Deliberation and polarization: a multi-disciplinary meta-analysis

Deliberative democracy has often been heralded as a cure for societal and political polarization. By allowing citizens to deliberate in group on salient political issues, democratic deliberation is claimed to have strong conflict-mitigating effects: deliberative mini-publics are often considered learning schools for democracy, in that they can foster both opinion and attitude changes (Grönlund, Bächtiger & Setälä, 2014; Smith & Setälä, 2018), and through deliberation citizens are claimed to develop more cosmopolitan, inclusive, and moderate political orientations (Gastil et al., 2010). Even though research on this topic has gained momentum in recent years, studies on the effects of deliberation on polarization are scattered across disciplines (social psychology, political science, communication studies), and their results are often contradictory. Cass Sunstein’s (2002, 2009) work for instance has systematically warned about the polarizing effects of deliberation, whereas other studies find that deliberation fosters moderation (see e.g. Grönlund et al. 2015; Strandberg et al. 2017). A closer look at the literature reveals some potential explanations for these conflicting findings. Deliberation is often implemented in multiple ways, ranging from small three-person laboratory experiments, over exchanging opinions on social media, to face-to-face townhall meetings including hundreds of individuals. Moreover, polarization is often conceptualized in dissimilar ways. Some focus on partisan or opinion polarization, whereas others study attitude or affective polarization. Finally, there is much contextual variation in terms of topics (high vs. low salience issues), settings (hot vs. cold deliberative settings), design (with or without experts or expert information), duration (ranging from a few hours to several weeks), goals (mere deliberation vs. formulating specific policy recommendations) and interaction type (online vs. face-to-face deliberation). In response, this paper aims to take stock of the research on deliberation and polarization across multiple disciplines. It will map the different conceptualizations of deliberation and polarization and offer a meta-analysis of the strength and direction of the effects, and the conditions under which deliberation could mitigate polarization.