Law statements and the role of kinds

Scientific theories describe causal properties, concrete structures, and particulars such as objects, events, and processes. Semirealism maintains that under certain conditions it is reasonable for realists to believe that the best of these descriptions tell us not merely about things that can be experienced with the unaided senses, but also about some of the unobservable things underlying them. In charting the proposed conceptual foundations of this stance I have covered a lot of ground, from causal processes through de re necessity and now to one last crucial concept. There is one item on the inventory of realist commitment about which I have not yet had much to say. In addition to theorizing about and experimenting on instances of properties and relations, the sciences also describe kinds of particulars. Earlier I said that instances of causal properties regularly cohere to form units that are especially apt for scientific study, and it is precisely these groupings that make up the particulars described by theories. The time has come, finally, to consider the particulars of scientific discourse.

The idea of kinds of particulars is perhaps more in need of clarification than any other aspect of the metaphysics of semirealism. For here more than in any other place, scientific realists have allowed their position to become unfortunately entangled with the metaphysical speculations of past, systematic philosophers.