The implicit priorities of political science have been inhospitable to policy evaluation as a scholarly endeavor. Political science, at least as practiced in the West, draws from a philosophic tradition that views political life, and especially democratic political life, as a public objective, not principally as instrumental for other social ends. Comparative policy analysis deviated from that tradition by treating political conditions as independent variables. Policy evaluation, which assesses the social consequences of governmental actions, deviated even further, adopting an explicitly and totally instrumental approach. The argument ofthis essay is that the tactics of policy evaluation should be adopted for assessing the impact of policy patterns on democratic institutions and citizenship. This recommendation flows from a diagnosis of the division of labor in policy evaluation and an identification of the optimum entry points for political science.