research

Dissertation: Switch-reference and topicality.

Situation semantics

Non-canonical switch-reference and topic situations. My second generals paper (~qualifying paper) is in semantics. In it I argue that switch-reference can mark `identity' between two topic situations. This explains two major facts about SR: It does not always track subjects (= non-canonical SR); and non-canonical SR is not found in subordinating configurations. It also demonstrates that situations need not be spatio-temporally contiguous, a finding that proves useful in variable-based accounts of contextual restriction.

I presented a talk based on this fieldwork at the SULA 4 conference in May 2007. In it, I hypothesize that non-canonical switch-reference involves topic situations. Here is a draft of the write-up for the proceedings that have recently been published:

Non-canonical switch-reference and topic situations (pdf, 111 kb) bibtex

And here is the generals paper:

Paper: Non-canonical switch-reference and topic situations (pdf, 196 kb)

Sequence of tense and split readings. What happens when an embedding context is split between simultaneous and anterior readings? E.g., Five people told me my hat was on backwards is felicitous when three told me "your hat is on backwards," while two told me "your hat was on backwards." These sentences force a sequence of tense reading. A null-tense proposal derives split readings, while the standard bound-tense theory does not. Split readings in languages without sequence of tense require past tense. This emergent sequence of tense behavior is further evidence that sequence of tense is available in all languages, and shows that null tense cannot correspond to present-tense marking.

Negative indefinites. Turkish negative words like kimse are neither negative polarity items nor negative quantifiers. They are indefinites with an unpronounced negative operator that raises to NegP. This analysis provides diagnostics for discovering such items in other languages with no overt split between operators and indefinites, and provides an argument against mobile sentential negation (Kelepir 2001).

Fixing the Scope of Turkish Negation (pdf, 110 kb)

A syntactic argument for a semantic EPP. My first generals paper is on the syntax-semantics interface, with an emphasis on the syntax. The basic claim is that the EPP feature on the T head is restricted semantically. It can only attract phrases of semantic type e. Therefore, quantifiers cannot be interpreted in the specifier position of T. Combined with the copy theory of movement, we derive obligatory partial negation reading for subjects in French. You can read the slimmed down version I presented at ECO 5.

A syntactic argument for a semantic EPP (pdf handout, 50 kb)