In this chapter we start from the observation that in Romance languages, complementizers are nominal, belonging to the same argumental series as wh–phrases, although current theories treat them as functional projections of verbs, filling the same positions as verbs do. We argue that the Romance che-type complementizer is not a functional category of the verb, but rather a nominal head, which satisfies an argument slot of the matrix verb and which takes the embedded sentence as its complement (section 1.1). Both as a wh–phrase and as a complementizer, Italian che introduces a variable. If it introduces an individual variable, the wh–phrase reading arises; if it introduces a propositional variable (ranging over situations or possible worlds) it is read as a so-called complementizer.

In section 1.2, we argue that our proposal is compatible with fine distributional evidence relating to the left periphery of the sentence, in particular with the fact that elements such as topics or foci can occur both below the complementizer and above it. We capture this distribution by allowing them to occur at the left periphery of the embedded sentence (below the complementizer) or at the left periphery of the complementizer phrase (above the complementizer). This structural hypothesis predicts that occurrences of the relevant material below and above the complementizer can combine – a fact that requires quite complex functional hierarchies in alternative accounts.