I raise a puzzle case for “cluster” accounts of natural kinds—the homeostatic property cluster and stable property cluster accounts, especially—on the basis of their expected treatment of the metaphysics of certain disease kinds. Some kinds, I argue, fail to exhibit the co-instantiated property clusters these cluster views take to be constitutive of natural kinds. Some genetic diseases, for example, have archetypical instances with few or none of the pathological processes or symptoms associated with the kind: their instances are typified by (...) a single dispositional property. I dub such kinds ‘amorphic’, owing to their limited morphology, and try out a number of ways in which these kinds might be treated in terms of property clusters, adapting responses cluster theorists have offered to the problem of polymorphic species. Finding these responses wanting, I conclude that cluster accounts are unlikely to be the best account of the metaphysics of amorphic kinds. (shrink)

This paper offers an original solution to the problem of the many, built on a foundation of powers-based causation. At its most basic, the solution should be understood as a type of maximality response, and on those grounds its originality might be questioned. However, it is argued that novelty of the solution owes as much to the meta-metaphysical context in which the solution is framed as it does the model of causal powers. A discussion of paradigms in metaphysics is included.