Scholars have long recognized that the social composition of militaries can have important consequences for military effectiveness, among other outcomes of interest to political scientists. While this research agenda received renewed interest following the Arab Spring and has made great progress, the challenge of collecting data for a large universe of cases on a topic about which many governments are intentionally secretive has meant that statistical analysis has remained frustratingly out of reach.

By offering a complete initial step in such a cross-national data collection project, and the rigorous multi-method testing to go with it, Jason Lyall’s first book makes a major contribution to any literature that seeks to understand military design or behavior. In Divided Armies, Lyall explains why modern armies vary in their battlefield performance. His argument is that military inequality—or the extent to which a military includes ethnic groups that are disadvantaged or discriminated against in broader society—causes substantial problems on the battlefield. This is a notable adjustment to more common arguments that merely emphasize diversity or representation in the military. Lyall argues that in militaries characterized by greater inequality, soldiers who are second-class citizen