Andreas Haida (Hebrew University of Jerusalem)
On how the strongly exhaustive reading of wh-questions is derived
There is little controversy that wh-questions have a strongly exhaustive reading. At the same time, there's little agreement about how strong exhaustivity comes about when it is observed. In my talk, I will argue that exhaustification is performed by means of an operator that quantifies over formal alternatives of the question nucleus (as suggested in Nicolae 2013). Crucial support for this conclusion comes from the observation that in both syntactic and semantic terms (i-b) differs from (i-a) in the same way as (ii-b) differs from (ii-a).
a. Sue weiß, wie schnell sich Toms Auto bewegt hat.
'Sue knows how fast Tom's car was moving.'
b. Sue weiß, wie schnell sich Tom's Auto mindestens bewegt hat.
'Sue knows for which speed d Tom's car was moving at least d-fast.'
a. Sue knows that Tom's car was moving at 130 km/h.
b. Sue knows that Tom's car was moving at at least 130 km/h.