Abstract

There is evidence that mental illness is partly socially constituted: diagnoses are historically “transient” (Hacking, _Rewriting the soul: Multiple personality and the sciences of memory_. Princeton University Press, 1998a; _Mad travelers_. University of Virginia, 1998b) and culturally variable (Toh, _Nature Reviews Psychology_, _1_(2), 72–86, 2022). However, this view risks pernicious relativism. On most social constitution views, mental illness is what (some suitably expert part of) society takes it to be. But this has morally abhorrent implications, e.g., it legitimizes many spurious and harmful diagnoses (like “drapetomania”) and makes mental illness a matter of fashion rather than an objective challenge. This paper defends a conception of mental illness according to which it is partly socially constituted, yet which avoids such pernicious relativism: mental illness consists in an objective inability - a deficit in the skilled metacognitive self-regulation required to be rationally interpretable by one’s community, including oneself. Such reasons responsiveness requires skilled regulation of cognition, conation, and behavior, such that they respect relevant interpretive norms. Because such norms vary culturally, such skills are partly socially constituted.