Article Conflict Management and Peace Science 1–21 Ó The Author(s) 2018 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0738894218795134 journals.sagepub.com/home/cmp Selective or collective? Palestinian perceptions of targeting in house demolition Sophia Hatz Uppsala University, Department of Peace and Conflict Research, Sweden Abstract There is a growing consensus that repression and counter-insurgency can be effective when selec- tive. Yet the empirical evidence is mixed and theories specify that (unmeasured) perceptions of target selection matter. This article addresses this gap by directly measuring individuals’ interpre- tations of a coercive policy which varies in target selection. It employs original surveys with Palestinians on their exposure to house demolition, views on the policy and attitudes towards the Israel–Palestine conflict. The study finds that when interpreted as indiscriminate, house demolition increases opposition to compromise. The results are consistent when perceived tar- get selection is manipulated in an embedded survey experiment. Keywords Coercion, perceptions, Israel–Palestine, repression Introduction Under what conditions is state repression and counter-insurgency effective, and under what conditions is it counter-productive? There is a growing consensus among academics and pol- icymakers that a state’s use of threats and violence against non-state actors is effective when selective. In targeting militants and their supporters precisely, selective tactics signal punish- ment for engaging in armed resistance and generate clear incentives for compliance with a state. Indiscriminate targeting, on the other hand, is considered ineffective at best, and counter-productive at worst (Kalyvas, 2006:151). In targeting both militants and civilians, indiscriminate tactics fail to signal that punishment can be avoided and generate moral out- rage, motivating dissent (Kalyvas, 2006; Toft and Zhukov, 2015). At the same time as this is a widely accepted theory, the empirical evidence is mixed. In particular, a number of studies find that selective counter-insurgency measures, such as drone strikes and targeted assassinations, can be counter-productive when they result in Corresponding author: Sophia Hatz, Uppsala Universitet Institutionen for freds- och konfliktforskning, Box 514, Uppsala, 751 20, Sweden. Email: sophia.hatz@pcr.uu.se