Semantic externalism(s) and semantic deference

We examine various forms of content externalism and the role semantic deference plays in each form.

We will distinguish four variants of externalism, Wittgensteinian, Burgean, Putnamian, and extended-indexicalist. These labels are indicative, and reflect only our preferred interpretation of the views of the above authors. We take it that Wittgenstein (1953) defended an early variant of ‘social externalism’, as captured by his thesis that “meaning is use”. Consequently, word meaning is distributed across the whole set of linguistic practices of a community of speakers and is never encapsulated in a precise definition in terms of necessary and sufficient conditions. Tyler Burge’s (1979) anti-individualism embodies a more recent variety of social externalism. Burge notoriously maintained that the meaning of the word arthritis is captured by the best description offered by contemporary scientific experts, a description that is not available to non-expert speakers, who defer to the experts for the meanings of the words they do not master completely. A third variety, ‘physical externalism’, which had been advocated a few years before by Saul Kripke (1972) and Hilary Putnam (1975), held that the meanings of words like gold or water depend on the actual nature of the very substances the words are intended to denote, whatever the descriptions provided by the current scientific experts. The implication is that experts too fall short of perfect mastery of certain categories of terms used by the community.

We propose a fourth variety of externalism, dubbed ‘extended indexicalism’. Even though he argued for a “(social) division of linguistic labor”, thus apparently endorsing deference to current scientific experts, Putnam advocated physical externalism, i.e. an indexical rather than descriptive theory of reference for natural-kind terms (Liu 2002). Drawing upon him, we consider the possibility of doing justice to certain ‘realist intuitions’ of speakers by arguing that social externalism, in its two variants, might be made subordinate to some indexical externalism for many other terms than natural-kind terms. On this hypothesis, many descriptive terms (even sortal or abstract ones, say, freedom) are seen as intended to capture phenomena that have a certain nature, which might lend itself to rigorous scientific investigation. It is in this ‘nature’ of the phenomenon, of which putative future experts may improve our knowledge, that the meaning of these descriptive terms resides. Current usage, or, when available, descriptions provided by current experts only approximate this meaning.

Semantic deference lies in the fact that “Other people know what I mean better than I do”. It is indeed often the case that users intend to use words the way (some) other members of their linguistic community do. They are then disposed to be corrected if they use the words improperly (i.e. willing to accept that previous statements they made with misused words were false). In the case of social externalism, semantic deference to experts is rather straightforward. We will however show that there are other forms of semantic deference related to the other forms of externalism of contents.