Twin Earth thought experiment

It is meant to serve as an illustration of his argument for semantic externalism, or the view that the meanings of words are not purely psychological.

In its place there is a liquid that is superficially identical, but is chemically different, being composed not of H2O, but rather of some more complicated formula which we abbreviate as "XYZ".

Searle suggests that in the Twin-Earth example, the second seems more plausible, since if Twin Earth doesn't have water, then all its water-based products will also be different.

Here, privileged self-knowledge is taken to be the idea that one can know the content of one's thoughts without having to investigate the external world (for empirical evidence).

Although this type of argument does not directly disprove externalism, it is pressing because of the intuitive plausibility of privileged self-knowledge's existence.

Phil Hutchinson,[3] for example, notes that a) if one looks at Putnam's own later criticisms of others (for example his criticisms of Jaegwon Kim in his book The Threefold Cord) one finds that implicitly he criticises his own earlier self; and b) that the persuasive power of the Twin Earth thought experiment/intuition pump relies on our turning a blind eye to aspects of the experiment in order that it establish that which Putnam claims it to.

Phil Hutchinson has since argued that this concession to McDowell means that the distinction which Putnam wishes to operationalize, between intension and extension, is now problematized.

[6] John Dupré, in a number of papers and mainly in his paper "Natural Kinds and Biological Taxa",[7] has demonstrated that the theory of natural kinds, which many have taken to be established and/or supported by Putnam's Twin Earth thought experiment does not find support in the practice of scientific classification.

Avrum Stroll has produced probably the most comprehensive critique of the program of natural kind semantics (both Putnam's and Kripke's) in his book Sketches of Landscapes.