Representation and Explanation

Many agree that propositions are necessary existents that represent things as being a certain way. While there are detractors, it is relatively uncontroversial that propositions have these features. They are thus able to play important theoretical roles, such as being the objects of beliefs or being the fundamental bearers of truth or falsity. A theory of propositions should make sense of why propositions have these features. In Propositions, Trenton Merricks masterfully takes on the following central question: how do propositions manage to represent things as being a certain way? Merricks argues against the two leading views of propositions and proposes a bold alternative. The overall narrative is an argument by elimination. According to the first view (call it ‘WV’ for ‘Worlds View’), propositions are sets of possible worlds. Each proposition carves out the region of possibility space according to which it is true. Hence, the proposition that dogs bark is the set of worlds in which dogs bark. The second view (call it ‘SV’ for ‘Structured View’) holds that propositions are structured entities whose constituents are united in a certain way. For example, that dogs bark is a structured entity whose constituents may be the property of being a dog and the property of barking. But, Merricks argues, neither WV nor SV can explain how propositions manage to represent in the first place. Merricks’s own view is that there simply is no explanation. Propositions just are necessary existents that represent the world as being a certain way: We cannot say anything more about how they manage to represent what they do. I will call his view ‘primitivism’. I think that many of Merricks’s arguments against rival views succeed. However, I am skeptical about primitivism for two reasons, which I will elaborate on in the following sections. First, distinguish modal validities—arguments whose premises necessitate their conclusion—from logical validities—arguments that are truth-preserving in virtue of logical form. Merricks’s view, unlike those of his opponents, cannot explain something that cries out for explanation: why any logical validity guarantees the existence of a corresponding modal validity. Second, contra Merricks’s claims, there is a version of SV that