Making It Implicit: The Limits of Internalising Meaning

Abstract: The meaning of sentences is a notoriously tricky beast to nail down. In spite of the ease at which we manage to understand new and unfamiliar expressions every day most of us would be at a loss if asked what that understanding consists in. In contemporary philosophy linguistic meaning is often conceived of as a relation of referring or representing things in either some possible world or the actual one. Another strand of thought, often pursued in opposition to the first, instead puts the spotlight on the roles that sentences can play in reason and action. This second approach is the starting point of \textit{inferentialism} or \textit{inferential-role semantics}. Broadly speaking this view says that the meaning of a sentence is captured by its connection to other sentences in reasoning. A particularily detailed inferentialist theory of meaning has been put forward by Robert Brandom and has since been much debated. In this text I give a birds eye view of this brand of inferentialism as well as it's alleged shortcomings. In particular I focus on an avenue of attack, originally due to Kevin Scharp, which targets the tension between the goals of explaining the meaning of semantic vocabulary and being able to state the theory of meaning for a language inside that language. After reiterating the exchange between Brandom and Scharp I offer a sharpening of the original objection. The tension between these goals has so far been discussed in terms of an inferentialist theory of semantics but through a use of Lawvere's Fixed Point theorem can be shown to extend to all attempts of internalising theories of meaning.