Speaker: Tanya Bonderanko
Title: Hyperraising and Logical Form: evidence from Buryat
Time: Wednesday, 02/19, 1pm – 2pm
Location: 32-D461
Abstract: Languages differ in whether they allow hyperraising to object: movement of an argument of an embedded finite clause into the matrix clause. Languages like Buryat (Mongolic) allow such movement, languages like English don’t:
(1) a. bair badm-i:jɘ-1 [CP t-1 sajan-i:jɘ zura-xa gɘʒɘ] han-a:
Bair.NOM Badma-ACC Sajana-ACC draw-FUT COMP think-PST
`Bair thought that Badma will draw Sajana.’
b. *Bair thought Badma-1 [CP that t-1 will draw Sajana].
The question that arises is: what determines whether a language allows hyperraising to object?
In this work in progress, I would like to propose that the relevant factor is the semantic type of the clause. I adopt Kratzer’s (2013) approach to semantics of attitude verbs and follow Deal (2018) in analyzing hyperraising as (potentially covert) raising into a theta-position. I propose that CPs come in two kinds: some, like Buryat CPs, denote properties of events (<vt>-CPs), others, like English CPs, denote properties of individuals (<et>-CPs). I argue that only <vt>-CPs can be hyperraised out of: due to the semantics of movement into a theta-position I propose, hyperraising out of <et>-CPs creates a type mismatch. This account automatically captures such properties of hyperraised arguments as inability to undergo reconstruction, obligatoriness of de re interpretation, and impossibility of indexical shifting.