Abstract Although government use of militias during civil conflict can ultimately undermine state authority, governments still use militias for battlefield assistance. This paper examines the selectivity of government decisions to use militias by disaggregating civil conflict to the level of battle phases. Civil-conflict battles typically consist of four phases: preparation, clear, hold, and build. I argue that governments decide to use militias based on the strength of government security forces, operational advantages of militias, and the type of battle phase. Governments will limit the use of militias during key battle phases that are likely to receive increased media attention unless a victory secured by government security forces is unlikely or militias hold an operational advantage. A comparative analysis of the offensive operations in Tikrit and Ramadi during Iraq’s war against the Islamic State of Iraq and Syria (ISIS) lends initial support to this theory.

中文翻译：

---

不是是否，而是何时？政府在战争中使用民兵

摘要 尽管政府在国内冲突期间使用民兵最终会破坏国家权威，但政府仍然使用民兵进行战场援助。本文通过将国内冲突分解为战斗阶段的水平，研究了政府决定使用民兵的选择性。内战战斗通常包括四个阶段：准备、清除、保持和构建。我认为，政府决定使用民兵是基于政府安全部队的实力、民兵的作战优势以及战斗阶段的类型。政府将在可能受到更多媒体关注的关键战斗阶段限制使用民兵，除非政府安全部队不太可能取得胜利或民兵拥有作战优势。