Diffusion of Violent Political Protest - Geographic Evidence from Bangladesh
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Theories on authoritarian resistance and violent collective action posit that individuals are likely toengage in protest behavior when they assess that: (i) politically organizing will achieve their objectives;and (ii) the gains received will outweigh the risks of punishment. Initially, citizens have incompleteinformation to make these judgments successfully. To obtain the necessary information, protesters rely onexternal sources, such as news or social media networks. Most importantly, however, is the informationgenerated from neighbors and other nearby citizens.Social proof, where individuals seek optimal decision-making in times of uncertainty by observing theactions of others, is significant where criminal laws may be violated. Individuals’ apprehensions regardinggovernment reprisal are mitigated when observing successful protests with analogous law enforcementconditions and comprising similar participants. Previous studies have demonstrated that geography exertsa pivotal influence on the diffusion of protests.Despite increasing use of communication technologies, I hypothesize that geography remains animportant predictor of violent protest diffusion, specifically due to the ongoing necessity for social proof.This study provides both a substantive and methodological contribution. Substantively, this study analyzesa country that is in the process of democratic backsliding at an exponential rate, which is not consideredin other papers on protest diffusion. Methodologically, I employ a novel Hawkes process model to assessviolent protest data from Bangladesh and map the spatio-temporal clustering of events. Unlike prior uses of Hawkes processes in protest diffusion, this model accounts for continuous spatial and temporaldependencies simultaneously, avoiding the artificial boundaries imposed by traditional spatial weightmatrices.Ultimately, the findings demonstrate the enduring significance of geographic proximity in the diffusionof violent protests, highlighting the need for a nuanced understanding of the underlying mechanisms. Theanalysis unveils distinct patterns of diffusion at varying geographical scales, underscoring the importanceof administrative levels in discerning baseline intensities, triggering effects, and temporal decay rates.