People who take part in acts of civil disobedience or political violence are discontented about something. That is a truism. The interesting research problem is to identify particular kinds of discontent that are associated systematically with aggressive political behaviour-discontent that operates as a ‘precondition’ of aggressive behaviour in contrast to discontent that is peculiar to particular groups at particular times and places and serves only as a ‘precipitant’ of aggressive behaviour.

While much is known about the effects of the economy on the popularity and electoral fortunes of political leaders, political scientists know very little about how economic decline and political performance influence support for the political regime and the stability of democratic systems. We use three cross-national longitudinal surveys to address this issue: two collected in Costa Rica in the midst of a severe economic crisis in the late 1970s and early 1980s; and one in West Germany during the recession of the mid-1970s. We show that in both countries, overall support for the political regime remained extremely high during the economic decline, while satisfaction with incumbent performance fluctuated much more sharply. Moreover, at the individual level, changes in satisfaction with incumbent performance were only weakly related to changes in regime support. These results provide strong evidence suggesting that if democracies enter economic downturns with initially high levels of regime support, they will be able to withstand even severe, prolonged crises of economic performance.

Causal inference in research testing rational choice models of unconventional political behavior has been hampered by the inability to use perceptions of the costs and benefits of participation at a given time to predict behavior that necessarily occurred in the past and by ambiguities associated with analyzing behavioral intentions instead of actual participation. Using panel data collected on a national sample in West Germany between 1987 and 1989, we show that variables from a “collective interest” model measured in 1987—individuals' dissatisfaction with the provision of collective goods, beliefs that group actions can be successful, and beliefs in the importance of their own participation—predict subsequent participation in collective protest activities. Variables corresponding to the private “selective incentives” associated with protest are found to be less relevant. Furthermore, we find that engaging in protest changes many of the perceptions that influence future participation. We discuss the implications for theories of political mobilization.

A causal model of relationships between structural properties of states, civic culture attitudes of the general public, and change in level of democracy is tested with cross-national data. The model permits inferences about the possibility of unidirectional or reciprocal causation between civic culture attitudes and democracy, controlling for macrosocietal variables such as economic development, income inequality, and subcultural pluralism. Most civic culture attitudes do not have any significant impact on change in democracy. One of them, interpersonal trust, appears clearly to be an effect rather than a cause of democracy. The exception is the percentage of the general public that prefers gradual reform of society instead of revolutionary change or intransigent defense of the status quo. Support for gradual reform has a positive impact on change in democracy, and it is unrelated to a country's years of continuous democracy—findings that support the hypothesis of a unidirectional civic culture effect on democracy.

In their 1987 article in this Review, Muller and Seligson used logged ordinary least-squares (LOLS) to estimate the effect of income inequality on cross-national levels of deaths by political violence. T. Y. Wang challenges the robustness of the main conclusion and argues for the application of a maximum likelihood approach—the exponential Poisson regression (EPR) model—rather than LOLS. He concludes that the widely used LOLS approach yields misleading conclusions when applied to event count data. Dixon, Muller, and Seligson replicate previous work using both LOLS and EPR approaches and conclude that in most—but not all—respects the two approaches yield similar results, supporting the effect of inequality when the specifications are identical. They also argue (in response to concerns expressed by Brockett 1992) that the inequality results are robust when account is taken systematically of the best information on underreporting of deaths.

Discontent theories of rebellion postulate that politicized discontent will have a strong independent effect on individuals' potential to participate in rebellious political action. Expected utility theories postulate that participation in rebellious action is motivated by expectation of reward and that discontent is relevant at most only insofar as individuals expect that collective action can be successful and that their participation is important to that end. We test these theories with data from a national sample and a sample of students at a protest-prone university in Peru, a country with significant objective conditions of discontent and a high incidence of rebellious political conflict. The results provide no evidence for the discontent models but strong support for the expected utility models. The potential for participation in rebellious political action proves to be a function primarily of discontent weighted by the expectancy of the action's success and the perceived importance of personal participation. Private social and normative rewards and costs also are relevant—but to a lesser extent—for the individual's calculation of the expected utility of participation.

We propose two models to explain why individuals participate in collective political action—a personal influence model and a collective rationality model. Each model overcomes the free-rider problem posed by conventional rational choice theory and left unresolved in previous research. The models are tested for legal and illegal protest behaviors, using data from a national sample and two samples of protest-prone communities in the Federal Republic of Germany. The personal influence model is supported for both forms of participation, while the collective rationality model is supported for legal protest. We discuss implications of the results for grievance and rational choice theories of collective political action.

Considerable research effort has been invested in establishing the appropriate relationship between patterns of land distribution and political violence. In an article in the June 1988 issue of the Review, Manus I. Midlarsky proposed and tested a new measure of the distribution of land, which he called “patterned inequality.” He presented supporting evidence with data from Latin American and Middle Eastern countries. In this controversy, Midlarsky's analysis is challenged by Edward N. Mutter, Mitchell A. Seligson, and Hung-der Fu. They advocate an alternative measure of land inequality, test its effect on levels of political violence in Latin America, and find it wanting. In his rejoinder, Midlarsky offers new analytical support for his claims.

Maldistribution of land in agrarian societies is commonly thought to be an important precondition of mass political violence and revolution. Others argue that because of the difficulty of mobilizing rural populations for political protest, land maldistribution is irrelevant except as part of an inegalitarian distribution of income nationwide. These rival inequality hypotheses have significant implications with respect to the kinds of reforms likely to reduce the potential for insurgency in a society. They are tested using the most comprehensive cross-national compilation of data currently available on land inequality, landlessness, and income inequality. Support is found for the argument that attributes the greater causal import to income inequality. Moreover, the effect of income inequality on political violence is found to hold in the context of a causal model that takes into account the repressiveness of the regime, governmental acts of coercion, intensity of separatism, and level of economic development.

Why does it happen that ordinary people can come to participate in rebellious collective action? In the June 1986 issue of this Review, Edward N. Muller and Karl-Dieter Opp argued a public-goods model to account for why rational citizens may become rebels. They offered empirical data drawn from samples in New York City and Hamburg, Germany in support of the public-goods model. George Kolsko takes issue with the rationale of Muller and Opp, arguing that their public-goods model is not a rational-choice explanation of rebellious collective action. In response, Muller and Opp clarify their theory and further elaborate its assumptions.

A basic problem for a rational choice theory of rebellious collective action is to explain why average citizens would participate in such behavior, since they have nothing to gain (they will receive benefits of successful rebellion, in terms of public goods, regardless of whether they take part or not), but much to lose (rebellious behavior may be quite costly). According to the conventional private interest or “by-product” theory, the incentive to participate must come from the expectation of receiving selective benefits; but since average citizens in a general case cannot expect substantial private material rewards, the relevant selective benefits must be psychological in nature. In contrast to the model of private interest theory, a public goods model is proposed, stipulating that the value of rebellion in terms of public goods can be a relevant incentive for participation. Using data from surveys conducted in New York City and Hamburg, West Germany, we investigate the relationship between participation in rebellious political behavior and measures of the incentives of public goods and private interest. The results do not support predictions of the private interest model in regard to nonmaterial selective incentives. Hypotheses of the public goods model are supported.

The incumbent vs. system affect distinction is basic in the conceptualization of political support. It is based on the premise that system affect is a more important antecedent of aggressive political behavior than incumbent affect. The data reported here show that it is possible to distinguish incumbent from system affect empirically, and also theoretically important to make the incumbent-system distinction. Measures especially sensitive to incumbent affect correlate differently with ideology than does a measure especially sensitive to system affect. Byvariate correlations between measures of incumbent affect and a measure of aggressive political behavior are shown to be either spurious or indirect, due to the fact that incumbent affect is correlated with what appears to be a more powerful and direct antecedent of aggressive political behavior, namely, system affect. The theory behind the incumbent-system distinction is expressed in four propositions. In general, the data conform to it, but each prediction is qualified according to whether ideology and community context are inhibitory or facilitative.

Theories of the behavioral consequences of political support coincide in the prediction that political support will correlate positively with indices of conventional behavior, negatively with indices of unconventional behavior. Survey data drawn from three communities in the Federal Republic of Germany show that an index of support for the structure of political authority is negatively correlated both with an index of actual participation in aggressive political behavior and with an index of participation in conventional electoral/pressure-group politics. Since the political behavior indices are themselves positively correlated, it is useful to construct a typology which differentiates between “pure” types – no participation, participation only in conventional, participation only in aggressive – and “mixed” types – participation in conventional and moderately aggressive, participation in conventional and highly aggressive. When the relationship between political support and the political-action type index is examined, it turns out that two of the types are associated with medium political support, while four of them occur at low support. To achieve more accurate explanation of types of political behavior, a model for prediction of each action type is proposed, taking into account interaction between political support, sense of personal political influence, and belief in the efficacy of past collective political aggression. The test of the model yields positive results, suggesting that it represents a fruitful beginning toward development of a theory of behavioral consequences of political support.

Perception of discrepancy between optimum level of achievement with respect to desired values and actual level of achievement is a concept that has figured importantly in explanations of collective violence and its subset, political violence (approval of and readiness to engage in behaviors which constitute progressively greater challenge to a political regime). Hypotheses about relationships between a number of static and dynamic achievement discrepancy constructs (labeled “relative gratification,” and built from a variant of the Cantril Self-Anchoring scale) are tested. The achievement discrepancy constructs generally show only a weak degree of association with potential for political violence. However, measures of shift over time in discrepancy show an unexpected and intriguing relationship with potential for political violence: individuals who perceive negative change and individuals who perceive positive change show the highest potential for political violence, while individuals who perceive no change show the lowest potential for political violence; and this V-Curve relationship persists in the presence of various control variables. Moreover, absolute magnitude of shift in discrepancy from present to future shows a moderate degree of correlation with potential for political violence, and makes an independent contribution to a linear additive model. The data base is a sample of a population in which instances of political violence have been relatively frequent in the past.

Potential for political violence is defined by a summated scale built from two cumulative scales measuring approval of and readiness to engage in behaviors which constitute progressively greater challenge to a political regime. A prevalent explanation of potential for political violence proposes that the major direct antecedent is relative deprivation. The partial theory proposed here does not assign a central role to relative deprivation; rather, diffuse support for the political authority structure, and belief that political violence has led to goal attainment in the past, are proposed as major direct antecedents. Relative deprivation is denned by position on the Cantril Self-Anchoring scale with respect to three welfare values. The measure of diffuse affect for the political authority structure is a weighted summation of items measuring the degree to which political authorities are believed to wield power honestly, benevolently, and justly. Belief that political violence has led to goal attainment is denned as a summation of items measuring the degree to which the use of political violence by dissident groups is thought to have been helpful. The data base is a sample of a population in which instances of political violence have been frequent in the past. A linear additive model of Potential for Political Violence, with Trust in Political Authorities and Efficacy of Past Violence as describing variables, shows an accuracy of prediction which is satisfactory and superior to that yielded by a multiplicative model.

Representation is a matter of linkage. In this paper it is argued that a useful handle can be gained on the problem of identifying and measuring representational relationships empirically by viewing representation as a type of support linkage between members of political systems and the authorities.

To conceive representation as a type of support linkage is to direct attention primarily to the represented rather than the representative. Representational relationships have functional significance for political systems particularly because they are linkages which involve members' satisfaction-dissatisfaction with the behavior of the political authorities—linkages which reflect the degree to which members feel that the performance of the authorities “stands for” or “re-presents” their own interests; and this performance satisfaction-dissatisfaction presumably makes a contribution to more general support for the political system. In contrast to legitimacy sentiments, which are independent of immediate outputs from political authorities, members' perceptions of representational linkages between themselves and the authorities depend on their affective responses to outputs, encompassing not only instrumental performance satisfactions, but (and most commonly among the membership in general) symbolic performance satisfactions as well.

This paper reports an exploratory investigation of a construct for measuring sense of representation. The sample consists of a group of students enrolled at the University of Iowa—including, in order to ensure adequate variation on the support scales, a number of students arrested for participating in a protest demonstration against the presence of Marine recruiters on the University of Iowa campus.

Political efficacy, the belief that the ruled in a political system have some capacity for exercising influence over the rulers, has been studied extensively by political researchers. A selected bibliography compiled by Easton and Dennis in early 1967 contains some thirty books and articles which have dealt in one way or another with political efficacy and its correlates. And this bibliography could be updated considerably.

Substantial theoretic import has been attributed to political efficacy. Easton and Dennis consider the SRC sense of political efficacy construct to be an important determinant of the persistence of democratic regimes. They argue that beliefs in political efficacy provide “a reservoir of diffuse support upon which the system can automatically draw in normal times, when members may feel that their capacity to manipulate the environment is not living up to their expectations, and in special periods of stress, when popular participation may appear to be pure illusion or when political outputs fail to measure up to insistent demands,” A related construct, termed “subjective competence” by Almond and Verba, is based on different indicators but interpreted as substantively equivalent to the SRC construct. On the basis of their analysis of the Five-Nation data, Almond and Verba arrive at the general conclusion that “the self-confident [subjectively competent] citizen appears to be the democratic citizen.” The concept of political competence, as formulated by Barnes, subsumes political efficacy under the aegis of an individual attribute consisting of “political skills plus the sense of efficacy necessary for effective political action.” Barnes contends that high levels of political competence dispose individuals to prefer democratic styles of leadership, while low levels dispose individuals to prefer authoritarian styles. On these grounds, he concludes that relatively high levels of political competence are a necessary condition of political democracy.