This paper explores ways in which civil conflict can be simulated using numerical methods. A general two-party model of conflict is developed by extending an approach proposed by [Christia, F., (2012), Alliance Formation in Civil Wars, Cambridge University Press, New York], which is based on a metric of the 'relative power' that exists between the state and a rebel group. Various definitions of relative power are considered and one of these is chosen to illustrate different types of two-sided armed conflict, namely direct-fire, guerrilla and asymmetric warfare. The additional suggestion of Christia that random or stochastic events can lead to unexpected conflict outcomes is also further extended in this paper. The inclusion in the model of terms describing concurrent rebel recruitment of civilians and state deployment of troops are then described. Examples are presented for various hypothetical cases. It is demonstrated that numerical simulation techniques have great potential for modelling civil war. The Christia approach is shown to provide an excellent basis from which numerical models of civil conflict can be built and from which the progress of a conflict can usefully be visualised graphically.

The effects of unemployment on the genesis of civil conflict are examined as both a social and economic factor, with particular emphasis on civil conflict in ethnically heterogeneous nations. A logit statistical analysis of a data set indicates that increased unemployment rates do contribute to the onset of civil conflict.

Studies indicating that women as leaders and negotiators have a pacifying effect on interstate conflict stand in contrast to the reality of women's active involvement in civil conflict through armed rebel groups and insurgencies. This dissertation seeks to provide insight into this apparent paradox by analyzing how and why women become involved in rebel groups, drawing on insights from feminist and IR theories to create a gendered theory of rebellion. Hypotheses developed from this theory are examined using new data on women's participation in rebel groups from 1990-2008. These tests are supplemented with qualitative analysis focusing on the Farabundo Martí National Liberation Front (FMLN) in El Salvador. Among the findings, data on rebel organizations in the post-Cold War era show that women are active in over half of all armed insurgencies, a level of activity much greater than what is recognized by current scholarship in international relations. The analysis also indicates that economic and ethnic- or religious-based grievance motivates women's participation, but disputes theories that portray rebels as profit- or power-seekers.

There has been a great deal of discussion about the large volumes of foreign fighters involved in civil conflicts in Syria and Iraq over recent years. Yet, there remains little systematic evidence about the effect, if any, that foreign fighters have upon the conflicts they join. Existing literature distinguishes between the resources fighters bring to rebel groups and the liability they represent in regards to campaign cohesion. We seek to establish preliminary evidence as to whether or not foreign fighters contribute to the success of the campaigns they join. Our multinomial logistic and competing risks regression analyses of civil conflicts between 1946 and 2013 suggest that foreign fighters are associated with a decreased likelihood of government victory. Furthermore, we offer partial evidence to suggest that foreign fighters from non-contiguous countries are more likely to help rebels achieve a negotiated settlement or to continue their struggle against the government, but not to directly help them achieve victory.

Although largely overlooked in much of the previous research on political tolerance, I argue that contextual factors, specifically state-level features, play a significant role in influencing individual tolerance judgments. Drawing from extant theories of public opinion, international conflict, and political institutions, I seek to further our understanding of the determinants of political tolerance by trying to answer the following question: What accounts for the significant differences in political tolerance levels across countries? While models using individual-level predictors account for some of the disparity in tolerance levels, a substantial amount remains unexplained. I assert that several macro-level theoretical frameworks offer compelling explanations for the marked difference in tolerance levels across countries. Specifically, I examine the effect of state-level external threats, internal threats, and the role of domestic political institutions in shaping individual attitudes towards unpopular groups. To test my propositions, I use data from the 1995-1997 World Values survey as well as multi-level statistical modeling to estimate the aggregate effects of state-level factors on political tolerance levels across 33 countries while also controlling for individual-level predictors.This dissertation demonstrates that elevated objective threats to the state, whether international disputes or incidents of civil conflict, serve to dampen overall tolerance levels. In doing so, this study also highlights that not all types of external threat resonateequally amongst the public. Individuals in countries involved in territorial disputes or countries targeted in international disputes are generally less tolerant overall than those in countries involved in disputes over other issues. In terms of domestic political institutions, I find that electoral rules designed to build consensus and ameliorate societal tensions among groups may actually serve to foster intolerance in countries under certain conditions. Finally, my analyses reveal that the effect of democratic longevity on political tolerance levels is actually conditional based on the type of political institutions that exist in a country. Overall, the findings discovered here underscore the importance of contextual factors in shaping political tolerance levels across countries and stresses the need for this type of analysis in future studies of political tolerance.

<p>In situations of armed violence and insecurity, how do civilians influence the political environment around them? In this dissertation, I present three different studies that broadly engage this question. In concert, the papers presented herein offer new insights on civilians' relationship to armed conflict through a focus on victimization, participation, attitudes on resistance, and the effects of civilian resistance on aggregate levels of violence. </p><p>The first study explores the effects of victimization on political participation. I argue that previous answers to this question have overlooked a key variable for predicting civilian behavior: individual level social context. As a step forward in connecting social support networks to behavioral outcomes, I present the kinship network as a novel measure that captures an individual's valuable and private social interactions. I find that survivors of criminal violence with strong ties to kinship networks are most likely to attend political meetings. By highlighting variation in behavior across victims, I challenge previous work which only examines differences in participation between victims and non-victims. </p><p>Motivated by the assumption that attitudes are a precursor to action, my second study examines civilian attitudes on the efficacy of resistance in regions of protracted violence. Using an original survey fielded in Mexico, I explore the conditions under which civilians are likely to view nonviolent or violent methods as useful tools for change. I first test whether several demographic factors--age, gender, income, knowledge about civil resistance, and media consumption--influence attitudes toward resistance. Moving beyond these variables, I then test whether perceptions about government responsibility affect these attitudes. Specifically, I argue that civilian attitudes towards resistance methods are informed by which political actor civilians view as responsible for their security problems. I find that the predicted probability of viewing nonviolent action as more effective than violent action increases by 20.8% for those who attribute security responsibility to local authorities, compared to other actors. Using an embedded survey experiment, I then address the empirically relevant question of whether these attitudes about resistance correlate with action. I find that compared to those who do not view resistance as useful, respondents who view nonviolence as effective are "supportive types'' who are more willing than others to support local resistance groups, regardless of the methods these groups employ. Together, these analyses provide important information for civilian organizers seeking to mobilize latent support for resistance. Moreover, they enrich our understanding of the ways in which communities can reduce violence in order to reclaim political control during armed conflicts.</p><p>Last, I present an aggregated analysis on the evolution of armed conflict in Mexico. The criminal war in Mexico is extremely complex: Drug Trafficking Organizations, citizens, government agents, amongst others, are all relevant actors within the dynamic evolution of the conflict. Existing research, however, typically ignores the interdependencies inherent to these networks. Using a new collection of machine-coded event data, I generate conflict networks for each year from 2004 to 2010. In doing so, I make three major contributions. First, I offer insights into the potential promise and pitfalls of using machine-coded data for country-level analysis. Next, after cleaning and improving upon the original data, I generate yearly networks, which capture a wide range of violent-related actors. Importantly, I demonstrate how these networks illustrate the interdependent nature of the Mexican conflict and present new insights, such as how government coordination changes in response to cartel violence over time. Finally, I use a latent space approach to identify previously unobservable violence between government actors, criminal groups, and civilians. This research design serves as a platform for future research to investigate the effects of major civilian-led events--such as mass protests--on the evolution of armed conflict.</p> / Dissertation

This study focuses on how perceived security environment affect U.S. unilateral, military intervention in civil conflicts, using the concept of Bayesian learning to illustrate how threat perceptions are formed, how they change, and how they affect the U.S. decision to intervene militarily in civil conflicts. I assess the validity of two primary hypotheses: (1) the U.S. is more likely to intervene in civil conflicts with connections to a threatening actor or ideology; and (2) the U.S. is more likely to intervene in civil conflicts for humanitarian motives in a less threatening security context. To test these hypotheses, I compare U.S. military intervention in three temporal contexts reflecting more threatening security contexts (Cold War and post-9/11) and less threatening security contexts (1992-2001). Results of logit regression analysis reveal that a conflict’s connection to a threatening actor or ideology is the most statistically and substantively significant determinant of U.S. military intervention in civil conflicts, both in more and less threatening security contexts. They also indicate that humanitarian motives are not a statistically significant determinant of U.S. military intervention in civil conflicts, even in a more benign security environment. These findings imply that U.S. unilateral military intervention is reserved for more direct national security threats, even those that are less grave, and that the perception of the U.S. as “global cop” may be misleading, at least in terms of unilateral military intervention.

Previous civil war analyses have approached conflict as a single category with limited exceptions, and this thesis project assesses whether differentiating conflicts by their type and intensity using a local-level geo-referenced analytical approach produces differing results for sub-groups of conflicts. The conflicts are divided into 1) governmental hostilities, where the aim of the armed non-state group is to capture the state, and 2) territorial hostilities, where the aim of the armed non-state group is to capture increased autonomy or secession for a territorial claim. The conflicts are also differentiated by intensity into 1) low-intensity conflicts, with fewer than 1000 battle-related deaths per year, and 2) civil wars, with 1000 or more battle-related deaths per year. The results demonstrate that conflicts with differing insurgent goals and intensities of battle are correlated with markedly different factors. There are three factors – local population density, change in local rainfall, and statewide GDP growth – that are significant to both governmental and territorial hostilities but have opposite signs for the two sets. Only one variable – Polity IV scores – showed a consistently significant correlation for governmental and territorial hostilities. There are no factors that are significant to both low-intensity conflict and higher-intensity civil war. These findings suggest that approaching all conflicts as a single class, particularly at the local level, may not reveal significant differences in factors correlated with conflict. Modeling of local conflict will require differentiation of conflicts into salient sub-groups. For policymakers and practitioners, this research suggests that there is not a one-size-fits-all approach for conflict prevention but that strategies need to be targeted to specific types of conflict.

This thesis investigates federalism and civil conflict. Past work linking federalism and civil conflict has investigated the factors that pacify or aggravate conflict, but most such studies have examined the effect of decentralization on conflict onset, as opposed to the form federalism takes (such as congruent vs incongruent forms, for example). I collect data on civil conflict, the institutional characteristics of federalist states and fiscal decentralization. My theoretical expectations are that federations who treat federal subjects differently than others, most commonly in an ethnically based manner, are likely to experience greater levels of conflict incidence and more severe conflict. I find support for these expectations, suggesting more ethnically based federations are a detriment to peace preservation. I close with case studies that outline three different paths federations have taken with regards to their federal subunits.

Studies of terrorism have explored a number of factors thought to drive the phenomenon. Authors often tie socioeconomic development to reducing terrorism. Among structural explanations of terrorism, however, authors generally neglect the effect of gender inequality, though studies show that gender inequality increases the risk of international and civil conflict. Therefore I explore the impact of gender inequality in important socioeconomic issues on terrorism for 143 countries from 1998-2009. I argue that socioeconomic gender inequality reflects poor state capacity, resulting in grievances that contribute to domestic non-suicide and suicide terrorism. I study gender inequality in the areas of education, labor participation, and life expectancy. Results indicate that education and life expectancy inequality increase the risk of terrorism, while labor inequality is unrelated. While the time frame and data used in this study limit generalizability, results indicate that states should provide socioeconomic gender parity to reduce the risk of domestic terrorism.