I am grateful to Patricia Owens for her incisive commentary, and for posing provocative questions to frame this response. First, Owens questions my categorization of government and rebel-group violence against civilians—describing as “low-casualty cleansing and terrorism” the Turkish government’s evacuation and destruction of Kurdish villages, and as “high-casualty terrorism” the PKK’s use of bombs to attack public civilian targets. Owens argues that this involves normative and political judgments about particular types of violence. I agree that even behavior involving restraint can induce terror. I refer to the Turkish government’s strategy as “terrorism” for precisely this reason: the purpose was to induce terror among the Kurdish population, thereby pressuring the opponent to make concessions.
However, the categorization of violence is descriptive and analytical, not normative or political. Although the terms “low and high casualty” describe body counts, my focus is on the belligerents’ intent in choosing a strategy of low- or high-casualty control, cleansing, or terrorism. The Turkish government’s strategy is low-casualty cleansing and terror compared to other belligerents that use high-casualty cleansing and terror. For example, the Sudanese government’s strategy in Darfur was high-casualty cleansing: the government intentionally killed several hundred thousand civilians to clear territory of ethnic groups believed to be aiding insurgents. The Turkish government’s strategy was low-casualty cleansing and terrorism because it used violence to forcibly expel civilians from territory and to impose costs on the PKK; however, the Turkish government also limited violence, killing many fewer civilians than the Sudanese government did and killing fewer civilians than its own well-equipped military was capable of killing. Meanwhile, the PKK strategy in Turkey was high-casualty terrorism because they deliberately killed large numbers of civilians to induce government concessions. In contrast, the Farabundo Martí National Liberation Front (Frente Farabundo Martí para la Liberación Nacional, FMLN) in El Salvador also used terror to coerce the government, but chose targets to minimize civilian deaths, despite having the capacity to inflict greater casualties.
Second, Owens inquires about the impact of military and financial assistance from hegemonic powers on incentives for restraint—for example, U.S. military and financial support for the Indonesian government. U.S. complicity in Suharto-era atrocities and in human-rights abuses after Indonesia’s democratic transition is important, and I do not intend to portray liberal governments as always upholding humanitarian standards in their support for belligerents. The premise of my argument is that Western states, more so than non-Western states, incur costs in departing from humanitarian norms, not that every Western government is perfectly consistent in its rhetoric and behavior.
Additionally, my focus in the book is on belligerents’ attempt to win diplomatic support. I argue that belligerents exercise restraint to garner support from Western actors, whom they perceive to care about violations of international humanitarian norms. In East Timor, for example, the Revolutionary Front for an Independent East Timor (Frente Revolucionária do Timor Leste Independente, Fretilin) lobbied for a shift in international attitudes, in part by contrasting its own restraint with Indonesia’s abuses. Although Australia and the United States could have done more to pressure Indonesia to halt abuses, they did pull back from their long-standing support for Indonesia following the 1991 Dili massacre. This loss of international diplomatic support was instrumental in inducing Indonesia to alter its East Timor policy.
Ultimately, my book shares with Owens’s book a desire to explain the devastation that war participants often inflict on civilians. Pushing back against the tendency to analyze government and rebel-group behavior in isolation, I show that governments and rebel groups often have similar intentions in directing violence against civilians; face similar constraints on violence, in the form of pressure from domestic and international audiences; and make similar calculations about the strategic value of restraint.
