1. The cases I have in mind are those where it is suddenly or gradually made clear that a “natural kind” is as much a product of our collective imagination as it is a feature of the world about us. We might suppose that our idea of fire originates in the natural element by that name (among the four “elemental” components of the natural order; the other three being earth, wind, and water). When we discover that this “element” is variously composed of other, materially prior, processes, and that there is indeed no single matter which “fire” identifies, the idea of fire can no longer be considered both simple and natural. One of these has to be abandoned in the face of a more complex natural science. Usually we abandon the notion that ‘fire’ is a simple idea. But we might as well abandon the notion that ‘fire’ is representative of a materially simple kind, process, or object, while maintaining that it is somehow experientially simple. There would then, in this case, be a mismatch between the natural order, and our representations of it.

2. See Fodor and Lepore, pp.1-5, for a detailed exposition of the terms.

3. Fodor and Lepore explain, “The atomistic tradition proceeds from the likes of the British empiricists, via such of the pragmatists as Peirce and James. The locus classicus is the work of the Vienna Circle, but see also the Russell of The Analysis of Mind. The contemporary representatives of this tradition are mostly model theorists, behaviorists, and informational semanticists. Whereas people in this tradition think that the semantic properties of a symbol are determined solely by its relations to things in the nonlinguistic world, people in the second tradition think that the semantic properties of a symbol are determined, at least in part, by its role in the language. Languages are, inter alia, collections of symbols; so, if what a symbol means is determined by its role in a language, the property of being a symbol is anatomic. This second tradition proceeds from the likes of the structuralists in linguistics and the Fregeans in philosophy [with the possible exception of Frege himself].” p.7.

4. Borrowing a euphemism from the Port-Royal Logic.

5. Fodor and Lepore, p.14.

6. Fodor and Lepore state: “The problem isn’t, notice, that if holism is true, then the conditions for belief identity are hard to meet; it’s that, if holism is true, then the notion of ”tokens of the same belief type“ is defined only for the case in which every belief is shared. Holism provides no notion of belief-type identity that is defined for any other case and no hint of how to construct one. But if there is no construal of the claim that two beliefs are tokens of the same type in cases where belief systems fail to overlap completely, how in such cases, are we to construe the notion of two beliefs being tokens of almost the same type?” p.19.

7. Fodor and Lepore, p.21.

Chomsky, Noam. Cartesian Linguistics: A Chapter in the History of Rationalist Thought, First Edition. 1966. Harper & Row, Publishers / New York.

Davis, Wayne. Meaning, Expression, and Thought. 1994. (Author’s Draft)

Donnellan, Keith. “Reference and Definite Descriptions” (1966), as reprinted in The Philosophy of Language, Second Edition (edited by A.P. Martinich). 1990. Oxford University Press / Oxford. pp.235-247.

Fodor, Jerry and Lepore, Ernest. Holism: A shopper’s guide. 1992. Basil Blackwell Ltd. (T.J.Press Ltd.) / Oxford, UK.

Kripke, Saul. “Speaker’s Reference and Semantic Reference” (1977), as reprinted in The Philosophy of Language, Second Edition (edited by A.P. Martinich). 1990. Oxford University Press / Oxford. pp.248-266.