My primary research concerns linguistic and mental representation, especially as viewed through the semantics of attitude ascriptions in natural language.
In my dissertation and published work, I defend a new view of the propositional attitudes and their objects. I also show how complex semantics can be done under the assumption of structured propositions.
I'm increasingly interested in the philosophy of artificial intelligence, especially as it pertains to mental states and their ascriptions.
I present a challenge for the traditional view of belief as a two-place relation we bear to propositions. The central observation is that it makes sense to say (for example) that what Peter believes about Paris is what Lewis believes about London, despite no proposition being coherently identifiable as what each person believes about his respective city. Exceptional cases like this, I argue, demonstrate that sometimes we need something akin to a property to play the role originally assigned to a proposition. When I utter, ‘Peter believes about Paris that it’s pretty,’ for example, the ‘that’-clause must designate the property of being pretty; unlike any proposition, this property is the kind of thing that, roughly speaking, different people can believe about different cities. I end this work by showing how to salvage the traditional, binary view of belief through a complex lexical meaning for the word ‘about.’
Chair: Josh Dever
Committee: Hans Kamp, Ray Buchanan, Harvey Lederman, Jeff Speaks
My dissertation concerns the semantics of propositional attitude reports and the metaphysical nature of truth-evaluable content. It begins with novel linguistic data that challenge the standard view of the attitudes, such as belief, as two-place relations we bear to propositions. These data suggest that certain `that'-clauses designate properties, such as being pretty, that are not propositions; a formal semantic account demonstrates the complexity required in order to salvage the standard view. In light of this result, I propose a simpler account of the data according to which belief is a three-place relation we bear to things and properties, where propositional belief is understood as a relation we bear to possible worlds and their properties. The view is formally implemented within an intensional semantic framework that allows advocates of structured propositions to account for key features of linguistic intensionality.
I develop a compositional semantics for structured propositions that incorporates possible world variables at logical form in order to correctly predict challenging cases of transparency. This work contributes to a growing literature aimed at legitimizing structured views of propositions by implementing them in accordance with contemporary theories of formal semantics.
[Draft]
I argue that the data presented in 'A Puzzle about Belief-about' is far simpler to explain by taking belief to be a three-place relation that we bear to things, such as cities, and properties, such as the property of being pretty. Paradigmatic cases of propositional belief, such as when Lewis believes that London is pretty, can be assimilated into the new theory by taking propositions to be properties of (covertly expressed) possible worlds. On this view, for example, the logical form of ‘Lewis believes that London is pretty’ is better revealed by rewriting it as ‘Lewis believes about w that London is pretty in it,’ where ‘w’ refers to the relevant circumstance of evaluation, and ‘that London is pretty in it’ designates a property of possible worlds. A variety of linguistic and metaphysical considerations are relevant in evaluating this view, situating it in the literature, and defending it from objections.
I argue against a Geach-quantifier analysis of donkey anaphora proposed by Adam Morton in 'A solution to the donkey sentence problem' (Analysis, 2015).