ComSyn Lectures

Speaker: Thomas McFadden (ZAS, Berlin) & Sandhya Sundaresan (U. of Göttingen)
Title: Unifying species of C-agreement
Date: Thursday 4 March
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract

In this talk, we will undertake one of the first detailed comparisons of clause-peripheral complementizer agreement (“C-agreement”) and argue that it is splintered across distinct heads which stand in agreement with dedicated extra-clausal arguments. In this, the heterogeneity of C-agreement parallels that of clause-internal agreement, which is also typically understood to involve distinct functional heads (e.g. T or v) in agreement with (often distinct) nominal arguments (e.g. a subject or object). “C-agreement” is thus a misnomer, masking a slew of disparate agreement phenomena that are rarely discussed in unison or compared (but see Baker, To Appear).

We will compare and contrast three distinct types of agreement on C: (i) downward complementizer agreeement (DCA) with an embedded subject, as seen in many West Germanic languages (van Koppen, 2017); (ii) upward complementizer agreement (UCA) with a matrix subject as in many Bantu languages (Diercks, 2013; Carstens, 2016, a.o.); and (iii) allocutive agreement (AA) with an extra-argumental Addressee, as seen e.g./ in Basque and Tamil (Oyharçabal, 1993; Miyagawa, 2017; McFadden, 2020). We will show that these patterns differ along distinct parameters which do not straightforwardly facilitate a unified analysis. Instead, we will propose that the heterogeneity of C-agreement is merely epiphenomenal of the CP itself being articulated across a sequence of C-heads in a rigid, monotonic order (cf. the functional sequence in Cinque, 1999, a.o.) — again paralleling the standard notion of a functional sequence within the TP (minimally, T > v> V). Differences in C-agreement, we will argue, fall out solely from differences, parametrized across individual structures and languages, wrt.: (i) the presence vs. absence of a probe; (ii) the height of a probe relative to the embedded CP phase; and (iii) the structure of the CP which, in turn, influences the availability of certain goals. We will conclude by showing that our model makes the right empirical predictions wrt. the distribution and typology of C-agreement, both across languages and across individual structures.

Speaker: Olaf Koeneman (Radboud University) & Hedde Zijlstra (Georg-August-University Göttingen)
Title: Deriving pro drop in a non-paradigmatic way
Date: Thursday 10 December
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract:

Apart from incidental contexts in which Germanic varieties allow partial pro drop (cf. Fuss 2005, Rosenkvist 2007), a categorical difference between Germanic and Romance languages is that none of former allows full pro-drop, whereas many of the latter do. This is standardly explained by reference to the whole paradigm (cf. Taraldsen 1978, Rizzi 1982, Jaeggli & Safir 1989): A language like Italian is rich overall, so it has pro drop across the board, whereas English is poor overall, so it lacks pro drop overall. Such a paradigmatic approach accounts for the fact that English does not even have pro drop in the 3rd person singular, despite the unique -s form, but it faces several problems. One of them is the existence of partial pro drop languages, which suggest that null subjects can in principle be contextually licensed, for instance in the 2nd person singular. However, a contextual approach to understanding pro drop massively overgenerates. It for instance expects English to have pro drop in the 3rd person singular. To conclude, there are two ways of expressing a link between pro drop and rich inflection (paradigmatically or contextually) and both fail. In this talk, we will sketch a way out of this conundrum and show how paradigmatic effects can be derived within a contextual approach.

Speaker: Richard Stockwell (Oxford University)
Title: Contrast and Verb Phrase Ellipsis: Triviality, Symmetry, and Competition
Date: Thursday 26 November
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract
This talk, based on my recent UCLA dissertation, argues that contrast is crucial to verb phrase ellipsis (VPE). The contrast requirement comes to light from studying the interaction of VPE with triviality, symmetry, and competition. In some trivial sentences (1), VPE is ungrammatical for lack of contrast. With symmetrical predicates (2), symmetry creates semantic identity that overcomes non-identity in form, but can also lead to contrast failures. And in MaxElide effects (3) (Schuyler 2001, Merchant 2008), contrast (cf. Griffiths 2019) offers a more promising account for the ungrammaticality of VPE than competition from sluicing:

(1) If John is wrong, then he is *(wrong).

(2) John wanted to dance with Mary, but she didn’t want to (dance with him).

(3) John ate something, but I don’t know what (*he did).

Speaker: Alexandra Vydrina (CNRS)
Title: From addressee-oriented deixis to a reflexive: the case of the pronoun ì in Kakabe (Mande)
Date: Thursday 12 November
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract

It is not uncommon cross-linguistically to find a polysemy pattern whereby the same pronoun stands for the addressee and can be used as a generic pronoun. This is, for example, the case of the English pronoun you can be used deictically to refer to the addressee, but also as a generic pronoun as in (1).

(1) You don’t drink and drive.

The Kakabe pronoun ì, just like the English you, has both the deictic 2^nd person use and the generic use as in (1). What is more surprising, ì can also appear in contexts of the so-called restricted genericity where the variable introduced by it has a generic or a universally-quantified DP as antecedent. In this case, it patterns with the French generic-reflexive pronoun soi (Charnavel 2018); cf. (2) where soi  has a distributive universal DP as antecedent:

(2)  [Tout linguiste]_i parle de soi_i. Every linguist talked about himself.

Finally, and this type of use is the most unexpected one, ì can be used as a reflexive pronoun, but only if its antecedent is PRO in an infinitive clause or a relativized DP.

I will argue that, in its generic, distributive and reflexive-like uses, the Kakabe pronoun ì is a morphological instantiation of the zero pronoun, a semantic entity described by (Kratzer 1998). Relying on the evidence of other Mande languages, a diachronic scenario will be proposed whereby the 2nd person pronoun has developed into a dedicated morphological expression of the zero pronoun.

Speaker: Yining Nie (NYU)
Title: Two ways of forming causatives and their implications for recursion
Date: Thursday 29 October
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract:

Syntactic approaches to the causative alternation disagree as to whether causatives differ from their anticausative counterparts crucially in the addition of a causing event (Harley 1995, 2013; Pylkkänen 2008) or in the addition of a causer participant (Alexiadou et al. 2006, 2015; Schäfer 2008). In this talk, I focus on productive causative constructions of the “X made Y do Z” type and demonstrate that both strategies are in fact attested in the formation of productive causatives. While productive causatives involve added events in Japanese and Turkish, they only involve added participants in Tagalog. I show that the choice of strategy also correlates with a previously unexplored property: the ability to recurse causatives (“X made Y made Z … ”). Causatives which add events can recurse, while causatives which add participants cannot.

Speaker: Melissa Farasyn (Ghent)
Title: Nature and origin of v>2 in the French Flemish dialects: archaisms and novelties in a split left periphery
Date: Thursday 15 October
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract:

In this talk I focus on the syntactic properties of verb-later-than-second (V>2) word order in the moribund French Flemish dialects, which are the most (south)western continental West Germanic dialects, roofed by a Romance language.

French Flemish, in a similar fashion as the West Flemish dialects roofed by the Dutch standard language, allows V>2 word order, though earlier research suggests this happens way more often than in the other West Flemish dialects. While there has been some research on V>2 in French Flemish, these studies are, especially with respect to the number of investigated locations, insufficient to give a clear picture of the origin of the structure and the role of language contact. The research I present is intended to fill this gap, using time-aligned transcriptions of spoken speech recorded in the 1960s of over 50 different locations in French Flemish.

Departing from the remarkable socio-historical and extralinguistic context which shapes the French Flemish dialects, I will present my ongoing research on the frequency and the properties of V>2 in French Flemish, considering the impact of language contact. I argue that the different incidences are the result of both archaisms and novelties, such as a higher incidence of non-integrated initial adverbial clauses (without resumptive adverbials) in FrameP, and novelties, such as clause introducing ‘t maakt, which I analyse as a grammaticalised discourse marker in the left periphery (possibly DeclP). Such analysis inevitably leads to a theoretical framework with a split left periphery, in which many of the initial constituents and clauses in V>2 patterns interact with the discourse.

Speaker: András Bárány (Bielefeld/LUCL) and Jenneke van der Wal (LUCL)
Title: We don’t Agree (only) upwards
Date: Thursday 8 October
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

Abstract:
Bjorkman and Zeijlstra (2019) argue that Agree involves two operations:
checking and valuation. Checking always happens upwards, in a configuration in
which the checker of a feature, carrying an interpretable feature [iF], c-
commands the checkee, carrying an uninterpretable feature [uF]. Valuation
generally happens downwards, after the valuer has moved to the specifier
of the valuee. This makes very clear predictions, which we show are not borne
out.

We discuss several configurations involving in situ agreement controllers which
have not featured in this debate: subject agreement with in-situ subjects in
Matengo and German, object agreement in Sambaa and Liko, and complementiser
agreement in Nez Perce. We argue that such φ-agreement phenomena indicate that
Bjorkman and Zeijlstra’s (2019) proposal is empirically inadequate. Adding
these data to the ongoing debate on the directionality of Agree (see Zeijlstra
2012, Preminger 2013, Bjorkman and Zeijlstra 2014, Preminger and Polinsky 2015)
is important because if a given approach is not empirically adequate, the
question whether it has conceptual advantages is moot.

The new ComSyn season is starting! Due to the coronavirus situation, all talks this season will be held online via Skype. Please sign up for the mailing list / contact the ComSyn organizers for links to the talk (note that we will be using a different link from last semester!). The full programme can be found on the right-hand side bar; the first talk will be by Maria Kouneli:

Speaker: Maria Kouneli (Leipzig)
Title: The causative alternation in Kipsigis
Date: Thursday 1 October
Venue: Online (sign up for our mailing list or contact one of the organizers to get access to the talk)
Time: 15.15 – 16.30 hrs (CEST)

Abstract:

Recent syntactic approaches to the causative alternation (e.g. The cup broke vs. Sue broke the cup) treat it as a Voice alternation: the causative and anticausative variants have the same vP (event) layer, but differ in the presence vs. absence of an external argument-introducing Voice head (e.g. Marantz 2013, Alexiadou et al. 2015, Wood 2015, Kastner 2020). In this family of approaches, transitivity alternations cross-linguistically arise from different types of Voice heads (Alexiadou et al. 2015 a.o.). In this talk, I provide a description and analysis of the causative alternation in Kipsigis (Nilotic, Kenya), and I show that it is not a Voice alternation in the language: while it is the position of the external argument that regulates transitivity alternations in languages like English, it is the syntactic position of the internal argument instead that determines (in)transitivity in Kipsigis. Thus, not all transitivity alternations cross-linguistically can be explained by variation in Voice heads.

Speaker: Sjef Barbiers and Astrid van Alem (Leiden)
Title: Poor weak ‘t (it) and the syntax of pronominal clefts
Date: Thursday 18 June
Venue: Skype (contact us to get access to the meeting)
Time: 15.15 – 16.30 hrs

This paper derives the well-known 1/2 vs. 3 person split in Dutch pronominal clefts (Ackema and Neeleman 2018; cf. (1)) from the feature specification of ‘t ‘it’.

(1) a. * ‘t ben ik/jij die het antwoord weet.
it am I/you who the answer knows

b. ‘t is zij die het antwoord weet.
it is she who the answer knows

Based on a study of pronominal forms in Standard Dutch and Dutch dialects, we argue that free ‘t ‘it’ is not a pronoun but a determiner that has a 3p feature but no case or number feature. This underspecification explains the split in (1) and the fact that this split is neutralized in the plural and the past. In a questionnaire study, we furthermore observe that there are two grammars of Standard Dutch: A grammar Dutch A that allows (1b) with 1p and 2p pivots and a grammar Dutch B that does not. This follows from a restriction on nominative case checking. Nominative case checking in Dutch B is only possible under person agreement.

Speaker: Nikos Angelopoulos (CRISSP, KU Leuven)
Title: Complementizers as probes
Date: Thursday 4 June
Venue: Online (sign up for our mailing list or contact one of the organizers to get access to the talk)
Time: 15.15 – 16.30 hrs

Abstract