This project introduces a new framework for thinking about ethnic cleansing in civil war and tests this theory on the Lebanese civil war of 1975-1990 using an original quantitative data set with information on demographic, migration and violence for over 1,400 villages or urban neighborhoods as well as about 60 interviews gathered during 14 months of fieldwork. Ethnic cleansing can be a powerful military strategy for winning a quick and decisive victory yet I argue that, absent political goals such as secession, armed groups have strong political and economic incentives to moderate the amount of wartime displacement. For these reasons armed groups have an incentive to acquire more fine-grained information about resident non-coethnics to allow displacement of only those who are actively disloyal. This is easiest in areas where different ethnic groups mix socially, such as mixed residential neighbourhoods, which are therefore more likely to witness selective violence but less likely to experience full collective ethnic cleansing. This hypothesis is the direct opposite of the conventional wisdom yet finds strong support in the data which stems from voter registration rolls, government documentation, NGO reports, and other sources and allows for a rich set of quantitative tests.

The control of territory is fundamental to the power of governments and the opponents they confront in civil wars. And yet considerable variation exists not only in the extent to which armed groups control territory over the course of a conflict, but, more importantly, the extent to which they engage in military operations with the intention of taking and holding territory in the first place. The literature on ethnic and sectarian conflict explains patterns of violence according to the identify-based division that defines the war. Fighting to control territorial is predicted to occur largely at the outset of such conflicts, generating sorting into homogenous enclaves; subsequent violence is largely non-territorial and linked to a bargaining process between the warring sides. Lebanon's 1975-1990 civil war is thought to be a case in point. According to common accounts, armed groups in Lebanon fought to create, enlarge, and then defend sectarian enclaves; indeed, the front lines became largely static after the war's first two years. I subject these accounts to theoretical and empirical scrutiny. I provide a theory of the operational goals of violence that captures the trade-offs that armed groups face between responding to tactical pressures in combat and the pursuit of post-conflict political objectives. I then use interviews with former Lebanese commanders and a unique dataset created from local newspapers covering events from 1975-76 at the neighborhood level to test the theory's predictions in Lebanon. I assess the extent to which patterns of fighting in Lebanon resulted from the theorized trade-offs between short-term military necessity and long-term political agendas. The results show that a purely sectarian logic has limited ability to explain patterns of violence; violence that appears sectarian in nature can only be fully understood when placed in the context of armed groups' political aspirations.

In this paper, I explore the effects of civil war territorial control on post-war local politics in Lebanon. Why have pre-war local elites regained their power in certain neighborhoods while others continue to be dominated by wartime players? I argue that consolidating control over territory is a strategic imperative for armed groups contesting a civil war. In order to accomplish this, armed groups can change the composition of the population within the territory, displacing out-group members and attracting displaced in-group members with the promise of security. These vulnerable in-group IDPs have been detached from their local networks, making them likely supporters of groups that offer security. An armed group can also try to win over the remaining population living in the territory, providing security and services in exchange for loyalty and local information. These relationships of exchange are established through the cooptation of local elites and their networks. I argue that these original residents (in contrast to IDPs), while they may also support the armed group, have strong attachments to pre-war local elites. These strategies have consequences for the post-war local balance of power. If the armed group that controls a territory wins the war, it monopolizes the political space. Local elites must either be subordinate to its political directives or risk marginalization. Displaced “new” residents that have direct links to the party will be relatively more empowered. In contrast, if the controlling armed group is defeated, it will be unable to continue monopolizing the locality’s political space. In the vacuum left after the group’s defeat, local elites will function more independently. In this context, “original” residents that have strong links to local elites will be just as, if not more, empowered than displaced “new” residents. I test these hypotheses with data from Lebanon. First, I investigate the strategies used by armed groups to consolidate territorial control. I examine whether there is a relationship between displacement, resettlement, and assistance from armed groups. I use results from a nationally representative survey of 50,000 households conducted in 1987. Then, I undertake a micro-level study of two Beirut suburbs. Both came under armed group control during the war. One suburb’s controlling armed group won the the war. The other lost. Through thirty in-depth interviews with local elites and residents, I investigate the relationship between wartime players and local elites and compare the role of “new” and “original” residents in both communities.

Power-sharing is a common institutional design prescribed for post-conflict societies throughout the world. Within the Middle East, Lebanese “consociationalism” is a longstanding example of such a system while Iraq developed de facto power-sharing arrangements in the aftermath of the U.S. invasion in 2003. The approach is now frequently invoked by scholars and policymakers alike as the primary or sole means of restoring peace and stability in Syria by providing the warring parties with a stake in the post-conflict political order and share of resources. As some studies have shown, however, power-sharing may be effective at “stopping the guns” but often falls short in promoting effective governance and, as a result, can lead to the frequent recurrence of conflict and even all-out war (Cammett and Malesky 2012, Roeder and Rothschild 2005). This paper builds on this research by tracing the effects of power-sharing in Lebanon on public goods provision, an outcome that has received scant attention in the power-sharing literature. The analyses focuses most explicitly on the health and education sectors, showing how the power-sharing system in place even before independence has incentivized high levels of spending and inefficiencies while leading to mediocre human development outcomes vis-à-vis countries at comparable levels of development. To show how these findings hold more generally, the case of Lebanon is then situated in a larger set of countries based on time series cross-sectional analyses of a global dataset of countries.