Putnam, Hilary (1926–2016)

3. The meaning of meaning

Putnam faults the traditional theory of meaning for being individualistic rather than social, and for neglecting the contribution of external reality to meaning. He construes the standard theory as based on the two assumptions that to know a term’s intension is to be in a particular psychological state and that intension determines extension. Hence, it is argued, speakers who are in the same mental state when uttering a word share both its intension and its extension.

Putnam uses the twin earth (TE) thought experiment to attack this theory. TE resembles earth down to the smallest detail except that the liquid functioning as water, and called ’water’, on TE is not ${\mathrm{H}}_2\mathrm{O}$ , but a different chemical compound. Since there is no reason to ascribe different mental states to a person using ’water’ on earth, and their counterpart on TE, or, at least, there was no such difference prior to the emergence of chemistry as a science, this constitutes an example of people using the same word when in identical mental states, but ascribing different meanings to it: one refers to water, the other to the TE equivalent. One need not, however, travel as far as TE to find examples of similar mental states differing in extension. Putnam testifies that he is unable to tell a beech from an elm: his mental image of the two is the same, yet the extensions, and the meanings, of ’elm’ and ’beech’ are quite different in his idiolect. Can Putnam refer to an elm though unable to identify one? His answer introduces the concept of the division of linguistic labour: it is sufficient that experts can distinguish an elm from a beech, and unnecessary that each member of the linguistic community be able to do so.

The assumption underlying the TE example is that words like ’water’ always refer to the stuff we call ’water’ in the actual world. Though it is neither analytic nor even irrevisable that water is ${\mathrm{H}}_2\mathrm{O}$ , the extensions of ’water’ on TE and on earth cannot be identical. To complete the account of meaning, one has to address the question of how extension is actually fixed. These aspects of Putnam’s conception are related to Kripke’s work on reference and rigid designation (see Kripke, S.A.; Proper names; Reference §2), but while Kripke was thinking mainly of proper names, Putnam’s concern is the meaning of scientific terms. If the extensions of these terms are fixed by the theories in which they figure, then extensions will be liable to change with theoretical change. This argument can be taken as an encapsulation of Kuhn’s relativism: different theories refer to different entities, and are therefore incommensurable (see Incommensurability). Viewing it as a reductio ad absurdum of the theory of meaning on which it rests, Putnam recommends replacement of this theory with a variant of Kripke’s causal theory of reference that emphasizes not only the causal relation between speakers and what they refer to, but also such social and pragmatic factors as shared stereotypes, reliance on experts, discretion and charity (see Content: wide and narrow).