On the left periphery of ‘optional’ wh-in-situ: Evidence from Greek

A recent generalization about ‘optional’ wh-in-situ states that wh-in-situ questions are blocked in selected contexts, where the only grammatical option is wh-fronting. In this squib, I show that the generalization is misplaced: Wh-in-situ is not actually blocked in selected contexts, it is simply not visible to selection as a wh-construction. To this end, I provide empirical evidence from Greek wh-in-situ questions, concentrating on issues revolving around selection, movement and the discourse. The analysis proposed treats wh-in-situ as a construction in its own right, being open to labeling possibilities that its wh-fronting counterpart cannot have.