Patchwork War: Command, Human Capital, and Counterinsurgency

Abstract

Insurgencies are historically protracted, decentralized, and complex and thus provide a unique environment where individual local commanders exercise considerable autonomy and authority. This decentralized environment then affords commanders the opportunity to conduct operations in significantly different ways. This dissertation proposes that differences in commanders’ human capital accumulation predict unit-level COIN variation during training and employment that is not fully explained by their environment. These different ways to wage counterinsurgency produce measurably different short and long-term violence levels, which contribute toward the larger COIN outcome.

To test this theory I examine brigade and battalion counterinsurgency by the U.S. Army in the Iraq War. An original survey was conducted of over 7000 Army officers to proxy unit-level priorities and strategies. A unique database compiled all unit rotations and their location for the entire war, and at the unit-level, matched survey responses, incidents of violence, and commander’s human capital variables. Local district level data for terrain, ethnicity, and population density accounts for contextual differences for each unit.

The analysis reveals that while unit structure and local environments explain to some extent why units trained and operated differently, commanders’ human capital further predicts significant variation in COIN training, operational employment, and violence outcomes and these findings obtain with and without survey-related measures. Specifically, commanders’ latent rather than developed human capital, most significantly predicts this variation, whether measured as source of commission and level of scholarship or the selectivity of the commanders’ undergraduate institution. The effect of unit structure is not consistent with mechanization theory and can be directly influenced by the unit commander. Ultimately, this variation persists throughout the war, despite exogenous national level changes such as doctrine, troop levels, or strategy. The implications of these findings confront military unit rotation policy, the utility of professional military education, and wartime assessment, adaptation, and innovation. While additional data for local insurgent and counterinsurgent activity would bolster these findings, the scope of this unique data and the significant findings further the field of civil wars, counterinsurgency, and military innovation.