Identifying a clear relationship between rebel group structures and the use of violence faces the challenge that group structures rarely change over time. We exploit the analytical advantage provided by long religious holidays to address this issue using the principal-agent framework. Religious holidays serve as a focal point and reduce group coordination costs, but also raise the societal costs of violence. We argue the principal of rebel groups is more sensitive to the increased societal costs than the agents and thereby attempts to restrain the agents from attacking during religious holidays. However, the success of these attempts depends on the group’s institutionalization level. Long religious holidays exacerbate the principal-agent problem due to the high costs of restraining agents from attacking for an extended period, making them the most likely case for our analysis. We test the theory by first conducting microlevel analysis of Islamic separatist groups in three Southeast Asian countries and then analyzing a cross-sectional sample of Islamic rebel groups. Results show that highly institutionalized groups that have a central command system and control over constituent groups are less likely to attack during long religious holidays than on other days, and vice versa for weakly institutionalized groups.

中文翻译：

---

反叛机构化、宗教节日和政治暴力

确定反叛团体结构与使用暴力之间的明确关系面临着一个挑战，即团体结构很少随着时间的推移而改变。我们利用长宗教节日提供的分析优势，使用委托代理框架来解决这个问题。宗教节日作为一个焦点，降低了群体协调成本，但也增加了暴力的社会成本。我们认为，反叛组织的负责人比特工对增加的社会成本更敏感，因此试图限制特工在宗教节日期间发动袭击。然而，这些尝试的成功取决于该团体的制度化水平。较长的宗教假期加剧了委托代理问题，因为长时间限制代理人发动攻击的成本很高，这使它们成为我们分析的最有可能的情况。我们首先对三个东南亚国家的伊斯兰分离主义团体进行微观分析，然后分析伊斯兰反叛团体的横截面样本，以此来检验该理论。结果表明，拥有中央指挥系统并控制组成团体的高度制度化团体在较长的宗教节日期间发动袭击的可能性低于其他日子，对于制度化较弱的​​团体反之亦然。