharp distinction between long-run, underlying causes – the preconditions, which create a potentially explosive situation and can be analysed on a comparative basis – and immediate, incidental factors – the precipitants, which trigger the outbreak and which may be nonrecurrent, personal, and fortuitous. This effectively disposes of the objections of those historians whose antipathy to conceptual schematization takes the naïve form of asserting the uniqueness of each historical event.
One of the first in the field of model-building was Crane Brinton, who as long ago as 1938 put forward a series of uniformities common to the four great Western revolutions: English, French, American, and Russian. These included an economically advancing society, growing class and status antagonisms, an alienated intelligentsia, a psychologically insecure and politically inept ruling class, and a governmental financial crisis.12
The subjectivity, ambiguity, and partial self-contradiction of this and other analyses of the causes of specific revolutions – for example the French Revolution – have been cruelly shown up by Harry Eckstein.13 He has pointed out that commonly adduced hypotheses run the spectrum of particular conditions, moving from the intellectual (inadequate political socialization, conflicting social myths, a corrosive social philosophy, alienation of the intellectuals) to the economic (increasing poverty, rapid growth, imbalance between production and distribution, long-term growth plus short-term recession) to the social (resentment due to restricted élite circulation, confusion due to excessive élite recruitment, anomie due to excessive social mobility, conflict due to the rise of new social classes) to the political (bad government, divided government, weak government, oppressive government). Finally there are explanations on the level of general process, such as rapid social change, erratic social change, or a lack of harmony between the state structure and society, the rulers and the ruled. None of these explanations are invalid in themselves, but they are often difficult or impossible to reconcile one with the other, and are so diverse in their range and variety as to be virtually impossible to fit into an ordered, analytical framework. What, then, is to be done?
Fundamental to all analyses, whether by historians like Brinton and Gottschalk or by political scientists like Johnson and Eckstein, is the recognition of a lack of harmony between the social system on the one hand and the political system on the other. This situation Johnson calls dysfunction, a word derived from the structural-functional equilibrium model of the sociologists. This dysfunction may have many causes, some of which are merely random or cyclical, such as may develop because of personal weaknesses in hereditary kingships or single-party regimes. In these cases, the revolution will not take on serious proportions, and will limit itself to attacks on the governing élite, or at most the governing institutions, leaving regime and society intact. In most cases, however, including all those of real importance, the dysfunction is the result of some new and developing process, as a result of which certain social sub-systems find themselves in a condition of relative deprivation. Rapid economic growth, imperial conquest, new metaphysical beliefs, and important technological changes are the four commonest factors involved, in that order. If the process of change is sufficiently slow and sufficiently moderate, the dysfunction may not arise to dangerous levels. Alternatively, the élite may adjust to the new situation with sufficient rapidity and skill to ride out the storm and retain popular confidence. But if the change is rapid and profound, it may cause the sense of deprivation, alienation and anomie and spread into many sectors of society at once, causing what Johnson calls multiple dysfunction, which may be all but incurable within the existing political system.
In either case the second vital element in creating a revolutionary situation is the condition and attitude of the entrenched élite, a factor on which Eckstein rightly lays great stress. The élite may lose its manipulative skill, or its military superiority, or its self-confidence, or its cohesion; it may become estranged from the non-élite, or overwhelmed by a financial crisis; it may be incompetent, or weak or brutal. Any combination of two or more of these features will be dangerous. What is ultimately fatal, however, is the compounding of its errors by intransigence. If it fails to anticipate the need for reform, if it blocks all peaceful, constitutional means of social adjustment, then it unites the various deprived elements in single-minded opposition to it, and drives them down the narrow road to violence. It is this process of polarization into two coherent groups or alliances of what are naturally and normally a series of fractional and shifting tensions and conflicts within a society that both Peter Amman and Wilbert Moore see as the essential preliminary to the outbreak of a Jacobin revolution.14 To conclude, therefore, revolution becomes possible when a condition of multiple dysfunction meets an intransigent élite: just such a conjunction occurred in the decades immediately before the English, the French, and the Russian Revolutions.
Revolution only becomes probable (Johnson might say ‘certain’), however, if certain special factors intervene: the ‘precipitants’ or ‘accelerators’. Of these, the three most common are the emergence of an inspired leader or prophet; the formation of a secret, military, revolutionary organization; and the crushing defeat of the armed forces in foreign war. This last is of critical importance since it not only shatters the prestige of the ruling élite, but also undermines the morale and discipline of the soldiers and thus opens the way to the violent overthrow of the existing government.
The defects of Johnson’s 1964 model were partly overcome by his revision of 1966. The main defect was that it concentrated too much on objective structural conditions, and attempted to relate these conditions directly to political actions. In the new version far greater scope is allowed for ideas and values as the mediating factor between the conditions and the response. Johnson now recognizes that a social system may be upset just as much by changes in values – for example, the spread of new ideas either coming from ‘marginal men’ or from innovating élites – as by changes in the environment – for example, changes in industrial techniques and organization, modern medicine or foreign trade. He unnecessarily complicates his categories by sub-dividing them into exogenous and indogenous, which is largely meaningless. For what does it matter whether the technology or the idea is home grown or imported? All the same, this new emphasis on both environment and values makes it possible to explain why in the past similar political activity has occurred in different environmental conditions, and different activity in similar conditions. An approach which lays equal stress on such things as anomie, alienation of the intellectuals, frustrated popular aspirations, élite estrangement, and loss of élite self-confidence, is more likely to produce a satisfactory historical explanation than is one that sticks to the objective social situation.
In his later book Johnson goes far to remedy the second major defect of his 1964 model, namely its determinist neglect of the role of the unique and the personal. He seemed to regard his accelerators as automatic triggers, ignoring the area of unpredictable personal choice that is always left to the ruling élite and to the revolutionary leaders, even in a situation of multiple dysfunction exacerbated by an accelerator. Revolution is never inevitable – or rather the only evidence of its inevitability is that it actually happens. Consequently the only way to prove this point is to indulge in just the kind of hypothetical argument that historians prudently try to avoid. Thus it is still just possible that modernization may take place in Morocco and India without revolution. The modernization and industrialization of Germany and Britain took place without revolution in the nineteenth century (though it can be argued that in the latter case the process was slow by twentieth-century standards, and that, as is now becoming all too apparent, the modernization was far from complete). Some think that a potentially revolutionary situation in the United States in the 1930s was avoided by shrewd political action. In his 1966 model Johnson recognized the role of the personality as a semi-independent variable in the social system, free to operate within the limits roped off and labelled ‘crime’ or ‘mental sickness’.
One of the unresolved difficulties of the Johnson model is that it makes no allowance for the fact that political actions taken to remedy dysfunction often precipitate change themselves. This produces the paradoxical hypothesis that measures designed to restore equilibrium in fact upset equilibrium. Because he begins with his structural-functional equilibrium model, Johnson is a victim of the fallacy of intended consequences. As often as not in history it is the unintended consequences that really matter: to mention but one example, it was Louis XVI’s belated and half-hearted attempts at reform that provoked the aristocratic reaction, which in turn opened the way to the bourgeois, the peasant, and the sans-culotte revolutions. Finally the dysfunction concept is not altogether easy to handle in a concrete historical case. If societies are regarded as being in a constant state of multiple tension, then some degree of dysfunction is always present. Some group is always in a state of relative deprivation due to the inevitable process of social change.
Recognition of this fact leads Eckstein to point out the importance of forces working against revolution. Historians, particularly those formed in the Western liberal tradition, are reluctant to admit that ruthless efficient repression – as opposed to bumbling, half-hearted repression – involving the physical destruction of leading revolutionaries and effective control of the media of communication, can crush incipient revolutionary movements. Repression is particularly effective when governments know what to look for, when they have before their eyes the unfortunate example of other governments overthrown by revolutionaries elsewhere. Reaction, in fact, is just as infectious as revolution. Moreover, diversion of energy and attention to successful – as opposed to unsuccessful – foreign war can ward off serious internal trouble. Quietist – as opposed to activist – religious movements may serve as the opiate of the people, as Halévy suggested about Methodism in England. Bread and circuses may distract popular attention. Timely – as opposed to untimely – political concessions may win over moderate opinion and isolate the extremists.
Basing himself on this suggestive analysis, Eckstein produces a paradigm for universal application. He sees four positive variables – élite inefficiency, disorienting social process, subversion, and available rebel facilities – and four negative variables – diversionary mechanisms, available incumbent facilities, adjustive mechanisms, and effective repression. Each type of internal war, and each step of each type, can, he suggests, be explained in terms of these eight variables. While this may be true, it is fair to point out that some of the variables are themselves the product of more deep-seated factors, others mere questions of executive action that may be determined by the accidents of personality. Disruptive social process is a profound cause; élite inefficiency a behaviour pattern; effective repression a function of will; facilities the by-product of geography. One objection to the Eckstein paradigm is therefore that it embraces different levels of explanation and fails to maintain the fundamental distinction between preconditions and precipitants. Second, it concentrates on the factors working for or against the successful manipulation of violence rather than on the underlying factors working to produce a revolutionary potential. This is because the paradigm is intended to apply to all forms of internal war rather than to revolution proper, and because all that the various forms of internal war have in common is the use of violence. It is impossible to tell how serious these criticisms are until the paradigm has been applied to a particular historical revolution. Only then will its value become apparent.
If we take the behaviourist approach, then a primary cause of revolutions is the emergence of an obsessive revolutionary mentality. But how closely does the subjective mentality relate to the objective material circumstances? In every revolutionary situation one finds a group of men – fanatics, extremists, zealots – so convinced of their own righteousness and of the urgent need to create a new Jerusalem on earth (whether formally religious or secular in inspiration is irrelevant) that they are prepared to smash through the normal restraint of habit, custom, and convention. Such men were the seventeenth-century English Puritans, the eighteenth-century French Jacobins, the early twentieth-century Russian Bolsheviks, and the late twentieth-century Maoists in Pekin, Paris and New York. But what makes such men is far from certain. What generates such ruthlessness in curbing evil, such passion for discipline and order? Rapid social mobility, both horizontal and vertical, and particularly urbanization, certainly produces a sense of rootlessness and anxiety. In highly stratified societies even some of the newly-risen elements may find themselves under stress.15 While some of the arrivistes are happily absorbed in their new strata, others remain uneasy and resentful. If they are snubbed and rebuffed by the older members of the status group to which they aspire by reason of their new wealth and position, they are likely to become acutely conscious of their social inferiority, and may be driven either to adopt a pose plus royaliste que le roi or to dream of destroying the whole social order. In the latter case they may try to allay their sense of insecurity by imposing their norms and values by force upon society at large. This is especially the case if there is available a moralistic ideology like Puritanism or Marxism to which they can attach themselves, and which provides them with unshakable confidence in their own rectitude.
But why does the individual react in one particular way rather than another? Some would argue that the character of the revolutionary is formed by sudden ideological conversion in adolescence or early adult life (to Puritanism, Jacobinism, or Bolshevism) as a refuge from this anxiety state.16 What is not acceptable is the fashionable conservative cliché that the revolutionary and the reformer are merely the chance product of unfortunate psychological difficulties in childhood. It is possible that this is the mechanism by which such feelings are generated, though there is increasing evidence of the continued plasticity of human character, until at any rate post-adolescence, and of the happy and well-adjusted childhood of most of the modern radicals. In any case, this crude psychological reductionism cannot explain why some individuals respond to stress by adopting radical ideas and others live blamelessly conformist lives. The other main objection to this theory is that it fails to explain why these particular attitudes become common only in certain classes and age groups at certain times and in certain places. This failure strongly suggests that the cause of this state of mind lies not in the personal maladjustment of the individuals or their parents, but in the social conditions that created that maladjustment. Talcott Parsons treats disaffection or ‘alienation’ as a generalized phenomenon that may manifest itself in crime, alcoholism, drug addiction, daytime fantasies, religious enthusiasm, or serious political agitation. To use Robert Merton’s formulation, ritualism and retreatism are two possible psychological escape-routes; innovation and rebellion two others.17 Gripped by frustration generated by environmental circumstances, the individual may get drunk, beat his wife, retire to a monastery, write a book, throw a bomb, or start a revolution.
Even if we accept a generally behaviourist approach (which I do), the fact remains that many of the underlying causes both of the alienation of the revolutionaries and of the weakness of the incumbent élite are economic in origin; and it is in this area that some interesting work has centred. In particular a fresh look has been taken at the contradictory models of Marx and de Tocqueville, the one claiming that popular revolution is a product of increasing misery, the other that it is a product of increasing prosperity. Two economists, Sir Arthur Lewis and Mancur Olson, have pointed out that because of their basic social stability, both pre-industrial and highly industrialized societies are relatively free from revolutionary disturbance.18 In the former societies people accept with little question the accepted rights and obligations of family, class, and caste. Misery, oppression, and social injustice are passively endured as inevitable features of life on earth. It is in societies experiencing rapid economic growth that the trouble usually occurs. Lewis, who is thinking mostly about the newly emerging countries, primarily of Africa, regards the sense of frustration that leads to revolution as a consequence first of the dislocation of the old status patterns by the emergence of four new classes – the proletariat, the capitalist employers, the urban commercial and professional middle class, and the professional politicians; and second of the disturbance of the old income patterns by the sporadic and patchy impact of economic growth, which creates new wealth and new poverty in close and conspicuous juxtaposition. Both phenomena he regards as merely transitional, since in a country fully developed economically there are strong tendencies toward the reduction of inequalities of opportunity, income, and status.
This model matches fairly well the only detailed analysis of a historical revolution in which a conscious effort has been made to apply modern sociological methods. In his recent study of the Vendée, Charles Tilly argues that a counter-revolutionary situation was the consequence of special tensions created by the immediate juxtaposition of, on one hand, parish clergy closely identified with the local communities, great absentee landlords and old-fashioned subsistence farming; and, on the other, a large-scale textile industry on the putting-out system and increasing bourgeois competitition.19 Though the book is flawed by a tendency to take a ponderous sociological hammer to crack a simple little historical nut, it is none the less a suggestive and successful pioneering example of the application of new hypotheses and techniques to historical material.
Olson has independently developed a more elaborate version of the Lewis theory. He argues that revolutionaries are déclassé and freed from the social bonds of family, profession, village or manor; and that these individuals are the product of rapid economic growth, which creates both nouveaux riches and nouveaux pauvres. The former, usually middle-class and urban artisans, are better off economically, but are disoriented, rootless, and restless; the latter may be workers whose wages have failed to keep pace with inflation, workers in technologically outdated and therefore declining industries, or the unemployed in a society in which the old cushions of the extended family and the village have gone, and in which the new cushion of social security has not yet been created. The initial growth phase may well cause a decline in the standard of living of the majority because of the need for relatively enormous forced savings for reinvestment. The result is a revolution caused by the widening gap between expectations – social and political for the new rich, economic for the new poor – and the realities of everyday life.
A sociologist, James C. Davis, agrees with Olson that the fundamental impetus towards a revolutionary situation is generated by rapid economic growth but he associates such growth with a generally rising rather than a generally falling standard of living, and argues that the moment of potential revolution is reached only when the long-term phase of growth is followed by a short-term phase of economic stagnation or decline.20 The result of this ‘J-curve’, as he calls it, is that steadily soaring expectations, newly created by the period of growth, shoot further and further ahead of actual satisfaction of needs. Successful revolution is the work neither of the destitute nor of the well-satisfied, but of those whose actual situation is improving less rapidly than they expect.
These economic models have much in common, and their differences can be explained by the fact that Lewis and Olson are primarily concerned with the long-term economic forces creating instability, and Davis with the short-term economic factors that may precipitate a crisis. Moreover their analyses apply to different kinds of economic growth, of which three have recently been identified by W. W. Rostow and Barry Supple: there is the expansion of production in a pre-industrial society, which may not cause any important technological, ideological, social, or political change; there is the phase of rapid growth, involving major changes of every kind; and there is the sustained trend towards technological maturity.21 Historians have been quick to see that these models, particularly that of Rostow, can be applied only to a limited number of historical cases. The trouble is not so much that in any specific case the phases – particularly the last two – tend to merge into one another, but that changes in the various sectors occur at irregular and unexpected places on the time-scale in different societies. In so far as there is any validity in the division of the stages of growth into these three basic types, the revolutionary model of Olson and Lewis is confined to the second; that of Davis is applicable to all three.
The Davis model fits the history of Western Europe quite well, for it looks as if in conditions of extreme institutional and ideological rigidity the first type of economic growth may produce frustrations of a very serious kind. Revolutions broke out all over Europe in the 1640s, twenty years after a secular growth phase had come to an end.22 C. E. Labrousse has demonstrated the existence of a similar economic recession in France from 1778,23 and from 1914 the Russian economy was dislocated by the war effort after many years of rapid growth. Whatever its limitations in any particular situation, the J-curve of actual satisfaction of needs is an analytical tool that historians can usefully bear in mind as they probe the violent social upheavals of the past.
As de Tocqueville pointed out, this formula of advance followed by retreat is equally applicable to other sectors. Trouble arises if a phase of liberal governmental concessions is followed by a phase of political repression; a phase of fairly open recruitment channels into the élite followed by a phase of aristocratic reaction and a closing of ranks; a phase of weakening status barriers by a phase of reassertion of privilege. The J-curve is applicable to other than purely economic satisfactions, and the apex of the curve is the point at which underlying causes, the preconditions, merge with immediate factors, the precipitants. The recipe for revolution is thus the creation of new expectations by economic improvement and some social and political reforms, followed by economic recession, governmental reaction, and aristocratic resurgence, which widen the gap between expectations and reality.
All these attempts to relate dysfunction to relative changes in economic prosperity and aspirations are hampered by two things, of which the first is the extreme difficulty in ascertaining the facts. It is never easy to discover precisely what is happening to the distribution of wealth in a given society. Even now, even in highly developed Western societies with massive bureaucratic controls and quantities of statistical data, there is no agreement about the facts. Some years ago it was confidently believed that in both Britain and the United States incomes were being levelled, and that extremes of both wealth and poverty were being steadily eliminated. Today, no one quite knows what is happening in either country.24 And if this is true now, still more is it true of societies in the past about which the information is fragmentary and unreliable.
Second, even if they can be clearly demonstrated, economic trends are only one part of the problem. Historians are increasingly realizing that the psychological responses to changes in wealth and power are not only not precisely related to, but are politically more significant than, the material changes themselves. As Marx himself realized at one stage, dissatisfaction with the status quo is not determined by absolute realities but by relative expectations. ‘Our desires and pleasures spring from society; we measure them, therefore, by society, and not by the objects which serve for their satisfaction. Because they are of a social nature, they are of a relative nature.’25 Frustration may possibly result from a rise and subsequent relapse in real income. But it is perhaps more likely to be caused by a rise in aspirations that outstrips the rise in real income; or by a rise in the relative economic position in society of the group in question, followed by a period in which its real income continues to grow, but less fast than that of other groups around it. Alternatively it may represent a rise and then decline of status, largely unrelated to real income; or if there is a relationship between status and real income, it may be an inverse one. For example, social scientists seeking to explain the rise of the radical right in the United States in the early 1950s and again in the early 1960s attribute it to a combination of great economic prosperity and an aggravated sense of insecurity of status.26 Whether or not this is a general formula for right-wing rather than left-wing revolutionary movements is not yet clear.
Moreover the problem is further complicated by an extension of reference-group theory.27 Human satisfaction is related not to existing conditions but to the condition of a social group against which the individual measures his situation. In an age of mass communications and the wide distribution of cheap radio receivers, even among the impoverished illiterate of the world, knowledge of high consumption standards elsewhere spreads rapidly, and as a result the reference group may be in another, more highly developed, country or even continent. Under these circumstances revolutionary conditions may be created before industrialization has got properly under way.
All modern theories of revolutionary behaviour are based on the ‘relative deprivation’ hypothesis that it is generated by the gap between expectations and the perceptions of reality. The psychological chain of causation links the gap to frustration, and frustration to aggression. This is the central theme of Professor Gurr’s study of revolutions.28 Although darkened by obfuscating jargon, much of which conceals solemn statements of the obvious, although peppered with a maddening array of numbered hypotheses and corollaries, and although disfigured with bewildering charts full of boxes and arrows running every way, the book nevertheless has interesting things to say. Relative deprivation may take the form of a gap between expectations and perceived capabilities over three general sets of values: ‘welfare’, meaning economic conditions; political power; and social status – the three old Weberian categories. The gap may open up through decremental deprivation, involving stable expectations and a decline of capabilities. This may be caused by a general economic depression, or more likely by selective economic impoverishment of a specific group, for example, English hand-loom weavers in the early nineteenth century, or parish clergy in the twentieth. Or it may be caused by political repression and the exclusion of a certain group from political participation; or it may be caused by a relative status decline due to the rise of competing groups. The second type is progressive deprivation, in which a long period of improvement is followed by a sudden downward turn in capabilities, while expectations go on rising because of past experience. This is Davis’s ‘J-curve’, now extended to more than merely material circumstances. The third is aspirational deprivation, in which expectations rise while capabilities remain the same. This is fairly rare in the past, although the Reformation may be looked at in this light, in the terms of rising lay expectations of spiritual zeal from the clergy, which the latter could not satisfy. Today it is common among under-developed countries, as populations learn about the benefits of modernization elsewhere. The intensity of the deprivation varies with the size of the gap; with the importance the group ascribes to the value affected – the poor attach most value to economic conditions, the rich to political participation; with the extent that alternative satisfactions are available; and with the duration of the existence of the gap.
The factors causing a rise of expectations are identified as urbanization, literacy, the demonstration effect of striking improvement elsewhere, new ideologies, especially of a chiliastic character, and an inconsistency between the three variables of wealth, power and status. G. Lenski has laid great emphasis on this last factor, itself a product of social mobility.29 The upwardly mobile tend to find that their status and power are not commensurate with their wealth, since the old élites will not accept them as social equals or share power with them. The downwardly mobile retain their status for a while, but lose wealth and power. In either case the discrepancy breeds anxiety, the anxiety resentment, and the resentment aggression which may find an outlet in radical ideologies and actions. Gurr points to the ‘zero-sum’ aspect of the distribution of all goods. Even economic resources are valued largely in relative terms, compared with the wealth of close reference groups, and power and status are relative by definition; the more of them obtained by one person or group, the less of them is available for others. On the other hand moderation in the appetite of the élite, some modesty in the display of affluence, some restraint in its assertions of social superiority, some tolerance of élite circulation from below, all help to dampen the fires of resentment.
All this is quite helpful as far as it goes, in identifying the causes of resentment in a social system and in devising ways of assessing its intensity and spread. But Gurr is not particularly helpful on the coercive balance of physical force between élite and dissidents, or on the role of ideology, or on the complex interaction of all these factors to generate a revolution. In particular he ignores the ‘X factor’ of personality and choice, and his model is so mechanical and schematized as to be hard to apply to any given situation. His main contribution is his analysis of the types and causes of relative deprivation.
Another recent theory regards the ‘Great Revolutions’ as stages in the modernization of under-developed societies.30 The central feature of a pre-revolutionary situation is the unwillingness or incapacity of the ruling élite to carry out the transformation needed to raise productivity, increase trade, expand education, redistribute and exploit capital resources. The principal consequence of the revolution is to strengthen the governmental machinery to carry out these tasks. The modernization theory of revolution looks suspiciously like the Marxist bourgeois revolution in a new guise. It is a theory which fits late twentieth-century revolutions in backward countries well enough, but it is difficult to accommodate the older revolutions to the model. In what ways were the pre-revolutionary governments of Charles I and Louis XVI failing in their modernization functions? In what ways were the Parliamentary leaders of 1640 or 1789 trying to innovate with this purpose in mind? The recent stress on the political and modernizing aspects of the causes and consequences of revolution is a healthy antidote to the excessive social and economic determinism of some earlier theorists, but it under-plays the very important components of social change and ideological innovation.31 Nor does it explain the strong egalitarian component in past ‘Great Revolutions’, including the Bolshevik, with insistent demands for greater participation in decision-making, which is hardly the way to stream-line the state bureaucracy for decisive, modernizing action. One-party dictatorship is a surer way to achieve results than the fumblings of democracy in its infancy.
The last area in which some new theoretical work has been done is in the formulation of hypotheses about the social stages of a ‘Great Revolution’. One of the best attacks on this problem was made by Crane Brinton, who was thinking primarily about the French Revolution, but who extended his comparisons to the three other major Western revolutionary movements. He saw the first phase as dominated by moderate bourgeois elements; their supersession by the radicals; a reign of terror; a Thermidorian reaction; and the establishment of strong central authority under military rule to consolidate the limited gains of the revolution. In terms of mass psychology he compared revolution with a fever that rises in intensity, affecting nearly all parts of the body politic, and then dies away.
A much cruder and more elementary model has been advanced by an historian of the revolutions of 1848, Peter Amman.32 He sees the modern state as an institution holding a monopoly of physical force, administration, and justice over a wide area, a monopoly dependent more on habits of obedience than on powers of coercion. Revolution may therefore be defined as a breakdown of the monopoly due to a failure of these habits of obedience. It begins with the emergence of two or more foci of power, and ends with the elimination of all but one. Amman includes the possibility of ‘suspended revolution’, with the existence of two or more foci not yet in violent conflict.
This model admittedly avoids some of the difficulties raised by more elaborate classifications of revolution: how to distinguish a coup d’état from a revolution; how to define the degrees of social change; how to accommodate the conservative counter-revolution, and so on. It certainly offers some explanation of the progress of revolution from stage to stage as the various power blocks that emerge on the overthrow of the incumbent regime are progressively eliminated; and it explains why the greater the public participation in the revolution, the wider the break with the habits of obedience, and therefore the slower the restoration of order and centralized authority. But it throws the baby out with the bathwater. It is impossible to fit any decentralized traditional society, or any modern federal society, into the model. Moreover, even where it might be applicable, it offers no framework for analysing the roots of revolution, no pointers for identifying the foci of power, no means of distinguishing between the various revolutionary types, and its notion of ‘suspended revolution’ is little more than verbal evasion.
Though it is set out in a somewhat confused, over-elaborate, and unnecessarily abstract form, the most convincing description of the social stages of revolution is that outlined by Rex D. Hopper.33 He sees four stages. The first is characterized by indiscriminate, uncoordinated mass unrest and dissatisfaction, the result of dim recognition that traditional values no longer satisfy current aspirations. The next stage sees this vague unease beginning to coalesce into organized opposition with defined goals, an important characteristic being a shift of allegiance by the intellectuals from the incumbents to the dissidents, the advancement of an ‘evil men’ theory, and its abandonment in favour of an ‘evil institutions’ theory. At this stage there emerge two types of leaders: the prophet, who sketches the shape of the new utopia upon which men’s hopes can focus, and the reformer, working methodically toward specific goals. The third, the formal stage, sees the beginning of the revolution proper. Motives and objectives are clarified, organization is built up, a statesman-leader emerges. Then conflicts between the left and the right of the revolutionary movement become acute, and the radicals take over from the moderates. The fourth and last stage sees the legalization of the revolution. It is a product of psychological exhaustion as the reforming drive burns itself out, moral enthusiasm wanes, and economic distress increases. The administrators take over, strong central government is established, and society is reconstructed on lines that embody substantial elements of the old system. The result falls far short of the utopian aspirations of the early leaders, but it succeeds in meshing aspirations with values by partly modifying both, and so allows the reconstruction of a firm social order.
Some of the writing of contemporary social scientists are ingenious feats of verbal juggling in an esoteric language, performed around the totem pole of an abstract model, surrounded as far as the eye can see by the arid wastes of terminological definitions and mathematical formulae. Small wonder the historian finds it hard to digest the gritty diet of this neo-scholasticism, as it has been aptly called. The more historically-minded of the social scientists, however, have a good deal to offer. The history of history, as well as of science, shows that advances depend partly on the accumulation of factual information, but rather more on the formulation of hypotheses that reveal the hidden relationships and common properties of apparently distinct phenomena. For, all their faults, social scientists can supply a corrective to the antiquarian fact-grubbing to which historians are so prone; they can direct attention to problems of general relevance, and away from the sterile triviality of so much historical research; they can ask new questions and suggest new ways of looking at old ones; they can supply new categories, and as a result may suggest new ideas.
1 Chalmers Johnson, Revolution and the Social System, Hoover Institution Studies, 3, Stanford, 1964.
2 Sheldon S. Wolin, ‘Violence and the Western political tradition’, American Journal of Orthopsychiatry, 33, January 1963, pp. 15–28.
3 C. Brinton, The Anatomy of Revolution, New York, 1938; L. Gottschalk, ‘Causes of revolution’, American Journal of Sociology, 50, July 1944, pp. 1–8; G. S. Pettee, The Process of Revolution, New York, 1938.
4 ‘Toward a theory of power and political instability in Latin America’, Western Political Quarterly, 9, 1956. See also D. Bivy, ‘Political instability in Latin America: the cross-cultural test of a causal model’, Latin American Research Review, 3, 1968; W. Dean, ‘Latin American Golpes and economic fluctuations, 1823–1966’, Social Science Quarterly, June 1970.
5 Harry Eckstein, ed., Internal War, New York, 1964; ‘On the etiology of internal war’, History and Theory, 4, no 2, 1965, pp. 133–63. I am grateful to Mr Eckstein for allowing me to read this article before publication.
6 The formula has been used by an historian, Peter Paret, in Internal War and Pacification: The Vendée, 1793–96, Princeton, 1961.
7 Barrington Moore, ‘The strategy of the social sciences’, in his Political Power and Social Theory, Cambridge, Mass., 1958; Ralph Dahrendorf, ‘Out of utopia: towards a reorientation of sociological analysis’, American Journal of Sociology, 64, September 1958, 115–27; C. Wright Mills, The Sociological Imagination, New York, 1959; Wilbert E. Moore, Social Change, Englewood Cliffs, 1963; T. R. Gurr, Why Men Rebel, Princeton, 1970. It should be noted that both the equilibrium and the conflict views of society have very respectable ancestries. The equilibrium model goes back to Rousseau – or perhaps Aquinas; the conflict model to Hobbes, Hegel, and Marx.
8 Quoted in Edward Mead Earle, ed., Makers of Modern Strategy, Princeton, 1943, pp. 104–5.
9 Revolution and the Social System, Stanford, 1964; a lengthier and modified version of some of these ideas is contained in his Social Change, Boston, 1966.
10 N. R. C. Cohn, Pursuit of the Millennium, New York, 1961; Eric J. Hobsbawm, Primitive Rebels, Manchester, 1959; S. L. Thrupp, ‘Millennial Dreams in Action’, Supplement 2, Comparative Studies in Society and History, The Hague, 1962; A. J. F. Köbben, ‘Prophetic movements as an expression of social protest’, Internationales Archiv für Ethnographie, 49, no. 1, 1960, pp. 117–64.
11 The phrase is not used in his Social Change.
12 Anatomy of Revolution, op. cit.
13 ‘On the etiology of internal war’, op. cit.
14 P. Amman, ‘Revolution: a redefinition’, Political Science Quarterly, 77, 1962.
15 Emile Durkheim, Suicide, The Free Press, Chicago, 1951, pp. 246–54; A. B. Hollingshead, R. Ellis and E. Kirby, ‘Social mobility and mental illness’, American Sociological Review, 19, 1954.
16 M. L. Walzer, ‘Puritanism as a revolutionary ideology’, History and Theory, 3, 1963.
17 T. Parsons, The Social System, The Free Press, Chicago, 1951; R. K. Merton, Social Structure, The Free Press, Chicago, 1957, ch. 4.
18 W. A. Lewis, ‘Commonwealth address’, in Conference Across a Continent, Toronto, 1963, pp. 46–60; M. Olson, ‘Rapid growth as a destabilizing force’, Journal of Economic History, 23, December 1963, pp. 529. I am grateful to Mr Olson for drawing my attention to Sir Arthur Lewis’s article, and for some helpful suggestions.
19 The Vendée, Cambridge, Mass., 1964.
20 ‘Toward a theory of revolution’, American Sociological Review, 27, February 1962, pp. 1–19, especially the graph on p. 6.
21 W.W. Rostow, The Stages of Economic Growth, Cambridge, Mass., 1960; B. Supple, The Experience of Economic Growth, New York, 1963, pp. 11–12.
22 E. J. Hobsbawm, ‘The crisis of the seventeenth century’, in T. H. Aston, ed., Crisis in Europe, 1560–1660, London, 1965, pp. 5–58.
23 La Crise de l’économie française à la fin de l’ancien régime et au début de la révolution, Paris, 1944.
24 Gabriel Kolko, Wealth and Power in America, New York, 1962; Richard M. Titmuss, Income Distribution and Social Change, London, 1962.
25 Davis, op. cit., p. 5, quoting Marx, Selected Works in Two Volumes, Moscow, 1955, 1, p. 947.
26 Daniel Bell, ed., The Radical Right, Garden City, 1963.
27 Merton, op. cit., ch. 9.
28 Why Men Rebel.
29 G. E. Lenski, ‘Status crystallization: a non-vertical dimension of social status’, American Sociological Review, 19, 1956.
30 C. E. Black, The Dynamics of Modernization: A Study of Comparative History, New York, 1966.
31 J. R. Gillis, ‘Political decay and the European revolutions, 1789–1848’, World Politics, 22, 1970.
32 ‘Revolution: a redefinition’, 22, op. cit.
33 ‘The revolutionary process’, Social Forces, 28, March 1950, pp. 279–97.