[PLing] Talk by Boban Arsenijević on January 18, 13:15 CET

[Madeleine Butschety]

Dear colleagues, 

we cordially invite you to a talk by Boban Arsenijević (University of Graz) as part of the Jezik & Linguistics Colloquia series. His talk entitled "The size of complements and the height of finiteness: an analysis based on Torlakian Serbo-Croatian complementizer-less constructions" (abstract see below) will take place next Thursday, January 18, 13:15 CET at lecture room P5 at the University of Nova Gorica, and will also be streamed on Zoom. 

To join us via Zoom, please use the following link: https://ungsi.zoom.us/j/63750587041?pwd=NzhONHdOcEY0RnM0MUdyMFA1dW5jUT09 

We hope to see many of you there! 

Kind regards, 
Madeleine Butschety, on behalf of the Center for Cognitive Science of Language 

>>BOBAN ARSENIJEVIĆ: 
"The size of complements and the height of finiteness: an analysis based on Torlakian Serbo-Croatian complementizer-less constructions" 

Based on the Standard and Torlakian Serbo-Croatian (SC) dialects, in particular the complementizer-less verbal complements in the latter (Sobolev 2004), I show that 'da'-complements in SC cannot be smaller than the AspP and develop an analysis for verb agreement in each of Todorović & Wurmbrand (2020) and Wurmbrand et al. (2020)’s three types of finite complements matching the CP, the TP and the VP. I argue that 'da' takes a predicate over situations and abstracts an ordering source. It can thus be generated in any of the projections which otherwise provide such ordering sources, in particular AspP, EpistP, EvidP and ForceP. The embedded clause is then a predicate over ordering sources and acts as a modifier of a verb expressing an ordering source: a temporal one in AspP, epistemic in EpistP, evidential in EvidP or speech-act-related in ForceP. When da is derived in AspP, the complement verb must be in the present – the barest verb form in SC, and cannot have an own subject, because these are specified in AspP or higher. Contra Wurmbrand and colleagues, 'da' cannot be generated below AspP, as there no ordering source is supplied. VP complements hence lack a functional projection and cannot be spelled out on their own. They thus have access to the selecting verb and agree with it directly in phi-features. The relativizing capacity of 'da' is represented in syntax as an uninterpretable finiteness feature, which probes up and agrees with the closest topic situation – the one specified in the matrix AspP. Thus, 'da' always underlyingly carries the finiteness values of the matrix verb – its person, number and gender. When the complement is larger than AspP, its verb has own specification of these features, acquired by agreement with AspP or other projection in which it is specified. When the complement is AspP, the probe instead finds 'da' there, and inherits the features of the matrix verb (cf. Zagona 2014). 

References: Sobolev, A. 2004. On the areal distribution of syntactic properties in the languages of the Balkans. In O. Tomić (ed.), Balkan Syntax and Semantics. 59-100. Amsterdam: Benjamins. Todorović, N./Wurmbrand, S. 2020. Finiteness across domains. In: T. Radeva-Bork/P. Kosta (eds), Current Developments in Slavic Linguistics. Twenty Years After. Frankfurt/Main: Peter Lang, 47 –66. Wurmbrand, S./I. Kovač, M./Lohninger, C./Pajančič/N. Todorović. 2020. Finiteness in South Slavic complement clauses: Evidence for an Implicational Finiteness Universal. Linguistica 60(1). 119–137. 


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.univie.ac.at/pipermail/pling/attachments/20240108/a1fbad12/attachment.html>