An epistemic theory of democracy

Abstract

One attractive feature of democracy is its ability to track the truth by information aggregation. The formal support for this claim goes back to Condorcet’s famous jury theorem. However, the theorem has often been dismissed as a mathematical curiosity because the assumptions on which the theorem is based are demanding. Such quick dismissals tend to misunderstand the original theorem. They also fail to appreciate how Condorcet’s assumptions can be weakened to obtain jury theorems that are readily applicable in the real world. The first part of the book explains the original theorem and its various extensions and introduces results to deal with the challenge of voter dependence. Part II considers opportunities to make democracies perform better in epistemic terms by improving voter competence and diversity, by dividing epistemic labour, and by preceding voting with deliberation. In the third part, political practices are looked at through an epistemic lens, focusing on the influence of tradition, following opinion leaders or cues, and on settings in which the electorate falls into diverging factions. Part IV analyses the implications for the structures of government. While arguing against the case for epistocracy, the use of deliberation and expert advice in representative democracy can lead to improved truth-tracking, provided epistemic bottlenecks are avoided. The final part summarizes the results and explores how epistemic democracy might be undermined, using as case studies the Trump and Brexit campaigns.