No abstract

Contradictory findings, that economic inequality may have a positive, negative, or no impact on political conflict, are a puzzle for conflict studies. Three approaches have been used t o explain the inconsistent findings of the EI-PC (Economic Inequality-Political Conflict) nexus: statistical modeling, formal modeling, and theory building. Because analysts have tended to possess different research skills, these three approaches have been employed in isolation from one another. Singly, however, all three approaches have proved deficient and are unlikely to solve the EI-PC puzzle. The most fruitful approach is to combine the assumptions of the theory builders and the deductive approach of the formal modelers with the various empirical tests of the statistical modelers. Such an approach to the EI-PC puzzle produces a crucial test of the Deprived Actor and Rational Actor theories of conflict. The approach is also our best hope for solving the other long-standing puzzles in conflict studies.

Aggregate data studies of domestic political conflict have used an Action-Reaction (AR) model that has produced contradictory findings about the repression/dissent nexus: Repression by regimes may either increase or decrease dissent by opposition groups. To clarify these findings I propose an alternative Rational Actor (RA) model from which are derived three propositions. (1) An increase in a government's repression of nonviolence will reduce the nonviolent activities of an opposition group but increase its violent activities. (2) The balance of effects, that is, whether an increase in the regime's repression increases or decreases the opposition group's total dissident activities, depends upon the government's accommodative policy to the group. (3) Consistent government accommodative and repressive policies reduce dissent; inconsistent policies increase dissent. The RA model thus accounts for the contradictory findings produced by the AR-based aggregate data studies of repression and dissent.

Peasant upheavals are studied from the perspective offered by the selective incentives solution to Olson's collective action problem. This article presents much evidence from three different forms of peasant struggles—everyday forms of peasant resistance, unorganized rural movements, and organized peasant rebellions—that demonstrates the widespread existence of selective incentives. Questions about the causes and consequences of selective incentives are then examined. First, what are the conditions under which peasant struggles emphasize material selective incentives rather than nonmaterial altruistic appeals? The level of selective incentives in any peasant upheaval is a function of demand and supply considerations. Peasants demand selective incentives. The suppliers include one or more dissident peasant organizations, the authorities, and the allies of both. A political struggle ensues as the suppliers compete and attempt to monopolize the market. Second, what are the conditions under which the pursuit of material self-interest hurts rather than helps the peasantry's collective cause? Selective incentives supplemented by ideology can be effective; selective incentives alone are counterproductive.These questions and answers lead to the conclusion that the selective incentives solution reveals much more about peasant upheavals than simply that peasants will often be concerned with their own material self-interest. It is therefore important to study the following three aspects of peasant collective action: the dilemma peasants face, or how peasant resistance is in the interest of all peasants but in the self-interest of none; the paradox peasants face, or that rational peasants do solve their dilemma (for example, with selective incentives) and participate in collective action; and the irony peasants face, or that self-interest is both at the root of their dilemma and at the foundation of a solution to their paradox.

This article proposes and tests a self-generative theory of conflict processes within nations. We dissect the "conflict breeds conflict" truism into three hypotheses: (1) the present extent of conflict simultaneously determines its intensity, while the present intensity of conflict determines its future extent; (2) the present extent of protest determines the present extent of rebellion and vice versa; and (3) the extent and intensity of both protest and rebellion persist over time. Our principal findings are: (1) man-days of protest is a weak positive and linear function of simultaneous man-days of rebellion and lagged mandays of protest; (2) deaths from protest is a strong curvilinear function of simultaneous man-days of protest; (3) man-days of rebellion is a weak positive and linear function of simultaneous man-days of protest and lagged man-days of rebellion, and a U-shaped function of lagged deaths from rebellion; (4) deaths from rebellion is a strong exponential function of present man-days of rebellion, and a linear and positive function of lagged deaths from protest and from rebellion. We conclude that the self-generative model provides a less-than-sufficient explanation of variations in internal conflict.The purpose of this article is to specify and test with cross-national data the common-sense argument that conflict breeds conflict. The underlying assumptions are that societies and social groups have characteristic patterns of conflict behavior which exist in complex