Do Psychiatric Disorders Need Types?: Case Identity, Naturalism, and the Limits of Biological Grounding

Psychiatric classification has long struggled with a fundamental question: do diagnostic categories like “depression” or “schizophrenia” correspond with distinct diseases with unified biological causes, or are they pragmatic groupings of heterogeneous conditions? Recent reform proposals, exemplified by Steven Hyman’s influential work, reject the idea that current categories in the Diagnostic and Statistical Manual of Mental Disorders are natural kinds while maintaining that future research will eventually reveal the true biological structure of mental disorders. This article identifies a problematic pattern of reasoning in such proposals. I argue that reformist naturalism conflates two distinct requirements: the practical need to identify and track cases of “the same condition” across patients and studies, and the much stronger claim that all such cases share a single underlying biological mechanism. This conflation leads to what I call “promissory essentialism”: the assumption that psychiatric science must ultimately discover unified biological types, even as current evidence points toward heterogeneity. I show how this assumption inverts the appropriate burden of proof: instead of treating biological unity as a hypothesis to be tested, reformist programs treat it as a precondition of scientific legitimacy. Once this conflation is diagnosed, space opens for alternative approaches that take seriously the possibility that psychiatric categories may be irreducibly plural, grounded in overlapping mechanisms rather than unified essences.