War, casualties, and public opinion

The authors begin the construction of a generalizable theory of casualties and opinion, reexamining the logic employed by Mueller and showing that although human costs are an important predictor of wartime opinion, Mueller's operationalization of those costs solely as the log of cumulative national casualties is problematic and incomplete. The authors argue that temporally proximate costs, captured as marginal casualty figures, are an important additional aspect of human costs and a critical factor in determining wartime opinion. Using Mueller's data on opinion in the Vietnam and Korean wars, the authors find that marginal casualties are important in explaining opinion when casualty accumulation is accelerating, and earlier findings about the importance and generalizability of the log of cumulative casualties as the sole casualty-based predictor of opinion are overstated. Finally, the authors offer some thoughts about other factors that should be considered when building a model of war deaths and domestic opinion.