The limits of variability in extraction asymmetries

• project C05 of the CRC 1287 Limits of Variability in Language: Cognitive, Grammatical, and Social Aspects(1st funding phase) at the University of Potsdam
• funding period: July 2017 -- June 2021, funded by the DFG (German Research Foundation)
• researchers: Doreen Georgi (PI), Johannes Hein (postdoc)

The project studies displacement asymmetries from a cross-linguistic perspective, in particular asymmetries between subjects and non-subjects. Subject displacement tends to be more restricted than non-subject displacement in that the former often requires special morphological devices or special syntactic constructions (e.g. valency-changing operations) to be expressed. Relatively well-studied asymmetries of this kind are the that-trace effect and the ban on ergative movement. This project focuses on a further instance, the "anti-agreement effect" (AAE) but also invetsigates other extraction asymmetries. In languages with the AAE, argument encoding morphology (case, agreement, argument doubling devices) are either supressed or reduced when the corresponding argument undergoes extraction (viz., A-bar-movement, such as wh- or focus movement) - at least according to the traditional descriptions of the phenomena in Ouhalla (1993).

The overall goal of this project is to find out what causes the AAE and extraction asymmetries more generally. Is there a (potentially universal) constraint in the grammer - whatever its nature - that disfavors subject displacement and that underlies all subject/non-subject extraction asymmetries in the languages of the world (these would then simply be different repair strategies applied to avoid violations of the same general constraint)? Are there certain properties in the grammars of individual languages that can be considered strong predictors for the occurrence of the AAE (because not all languages exhibit this phenomenon)? Are there implicational relations between the factors (if the AAE is triggered in context A, it will also be triggered in context B)? The latter question will inform us about the interaction of grammatical processes that lead to the AAE and hence about its causes. Finally, we intend to provide formal analyses of the AAE in the indivudual languages and evaluate existing approaches to the AAE on the basis of our findings.

To answer the above questions, we conduct a cross-linguistic in-depth study in about 15 languages from different language (sub)families with the AAE. We collect new data with native speakers in questionnaires and interviews in order to identify the factors that trigger or block the AAE. Then we compare whether (i) these factors are stable or highly variable across different languages that exhibit the AAE, and (ii) whether different types of extraction asymmetries have the same or different triggers.

AAE languages: Awing & Limbum (Niger-Congo, Grassfields Bantu) Asante Twi (Niger-Congo, Kwa) Kikuyu (Niger-Congo, Bantu) Nuer (Niger-Congo, Nilotic) Berber languages (Afro-Asiatic, Berber) Somali (Afroasiatic, Cushitic) Fiorentino (Indo-European, Romance) Welsh & Breton (Indo-European, Celtic) Turkish (Turkic) Tundra Nenets (Uralic, Samoyedic) Adyghe (North West Caucassian, Circassian)

Other asymmetries / languages studied: past participle agreement in French (with Elisabeth Stark, Univerity of Zurich) ATB-movement / reflexes of movement in Duala, Buli, Kitharaka and Ewe Focus marking and focus movement in Igbo (with Mary Amaechi) verb fronting/verb doubling

General results so far (wrt. the AAE)):
We found that - at least in some of the languages we study - the AAE can be explained by independent properties of the individual languages combined with the interaction of general syntactic and morphological operations. Given that these properties are language-specific, the explanations also differ from language to language, and the varibility in the behavior of the AAE across languages is in fact expected. As for the explanations, we propsoe that in some languages the AAE is the result of contextual allomorphy, in others it is related to the way in which the language forms A-bar-dependencies (via movement or base-generation, clause-bound vs. unbounded dependencies) and the nature of the argument encoding devices (agreement vs. pronominal clitic). Put differently, we claim that although the phenomenon looks very similar on the surface across languages (i.e., seems to be stable), it has different causes in different languages. Importantly, these findings suggest that it is not necessary - and in fact superfluous - to postulate construction-specific rules related either to A-bar-movement, A-bar-binding or A-bar-features of verbal arguments, as has been done in the previous literature. The core observation that leads to this conclusion is the fact that the AAE also arises in contexts in which the respective arguments are neither A-bar extracted nor hosts of A-bar-feature. Postulating A-bar-related analyses will not have anything to say about these contexts and must find a different explanation for why the same effect occurs there - these approaches thus miss a broader generalization about argument encoding in the respective languages. We offer analyses that capture all contexts, and, as a result, do not need to say anything special at all about the A-bar-contexts in which the AAE surfaces as well; these simply fall out from the morphosyntactic analyses of the argument encoding strategies in the language. Finally, we find further supporting evidence - highlighted in the recent literature on the AAE (see Baier 2017) - that the AAE does not only affect subjects but also objects (so there is in fact no asymmetry between subjects and non-subjects) and is not causally connected to movement of the respective arguments.

Papers:

• Georgi, Doreen (to appear): On the nature of ATB-movement: insights from reflexes of movement. To appear in Proceedings of NELS 49.
• Georgi, Doreen and Elisabeth Stark (to appear): Past participle agreement in French - one or two rules? In: M. Hinzelin, N. Pomino and E.-M. Remberger, eds, Formal Approaches to Romance Morphosyntax: Linking Variation to Theory. Linguistische Arbeiten, de Gruyter, Berlin / New York.
• Hein, Johannes (2019): Subject asymmetries in Limbum: Antiagreement, resumption and focus marking.
• Fominyam, Henry & Doreen Georgi (2019): On subject marking in Awing. Ms., University of Potdam.
• Georgi, Doreen & Johannes Hein (2019): On the interaction of extraction and argument marking in Asante Twi. Ms. in preparation, University of Potsdam.
• Amaechi, Mary & Doreen Georgi (2019): Quirks of subject (non-)extraction in Igbo. Glossa: a journal of general linguistics 4(1): 69, 1-36. DOI: https://doi.org/10.5334/gjgl.607 (Glossa Special Collection).
• Hein, Johannes (2018): Deriving the typology of verbal fronting. In: S. Hucklebridge and M. Nelson, eds, Proceedings of the Forty-Eighth Annual Meeting of the North East Linguistic Society. Vol. 2, GLSA Publications, Amherst, MA, pp. 29-38
• Amaechi, Mary & Doreen Georgi (2018): Focus marking strategies in Igbo. Submitted to Proceedings of ACAL 49, Michigan
• Hein, Johannes (in revision): On the lack of verb-doubling VP-topicalization in Germanic, Journal of Comparative Germanic Linguistics.
• Hein, Johannes (to appear): V(P)-fronting in Asante Twi and Limbum. In: V. Lee-Scheonfeld and D. Ott, eds, Parameters of predicate fronting: A crosslinguistic look at V(P)-initial constructions. Oxford University Press, Oxford.
• Hein, Johannes and Andrew Murphy (to appear): Case matching and syncretism in ATB dependencies, Studia Linguistica.
• Hein, Johannes and Katja Barnickel (2018): Replication of R-pronouns in German dialects, Zeitschrift für Sprachwissenschaft 37 (2), 171-204.

Talks:

• Hein, Johannes (July 2019): Subject asymmetries in Limbum. GGS 45, Goethe University of Frankfurt.
• Hein, Johannes (May 2019): Subject asymmetries in Limbum. ACAL 50, UBC Vancouver.
• Georgi, Doreen (May 2019): Reflexes of movement and locality. Invited talk at CamCos 8, Cambridge, UK.
• Fominyam, Henry & Doreen Georgi (May 2019): On the nature of subject marking in Awing . ACAL 50, UBC, Vancouver.
• Fominyam, Henry & Doreen Georgi (April 2019): On the role of referentiality in subject marking in Awing. DISCO workshop, Leipzig.
• Amaeci, Mary & Doreen Georgi (March 2019): On "optional" wh-/focus fronting in Igbo: a SYN-SEM-PHON interaction. Workshop Who cares? Contrast and opposition in 'free' phenomena at the 41st Annual Conference of the German Linguistic Society (DGfS), Bremen.
• Georgi, Doreen (October 2018): On the nature of ATB-movement: insights from reflexes of movement. NELS 49, Cornell University, Ithaca.
• Hein, Johannes (September 2018): On the interaction of head movement, ellipsis, and copy deletion: The case of Mainland Scandinavian. CGSW 33, Göttingen.
• Georgi, Doreen (April 2018): On the nature of ATB-movement: insights from reflexes of movement. GLOW 41, Budapest.
• Amaechi, Mary & Doreen Georgi (March 2018): On the source of subject / non-subject asymmetries in Igbo wh-/focus constructions. ACAL 49, Michigan.
• Amaechi, Mary & Doreen Georgi (March 2018): On subject / non-subject extraction asymmetries in Igbo. Workshop Referential and relational approaches to syntactic asymmetries at the 40th Annual Conference of the German Linguistic Society (DGfS), Stuttgart.
• Amaechi, Mary and Doreen Georgi & (November 2017): On subject / non-subject extraction asymmetries in Igbo. Invited talk at the SVM lecture series, University of Potsdam.
• Hein, Johannes (October 2017): Deriving the typology of predicate fronting (Poster presentation). NELS 48, Reykjavik.
• Hein, Johannes (October 2017): On the correlation of V(P) fronting and verb doubling/do-support. SinFonIJA 10, Dubrovnik.
• Hein, Johannes (September 2017): Why Germanic VP-topicalization does not induce verb doubling. CGSW 32, Trondheim.
• Amaechi, Mary and Doreen Georgi & (August 2017): On subject / non-subject extraction asymmetries in Igbo. Invited talk at the Workshop on Quirks of Subject Extraction, National University of Singapore.