The Syntax-Semantics/MULL group will meet at 3pm on October 21 in Room 002 of the McGill linguistics department. The upcoming meeting will feature two shorter presentations. Sama’a Salama (McGill) will present a talk “Direct Evidence isn’t Always it!” and Chase Boles (McGill) will present a talk “‘Bleeding’ Condition C in Kanien’kéha.” Both of their abstracts are below:
Direct Evidence isn’t Always it!: The enclitic =mi in Imbabura Kichwa and other Quechuan languages has been argued to be a marker of direct evidence source (ex. Muysken, 1995). A feature of the family found across its different variants is the overlap between the marking of evidential and focus categories (Cole, 1982; Faller, 2002; Grzech, 2020; Muysken, 1995; Sánchez, 2010 etc.). Through a variety of collected data including natural speech, questionnaires, and designed tasks (ex. Map Task (Anderson et al., 1999)), this work finds rather that: (i) =mi in Imbabura Kichwacannot only mark direct evidence, even if it is compatible with it, and (ii) evidentiality more generally cannot be properly accounted for as only encoding information source. =mi is compatible with various sources of evidence but is only felicitous if the speaker is committed to the truth of the prejacent. =mi does not encode for direct evidence nor best possible grounds, but rather the epistemic stance of the speaker, therefore ending the line of inquiry and answering the current question under discussion. This reassessment of =mi opens the door to a revaluation of its focus role, lending itself to the possibility that these enclitics are not markers of focus but rather evidential markers at their core that associate with focus.
“Bleeding” Condition C in Kanien’kéha: Kanien’kéha (Mohawk; Northern Iroquoian) exhibits an odd combination of binding effects: Condition C appears to hold across clauses but not clause-internally. Baker (1996) conducts a barrage of tests to argue that certain R-expressions are object-internal and bound by a coreferring pro, concluding that all overt arguments must be high-adjoined to account for Condition C effects. In this talk, I argue that the apparent bleeding of Condition C effects is due to a parsing ambiguity arising from robust pro-drop and “free” word order. I discuss Baker’s tests as well as novel data from conjoined objects, showing Baker’s analysis to be untenable. Building on recent work on Condition C effects being obscured by other mechanisms (e.g., Royer 2024), I argue that the Kanien’kéha binding operates as expected and does not require radical new argument structure proposals.
Online participants can join with this link: https://mcgill.zoom.us/meeting/register/tZYvfu6orzIrHtEHSdthyymSx50ZHxlHqvwu.