Peer-Reviewed Publications
“Democracy Dismissed: When Leaders and Citizens Choose ElectionViolence,” with Kathleen Klaus. 2025. Journal of Peace Research, 62 (5): 1447-1462.
In democratic settings, election violence is often jointly produced: it relies not only on elite incentives and capacities to deploy violence, but also on the willingness of ordinary actors to participate. Yet many studies of election violence overlook this elite–citizen interaction, effectively black-boxing the process through which elites mobilize people to fight. This article introduces and advances the concept of the joint production of election violence – a relatively common but undertheorized process through which political elites rely, not on their own militias or state security forces, but on the collaboration and participation of ordinary citizens. Such violence is especially puzzling in democracies, where citizens ostensibly have nonviolent avenues for political claim-making. To help explain how such violence becomes possible and how it unfolds, the article develops a framework that emphasizes two central components: (1) the circulation and resonance of threat-based and victimhood narratives that legitimize political violence, and (2) the social infrastructure – networks and organizational linkages – that facilitate the organization and coordination of violence. We draw on two cases of jointly produced election violence – Nigeria in 2003 and the United States in 2021 – to demonstrate how the framework can be applied across democracies at distinct stages of consolidation. Broadly, by developing the concept of jointly produced violence and offering a framework for its study, we aim to facilitate more systematic and comparative analyses of elite–citizen interactions in the context of electoral violence, helping to render visible a process that is often invisible in existing studies, while also bridging theories of election violence, democratic erosion, and right-wing extremism.
“Couples Therapy for a Divided America: Assessing the Effects of Reciprocal Group Reflection on Partisan Polarization,” with Hannah Baron, Robert A. Blair, Donghyun Danny Choi, Laura Gamboa, Jessica Gottlieb, Amanda Lea Robinson, Steven C. Rosenzweig, and Emily A. West. 2025. Political Behavior, 47: 1363-1388.
Overcoming America’s deep partisan polarization poses a unique challenge: Americans must be able to sharply disagree on who should govern while agreeing on more fundamental democratic principles. We study one model of depolarization—reciprocal group reflection—inspired by marital counseling and implemented by a non-partisan non-governmental organization, Braver Angels. We randomly assigned undergraduates at four universities either to participate in a Braver Angels workshop or simply to complete three rounds of surveys. The workshops substantially reduced polarization according to explicit and implicit measures. They also increased participants’ willingness to donate to programs aimed at depolarizing political conversations. These effects are consistent across partisan groups, though some dissipate over time. Using qualitative data, and building on contact and deliberative theories, we argue that depolarization is especially effective when it includes both informational and emotional components, such that citizens who are moved to empathize with outgroup members become more likely to internalize new information about them.
“Elites and Arbitrary Power: Ethical Challenges and Guiding Principles for Research with Violent Political Actors,” with Niloufer A. Siddiqui. 2024. Conflict, Security, and Development, 24 (6): 599-619.
Many approaches to and recommendations for ethical research assume a power dynamic between the researcher and the researched that favors the researcher. When study participants include political elites however, the power dynamic is often reversed. This power differential in favor of political elites can render existing guidelines unhelpful or inappropriate, especially where such actors have individualized discretion over coercion and violence. We argue that paying particular attention to how political elites wield arbitrary power can help researchers navigate ethical challenges that arise when political elites are included in the study. We focus on three topics in particular: researchers’ interactions with political elites outside of formal data collection, anticipating and reducing risks for non-elite study participants, and sharing and contextualizing research findings with scholars and a general audience. In doing so, this paper contributes to a growing conversation on the thorny ethical challenges that come with studying violent political elites.
“Nigerian Youth Engagement in Violent Electoral Environments: Political Apathy or ‘Constrained Optimism’?“ with Justine M. Davis. 2024. African Affairs, 123 (491): 133-164. Summary of focus group findings available here.
How do young ordinary citizens engage with politicians and their political environment where elections are frequently affected by violence? We explore this question with focus groups in Nigeria, a country that has some of the highest rates of election violence in Africa. While participants largely condemned violence, they were divided on whether they would continue to support their preferred candidate after hearing that they had engaged in violence, with some participants justifying support for violent candidates if they showed remorse, use violence defensively, or when information about the veracity of violence use was dubitable. Despite participants expressing that violence discouraged participation, we show that they also felt a determination to elect a better government and a strong sense of a civic duty to vote and encourage others to do so as well. Taken together, we argue that young voters in violent contexts operate with ‘constrained optimism’ where they perceive low levels of external efficacy, but possess high levels of internal efficacy: remaining committed to democracy, but facing constraints to who they support and how they participate. Our findings nuance expectations of the effects of electoral violence on political participation and better help us understand the challenges facing voters in contexts where violence is rife.
“Incumbent Responses to Armed Groups in Nigeria and Kenya.” 2024. Studies in Comparative International Development, 59: 56-85.
Under what conditions do incumbents support, tolerate, or try to rein in armed groups within their borders? The paper argues that incumbents strategize toward armed groups in ways that help them manage the ruling coalition and remain in power. I find that incumbents support armed groups in provinces where provincial politicians have defected from the ruling coalition to opposition parties, recruiting armed groups to deliver the province in national elections with fraud and violence and punish elite defectors. Where incumbents are not fighting to keep provinces in the ruling coalition, I show that they tolerate non-dissident groups that enjoy social contracts with local communities, and try to contain their dissident counterparts. Incumbents are likely to repress predatory armed groups, dissident and non-dissident alike. Doing so helps boost elite and mass support for incumbents and project central government power into the province. The argument was inductively built with comparative case studies from Nigeria and then evaluated in Kenya. The findings contribute to an important research agenda on government-armed group relationships and carry implications for security sector aid and reform.
“When Armed Groups Refuse to Carry Out Election Violence: Evidence from Nigeria.” 2021. World Development, 146: 1-15.
Social science research on election violence shows that incumbents regularly turn to different nonstate armed groups to organize violence during elections, including ethnic militias, gangs, criminal organizations, and paramilitaries, among others. Less well known are the motivations of these different actors, what they seek to gain from election violence, and when they turn down incumbents’ overtures. From a practitioner perspective, understanding when armed groups supply election violence for incumbents is important because of the severe consequences of such acts, including economic hardship, forced displacement, damaged infrastructure, and costs to human life. The paper asks: under what conditions do armed groups agree or refuse to perpetrate election violence for incumbents? Drawing on most similar case studies of the Ijaw Youth Council and the O’odua People’s Congress in Nigeria, we find that internal armed group politics help to explain how these actors respond to incumbent governors’ demands for election violence. Specifically, groups divided by leadership rivalries agree to perpetrate election violence for incumbents whereas those with cohesive leadership refuse to do so. Leaders of rival factions accept money and arms from incumbents to try to eliminate their competitors, and in exchange, agree to organize violence during elections for their incumbent sponsors. In contrast, groups with cohesive leadership turn down incumbents’ overtures given the risks of cooptation and weakened community support. The findings contribute to our understanding of how election violence is co-produced by elites and nonstate armed groups by explaining the motivations and decisions of the latter. More broadly, the paper speaks to larger questions about security challenges in developing democracies. The findings also highlight the need for greater attention to interventions to prevent armed groups from engaging in election violence on behalf of political elites.
“Elite Competition, Social Movements, and Election Violence in Nigeria.” 2020/21. International Security, 45(3): 40-78.
Election violence varies significantly within countries, yet how and why are undertheorized. Although existing scholarship has shown how national-level economic, institutional, and contextual factors increase a country’s risk for violence during elections, these studies cannot explain why elites organize election violence in some localities, but not others. An analysis of gubernatorial elections in Nigeria reveals the conditions under which elites recruit popular social movement actors for pre-election violence. Gubernatorial elections are intensely competitive when agreements between governors and local ruling party elites over the distribution of state patronage break down. To oust their rivals and consolidate power, elites recruit popular reformist groups for pre-election violence and voter mobilization. Conversely, when local ruling party elites are aligned over how state patronage is to be distributed, the election outcome is agreed to well in advance. In this scenario, there is little incentive to enlist social movement actors for violence. Case studies of the Ijaw Youth Council and Boko Haram provide empirical support for the argument. The theory and evidence help to explain subnational variation in election violence, the relationship between intraparty politics and violence during elections, and speak to broader questions about political order and violence.
“When Militias Provide Welfare: Lessons from Pakistan and Nigeria.” With Yelena Biberman, 2018. Political Science Quarterly 133(4): 695-727.
Militias are often noted and studied for their violent behavior towards civilians. Still underexplored and poorly understood is their provision of welfare to local communities. Why do some militias provide welfare while others do not? We argue that the timing of access to material resources plays an important role in the militia’s willingness and capacity to provide welfare. While wealthy militias lack the incentives and poor militias lack the capacity to provide public goods, “nouveau riche” militias – which begin poor and acquire wealth later on – are most likely to provide welfare. The empirical implications of the argument are explored with a structured comparison of six militias in Pakistan and Nigeria. Our findings help to illuminate the broad range of militia behavior and bridge three burgeoning literatures – on militias, non-state welfare, and insurgent behavior in civil war.
Works-in-Progress
“Virtual Focus Groups on Sensitive Topics in the Global South: Opportunities and Challenges,” with Justine M. Davis, Juliet Ilori, and Alaba Uthman. Revised and resubmitted to Qualitative and Multi-Method Research
During the COVID-19 pandemic, many researchers turned to virtual data collection techniques, including ourselves. In this research note, we share our experience of organizing virtual focus groups with ordinary Nigerians. We discuss the lessons learned, advantages for researchers and study participants, and reflect on the distinct risks and limitations of virtual data collection. We conclude that, overall, virtual focus groups carry notable benefits and suggest different protocols to address important concerns regarding privacy, data management, and unique challenges for virtual discussions.
“Authenticity and Election Violence: How Candidate Narratives Shape Political Participation,” with Justine M. Davis. Under review; pre-analysis plan available here.
How do ordinary citizens engage with politicians and their political environment, where elections are frequently manipulated with fraud and violence? We explore this question with focus groups and a conjoint experiment in Nigeria, a country that has some of the highest rates of election violence in the world. We argue that in contexts where violence is expected and where many candidates either implicitly or explicitly condone or justify it, voters still have preferences over candidates based on how they talk about violence. We theorize that “authentic” justifications for violence use can shape whether voters see candidates as trustworthy or representative, even when candidates admit to using violence. We test how narratives (remorse, endorsement, resignation, and defensive versus offensive violence use) shape vote choice and broader political engagement. Our findings nuance expectations of the effects of electoral violence on political participation and help us better understand the challenges facing voters in contexts where violence is rife.
“Does Fractured Authority Fracture Legitimacy Beliefs? Evidence on (Non)State Armed Security Provision from Nigeria,” with Hanne Fjelde, Kristen Kao, and Tore Wig. Under review; pre-analysis plan available here.
What underpins the legitimate use of violence? Existing research points to security provision, impartiality, popular approval, social contracts, and the formal status of armed actors. This paper develops an experimental vignette to test these logics of legitimation in Nigeria, a context where state and non-state security providers have long coexisted. Using a survey experiment supported by observational analyses and focus groups, we find that citizens evaluate state and non-state actors based on procedural fairness, suggesting a shared framework for judging armed authority regardless of formal status. State actors are held to higher performance standards, however, which focus group discussions suggest derive from their constitutional mandate to protect citizens. Overall, we conclude that while authority may be fragmented across state and non-state armed actors, beliefs about legitimacy remain coherent. The findings contribute to scholarship on state building, armed groups, and political violence.
“Social Capital and Communal Violence in Nigeria,” with Joshua R. Gubler and Ashutosh Varshney. Writing in progress.
Studies of ethnic violence have focused a great deal on “bridging” social capital and less on “bonding” social capital. The former term refers to ties across group boundaries, and the latter to networks within the confines of a group. While some studies suggest that bonding prevents inter-group violence via in-group policing, others note that these same ties can provide a social infrastructure to coordinate and carry out violence. Under what conditions is bonding more conducive to violence or peace? Closely investigating the evolution of two cities in central Nigeria, one endemically violent in recent years (Jos) and the other relatively peaceful (Zaria), we show that bonding religious networks are linked both to communal peace and violence. A key difference between these two cities, we argue, is that in Jos, bonding social capital within religious communities became entangled with ethnic power struggles around the time of Nigeria’s 1999 democratic transition, while religious networks in Zaria have largely remained separate from ethnic tensions. The intersection of ethnic and religious identity in Jos, in other words, has led to the repurposing of communal bonds within religious communities for the organization of violence. In contrast, bonding social capital in Zaria has led to in-group policing and restraint. We use novel survey data and qualitative evidence to support our claims. Our findings contribute to scholarship on social capital and democratization, identity politics, and the social organization of violence.
