¹ Cohen, Raymond, ‘Pacific Unions: A Reappraisal of the Theory that “Democracies Do Not Go to War with Each Other”’, Review of International Studies, vol. 20 (1994), p. 223CrossRefGoogle Scholar. We must acknowledge that Professor Cohen's critique is not the only one of its kind. See two articles in International Security, vol. 19 (1994)Google Scholar: Christopher Layne, ‘Kant or Cant: The Myth of Democratic Peace’, and David Spiro, ‘The Insignificance of the Liberal Peace’. Bruce Russett responds to them, however, in ‘The Democratic Peace: “And Yet It Moves”’, International Security, vol. 19 (1995)Google Scholar.

² This characterization of the democratic peace proposition is from Russett, Grasping the Democratic Peace: Principles for a Post-Cold War World (Princeton, NJ, 1993)Google Scholar, ch. 1, and the most common version among its proponents. The caution is deliberate. It does not say that democracies never have made or will make war on each other, though with reasonable and consistent definitions for democracy and war it is impossible to find a clear-cut example of such in the twentieth-century international system. Nor does it say democracies never use force or threaten to use force against each other-clearly they do, but much less often than do other kinds of states.

³ There is, however, evidence that it applies even in very different ‘non-western’ domains, notably in pre-industrial societies. See the ethnographic evidence in Russett, Grasping, ch. 5, and Crawford, Neta, ‘A Security Regime among Democracies: Cooperation among Iroquois Nations’, International Organization, vol. 48 (1994)CrossRefGoogle Scholar. Weart, Spencer, ‘Peace among Democratic and Oligarchic Republics’, Journal of Peace Research, vol. 31 (1994)CrossRefGoogle Scholar, extends the pattern to a variety of historical settings, including medieval Switzerland.

⁴ It is not clear whether he means to include the five named exceptions (noting that they ‘inherited their system of government from the Anglo-American tradition’) a s honorary North Atlantic/Western Europeans or not. In later discussion (e.g., p. 220) he explicitly calls Israel a ‘democratic state’, but also again (on p. 222) declares that a narrow definition of democracy ‘limits the [democratic peace] generalization to the North Atlantic/West European area after 1945’. Because of difficulties like this, and for strong conceptual reasons, Russett's analysis used both a binary definition of democracy/autocracy (not identified by geography) and a continuous definition measuring points along a spectrum. The results-identifying much broader applicability for the democratic peace phenomenon-did not differ very much either way.

⁵ Cohen cites Ray's convention paper, which was published in revised form as ‘Wars between Democracies: Rare, or Nonexistent’, International Interactions, vol. 18 (1993)Google Scholar.

⁶ In talking about the West European and South American ‘pacific unions’ as attributable to influences other than shared democracy he uses a description (p. 221) which we adapt t o Eastern Europe durin g the Cold War years. (We change only proper names and make linguae francae singular for the purpose.) This pacific union appeared in a part of the world ‘bound together by ancient ties of civilisation and culture, possessing a diplomatic lingua franca (Russian) and well-established traditions. Communism was a shared feature. Governing elites had much in common …. [The region] had acquired a strong sense of shared identity enshrined in communal organizations and legislation. Both regions had com e to accept the norm outlawing war as a legitimate instrument of statecraft within the community and had developed (albeit wit h varying degrees of success) mechanisms for the peaceful resolution of international conflict.’ In this variant the words more or less fit, but the tune is sour.

⁷ Maoz, Zeev and Russett, Bruce, ‘Normative and Structural Causes of Democratic Peace, 1946–1986’, American Political Science Review, vol. 87 (1993)CrossRefGoogle Scholar; Russett, Grasping, appendix to ch. 4. On p. 213, however, he promotes Peru (also Ecuador-which we did not code as democratic) to quasi-democratic as examples of ‘quasi-democratic’ states ‘fighting’ each other. (As for fighting, on p. 215 this instance becomes only a ‘limited 1981 border incident’.) One more example of Cohen's flexibility is the inclusion of Lebanon among quasi-democratic states on p. 213, whereas on p. 212 it was ‘more akin to a medieval oligarchy’.

⁸ Examples: Argentina and Chile went very close to war over some islands in 1977 when both were governed by military juntas; since 1991 under democratic rule they have settled twenty-two border disputes amicably; the military of Argentina and Brazil have, under democratic governments, ‘junked’ the idea of the other as a military threat; under democratic rule military spending by these three countries dropped by 25 per cent between 1985 and 1992; more than a half-century of antagonism between Bolivia and Paraguay (including the Chaco War) has been eased by the passing of the Stroessner dictatorship in Paraguay. See James Brooke, ‘The New South Americans: Friends and Partners’, New York Times, 8 April, p. A3.

⁹ Maoz and Russett, ‘Normative and Structural’; Russett, Grasping, p. 122.

¹⁰ Never mind that ‘democratic’ France had not yet held a parliamentary election after the occupation. Nor is Cohen's assertion that mutual perceptions of democratic constraints played no part in avoiding an Anglo-French war over the very dangerous Fashoda crisis incontestable. See the historical analysis in Owen, John M., ‘How Liberalism Produces Democratic Peace’, International Security, vol. 19 (1994)CrossRefGoogle Scholar, and Ray, James Lee, Democracy and International Conflict: An Evaluation of the Democratic Peace Proposition (Columbia, SC, 1995)Google Scholar, ch. 5.

¹¹ Russett, Grasping; Stuart Bremer, ‘Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 1816–1965’, Journal of Conflict Resolution (1992)Google Scholar; Bremer, , ‘Democracy and International Conflict, 1816–1965’, International Interactions, vol. 18 (1993)CrossRefGoogle Scholar. Cohen acknowledges on p. 210 that this kind of exercise has been done, but misses its implications regarding alternative hypotheses.

¹² The statement about no causal mechanism (‘necessary link’ aside) having been developed is itself erroneous. Cohen's standards are unspecified, but he could choose among at least the following complementary efforts. Russett, Grasping, ch. 2, discusses two inter-related sets of causal propositions at length, an d Maoz and Russett, ‘Normative and Structural’, give a slightly different version; Bruce Bueno de Mesquita and Lalman, David, War and Reason (New Haven, CT, 1992)Google Scholar develop and test an elaborate game-theoretic explanatory process, as does Kilgour, D. Marc, ‘Domestic Political Structure and War: A Game-Theoreti c Approach’, Journal of Conflict Resolution, vol. 35 (1991)CrossRefGoogle Scholar. See also Rummel's, R. J. five-volume Understanding Conflict and War (Newbury Park, CA, 1975-1981)Google Scholar, and Lake, David, ‘Powerful Pacifists: Democratic States and War’, American Political Science Review, vol. 86 (1992), pp. 24–37CrossRefGoogle Scholar. Since Cohen wrote, the process has been specified in John Owen's careful case-study analysis, ‘How Liberalism’.

¹³ Reiss, Hans Siegbert (ed.), Kant's Political Writings (Cambridge, 1970), p. 100Google Scholar. Ironically, Kant himself warned against confusing ‘republicanism’ with ‘democracy’. ‘Democracy, in the truest sense of the word, is necessarily a despotism’ (ibid, p. 101).