Too brutal for war: comparing rationales for weapon taboos

Abstract What makes a weapon too horrific to use? This study investigates public attitudes toward the use of different weapon types using a conjoint survey experiment fielded on a U.S. sample. By randomly varying key attributes of military strikes—expected civilian casualties, operational effectiveness, and weapon type—we isolate the causal effects of each on public support. Casualty estimates exert the strongest influence, and effectiveness also matters. But we find that respondents rely heavily on powerful categorical heuristics about weapon types. These preferences persist even when other strike characteristics are held constant. The results reveal a robust hierarchy: cyber attacks were most favored, followed by conventional strikes, then cluster munitions over chemical, biological, and finally nuclear weapons. In fact, military operations using more favored weapons were often supported over more effective, or less lethal, ones employing disfavored armaments. These patterns reflect public intuitions about which weapons are perceived as brutal, indiscriminate, or illegitimate. Our study therefore has broad theoretical implications for understanding how weapon taboos interact with instrumental factors. It offers vital insights into the complex interplay of public preferences for military operations, weapon transfers, and arms control.