Reliable Partners: How Democracies Have Made a Separate Peace.

By Charles Lipson. Princeton: Princeton University Press, 2003. 288p. $32.50.

Here is the message of this important and engagingly written book: Democracies almost never fight one another. The evidence for the existence of a democratic peace is solid and persuasive. To date, no theory to explain this is satisfactory. But the explanation is that wars are caused by problems of incomplete information and states' inability to commit reliably to carry out their promises and threats. Contracts and transparency of behavior provide this ability, and democracies are uniquely suited to provide them. So, at least between themselves, democracies can overcome distrust and avoid war.

After Charles Lipson opens with a brief overview of the volume, he reviews the historical evidence in many “hard cases” of near wars between democracies during the nineteenth and twentieth centuries, concluding that genuinely democratic elements are at work in their ability to settle conflicts without war. He then reviews a long list of theories that have attempted to explain this result, concluding that each is unsatisfactory or at best a partial explanation. His alternative, then, is a “contracting theory”: Constitutional procedures make promises easier to carry out and harder to reverse by successor governments; public debates and relatively open decision making assure democracies' negotiating partners that the commitment is reliable. Democratic government is not totally advantageous to striking bargains, since its procedures are time-consuming and sometimes messy. But what is lost in simplicity is made up by the gravity of the commitment once made. In effect, democracies solve the problem of Hobbesian anarchy defined as the lack of higher authority to enforce binding commitments. The transparent structure of democratic institutions, coupled with the equally transparent transnational institutions they build for dispute settlement among themselves, provide the instruments for self-binding and thus the avoidance of military violence between them.

Throughout Reliable Partners Lipson melds theory with historical examples. Although he presents no new statistical analyses, he reviews existing studies in a confident and nontechnical manner. In good Lakatosian fashion, he extends his contracting theory with evidence for new related hypotheses, such as that when democracies seek agreements with autocratic states, they will require stronger and more extensive self-protecting conditions in the absence of the kind of secure contracting commitment that democracy can provide. He expects longer periods of democracy to produce more stable peace, and that the frequency and severity of militarized disputes between democracies will diminish over time as their contracting credibility rises. On the contested question of whether democratizing states will be more or less prone to disputes and war, he notes that even if they were, it is to be expected from his theory that such states, unable in the early stages to firmly establish the basis for making credible commitments, would for a while experience more disputes.

The author is less enamored with multilateral institutions than with the commitments democracies can make bilaterally. Consequently, he explicitly doubts the ability of Kant's “pacific union” in a weak confederation to act as a credible agent for deterring aggression against its members, especially if it lacks a single dominant member able to transcend the collective action problem. As he admits, the evidence on this is not very systematic, but he does not mention that part of the reason lies in testing that does not distinguish the game theoretic implications of the difference between some alliances' success in deterring attacks (manifest as the dogs not barking) and others' inability to defend collectively against an attack that occurred because the commitment was not very credible. Also, increasing evidence that democracies are more peaceful in general, not just with each other, cannot easily be explained by his contracting theory.

A more serious criticism is that his review of competing theories is incomplete, and thus overstates the power of his theory to explain better than its predecessors. He gives but scant discussion of the following chain of institutional reasoning: Democratically elected leaders must satisfy a very large segment of the populace in order to retain power in the next election. Long and expensive wars will alienate much of the electorate. Thus, democratic leaders will avoid wars they cannot expect to win easily. Leaders of other democracies will have the same incentives. Thus, both sets of democratic leaders are likely to settle differences with each other peacefully, sometimes by the weaker party making disproportionate concessions. Autocratic leaders, needing only to satisfy a relatively narrow circle of cronies, have more to gain from war and accept less risk of losing office from a war that goes bad. Bruce Bueno de Mesquita, Alastair Smith, Randolph Siverson, and James Morrow, in The Logic of Political Survival (2003), have laid out the reasoning and evidence for this in carefully elaborated form, as have Dan Reiter and Allan Stam (Democracies at War [2002]) in a variant. While it is unfair to expect a book published in the second half of 2003 to cite those particular works, both variants have been around for a while in articles, and a discussion of institutional theory about the democratic peace without them misses a key piece.

More fundamentally, I still doubt that a single deductive theory is likely to explain satisfactorily all instances and varieties of the democratic peace phenomenon. In this I cling to the idea that several strains of theory, emphasizing aspects of accountability as well as transparency induced by institutions, and norms as well as institutions, will show different degrees of explanatory power in different contexts. In short, such a complex social phenomenon may be reached, equifinally, by multiple paths. Lipson has developed one of those explanatory paths with a detailed and balanced analysis. It is surely a big part of the picture, and scholars of conflict and of institutions should read it. Moreover, his presentation will be accessible to an audience of students and policymakers that too often finds formal theoretic or large-scale statistical analysis opaque. If you want a book that takes theory seriously yet will engage students on fundamental aspects of international politics, this is one on a short shelf.