I'm interested in the syntax-semantics interface, with regards to temporal and aspectual properties cross-linguistically.
I am currently looking at how these properties manifest themselves in Gitksan, a Tsimshianic language of British Columbia, Canada. I'm an active member of Gitlab at UBC.
Current Gitksan-related projects:
1) Agreement in embedded clauses: agreement in Gitksan shows the split between independent and dependent marking: root clauses can have either and embedded clauses can only have the dependent marking. The picture gets seemingly messy once we start looking at the embeddings of different verbs - both agreement patters are allowed in the complements. But, once we analyze some of these embeddings as forming a single clausal domain with the matrix clause, everything falls into place.
2) 2 future forms in Gitksan behave exactly the same as English will and be going to. Why is that interesting? Gitksan forms combine independently motivated progressive and prospective; this contributes to the debate about the nature of the English forms, but also to the idea of combining semantic building blocks. This is a joint project with Lisa Matthewson and Michael Schwan published in Glossa.
3) Sequence-of-Tense: Gitksan has been argued to lack tense only superficially. Spoiler alert: Gitksan shows exactly the same readings as English embedded past, which can only be captured if there is Tense.
Other ongoing projects:
1) Is there really a difference between languages with and without temporal morphology when it comes to future reading? Not really - in all of them, something needs to structurally license the future-denoting element.
2) Do (negative) questions in Serbian divide the pie in the same way as English, when it comes to bias (both speaker and evidential)? Do some of the question-related morphemes in Serbian come with a bias? Is there a difference between polar and wh-questions when no answer is expected from the interlocutors? I explore these and similar questions (!).