ComSyn Lectures

Speaker: Hossam I. Ahmed (Leiden Institute for Area Studies)

Title: Reconstructing Verbal Complementizers in Arabic

Date: Thursday 11 February

Venue: P.N. van Eyckhof 2/003

Time: 15.15-16.30 hrs

Abstract: A class of Modern Standard Arabic complementizers known as ‘ʔinna and its sisters’ demonstrate unique case and word order restrictions. While CPs in Arabic allow both Subject-Verb (SV) and Verb-Subject (VS) word order and their subjects show nominative morphology, CPs introduced by ʔinna ban a verb from directly following the complementizer. Preverbal subjects in ʔinna clauses show accusative case marking, while postverbal subjects show nominative morphology. Previous research explains these restrictions as default case or Multiple Case Assignment, both problematic for Case Theory as they violate the Activation Principle.

I explain word order and case effects of ʔinna within the framework of Phase Theory and Feature Inheritance (FI). Morphological, historical, and usage evidence point out that ʔinna-type complementizers have verbal properties similar to illocutionary verbs. Taking Case to be a reflection of phi features that T heads receive from higher heads (e.g. Complementizers) via Feature Inheritance, the nominative-accusative alternation on preverbal subjects can be attributed to the selection of C heads: phi features on null complementizers and conditionals reflect as NOM, while phi features on Verbal Complementizers (VCs) reflect as ACC. Nominative postverbal subjects in ʔinna clauses are explained as an effect of anti-agreement at Spell-Out. Postverbal subjects and the Case probe on T are PF local, allowing for impoverished case agreement. Preverbal subjects and the Case licenser belong to different Phonological Phrases. To satisfy the Recoverability Condition, full case agreement is required between T and the subject, resulting in accusative morphology on the subject. Finally, the requirement that ʔinna-clauses have an intervener between ʔinna and the verb is explained by associating the full phi features of ʔinna with EPP. As the phi set is inherited from ʔinna to T, the EPP property is satisfied by the preverbal subject or by adverbial intervening between ʔinna and the verb.

Speaker: Marieke Meelen (LUCL)

Title: Reconstructing the Rise of V2 in Middle Welsh

Date: Thursday 17 December

Venue: Van Wijkplaats 3/006

Time: 15.15-16.30 hrs

Abstract: Middle Welsh word order has been a ‘vexed’ problem for a long time. First of all, because the basic order was verb-second (called the ‘Abnormal Order’ by Welsh grammarians), whereas Old and Modern Welsh exhibit verb-initial patterns. Secondly, there is a wide range of V2 patterns (subject, object, adjunct or verbal-noun-initial) in Middle Welsh and the distribution of those is not at all clear.

Data from a syntactically and pragmatically annotated corpus show that the sentence-initial position in Middle Welsh could be occupied by contrastively focused constituents at first, but later also by contrastive topics and new information focus as well as Aboutness and Familiar topics. In this talk I show how these V2 patterns developed by carefully reconstructing their syntactic history from earlier patterns with hanging topics and focussed cleft constructions in Old Welsh and related Celtic languages. I argue that a combination of pragmatic and syntactic features explains which word-order pattern appears in which particular context, thus contributing to a better understanding of the origin and distribution of these ‘Abnormal’ patterns in Middle Welsh and knowledge about the development of V2-patterns in general.

Speaker: Christos Vlachos (Queen Mary University of London)

Title: Wh-inquiries revis(it)ed

Date: Thursday 3 December

Venue: Van Wijkplaats 3/006

Time: 15.15-16.30 hrs

Abstract A standard (hypo)thesis, within the framework of Generative Grammar, has been that a wh-parameter may broadly classify languages into two types with respect to how they form wh-questions: languages like English usually front wh-elements, while languages like Chinese usually realize wh-elements in situ. Wh-fronting languages may also feature wh-in situ arrangements, and a tacit (hypo)thesis, tied to the aforementioned one, is that fronting configurations map to information-seeking (“true”) questions, while the in situ counterparts correspond to echo (“pseudo”) questions. Neat as this taxonomy may appear to be, in Greek, which is a typical wh-fronting language, each wh-configuration may map to either meaning. This evidence, in turn, raises the question as to “how much” of the relevant semantics is registered in the relevant syntactic structures. This is the question that this talk will address, by capitalizing on already documented evidence from distribution, interpretation and intonation. The idea to be spelled-out is that part of the relevant semantics is calculated in syntax, while another part is computed by intonation.

Speaker: Anna Volkova (Higher School of Economics (Moscow) & Utrecht University)
Title: Locally bound possessives in Mordvin languages 
Date: Thursday 19 November
Venue: Van Wijkplaats 3/006
Time: 15.15-16.30 hrs

Abstract:
Reuland (2011) and Despic  (2011, forthcoming) propose the following correlation between the prevalence of dedicated reflexives and definiteness marking: Languages without prenominal definite articles employ dedicated reflexive possessives while languages with prenominal definite articles employ simple pronominals. The question is, then, why this would be so, and whether more basic properties of the computational system underlie this correlation. In my talk I contrast data on the use of locally bound possessives in two Mordvin languages: Shoksha Erzya and Moksha. Neither language has prenominal definite articles, hence in both one would expect to find a dedicated reflexive possessive. In Moksha, indeed, a dedicated reflexive possessive es’ is used to encode this relation (1), while in Shoksha Erzya a genitive form of a pronominal is employed (2).

(1)  Son      kel’k-si                            es’         c’ora-nc.
       S/he     love-npst-3sg.s.3sg.s  self        son-3sg.poss.sg.gen
She loves her son. (Moksha)

(2)  Son       aj-n’ij^h-sa-za                               sons’inde       brat-t.
       S/he     ipf-see-prs-3sg.o.3sg.s             s/he.intf.gen  brother-def.gen
He sees his brother. (Shoksha Erzya)

Moksha es’ is feature deficient in number and person (which is complemented by the obligatory presence of a possessive marker on the head noun), unlike the pronominal in Shoksha Erzya. The question is then, what could explain the difference, and what it entails for the Despic-Reuland conjecture. I argue that the relevant factor is not the presence of a DP shell on the object NP per se, but that a broader range of intervening factors enters into an explanation.

References
Despic, Miloje. 2011. Syntax in the Absence of Determiner Phrase. PhD dissertation, University of Connecticut.
Despic, Miloje. Forthcoming. Phases, Reflexives and Definiteness. Syntax.
Reuland, Eric. 2011. Anaphora and Language Design. Cambridge, MA: MIT Press.

Speaker: Jemes Griffiths (Utrecht University)
Title: The syntax of reformulative appositions 
Date: Thursday 5 November
Venue: Van Wijkplaats 3/006
Time: 15.15-16.30 hrs

Abstract:

Reformulative appositions (RAs) are appositions that provide alternative and often more informative names for the referents (1a) or concepts (1b) that their anchors denote. (In (1), appositions are italicised and anchors are boldfaced.)

(1)         a.       The Big Apple, or New York, is a huge city.

b.       Bren confusticates, i.e. perplexes, Swantje.

I claim that reformulative appositions are coordinated with their anchors in a ‘regular’ manner. In other words, I claim that (1a) fits the coordination schema in (2), which is identical to the schema for an utterance such as Bill and Ben slept.

(2)         [[The Big Apple], or [New York]], is a huge city.

I use a variety of arguments to support this claim. To provide two here: (i) I show that RAs are not opaque to c-command, as is typically assumed (for instance see (3), where ATB-movement is observed), and (ii) I show that an understudied subset of RAs that permit ellipsis license gapping, which is reserved for coordinative environments (see (4), where chevrons denote ellipsis).

(3)         [Which country]₁ do you hate the motorways of t₁, or as the Americans say the highways of t₁, the most?

(4)         He went there last month, Sting <went> to New York, I mean.

I also show that this analysis of RAs is superior to the ‘orphanage’ analysis, according to which RAs are remnants of elliptical clauses that are derived in syntactic isolation to their host clauses (5).

(5)         The Big Apple, [New York <is a huge city>], is a huge city.

Speaker: Mark de Vries (University of Groningen)
Title: The Proleptic accusative as an exceptional Exceptional Case Marking construction
Date: Thursday 29 October
Venue: Van Wijkplaats 3/006
Time: 15.15-16.30 hrs

Abstract
This talk (representing joint work with Marjo van Koppen and Lucas Seuren) is about the so-called proleptic accusative. This construction type is relatively rare, but attested in a number of different languages (e.g., Latin, German, Nahuatl). I will discuss data from Middle Dutch in particular, which enables us to shed new light on the matter. An example is the following:

Maer  die         serjanten       sijn     kenden          den     coninc            van         Israël,

but   the         sergeants         his      knew             the_acc    king              of        Israel

dat        hi        niet     was   harde   fel.

that       he_nom   not      was   very    fierce

lit. ‘But his sergeants knew the king of Israel, that he wasn’t very fierce.’ (Rijmbijbel,      v.12643, translation ours)

What is interesting is that there appear to be too many arguments in the main clause, one of which is thematically related to the subordinate predicate. It can be shown that the proleptic accusative is crucially different from complex cases of raising. Instead, we explore two novel and competing hypotheses about its structural properties. We first suggest that an analysis in terms of clausal coordination and ellipsis straightforwardly solves a number of syntactic complications, but this leads to problems with semantic interpretation in various cases. We then propose that the seemingly additional accusative argument in the matrix can be base-generated as an embedded hanging topic of the complement clause involved. This requires exceptional accusative case marking across a clause boundary by the matrix verb. The necessary combination of various prerequisites explains why the proleptic accusative is only sporadically found across languages. Importantly, however, it is now clear how this complex construction type can be decomposed into more basic syntactic ingredients without additional stipulations.

Speaker: Melissa Farasyn  (University Gent)

Title: Agreement patterns in Middle Low German non-restrictive relative clauses

Date: Thursday 15 October

Venue: Lipsius 235B

Time: 15:15 – 16:30 hrs 

Abstract

In this talk I will focus on agreement patterns in Middle Low German (MLG) non-restrictive relative clauses, particulary in relative clauses with a first or second person antecedent. This implies the following kinds of structures, in which some kind of agreement has to be achieved between antecedent, relative pronoun and VfinRel:

            O here de du my geschapen hefst

            O lord REL you me created have-2SG

‘O lord who has created me’

(Ey(n) Jnnige clage to gode, Münster, 1480)

            dat=tu mijn vader woldest wesen de mijn schepper bist

that=you my father would be REL [ ] my creator are-2SG

‘that thou wouldst be my father, who [thou] art my creator’

(Dat myrren bundeken, Münster, 1480)

The establishment of agreement between these elements can result in different agreement types, which will be called agreement patterns or agreement chains (Kratzer 2009). Starting point of detecting these patterns will be the ones found in High German (HG), since these were already largely described (f.e. Ito & Mester 2000, Kratzer 2009, Trutkowski & Weiß to app.). I will use the constraints for detecting the phenomenon in HG to explain how we can find these – not that common – patterns in a MLG corpus. Based on a focused corpus study, I will give an overview of the syntactic distribution of the agreement patterns and I will focus on the patterns themselves. Using a privative feature system, a preliminary analysis of the agreement chains and the mismatches in MLG will be given. Further evidence for this analysis will be shown by looking at a former pilot study about null pronominal subjects in MLG (Farasyn & Breitbarth to app.), revealing some remarkable similarities to phenomena found in this study as well, as for example a strong influence from the discourse and the frequent occurrence of deficient pronouns (Cardinaletti & Starke 1999) in the Wackernagel position .

Speaker: Seid Tvica (University of Amsterdam)

Title: Rich Agreement Hypothesis beyond Indo-European

Date: Thursday 1 October

Venue: Wijkplaats 3/ room 006

Time: 15:15 – 16:30 hrs 

Abstract

It is well-established in the literature that many Germanic and Romance languages differ in the placement of adverbs, appearing either before or after the finite verb. This typological distinction is standardly accounted for via v-to-I⁰ movement, arguably triggered by the subject agreement features that are assumed to be located at I⁰ (cf. Roberts 1985; Kosmeijer 1986; Rohrbacher 1994; Vikner 1995; Bobaljik and Thráinsson 1998; Koeneman and Zeijlstra 2014, among many others). The observed correlation between the properties of agreement morphology and verb movement gave rise to the so-called “rich agreement hypothesis” (RAH) which, in its strong version, states that in controlled environments the finite verb moves to a vP-external position if and only if the agreement morphology is rich (cf. Koeneman and Zeijlstra 2014).  Building on the work done so far in this talk I present the results of a typological investigation of RAH, showing that RAH holds across many languages, well beyond the Indo-European family. In particular, I will discuss verb movement in three unrelated non-Indo-European languages.

Speaker: Hadas Kotek (McGill University)
Title: Pervasive intervention and the architecture of grammar
Date: Wednesday 23 September
Venue: De Vrieshof 4/010
Time: 13.30-14.45 hrs (NB. Different day and time than the other lectures)

Abstract
I propose a new theory of so-called ‘intervention effects’ in wh-questions which views intervention as a systemic problem that results from an attempt to compute λ-abstraction inside regions where focus alternatives are being used. That λ-abstraction is not well-defined inside regions of focus alternative computation is a known property of simple-typed semantic systems, such as the one in Heim and Kratzer (1998), dating back to the very origins of the focus system in Rooth (1985). However, until now, this problem has been ignored by most major work in semantics, or has prompted researchers to adopt a variable-free semantics or a higher-typed semantic system (e.g. Shan, 2004; Novel and Romero, 2009).

I show instead several novel patterns of intervention effects in English that suggest that intervention effects surface precisely when movement must target a position inside a region where focus alternatives are being computed, and is avoided whenever this configuration can be undone. This sheds light on the architecture of grammar: to explain intervention, we must posit a simple typed system that allows overt and covert movement alongside the projection of focus alternatives, with movement as the default mode of scope-taking in English.

Speaker: Luis Vicente (the University of Potsdam)
Title: ATB extraction without coordination
Venue: Wijkplaats 3/006
Time: 15:15 – 16:00

Abstract
A substantial portion of the literature treats Across-the-Board (ATB) extraction as a process restricted to coordinate structures. However, our best accounts of ATB extraction (Citko 2003, 2005; Bachrach and Katzir 2007) make no special reference to coordination; rather, ATB extraction is simply a way of enabling linearization of trees containing multidominated constituents, regardless of whether extraction takes place from within a coordinate structure. The goal of this talk is to convince you that the lack of reference to
coordination is actually a good feature of these analyses. We will see that sentences like (1) and (2), which have been traditionally treated as instances of parasitic gaps, show a number of properties that point instead towards an analysis in terms of non-coordinate ATB extraction.

(1) This is the book that everyone who reads [__] raves about [__].
(2) Where did this train run to [__] from [__]?