Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. (Eck and Hultman, 2007;Hultman, 2007;Kalyvas, 2006). It is common for both government and rebel forces to target civilians during civil wars and insurgencies.
Terms of use:
Documents inThere are two reasons. Firstly, terror against civilians can sometimes reduce an opponent's ability to mobilize support, increasing one's chance of outright victory. Secondly, if there is no realistic chance of victory, one can use violence against civilians to create conditions in which an opponent prefers a negotiated settlement to continued fighting, and so improve one's bargaining position (Lichbach, 1998).Weak governments may try to consolidate their position by attacking their own citizens, so care is needed to ensure that the international response to a conflict avoids aggravating civilian suffering (Azam and Hoeffler;2002;Hultman, 2011). Analysis of the data reveals a strong 'cycle of violence': typically, when one side increased attacks against its opponents or its effort in terrorizing civilians, the other side responded in kind. Foreign interventions designed to strengthen the government militarily exacerbated the conflict. In contrast, the cycle of violence was attenuated by interventions which raised the opportunity cost of fighting or reduced the resources available to fund the war effort. These results are relevant to the planning of international responses to conflicts in which weak governments are seriously challenged by rebel movements.The next part of the paper outlines the history of the Peruvian conflict. This is followed by a review the existing literature on civilian abuse and civil war, which informs our specific hypotheses about the factors driving variations in the level of abuse over time. We then present the data used to test these hypotheses, our modeling strategy and results.1 This dataset is described in detail by .