Social Media and the Pursuit of Power in a Hybrid Regime: Evidence from Pakistan’s 2018 Election

Speaker(s)

Asfanydar Mir, Tamar Mitts & Paul Staniland

Date and Time

November 18, 2020 11:30 AM - 12:30 PM

Availability

RSVP Required.

Location

Virtual Seminar

* Please note all CISAC events are scheduled using the Pacific Time Zone.

 

Register in advance for this webinar: https://stanford.zoom.us/webinar/register/3816025578507/WN_AOSMVQyHT5OAlQ29pxeLsw

 

About the Event: Research on social media and politics has largely focused on two very different political contexts: authoritarian regimes and “normal” democratic polities. However, many countries’ political systems exist between these extremes: there is both “normal” online mobilization and efforts at manipulation that emanate in whole or in part from state-linked actors. In this article, we focus on a country with such a system: Pakistan. We investigate the politics of social media in the run-up to Pakistan’s 2018 general election. The campaign involved both intense, large-scale electoral mobilization and recurrent, credible allegations of influence by the country’s politically powerful army. We analyze millions of Twitter posts in English and Urdu by major political actors and their followers in Pakistan before and just after the 2018 election to identify patterns of 1) normal mobilization and 2) coordinated manipulation. Several findings emerge. First, the main political parties were highly active on social media, with the eventually-victorious Pakistan Tehreek-e-Insaf (PTI) of Imran Khan showing a noticeable edge in online enthusiasm that matches the substantial victory of the PTI in the election. Second, there was a noticeable “dissident sphere” on Twitter, seeking to get around a campaign of censorship and media influence by the military. However, dissidents’ messages were largely swamped by the broader party competition and narratives favorable to the PTI and the military. Third, we find evidence of coordinated activities. This appears to have largely favored the PTI and pro-military messages, which saw a substantially higher rate of amplification. Finally, we see evidence of narrative alignment between the PTI and the military – the clusters of their followers seemed to advance pro-PTI and anti-PML-N messages; pro-PTI and anti-PML-N narratives were pervasive in the PML-N and dissident clusters.

 

About the Speakers: 

Asfandyar Mir is a Postdoctoral fellow at the Center for International Security and Cooperation at Stanford University. His research interests are in international security with current work focusing on counterterrorism, counterinsurgency, drone warfare, US counterterrorism policy, South Asia security issues, misinformation dynamics, and Al-Qaida. Some of his research has appeared in peer-reviewed journals, such as International Security, International Studies Quarterly, and Security Studies. My commentary has appeared in Foreign Affairs, Foreign Policy, H-Diplo, Lawfare, and Washington Post Monkey Cage.

 

Tamar Mitts is Assistant Professor at the School of International and Public Affairs and a Faculty Member at the Data Science Institute and the Saltzman Institute of War and Peace Studies at Columbia University. Her research applies machine learning and text analysis methods to study political behavior in the digital age, and has been published in the American Political Science ReviewInternational Organization, the Journal of Economic Perspectives, and Political Science Research and Methods, among other outlets.

 

Paul Staniland is Associate Professor of Political Science at the University of Chicago and a nonresident scholar in the South Asia Program at the Carnegie Endowment for International Peace. His research focuses on political violence and international security in South Asia. Staniland’s first book, Networks of Rebellion: Explaining Insurgent Cohesion and Collapse, was published by Cornell University Press in 2014, and his second book, Armed Politics: Violence, Order, and the State in South Asia, will be published by Cornell in 2022.