“Liberalism and World Politics” had an impact because it is short and (by my standards) well written. I am sure its brevity has attracted anthologists. The more substantial reasons for its impact seem to me to be the following five:

First, the article addressed an important anomaly. Traditional theories of democratic foreign policy said that democracies would inherently be peaceful. The mainstream of the literature on war and peace refuted that claim. A group of carefully done empirical studies suggested that democracies were peaceful, but only with one another. This article resolved the anomaly by outlining the logic of how states that were both liberal and representative would maintain peace with one another, but not reliably with nonliberal, nonrepresentative states. It argued that although states throughout history have gone to war with one another to advance both their rational national and nonnational governmental interests, liberal democratic peoples have, by and large, respected one another's sovereignty and security.

This argument was inspired by Kant's “Perpetual Peace,” (Kant 1970) which held that liberal representative republics do not wage war against one another (though they sometimes do go to war against nonrepublican peoples). The reasoning was that liberal republics are responsible to the majority of electors, who, unlike monarchs and dictators, cannot regularly displace the costs of going to war on others. If, moreover, those electors respect liberal principles, then they will respect the rights of peoples similarly free to express their rights, and will negotiate rather than fight over differences of interest. Respect for rights of property and the benefits of commercial exchange will reinforce these moral commitments.

But the same respect and self-interest that generates a self-enforcing peace among liberal republics produces distrust of states that oppress their subjects and compound conflicts of interest between liberal republics and nonliberal autocracies that neglect property rights and restrict trade and other contacts. This is what makes for a peace among liberals, but not between liberal and nonliberal states.

Second, it was interdisciplinary. It combined the international relations of war and peace with political and social philosophy. It drew on Machiavelli, Schumpeter, and Kant to distinguish imperial,pacific, and internationalist variants of popular republicanism.¹

Third, it bridged methodological divides. It reflected my long fascination with Carl Hempel's (1942) “The Function of General Laws in History” and yet it gave a central place to interpreted, inter-subjective ideas and norms as causal factors. It lent itself to both quantitative and qualitative assessment. It argued that significant insights could be produced by the combination of institutions (representative republics), ideas (human rights), and interests (commerce, investment) that did not produce significant outcomes in isolation.

Fourth, it provided a framework for thinking about politics that linked to real world political discourse ranging from Madison to Gladstone to Wilson and Reagan (whom it criticized for being half-right). I sent memoranda—with no appreciable effect—to the Dukakis and Clinton campaigns.²

Fifth, and importantly, it has attracted a large legion of many of the most talented scholars in international relations. Some were convinced, and still are, that it was wrong; others, that it was right. They turned their energies and insights into demonstrating their convictions, which generated many citations. (I acknowledge that if more had been convinced that it was right—or wrong—it would have had less “impact,” as measured by the citations that have been tallied). Scholars contributed new insights on the relative weights of institutional, normative, and material influences, and the relative significance of disputes and armed conflicts, perceptions and realities, and transitional versus established democracies. It would exhaust my quota of words to begin a bibliography of the excellent literature. Rather than doing so, I will simply refer to the bibliography in a recent study by David Rousseau that both surveys and contributes to the debate that followed (2005, 351–73).