Democracy and Victory: Fair Fights or Food Fights?

Ajin Choi, David Lake, and Dan Reiter and Allan Stam have each provided useful rejoinders to the critique of democratic triumphalism in my recent article "Democracy and Victory: Why Regime Type Hardly Matters." ¹ In response, I begin by summariz-ing our arguments and pointing out several issues where we have little or no disagreement. I then examine our two major areas of contention: how best to test whether democracy matters much in explaining military outcomes, and whether the democratic triumphalists' proposed mechanisms convincingly explain why democracies frequently appear to win their wars.

The Arguments

Democratic triumphalists argue that democracies are more likely to achieve victory in warfare because of the nature of their domestic regimes. According to the triumphalists, democracies (1) start only wars they can win easily, and (2) enjoy important wartime advantages such as greater wealth, stronger alliances, better strategic thinking, higher public support, and more effective soldiers.

After examining the data and methods that underpin these findings, I concluded that "whether a state is democratic is not the most important factor to consider" in determining a state's likelihood of victory in war—hence the subtitle of my article "Why Regime Type Hardly Matters" (p.42, emphasis added). I do not argue that regime type plays no role—only that it appears to be modest compared with other factors.

Points of Agreement

The triumphalists and I recognize the striking empirical regularity that democracies have been on the winning side of most wars since 1815. If one looks at all the cases of wars involving democratic states (with a score of 6 or better on the 21-point [210 to 10] POLITY Democracy Index), democracies win more than 80 percent of the time. If one broadens the range of cases to look at whether the more democratic state prevailed irrespective of whether it had a democracy score of 6 or higher, the more democratic state prevails more than 70 percent of the time. And even my restrictive subset of "fair fight" cases still credits democracies with winning more than 60 percent of the time. Given this, the triumphalists and I also agree that on balance democracy is not an obstacle to the successful conduct of war.

We also accept that victory is likely to be the result of a variety of factors, including material power and other, nonmaterial, variables (Desch, p.23; Lake, p. 166; and Reiter and Stam, p. 172). Where we part company is over how much relative influence these factors have in explaining whether a state wisely chooses and then effectively wages war.

There are two other points where the triumphalists suggest that we have major disagreements when in fact we do not. First, Lake and Reiter and Stam argue that I advance a deterministic theory of military victory (i.e., "the more powerful side always wins its wars"), whereas they offer a more theoretically and empirically defensible probabilistic theory of victory (i.e., "more democratic states are more likely to win their wars") ² (Lake, pp. 158-161, and Reiterand Stam, p. 174). This is not an accurate portrayal of my position. When listing the alternative theories that I think better explain victory, I qualified almost every one with words such as "often," "sometimes," "likely," and "could" (Desch, p.7). Indeed there are few, if any, deterministic theories in the field of international relations. ³

Second, Lake argues that causation can be inferred only in the context of a prior theoretical framework and criticizes me for not having one. But elsewhere Lake and Reiter and Stam identify me as a realist based not only on my previous work but also, undoubtedly, on my conclusion that materialist variables such as power play a more important role in explaining victory than does democracy. In other words, they recognize that my explanation of military victory is not divorced from an overarching theoretical framework. Thus, we are in basic agreement about...