Behind Closed Doors: Publicity, Secrecy, and the Quality of Deliberation

Abstract

T HERE are many reasons to keep secrets.Some of these reasons will inevitably be better than others.With regard to public policy in a liberal democracy, all reasons for secrecy deserve some scrutiny and justification.Liberal democracy contains a presumption in favor of publicity although it in no way denies the necessity of secrecy.One of the more intriguing reasons for secrecy invokes it for the sake of publicity.This is what theorists are doing when they defend secrecy on the grounds that it enhances deliberation.When theorists argue that the "glare" of publicity makes it difficult to argue reasonably, they imply that deliberation in secret will result in more publicizable outcomes.1 This appears to contradict a cornerstone of deliberative democratic theory: secrecy is fertile ground for private or non-public reason, while public reason flourishes in the open spaces of a public sphere.In what follows, I outline this contradiction in more detail and offer a typology of public reason to sort out the positive effect of publicity from its harmful effect.I argue that the distinction between public reason and private reason has dominated debate within deliberative democratic theory to such an extent that the harmful effects of publicity have been under-theorized.In order to understand just what goes wrong with deliberation under the glare of publicity, I develop a concept of plebiscitory reason.Private reason fails a democratic accountability test.Plebiscitory reason fails a Socratic accountability test.Private reasons are not general enough to appeal to all those affected while plebiscitory reasons are too poorly argued, shallow or manipulative to serve as acceptable justifications of public policy.The question that should concern theories of deliberative democracy is: how can we mitigate or contain plebiscitory reason in the public sphere?The corollary to this is: how can we contain private reason behind closed doors?With regard to the first question, I argue that a clear