PhD thesis defence by Toine Fiselier
Rising polarization is widely regarded as a threat to democracies, yet the empirical evidence remains contested. This dissertation asks how different forms of polarization shape citizens' relationship with their political system: whether vertical affective polarization erodes mid-level political support, whether citizens prefer moderate ideological polarization over both convergence and extremes, and what drives interpersonal political evaluations. The approach is theory-driven with empirical analyses leveraging the internal validity of survey experiments and the external validity of panel data.
The first study draws on Easton's political support framework, expressive partisanship theory, and the competing account of partisan affect to hypothesise that affective polarization is less corrosive than assumed. Using British Election Study panel data, the study decomposes vertical affective polarization into in-party warmth and out-party hostility. Increases in affective polarization are positively associated with mid-level political support, but this association runs entirely through in-party warmth. The second study advances a sociotropic argument for moderate ideological polarization grounded in four normative standards (choice, representativeness, governability, and accountability). Combining a survey experiment with panel data analysis, it shows that citizens prefer moderate party system polarization over both convergence and extremes, independently of their ideological proximity to any party. The third study investigates the mechanisms driving horizontal affective polarization, juxtaposing social identity theory with belief congruence and social homophily in a pre-registered vignette experiment across six European democracies. Policy disagreement, more than partisan identity, drives negative evaluations of fellow citizens, while sociodemographic similarity sustains positive ones.
Taken together, these findings suggest that the democratic impact of any polarization score depends on the form of polarization and the mechanisms driving it. The alarm is warranted for some forms but overstated for polarization as such. By bundling components with different and sometimes opposing democratic consequences into a single concept, the literature inflates the threat it sets out to measure. Future research should therefore separate these components in measurement and theory, and reconsider when to sound the democratic alarm.
Toine Fiselier is a PhD researcher in the Department of Political and Social Sciences at the European University Institute, supervised by Filip Kostelka. His dissertation asks when political polarization actually threatens democracy, and finds that much of what is measured as polarization reflects attachment and substantive disagreement more than hostility. His work has appeared in the European Journal of Political Research, Election Law Journal and Acta Politica. He is currently a postdoctoral researcher at ESSEC Business School, where he studies how European democracies can better withstand strategic influence operations, from electoral interference to disinformation, within the interdisciplinary R-S3i project. Before the EUI, he completed an MSc in Comparative Politics at Radboud University.
Register