Natural Kinds

Carlos Santana (University of Utah)
Mineral species are, at first glance, an excellent candidate for an ideal set of natural kinds somewhere beyond the periodic table. Mineralogists have a detailed set of rules and formal procedure for ratifying new species, and minerals are a less messy subject matter than biological species, psychological disorders, or even chemicals more broadly — all areas of taxonomy where the status of species as natural kinds has been disputed. After explaining how philosophers have tended to get mineralogy wrong in discussions of natural kinds, I show how mineral species don't behave like natural kinds. They are defined on the basis of human intentionality, not merely natural distinctions. They aren't ideal grounds for inductive inference. And they don't form a system that divides nature along a set of equivalent joints. While this is a regrettable outcome to those of us who like the idea of science relying on natural kinds, I contend that mineralogy is doing just fine without a natural kind-based taxonomy, and may in fact be better off without one.

Jonathan Tsou (Iowa State University)
This paper addresses the question of how human science categories yield projectable inferences through a critical examination of Ron Mallon's 'social role' account of human kinds. Mallon contends that human categories are causally significant and projectable when a social role produces a homeostatic property cluster (HPC) kind. His account suggests that a human category is projectable when certain social mechanisms serve to stabilize and entrench that category. Contra-Mallon, I argue that human science categories are projectable when a human classification individuates an HPC kind underwritten by natural (i.e., biological) mechanisms. My argument suggests that there is a distinction to be drawn with respect to HPC kinds maintained by social versus natural mechanisms and that only the latter will yield robust projectable inferences. Classifications from psychiatry ('schizophrenia,' 'bipolar disorder,' 'hysteria') are discussed as examples.

Kikyung Lee (University of Pennsylvania)
The aim of this paper is to make a positive contribution to Slater’s stable property cluster account of natural kinds via two endeavors. First, I argue that Slater’s account is better than its two main rival theories (homeostatic property cluster theory and bare projectibilism) because it has both explanatory strength and generality. However, this is not to say that Slater’s account is fully tenable in its current formulation. Hence, my second endeavor is to strengthen the theory by suggesting a refinement that can protect the theory against a compelling counterexample from biochemistry: enantiomers. 

Devin Gouvea (University of Chicago)
Recent philosophical work on biological homology has generally treated its conceptual fragmentation as a problem to be solved by new accounts that either unify disparate approaches to homology or specify sharp constraints on its meaning. I show that several proposed solutions either misunderstand or ignore central features of comparative biological research, despite attempts to capture scientific practice. I conclude that the problem is incorrectly framed and that disagreements about homology may be epistemically fruitful. Empirically tractable debates are more likely to occur among biologists who share theoretical perspectives on homology. Philosophers should consider homology not merely as a generator of inductive generalizations but also as a scaffold for meaningful empirical comparisons.