A HOUSE DIVIDED: CIVILIAN DEFENSE POLICY BEHAVIORS IN A MULTIPLE-PRINCIPAL SYSTEM

In democracies, civilian political leaders have ultimate control over the military while the military has the coercive power to overthrow the government. Scholars tend to assume civilians behave as a unitary actor that consistently and uniformly develops defense policy for the military to execute. Yet, in the U.S., multiple civilian institutions possess distinct roles to set policy and a collective role to control the military. What explains variation in defense policy preferences among the institutions responsible for civilian control of the military? I argue that civilian homogenization obscures three salient factors that produce unique policy preferences across the institutions responsible for civilian control: multiple-principal system, party affiliation, and military experience.To answer this question, I use a nested analysis approach. First, I conduct descriptive and inferential statistical analyses using a novel dataset. I identify and assess defense policy variations across the three civilian institutions in the U.S. with civilian control responsibility: Congress, the presidency, the and defense secretary. Holding the military constant, I find that the multiple-principal system, party affiliation, and military experience of each institution affect its respective policy formulations through the annual defense budget. Second, I conduct a series of qualitative, model-building case analyses to identify omitted variables and interactions in my regression model. I find that broader, non-defense policy agendas as well as the interactions between Congress, the president, and the defense secretary their preferences to entrust foreign policy goals and their associated funding to the military. Civilian controllers are not passive observers nor blank slates subject only to military interloping. They represent various interests, perspectives, and preferences independent of their interactions with the military institution. Consequently, my research suggests that civil-military tensions are not limited to this divide, they may also be contingent upon these three institutional factors that alter civilian policy preferences. The assertion is that civilian control is an inherent civilian responsibility. Thus, the implication is that identifying the sources of policymakers’ preferences is imperative to understand what drives civilian institutions’ defense policy development and entrusting it to the military.