Doctoral Dissertation Research in Political Science: Why Do the Longest Civil Wars Exhibit Low-Level Violence?

虽然学者们在过去十年里对内战的研究取得了重大进展，但他们还没有解释一个重要的谜团。为什么持续时间最长的叛乱表现出较低的暴力程度？为什么这些持久但小规模的冲突发生在相对有能力的国家，比如印度、印度尼西亚、缅甸和哥伦比亚，如果这些国家把全部精力投入到反叛乱中，就可以消灭叛乱分子？为什么这些国家容忍这些反叛组织，只要他们使用小规模的暴力？该项目超越了通常基于承诺或信息问题、国家能力低下或政权类型的解释，并将重点放在国家动机上，以解释为什么持续时间最长的叛乱仍保持在低水平的暴力。该项目认为，某些发展中国家不愿意将全部精力投入到国家渗透和国家形成中，而愿意接受某些领域的低状态。如果叛乱分子在这些国家地位较低的地区，这些相对强大的国家愿意容忍这些反叛组织，只要暴力活动保持在阈值以下，并且不威胁到政治领导人。该项目使用多层嵌套分析研究设计来验证其论点，该设计在所有叛乱的跨国统计分析中嵌套了对此类叛乱的次国家比较分析。跨国分析表明，有多个叛乱的中等能力国家的叛乱时间往往最长。该项目考察了印度这样一个具有多种叛乱的中等能力国家，并对自1980年以来持续不断的小规模但持久的毛派叛乱进行了次国家比较分析。研究者在毛派叛乱事件数据集（涵盖2002-2008年）上测试了该论点的含义，该数据集是在之前的实地工作中创建的，并使用了受叛乱影响的所有州的地区级暴力数据。在这个全印度的定量分析中，研究人员对两个受叛乱影响的相邻邦——安得拉邦和恰蒂斯加尔邦的暴力和反叛乱模式进行了比较分析。虽然这两个邦的叛乱都发生在以前由英国间接统治、政府渗透程度较低的地区，但这两个邦的反叛乱和暴力模式完全不同。当安得拉邦在20世纪80年代末开始严肃地镇压叛乱时，恰蒂斯加尔邦直到2005年才升级镇压叛乱。利用在实地工作中收集到的数据，研究者证明了来自叛乱的政治家对威胁感知的差异解释了暴力模式的差异。这一发现支持了一个更广泛的论点，即国家动机解释了这些持久叛乱的低暴力模式。这个项目的学术价值在于，它强调有必要将各级政府的民选官员和政府雇员纳入对冲突强度和持续时间的分析。该项目还创建了第一个关于印度叛乱的事件数据集，从而做出了重大的实证贡献。在这个世界上最大的民主国家，这样一场从较低种姓和部落招募的革命叛乱实际上可以与民主制度共存，这应该引起民主学者、政治暴力分析人士和对种族、公共产品分配和冲突之间联系感兴趣的发展经济学家的兴趣。该项目的更广泛影响源于这样一个事实，即虽然这些小规模的持久冲突不会导致每年死亡人数很高，但它们的持续存在导致地方一级在健康、教育和环境方面付出高昂的人类安全代价。这项研究具有重要的政策意义，因为毛派叛乱最近升级，并构成了严重的安全威胁。对叛乱起源的研究可能有助于政策制定者了解如何改善毛派冲突地区的当地条件。该项目还应提高人们对反叛乱措施选择的理解。它将讲述印度社会中弱势群体的故事，以及一场直到最近才被隐藏起来的战争，从而有望提高国际政策团体和印度城市环境的认识。

While scholars have made significant advances in the last decade in the study of civil wars, they have not yet explained an important puzzle. Why do the longest insurgencies display low levels of violence? Why do many of these durable but small-scale conflicts occur in relatively capable states, such as India, Indonesia, Burma and Colombia, which could eliminate the rebels if they put all their effort into counterinsurgency? Why do these states tolerate these rebel groups as long as they use low-scale violence? The project moves beyond the usual explanations based on commitment or information problems, low state capacity, or regime type, and focuses on state motivations to explain why the longest insurgencies remain at low levels of violence. The project contends that certain developing states prefer not to put all their efforts into state penetration and state formation, and are willing to accept low stateness in certain areas. If rebels locate themselves in these areas of low stateness, these relatively strong states are willing to tolerate these rebel groups as long as violence remains below a threshold and does not threaten the political leaders. The project tests its argument using a multiple level, nested analysis research design, which nests a sub-national comparative analysis of one such insurgency within cross-national statistical analysis of all insurgencies. The cross-national analysis shows that medium-capacity states with multiple insurgencies tend to have the longest insurgencies. The project examines one such medium capacity state with multiple insurgencies-India-and conducts a sub-national comparative analysis on the small-scale but durable Maoist insurgency, ongoing since 1980. The investigator tests implications of the argument on an events dataset for the Maoist insurgency (covering 2002-2008), created during prior field work and using district-level violence data for all states affected by insurgency. Nested within this all-India quantitative analysis, the researcher carries out a comparative analysis of the patterns of violence and counterinsurgency of two insurgency-affected contiguous states, Andhra Pradesh and Chhattisgarh. While in both states the insurgency occurs in areas previously under indirect British rule with low state penetration, the patterns of counterinsurgency and violence are completely different for the two states. While Andhra Pradesh started serious counterinsurgency in the late 1980s, Chhattisgarh escalated counterinsurgency only in 2005. Using data gathered during field work, the investigator demonstrates that variation in threat perception by politicians from the insurgency explains variation in patterns of violence. This finding supports the broader argument that state motivations explain the low violence patterns of these durable insurgencies.The project's intellectual merit lies in its emphasis on the need to incorporate elected officials and state employees, at distinct levels of government, into the analysis of conflict intensity and duration. The project also creates the first events dataset on any insurgency in India, thus making a major empirical contribution. That such a revolutionary insurgency recruiting from lower castes and tribes can actually co-exist with democratic institutions in the world's largest democracy should interest scholars of democracy, analysts of political violence, and development economists interested in the links between ethnicity, distribution of public goods, and conflict. The broader impacts of the project stem from the fact that while these low-scale enduring conflicts do not lead to high annual fatality counts, their persistence leads to high human security costs in terms of health, education and environmental costs at the local level. The research is of high policy importance, since the Maoist insurgency has recently escalated and poses a grave security threat. The study of the origins of the insurgency may help policy makers understand how to improve local conditions in the areas with Maoist conflict. The project should also enhance understanding of choices of counterinsurgency measures. It will tell the story of underprivileged groups in Indian society and a war that until recently was hidden, and thus promises to increase awareness both in the international policy community and in the Indian urban milieu.