Kamil Bernaerts

It is often claimed that we are living in an age of increasing polarization. Political views, opinions, and worldviews become increasingly irreconcilable (idea-based polarization), while at the same time society appears to be getting fractured in antagonistic, opposing camps (identity-based polarization). However, a closer look at international datasets reveals that these forms of polarization do not affect all democracies to the same extent. Levels of identity-based and idea-based polarization strongly vary across countries. The question then becomes what can explain these diverging levels of polarization. In this paper, we hypothesize that the institutional design of a country impacts polarization, and that consensus democracies would display lower levels of polarization. Based on a quantitative analysis of the Comparative Political Dataset and Varieties of Democracy data in 36 countries over time (2000–2019), our results show that institutions did matter to a great extent, and in the hypothesized direction. Countries with consensus institutions, and more in particular PR electoral systems, multiparty coalitions, and federalism did exhibit lower levels of both issue-based and identity-based polarization, thereby confirming our expectations. Moreover, we found that consensus democracies tend to be better at coping with identity-based polarization, while the effect on idea-based polarization is smaller.

In recent years, deliberative democracy has drawn attention as a potential way of fighting polarization. Allowing citizens to exchange arguments and viewpoints on political issues in group, can have strong conflict-mitigating effects: it can foster opinion changes (thereby overcoming idea-based polarization), and improve relations between diametrically opposed groups (thereby tackling affective forms of polarization, such as affective polarization). However, these results conflict with social psychological and communication studies which find that communicative encounters between groups can lead to further polarization and even group think. The question therefore arises under which conditions deliberative interactions between citizens can decrease polarization. Based on a multidisciplinary systematic review of the literature, which includes a wide diversity of communicative encounters ranging from short classroom discussions to multi-weekend citizen assemblies, this paper reports several findings. First, we argue that the effects of communicative encounters on polarization are conditional on how those types of communication were conceptualized across disciplines. More precisely, we find depolarizing effects when group discussions adhere to a deliberative democracy framework, and polarizing effects when they do not. Second we find that the depolarizing effects depend on several design factors that are often implemented in deliberative democracy studies. Finally, our analysis shows that that much more work needs to be done to unravel and test the exact causal mechanism(s) underlying the polarization-reducing effects of deliberation. Many potential causal mechanisms were identified, but few studies were able to adjudicate how deliberation affects polarization.