Departmental talk, 2/7 – Emily Clem

The relationships between arguments that are morphologically tracked in switch-reference systems look challenging from the perspective of a constrained theory of syntactic dependency formation. In this talk, I argue that the challenge is only apparent. In particular, I propose that the adoption of Cyclic Agree (Rezac, 2003; Béjar and Rezac, 2009) provides the tools needed to handle the relevant syntactic dependencies in a strictly local way. Drawing on data from original fieldwork, the talk centers on a pattern of switch-reference in Amahuaca (Panoan; Peru), which is typologically unusual (and especially striking from a locality perspective) in that the reference of both objects and subjects in both matrix and dependent clauses is tracked. I argue that Amahuaca adjunct C, which is spelled out as a switch-reference marker, agrees directly with DPs in its own complement and with matrix DPs. This is possible because the maximal projection of this high adjunct C can probe its c-command domain––the matrix TP. I argue that this happens through cyclic expansion of C’s probe in a manner consistent with the predictions of Cyclic Agree and Bare Phrase Structure (Chomsky, 1995). Not only is this account based on cyclic expansion able to accommodate object tracking in switch-reference, but it also provides a straightforward way to capture this apparently non-local pattern of agreement without loosening the conditions on locality in Agree. I conclude with a look at the typology of switch-reference systems and the syntactic and morphological sources of diversity in this domain.