Funded by DFG (493809260)
SPARK analyses the short-term escalation dynamics in cultural identity conflicts on a global scale. Conflictive mass behaviour is
conceptualized as “cascades”, which are understood as propagations of self-organizing conflictive mass behaviour of varying intensity and extensity. In our model, an emboldening emotional climate
provides the “fuel” that is sparked by a triggering event. As an intermediate step, we expect that the activated potential translates into action through collective self-organization. By
combining triggering events, collective emotions and self-organization, SPARK investigates an innovative and comprehensive explanation of the non-linear short-term escalation dynamics of
collective mass behaviour in cultural identity conflicts. The project is structured in three work packages (WP). WP1 establishes the empirical database. In a second step (WP2), we subject the
model to a quantitative test. For this purpose, we measure collective self-organization and emotional climates by conducting network and sentiment analyses. The third step (WP3) consists of a
qualitative test of the model. Based on the statistical evaluation, we select typical and deviant cases for process tracing analysis. In cooperation with local case experts, conflict-specific
case studies are prepared and subjected to a systematic cross-case comparison.
This study investigates how different targets of state-sanctioned arrests shape the likelihood of collective action. We hypothesize that leader arrests are especially likely to result in backlash
protests. Leader arrests symbolize the suppression of social collectives, they create collective grievances, and constitute focal points for mobilization. Building on a global sample of arrests
of cultural identity group members, we qualitatively traced for each arrest whether it sparked a backlash protest. Drawing on coarsened exact matched models, we find that protests
are significantly more likely following leader arrests. In contrast, mass arrests are not significantly linked to backlash protests. Additional tests show that organizational membership does not drive this findings, whereas the symbolic value of leaders is linked to protest outbreaks. Our findings cast doubt on the narrow focus on quasi-constant structural variables and make the case for the disaggregation of repression and the importance of triggering events.