Whence Structured Propositions?

This thesis is a critical examination of Structured Propositionalism (SP), the view that propositions are complex entities composed of the semantic values of the (meaningful) parts of the sentences that express them. According to SP, propositions have constituents and are individuated by the identity and arrangement of their constituents.

Two questions face advocates of Structured Propositionalism. First, what does it mean to say that propositions are structured and have constituents? And second, why think that propositions have these properties? These are the primary questions I address in this thesis. Given the deficiency of contemporary attempts to answer these questions, three chapters are spent examining historical accounts of structured truth-bearers, going back to Plato's Theaetetus and Sophist. I argue that an important reason for SP's wide acceptance is the historical ubiquity of the view that the fundamental truth-bearers are sentential, and thus obviously structured.

After presenting historical accounts of structured truth-bearers, with a focus on Frege's and Russell's influential versions of SP, I present some objections to the application of mereological notions to abstracta, with an emphasis on the difficulties for providing a metaphysics of propositional constituency. I discuss five approaches and show that all of them face serious problems. These problems, while perhaps not insurmountable, shift the burden of argument to the proponents of SP.

I then argue that the required arguments have not been given. In the penultimate chapter, I evaluate six arguments for SP: the Compositionality Argument, the Logical Form Argument, the Argument from Analyticity, The Similarity Argument, the Direct Reference Argument, and the Language-based Argument. These arguments are based on suggestions in the historical literature' primarily from the early analytic philosophers' and enthymemes in the contemporary literature. I conclude that none of the arguments makes a compelling case for SP.

In the final chapter I examine two versions of SP recently advocated by Scott Soames and Jeffrey King. I argue that 'propositions' of the kind postulated by these theories are unable to play some of the fundamental roles that motivate the postulation of propositions in the first place.