Polysemy Across Languages and Lexical Externalism

The network model of polysemy is much discussed in linguistics but not in philosophy. Conversely, much ink has been spilt on semantic externalism in philosophy, while the topic evokes little interest in linguistics. This chapter demonstrates that the network model of polysemy represents a potential threat to semantic externalism and proposes to adopt the repetition-based conception of linguistic expression to find a middle ground. Words are defined in terms of repetition, which comprises both modulation and cognate relationships. Insofar as the speaker has the ability to modulate senses, the meaning of the term ‘water’ can be considered a function of the intrinsic properties of the speaker. Oscar’s and Toscar’s identical intrinsic properties determine the identical polysemous meaning of ‘water’, i.e. Oscar’s and Toscar’s mental lexicons contain essentially the same polysemy network. Nevertheless, they know distinct words, because their utterances of ‘water’ are not linked by any relation of repetition. Even though Oscar’s ‘water’ and Toscar’s ‘water’ are synonyms, they are still not cognates. This shows that the content of one’s mental lexicon constitutively depends upon one’s external environment. Although Oscar and Toscar can, if put in the same situation, mean the same thing by using the same words, they do not know whether the words they store in their lexicons are identical or not. Semantic externalism should give way to lexical externalism.