Skip to content


I am broadly interested in intrastate conflict dynamics and peace processes, including how multi-party conflict influences government strategy, the bargaining power of insurgent groups, and civilian perceptions of violence and state legitimacy. Please read below for information about my current book project, as well as other works in progress.

Book Project

Combatting more than one rebel group strains governments’ military abilities by dividing resources across multiple wars. Governments that would otherwise be able to defeat a rebel group may be unable to do so when tied down by multiple insurgencies.  Peace agreements are heralded as tools for ending civil war. I am currently working on a book manuscript that looks at exclusive “peace” agreements—those agreements governments that exclude at least one warring party in the conflict— and argues that these settlements function as a part of states’ counterinsurgency strategy.

Chapter 1 – Introduction.
Chapter 2 - A theory of peace agreements as a counterinsurgency strategy.
Chapter 3 - Cross-national quantitative test: Do warring parties sign exclusive peace agreements in response to the threat posed by other rebel groups?
Chapter 4 - Cross-national quantitative test: In multiparty conflicts, can a rebel group leverage the threat posed by other rebel groups to achieve a better bargain?
Chapter 5: Cross-national quantitative test: Does the threat posed by other rebel groups serve as an effective commitment device, creating durable peace between signatories?
Chapter 6: Experimental test: Do exclusive peace agreements change hearts and minds?
Chapter 7: Qualitative test: A case study of civil conflict in the Southern Philippines.
Chapter 8: Conclusion.

Papers Under Review

Divide and Conquer: Exclusive Peace Agreements as a Counterinsurgency Strategy

In civil wars involving more than one rebel group, how does war between one rebel group-government pairing affect conflict dynamics with another? As a government must spread resources across multiple fronts, fighting more than one rebel group strains a government's war-fighting capabilities. While peace agreements are often thought of as tools of peace, I argue governments may sign peace agreements as a counterinsurgency strategy: by signing a peace agreement that excludes one or more of the conflict’s rebel groups, a regime can reduce the number of fronts across which it must divide material and human resources. A government can then redirect previously encumbered weapons, troops, and equipment into the battles against the remaining rebel forces and increase the likelihood of a government military victory. As such, a government is more likely to sign an exclusive peace agreement with a given rebel group when other insurgent forces are extremely threatening. I present two sets of evidence to evaluate these arguments. First, I use a large cross-national dataset to test whether the threat posed by other rebel groups increases the likelihood of a government and rebel group signing a negotiated settlement. I find that the likelihood that a government and rebel group sign an exclusive peace accord is positively correlated with the strength and number of other active rebel groups in a conflict. Second, I use a case study of civil war in the Southern Philippines to identify the mechanisms behind these correlations. In line with the theory, the evidence shows that the threat posed by other rebel groups can jump-start stalled peace talks.

I’ve Got Friends in High Places: Intergovernmental Organizations and Ethnic Party Formation (with Victor Asal)

Why do some ethnic organizations participate in elections while others do not? The literature on ethnic political parties has identified institutional features and ethnic group characteristics that predict ethnic party formation – but these cannot explain which organizations represent an ethnic group as political parties. Scholars have analyzed how intergovernmental organizations (IGOs) promote democratization in a top-down fashion by pressuring and incentivizing regimes to adopt reforms. In this paper, we study the effect of a bottom-up approach to democratization – IGO support of ethnic organizations' participation in elections. Analyzing original data on ethnic organizations in Eastern Europe and Russia from 1991 to 2006, we test whether IGO support encourages ethnic organizations to act as ethnic political parties. Using a nested random-effects model to account for the demographic and institutional features, we demonstrate that ethnic organizations with the backing of an IGO are significantly more likely to participate in elections as a political party. A case study of IGO support for party development in post-war Bosnia and Herzegovina clarifies the theory's causal mechanisms. In studying how ties between IGOs and specific ethnic organizations alter the number and kind of ethnic organizations that participate in elections, this research models an important selection process that should inform the broader discussion on the consequences of emergent ethnic parties.

Works in Progress

Formalized Side Switching & Rebel Group Participation in Counterinsurgency

Peace agreement provisions for military integration are often seen as mechanisms to reduce rebel groups’ vulnerability during the implementation of a peace accord. I posit that military integration serves an additional function: to enhance governments’ counterinsurgency capabilities. Governments can supplement their counterinsurgency capabilities by incorporating rebel soldiers into the national armed forces. In agreeing to participate in governments’ counterinsurgency campaigns, rebel groups gain confidence that a government won’t renege on a deal. By consolidating war fronts and incorporating rebel soldiers into government counterinsurgency offensives, peace agreements in one conflict function as war-fighting tools for another. This article presents data that is a corrective to current data collection practices that obscure both multiparty nature of many conflicts as well as a primary mechanism by which rebel groups switch sides in civil war

#CivilWar: Using Social Media Data to Predict Conflict Onset and Intensity (with Håvard Hegre)

Observational civil war data often fails to capture important on-the-ground variables. Examples include: the opposition’s dedication to the cause, the public’s perceptions of injustice, and levels of civilian fear. These can be important weathervanes for predicting the course of political violence but are difficult to measure using traditional methods. This paper uses social media data to capture these variables by drawing on a sample of 1% of all Twitter posts geolocated in Africa. Topic modeling and sentiment analysis allow us to produce cross-national measures of these intangible variables that account for precise physical and temporal location. By analyzing this data, we examine whether the tone or topics of social media posts can predict civil war escalation at a specific place and/or time. We evaluate the value of social-media data for this purpose by incorporating them in the ViEWS forecasting model of organized political violence and assess the extent to which they improve the model’s ability to forecast the onset, escalation, and de-escalation of violence.

What’s in a Name? Media Coverage and Islamist Franchises

Terrorist organizations gain financial support, strategic guidance, and tactical knowledge by joining transnational terrorist networks such as Al Qaeda and the Islamic State (Mendelsohn 2016). This paper provides empirical support for an additional advantage garnered by groups that pledge allegiance to these organizations – notoriety. Gaining media attention is the principal motivation for many terrorist attacks (Hoffman 2006). Mapping all coverage of eight terrorist organizations in the New York Times and the Washington Post from 2002-2018, this paper shows these organizations receive more frequent and more dramatic media coverage after pledging allegiance to either Al Qaeda or the Islamic State, independent of the frequency and severity of the group’s terror attacks. In doing so this paper highlight an important and understudied benefit terrorist organizations gain by joining larger transnational organizations.

Understanding the Local Effect of UN Peace Operations: A Quasi-Experimental Spatial Impact Evaluation in Burundi
(with Susanna Campbell)

When do UN peace operations have the desired sub-national effect in post-conflict countries? The literature on UN peacekeeping argues that the presence of robust and well-resourced UN peacekeeping operations provides a security guarantee and deters fighting between warring factions, contributing to short- and long-term violence reduction (Walter 1997; Doyle and Sambanis 2000; Fortna 2004; Hultman et al. 2014). While these cross-national studies provide important insight into the influence of UN peacekeeping on civil war, they overlook crucial heterogeneity at the sub-national level. First, peacekeeping is no longer solely the realm of peacekeepers. Numerous civilian personnel work for and alongside UN peace operations to build security, governance, and economic institutions. Second, UN peacekeeping does not operate uniformly throughout a territory, instead targeting particular sub-national locations. A sub-national examination of multidimensional UN peace operations gives a very different picture of its effects and the causes of these effects. Based on a quasi-experimental impact evaluation of the UN Mission in Burundi – combining a large household level survey with almost 200 semi-structured interviews – we show that the way the mandate is implemented at the sub-national level has an impact on the sub-national and national-level success of the UN peace operation. By capturing the multidimensional work of UN peace operations at the sub-national level, this paper identifies a crucial omitted variable – sub-national mandate implementation – that has been overlooked by all previous studies of UN peacekeeping.

Action or Reaction: Violence Dynamics in the Israeli-Palestinian Conflict
(with Karsten Donnay and Ravi Bhavnani)

Building on studies of reactive or “retaliatory” violence in Gaza/Israel (Jaeger and Paserman 2006, Jaeger and Paserman 2008, Haushofer et al. 2010), as well as research that analyzes the use of selective vs. indiscriminate violence in the Israeli-Palestinian conflict (Bhavnani et al. 2011), we analyze the reactive nature of violence in Israel and the Palestinian Territories in the period from 2000-2014. Our analysis is based on detailed geo-coded event data of deadly violence and Palestinian rocket/mortar launches. Using a novel computational technique for robust causal inference in spatio-temporal event data (Schutte and Donnay 2014), we find that violence perpetrated by a group tends to increase as a result of previous violence targeting the same group. Our preliminary analysis suggests that these effects depend on geographical location: for example, reactive signatures differ if we only consider incidents in the Palestinian territories under incomplete Palestinian control versus the full conflict dynamics in Israel and the occupied territories. The analysis further shows that reactive signatures differ depending on the conflict period, illustrated here for the Al-Aqsa Intifada (2000-2005) as compared to the post-Intifada period (2005-2008).