Using quarterly data from 1968 to 1988, we analyze the time series properties of the various attack modes used by transnational terrorists. Combining vector autoregression and intervention analysis, we find strong evidence of both substitutes and complements among the attack modes. We also evaluate the effectiveness of six policies designed to thwart terrorism. The existence of complements and substitutes means that policies designed to reduce one type of attack may affect other attack modes. For example, the installation of metal detectors in airports reduced skyjackings and diplomatic incidents but increased other kinds of hostage attacks (barricade missions, kidnappings) and assassinations. In the long run, embassy fortification decreased barricade missions but increased assassinations. The Reagan “get tough” policy, which resulted in the enactment of two laws in 1984 and a retaliatory raid on Libya in 1986, did not have any noticeable long-term effect on curbing terrorist attacks directed against U.S. interests.

When scholars extend their models and hypotheses to encompass additional cases, they commonly need to adapt their analytic categories to fit the new contexts. Giovanni Sartori's work on conceptual “traveling” and conceptual “stretching” provides helpful guidance in addressing this fundamental task of comparative analysis. Yet Sartori's framework draws upon what may be called classical categorization, which views the relation among categories in terms of a taxonomic hierarchy, with each category having clear boundaries and defining properties shared by all members. We examine the challenge to this framework presented by two types of nonclassical categories: family resemblances and radial categories. With such categories, the overly strict application of a classical framework can lead to abandoning to category prematurely or to modifying it inappropriately. We discuss solutions to these problems, using examples of how scholars have adapted their categories in comparative research on democracy and authoritarianism.

A simple model is used to compare, under different electoral systems, the incentives for candidates to create inequalities among otherwise homogeneous voters, by making campaign promises that favor small groups, rather than appealing equally to all voters. In this game model, each candidate generates offers for voters independently out of a distribution that is chosen by the candidate, subject only to the constraints that offers must be nonnegative and have mean 1. Symmetric equilibria with sincere voting are analyzed for two-candidate elections and for multicandidate elections under rank-scoring rules, approval voting, and single transferable vote. Voting rules that can guarantee representation for minorities in multiseat elections generate, in this model, the most severely unequal campaign promises.

Norwegian party politics is characterized by coalition avoidance that defies conventional coalition theory. This failure of coalescence can be caused either by policy pursuit (preference-induced) or by institutional constraint (structure-induced). We test the explanatory power of policy-based and institutional explanations, relying on content analysis of authoritative party and government documents for our policy measures. The results show that the left–right policy dimension has powerfully constrained Norwegian interparty bargaining and that policy-based coalition theory can account for many apparent anomalies in Norwegian coalition politics. A permissive institutional environment has also fostered coalition avoidance. Although core-based coalition theory can thus be successfully adapted to the Norwegian case, it rests on a number of critical assumptions that limit its general applicability.

First, I will demonstrate that Machiavelli's conception of political phenomena is richer and more varied and conforms to an adequate understanding of political reality more closely than later, more “systematic,” or “regular,” conceptions often associated with the Enlightenment. Second, I find in Machiavelli strong grounds for resisting authoritarian claims that the variegated and unpredictable nature of political phenomena must be managed with unipartite and, especially, centralized practical political alternatives. I focus upon Machiavelli's frequent use of a particular word by which he refers to politically significant occurrences, accidente. I argue that his employment of this word, especially in the Discourses, serves to accentuate his sensitivity to the irregular, nonsystematic nature of political reality—to the possibility of “exceptions.” But I demonstrate that his practical response to this reality, is not a perpetually vigilant, all-powerful sovereign but is, rather, a far more moderate answer, the mixed regime.

On the philosophical plane, James Mill's political thinking began from a model of man quintessentially utilitarian in constitution. Starting with individual agents, it was to his account of the science of human nature that he turned in the quest for a science of politics suitable for the modern world. If James Mill's science of politics was individualist in character, it was neither automatically nor necessarily democratic in the practical political arena. On that subject, everything turned on the question of judging when (or if) individual capacity had reached an acceptable standard. This criterion proved to be sufficiently malleable to allow him to appear either expansive and democratic or narrow and elitist, as the case required.

The study of legal politics often attends to the description and explanation of the instrumental politics of legal change or to the mobilization of rights by individual citizens as an act of empowerment beneficial to their individual liberty. Drawing far less attention from the discipline is the privileged construction of a discourse or set of attitudes about rights by the judiciary. I present a case study of the First Amendment opinions of two members of the Supreme Court of the United States to criticize the range of their attitudes about rights and to illustrate how their opinions help construct and legitimate the disciplinary actions that provide order in the modern liberal regime. To preserve order, the justices are shown to use the language of rights as an instrument for the facilitation of violence, repression, and subjection against some litigants, rather than as an instrument for the enhancement of expressive liberties.

Most previous research on congressional career decisions has focused on one of two binary choices—between retiring and running for reelection, or between running for reelection and seeking higher office. But most of the time, representatives face all three choices simultaneously. Employing a “mother logit” model, we estimate the effects of relevant variables both on pairwise comparisons (conditional probabilities) and on the unconditional probabilities of choosing each one of these three alternatives. Probably most intriguing is our finding that a member's age has little or no effect upon the unconditional probability of running for reelection. The interrelatedness of career options is seen particularly clearly in the case of incumbents who had been redistricted out of their seats. When they had an opportunity to run for higher office, they were likely to take it. Only when they lacked such an opportunity were they more likely than other members to opt for retirement.

Alvarez, Garrett and Lange (1991) used cross-national panel data on the Organization for Economic Coordination and Development nations to show that countries with left governments and encompassing labor movements enjoyed superior economic performance. Here we show that the standard errors reported in that article are incorrect. Reestimation of the model using ordinary least squares and robust standard errors upholds the major finding of Alvarez, Garrett and Lange, regarding the political and institutional causes of economic growth but leaves the findings for unemployment and inflation open to question. We show that the model used by Alvarez, Garrett and Lange, feasible generalized least squares, cannot produce standard errors when the number of countries analyzed exceeds the length of the time period under analysis. Also, we argue that ordinary least squares with robust standard errors is superior to feasible generalized least squares for typical cross-national panel studies.

In their research note Robert Bates and Amy Curry (1992) have drawn our attention to the importance for political science of the challenges posed by the moral economic approach. Using their essay as a springboard and recasting their understanding of the idea of the moral economy, I show that the full significance of this controversy would be better illuminated by moving the debate on to a different terrain, one that captures the fundamental sources of the antagonism between the moral economic and competing approaches. I begin by setting out the central difficulty in Bates and Curry's reading of the moral economy argument and relate that to how the controversy has been framed in the recent political science literature. I then offer a sketch of the core claims of the moral economic approach, drawing out the radicalness of their challenge, and offer an estimate of their weaknesses and value. I conclude with some observations about political philosophy and the idea of the moral economy.

Electoral competition is a concept that has played a central role in much of the state politics literature. One commonly used indicator of competition in the states is the Ranney index. We offer an alternative indicator of competition, one based on district-level outcomes of state legislative elections. After evaluating both indicators in terms of validity and reliability, the analysis suggests that the district-level indicator is both empirically and intuitively superior as a measure of electoral competition. The implications of this finding are discussed.

When the United States began its overt military conflict with Iraq in January 1991, the news media focused unceasingly on the Gulf crisis. Using national survey data, we show that this emphasis altered the ingredients of Americans' assessments of George Bush's performance. After the war, assessments were based more on beliefs about Bush's effectiveness in managing the conflict and less on confidence in his handling of other foreign relations matters or the domestic economy. Consequently, Bush's overall performance ratings increased dramatically following the war. We also show that the media's impact on political judgments was regulated by citizens' levels of political knowledge, exposure to political news, and interest in the war. Greater impact was associated with higher levels of knowledge and lower levels of exposure and interest. These findings challenge traditional views of these dimensions of political involvement and support a view derived from contemporary psychological theories of information processing.

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.