Of passives and propositions

Must the structure of thought mirror the structure of language? Actives and passives are a test case. An active sentence such as ‘Annabel loves Piper’ and its passive counterpart ‘Piper is loved by Annabel’ are paradigm cases of sentences that differ in grammatical structure but have the same content. The passive therefore motivates approaches to semantic composition that differentiate the structure of the proposition from the grammatical structure of the sentence. Yet, a tradition in linguistics and philosophy says that active sentences and their passive counterparts – despite appearances – have the same grammatical structure. I argue that this tradition is mistaken. Standard syntactic theories posit a difference between an active and its passive counterpart. Therefore, the semantic theory should allow the structure of a sentence to differ from the structure of a proposition that it expresses.