Book excerpt: It is often argued that democratic institutions should be designed to produce good outcomes, assuming that we know what good outcomes are and which institutions will track them. This book denies both assumptions. The idea of the general interest is ill-defined and our understanding of social causality is fragile. Instead, one should reduce as much as possible the impact of self-interest, passion, prejudice, and bias on the decision makers, and then let the chips fall where they may. In addition to making novel theoretical proposals, this book discusses a welter of case studies and historical episodes.