Radicalization and Violent Extremism

From Isolation to Radicalization: Anti-Muslim Hostility and Support for ISIS in the West. Published in the American Political Science Review. [Paper] [Ungated] [Appendix] [Replication]

What explains online radicalization and support for ISIS in the West? Over the past few years, thousands of individuals have radicalized by consuming extremist content online, many of whom eventually traveled overseas to n join the Islamic State. This study examines whether anti-Muslim hostility might drive pro-ISIS radicalization in Western Europe. Using new geo-referenced data on the online behavior of thousands of Islamic State sympathizers in France, the United Kingdom, Germany, and Belgium, I study whether the intensity of anti-Muslim hostility at the local level is linked to pro-ISIS radicalization on Twitter. Results show that local-level measures of anti-Muslim animosity correlate significantly and substantively with indicators of online radicalization, including posting tweets sympathizing with ISIS, describing life in ISIS-controlled territories, and discussing foreign fighters. High-frequency data surrounding events that stir support for ISIS — terrorist attacks, propaganda releases, and anti-Muslim protests — show the same pattern.

Countering Violent Extremism and Radical Rhetoric. Published in International Organization. [Paper] [Ungated] [Appendix] [Replication]

How do extremist sympathizers respond to counter-radicalization efforts? Over the past decade, programs to counter violent extremism have mushroomed around the world, but little is known of their effectiveness. This study uses social media data to examine how counter-radicalization efforts shape engagement with extremist groups in the online world. Matching geolocated Twitter data on Islamic State sympathizers with granular information on counter-extremism activities in the U.S., I find that, rather than deradicalizing, these efforts led ISIS sympathizers to act strategically in order to avoid detection. After counter-extremism activities, the group’s supporters on Twitter who were in the vicinity of these events began self-censoring expressions of support for ISIS, altered profile images and screen names, and encouraged followers to migrate to Telegram – an encrypted network not viewable to the public. These findings reveal previously unknown patterns on the effects of counter-extremism programs in the digital era.

Studying the Impact of ISIS Propaganda Campaigns. With Barbara Walter and Gregoire Phillips. Forthcoming in the Journal of Politics. [Paper] [Ungated] [Appendix] [Replication]

Over the past decade, a large number of extremist and hate groups have turned to internet platforms to inspire mass violence. Currently, there is little reliable evidence on how such campaigns radicalize targeted audiences. We provide systematic, large-scale, micro evidence on the effect of Islamic State propaganda on social media. We use several machine learning algorithms to detect recruitment messages in online propaganda, identify their dissemination on Twitter, and quantify the reactions of exposed users. Analyzing content produced by the Islamic State between 2015 and 2016 shows that propaganda conveying the material, spiritual, and social benefits of joining ISIS increased online support for the group, while content displaying brutal violence decreased endorsement of ISIS across a wide range of videos. Only the group’s most extreme supporters reacted positively to violent propaganda.

Banned: How De-platforming Extremists Mobilizes Hate in the Dark Corners of the Internet. In Progress. [Paper]

In recent years, the world has seen a rapid increase in the use of social media platforms by violent extremists. Hate groups espousing radical ideologies have been using online platforms to communicate, disseminate propaganda, and in some cases, plan violent acts. In response, social media companies have upped their efforts to take down content and prevent the spread of hate speech on their platforms. While these actions reduced the availability of extremist content on mainstream social media, little is known about what happens to suspended individuals after being deplatformed. This project sheds light on the effects of deplatforming among online communities affiliated with the far-right in the United States. Analyzing cross-platform data that includes information on individuals who have accounts both on Twitter (a mainstream platform) and Gab (a fringe platform favored by far-right extremists), I find that Twitter suspensions increase engagement with hate speech on Gab. I discuss several approaches that can help mitigate radicalization on fringe platforms.  

The Causes and Consequences of Violent and Nonviolent Conflict

Effective for Whom? Ethnic Identity and Nonviolent Resistance. with Devorah Manekin. Forthcoming in the American Political Science Review. [Paper] [Ungated] [Appendix] [Replication]

A growing literature finds that nonviolence is more successful than violence in effect- ing political change. We suggest that a focus on this association is incomplete, because it obscures the crucial impact of ethnic identity on campaign outcomes. We argue that because of prevalent negative stereotypes associating minority ethnic groups with violence, such groups are perceived as more violent even when resisting nonviolently, increasing support for their repression, and ultimately hampering campaign success. We show that, cross-nationally, the effect of nonviolence on outcomes is significantly moderated by ethnicity, with nonviolence increasing success only for dominant groups. We then test our argument using two experiments in the U.S. and Israel. Study 1 finds that nonviolent resistance by ethnic minorities is perceived as more violent and requiring more policing than identical resistance by majorities. Study 2 replicates and extends the results, leveraging the wave of racial justice protests across the U.S. in June 2020 to find that white participants are perceived as less violent than Black participants when protesting for the same goals. These findings highlight the importance of ethnic identity in shaping campaign perceptions and outcomes, underscoring the obstacles that widespread biases pose to nonviolent mobilization.

The Politics of Allyship: Multiethnic Coalitions and Mass Attitudes Towards Protest. With Devorah Manekin and Yael Zeira. In Progress.

Recent work finds that nonviolent resistance by minority ethnic groups is perceived as more violent and requiring more repression than nonviolent resistance by ethnic majorities, reducing its impact and effectiveness. We ask whether allies – committed participants from majority groups – can mitigate these barriers, and at what cost to the movements’ core constituencies. We study this question in the context of Black Lives Matter, a movement that, in the summer of 2020, attracted unprecedented numbers of white participants. Employing an original survey experiment, we find that sizeable white presence at racial justice protests increases public approval for the protest, reduces perceptions of protester violence and reduces support for protest policing. We complement our experimental results with observational analysis of protest imagery shared in hundreds of thousands of tweets posted during the 2020 BLM protests. We find a substantial and significant association between the presence of whites in these images and user engagement, amplification, and positive sentiment. Taken together, these findings suggest that ally participation can be a powerful tool for countering the structural biases and obstacles facing minority group movements fighting for social and political change.

Can war foster cooperation?  With Michal Bauer, Christopher Blattman, Julie Chytilová, Joseph Henrich, and Edward Miguel. Journal of Economic Perspectives, 30(3). [Paper] [Ungated] [Appendix] [Replication]

In the past decade, nearly 20 studies have found a strong, persistent pattern in surveys and behavioral experiments from over 40 countries: individual exposure to war violence tends to increase social cooperation at the local level, including community participation and prosocial behavior. Thus while war has many negative legacies for individuals and societies, it appears to leave a positive legacy in terms of local cooperation and civic engagement. We discuss, synthesize and reanalyze the emerging body of evidence, and weigh alternative explanations. There is some indication that war violence especially enhances in-group or “parochial” norms and preferences, a finding that, if true, suggests that the rising social cohesion we document need not promote broader peace.

Terrorism and the Rise of Right-Wing Content in Israeli Books. Published in International Organization. [Paper] [Ungated] [Appendix] [Replication]

In the past few years the Western world has witnessed a rise in the popularity of right-wing political discourse promoting nationalistic and exclusionary worldviews. While in many countries such rhetoric has surfaced in mainstream politics only recently, in Israel, right-wing ideology has been popular for almost two decades. Explanations for this surge focus on Israeli citizens’ attitudinal change in the face of exposure to terrorism, but largely do not account for why such ideas remain popular over the long term, even after violence subsides. This study examines whether the long-lasting prominence of right-wing nationalistic politics in Israel is linked to the perpetuation of right-wing ideology in popular media. Analyzing the content of more than 70,000 published books, this study finds that content related to the political right has increased in Israeli books after periods of terrorism, a change that has become more pronounced over the years. 

Contested Ground: Disentangling Material and Symbolic Attachment to Territory. With Guy Grossman and Devorah Manekin. Published in Political Science Research and Methods.  [Paper] [Ungated] [Appendix] [Replication]

Territorial disputes are prone to conflict because of the value of territory to publics, whether due to its strategic and material worth, or to its intangible, symbolic value. Yet despite the implications of the distinction for both theory and policy, empirically disentangling the material from the symbolic has posed formidable methodological challenges. We propose a set of tools for assessing the nature of individual territorial attachment, drawing on a series of survey experiments in Israel. Using these tools, we find that a substantial segment of the Jewish population is attached to the disputed West Bank territory for intangible reasons, consisting not only of far-right voters but also of voters of moderate-right and centrist parties. This distribution considerably narrows the bargaining space of leaders regardless of coalitional configurations. Our empirical analysis thus illustrates how the distribution of territorial preferences in the domestic population can have powerful implications for conflict and its resolution.

Media and Politics

How the Ultra-Rich Use Media Ownership as a Political Investment. With Guy Grossman and Yotam Margalit. Forthcoming in the Journal of Politics. [Paper] [Ungated] [Appendix]

Can the ultra-rich shape electoral results by controlling media outlets that openly propagate their political interests? How consumers discount slanted media coverage is a question gaining urgency as a growing number of billionaires mix ownership of major media outlets with business interests and political agendas. We study this question in the context of Israel, where billionaire Sheldon Adelson launched in 2007 Israel Hayom, a right-leaning newspaper. Handed out for free, it soon became the most widely read newspaper nationally. Utilizing local media exposure data since the launch, our analysis indicates that the newspaper exerted significant electoral influence, primarily benefiting Netanyahu and his Likud party. This shift helped bring about a sea-change in the right’s dominance of national politics. The findings highlight the immense impact the ultra-rich can exert in shaping politics through media ownership.

Terrorism as a Stage for Far-Right Mobilization. [Paper]

Research on the electoral success of far-right parties has traditionally focused on demand-side explanations such as voters’ grievances or supply-side accounts highlighting opportunities. So far, there has been little empirical research on contexts that facilitate the interaction of the two. This paper shows that terrorist attacks move both demand and supply in favor of right-wing populism, making aspects of its ideology resonate with the media and other drivers of the social conversation while simultaneously increasing public sympathy for far-right narratives. Using multiple sources of textual data from national newspapers and social media published before and after terrorist attacks in Europe, I show that (i) news media in countries targeted by terrorism cover more topics related to far-right ideology; (ii) far-right politicians talk more about immigration, security, and national identity in these time periods; and (iii) far-right messages become popular among the public in the month after terrorist acts.

Regulating the Marketplace of Ideas. With Jack Snyder, In Snyder, Jack., Human Rights for Pragmatists. Princeton University Press.  [Google Survey]

Freedom of speech and the press are at the center of any democratic system. Especially in America, absolute free speech is believed to facilitate the discovery of truth through constructive public debate, regardless of the context. But as free speech has been expanding throughout the world on an unprecedented scale—largely thanks to rapid innovations in information and communication technologies—the dissemination of false information, hate speech, and polarization has accelerated. We argue that these outcomes do not reflect any basic flaw in liberal free speech principles, but point to the necessity of supportive institutions and norms that enable free speech to yield socially beneficial consequences. We develop theoretical arguments that account for the mechanisms that link media freedom to political outcomes, and empirically illustrate the relationship between media freedom and these outcomes by drawing on cross-national data and case studies.

Political Coalitions and Social Media: Evidence from Pakistan. With Asfandyar Mir and Paul Staniland. Accepted at Perspectives on Politics.

Social media is frequently an arena of intense competition among political actors. We argue that a fruitful way of understanding this competition is as coalitions among actors and their networks of followers. These coalitions can both advance a shared political message and target mutual rivals. Importantly, coalitions can be tacit or explicit, and they do not necessarily require direct manipulation or state repression, though they often do. This makes a coalitional framework particularly valuable for studying complex political environments with multiple actors blending cooperation and competition. Empirically, we show the value of the approach with novel data collection and analysis of Twitter and Facebook content from 2018-19 Pakistan, with a focus on the dynamics leading up to and after the controversial 2018 general election. We map out networks of narrative alignment and conflict on Pakistani social media, providing important insights into the relationships among the major political parties, military, media, and dissidents. Future research on social media can explore this intricate, often-murky world by examining dynamics of online alignment, competition, and coalition-building.

Removal of Anti-Vaccine Content Impacts Social Media Discourse. With Nilima Pisharody and Jacob Shapiro. In 14th ACM Web Science Conference 2022 (pp. 319-326).  [Paper]

Over the past several years, a growing number of social media platforms have begun taking an active role in content moderation and online speech regulation. While enforcement actions have been shown to improve outcomes within moderating platforms, less is known about possible spillover effects across platforms. We study the impact of removing groups promoting anti-vaccine content on Facebook on engagement with similar content on Twitter. We followed 160 Facebook groups discussing COVID-19 vaccines and prospectively tracked their removal from the platform between April and September 2021. We then identified users who cited these groups on Twitter, and examined their online behavior over time. Our findings from a stacked difference-in-differences analysis shows that users citing removed Facebook groups promoted more anti-vaccine content on Twitter in the month after the removals. In particular, users citing the removed groups used 10-33% more anti-vaccine keywords on Twitter, when compared to accounts citing groups that were not removed. Our results suggest that taking down anti-vaccine content on one platform can result in increased production of similar content on other platforms, raising questions about the overall effectiveness of these measures.