Whamit!

The Weekly Newsletter of MIT Linguistics

Issue of Monday, September 5th, 2022

Colloquium - Ryan Bennett (UC Santa Cruz)

Speaker: Ryan Bennett (UC Santa Cruz)
Title: Vowel Deletion as Grammatically-Controlled Gestural Overlap in Uspanteko
Time: Friday September 9, 3:30pm

Abstract: Uspanteko (Mayan) is spoken by ~5000 people in the central highlands of Guatemala. Unstressed vowels in Uspanteko often delete, though deletion is variable within and across speakers. Deletion appears to be phonological, being sensitive to phonotactics, foot structure, vowel quality, and morphology; and being largely insensitive to speech rate and style. But deletion also appears to be phonetic, being variable, gradient, insensitive to certain phonotactics, and opaque with respect to accent placement. Electroglottography data suggests that even apparently ‘deleted’ vowels may contribute voicing to [C(V)C] intervals, albeit inaudibly. We thus analyze deletion as grammatically-controlled gestural overlap, which masks vowels in [CVC] contexts, either in the phonology proper (e.g. Gafos 2002) or as part of a grammar of phonetic interpretation (e.g. Kingston & Diehl 1994).

Welcome, Ling-22!

Let us all give a big welcome to the incoming cohort, ling-22! 

Johanna Alstott: Hello MIT Linguistics! My name is Johanna Alstott (she/her). I’m originally from Connecticut, but Cambridge has already become a second home to me because I did my undergrad at Harvard. My main linguistic interests concern language acquisition as well as semantics, pragmatics, and their interface. I have a particular soft spot for experimental methods, with my main research to date consisting of experiments probing the semantics of adverbial quantifiers like sometimes and usually. My main hobby outside of linguistics is classical oboe, though I also love Dungeons & Dragons and consuming art in all its forms. I am really excited to get started at MIT!
 
Zachary Feldcamp: My name is Zachary Feldcamp (he/him/his). I’m from New Jersey and have recently completed a master’s in Linguistics at the University of Toronto. My primary research interest is the morphosyntax of the noun phrase. In particular, I have been working on a general account of linkers, drawing on data from Tshiluba and other languages. I am originally a classicist, so I am also interested in classical Latin and Greek. I have also worked on Middle Elamite, a language isolate from the ancient Near East with a unique linker system. Outside of linguistics, I enjoy playing and listening to jazz.
 
Yurika Aonuki: I’m Yurika Aonuki (she/her). I’m originally from Japan. I did my BA and MA at University of British Columbia, where I started doing fieldwork on Gitksan (Tsimshianic). My main area is semantics, and I have worked on tense and aspect in Gitksan, Japanese, and English. I also have interests and research experience in American Sign Language. Recently, I started eliciting degree constructions in Gitksan. Outside of linguistics, I like to dance (contemporary and ballet especially) and play with rabbits (and cats and dogs if they’re not intimidating).
 
Xinyue (Cynthia) Zhong (she/her): I grew up in Beijing, China and moved to California when I was 13. My current interests are in phonetics/laboratory phonology and bilingualism. Outside of linguistics, I enjoy playing/arranging music, doing translations, and video games :)
 
Bergül Soykan: I am Bergül Soykan and my pronouns are she/her/hers. I am coming from Turkey and I got my BA and MA in Boğaziçi University. I am mostly into semantics, particularly the semantics of conditionals and counterfactuals. Other than linguistics, I enjoy reading detective stories, watching action movies/series and traveling to new places.

Juan Cancel: Hi, I’m Juan D. Cancel (he/him/his) and I’m one of the new incoming graduate students here at MIT. I was born in Puerto Rico and lived there for many years, though for the past four years I’ve been living in the Philadelphia area. My linguistic interests span syntax, morphology and language typology, while my language interests revolve around Chukokto-Kamchatkan, Inuit-Yupik-Unangan, and Celtic. As for non-linguistic interests, I enjoy reading history and philosophy, playing grand-strategy games and watching series on Netflix.

Zhouyi Sun: I am from Yuyao, a city in eastern China. I got an MA in linguistics at Queen Mary University of London. My main interests lie in syntax and morphology. Outside of linguistics I enjoy watching comedy sketches, and one of my current favorite artists is none other than Liz Truss.

Runqi Tan: I’m from China. My research interests started with modelling phonological form and structure with optimization models, and I’m fascinated by the general principles that shape the language systems. I enjoy spending time with friends, reading science, history, biography, watching cartoons and going to musicals and concerts.

Haoming Li: My name is Haoming Li. My pronouns are he, him, his. I am from Chengdu, China. My fields of interest in linguistics are syntax and semantics, and in particular, Chinese (Mandarin) syntax and semantics. Outside of linguistics, I enjoy listening to classical music, recreational programming, and collecting mechanical keyboards.

Taieba Tawakoli

Ukhengching Marma

Staniszewski defends!

Congratulations to Frank Staniszewski, who successfully and excellently defended his dissertation on August 25, 2022, titled Modality and Time in Logical Context

The dissertation develops a theory of neg-raising that unifies the phenomenon with existing theories of free choice and negative polarity items. The empirical focus is on “until”-phrases and on the neg-raising predicates “want”, “should”, and “be supposed to”. Predictions of the formal account are then examined in a language acquisition experiment.

Baron defends!

Many congratulations to Christopher Baron, who successfully and excellently defended his dissertation on August 15, 2022, titled The Logic of Subtractives, or, Barely anyone tried almost as hard as me!

The dissertation analyzes the elements “almost and “barely”, proposing a formal analysis in which they are subtractive modifiers of quantifiers that via exhaustification result in exceptive meanings. The resulting theory is then used to examine the compositional structure of comparative and equative constructions as well as numeral constructions.

24.943, Fall 2022: “Aspects of Haitian Creole in minds, in history and in society”

24.943, Fall 2022: Aspects of Haitian Creole in minds, in history and in society

Michel DeGraff

Wednesdays 10AM–1PM, Room 32-D831

“In carrying out field research, linguists are inevitably responsible to the larger human community which its results could affect… What matters is eventual success, and that will be measured by the extent to which work on the language is integrated in a meaningful way into the life of the community of people who speak it.” (Ken Hale, 2001)

 

Background and Course Description:

First, some personal and historical background to unveil my positionality and situate the aspects of this course that are rooted in my biography and ongoing concerns as a linguist, educator, activist, Haitian and human being; then some details about topics, requirements, calendar, etc.

2022 is another eventful year for me as a Haitian linguist and for Haiti and Haitian Creole (“Kreyòl”) for a few reasons.  Here I’ll mention six of these.  The first one is the personal that’s political, the latter five bear on history, socio-linguistics and education—all of these contribute to the political context of the course, keeping in mind Ken Hale’s exhortation above:

  1. This year (2022) I’ve been both selected as a Fellow of the Linguistic Society of America (LSA) and nominated to stand for election for a seat on LSA’s Executive Committee.  I take both of these as kudos to my work on Kreyòl with the MIT-Haiti Initiative.  The challenges, opportunities and documents related to this Initiative will provide some of the backdrop and resources for this course.

 

  1. On February 22, 2022, DuoLingo released its Haitian Creole online course—which will be another important resource for this course.

 

  1. On April 1, 2022 (April’s Fool Day) Prof. Rochambeau Lainy, a linguist and a founding member of Akademi Kreyòl Ayisyen (AKA), made a public plea that French should remain as the primary language of instruction in Haiti, a country where most students (and teachers!) are fluent in Kreyòl only.  Later that year, AKA’s scientific commission has made proposals to alter the structure of Kreyòl’s phonemic alphabet and to adopt letter names that are derived from French, thus distancing the alphabet from its phonemic bases.

 

  1. On April 6, 2022, Ambassador Dominique Dupuy, Haitian Delegate at UNESCO in Paris, took a spectacular public stance against linguistic discrimination in UNESCO’s survey for their World Language Atlas whereby Creole languages (alongside Pidgins, Mixed Languages and Language Isolates) were mis-classified outside the “Language” category.

 

  1. It’s now been 40 years since the most transformative and (in?)famous reform in the history of education in Haiti: the 1982 reform by Mr. Joseph C. Bernard who was Minister of National Education in Haiti from 1979 to 1982.  He was fired in 1982 by Dictator / President-for-Life Jean-Claude Duvalier soon after he (Bernard) began efforts at implementing this reform whereby Kreyòl was to become a language of instruction in the first 10 years of schooling.  Now, Minister Nesmy Manigat, the current Minister of Education since November 2021, has valiantly relaunched efforts to implement the Bernard Reform.  So this course will be proceeding in parallel with Minister Manigat’s efforts toward the use of Kreyòl to open up quality education for all in Haiti, alongside often vociferous invectives against the claim that Kreyòl is a “normal” language instead of an expressively inadequate deformation of French. These invectives echo similar attacks, even among linguists, against the validity of Kreyòl and other Creole languages throughout their history.

 

  1. 2022 is also when the New York Times (on May 25, 2022) published a historic series of articles on the Ransom that King Charles X of France levied against Haiti in 1825 as an indemnity for repayment of the “lost property” of the white colonists who were ousted in 1804 as a result of the successful Haitian Revolution against colonization and slavery.  As the NYT documents, France’s syphoning Haiti’s financial reserves from the birth of our nation (a ransom that forced Haiti into endless cycles of debts and corruption) made it impossible for Haiti to ever have any opportunity for sustainable development and sovereignty.  The NYT articles also show that this ransom was orchestrated by France as one way to punish Haiti for having successfully defied white supremacy and having started the #BlackLivesMatter movement—long before the hashtag era.  This is the first time in history that an article in a major international newspaper was translated into Kreyòl.


In turn, the NYT “Ransom” articles raise questions that connect issues in all six bullet points above. To start with, let’s ask:   

 

Why would the Haitian élites in power in the early 19th century ever agree to, and even celebrate, Charles X’s ordinance for such an abominable ransom whose amount even included the “property” value of the enslaved Africans that liberated themselves?

 

The answer is very complex, but my contention, as I follow the analyses of anthropologists, sociologists and historians such as Michel Rolph Trouillot, Jean Casimir and Alex Dupuy, is that such consent on the part of the Haitian State in the 19th-century was also a tacit consent to participate in white supremacy against the Haitian nation and the Haitian Revolution’s promise of “equality for all.”  This Francophile élite did not recognize themselves in the language, culture and political and economic interests of the broader Kreyòl-speaking population.  Furthermore the racially motivated oppression against the Haitian nation is connected to the reasons why the 1982 Bernard Reform (for the benefit of the Kreyòl-speaking majority in Haiti) has been such a challenge to implement, up to today, in spite of its robust scientific foundations in linguistics and education.

 

In turn, the ambivalence toward Kreyòl as language of instruction is connected to the psychological and social consequences of the brutal transatlantic slave trade that triggered the formation of Creole languages in Africa and the Caribbean, including my native Haiti.  Some of my papers have analyzed how the hierarchies of power embedded in this colonial, then neo-colonial, history have shaped certain scientific claims around Creole languages, from the onset of Creole studies, as early as the very first description of Creole languages by European scholars in the 17th century.  Understanding these socio-historical connections will help us look at analyses of Haitian Creole and other Creole languages in a broader frame for social justice which can benefit the speakers of the languages that we linguists so love to study.

 

Indeed, the field of Creole studies is still discomforted by debates around the proper characterization of Creole languages and their formation—debates around a host of questions such as:

 

  • Are Creole languages “normal” / “regular” languages? 
  • Do Creole languages arise through “abnormal” processes of language evolution? 
  • Are Creole languages (part of) a family?
  • In the case of Caribbean Creole languages, are they genetically related to the Indo-European or Niger-Congo languages that were in contact during Creole formation?
  • Do Creoles belong to an “exceptional” typology? 
  • Can Creole languages be used to teach and learn science and other complex concepts?

 

The last question is one that is most relevant to ongoing debates in contemporary education circles in Haiti and most everywhere else in the Greater Caribbean and beyond in the Global South. And all these questions are part of the larger socio-historical and biographical backdrop of this seminar. But here we won’t spend too much time on these centuries-old debates about the development, structures and viability of Creole languages, even though these debates still infect most linguistic textbooks—as a banal reflex of unbroken transmission of biases via (neo-)colonial power/knowledge cycles in the human sciences.

 

What we’ll do then, in the first couple of sessions of the course, is to sample these debates and the biases therein.  Then, in the rest of the course, we’ll focus on  Kreyòl as a perfectly “normal”language, and we’ll study it with “uniformitarian”lenses—that is, we’ll enlist the toolbox of syntactic theory in order to examine Kreyòl as a language that is as “exceptional” as every other language. More generally, I’d like to assume that whatever tools linguistic theory gives us to understand the synchrony and diachrony of any non-Creole language will also help us understand the synchrony and diachrony of any Creole language. That is, our seminar will assume that there’s absolutely no need for a sui generis theory of Creole formation.

 

With these caveats in mind, I will invite students to take a stab at various puzzles in the diachrony and synchrony of Haitian Creole and other Creole and non-Creole languages.  I do have a preliminary menu of puzzles and theoretical proposals to analize some of the data in these puzzles. This menu comes mostly from my own work, over the past three decades, on Haitian Creole, alongside a manuscript in progress, from which I will share drafts of chapters.  But I also want to leave space for course participants to come up with their own favorite puzzles and challenges, especially if they might bear on puzzles that still resist neat explanations.

 

Course requirements and preliminary calendar of activities

 

Course participants will present and lead discussion on topics of their liking that connect with the afore-mentioned areas of syntax and that include relevant (Haitian) Creole data. So the formal course requirements will include regular weekly participation, questions and comments about assigned readings before each session, in-class presentations and a short paper (~10 pages) which may well be a draft of something publishable.

 

Meanwhile, on Wednesday, September 7, we’ll begin the seminar with a discussion of basic socio-historical and political issues (as sketched above), in order to clear up some muddy issues around general terminological and conceptual background about “Creole” languages and their history.  We will also use that the first session to survey participants’ backgrounds, interests and (potential) topics for research in the context of the cours.e

 

Then time permitting or perhaps next Wednesday, September 14, we’ll segue with my and Enoch Aboh’s views about “A null theory of Creole formation based on Universal Grammar.”

 

Meanwhile you may also want to look at another paper of mine that gives a general survey of Haitian Creole—from John Holm’s 2007 book Comparative Creole Syntax:

Course Requirements (equally weighted at 25% each):

 

  1. Regular weekly class participation
  2. Questions & comments on relevant readings before each session—by Monday, 8PM
  3. Presentation in class (toward the end of the semester)
  4. Short paper (~10 pages)

 

TENTATIVE Schedule of Classes with topics—subject to change based on participants’ evolving interests, new questions and new puzzles, etc.

 

Wed., 9/7:       General socio-historical introduction—“the politics of Creole studies” starting with the colonial invention of “race” and of “Creole languages” 

                        DeGraff (2020) “The politics of education in post-colonies: Kreyòl in Haiti as a case study of language as technology for power and liberation”
https://iacpl.net/wp-content/uploads/2020/07/DeGraff-1.pdf

 Wed., 9/14:     General background about “Creoles” — What’s in a label?  “Creole Exceptionalism” vs. “Uniformitarianism”

                        Aboh & DeGraff (2017) “A null theory of Creole formation based on Universal Grammar” http://linguistics.mit.edu/wp-content/uploads/aboh-and-degraff-2017-null-theory-of-creole-formation.pdf                       

                        DeGraff (2007) “Kreyòl Ayisyen or Haitian Creole” http://lingphil.mit.edu/papers/degraff/degraff2007hc-ccs.pdf

                        A peek at DuoLingo’s online Kreyòl course—and at some puzzles (glitches?) there in search of analyses https://www.duolingo.com/course/ht/en/Learn-Haitian%20Creole

                        [To be continued…]

24.956, Fall 2022: “Topics in Syntax”

Instructors: Danny Fox and Mitya Privoznov
Monday 2-5pm; 32D-461, https://canvas.mit.edu/courses/16353
 
This class will focus on restrictions on overt and covert A-bar movement and on what these might tell us about the principles that determine the nature of syntactic derivations.
 
We will begin with the study of three strong islands (the Subject Condition, the Coordinate Structure Constraint and the Adjunct Condition) and their apparent violations attested cross-linguistically. Each of these islands can be stated as an absolute configurational restriction along the lines of Huang’s (1982) Condition on Extraction Domain (CED), which in turn can be construed as a condition on the application of External Merge and Spell Out. However, the islands have also been argued to be selectively violated across languages. For example, the Subject Condition is claimed to be violable in Russian, Turkish and Japanese – see the debate in Jurka et al. (2011) vs. Fukuda et al. (2016, 2018). This creates a tension between the configurational approaches and various alternative weaker proposals (Takahashi 1994, Rizzi and Shlonsky 2007) We will go through argumentation on both sides for each of the island effects and discuss their wider implications, presenting arguments from Privoznov (2021) that the governing factor pertains to structural configurations determined prior to movement. [We might also discuss possible connections to the theory of discourse anaphora.] 
 
From there we will move to the study of the (locality) conditions that determine how multiple specifiers are stacked at the edge of various projections, with a focus on wh-movement – overt and covert – and parasitic gap licensing (Richards, 2000, 2003; Nissenbaum 2000, Fox and Nissenbaum 2018, Davis 2020). As usual for advanced seminars, we expect the discussion to place us in unanticipated locations, hence no syllabus at this stage. For some of the very initial readings, please visit the website.

 

 

24.S96, Fall 2022: “Methods in Computational Linguistics”

Instructor: Forrest Davis
Tuesday 10-1; 32-D461, 

https://canvas.mit.edu/courses/17408

 

 

Current models in natural language processing are trained on large amounts of text with a simple objective: predict the next word (or predict a word in context). These models have garnered a lot of attention, and there are claims that they can learn non-trivial aspects of human linguistic knowledge. A growing body of literature, framed as “model interpretability”, has attempted to address what exactly such computational models know about linguistic structure. Exploration of these linguistically “naive” models can be used to clarify claims about the nature (and origin) of linguistic knowledge.

 

In this course, we will survey papers and methods in computational linguistics and natural language processing with an aim towards understanding five key approaches to evaluating neural network models:

 

  • Targeted syntactic evaluations

  • Representational probing

  • Direct comparison to human behavioral measures

  • Priming/fine-tuning

  • Cross-linguistic comparison

 

The course is intended to be a hands-on experience. We will follow a cyclic pattern. First a model will be introduced with a particular approach to evaluation, with students implementing core aspects of both. Then, student led presentations will explore replications, extensions, and challenges to the existing empirical results, broadening our understanding of how to use and evaluate neural models, and how these findings may relate to a theory of language.

A tentative syllabus is attached. The first classes will introduce students to the relevant computational tools, so no background in computer science or machine learning is assumed.