RESEARCH INSTITUTE FOR LINGUISTICS (HAS)
PÁZMÁNY PÉTER CATHOLIC UNIVERSITY
BENCZÚR UTCA 33, BUDAPEST, 1068, HUNGARY
MIKSZÁTH TÉR 1, BUDAPEST, 1088, HUNGARY
Syntax, semantics, radically truncated clauses, dependent indefinites, argument structure, voice, grammaticalization, imperatives.
I consider myself a problem solver. What I enjoy most is identifying and cracking challenging and potentially rewarding problems: unexplained, often unnoticed yet striking linguistic patterns which have the potential to reveal something deep about language.
Radically Truncated Clauses
What do Radically Truncated Clauses (minimal clauses which lack tense and modal marking, phi feature agreement, accusative case assignment and higher functional projections such as focus) and similar constructions tell us about the fine structure of the minimal VP (and much else) in Hungarian and beyond?
2021. Radically Truncated Clauses in Hungarian and Beyond: Evidence for the Fine Structure of the Minimal VP. Syntax 24:3, 376–416. (supplementary material)
2022. Radically Truncated Clauses – The Key to Various Mysteries of Hungarian Syntax? CIFU13, University of Vienna. (presentation)
I try to keep track of RTC-ish constructions in various languages. The list is short as of now but I expect it to grow. Interestingly, it seems that the locus of truncation can vary significantly:
Hungarian: VPs (reported by Halm 2021)
German: vPs (the so-called Inflektiv, reported by Teuber 1998, Schlobinski 2001, Bücking & Rau 2013)
Amarasi: vP+TAM+Neg (as reported by Tan (2022), Copular Pronoun Doubling constructions in Amarasi may be analyzed as Radically Truncated Clauses, an option considered but not adopted by Tan (2022))
Virtual performatives (e.g. *jumps in excitement*, *sings myself happy birthday*) have unique syntactic characteristics such as a silent subject and split agreement phenomena. I argue that virtual performatives are full, non-truncated clauses, and their special behaviour is due to two factors: i) their subjects are first person singular personal pronouns modified by an adjectival temporal expression and ii) they are unspecified for tense.
2022. Modified pronouns and split agreement: the syntax of virtual performatives. In Breanna Pratley, Özge Bakay, Eva Neu & Peyton Deal (eds.): Proceedings of NELS 52, Vol 1, 319-329.
Are all phrases single-headed? Apparently not! In my talk, I argue that so-called cocompounds are in fact the complex heads of two-headed phrases. I will show that, while seemingly exotic, two-headed phrases are actually completely in line with our understanding of endocentricity and the projection principle, as long as they meet some fairly sensible conditions.
In prep. The curious case of the two-headed phrase: A generative approach to co-compounds. (abstract)
What are the semantic and syntactic configurations that condition the distribution of free-choice items (FCIs) in Hungarian? The study of Hungarian, a language with a rich and well-articulated left periphery, has the potential to significantly contribute to our cross-linguistic understanding of FCIs, especially by providing a more rigorous syntactic perspective.
In prep, with Ágnes Bende-Farkas. The Birth of an Epistemic Indefinite: Vaegy in Transylvanian Hungarian. (conference handout, see also paper version in Hungarian)
In prep. Free Choice Items: The View from Hungarian. Studies in Generative Grammar Series, Mouton de Gruyter.
2021. Want, unconditionals, ever-free-relatives and scalar particles: the sources of free-choice items in Hungarian. Formal Diachronic Semantics 6, University of Cologne. (conference handout, see also paper version in Hungarian)
2019. The Semantics of Weak Imperatives Revisited: Evidence from Free-Choice Item Licensing. Acta Linguistica Academica 66 (4), 445–489.
2016. The Syntactic Position and Quantificational Force of FCIs in Hungarian. Acta Linguistica Academica, 63 (2). 241-276.
2016. The Grammar of Free-Choice Items in Hungarian. PhD Dissertation, Pázmány Péter Catholic University.
2015. Free Choice and Aspect in Hungarian. In: É. Kiss, Katalin – Surányi, Balázs – Dékány, Éva (eds.) Approaches to Hungarian. Volume 14: Papers from the 2013 Piliscsaba Conference. John Benjamins Publishing Company, Amsterdam, 167-185.
2013. Free Choice and Focus: FCIs in Hungarian. In: Surányi, Balázs (ed.): Proceedings of the First Central European Conference in Linguistics for Postgraduate Students. Pázmány Péter Catholic University, Budapest, 109-121.
In joint work with Anna Huszár, we analyze expletive negation in exclamatives in Hungarian, contributing to the cross-linguistic understanding of expletive negation, especially in terms of syntax and the syntax-semantics interface.
2021. Expletive Negation in Exclamatives - Evidence from Hungarian. Acta Linguistica Academica 68:4, 553-583.
The morphosyntactic encoding of middle voice (anticausatives, dispositional middles, reflexives, reciprocals and antipassives) has changed radically in the history of Hungarian. I am looking at how this can inform our cross-linguistic generalizations about the mechanisms underlying grammaticalization.
2020. Grammaticalization without Feature Economy: Evidence from the Voice Cycle in Hungarian. Diachronica 37:1, 1-42.
Almost-approximators have been analyzed in many languages in terms of intensionality vs. scalarity. I show that in Hungarian, there is a neat division of labour. Majdnem 'almost' is an intensional approximator (one proposition, alternative possible worlds, distance between possible worlds); whereas szinte 'almost' is a scalar approximator (alternative propositions, one possible world, distance between precision standards). The diachronic semantic pathways behind these two approximators are also explored.
To appear. Why almost and almost are not even approximately the same: The diachronic semantics of approximatives in Hungarian. In: Nora Boneh and Elitzur Bar-Asher Siegal (eds.): Language Change: Theoretical and Empirical Perspectives. Jerusalem Studies in Philosophy and History of Science: Springer.
Case and Agreement
Personal names and nicknames often preserve archaic features of phonology, morphology, and even syntax. Hungarian has a set of personal names (nicknames) that seem to preserve something interesting: a non-casemarked object plus a verb which exhibits definite object agreement. In my paper, I explore how this can help us reconstruct an aspect of Old Hungarian syntax: the decoupling of case and agreement.
Under review. Vegans, Teetotalers and the Decoupling of Case and Agreement in Old Hungarian.
Possessedness morphology is known to have an extra, non-possessive function in many languages: that of expressing definiteness/identifiability (without there being any possessor in sight). In Hungarian, the POSS.3SG suffix is found in affective demonstratives. I argue that this additional function can be shown to have developed via grammaticalization from a bona fide possessive construction.
2018. From Possessive Suffix to Affective Demonstrative Suffix in Hungarian: a Grammaticalization Analysis. Morphology 28 (4), 359-396. (supplementary material)
POSS.3SG suffixes are used to express various flavours of definiteness/identifiability in several Uralic languages. In this squib, I propose an acquisition-based account for this pattern, capitalizing on unique features of Uralic (possessedness suffixes, widespread pro-drop, no grammatical gender) and on general characteristics of child-directed speech strategies (the frequent use of imposters, relational constructions and physical deixis).
2021. Non-possessive use of POSS3SG in Uralic languages: an acquisition account. Squib. [unfortunately Hungarian is actually not the best language through which to explore my hypothesis, so I would be happy to team up with someone with expertise in other Uralic languages and language acquisition to develop this idea further]
Verbs of involuntary perception, recognition and physical disposition in Hungarian pattern together across four, seemingly unrelated, phenomena: 1) they cannot felicitously combine with an ability modal auxiliary, 2) they appear in dispositional middles, 3) they appear in so-called root infinitivals of circumstantial modality and 4) a large subset of them can function as modal auxiliaries. Connecting all these pieces of novel empirical evidence, I argue that in Hungarian, verbs falling into these classes have an enriched semantics: they are lexically specified as modal, with ability modality hard-wired into their semantics:
In prep. If you can see it you can’t help seeing it: On the inherent modality of verbs of involuntary perception, (re)cognition and physical disposition in Hungarian. [handout] [supplementary material][paper in Hungarian - English version coming soon-ish]
Semelfactive verbs in Hungarian show a mixed behaviour in terms of standard unaccusativity/unergativity diagnostics. I argued that this can be derived from the interplay of their argument structure (low agentivity, high affectedness) and their event structure (dynamicity).
2012. Unergative and/or Unaccusative: On the Argument Structure, Semantics and Syntax of Semelfactives in Hungarian. (2012) In: Surányi, Balázs - Diána Varga (eds.): Proceedings of the First Central European Conference in Linguistics for Postgraduate Students. Pázmány Péter Catholic University, Budapest, 104-117.
My Erdős number is 6. (Ágnes Bende-Farkas < Hans Kamp < Robin Cooper < Jon Barwise < Kenneth Kunen < Paul Erdős)