This week the CPS reading group will be reading Risalatul Hukmi's article "Natural Kinds Pluralism Refined: A Meta-Level Defense". You can find the article attached to this email, the abstract below, and the zoom link and meeting info below that.
As always, we will meet on Wednesday, in room 6.71, from 10:30 to 11:30.
Cheers,
Tom
Abstract:
The collapse of essentialist theories of natural kinds has left philosophy of science facing a crucial choice: eliminativism or pluralism. I defend a third option, a refined pluralism. I argue that we should keep the natural kind category, but only as a meta-level epistemic norm that coordinates heterogeneous classificatory practices. On my view, natural kinds are not unified by a common ontology but by satisfying a principle of generality: they are classifications built and maintained so that their inferential roles and evidential standards can be exported, translated, and integrated across distinct contexts of inquiry. This principle is more than sociological description. First, I diagnose the limitations of meta-monism, meta-eliminativism, and merely descriptive meta-pluralism. Second, I articulate generality as a necessary and defeasible criterion that explains why some classifications count as scientific kinds while others remain merely local. Third, I model coordination in terms of the preservation of inferential roles under cross-context uptake, and show how worldly structures constrain which classificatory projects can successfully meet the generality demand. The resulting position is a pluralist realism: there is no single true theory of kindhood, but there are worldly regularities that make generality a substantive, non-optional constraint on scientific classification.
