Abstract

This dissertation is about Semantic Uniformity. Semantic Uniformity is the claim that what is true for some expressions is true for them all—at least, when it comes to semantics. In particular, I defend three claims in three chapters, in this order: First, all simple linguistic expressions, and not just some, are non-descriptive. That is, their referents are not determined by fit with our beliefs. Second, all simple linguistic expressions are rigid. Relative to each possible world, construed as a world of evaluation, every expression has one and the same referent. These two claims point toward a general picture of why simple expressions refer to what they do. Simple expressions have their referents determined by the causal, informational, or lawful connections they bear to objects and properties in the world. Furthermore, this referential connection is 'direct‘ in that it doesn‘t depend on the distribution of objects and properties at a world; simple expressions simply refer, and that explains their rigidity. In the third chapter I defend the view that the same holds for complex expressions. Consensus has it that complex expressions derive their meanings from the meanings of their parts and how those parts are combined. But, I argue, a better account assumes that complex expressions have their referents determined by the causal, informational, or lawful connections they bear to objects and properties in the world, just like simple expressions. In the background of all this is the Referentialist thesis, that there is no further aspect to meaning than reference. If Referentialism is accepted, then the argument of the dissertation is that we can have a "complete" theory of meaning simply by extending a direct causal theory of reference to all expressions.