![]()
PART I
Theories of rebellion, repression, and responses to scarcity
These three essays summarize my main theoretical contributions to understanding why rebels and governments resort to violence for political objectives. Essay 1 is a revision of an article published in 1968 in World Politics (Gurr 1968b), which proposes that the violence of rebels is not only capable of general explanation, but that we know enough about the sources of human violence to specify in general theoretical terms some of the social psychological dynamics, and social patterns, that dispose people to collective violence. It sketches the essential arguments of my book, Why Men Rebel (1970, 2010) and is based on the fundamental assumption that explanation of political violence, in all its forms, should begin with an analysis of people’s hopes and grievances. We need to know what they think and believe if we are to understand why they participate in risky acts of protest and rebellion.
The revisions have streamlined the presentation and updated some of its terminology. The essence of the theoretical argument has not been altered. Moreover the research evidence cited, both on psychological dynamics and on civil violence, is that which was available to me in the mid-1960s. A few later bodies of work are cited in the footnotes.
The arguments of this essay prompted a great deal of empirical research, discussion, and criticism by the scholars who contributed to the explosion of systematic research on violent conflict after the mid-1960s. My own thoughts about what was valid about the theory, and how to modify it, are summarized in essay 14, Why Men Rebel Revisited: Observations on Revolution in Contemporary Africa.
A recurring theme in the criticisms of Why Men Rebel was that it seemed to assume a linear process in which people rebelled and governments responded. This raised the more general issue of why some governments base their rule and responses on varying degrees of repression and violence. The views of Terry Nardin (1971), Michael Stohl (1976), and Charles Tilly (1978) especially shaped my thinking about this issue, and my theoretical responses are embodied in essay 2, War, Revolution, and the Growth of the Coercive State, in which I show systematically why a state’s involvement in war or revolution underlies the development of what I call garrison and police states whose leaders are disposed to respond coercively to challenges. The analysis here gives little attention to the psychological dynamics of elites—it assumes they are shaped mainly by historical experiences and institutional context.
In an essay not included in this volume, Foundations for a General Theory of Political Coercion (Gurr 1986b), I used the arguments of the first two essays to develop a single conceptual scheme that accounts for both rebels’ use of violent political action and the reliance of regimes on agencies and policies of repression. One general postulate developed in this framework is that there is a general tendency toward equilibrium in the intensity of coercion used by regimes and their opponents; the equilibrium will be at a high level in societies that are sharply differentiated along segmental and class lines.
Essay 3 in this section assumes that scarcity and economic crises are recurrent features of human experience. The key questions concern their social and political consequences, based on theoretical arguments derived from the two previous essays. Some historical and contemporary evidence also is cited. The essay zeros in on three common outcomes: increasing inequality within and among countries; intensified political conflict about who gets what, when and how; and a retreat from democratic norms and practices. Some of the arguments may seem dated and irrelevant given the fracking-fueled exuberance of the 2010s. Yet regional scarcity and decline are everpresent in parts of Africa and Asia and may become much more pervasive in consequence of global climate change. Ecologists think systematically about environmental futures; readers should do the same about scenarios of conflict and repression in ecologists’ projected futures.
![]()
1
PSYCHOLOGICAL FACTORS IN POLITICAL VIOLENCE1
Until the 1960s many political scientists tended to regard violent civil conflict as a disfigurement of the body politic, neither a significant nor a proper topic for their empirical inquires. The attitude was part of our legacy from Thomas Hobbes’s contention that violence is the negation of political order, a subject fit less for study than for admonition. Moreover, neither the legalistic nor the institutional approaches that dominated traditional political science could provide much insight into group action that was regarded by definition as illegal and the antithesis of institutionalized political life. Representative of these attitudes was Arnold Forster’s judgment in 1966 that political violence “by its very nature [is] beyond any simple or reasonable laws of causation” (1966, 142). Lawrence Stone, a prominent historian, wrote an essay on theories of revolution in which he made the curious argument that “collective violence generally cannot be the object of useful theorizing because it is at the same time both pervasive and somewhat peripheral” (Stone 1966).
Revolutions have traditionally been regarded as the most significant form of political violence, because the universe of such events has been defined by reference to their consequences rather than their common characteristics or preconditions, and because the older theoretical generalizations have emphasized primarily the processes of such events and categorization of their concomitants at a low level of generality (see Edwards 1927; Brinton 1938; Stone 1966). But the evidence both of recent history and of systematic attempts at specifying the evidence of political violence suggests that revolutions are but one of an extraordinarily numerous variety of interrelated forms of strife; that some of these forms, among them coup d’état, guerrilla war, and massive rioting, can alter political processes and social institutions as drastically as any of the classic revolutions; and that the forms themselves are mutable, or rather, that by reifying our arbitrary distinctions among forms of strife we have overlooked some fundamental similarities. The “French Revolution” was a series of events that would now be characterized as urban demonstrations and riots, peasant uprisings, and a coup d’état. It is called a revolution in retrospect and by virtue of the Duc de Liancourt’s classic remark to Louis XVI. The American Revolution began with a series of increasingly violent urban riots and small-scale terrorism that grew into a protracted guerrilla war. Examination of those special conditions and processes that lead from turmoil to revolution provides a partial understanding of revolution per se, but for a sufficient explanation we require a more general theory, one capable of accounting for the common elements of that much larger class of events, civil strife.2
The resort to illicit violence is the defining property that distinguishes these collective events from others. We can regard this as just a definitional point, but it has a crucial theoretical consequence: to direct attention to psychological theories about the sources of human aggression.
Some types of psychological theories about the sources of aggressive behavior can be eliminated at the outset. There is little value in pseudo-psychological speculation about revolutionaries as deviants, fools, or the maladjusted. Psychodynamic explanations of the “revolutionary personality” may be useful for microanalysis of particular events but scarcely for general theory. Aggression-prone victims of maladaptive socialization processes are found in every society, and among the actors in most outbreaks in political violence, but they are much more likely to be mobilized by strife than to be wholly responsible for its occurrence. Nor can a general theory of civil strife rest on culturally specific theories of modal personality traits, though it might well take account of the effects of these traits. Some cultures and subcultures produce significantly more aggression-prone than cooperative personalities, but an explanation of this order says little of the societal conditions that elicit aggression from the aggression-prone, and nothing at all of the capacity for political violence of even the most apparently quiescent populations.
The only generally relevant psychological theories are those that deal with the sources of characteristics of aggression in all people, regardless of culture. Such psychological theories do not directly constitute a theory of civil strife. They do offer alternative motivational bases for such a theory and provide means for identifying and specifying the operation of some crucial explanatory variables. As is demonstrated in the following section, one or another of these theories is implicit in most theoretical approaches to civil strife that have no explicit motivational base although one of them appears highly plausible in the light of empirical evidence.
Psychological theories of aggression
There are three distinct psychological assumptions about the generic sources of human aggression: that aggression is solely instinctual, that it is solely learned, or that it is an innate response activated by frustration. The instinct theories of aggression, represented, among others, by Freud’s attribution of the impulse to destructiveness to a death instinct and by Lorenz’s view of aggression as a survival-enhancing instinct, assume that most or all people have within them an autonomous source of aggressive impulses, a drive to aggress that, in Lorenz’s words, exhibits “irresistible outbreaks which recur with rhythmical regularity” (Lorenz 1966, xii). Although there is no definitive support for this assumption, and much evidence to the contrary, its advocates, including Freud and Lorenz, have often applied it to the explanation of collective as well as individual aggression (Freud 1930; Lorenz 1966, chs. 13, 14; also see Berkowitz 1962, ch. 1). The assumption is evident in Hobbes’s characterization of man in the state of nature and is perhaps implicit in Nieburg’s concern for “the people’s capability for outraged, uncontrolled, bitter, and bloody violence” (1962, 870), but plays no significant role in contemporary theories of civil strife.
Just the opposite assumption, that aggressive behavior is solely or primarily learned, characterizes the work of some child and social psychologists, whose evidence indicates that some aggressive behaviors are learned and used strategically in the service of particular goals—aggression by children and adolescents to secure attention, by adults to express dominance strivings, by groups in competition for scarce values, by military personnel in the service of national policy (see Bandura and Walters 1963). The assumption that violence is a learned response, rationalistically chosen and dispassionately employed, is common to a number of recent theoretical approaches to civil strife. Johnson repeatedly, though not consistently, speaks of political violence as “purposive,” as “forms of behavior intended to disorient the behavior of others, thereby bringing about the demise of a hated social system” (Johnson 1966, 12, 13). Parsons attempts to fit political violence into the framework of social interaction theory, treating the resort to force as a way of acting chosen by the actor(s) for purposes of deterrence, punishment, or symbolic demonstration of their capacity to act (Parsons 1964, 34–35). Schelling is representative of the conflict theorists: he explicitly assumes rational behavior and interdependence of the adversaries’ decisions in all types of conflict (1960, 4). Stone criticizes any emphasis on violence as a distinguishing or definitional property of civil strife on grounds that it is only a particular means, designed to serve political ends (Stone 1966, 15).
The third psychological assumption is that it occurs primarily as a response to frustration. A “frustration” is an interference with goal-directed behavior; “aggression” is behavior designed to injure, physically or otherwise, those toward whom it is directed. The disposition to respond aggressively when frustrated is considered part of our biological makeup; there is an innate tendency to attack the frustrating agent. Learning can and does modify the tendency: what is perceived to be frustrating, modes of aggressive response, inhibition through fear of retaliation, and appropriate targets are all modified or defined in the learning process, typically but not solely during socialization.
Frustration-aggression theory is more systematically developed, and has had more empirical support, than theories that assume either all men have a free-flowing source of destructive energy or that all aggression is imitative and instrumental. Moreover, the kinds of evidence citied in support of theories of the latter type appear to be subsumable by frustration-aggression theory, whereas the converse is not the case.
One crucial element that frustration-aggression theory contributes to the study of political violence concerns the drive properties of anger. In a reformulation of the theory by Berkowitz, the perception of frustration is said to arouse anger, which functions as a drive. Aggressive responses tend not to occur unless evoked by some external cue, but their occurrence is an inherently satisfying response to that anger (Dollard et al. 1939; Yates 1962; Berkowitz 1962). Similarly, Maier has amassed extensive evidence that the innate frustration-induced behaviors (including regression, fixation, and resignation, as well as aggression) are for the actor ends in themselves, unrelated to further goals and qualitatively different from goal-directed behavior (Maier 1949).
To argue that aggression is innately satisfying is not incompatible with the presence of learned or purposive components in the acts of individual or collective aggression. Cues that determine the timing, forms, and objects of aggression are learned, just as habits of responding aggressively to moderate as well as severe frustration can be learned. The sense of frustration may result from quite rational analysis of the social universe. Leaders can put their followers’ anger to rational or rationalized uses. If anger is sufficiently powerful and persistent it may function as an autonomous drive, leading to highly rational and effective efforts by both leaders and the led to satisfy anger aggressively. The crucial point is that rationalization and organization of illicit violence are typically subsequent to, and contingent upon, the existence of frustration-induced anger. Collective violence may be a calculated strategy of dispassionate elite aspirants, and expectations of gains to be achieved through violence may be present among many of its participants. Nonetheless the implication of frustration-aggression theory is that political violence almost always has a strong “appetitive,” emotional base and that the magnitude of its effects on the social system is subsequently dependent on how widespread and intense anger is among those it mobilizes.
If anger implies the presence of frustration, there is compelling evidence that frustration is all but universally characteristic of participants in civil strife: discontent, anger, rage, hate, and their synonyms are repeatedly mentioned in studies of strife. Moreover, the frustration assumption is implicit or explicit in many theoretical analyses of the subject. Ridker characterizes the consequence of failure to attain economic expectations as “discontent,” analogous in source and consequence to anger (Ridker 1962). In Davies’ theory of revolution, the reversal of a trend of socioeconomic development is said to create frustration, which instigates revolution (Davies 1962). Galtung’s theory of both intranational and international aggression recognizes that “the external conditions leading to aggression … probably have to pass through the minds of men and precipitate as perceptions with a high emotive content before they are acted out as aggression” (Galtung 1964, 95).
In few approaches to the theory, however, has frustration-aggression theory been systematically exploited nor have its variables been taken into account (an exception is Feierabend and Feierabend 1966). The primary object of this essay is to demonstrate that many of the variables and relationships identified in social psychological research on the frustrationaggression relationship appear to underlie the phenomenology of political violence. Juxtaposition of these two diverse types of material provides a basis for an interrelated set of propositions that is intended to constitute the framework of a general theory of the conditions that determine the likelihood and magnitude of political violence. These propositions are of two types, whose proposed relationships are diagrammed in Figure 1.1: First are propositions about the operation of instigating variables, which determine the magnitude of anger; second are propositions about the mediating variables, which determine the likelihood and magnitude of overt violence as a response to anger.
This approach does not deny the relevance of aspects of the social structure, which many conflict theorists have held to be crucial. The supposition is that theory about political violence is most fruitfully based on systematic knowledge about those properties of people that determine how they react to certain characteristics of their societies.
My basic premise is that the necessary precondition for violent conflict is relative deprivation, defined as actors’ perception of discrepancy between their value expectations and their environment’s apparent value capabilities.3 Value expectations are the goods and conditions of life to which people believe they are justifiably entitled. The referents of value capabilities are to be found largely in the social and physical environment: they are conditions that determine people’s perceived chances of getting or keeping the values they legitimately expect to attain. In a comparable treatment, Aberle (1962) defines relative deprivation as “a negative discrepancy between legitimate expectation and actuality,” viewing expectations as standards, not mere prophecies or hopes. For purposes of general theoretical specification I assume that perceived discrepancies between expectations and capabilities with respect to any collectively sought value—economic, psychosocial, political—constitute relative deprivation. The extent to which some values may be more salient than others is a subject of theoretical and empirical inquiry not evaluated here.
FIGURE 1.1 ...
