D. Scott Bennett

The Duration of Interstate Wars, 1816–1985

We present a model of war duration which incorporates both realpolitik and domestic political variables. We hypothesize that strategy, terrain, capabilities, and government type, among other variables, will play key roles in determining the duration of war. We test these hypotheses using hazard analysis and find empirical support for our key arguments. We find that the realpolitik variables play a greater role than regime behavior and type in determining war duration. We also find that historically, on average, mobilization and strategic surprise have little effect on war duration and that wars are not duration dependent.

Research Design and Estimator Choices in the Analysis of Interstate Dyads

Researchers face three basic questions when testing theoretically driven hypotheses. First is research design: for example, what population should be analyzed, what sample should be drawn from that population, and what cases should be excluded from the sample? What statistical estimator should one use? What set of control variables should one employ? Results drawn from several permutations of the Correlates of War Militarized Interstate Dispute data set are compared using three statistical models. For some theories and variables (international institutionalism—intergovernmental organization and alliance membership; realism—balance of power; expected utility theory—international interaction game equilibria), research design and estimator choices substantially influence the findings. For others (trade and democratic peace theory), the relationship between the dependent and independent variables is relatively impervious to research design and estimator choices.

A Landscape Theory of Aggregation

Aggregation means the organization of elements of a system into patterns that tend to put highly compatible elements together and less compatible elements apart. Landscape theory Predicts how aggregation will lead to alignments among actors (such as nations), whose leaders are myopic in their assessments and incremental in their actions. The predicted configurations are based upon the attempts of actors to minimize their frustration based upon their pairwise Propensities to align with some actors and oppose others. These attempts lead to a local minimum in the energy landscape of the entire system. The theory is supported by the results of two cases: the alignment of seventeen European nations in the Second World War and membership in competing alliances of nine computer companies to set standards for Unix computer operating systems. The theory has potential for application to coalitions of political Parties in parliaments, social networks, social cleavages in democracies and organizational structures.

Testing Alternative Models of Alliance Duration, 1816-1984

Theory: At least four models of alliance duration exist, namely the capabilityaggregation, security-autonomy, domestic politics, and institutionalization models of alliances. The independent tests of these models have been underspecified, however, as no simultaneous comparative testing of the hypotheses from these models has been conducted. Hypothesis: Alliance duration may be a function of security threats, changes in national capabilities, the security-autonomy tradeoffs between larger and smaller states in an alliance, regime type and changes in regime, and the time the alliance has had to become institutionalized. Methods: A hazard model employing time-varying covariates applied to 207 alliances from 1816 to 1984. Results: Key elements of the security-autonomy and domestic politics models have effects on alliance duration as expected. It does not appear, however, that institutionalization over time has an effect on alliance duration, as alliances appear to become more likely to dissolve the longer they last. Additionally, while one hypothesis advanced by the capability-aggregation model receives some support, a second variable is statistically significant in the opposite direction of the models expectation, suggesting possible problems with the models validity.

System Uncertainty, Risk Propensity, and International Conflict among the Great Powers

The authors investigate the relationship between system structure and the initiation of militarized disputes among Great Powers. The central hypotheses concern the interaction between system uncertainty and the risk propensity of national decision makers. The authors employ a research design that enables them to incorporate explanatory variables from various levels of analysis into their theoretical model. The model is tested by probit analysis on a pooled time series of Great Power rival dyads from 1816 to 1975. The empirical results support the central hypothesis that the effects of the international system are mediated by the risk propensity of decision makers. In addition, the authors find that dyadic- and unit-level variables such as arms races, power transitions, and the current and past dispute behavior of rivals also have significant effects on conflict behavior. Finally, the results indicate that nuclear weapons do not seem to have a systematic impact on the initiation of militarized disputes among Great Powers.

Integrating and Testing Models of Rivalry Duration

Theory: Previously separate explanations of rivalry termination based in political shocks, security, and domestic politics can be integrated in a single model of bargaining between rational decision-makers who have multiple goals. Hypotheses from the previously separate models should be tested together. Hypotheses: Security concerns, issue salience, democracy and democratization, regime change, and political shocks affect the probability of rivalry termination, and, in turn, rivalry duration. Duration dependence may also affect rivalry duration. Methods: A parametric hazard model with a Weibull specification and time-varying covariates is estimated on a data set of sixty-three enduring rivalries generated from the latest militarized interstate dispute data. Results: Domestic political factors and issue salience are the most likely causes of rivalry termination. Rivalries also appear strongly and positively duration-dependent, meaning that the hazard rate in rivalries increases over time, and thus that rivalry termination tends to accelerate as rivalries continue. Findings on political shocks and security concerns are sensitive to operationalization. Some causal relationships appear to differ between preand post-WWII rivalries.

Measuring Rivalry Termination, 1816-1992

The predominant operationalizations of enduring rivalries use a period of time without a militarized dispute to identify the termination of such rivalries. The author argues that this misses the true termination date of most rivalries because it does not identify when the underlying disputed issues in the rivalry are resolved. The author suggests an operationalization that identifies rivalry termination dates based on a combination of the absence of militarized disputes and the use of public documents and statements that show issue settlement. Such an operationalization can result in significant changes in rivalry termination dates. The 1996 version of the Correlates of War militarized dispute data set is used to apply this new measure to Goertz and Diehls (1995) set of enduring rivalries to produce new termination dates and show that tests of a model of rivalry termination produce different results when applied to this new data set.

Foreign Policy Substitutability and Internal Economic Problems in Enduring Rivalries

The authors advance a general perspective on how to incorporate the notion of foreign policy substitutability in probabilistic models of international politics. They suggest that substitutability may be dealt with in one of two ways, namely, (1) through better specification or (2) through the use of multinomial logit analysis techniques. Their solution is applied to the substantive problem of how internal economic conditions affect the behavior of states in enduring rivalries. States in rivalries might react to worsening economic problems by either (1) escalating conflict within the rivalry to gain the benefits of a diversionary conflict or (2) seeking to settle the rivalry to free up resources that can be directed toward dealing with economic problems. The possibility that seemingly contradictory policies may be undertaken requires the use of a model for analysis that takes into account substitutability. The authors perform empirical analysis to determine how economic conditions may simultaneously affect the probabilities of dispute initiation and rivalry termination.

A Universal Test of an Expected Utility Theory of War

Bueno de Mesquita and Lalmans version of an expected utility theory of war has become one of the most widely cited theories of international conflict. However, the testing of the theory has lagged behind its theoretical development. In its most sophisticated formulation, the theory has been tested on only 707 dyad-years, all drawn from Europe between 1816 and 1970. We present a test of the expected utility theory of war (as developed in War and Reason ) on the international system from 1816 to 1984. Specifically, we examine the relationship between the main equilibrium variables derived under the “domestic variant” of the international interaction game and behavioral outcomes using multinomial logit. We find that the equilibria correlate with actual behavior in both the set of all dyads and a subset of politically relevant dyads, even after including a set of control variables. The relationship is somewhat less clear among the population of all interstate-directed dyads, however, with key equilibrium variables having smaller effects at increasing the odds of interstate conflict among all dyads. We also present a new software program, EUGene , which generates expected utility data and can serve as an important data management tool for international relations researchers.

Democracy, regime change, and rivalry termination

I examine how certain domestic political conditions affect if and when interstate rivalries end. I argue that rivalries are more likely to end (1) when regime changes occur, (2) when a rivalry involves democratic states, and (3) when rivals are democratizing. I also test the proposition that both states in a rivalry must be democratic in order for the probability of rivalry termination to increase. I assess the argument using a hazard analysis of rivalry duration, using two data sets of rivalries from 1816 to 1988, and using different measures of democracy. I find strong support for the existence of a relationship between polity change and the end of rivalries, and somewhat weaker support for the existence of a relationship between democracy and the end of rivalries. It appears that it is in fact the joint presence of two democracies in a rivalry that helps lead to their termination. However, the process of democratization appears to have little effect, either positive or negative, on rivalry duration.