Whamit!

Speaker: Adam Przepiórkowski (Warsaw / MIT)
Title: Concord as Insatiable Agree
Time: Thursday, April 24th, 12:30pm - 2pm
Location: 32-D461

Abstract: The aim of this talk is to present an analysis of concord – nominal agreement in case and phi features – as insatiable Agree (e.g., Deal 2023, 2024; Clem and Deal 2024). On this approach, typical modifiers do not agree with nominal heads directly but are flagged with the relevant features by a higher functional head (e.g., D or K). This provides an immediate account of some otherwise puzzling facts about the number of exponed cases in overt case-stacking languages (e.g., as discussed in Richards 2013 and Pesetsky 2013 with respect to Lardil) and allows for a straightforward analysis of heterogeneous case patterns in Slavic and Finnic. I will argue that this approach to concord is theoretically superior to the current alternatives (Norris 2014, 2018; Bayırlı 2017), which require ad-hoc feature percolation principles. Moreover, given an appropriate rule of exponence of stacked feature bundles, it is immune to Norris’s 2018 Estonian challenge for case-stacking approaches to nominal concord. Empirically, this approach predicts some of the typological generalizations discussed in Bayırlı 2017 (when a language has both phi features and case, it may exhibit concord in phi features alone, but not in case alone) and naturally accounts for cross-linguistic observations in Norris 2018.

A draft manuscript on which this talk is based is available at https://tinyurl.com/4k97try2; at one point the slides will appear there as well. (There’ll be no paper handouts.)

Speaker: Zachary Feldcamp and Ido Elhadad-Benbaji (MIT)
Title: Structure matters: missing implicatures and their consequence for the theory of alternatives
Time: Wednesday, April 23rd, 1pm - 2pm
Location: 32-D461

Abstract: A central component to theories of conversational implicature is the delimitation of the set of alternative sentences that could have been uttered, but were not. Yet there is considerable disagreement about the mechanism underlying the computation of alternative utterances and the kind of objects that alternatives are claimed to be. Examining the domain of English aspect marking, we provide a novel case of implicatures disappearing when syntax independently rules out the alternatives required to generate them, and argue that this data supports a theory in which alternatives are syntactic objects, computed by an algorithm that appeals to the notion of structural complexity (Katzir 2007; Fox & Katzir 2011) but challenges theories in which alternatives are semantic objects, as in the focus theory (Rooth 1992), and the conceptual alternatives theory (Buccola et al. 2022).

Speaker: Philip Shushurin (Higher School of Economics, Moscow)
Title: How to do voice restructuring in a language without voice: Transitivity Concord in Chechen.
Time: Tuesday, April 22nd, 1pm - 2pm
Location: 32-D461

Abstract: The paper presents an analysis of a construction known as Transitivity Concord in Chechen. I argue that this construction presents a novel and previously undescribed type of Voice Restructuring. In the Transitivity Concord Construction, certain matrix transitive verbs impose a requirement to embed a strictly transitive structure.

(1) Asa kniga j-esa d’a-j-ol-ira
I.erg book f-read prf-f-begin-rpst
‘I began reading the book.’

(2) *So/*Asa ida d’a-d-oli-ra
I.abs/I.erg run prf-n-begin-rpst
‘I began running.’

I argue that this condition is in fact a requirement on the matrix verb to contain an agreeable (Absolutive) Direct Object in the embedded clause with which it establishes a long-distance Agree relation. The embedded verb in such a construction is shown to be syntactically deficient, despite bearing agreement morphology. The construction is analyzed in terms of Pietraszko (2021), Bryant, Kovac and Wurmbrand’s (2023) system whereby the embedded structure is featurally dependent on the matrix structure in terms of the ID feature. I depart from this system in assuming that (a) the embedded verb’s deficiency needs not be conditional on its case assigning potential (Absolutive assignment is very low in Chechen) (b) the voice dependency might be encoded as just ϕ. Furthermore, I assume that v and Voice are featurally dependent in Chechen which results in automatic feature transmission between the two heads.
I discuss the TCC’s implications for both the theory of restructuring and infinitives.”