Title: Targeting Civilians in War
Name: Alexander B. Downes
Assistant Professor of Political Science
Duke University
301A Perkins Library
Durham, NC 27708
(919) 688-2225
Year: 2003
Type: Dissertation Fellowship
Summary: The central question of my research is why do states use military strategies that target enemy civilians in war? Despite the widespread belief that killing noncombatants is both morally wrong and militarily ineffective, states have targeted civilians in one-third of all interstate wars since 1815. Several hypotheses have been suggested to explain variation in civilian victimization, including those based in regime type, perceptions of the adversary’s identity, and organization theory. Some argue, for example, that democracies are uniquely restrained by their domestic norms and institutions from targeting civilians. Other regime type theorists, however, contend that democracies’ cost-aversion and need to win wars quickly may increase the likelihood of civilian victimization. A second school of thought emphasizes the “barbaric” identity of the enemy: civilian victimization results from the belief that one is fighting an uncivilized opponent. The choice of strategy, in other words, depends on one’s view of the adversary: the laws of war only apply in wars against “civilized” opponents, not “barbarians” or racial inferiors. Organization theorists, finally, stress that civilian victimization can result when a military’s culture favors strategies that punish civilians, or when parochial organizational interests—such as attaining greater resources or organizational independence—are served by targeting civilians.

My argument, by contrast, focuses on two variables: desperation to win and to save lives on one’s own side in lengthy wars of attrition, and the intention to conquer and annex enemy territory in wars of expansion. According to the desperation logic, costly and protracted wars cause states to become increasingly anxious to prevail and to reduce their losses. Targeting civilians allows states to continue the war while managing their losses and hopefully coercing the adversary to concede. In wars of territorial annexation, on the other hand, belligerents specifically intend to seize and permanently appropriate territory belonging to another state. Attackers in this model utilize civilian victimization to eliminate enemy noncombatants, who can threaten the aggressor’s immediate military prospects and present a threat of rebellion in the future.

I test my argument against the competing hypotheses using both quantitative and qualitative evidence. First, I conduct a multivariate statistical analysis of civilian victimization by belligerents in interstate wars from 1816-2003. This analysis strongly supports the desperation and annexation mechanisms: states involved in costly wars of attrition as well as those intent on annexing enemy territory are more likely to target civilians and kill larger numbers of noncombatants. Regime type and perceptions of the adversary’s identity by themselves exert little influence on civilian victimization, but democracy in combination with desperation is important: democracies in wars of attrition are more likely than non-democracies to target noncombatants, a finding that begs further research.

Second, I examine a number of cases in depth to show how desperation and annexationist aims actually caused civilian victimization. Among those examined are the starvation blockades of World War I; strategic bombing in World War II; counterinsurgency by the British in the Boer War (1899-1902) and Italy in Libya (1923-32); and the flight and expulsion of the Palestinian Arabs in the War of Israeli Independence.
Bibliography: I completed the dissertation at the University of Chicago in 2004, and have revised it into a book manuscript that is currently under review at a university press. For an article based on my research, see:

Downes, Alexander B. “Desperate Times, Desperate Measures: The Causes of Civilian Victimization in War.” International Security (Spring 2006): 152-195.