At this week’s Montreal Language Modeling Lab meeting (Tues Nov 14 at 5:30-7:30pm in Room 117), Emily Mulhall will present her replication of the Rational Speech Act model of language understanding in Goodman & Stuhlmuller (2013). “Knowledge and Implicature: Modeling Language Understanding as Social Cognition.” Topics in Cognitive Science, 5(1):173-184. She will also review the RSA framework in general and alternatives to it. Light food provided. Everyone is welcome; please RSVP to firstname.lastname@example.org if not on the lab mailing list.
The WORDS Group will be meeting on Friday 17th November, at McGill, Dr. Penfield Ave. 1085 (room 117) at 1pm-2.30pm. Tim O’Donnell will present “Productivity and Reuse in Language”:
A much-celebrated aspect of language is the way in which it allows us to express and comprehend an unbounded number of thoughts. This property is made possible because language consists of several combinatorial systems which can be used to productively build novel forms using a large inventory of stored, reusable parts: the lexicon. For any given language, however, there are many more potentially storable units of structure than are actually used in practice — each giving rise to many ways of forming novel expressions. For example, English contains suffixes which are highly productive and generalizable (e.g., -ness; Lady-Gagaesqueness, pine-scentedness) and suffixes which can only be reused in specific words, and cannot be generalized (e.g., -th; truth, width, warmth). How are such differences in generalizability and reusability represented? What are the basic, stored building blocks at each level of linguistic structure? When is productive computation licensed and when is it not? How can the child acquire these systems of knowledge? I will discuss a theoretical framework designed to address these questions. The approach is based on the idea that the problem of productivity and reuse can be solved by optimizing a tradeoff between a pressure to store fewer, more reusable lexical items and a pressure to account for each linguistic expression with as little computation as possible. I will show how this approach addresses a number of problems in English inflectional and derivational morphology, and briefly discuss its applications to other domains of linguistic structure.
Join us this Friday at 10am in room 117 of the Linguistics building for our meeting where Nico Baier will be presenting his paper on “Anti-agreement in non-local contexts.” All are welcome!
The Prosody & Meaning Reading group will discuss Cristian DiCanio’s “The phonetics of information structure in Yoloxóchitl Mixtec” on Monday, Nov 6 (12-1pm, Room 117).
At this week’s Montreal Language Modeling Lab meeting (Tues Nov 7 at 5:30-7:30pm in Room 117), Chris Bruno will present work on the generalized parser project, a general framework capable of parsing many different grammar formalisms. Light food provided. Everyone is welcome; please RSVP if not on the lab mailing list.
In this week’s P* Reading Group on Wednesday (Nov. 8) 11am-12pm in Room 117, Yeong will lead a discussion of DiCanio et al. (2015). Vowel variability in elicited versus spontaneous speech: evidence from Mixtec. Journal of Phonetics, 48, 45-59. This discussion is especially relevant because Christian DiCanio will be the colloquium speaker on Friday. Everyone is welcome!
Christian DiCanio from the University at Buffalo will giving a talk entitled “Phonetic variation and the construction of a Mixtec spoken language corpus” as part of the McGill Linguistics Colloquium Series on Friday, November 10th at 3:30pm in room 433 of the Education Building. All are welcome to attend! For the abstract and for any other colloquium information, please visit the Colloquium Series web page: https://www.mcgill.ca/linguistics/events/colloquium-series.
The Synt-ex reading group (experimental syntax reading group) is an interdisciplinary reading group that meets up every two weeks. Our next meeting will be November 14th, 3-4pm at the Rabinovitch House (3640 rue de la Montagne). There will be snacks provided by CRBLM. Anyone interested is welcome to join, if you want to be added to the list send an e-mail to email@example.com or check out the website https://syntex-mcgill.github.io/welcome/.
A paper by Hye-Young Bang and co-authors (Morgan Sonderegger, Yoonjung Kang, Meghan Clayards, Taejin Yoon), “The emergence, progress, and impact of sound change in progress in Seoul Korean: Implications for mechanisms of tonogenesis”, has just appeared in Journal of Phonetics. Congratulations!
This study examines the origin, progression, and impact of a sound change in Seoul Korean where the primary cue to a stop contrast in phrase-initial position is shifting from VOT to f0. Because it shares similarities with the initial phase of tonogenesis, investigating this “quasi-tonogenetic” sound change provides insight into the nature of the emergence of contrastive f0 in “tonogenetic” sound changes more generally. Using a dataset from a large apparent-time corpus of Seoul Korean, we built mixed-effects regression models of VOT and f0 to examine the time-course of change, focusing on word frequency and vowel height effects. We found that both VOT contrast reduction and f0 contrast enhancement are more advanced in high-frequency words and in stops before non-high vowels, indicating that the change is spreading across words and phonetic contexts in parallel. Furthermore, speakers suppress non-contrastive variation in f0 as f0 emerges as a primary cue. Our findings suggest that one impetus for tonogenetic change is production bias coupled with an adaptive link between the cues. We further discuss the role of Korean intonational phonology on f0 which may help explain why the phonetic precondition leads to change in Seoul Korean but not in other languages.
McGill Linguistics had a table at McGill’s Open House this past Sunday. Thanks to Claire Bautista, Shannon Fiedler, Fiona Higgins, Hayley Ostrega, Alele Rangel, Nicole Ryan (pictured left), Vicky Svaikovsky and Tea Vincic (pictured right) who volunteered at the event — it was a great success!
On Monday, October 30, Ruveneko Ferdinand-Peterkin will lead the discussion about Goldrick et al. (2016): Automatic analysis of slips of the tongue…, and a related production experiment (Monday Oct 30, 12-1pm, Room 117, note shorter time due to Aron Hirsch’s minicourse).
At this week’s Montreal Language Modeling Lab meeting (Tues Oct 31 at 5:30-7:30pm in Room 117), Arlie Coles will be presenting on her work implementing a neural network model for the Montreal Forced Aligner. Light food provided. Everyone is welcome; please RSVP if not on the lab mailing list.
In this week’s P* Reading Group on Wednesday (Nov. 1), 11am-12pm in Room 117, Gasper Begus, a PhD student visiting from Harvard University, will present a talk entitled, “What can unnatural processes tell us about typology? ” The abstract is below. Everyone is welcome!
One of the most contested debates in phonology concerns identifying factors that affect typology. Two lines of thought emerge in this discussion: Analytical Bias (AB) and Channel Bias (CB) approach. Disambiguating between Analytic and Channel Bias influences on typology is complicated by the fact that several proposals assume learning biases (AB) crucially influence the frequency and directionality of sound change (CB). In this talk, I argue that this “duplication problem” is substantially reduced in the case of unnatural alternations. I present a model that estimates CB influences on typology based on a statistical technique non-parametric bootstrap called Bootstrapping Sound Changes (BSC). For any synchronic alternation, the BSC technique estimates the probability that the alternation arises based on the number of sound changes it requires and their respective probabilities. With the BSC technique, we can compare Historical Probabilities of attested and unattested alternations and perform inferential statistics on the comparison, predict (un)attestedness in a given sample for any alternation, and derive quantitative outputs for a typological framework that models both Channel Bias and Analytical Bias influences together. The BSC technique also identifies several mismatches in typological predictions of Analytic and Channel Bias approach. By comparing these mismatches with the observed typology, the paper attempts to quantitatively evaluate the distinct contributions of diachronic and synchronic factors on phonological typology.
Join us this Friday at 10am for our Syntax meeting in room 117 of the Linguistics building. Jessica Coon will be presenting joint work with Stefan Keine in their paper “Feature Gluttony and Hierarchy Effects”.
This paper offers a new take on a family of hierarchy-effect inducing configurations, including (i) PCC effects (Anagnostopoulou 2005; Nevins 2007), (ii) dative-nominative configurations (Sigurdsson & Holmberg 2008), and (iii) certain copula constructions (Coon, Keine, & Wagner, to appear). Following previous work, we take these configurations to arise in contexts in which two accessible DPs are in the same domain as a single agreeing probe (Béjar & Rezac 2003; Anagnostopoulou 2005). Standard accounts of these hierarchy effects attribute them to failures of nominal licensing, in particular, a Person Licensing Condition (Béjar & Rezac 2003; Preminger 2017). We argue instead that these effects are better understood as arising from properties of probes. We offer a new account which captures commonalities and differences across these constructions, both in terms of the types and specifications of the features involved, as well as in the result of hierarchy violations and their possible repairs.
All are welcome!
A sizeable contingent of McGill related linguistics attended the 48th Annual Meeting of the North East Linguistic Society (NELS 48) this past weekend. For the first time, the conference took place outside North America, viz. at the University of Iceland, Reykjavík. As the organizers emphasized, Iceland now easily holds the record as the smallest host country for NELS in terms of both population size (previously: Canada) and land area (previously: USA).
“Cross-categorial” operators — notably, the conjunction and and focus operator only — appear in a broad range of environments. And occurs, for instance, between full clauses in (1a) and DPs in (1b). Likewise, only occurs pre-vP in (2a) and pre-DP in (2b).
(1) a. John saw every student and Mary saw every professor.
b. John saw every student and every professor.
(2) a. John only learned oneF language.
b. John learned only oneF language.
Given their broad distribution, these operators seem to require a flexible semantics. In (1a), and operates on truth-values, like the & connective of propositional logic: (1a) is true iff both conjoined clauses are true. Yet, in (1b), and seems to have a different meaning which composes with quantifiers. A range of semantic mechanisms have been proposed to achieve the necessary flexibility (e.g. Keenan & Faltz 1978, 1985, Gazdar 1980, Partee & Rooth 1983, Jacobson 1999, 2015). One approach draws on type-shifting rules: and is stored in the lexicon as &, but type-shifted to compose with quantifiers in (1b). Only receives a similar analysis, through type-shifting (Rooth 1985).
The aim in this mini-course is to challenge the idea that these operators have a flexible semantics, pursuing instead the Semantic Inflexibility Hypothesis (‘SIH’). Under the SIH, and always operates on truthvalues (following Schein 2017), and only again patterns in kind. The viability of the SIH for data like (1b) and (2b) depends on covert syntax: the underlying structure must be richer than it appears from the surface string so that it includes a truth-value denoting scope site for the operator. The course will build a case the SIH. First: we will see that semantic flexibility approaches have overgeneration problems, providing initial motivation for the SIH. Second: we will diffuse some counterarguments to covert syntax with and from the prior literature (e.g. Partee 1970). And, third: we will provide a range of novel evidence that covert syntax is in fact present with both and and only in a fragment of data. The SIH, if successful, leads us to constrain the availability of type-shifting, and the expressive power of the semantic grammar more generally (cf. Heim 2015). Class 1: The Semantic Inflexibility Hypothesis October 30, Monday, 10:30-12:00 – Room 117 Class 2: Apparent DP conjunction November 2, Thursday, 11:30-13:00 – LEACOCK 14 Class 3: November 3, Friday, 15:00-16:30 – Room 117 Apparent NP conjunction Class 4: November 6, Monday, 10:30-12:00 – Room 117 Focus operators Class 5: November 9, Thursday, 11:30-13:00 – LEACOCK 14 Consequences for the grammar
The Montreal Language Modeling Lab is holding weekly meetings starting this semester to discuss topics related to computational and quantitative linguistics. Meetings are held on Tuesday evenings 5:30pm-7:30pm in Room 117, and light food is provided. Email Emily (firstname.lastname@example.org) to be added to the mailing list. In this week’s meeting on Tuesday (Oct. 24), Bing’er will present the 10-minute version of her first eval paper on the perception of tonal register contrast in Chinese Wu dialects, followed by a discussion of Kleinschmidt et al. (2011), “A Bayesian belief updating model of phonetic recalibration and selective adaptation,” Association for Computational Linguistics. All are welcome, but please RSVP if not on the mailing list.
In this week’s P* Reading Group on Wednesday (Oct. 25), 11am-12pm in Room 117, Bing’er will lead a discussion of Richter et al. (in press). “Evaluating Low-Level Speech Features Against Human Perceptual Data”. Transactions of the Association for Computational Linguistics. Everyone is welcome!
The WORDS Group will be meeting on Friday 27th October, at McGill, Dr. Penfield Ave. 1085 (room 117) at 1pm-2.30pm. This session will take place in the shape of a Mini Workshop on Person, where we will discuss various examples of morphologically complex pronouns that we have come across.
Everyone is welcome to attend!