The Behavioral Origins of War / Edition 1

Paperback (Print)
Buy Used
Buy Used from
(Save 28%)
Item is in good condition but packaging may have signs of shelf wear/aging or torn packaging.
Condition: Used – Good details
Used and New from Other Sellers
Used and New from Other Sellers
from $4.41
Usually ships in 1-2 business days
(Save 85%)
Other sellers (Paperback)
  • All (12) from $4.41   
  • New (2) from $63.97   
  • Used (10) from $4.41   


In The Behavioral Origins of War, D. Scott Bennett and Allan C. Stam analyze systemic, binary, and individual factors in order to evaluate a wide variety of theories about the origins of war.
Challenging the view that theories of war are nothing more than competing explanations for observed behavior, this expansive study incorporates variables from multiple theories and thus accounts for war's multiplicity of causes. While individual theories offer partial explanations for international conflict, only a valid set of theories can provide a complete explanation. Bennett and Stam's unconventional yet methodical approach opens the way for cumulative scientific progress in international relations.
D. Scott Bennett is Professor of Political Science at the Pennsylvania State University. Allan C. Stam is Associate Professor in the Government Department at Dartmouth College.
Read More Show Less

Editorial Reviews

From the Publisher

". . . an extraordinary book that takes the study of the causes of war to new levels of analytical sophistication and scientific advancement. Other scholars have surveyed and synthesized voluminous empirical studies of war. . . . But Bennett and Stam created a special testing methodology and publicly accessible software, and 'scientifically evaluated the relative explanatory power' of 16 theories of war at the state, dyad, and international system levels of analysis. . . . Underscoring the complex paths to conflict onset and escalation, their model predicts both system-level war frequency and 'dangerous dyads' exhibiting specific patterns including geographical contiguity, capability imbalance, arms races, autocracy, and recent history of conflict. A must for college university libraries and relevant specialized collections."

Read More Show Less

Product Details

  • ISBN-13: 9780472068449
  • Publisher: University of Michigan Press
  • Publication date: 12/11/2003
  • Edition description: New Edition
  • Edition number: 1
  • Pages: 280
  • Product dimensions: 5.90 (w) x 8.90 (h) x 0.70 (d)

Read an Excerpt

The Behavioral Origins of War

By Allan C. Stam

University of Michigan Press

Copyright © 2003 Allan C. Stam
All right reserved.

ISBN: 9780472098446


I pass with relief from the tossing sea of cause and theory to the arm ground of result and fact.

—Winston Churchill, 1898
Social science, like any science, achieves progress through the accumulation of systematic knowledge. To improve our collective understanding of international politics we need to regularly assess both the empirical regularities observed in the world around us and the theories that profess to explain these facts as we have come to know them. Political science in general and the study of international politics in particular suffer from having numerous competing theories, assertions, and conjectures describing the same phenomena. Although new explanations and descriptions of interstate war appear almost as frequently as the events themselves, new ideas rarely supersede those previously developed. Instead, they simply accumulate with little regard paid to the explanatory power of new accounts versus those previously advanced. This has resulted in the multitude of models, untested hypotheses, conjectures, and normatively grounded assertions that constitute the discipline.

The study of international conbict exempliaes these problems.Over the last forty years, the development of models and conjectures purporting to explain the incidence and escalation of international conbict has proceeded at a rapid pace. In their 1990survey, for instance, Dougherty and Pfaltzgraff identiaed over thirty examples of what they characterize as theories of international relations. If we counted variations on familiar themes and arguments developed since, the number would be much higher. Having an array of theoretical approaches and empirical conjectures is not a problem if scholars have a well-founded sense of which explanations or descriptions account for the greatest proportion of events or facts or, more colloquially, which theories or hypotheses work best. While political scientists have been very successful in developing numerous interesting stories about the nature of international conbict, we have been much less successful at conducting the type of rigorous analysis that would allow us to (1) evaluate the relative explanatory power of these various descriptions and (2) reach some sense of consensus about which stories are the most useful or valuable in understanding international political processes. The abandonment of "failed" theories or presumed empirical regularities is rare in political science—and particularly so in the study of international politics.

Apologists for the current state of affairs might claim that, compared to other disciplines, cumulative progress in political science may be particularly difacult to achieve for at least two principle reasons. One has to do with the limited availability of some critical data. For some problems in political science, there simply is insufacient data for researchers to test their hypotheses or arguments. For example, the complete absence of wars between modern liberal democracies after World War II makes it difacult to sort out the competing and often collinear explanations for this important fact. Similarly, tests of explanations focusing on human misperception or the role of private information require data that are notoriously difficult to assemble. Another problem we face, and one that is potentially damning, lies in the possibility that in the realm of politics there may be no fundamental regularities, or equilibria, to predict. William Riker (1980, 443) suggests that the prospects for theoretical advancement in political science are quite bleak by concluding that "politics is the dismal science because we have learned from it that there are no fundamental equilibria to predict." From this hypothesized lack of formal equilibria, Riker then concludes that "In the absence of such equilibria we cannot know much about the future at all, whether it is likely to be palatable or unpalatable, and in that sense our future is subject to the tricks and accidents of the way in which questions are posed and alternatives are offered and eliminated."1Of course, Riker's gloomy conclusion has not prevented others from (1) vigorously disagreeing with him, (2) continuing to develop new explanations of the political world, and (3) claiming the discovery of new empirical regularities of normative or substantive significance.

These protestations notwithstanding, it is probably premature to conclude that there are no predictable international political events or that there are insufficient data to assess the validity of carefully speciaed arguments. Nevertheless, we believe that there is a third critical problem hindering scientiac advancement in the study of international conflict behavior, namely, the paucity of systematic comparative testing that would allow us to examine, judge, and occasionally abandon explanations and descriptions of the origins of international disputes and wars. Without testing multiple explanations of organized international violence in a simultaneous framework, we cannot judge the relative power of competing explanations or predictions. We believe this lack of comparative testing to be a main culprit for our current state of affairs. In this volume, we turn to developing such comparative tests.


When testing an empirical conjecture or the observable implications of one theory versus another (that is, conducting comparative hypothesis testing), one of the arst tasks is to choose from the variety of standards one could use when deciding that one model of the world supersedes or replaces another. For example, one method we might apply is Popper's (1968) "method of elimination." Using his approach, we end up with relatively more powerful theories, as stronger theories replace weaker ones through a process of "dualistic elimination." An example of dualistic elimination—the process of pitting one "theory" against another— appears in the simulation used by Robert Axelrod in The Evolution of Cooperation (1984). There, in round-robin fashion, Axelrod pitted one computer simulation strategy after another against the competition until an overall winner emerged after thousands of iterations. In this case, Popper's method provided a suitable approach because each strategy was an exclusive set of rules to play the game of interest (Prisoners' Dilemma), with each strategy having a clear and unique payoff.

Although simple, Popper's method is not appropriate for testing most theoretical explanations or empirical conjectures in international politics. He notes an important caveat to applying his method of elimination, arguing that it is appropriate only if a theory may be demonstrably falsiaable, that is, if the theory is "sufficiently precise to be capable of clashing with observational experience" (Popper 1968, 131; italics in original). The use of such rigid tests is only appropriate when the theories themselves and the data used to test them are well speciaed and precise. In his work, Popper presents an example of disproving Kepler's description of circular, rather than elliptical, planetary motion. Unfortunately, the level of precision found in most theories and empirical conjectures of international politics is not nearly as high as in the planetary sciences, and the nature of the objects under study is rather different. In fact, most so-called theories of international politics are not really theories at all.

Here, by "theory" we mean a logically consistent and empirically falsiaable causal explanation of why some event or set of events occurred in the past and, given similar conditions, will occur in the future. Most international relations "theories" simply describe conditions where some factor supposedly influenced the likelihood or nature of some past international conbict. In other cases, we see arguments about some factor (such as economic interdependence) that supposedly affects the likelihood of conbict without deaning precisely what sort of conflict is in question (e.g., small disputes, crises, or large-scale wars). Other "theories" such as balance of power theory suffer from logical inconsistencies when moving from cause to effect. Moreover, since many of these causal factors that are linked to increased risk of disputes or war are not mutually exclusive—that is, multiple factors associated with various theoretical perspectives may simultaneously inbuence the conbicts we are trying to understand—dualistic elimination is inappropriate.

In addition, because we view human behavior as probabilistic rather than deterministic, we often will not be able to conclude that one explanation is clearly superior in all circumstances, as no single observation is sufficient to falsify a theory in a probabilistic world. By "probabilistic" we mean that there is a stochastic component to human behavior, that under apparently identical conditions, state leaders might choose to do one thing at one time and something quite different at another, but with some predictable probability of doing each. By contrast, a deterministic relationship would be one such as Newton's theory of force, mass, and acceleration, F MA. In the context of classical mechanics it would be nonsensical to assert that a given force would accelerate a known mass with some acceleration 70 percent of the time. Theories of political behavior are more akin to meteorology, where forecasters hope to predict the probabilities of various weather outcomes occurring, given a set of observed meteorological conditions.

In the political world there may be many factors leading to the probabilistic nature of political behavior, ranging from the tiny inbuences of unmeasured or immeasurable factors to conscious or unconscious strategies of randomized behavior, irrational or impulsive behavior, strategic decision making, and, anally, to the effects of actors making their decisions based on rational expectations in the setting of strategic choice. From this perspective, where chance plays a powerful role in determining the events we ultimately study, we become engaged in the task of forecasting tendencies and inbuences of a set of observable conditions on the relative likelihood that a decision maker will behave in a particular way. In this book, we focus solely on assessing the empirical content of various arguments purporting to predict or to explain the relative likelihood of war.

Taking a slightly different approach to the problem of falsiacation and theory rejection than Popper, we generally follow Lakatos's (1978, 32) standard that we should reject an explanation of past events or a prediction of future events if and only if another explanation predicts everything and more that the arst one does and that this new empirical content is verifiable. We implement what we see as a statistical version of Lakatos's perspective. By adopting the maximum likelihood logic of inference (King 1989) we are able to make two sets of judgments. First, we can assess whether a set of indicator variables derived from an argument about the relative likelihood of war makes novel contributions to the at of our statistical models. Second, we can evaluate the relative predictive power associated with each of these variables. This approach allows us to demonstrate that multiple factors simultaneously inbuence conflict. It also allows us to judge whether or not some conjecture is consistent with more than one unique event when controlling for other explanations, thereby suggesting elimination of this factor as a systematic predictor of war.

Following this perspective, with our analyses we do not attempt to eliminate conjectures based on ontological rigor or the internal logical consistency of the arguments. Rather, we focus solely on their empirical content. A quite different form of comparative analysis might carefully examine the internal logic and assumptions of multiple theories of international conbict and reject those that are logically contradictory or inconsistent (e.g., Zinnes 1967; Niou, Ordeshook, and Rose 1989). While there is much to be said for that approach, all too frequently inductively driven or normatively motivated stories remain at the center of academic and policy debate even after compelling demonstrations of the deductive baws in the "theory's" logic. For example, consider balance of power theory. Niou, Ordeshook, and Rose (1989) conducted an excruciatingly careful and nuanced evaluation of the logical underpinnings of classical balance of power arguments and found them sorely lacking. However, their formal mathematical analysis did little to change the beliefs of those supporting the view that an equitable balance of power or capabilities will keep the peace between potentially warring nations. Balance of power and other realpolitik arguments appear before students in much the same way as they have for decades, unaffected in many classrooms by careful logical analysis demonstrating the logical baws of the "theory." With the hope that unimpeachable empirical regularities may be persuasive where elegant logic has not always been, we focus on the empirical content of our models, hypotheses, and conjectures, represented by their associated variables and operational indicators.


Clearly, evaluating the relative explanatory power of different empirical assertions and dropping or modifying those that receive little or no empirical support is an important part of the scientiac enterprise. Our beliefs about the power of various empirical conjectures drive our collective research agendas and often mark the starting point for policy prescriptions. To date, however, only limited efforts exist to compare systematically the predictive power of the myriad different explanations of international conbict—or, more precisely, the explanatory power of the independent variables expected to correlate with conbict behavior. While many researchers pay lip service to Lakatos and his principles of progressive scientiac research based on careful theory development and testing, in practice most studies of international politics have failed to follow this model. Most so-called theories of international politics are simply broad-brush descriptions based on the observation of small numbers of events rather than carefully deduced explanations of political behavior.

Many of the existing empirical studies fall short in another dimension as well. Studies seeking to compare or cross-validate existing empirical claims all too often use different subsets of data, data cast at different units of analysis, and data sets with different dependent variables. Rather than conducting broad tests of multiple theories, most existing tests of various explanations of international politics assess new explanations against a null model or a small and carefully selected set of competing claims. Typically, an author presents a contrasting set of explanations where one argument is of primary interest with competing explanations presented as control variables. We and, for example, rational deterrence hypotheses compared to variables drawn from psychological approaches (Huth and Russett 1993) or a selected set of international system structure variables compared to a set of variables drawn from a dyadic perspective (Bueno de Mesquita and Lalman 1988). Maoz and Russett launched a veritable cottage industry based on pitting the democratic peace proposition against a variety of control variables (Maoz and Russett 1993; Russett and Oneal 2001). Bueno de Mesquita and Lalman (1992) tested their expected utility predictions along with other predictions drawn from power-based stories but failed to include other non-power-based stories. A few broader studies move beyond a set of closely related alternative explanations to examine several "likely suspects" in the hunt for the correlates of international conbict behavior (Bremer 1992; Huth, Bennett, and Gelpi 1992; Oneal and Russett 1999a). In none of these cases, however, do the authors attempt a comprehensive examination of competing explanations, nor do they systematically assess the strength of the various arguments' predictive power.

In the domain of formal rational choice theory, the lack of comparative testing is particularly noticeable and unfortunate. While the expected utility approach has been the target of both theoretical and empirical criticism, using both normative arguments and empirical case studies (e.g., Jervis, Lebow, and Stein 1985), formal models of international conbict have received remarkably little empirical evaluation with large-n statistical tests, particularly against a wide range of alternative explanations or predictors. This dearth of empirical testing led, in part, to Green and Shapiro's blistering critique of rational choice models and their advocates in Pathologies of Rational Choice Theory (1994, 7). There they and weak empirical support for rational choice theory generally and suggest that, "Despite its enormous and growing prestige in the discipline, rational choice theory has yet to deliver on its promise to advance the empirical study of politics. . . . we believe that this claimcan be defended across the board."

While formal rational choice models are not the only arguments that suffer from a lack of systematic empirical testing, advocates of the rational choice approach make particularly strong claims about the power of expected utility theory. Green and Shapiro also make sweeping and speculative claims about the (lack of) explanatory or predictive power of rational choice theories in all aelds of political science. Ironically, they offer no compelling evidence that alternative approaches might perform any better. Their study also suffers from a notable omission—they do not address the rational choice literature on international politics at all, where there have been a few serious efforts to conduct some rigorous tests (Bueno de Mesquita 1980; Bueno de Mesquita and Lalman 1992; Smith 1996a, 1999; Signorino 2000; Filson and Werner 2001).

There are a few noteworthy attempts to execute a comparative analysis of the empirical literature on international conbict. For example, in Gurr's Handbook of Political Conflict (1980) Bueno de Mesquita presents a review of the theoretical and empirical claims of balance of power arguments, power transition models, system structure conjectures, status inconsistency, arms race models, deterrence, and the externalization of domestic conbict. He rejects some approaches (such as balance of power) because of contradictions in their internal logic. His review, while helpful for understanding the reasoning underlying a variety of common explanations for conbict, does not provide any systematic empirical evidence directly comparing the various arguments' relative explanatory or predictive power. More recently, John Vasquez's The War Puzzle (1994) provides a self-described meta-analysis of the literature on interstate war. Vasquez similarly makes no overall empirical comparison of the various theories or conjectures that he identiaes. Numerous other edited volumes on international conbict exist as well that take a generally similar approach, where each chapter in a volume takes a different tack on the problem of international conbict by focusing on a single argument or approach (e.g., Midlarsky 1992, 2000). While these works help us understand the logic and possible strengths of the various arguments in isolation, they do not provide empirically based comparative hypothesis testing.


Excerpted from The Behavioral Origins of War by Allan C. Stam Copyright © 2003 by Allan C. Stam. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Read More Show Less

Table of Contents

Preface & Acknowledgments
1 Toward a Better Understanding of Theories of International Conflict 1
2 Comparative Hypothesis Testing and Some Limits to Knowledge 15
3 Choosing and Testing the Arguments: The Practice and Pitfalls of Comparative Hypothesis Testing 35
4 Arguments and Operational Measures 70
5 Findings 107
6 Assessing a Model's Reliability across Space and Time 165
7 Conclusion 200
App. A Data Development and Cumulation for International Relations: EUGene 223
App. B Measuring Expected Utility 232
Notes 249
Bibliography 257
Index 277
Read More Show Less

Customer Reviews

Be the first to write a review
( 0 )
Rating Distribution

5 Star


4 Star


3 Star


2 Star


1 Star


Your Rating:

Your Name: Create a Pen Name or

Barnes & Review Rules

Our reader reviews allow you to share your comments on titles you liked, or didn't, with others. By submitting an online review, you are representing to Barnes & that all information contained in your review is original and accurate in all respects, and that the submission of such content by you and the posting of such content by Barnes & does not and will not violate the rights of any third party. Please follow the rules below to help ensure that your review can be posted.

Reviews by Our Customers Under the Age of 13

We highly value and respect everyone's opinion concerning the titles we offer. However, we cannot allow persons under the age of 13 to have accounts at or to post customer reviews. Please see our Terms of Use for more details.

What to exclude from your review:

Please do not write about reviews, commentary, or information posted on the product page. If you see any errors in the information on the product page, please send us an email.

Reviews should not contain any of the following:

  • - HTML tags, profanity, obscenities, vulgarities, or comments that defame anyone
  • - Time-sensitive information such as tour dates, signings, lectures, etc.
  • - Single-word reviews. Other people will read your review to discover why you liked or didn't like the title. Be descriptive.
  • - Comments focusing on the author or that may ruin the ending for others
  • - Phone numbers, addresses, URLs
  • - Pricing and availability information or alternative ordering information
  • - Advertisements or commercial solicitation


  • - By submitting a review, you grant to Barnes & and its sublicensees the royalty-free, perpetual, irrevocable right and license to use the review in accordance with the Barnes & Terms of Use.
  • - Barnes & reserves the right not to post any review -- particularly those that do not follow the terms and conditions of these Rules. Barnes & also reserves the right to remove any review at any time without notice.
  • - See Terms of Use for other conditions and disclaimers.
Search for Products You'd Like to Recommend

Recommend other products that relate to your review. Just search for them below and share!

Create a Pen Name

Your Pen Name is your unique identity on It will appear on the reviews you write and other website activities. Your Pen Name cannot be edited, changed or deleted once submitted.

Your Pen Name can be any combination of alphanumeric characters (plus - and _), and must be at least two characters long.

Continue Anonymously

    If you find inappropriate content, please report it to Barnes & Noble
    Why is this product inappropriate?
    Comments (optional)