Natural Kinds: Rosy Dawn, Scholastic Twilight by Ian Hacking.
Basically and simply: there are no such things as “natural kinds.” Despite the fact that the concept has an endowed history and has functioned well for certain conceptual purposes (eg. as a mode of clear-cut classifying such as is required by the scholastic tradition), the concept is so theoretically saturated by varieties that it is now impotent.
The concept of natural kinds has not denigrated simply because continued research has grown the theory exponentially causing the plethora of strains to overshadow the original seeded theory and thus requiring a higher level theoretical noose. Rather, according to Hacking, the concept of natural kinds has within its essence a self-destructing thread.
Somehow, the hierarchical structuring of natural kinds (species, genera. families, orders) requires an “artifical” construction of grouping practices which precludes the possibility of natural kinds being natural in the first place. In this paper there is no reasoning as to why, when something is constructed, it is then not natural. This natural category is always problematic.
“Kinds” then are those entities which can be grouped into some set with a specific name. Thus, natural kinds is a a nomilistic convention.