Online research in philosophy

We talk as if there are natural kinds and in particular we quantify over them. We can count the number of elements discovered by Sir Humphrey Davy, or the number of kinds of particle in the standard model. Consequently, it looks at first sight at least, that natural kinds are entities of a sort. In the light of this we may ask certain questions: is the apparent existence of natural kinds real or an illusion? And if real, what sort of entity are natural kinds? Are they sui generis? Or can they be identified with or reduced to some other kind of entity? In this essay I shall look at possible reasons for asserting that either kinds are no sort of entity, or, if they are entities, their existence is equivalent to some fact not involving kinds. Richard Boyd seems to take the view that the apparent existence of natural kinds is an illusion.

Some things, argues Lewis, are just better candidates to be referents than others. Even at the cost of attributing false beliefs, we interpret people as referring to the most interesting kinds in their vicinity. How should this be accounted for? In section 1, I look at Lewis’s interpretationism, and the reference magnetism it builds in (not just for ‘perfectly natural’ properties, but for certain kinds of auxiliary apparatus). In section 2, I draw on (Field, 1975) to argue that what properties are reference magnetic may be an ultimately conventional matter—though in the Lewisian setting, there may be an objectively best conventional choice to make. But Lewis’s own account has implausible commitments, so in section 3 I consider variations and alternatives, all of which have problems. In section 4, I look in more detail at eligibility-based interpretationism that do not appeal to naturalness, arguing that there are credible metasemantic theories of this form.

The vision of natural kinds that is most common in the modern philosophy of biology, particularly with respect to the question whether species and other taxa are natural kinds, is based on a revision of the notion by Mill in A System of Logic. However, there was another conception that Whewell had previously captured well, which taxonomists have always employed, of kinds as being types that need not have necessary and sufficient characters and properties, or essences. These competing views employ different approaches to scientific methodologies: Mill’s class-kinds are not formed by induction but by deduction, while Whewell’s type-kinds are inductive. More recently, phylogenetic kinds (clades, or monophyletic-kinds) are inductively projectible, and escape Mill’s strictures. Mill’s version represents a shift in the notions of kinds from the biological to the physical sciences.

Natural kinds have been a constant topic in philosophy throughout its history, but many issues pertaining to natural kinds still remain unresolved. This paper considers one of these issues: the epistemic role of natural kinds in scientific investigation. I begin by clarifying what is at stake for an individual scientific field when asking whether or not the field studies a natural kind. I use an example from life science, concerning how biologists explain the similar body shapes of fish and cetaceans, to show that natural kinds play a central epistemic role in scientific explanations that cannot be delegated to other explanatory factors. A task for philosophy, then, is to come up with a theory of natural kinds that adequately accounts for the epistemic role of natural kinds in science. After having sketched the spectrum of available philosophical theories of natural kinds, I argue that none of the available theories adequately performs this task and that therefore the search is still open for a theory that does.

The paper discusses reference determination from the point of view of conceptual change in science. The first part of the discussion uses the homology concept, a natural kind term from biology, as an example. It is argued that the causal theory of reference gives an incomplete account of reference determination even in the case of natural kind terms. Moreover, even if descriptions of the referent are taken into account, this does not yield a satisfactory account of reference in the case of the homology concept. I suggest that in addition to the factors that standard theories of reference invoke the scientific use of concepts and the epistemic interests pursued with concepts are important factors in determining the reference of scientific concepts. In the second part, I argue for a moderate holism about reference determination according to which the set of conditions that determine the reference of a concept is relatively open and different conditions may be reference fixing depending on the context in which this concept is used. It is also suggested that which features are reference determining in a particular case may depend on the philosophical interests that underlie reference ascription and the study of conceptual change.

In this paper, I argue that the causal and description theories of natural kind term reference involve certain psychological elements. My main goal is to refine these theories with the help of empirical psychology of concepts, and to argue that the refinement process ultimately leads to the dissolution of boundaries between the two kinds of theories. However, neither the refined theories nor any other existing theories provide an adequate answer to the question of what makes natural kind terms rigid. To provide an answer to this question I conclude my paper by introducing a framework of a unified theory of natural kind term reference that is built on the empirical psychology of concepts.

Natural Kinds and Conceptual Change is a refreshingly direct book that challenges a range of orthodox views in the philosophy of science (especially biology), the philosophy of language, and metaphysics. Amongst these are the views that species are individuals rather than natural kinds; that scientists discover the essences of natural kinds; that the causal theory of reference has commonly-ascribed implications for realism and analyticity; that there is an unacceptable form of incommensurability entailed by descriptivism about reference; and that there are good grounds, familiar since Quine, for thinking that there is no distinction of significance to be drawn between changes in meaning and changes in theory. LaPorte argues against all of these claims, and if you are curious about just how he does it, then this is a book for you.