Democracy isn't that smart (but we can make it smarter): On landemore's democratic reason

Abstract

In her recent book, Democratic Reason, Hélène Landemore argues that, when evaluated epistemically, a democratic decision procedure is likely to be a better decision procedure than any non-democratic decision procedures, such as a council of experts or a benevolent dictator (p. 3). Landemore's argument rests heavily on studies of collective intelligence done by Lu Hong and Scott Page. These studies purport to show that cognitive diversity - differences in how people solve problems - is actually more important to overall group performance than average individual ability - how smart the individual members are. Landemore's argument aims to extrapolate from these results to the conclusion that democracy is epistemically better than any non-democratic rival. I argue here that Hong and Page's results actually undermine, rather than support, this conclusion. More specifically, I argue that the results do not show that democracy is better than any non-democratic alternative, and that in fact, they suggest the opposite - that at least some non-democratic alternatives are likely to epistemically outperform democracy.