Whamit!

Speaker: Si Berrebi (MIT)

Title: Bi-dialectal speakers store a sound category they never produce: An investigation of an ongoing merger in Modern Hebrew

Time: Thursday, April 3rd, 12:30pm - 2pm

Location: 32-D461



Abstract: Bi-dialectal speakers typically end up producing the more prestigious dialect within their community and showing a processing advantage for words when spoken in that dialect. This was shown for case studies in which the prestigious dialect preserves more segmental contrasts than the non-prestigious one. We study a case in Modern Hebrew in which two consonants are merged in the prestigious dialect while remaining distinct in a stigmatized dialect. Also, as the stigmatized dialect is consistent with the orthography, all speakers know which word contains which sound category in this dialect. We tested three groups of speakers: mono-dialectal merging speakers, bi-dialectal speakers who produce the distinction and bi-dialectal speakers who do not. Using a lexical decision task in which the sound categories had been switched in some words, we show that (i) bi-dialectal speakers maintain the non-merged contrast in their lexical representations, even if they do not produce the distinction; (ii) bi-dialectal speakers who produce the distinction are additionally sensitive to mismatches between talker identity and heard variant.

Speaker: Adam Przepiórkowski (Warsaw/MIT)

Title: On the origin of cases

Time: Tuesday, April 1st, 1pm - 2pm

Location: 32-D461



Abstract: Do grammatical cases exist, or are they just an epiphenomenon, a perceived result of flagging an NP with an additional category (Pesetsky and Torrego 2001, Pesetsky 2013) or with an additional set of ? features (Clem and Deal 2024)? If they exist, what is their origin on NPs? In this talk I’ll argue that cases do exist but may sometimes originate from within NPs, rather than being assigned externally. I will in particular propose that this explains the long-standing puzzle of the source of accusative on numeral subjects in Polish.

Speaker: Christopher Legerme (MIT)

Title: Non-Optimization and Locality in Phonology

Time: Monday, March 31st, 5pm - 6:30pm

Location: 32-D831



Abstract: Drawing on recent proposals from Storme & Otilien (2022) and Lahrouchi & Ulfsbjorninn (2024), I discuss various linguistic patterns of Haitian Creole (HC) to illustrate how different grammatical architectures account for “null alternations” and apparent “non-optimizing” behavior in the language. For instance, in HC the alternation between the two forms of the definite determiner, la and a, has traditionally been analyzed as a case of phonologically conditioned suppletive allomorphy: la surfaces after consonant-final stems, while a appears after vowel-final stems. However, it is often pointed out that this distribution fails to align with phonotactic optimization, as it results in marked VV and CC sequences that contrast with the language’s strong preference for CV syllables. The la ~ a alternation resists resolution when treated in isolation; instead, it forces us into a broader look at the nature of grammatical representation and the theoretical constructs invoked to explain the interaction of phonetic silence and phonological locality. Therefore, we will turn to the broader landscape of null alternations, patterns in HC where phonological material is variably silenced depending on structural context. Along the way, we consider two competing approaches: a parallel OT architecture that invokes morphological optimization (Storme & Otilien 2022), and a representational framework rooted in skeletal licensing and floating segments (Lahrouchi & Ulfsbjorninn 2024). The la ~ a alternation, when situated among the broader set of null alternations in the language, offers a critical vantage point for evaluating competing theoretical commitments. While the theories to be discussed each capture important properties of the puzzles for which they were developed to explain, I argue that a phonological rather than morphological or “morphonological” account offers a better unified explanation for locality disruption and the structural conditions on segment realization in HC.

The 22nd annual Computational and Systems Neuroscience (COSYNE) conference was held at Montreal, and Mont Tremblant, Quebec, Canada from 27 March to 1 April 2025. Our first-year PhD student Heidi Durresi presented her work with others (Authors: Linghao Kong, Heidi Durresi, Lu Mi, and Nir Shavit) at the conference, entitled “Presynaptic input synchrony at scale”! You can read the abstract here.

Our sixth-year PhD student Margaret Wang was an invited speaker at the 3rd workshop on Charting Honorific and Addressee Morphosyntactic Processes (CHAMP3) held at Universitat Pompeu Fabra from March 28 to 29, 2025. Her talk was entitled “Politeness as a part-time vocation for vacuous linguistic forms”, and here is the abstract:

This talk will describe and explain how various expressions of politeness, whether phi-featural or
not, are recruited as such. Presenting various diagnostics, I argue that semantic vacuousness is a
necessary condition for forms recruited for the task of politeness. Invoking typological
asymmetries and economy principles, the talk will also make the case for politeness necessarily
being a part-time vocation (arising only via pragmatic principles), instead of a full-time one
(which would involve dedicated linguistic features).

The 47th meeting of Generative Linguistics in the Old World (GLOW) was held at the Goethe University in Frankfurt and the University of Göttingen on March 25-28, 2025. Here are the presentations made by current students, faculty and alumni:

• Yiannis Katochoritis (2nd year): If Distributivity is Variable Binding, Scope Reconstruction must be Syntactic (abstract)
• Gianluca Porta & Elise Newman (faculty, PhD 2021): Ne-cliticization and the DP/PP distinction: A case for Q (abstract)
• Shoichi Takahashi (PhD 2006)[Aoyama Gakuin University]: What Prepositional Object Gaps Tell Us about Merge and Linearization (abstract)
• Isabelle Charnavel & Dominique Sportiche (PhD 1984)[UCLA]: On Semantic Agreement (abstract)
• Isabelle Charnavel, Tom Meadows, Dominique Sportiche (PhD 1984)[UCLA]: Interpreted Agreement in Indexical Binding (abstract)
• Silvia Silleresi, Itai Bassi (PhD 2021)[ZAS/Ben Gurion University], Abigail Anne Bimpeh, Imke Driemel, Johnson Folorunso Ilori & Anastasia Nuworsu: Pronoun preferences unmasked: an experimental study on Ewe and Yoruba (abstract)
• Colin Davis (PhD 2020)[Nord University]: Punctuated movement facilitates (multi-)clausal pied-piping in Albanian (abstract)
• Danfeng Wu (PhD 2022)[Oxford]: Two paths to correction (abstract)
• Cory Bill, Imke Driemel, Andreea Nicolae, Kazuko Yatsushiro, Napoleon Katsos & Uli Sauerland (PhD 1998)[ZAS]: Not all children find no subjects hard: A cross-linguistic investigation of children’s negative indefinite production (abstract)

Our colleague Norvin Richards had not one but two papers published last week!

The first, entitled “Reconstructing Stress in Wôpanâak”, appeared in the International Journal of American Linguistics. Here is the abstract:

Wôpanâak (also called Wampanoag, Massachusett, Natick), the original language of much of eastern Massachusetts, is known to us from a variety of seventeenth, and eighteenth-century documents, which mostly do not indicate the position of word-level stress. We do have one poem in Wôpanâak, a translation of the book of Psalms into verse. In this paper I present a description of the rules for stress in Wôpanâak, based largely on an analysis of the metrical version of Psalms.

In social media comments, Norvin thanked the members of the Wôpanâak Language Reclamation Project “whose work on this language has been a constant inspiration for me” — and described the paper more informally this way:

It’s about how to figure out which syllable to stress in words in Wôpanâak, the language of the traditional owners of a lot of eastern Massachusetts. Figuring this out is a challenge, because the language spent over a century not being spoken at all, so we know it from 17th- and 18th- century documents, mostly religious documents, which don’t indicate stress. Fortunately, there is one surviving poem in Wôpanâak. Unfortunately, the poet in question was not a terrific poet, and the meter is very uneven. Fortunately, the poem is also very long! So if you are sufficiently stubborn, you can scrape together enough information to figure out the system.

Norvin’s second paper to appear last week is entitled “Finding something to lean on”, and was published in Language, the flagship journal of the Linguistic Society of America.  Here’s the abstract:

Phrases in a number of syntactic contexts are required, in a variety of languages, to end in their heads. This article offers a unified theory of the relevant properties of these contexts and of why the phenomena in question, while widespread, are not completely universal. The theory makes use of proposals made independently in contiguity theory (Richards 2010, 2016): the relevant syntactic contexts are argued to involve a prosodically dependent element that must attach prosodically to the head of the phrase to its immediate left, and this attachment is often blocked if the phrase in question is not head-final.

Congratulations, Norvin!