ComSyn Lectures

Speaker:  Molly Rolf (University of Konstanz)
Title: The functional load shift from case to adposition: the role of L2-difficulty.
Date: Thu 11 May
Location: Lipsius 2.23
Time: 16:15-17:30

Abstract:

This talk proposes that a mixed realisation of case features in the KP (case phrase) is more ‘L2-difficult’ than a harmonic realisation, accelerating, in the correct sociolinguistic scenario, to a shift in the functional load expression in the case and adposition systems.

I take as my theoretical background the Final-Over-Final Condition (FOFC) (Biberauer, Holmberg and Roberts (2014) and the exploded KP approach adopted by Caha (2009), highlighting how the two give the same synchronic predictions concerning both case and adposition inventories and syncretisms. The diachronic predictions made by the FOFC mechanism can be straightforwardly applied to the KP, eliciting two directionality pathways of change (i.e. more or less head-initial).

It is furthermore suggested that L2-difficulty can play an important role in this diachronic change: one of the alternative grammars L2 adults can postulate when faced with a mixed realisation grammar is in line with the diachronic pathway we see in the history of Bulgarian and Macedonian and indeed, many Indo-European languages: that of a loss of morphological case and a rise of adpositions. I argue following Lindstedt (2000, 2018) that it is the historic sociolinguistic make-up of the Balkan Sprachbund that has led to the varying degrees of simplification in the case paradigms of the Balkan Sprachbund languages.

Finally, I show how corpus data can be used to track this functional load shift among case and adposition systems and discuss in which ways Balkan Slavic data supports the diachronic predictions made.

References:
Biberauer, Theresa, Anders Holmberg & Ian Roberts. 2014. A syntactic universal and its consequences. Linguistic Inquiry 45. 169–225
Caha, P. 2009. The Nanosyntax of Case. Ph.D. thesis, University of Tromsø
Holmberg, Anders. 2000. Deriving OV order in Finnish. In Peter Svenonius (ed.), The derivation of VO and OV, 123–152. Amsterdam: John Benjamins
Lindstedt, J. (2000). Linguistic Balkanization: Contact-induced change by mutual reinforcement
Lindstedt, J. (2018). Diachronic Regularities Explaining the Tendency towards Explicit Analytic Marking in Balkan Syntax, in Balkan Syntax and (Universal) Principles of Grammar
Roberts, Ian. 2019. Parameter Hierarchies and Universal Grammar. Oxford: Oxford University Press
Walkden, George and Breitbarth, Anne. 2019. “Complexity as L2-difficulty: Implications for syntactic change ” Theoretical Linguistics, vol. 45, no. 3-4, 2019, pp. 183-209.
Wahlström, Max. 2015. The loss of case in Bulgarian and Macedonian. Doctoral thesis, University of Helsinki, Helsinki, Finland.

Speaker: Iva Kovač (UniWien)
Title: Dissecting habituality: The Croatian know and its kin
Date: Thu 4 May
Location: Lipsius 2.23
Time: 16:15-17:30

Abstract:
Habituality is a versatile concept bordering categories such as aspect, modality, and iterativity, and has been used to describe a range of diverse meanings and configurations (see Boneh & Jędrzejowski 2019 for an overview). I investigate a habitual-like configuration in Croatian in which the verb znati ‘know’ embeds an infinitive (see also Hellman 2005) and discuss its syntactic and semantic properties. Comparing znati with expressions from other languages that have been described as habitual requires dissecting the notion of habituality and reveals three broad subcategories.

As for znati in particular, I argue that it is semi-lexical (Emonds 1985, Corver & Van Riemsdijk 2001, Cavirani-Pots 2020). Iterativity figures prominently in its meaning, pointing into a functional direction. At the same time, even though znati is a raising verb (and the subject thus need not possess knowledge of any kind), the core knowledge component applies to the speaker instead, invoking an evidential-like flavour. Since verbs denoting knowledge are known to grammaticalize not only into habitual but also into evidential markers (Kuteva et al. 2019), this comes as no surprise.

Finally, znati is also used as a factive verb and an ability modal. I explore a possible direction towards connecting these different meanings, especially in light of the finding that other languages (e.g., Norwegian, English) employ know in a similarly diverse manner.

Speaker: Charlotte Sant (Arctic University of Norway)
Title: Comparing apples and oranges: What grinding and portioning can tell us about gender and atomicity
Date: Wednesday 26th April
Location: Lipsius 1.30 (note day and room change)
Time: 16:15-17:30

Abstract:
It is not uncontroversial whether or to what extent gender and classification are connected (Corbett, 1991; Fedden & Corbett, 2017; Audring, 2016; Dixon, 1986; Aikhenvald, 2000), or whether a language can embody both systems (described by Fedden & Corbett, 2017). In this talk I will argue that Norwegian nouns vary with regards to whether they are lexically specified for a gender or atomicity value, meaning that these features are found within the same language but in complementary distribution.

Norwegian is consistently reported to have three genders: masculine, feminine and neuter. Using novel data, I will show that neuter is not a gender, but rather a signal that the noun marker carries a +/-ATOMIC feature rather than a gender feature. I observe that neuter nouns are more restricted for grinding and portioning. Neuter nouns can be ground if the noun is bare but not if the NP is definite. Compare the masculine noun ‘appelsin’ to the neuter noun ‘apple’ below. Imagine a context in which you are being trained to work in a cake factory and the person training you points to a barrel of mashed oranges or apples and tells you that this mass will be added to some mixture.

For the indefinite portioned form, masculine and feminine nouns show article agreement according to their gender, but neuter nouns cannot be portioned using the neuter article:

The data shows that atomicity is intrinsic to neuter nouns, and so, I argue, it must be specified in the lexicon. This is represented in the syntax through a Noun Marker (NM) head (equivalent to other categorizing heads such as n, see Harris, 1991; Kramer, 2016), which carries the gender feature for gendered nouns, but a +/-ATOMIC feature when a gender feature is not present. My analysis implies that atomicity is in the lexical entry of some nouns but not others, within the same language. Gendered nouns are able to grind and portion via context and are free to do so because atomicity is not part of the computation (see Landman, 2011; Falkum, 2010; Pelletier, Kiss, Husić, 2021; Kiss, Pelletier, Husić, & Poppek, 2017; Cheng, Doetjes, & Sybesma, 2008; Ramchand & Svenonius, 2008). The analysis also somewhat unifies gender and classification by placing the two features in the same noun-marking head, while still treating them as different features with different properties and semantic content.

Speaker: Tamirand Nnena de Lisser (University of Guyana)
Title: The Three Phases of Early Missing Subjects: Evidence from Creole Language Acquisition
Date: Thursday 6th April
Location: Lipsius 2.23/Zoom [online talk]
Time: 16:15-17:30

Abstract:
This talk will explore the null subject phenomenon in L1 acquisition of two Creole languages, Jamaican, a non-null subject language, and Morisyen, a language that allows null subjects in certain contexts. According to Bickerton (1981; 1984; 1999; 2014), the grammar of Creole languages may give an insight in the default parameter settings of the language faculty, and as such Creole-speaking children’s production should be errorless. Results from the cross-linguistic comparative analysis reveal striking similarities – Both Jamaican and Morisyen speaking children start with a system of target inconsistent missing subjects (Phase 1). This is however not related to Parameter (mis)setting or Bickerton’s Language Bioprogram Hypothesis, but rather a principle of economical structure building, accounted for by the modified version of the Truncation approach in terms of the Spell-Out mechanism (De Lisser et al., 2016). Around age three, when this system is accurately regulated by children, we note a shift to an overwhelming number of overt subjects in the children’s grammar (Phase 2). This developmental phase can arguably reflect what appears to be the default parametric option which is explored, regardless of the target being a null or non-null subject language. It is only later, based on evidence from the input, that children acquiring null subject languages fix the null subject parameter to reflect a target consistent grammar (Phase 3). The findings from the post-truncation phase of Creole language development (Phases 2 and 3) can therefore be argued to be supportive of the assumption that overt subject is the default unmarked option, and may contribute to discussion on Bickerton’s Language Bioprogram Hypothesis and parameter (mis)setting.

Speaker: Elisabeth J. Kerr (LUCL)
Title: Tunen syntax within a structural typology of Aux-O-V word orders
Date: Thursday 16th March
Location: Lipsius 2.23
Time: 16:15-17:30

Abstract:
Aux-O-V word order is an interesting case of disharmonic word order, where a head-initial TP (Aux-V) dominates a head-final VP (O-V). Aux-O-V disharmony has attracted attention in the theoretical syntax literature as an instantiation of disharmonic word order compatible with the Final-Over-Final Condition (FOFC; Biberauer, Holmberg & Roberts 2014). The Aux-O-V type of disharmony has been analysed in such Kaynesian frameworks as derived through roll-up movement triggered by a formal feature on heads within the extended projection of the verb, e.g. the ^-feature in Biberauer, Holmberg & Roberts (2014), as discussed predominantly for Germanic and Uralic varieties (see a.o. Sheehan et al. 2017, Roberts 2019). Other analyses of Aux-O-V are found in the Africanist literature, where most work has been on West African languages with S-Aux-O-V-X word order, often existing in alternation with SVO (see e.g. Sande et al. 2019 for a generative overview and Creissels 2005, Güldemann 2018 a.o. for descriptive overviews).

In this talk I present my thesis work on the analysis of Aux-O-V in Tunen, the only Bantu (Niger-Congo) language with S-Aux-O-V-X as the canonical word order, found in matrix (and embedded) clauses, across tense/aspect contexts, and in multiple information-structural contexts. The empirical foundation is field data I collected in Cameroon between 2019 and 2022 as part of the Bantu Syntax and Information Structure project. I discuss the three main types of formal analyses: first, a roll-up account as previously applied to Germanic, second, an account which generates Bantu verbal morphology via head movement (e.g. Zeller 2013, van der Wal 2022), which must then be modified via object shift in order to additionally derive Tunen’s OV order (in contrast to canonical Bantu SVO), and third, a non-Kaynesian account allowing for base-generation of OV order within the VP. I then raise some challenges for discussion in choosing between these analyses, namely (i) the proper derivation of Bantu verbal morphology, (ii) accounting for Tunen’s O-V-X order, (iii) the position of subjects, and (iv) the availability of postverbal discontinuous nominal modifiers in Tunen. In showing how previous analyses of Aux-O-V must be adapted to fit the Tunen data, I reflect on the structural typology of S-Aux-O-V(-X) derivations crosslinguistically, which has implications both for understanding word order change in West/Central Africa and word order variation more broadly.

References
Biberauer, Theresa, Holmberg, Anders & Ian Roberts. 2014. A syntactic universal and its consequences. Linguistic Inquiry, 45(2): 169-225.
Creissels, Denis. 2005. S-O-V-X constituent order and constituent order alternations in West African languages. In Rebecca T. Cover & Yuni Kim (eds.), Proceedings of the annual meeting of the Berkeley linguistics society, 31:2, 37-52.
Güldemann, Tom. 2008. The Macro-Sudan Belt: Towards identifying a linguistic area in northern sub-Saharan Africa. In Heine, Bernd and Derek Nurse (eds.). A Linguistic Geography of Africa, 151-185, Cambridge: Cambridge University Press.
Sheehan, Michelle, Biberauer, Theresa, Holmberg, Anders & Ian Roberts (eds.). 2017. The Final-Over-Final Condition: A syntactic universal. Cambridge, Mass/London: MIT Press.
Roberts, Ian. 2019. Parameter hierarchies and Universal Grammar. Oxford: OUP.
Sande, Hannah, Baier, Nico & Peter Jenks. 2019. The syntactic diversity of SAuxOV in West Africa. In Clem, Emily, Jenks, Peter and Hannah Sande (eds.). Theory and description in African linguistics: Selected papers from the 47th Annual Conference on African Linguistics, 667-701, Berlin: Language Science Press.
van der Wal, Jenneke. 2022. A featural typology of Bantu agreement. Oxford: OUP.
Zeller, Jochen. 2013. In defence of head movement: Evidence from Bantu. In Cheng, Lisa and Norbert Corver (editors), Diagnosing syntax, 87-111, Oxford: OUP

We are looking forward to a full programme of ComSyn talks in the upcoming season, running from March-June. All talks will be held at the usual timeslot of Thursdays, 16:15-17:30 CET (aside from 26th April; see below).  Here is the list of dates and speakers:

*Note that the talk on 26th April is held on a Wednesday, in order to avoid the King’s Day holiday on 27th April, and therefore is in a different room. The timeslot is unchanged (16:15-17:30).

Looking forward to seeing you there!

– Lis (on behalf of the ComSyn organising committee)

Speaker: Leonel Tadjo Fongang (Universität Leipzig)
Title: Towards a Unified Theory for Noun Class Agreement in Grassfields Bantu
Date: Thursday 8 December
Location: Lipsius 208
Time: 16:15 – 17:30

Abstract
DP-internal agreement in Grassfields Bantu languages exhibits cross-linguistically rare patterns. In canonical order, where the agreeing modifiers (possessive and demonstrative pronouns, for example) generally follow the noun, we find regular agreement in noun class. In the reverse order, we get what, on the surface, looks like (1) regular agreement (Yemba and Medumba), (2) reduced/impoverished agreement (Ngemba) (3) zero agreement (Shupamem?) or (4) more agreement than expected (Aghem and Nweh). In this talk, I argue that Feature Gluttony (Coon & Keine 2021) may provide the correct setting for proposing a unified account of the observed patterns. I show, specifically for Ngemba, that noun class probes in surface N-POSS/DEM order (canonical order) agree with and copy exactly the features they need for vocabulary insertion in the post-syntax. In POSS/DEM-N order, they agree with and copy more gender features than they need for vocabulary insertion. This, I claim, creates a conflict in the morphology, as there is no vocabulary item that can spell out all the features on the terminal node made available to the morphology by syntax (Morphological Ineffability; Coon & Keine 2021). I adopt a repair mechanism that necessarily assumes Impoverishment before vocabulary insertion for Ngemba. Moreover, I provide empirical evidence from Nweh that Feature Gluttony might be the correct way to look at intricate patterns of noun class agreement in Grassfields Bantu in general. languages only differ in the type of repairs they apply to Feature Gluttony. While Ngemba, for example, does Impoverishment, Nweh does Fission which, in turn, presupposes multiple exponence.

Speaker: Zhen Li (LUCL)
Title: Word order, information structure and agreement in Teke (Bantu B70)
Date: Thursday 1 December
Location: Lipsius 208
Time: 16:15 – 17:30

Abstract
In this talk, I will discuss the information structure and its interactions with syntax in Teke-Kukuya, which is a Narrow Bantu language (Guthrie code B77) spoken in the Republic of Congo. I will identify a dedicated immediate-before-verb (IBV) focus position in this language and show its interpretational properties. Based on some segmental and tonal evidence, I suggest that the IBV focus construction has its historical origin in a cleft. I will also propose the syntactic derivation of this preverbal focus strategy. There is also an intriguing a-/ka- class 1 subject marking alternation which co-varies with preverbal focus and word order, and I will provide an approach to explaining this phenomenon. The data presented is based on my field work in 2019 and 2021.

Speaker: Marijke De Belder (Universiteit Utrecht)
Title: Dutch primary compounding: towards a new inventory
Date: Thursday 17 November
Location: Lipsius 208
Time: 16:15 – 17:30

Abstract
I present an overview of my work on Dutch primary compounding. I show that there are three basic types, which are defined by a different syntax for the compound’s left-hand part (i.e. the non-head): there are bare roots as non-heads, roots which are nominalized by means of a nominal class marker and roots which incorporate into a D-layer, resulting in referential semantics for the non-head, which I argue is an option. I argue that the following facts are captured by the proposal: the distribution of so-called linking elements, the resemblance between linking elements and plural markers and ‘erroneous reference’ in Dutch. The proposal also brings the morphosyntax of Dutch NN compounding closer to what is known on NV incorporation (see Baker 1988).

Speaker: Romane Pedro (Nantes Université)
Title: The impact of the French wh-in-situ option in the acquisition of L2 English questions: An analysis of transfer
Date: Thursday 3 November
Location: Lipsius 208
Time: 16:15 – 17:30

Abstract
When learning a new language, being able to produce questions is essential. Question formulation and the linguistic similarities and differences it demonstrates across languages is thus becoming a crucial issue to Second Language Acquisition (Omane & Höhle 2021). Studying the case of L1 French learners of L2 English, my study deals with the acquisition of this phenomenon. The two languages under study involve differences in the production of questions, in terms of morpho-syntactic patterns, and in the information structure of questions, especially in the case of wh-questions (Engdahl 2006). While English wh-questions demonstrate an almost obligatory wh-fronting, the French language offers a wh-in-situ option when asking for information (Glasbergen-Plas 2021). In my talk, I will discuss the early stages of my investigation on question formation in L1 French and L2 English with a special focus on wh-in-situ questions, as in (1).

I will discuss primary results of the oral production experiment I conducted on native speakers and learners of the two languages examined. I will argue that the cross-linguistic differences between French and English questions may cause negative transfer from L1 French. I may also introduce the upcoming developments to my research project, including the addition of a third language for a better comparison and conclusion on the effect of the L1 in formulating questions in L2 English.