David Estlund

The papers published in this special issue can fairly be unified under the heading “Epistemic Democracy,” but there is more variety among them than this might indicate. They exhibit the broad range of ways in which epistemological considerations are figuring in contemporary philosophical discussions of democracy. The authors range from young and promising to established and distinguished. I'd like to introduce a few of the issues that run through the papers, sprinkling references to the actual papers along the way. From the beginning, democratic forms of government have included discussion and debate. In real life the value of democracy can hardly be separated from the value of free public discussion, prior to voting, about the issues and candidates. This is not to say that either the discussion or the vote have always been inspiring, but whatever value democracy is thought to have, it seems inseparable from public political discussion. One way of accounting for the value of the discussion is to suppose that voters exchange reasons (not always cooperatively) about what to do. Even a quick look at the content of political debate seems to confirm that it is mostly about which decision would be best for the country or city whose laws or leaders are in question

Under the assumptions of the standard Condorcet Jury Theorem, majority verdicts are virtually certain to be correct if the competence of voters is greater than one-half, and virtually certain to be incorrect if voter competence is less than one-half. But which is the case? Here we turn the Jury Theorem on its head, to provide one way of addressing that question. The same logic implies that, if the outcome saw 60 percent of voters supporting one proposition and 40 percent the other, then average voter competence must either be 0.60 or 0.40. We still have to decide which, but limiting the choice to those two values is a considerable aid in that. Key Words: Condorcet Jury Theorem • epistemic democracy • voter competence.

There is a growing literature under the banner of "deliberative democracy," and Paul Weithman suggests that much of it is based on, or at least implies, a critique of the kind of theory of justice pioneered by Rawls 1. The issue at stake is whether a democratic political theory can admit independent normative standards that apply to and constrain democratic decisions. A certain kind of critic thinks independent standards are anti-democratic. Weithman's defense of Rawlsian theory against this charge is, I believe, instructively mistaken. Rawlsian theory escapes the charge but not for Weithman's reasons. This dispute might reasonably hope to shed light on the interrelations not only between democracy and justice, but also between these and political legitimacy.

Sunstein argues that democratic theory has recently rested its normative claims on a vast but empirically uninformed optimism about the ability of collective deliberation to lead to morally and rationally better decisions. Once that question is considered empirically, he argues, deliberation turns out to be mixed at best, and a disaster at worst. I want to suggest that Sunstein exaggerates the claims of the deliberative democrats, and interprets the empirical literature against deliberation in a way that appears, even based on his own descriptions of the studies, to be unfairly biased against the value of deliberation.

Waldron argues that recent treatments of justice have neglected reasonable disagreement about justice itself. So Waldron offers a procedural account of democratic legitimacy, in which contending views of justice can be brought together to arrive at a decision without deciding which one is correct. However, if there is reasonable disagreement about everything, then this includes his preferred account of legitimacy. On the other hand, it is not clear that Waldron is right to count so much disagreement as reasonable. But then Waldron has not undermined the view he opposes in which some prevailing disagreement about justice is held to be unreasonable