Abstract
This article asks whether an interdisciplinary "critical science studies" (CSS) is possible between a critical theory in the Frankfurt School tradition, with its commitment to universal standards of reason, and relativistic sociologies of scientific knowledge (e.g., David Bloor's strong programme). It is argued that CSS is possible if its practitioners adopt the epistemological equivalent of Rawls's method of avoidance. A discriminating, public policyrelevant critique of science can then proceed on the basis of an argumentation theory that employs an immanent standard of relevance, which is illustrated by drawing on Helen Longino's critique of behavioral theory.