Internalist Deflationism: On the Limits of Ontological Investigation

Abstract

Since Frege(1879), the history of semantics identifies the meanings of natural

language expressions with the mind external things they denote, be they pedestrian

objects (e.g., cows and chairs), less pedestrian objects (e.g. mereological sums), or

abstracta (e.g., sets of possible worlds). For the Quinean Realist, a language with

such a semantics is fruitful for ontological investigation, insofar as analyzing the denotational meanings of (the constituents of) sentences in that language reveals which

objects populate the (external) worldly domain. However, consigning meaning over

to truth in this manner comes at a cost. The externalist thesis is only had by sacrificing the explanatory adequacy of our theory of meaning. Three arguments suggest

this: first, facts about the rapid human acquisition of natural language suggests

that languages are internal to the human mind, as an innate module in cognitive

architecture; second, naturalist commitments suggest that there is no sui generis,

mind-independent kind `word' to stand in the word-to-world relations posited by

the externalist; third, natural languages exhibit lexical flexibility, as manifest in the

distribution of natural language speaker judgments, and this property cannot be

easily explained by an externalist semantics. The Realist might respond to these

arguments by appealing to the languages utilized to express our best scientific theories, using those invented languages as ontological guides. Since these scientific

languages are constructed with the expressed purpose of perspicuously describing

reality, the Realist could contend that expressions in those languages have an externalist semantics. I argue, using examples from evolutionary biology, that scientific

languages exhibit lexical flexibility as well, casting doubt on the claim that these

languages have meanings that admit to externalist treatment. The Realist then

should reject the metaphysical methodology which assumes the externalist thesis

that the meaning of a linguistic expression determines its truth-conditions.