ComSyn Lectures

Speaker: Jenneke van der Wal (Leiden University)
Title: Topic-based flexible nominal licensing in Bantu
Date: Thursday 15 March
Venue: Van Wijkplaats 2/002
Time: 15.15-16.30 hrs

Abstract:
Two typical Bantu syntactic characteristics are subject inversion and symmetrical double objects, which are both problematic for current theories of Case and agreement. I propose that both can be captured using a new concept of flexible licensing by low functional heads (Appl and v), connected to topicality.

In subject inversion the logical subject appears postverbally and a preverbal locative/patient/instrument agrees with the verb (-zi- in 1):
(1) Inzogá nti-zi-nywá abáana.
9.alcohol NEG-9SM-drink 2.children
‘It’s the children who do not drink alcohol.’ Kinyarwanda (Ngoboka 2016: 356)

In double object symmetry, either object of a ditransitive (recipient or theme) can be passivised and either can be object-marked (OM) on the verb (2):
(2) a. U-mama u-ba-nik-e in-cwadi (aba-ntwana).
1a-mama 1SM-2OM-give-PFV 9-book 2-children
‘Mama gave them a book (the children).’
b. U-mama u-yi-nik-e aba-ntwana (in-cwadi).
1a-mama 1SM-9OM-give-PFV 2-children 9-book
‘Mama gave the children it (a book).’ Zulu (Adams 2010: 11)

The challenge is how T (for inversion) or v (for object marking) can agree with a lower Goal (theme) across a higher one (subject or benefactive), given standard non-intervention locality restrictions on Agree. I explore the hypothesis that the intervening argument can be licensed by the head that introduces it (v or Appl), if and only if it is non-topical. This allows the higher head to probe past it (assuming no defective intervention).

Speaker: Bernat Bardagil-Mas (Rijksuniversiteit Groningen/UC Berkeley)
Title: Multiple exponence and patchwork paradigms
Date: Thursday 1 March
Venue: Van Wijkplaats 2/002
Time: 15.15-16.30 hrs

Abstract

This talk discusses participant cross-reference in Panará, a Northern Jê language spoken in central Brazil. Previous work on the language (Dourado 2001, 2003) concluded that Panará presents a partial alignment split in the exponence of case. While DPs (both pronominal and lexical) are case-marked in an accusative pattern, the pronominal clitic system works in an alignment split: in realis mood, clitics mirror the case exponed on DPs, but in irrealis mood a different accusative alignment system emerges with the irruption of a different nominative clitic paradigm.

After introducing the existing narrative and comparing the Panará case with the existing claims for other Jê languages, I will challenge the two basic claims in the existing view of Panará alignment: there is no clitic nominative paradigm, and there is no clitic alignment split. I will argue that the illusion of a nominative clitic paradigm results from one of the instances of multiple exponence (Campbell 2012) in the Panará participant cross-reference system, and from a shift in the person features targeted by person exponence in irrealis mood.

Speaker: Hans Broekhuis (Meertens Instituut)
Title: Asymmetrical coordination
Date: Thursday 15 February
Venue: Van Wijkplaats 2/002
Time: 15.15-16.30 hrs

Click here for the abstract 

Speaker: Norbert Corver (UiL-OTS, Utrecht University)
Title: Getting imaginative with a brain full of adverbs
Date: Thursday 7 December
Venue: Van Eyckhof 3/002
Time: 15.15-16.30 hrs

Abstract:

Adverbs and adverbial expressions have played a somewhat subordinate role in the history of generative linguistics; see, though, Cinque’s (1999) seminal study on adverbs. Plausibly, this is due to their somewhat “elusive” nature. As has become clear from studies on adverbial syntax, the boundaries of the concept “adverb”, its grammatical characterization and its internal syntax are often unclear and not agreed upon by researchers (see among others Bowers 1975, Jackendoff 1977, Emonds 1985, Travis 1988, Baker 2003, Alexiadou 2013).

Many adverbs and adverbial expressions display “small elements” (functional material) whose status is not always immediately clear. In Dutch, for example, we find the element -s in adverbial expressions such as anders (other-s, ‘differently’) and straks (soon-s, ‘soon’). This -s also shows up in adverbial expressions such as voorzichtigjes (careful-dim-s, ‘carefully’), which features the diminutive morpheme -je besides the “adverbial” morpheme -s. And what about expressions like op z’n Trumps (at his Trump’s, ‘in a Trump-like way’), which display a possessive pronoun in combination with what superficially looks like a possessor noun phrase? Trying to be (as) imaginative (as rapper Eminem), I hope to give some insight into the linguistic nature and behavior of these small functional elements and this way also into the inner structure of adverbs/adverbial expressions.

Speaker: Rint Sybesma (Leiden University)
Title: VO-OV in Chinese languages and Voice and little v
Date: Thursday 23 November
Venue: Van Eyckhof 3/002
Time: 15.15-16.30 hrs

Abstract:
In Chinese languages, both SVO and SOV surface orders are common. We see variation both within and between languages. Some languages, like Cantonese opt almost exclusively for SVO, while languages from the Wu and Min families tend to have SOV as soon as either the V or the O is more than just a bare V or a bare O. Mandarin, finally, shows both orders in several contexts. In this talk I try to find out whether we can understand what is going on if our analysis of these patterns involves Voice and little v, which may or may not be split or bundled all the time.

Speaker: Norma Schifano (University of Cambridge)
Title: The expression of progressive aspect in an endangered variety: the case of Grico
Date: Thursday 9 November
Venue: Van Eyckhof 3/002
Time: 15.15-16.30 hrs

Joint work with Adam Ledgeway and Giuseppina Silvestri (University of Cambridge)

Abstract:
The aim of this talk is to investigates the expression of progressive aspect by means of verbal periphrases in the Italo-Greek variety known as Grico, spoken in Salento (southern Italy). Building on the extremely valuable, yet out-dated, description of Rohlfs (1977), we first present an overview of the array of different patterns brought to light by our recent fieldwork and through a survey of a selection of both early and contemporary sources which include combinations of (non-)inflected STAND with (non-)finite forms of a lexical verb, optionally linked by functional elements. After describing the empirical scenario, we assess the degree of grammaticalization of the patterns which are still productive today, reconstructing their evolution from earlier periphrases and paying particular attention to the grammaticalization of the ambiguous element pu ‘where; from; that’. Finally, we analyse a hybrid structure currently consistently produced by semi-speakers from different villages, which seems to instantiate a new ‘third’ option within the local repertoire. We conclude with of a number of observations about the role of this case study for our knowledge of diatopic morphosyntactic microvariation in Grico and for the nature of language contact and language change.

Speaker: Marijana Marelj (Utrecht University)
Title: The vagueness of HAVE(ing)
Date: Thursday 26 October
Venue: Eyckhof 3/002
Time: 15.15-16.30 hrs

Abstract:

Whether characterized as “semantically impoverished” (Butt 1995), “semi-lexical” (Corver and van Riemsdijk 2001), “vague” (Bruening 2015), or as “having no independent semantic content” (Ritter and Rosen 1997), different researchers seem to agree on the intuitive description of what the notion “light verb” is about. The formalization of what it means to be “light” or “semantically vacuous”, however, seems to elude us.

Taking the empirical domain of HAVE as a testing ground and the work of Ackema and Marelj (2012) as a starting point, I pursue the formalization of thematic lightness while maintaining the lex parsimony, i.e., having one single entry for all the guises of HAVE. This will further lead us to discuss the purported “specialness” of the complements of HAVE and probe into the relation between the Thematic and the Aspectual realms.

The upshots of the talk are the following:

1. HAVE: <[ ]> , where [ ] is an interface, unvalued feature.
2. The complements of light verbs are not per se aspectually “special”. It is the aspectual “scaffolding” they are embedded in that makes them look “special”.
3. Thematic & Aspectual roles cannot and should not be collapsed into each other.

Selected  Bibliography

Ackema, Peter (1999). Issues in Morphosyntax. Amsterdam: John Benjamins.
Corver, Norbert & Henk van Riemsdijk (2001). Semi-Lexical Categories. The function of Content Words and the Content of Function Words. Berlin/NY: Mouton de Gruyter.
Marelj, Marijana (2004). Middles and Argument Structure across Languages. Ph.D. dissertation, Utrecht University.
Marelj, M. (2012). To have the Empty Theta Role.  The Theta System: Argument Structure at the Interface Oxford: Oxford University Press. Met Martin Everaert en Tal Siloni. Met Peter Ackema.
Pesetsky, David & Esther Torrego (2007). The syntax of valuation and the interpretability of features. In Phrasal and Clausal Architecture: Syntactic derivation and interpretation in honor of Joseph E. Emonds, ed. Simin Karimi et al. Amsterdam: Benjamins, 262-294.
Ritter, Elizabeth and Sara Thomas Rosen (1997). ‘The Function of Have’. Lingua 101, 295-321.

Speaker: Cora Pots (KU Leuven)
Title: Te wel of niet (te) hoeven (te) plaatsen:
Variation in te-placement in Dutch non-finite verb clusters
Date: Thursday 12 October
Venue: Eyckhof 3/002
Time: 15.15-16.30 hrs

Abstract:
The morphosyntactic variation in Dutch finite verb clusters has been studied extensively (Barbiers et al. 2005; Wurmbrand 2017), in contrast to their non-finite counterparts. This talk focuses on regiolectal variation in verb clusters consisting entirely of non-finite verbs, and in which the infinitival marker te ‘to’ is required to appear based on selection requirements. I conducted a large-scale questionnaire study of three types of non-finite three-verb clusters, in which selection requirements dictate te should appear on V1 (te-V1-V2-V3) on V2 (V1-te-V2-V3; cf. (1)), and on V3 (V1-V2-te-V3). The data show that there is variation among speakers regarding the presence/absence of te and the placement of te. That is, there are speakers who: (i) allow or need te to be absent, contrary to selection requirements, (ii) allow or need te to be raised, i.e. to appear on a higher verb in the cluster than is required by selection requirements, and (iii) allow te to be doubled, i.e. to appear twice, when only one te is required.

(1) Koen zal    vandaag  niet [ (te)  hoeven₁   (te)   gaan₂   voetballen₃ ].
Koen will   today       not     to    need.INF  to     go.INF  play.football.INF
‘Koen won’t have to go play football today.’

In (1), the highest verb within the non-finite verb cluster, hoeven ‘need’, selects a te-infinitive: selection requirements thus dictate te to appear on V₂ gaan ‘go’. However, the data show that te can be either completely absent (te-drop), appear on V₁ instead of V₂ (te-raising), or appear on both (te-doubling).
I analyze Dutch non-finite verb clusters are cases of functional restructuring (Wurmbrand 2001; IJbema 2001; Ter Beek 2008). I argue that variation in te-drop is due to differences in structural complement size of the verb selecting the te-infinitive. Furthermore, I argue that te-raising is a case of clitic climbing, a well-known restructuring phenomenon of other restructuring languages such as Italian (Rizzi 1982; Cinque 2001). Te-doubling is analyzed as spell-out of both copies of raised te. By showing that clitic climbing is also attested in Dutch non-finite verb clusters, this talk fills a previously unexplained gap in the cross-linguistic distribution of restructuring phenomena across Germanic and Romance.

Speaker: Gertjan Postma (Meertens Instituut)
Title: Loss of the infinitival marker tau ‘to’ in Brazilian Pomeranian: language converence or accommodation to Portuguese?
Date: Thursday 28 September
Venue: Eyckhof 3/002
Time: 16.00-17.15 hrs (Different time than usual)

Abstract:
When various dialects enter in intense and prolonged mutual contact in a new sociological setting, they may converge (Weinreich 1954, Trudgill 1986) in a process of koineisation (Chambers & Trudgill 1980). This situation occurs with enhanced intensity in newly-colonized areas, in so-called language islands (Rosenberg 1994): convergence of a conglomerate of mutually intelligible dialects towards this new koinè. In this talk we study a well-known, but often ignored mechanism and outcome: retreat to default settings, the rise of the unmarked. The case study pursued here is the complete loss of the infinitival prefix tau ‘to’ in Pomeranian, a West Germanic language, extinct in Europe, but still spoken in isolated communities in Brazil. While the original Pomeranian dialects in Europe had a considerable amount of variation in this particular domain, Pomeranian in Brazil has converged to a remarkably uniform new construction, which was not present in Europe in the days of emigration. We will evaluate the koineisation hypothesis vis-a vis the alternative that language contact with Portuguese be at stake.

Speaker: Hiromu Sakai (Waseda University)
Title: Typological variation in the time-course of sentence production. A view from eye-tracking studies on Kaqchikel, Spanish, and Japanese.
Date: Thursday 21 September
Venue: Eyckhof 2/004 (room changed!)
Time: 15.15-16.30 hrs

Abstract:
Native speakers can produce sentences describing a simple event in a fast and accurate manner. Given a wide range of typological variation in the world languages, the time course of sentence production should differ significantly from one language to another. In a series of studies using the eye-tracking during sentence production (Griffin and Bock, 2000) paradigm, we examined how much differences exist in production processes of languages with different word orders and when the processes diverge in their time-course.