top of page

Research in Progress

Book Project

Refusing to Cooperate with Armed Groups

Civilian Agency & Collective Action in the Colombian Civil War

My book deals with the collective roles that civilians come to play in the context of civil war. Concretely, it documents and analyzes a little-studied pattern of civilian agency: civilian noncooperation with armed groups. It develops a theory that specifies where and when civilians are more likely to organize themselves to refuse non-violently to cooperate with armed organizations.  Where territorial control is shifting, where violence against civilians has recently spiked, and where targeting is perceived as unavoidable, a desire for noncooperation is likely to evolve. However, this desire is not enough for us to observe organized noncooperation. Campaigns of noncooperation are likely to emerge when desire meets capacity for collective action. Localities with a prior history of mobilization and/or with the support of external actors are more likely to count on the leadership and the associational space needed for organizing action. 

Civilian noncooperation is proposed both as a strategy of community self-protection and a form of contentious politics. In this sense, I bridge scholarship on the micro-dynamics of civil war, civil resistance, social movements/collective action and civilian protection. The analysis is embedded in a three-stage research design that combines within-case analysis, cross-case structured and focused comparisons, and paired comparisons of positive and control observations. The empirical data, both qualitative and quantitative, was gathered during two separate waves of field research in multiple warzones in the Colombian civil war using different techniques of data collection. 

The goal of this book is accomplished to the extent that it succeeds in the art of combining parsimonious theorization of an outcome with the smells and sounds of the complex processes that give life to that outcome. In other words, providing sensitive simplification and empirically falsifiable claims is as important as offering a realistic and fair account of the lives of the communities I lived and worked with over the past years. Ultimately, it is for the reader to judge.

The Human Costs of the War on Drugs​

Attitudes towards Militarization of Security in Mexico

[with Davide Morisi, Collegio Carlo Alberto]

What determines citizens’ support for the militarization of security?  Do people citizens care about the human costs of the “war on drugs”?  While evidence suggests that the militarization of security might not be effective in curbing crime, “iron-fist” approaches to security are widely used and enjoy strong popular support.  In this project, we address this puzzle by exploring whether awareness of human costs shapes people’s attitudes towards the militarization of security in Mexico’s “war on drugs”. We conducted a laboratory experiment (Study 1) and a survey experiment with a nationally representative sample (Study 2) in which we manipulated the presence of human costs related to a military operation against a drug lord, and presented participants with arguments either justifying or condemning these costs. Results consistently show that support for militarization decreases when individuals are made aware of human costs – even in successful military operations. However, this is the case only when casualties are civilian (i.e., people not associated with drug cartels), and among those who have not been victims of crime. Furthermore, we find that while arguments do not sway respondents’ opinions substantially, consequentialist arguments can offset the negative effect of human costs. Our findings shed light on the “public opinion side” of militarization and have important implications for policy debates on how to tackle organized crime.

[OSF Pre-print here]

Media coverage:

- "Hugs, not Gunshots": How can ALMO convince Mexicans of a new approach to security. Political Violence @ a Glance October 18, 2018

- "Abrazos, no balazos": ¿Cómo puede AMLO conseguir apoyo popular para un cambio en la política de seguridad mexicana? Oraculus. February 12, 2019

Brazil's Fight Against Drug Syndicates

Assessing Support for conditional security policies 

[with Davide Morisi, Collegio Carlo Alberto]

Since Brazil's transition to democracy in the mid-1980s, drug-related violence has been a central threat to citizens' security, especially in the hundreds of favelas (slums) in the country's main cities. In an effort to fight drug syndicates controlling large portions of these cities, for over 30 years state governments have privileged an "iron-fist" approach based on unconditional crackdowns on criminal organizations. This approach has had dramatic human consequences – e.g., between 2003 and 2017 Rio's military police have killed over 13,000 civilians (Magaloni et al. 2018) – and has proved largely unsuccessful in both curbing drug trade and preventing violence. Contrary to this approach, recent research and policy experiments in US cities, show that conditional policies – i.e., state repression conditional on criminal groups' behavior – are likely to be more effective in bringing citizen security (e.g., Lessing 2018). Yet, despite their potential effectiveness, these policies have been rarely implemented.

Allegedly, this is the case because, unlike unconditional repression, conditional policies face strong “acceptability constraints” among the general public, making them politically “toxic” for politicians. By means of field work, surveys and experiments in various Brazilian cities – including Sāo Paulo and Rio de Janeiro, the country's two major cities –, we empirically examine: (i) whether it is the case that Brazilians are not willing to support conditional policies to combat crime; and, (i) under what conditions Brazilians are more likely to move away from support for “iron first” approaches and give conditionality a chance.

We will be fielding our first round of data collection in early 2020.


Institute of Political Science

Leiden University
Turfmarkt 99

2511 DP The Hague
Room  6.06

+31 70 800 3879

bottom of page