Justifications of democracy rest in large part on unacknowledged empirical assumptions regarding the cognitive, informational, and behavioral capacities of individuals and voting populations. The goal of this project was to identify those assumptions and examine them in light of data from the social sciences. To the extent that these assumptions are undermined by empirical evidence, the normative legitimacy of democracy as a system of rule is weakened. Theories of democracy were organized along a continuum from purely instrumental to purely intrinsic or procedural, and a representative sample of theories from along this spectrum were analyzed in order to identify their core empirical assumptions. Interest-based, deliberative, and egalitarian theories of democracy were each demonstrated to be predicated on substantive empirical assumptions which were contradicted by the available evidence. A sophisticated hybrid account incorporating aspects from along the spectrum of available theories was likewise demonstrated to be predicated on unsubstantiated assumptions regarding human capacities. A concluding analysis of the circumstances which undermine the assumptions of democratic theory demonstrated the limited tractability of these circumstances, leading to my assertion that a new conception of what democracy is and what purposes it should serve is warranted, and that in the interim, contemporary attempts to justify the dissemination of democracy are undermined.