The Weekly Newsletter of MIT Linguistics

Issue of Monday, November 18th, 2019

Phonology Circle 11/18 - Sherry Yong Chen & Michael Kenstowicz (MIT)

Speakers: Sherry Yong Chen & Michael Kenstowicz (MIT)
TitleTones and Tone Sandhi in Longyou (Wu) Chinese
Date/Time: Monday (11/18), 5:00pm-6:30pm
Location: 32-D831
AbstractThis presentation is a progress report on our investigation of the tones and tone sandhi of the Southern Wu dialect of Longyou (Western Zhejiang Province). We document the Longyou tonal correspondences with Middle Chinese, their F0 shapes in citation form as well as four major sandhi changes appearing in disyllabic compounds. Longyou appears to be very conservative: the eight-tone system from Middle Chinese which splits into higher and lower registers, and the correlation between tonal registers and onset voicing, are both preserved intact. Time permitting, the data are considered in the light of recent typological studies of tone sandhi and tone change in the East Asian languages. ​

Syntax Square 11/19 - Elise Newman (MIT)

Speaker: Elise Newman (MIT)
Title: Mayan agent focus and the interaction between merge and agree (continued)
Time: Tuesday, November 19th, 1pm - 2pm
Location: 32-D461

Abstract: (Continued from 11/5) In this week’s edition of Syntax Square, I will discuss Mayan Agent Focus from the perspective of Coon, Baier, Levin 2019. In their paper, they propose that examples like (1), where a subject has been wh-extracted, are ungrammatical due to the so called Ergative Extraction Constraint (EEC).

(1) *Maktxel max y-il ix ix? who pfv A3s-see clf woman intended: ‘Who saw the woman?’

They propose that the EEC is active in some Mayan languages due to the fact that the object moves higher than the subject, and is a more local target for agree by a higher probe. While typically A’-movement is insensitive to intervening nominals, they argue that Mayan A’-probes are relativized to seek D features as well as wh/focus features (a mixed A/A’ probe). The result is that whenever a subject is marked with wh/focus features, the relevant probe searching for those features first agrees with the internal argument before finding the subject, and thus becomes gluttonous, which leads to a crash. The only way to pronounce (1) is to insert the agent focus morpheme in place of agreement with the moved subject.

(2) Maktxel max-ach il-on-i? who pfv-b2s see-AF-itv ‘Who saw you?’

On their approach, the agent focus morpheme licenses extraction of an ergative subject by blocking movement of the internal argument to a higher position, thus preventing the object from ever c-commanding the subject. In other words, the EEC is active in some Mayan languages, and agent focus appears in these languages only to prevent violations of the EEC.

This locality approach to the EEC and agent focus in Mayan is attractive because it builds on structural considerations that are well motivated by the Mayan literature and accounts for the fact that agent focus is sensitive to certain properties of the internal argument. However, their analysis of agent focus struggles to handle cases of multiple extraction where agent focus is present, thus suggesting that agent focus and the EEC may not be so tightly related. I will therefore propose the beginnings of a reanalysis of agent focus (very much still in progress) that builds off of their structural assumptions but disentangles agent focus from the EEC.

My proposal assumes a theory of merge and agree along the lines of Longenbaugh (2019), with the additional assumption that all merge tucks in (Richards 2005). These assumptions tightly constrain the order in which multiple specifiers may appear, given a head with particular selectional and EPP requirements. The result is that the subject is always the outer specifier of vP, unless it is more featurally specified than the internal argument, in which case the order of specifiers becomes reversed. In exactly these cases of reversal, both v and T end up agreeing with the same DP, namely the internal argument. I argue that these cases co-occur with the presence of agent focus because the morphology rejects haplology.

LF Reading Group 11/20 - Dóra Kata Takács (MIT)

Speaker: Dóra Kata Takács (MIT)
Title: A half baked Hungarian bagel of scalar additive particles
Time: Wednesday, November 20th, 1pm - 2pm
Location: 32-D461

Abstract: In this talk I look at the distribution of two Hungarian particles akár and is, which can separately and together give rise to scalar additive inferences. I am particularly concerned with the question what kind of NPI akár is.

MorPhun 11/20 - Stanislao Zompi’ (MIT)

Speaker: Stanislao Zompi’ (MIT)
Title: Božič (2019): “Strictly local Impoverishment: An intervention effect”
Time: Wednesday, November 20th, 5pm - 6:30pm
Location: 32-D831

Abstract: Languages that exhibit systematic patterns of morphological syncretism must involve a rule that derives such syncretism as a `deep’ property of the grammar, according to Harley (2008) and Nevins (2011). They show that, within Distributed Morphology (Halle & Marantz 1993), this needs to be derived by Impoverishment (Bonet 1991, Noyer 1992, Halle & Marantz 1994), which as a context-sensitive operation deletes feature F in the context of F. Nevins (2011) discusses Ljubljana Slovenian, and posits Impoverishment of the DUAL-number contrasts in the context of feminine gender. However, Nevins’ argument is only based on the relevant morphological paradigms in isolation and only their nominative Case forms. This papers provides more empirical context, viz. entire morphological paradigms from Ljubljana Slovenian, and also the interaction of the relevant syncretism with agreement patterns. While the agreement patterns confirm the post-syntactic nature of Impoverishment, the full morphological paradigms show that Impoverishment is systematically blocked in certain Case forms: while Impoverishment applies in the context of flexional morphology, it fails to do so in the context of agglutinative morphology. This pattern of blocking can be captured as an intervention effect if Impoverishment is limited to considering a strictly local Xº as context, viz. the closest Xº available in the c-command domain.

LingLunch 11/21 - Damian Blasi (Harvard Radcliffe Institute for Advanced Study)

Speaker: Damian Blasi (Harvard Radcliffe Institute for Advanced Study)
Title: The case for next-gen models of language change
Time: Thursday, November 21st, 12:30pm - 1:50pm
Location: 32-D461

Abstract: Drawing robust generalizations from language change data has been, with a few exceptions, a challenging task riddled with concerns about generalizability. A number of models of language change (inspired on diverse dynamics and frameworks from evolutionary dynamics to Bayesian learning) have opened the door to a particularly elegant solution to this problem, which I refer to as the bias-to-structure model. This general framework consists in detecting instantaneous biases in humans when learning, using or transmitting language or language-like behavior, and using the direction of the bias jointly with the worldwide distribution of the relevant linguistic structure as a way of arguing for robust history-independent pathways for language change. This popular approach has been deployed for explaining patterns involving trade-offs between morphological and syntactic marking of grammatical functions, the linear order of NP modifiers and the emergence of compositionality and regular morphological paradigms, among others. In this presentation I will summarize a number of challenges associated with this approach, ranging from the empirical adequacy of these language change models, to the generalizability of linguistic biases in the laboratory and the reliability of cross-linguistic frequency as an indicator of species-wide preferences. I conclude that, in spite of the fact that these approaches have helped moving forward our discussions and have yielded a plethora of interesting observations, concerns about ecological validity merit a re-examination of alternative models of language change.

Experimentalist Meeting 11/22 - Masoud Jasbi (Harvard)

Speaker: Masoud Jasbi (Harvard)
Title: Nativism vs. Constructivism: The Case of Disjunction
Time: Friday, November 22th, 2pm - 3pm
Location: 32-D461

Abstract: Disjunction has been a major source of insight for theories of meaning and language acquisition. Both nativist and constructivist theories have claimed that children’s development of disjunction conforms to their predictions. What are nativist and constructivist accounts of disjunction acquisition? How often do children hear disjunction in their parents’ speech? in what contexts? What type of learning model can succeed in learning the interpretations of disjunction from child-directed speech? In this talk, I review previous theories of disjunction acquisition and present the results of a study on naturalistic recordings of parent-child interactions. The results suggest that children may learn to interpret a disjunction by partitioning their form-meaning mappings based on salient cues that accompany a disjunction in child-directed speech. In order to better understand the distribution of “or” in parents’ and children’s speech, I collected statistics of its use across speakers, ages, and contexts. The results show that children start producing “or” between 18-30 months and by 42 months their productions plateau at a constant rate. I also show that the most likely interpretation of “or” in child-directed speech is exclusive disjunction. However, exclusive interpretations correlated with a rise-fall intonation, and logically inconsistent propositions. In the absence of these two cues, “or” was commonly not exclusive. Our computational modeling suggests that a hypothetical learner can successfully interpret an English disjunction by mapping forms to meanings after partitioning the input using the set of salient cues in the context of the utterance. I discuss the implications of this work for current theories of word learning and language acquisition.

Colloquium 11/22 - Ezra Keshet (University of Michigan)

Speaker: Ezra Keshet (University of Michigan)
Title: Pronouns in 3-D
Time: Friday, November 22nd, 3:30pm - 5pm
Location: 32-155

Abstract: This talk aims to demystify dynamic logic by tracing the development of a new plural logic step-by-step through 3 dimensions of meaning:

  1. Storing and retrieving single discourse referents (akin to names), each suitable for later reference via pronouns:
    Arthur saw Beth. She waved to him.
  2. Repeating this process along multiple parallel paths to explain pronoun reference to an antecedent indefinite (cf. Groenendijk & Stokhof 1991):
    A man saw a woman. (= Arthur saw Beth or Arthur saw Dara or Charlie saw Beth or Charlie saw Dara …)
    She waved to him.
  3. Accessing values along multiple paths at once to derive plural pronoun values and other more exotic effects (cf. van den Berg 1996):
    Every student donated a book. (= Arthur donated W&P and Beth donated C&P and Charlie donated P&P …)
    They are on that shelf. [they = W&P, C&P, P&P, …]

I will argue, perhaps unsurprisingly, that my new logic is simpler than existing analyses, while handling the same data plus new empirical cases. For instance, the logic can handle a class of sentences where a plural pronoun in the nuclear scope of a quantifier seems to refer to the very value being constructed by that nuclear scope, as in (4). I will propose that such cases relate to a straightforward account of reflexives such as each other.

(4) Almost every North Atlantic country agreed in a treaty that an attack on one of them constitutes an attack on all of them. [them = only the treaty signatories]


The Language Acquisition Workshop in New England took place at MIT last Sunday. Two groups of MIT students presented their work:


Sherry Yong Chen & Filipe His Kobayashi: Comprehending and: Lessons from Children’s Understanding of English Conjunction

Fulang Chen & Dóra Kata Takács: Interaction of negation and universal quantification in the grammar of 4-year-olds

Spadine defends!

Congratulations go to Carolyn Spadine, who defended her dissertation entitled “The structure of attitude reports: representing context in grammar” last week. The empirical core of her dissertation is a set of fascinating findings concerning Tigrinya, a Semitic language spoken in Eritrea. Tigrinya speakers can introduce a clause with an inflected element “ʔil-” that can translate the English verb ‘say’ (or ‘believe’) and take a subject of its own — but turns out not to be a verb at all, but a perspectival complementizer (a kind of subordinating conjunction) with a very different syntax and semantics. Clauses introduced by “ʔil-” also show the phenomenon called “indexical shift”, by which the meaning of pronouns such as “I” and “you” is not fixed as in English, but varies with context. Most excitingly, the syntax of indexical shift in Tigrinya can yield agreement mismatches, with a first or second person subject cooccuring with a third person verb — but only under very particular circumstances. Carrie’s dissertation shows how the details of these phenomena in Tigrinya support a novel theory of how pronouns come to be first or second person in the first place. Congratulations, Carrie!!


Abramovitz @ Petropavlovsk-Kamchatsky conference

Last weekend, fifth-year student Rafael Abramovitz was an invited plenary speaker at an international conference in Petropavlosk-Kamchatsky (the administrative center of Kamchatka) devoted to “The Preservation and Development of the Native Languages and Culture of Indigenous Minorities of the North, Inhabiting the Territory of the Kamchatka Region”. Rafael’s research focuses on the syntax, morphology, and phonology of Koryak, an endangered Chukotko-Kamchatkan language with about 1,700 speakers. The title of Rafael’s plenary talk (in Russian) was “Formal Linguistics and the Koryak Language” (Формальная лингвистика и корякский язык), and he also presented a second talk entitled “Problems of Koryak Language Instruction:” (Проблемы обучения корякскому языку).

While on Kamchatka, Rafael gave a fantastic interview to local media, in Russian, in which he describes how his interest in the language was awakened by the discovery of a Russian-language Koryak grammar in the library at the University of Chicago (where he received his undergraduate degree), and stresses the importance of studying and documenting languages like Koryak both for for the benefit of its speakers and for the benefit of the science of language. Watch his interview at either of the following links:


Rafael tells us that the structure behind him during this interview is a yayanga, the traditional dwelling of the reindeer-herding Koryaks.