Whamit!

Speaker: Steve Piantadosi (Gibson lab)
Title: What is the language of thought?
Time: Tues 3/29, 12pm
Location: 46-3310 (note the change from our usual location)

We present a computational model of learning compositional, structured concepts in a “language of thought.” We show that language of thought models out-perform alternatives such as exemplar models in capturing people’s patterns of generalization in a massive, online, concept-learning experiment. We additionally develop a method for formally evaluating different representation languages using the experimental results. We compare 9 different representation languages for simple boolean concepts, and 12 for languages for concepts that move beyond boolean logic, towards the types of representations necessary for natural language semantics.

Speaker: Sverre Stausland Johnsen (Harvard University)
Title: Rhyme acceptability determined by perceived similarity
Time: Tuesday 3/29, 5-6pm, 32-D831

Intuitively, a-b is a better rhyme than a-c when a and b are more ‘similar’ to each other than a and c are. But how do we measure this similarity? I show in this talk that the perceptual similarity between segments as calculated from confusion matrices is a much better predictor of people’s judgments of rhymes than are similarity measures based on articulatory or acoustic feature systems.

Upcoming talks:
Apr 5: Anne-Michelle Tessier
Apr 12: Ricardo Bermudez-Otero
Apr 13, 3-5pm: WCCFL Practice Talks ***Note Special Day and Time
Apr 26: Jongho Jun
May 3: Nina Topintzi
May 10: RUMMIT practice talks

You can view the current, up-to-date version of the schedule here (click ‘agenda’ to see the schedule as a list), or subscribe via iCal here.

• Gillian Gallagher (PhD 2010) has officially accepted a tenure track phonology position at NYU, starting Fall 2011; Gillian has already been teaching at NYU this year
• Jessica Coon (PhD 2010) has officially accepted a tenure-track syntax position at McGill; Jessica is currently a post-doc in the Polinsky lab at Harvard
• Luka Crni? has accepted a one-year post-doc position at the Hebrew University in Jerusalem
• Patrick Grosz has accepted a post-doctoral position (Akademischer Mitarbeiter), at the University of Tübingen.

Congratulations, Gillian, Jessica, Luka and Patrick - and congratulations also to NYU, McGill, Tübingen and Hebrew University!

This week, we have two LFRG meetings: one on Wednesday at 4pm, another one at the usual 2pm on Friday. Note the change of room on Friday (and perhaps on Wednesday too.)

WHO: Hadas Kotek & Yasutada Sudo
WHAT: A superlative reading for Most_PROP (practice talk for CLS)
WHEN: March 30, 4:00PM-5:30PM
WHERE: 32-D461 or 32-D831 - please stay tuned

WHO: Ben George (UCLA)
WHAT: Which answers matter, and how?
WHEN: April 1st, 2:00PM-3:15PM
WHERE: 32-D461 (CHANGE OF ROOM!)

[Click on the talk titles for abstracts]

Upcoming meetings:

4/06 Alya (special extra Wedn. meeting)
4/08 Guillaume
4/15 Igor
4/22 Sarah Ouwayda
4/29 Eva Csipak
5/06 UP FOR GRABS

On Saturday, April 2 MIT will host this year’s ECO5 student syntactic workshop. It is a yearly small conference where graduate students from five East Coast departments (Harvard, UConn, UMass, UMaryland and MIT) can present their ongoing or completed work on syntactic issues to a friendly crowd of faculty and students, which rotates between the five co-organizing departments.

This year it is MIT’s turn to run things, and you can find the tentative schedule here.

The workshop starts at 11am on Saturday, and will continue until around 6pm.

See you there! Coppe and Igor, MIT ECO5 representatives

Wayne O’Neil has recently returned from his annual (since 2000) March stay in Seattle, where he and his partner Maya Honda spent several days doing linguistics with David Pippin’s fourth- and fifth-graders at St Thomas School in Medina WA. On this occasion, ‘doing linguistics’ focused on some aspects of English orthography: why, example, civil, solemn, and moral are spelt the way they are. However, Wayne was also required to give a spirited reading of the opening lines of Beowulf as a lead-in to a discussion of the history of the English language.

Some of the work that Wayne and Maya have done in Seattle over the past decade is discussed in Honda, O’Neil, and Pippin’s paper “On promoting linguistics literacy: Bringing language science to the English classroom”, in Kristin Denham and Anne Lobeck (ed.) Linguistics at school: Language awareness in primary and secondary education, 175-188 (Cambridge University Press, 2010).

Jay Keyser was interviewed last week on the Callie Crossley Show on WGBH. Full audio of the 24 minute interview is available on the show’s website. Here’s a direct link to the mp3 stream.

The show’s description:

The history of hacks and pranks at MIT dates back almost as far as the venerable institution itself. Students with expertise in engineering, computer science, robotics, and math — and presumably with a little extra time and brainpower to spare — have taken pranksterism to the level of high art. Their hijinks have included a firetruck, police cruiser, and biplane replica, alternately hoisted atop MIT’s Great Dome; a fully appointed room — including a billiards table, a cat, chairs, and an illuminated lamp — hung upside down from the Media Lab; numerous interruptions staged during the annual Harvard - Yale football game; and a cross-country hacking war with rival institution, Caltech. MIT Professor Emeritus Jay Keyser, has seen a lot of it in his time, and he’ll talk about the school’s secret hacking society, the best hacks, and why bright students at a world-class institution can still find time to put one over on faculty.

Mentioned during the interview is Jay’s forthcoming new book: Mens et Mania: The MIT Nobody Knows.

We are deeply saddened by the news that our former student and colleague in linguistics Maria Babyonyshev has died at the untimely age of 44, from complications of a devastating 2006 car accident. Masha received not one, but two degrees from MIT. She received her undergraduate degree in linguistics from MIT in 1990 (Phi Beta Kappa) and her PhD from our department in 1996, with a dissertation entitled Structural Connections in Syntax and Processing: Studies in Russian and Japanese. In fact, the word “connections” aptly describes Masha’s entire varied and influential research career, which steadfastly and continually charted and explored the links between linguistic theory, language acquisition and processing - as well as the links between the structure of her native language, Russian, and the other languages of the world. In her own papers and her many collaborations, Masha made some of the most interesting and cleverest connections we know between the theoretical and experimental sides of linguistics. Most recently, working with students and colleagues at Yale (where she was a member of the linguistics faculty) Masha had started investigating a striking pattern of language impairment widespread in one village in the Arkhangelsk region of Russia. As a researcher, as a member of our field, and as a friend, she will be badly missed.

Masha’s family writes us that they are “creating a fund to complete a project that Masha was pursuing over the last year out of her love of plants, flora and trees. The Antonovka Apple was her favorite apple growing up in Russia, but it is rarely used as an eating apple here and therefore is essentially not grown. She had recently imported seeds of the apple and was working on the plant science of growing seedlings. We wish to hire a plant scientist and nursery to complete the project and provide seedlings to our friends and family to plant where they would like. Masha’s living legacy will be the trees and apples. Please send contributions to: The Antonovka Fund, c/o Ted Walls and Nadia Babyonyshev, 67 Prospect Avenue, North Kingstown, RI 02852.”

MIT will be well represented at the 47th meeting of the Chicago Linguistic Society, to be held next week (April 7-9). In addition to plenary talks by Colin Philips (PhD 1996) and faculty member Norvin Richards, the program includes talks by:

• Ezra Keshet (PhD 2008): “Contrastive focus and paycheck pronouns”
• Giorgio Magri (PhD 2009): “Correctness of OT online algorithms on Prince and Tesar’s (2004) test cases”
• Hadas Kotek, Yasutada Sudo, Edwin Howard, and Martin Hackl: “A superlative reading for most_prop”
• Ora Matushansky (PhD 2002) and Tania Ionin (BCS PhD 2003): “More than one solution”
• Young Ah Do and Seunghun J. Lee: “Acoustic bases of emotion related sound symbolism”
• Peter Graff and Jeremy Hartman: “Constraints on predication”
• Gillian Gallagher (PhD 2010): “Auditory features: the case from laryngeal cooccurrence restrictions”
• Jessica Coon (PhD 2010) and Omer Preminger: “Towards a unification of person splits”

Thank you to David Pesetsky for compiling this information.

Pritty Patel-Grosz recently presented her work on Agreement at Rajesh Bhatt’s South Asian syntax-semantics workshop which took place at UMass on March 19th-20th. The program can be found here.

Speaker: Micha Breakstone
Title: Inherent Evaluativity
Time: Thursday, March 31, 12:30-1:45pm
Location: 32-D461

Two related challenges have been the focus of research on the semantics of degree constructions. The first is the distribution of Measure Phrases (MPs), and the second concerns the distribution of evaluative readings. A theory of MP distribution should account for e.g. (1).

1. a. John is 6 feet tall.           b. *John is 3 feet short.

Evaluativity (aka ‘norm-relatedness’) may be defined as the phenomenon in which the interpretation of an adjective in a given construction is dependent on a contextual standard. A theory of evaluativity should explain the pattern which emerges in (2) [(+E) denotes Evaluative].

2.	a. John is tall. (+E)	e. John is short. (+E)
	b. How tall is John? (-E)	f. How short is John? (+E)
	c. John is as tall as Mary. (-E)	g. John is as short as Mary. (+E)
	d. John is taller than Mary. (-E)	h. John is shorter than Mary. (-E)

It has been noted (Bierwisch 1989, Sassoon 2008) that these challenges are related: Adjectives that do not license MPs (1b) are evaluative in equatives and degree questions (2g,f). The goal of this talk is to meet these two challenges and account for the correlation noted by Bierwisch.

Previous approaches assume a semantics for gradable adjectives under which evaluativity must enter independently (see Rett (2008), Cresswell (1976)) and for which explaining MP distribution entails additional assumptions (with Bierwisch’s correlation left unexplained). I will present an alternative under which evaluativity is an inherent part of the semantics. Instead of accounting for evaluativity, the challenge will be to do away with it. As it turns out this simple reflective move holds promise of significant explanatory power.

Speaker: David Adger, Queen Mary, University of London
Time: Friday, April 1, 2011, 3:30pm-5pm
Location: 32-141 (PLEASE NOTE ROOM)
Title: The syntax of (some) apparent complements to N

Abstract:

The usual view of the syntax of the title of this talk is the following:

(1) [the [ syntax [ of [ some [ apparent [ complements [ to N ]]]]]]]

where the arguments of the nominal heads are introduced as complements. I will argue that at least PP arguments of Ns are introduced much higher up in the structure. The main argument is cross-linguistic and relies on the following:

(2) In a structure where an N is modified by an intersective AP and takes a PP complement, the unmarked order is one where the AP intervenes between N and PP (so we have [PP AP N] and [N AP PP]).

The standard view makes this generalization difficult to derive. What is required is that the surface position of the PP is high. But then the question is why? Standard views of the syntax/semantics interface (as well as of the linearization properties of specifiers) would suggest that the PP moves to a high position (e.g. Kayne 2004). I argue instead that it is Merged in a high position and that its semantics are negotiated directly by functional structure (cf Borer 2005, Ramchand 208 etc for `complements’ to V). I propose that the linear order of this high PP with respect to the N arises because the syntax allows part of the extended projection of N to be generated as a high left daughter, as well as as a lower right daughter, of the head in whose specifier the PP sits (cf. Brody and Szabolcsi 2003). This gives us:

(3) [ the [ [ alternative syntax] [ [ of [ some [ apparent [ complements ]] [ [ to N ] ]] ]]

Last week, on March 4, Donca Steriade gave a colloquium talk on “Rhyming Evidence for Intervals” at UCLA. That same week, David Pesetsky gave a series of talks at Newcastle University (including a public lecture on his joint work with Jonah Katz on the relation of music and language). Last Friday, Norvin Richards spoke on the topic of “Generalized Contiguity” in the CrISP Distinguished Visitors Series at UC Santa Cruz. Finally, this week, Sabine Iatridou will be teaching a short course at the University of Athens.

In the aftermath of this week’s terrible earthquake and tsunami we have some news to share about former students, colleagues and friends who are linguists in Japan - all of it reassuring, we are relieved to say. Nobuko Hasegawa, a friend to many of us, and a recent visitor to our department, is fine, and writes that Shoichi Takahashi and his wife Yoko are fine. According to Nobuko, Shoichi says “that the inside of the apartment is a mess and the water line stopped but they are physically ok. He is in Fukushima prefecture, which is between Tokyo and Sendai. His place is far from the nuclear plant and I don’t think it affects him.” Masatoshi Koizumi, who works in the seriously affected Sendai area, is in Guatemala doing fieldwork at the moment, and therefore far from the quake. He writes that his family is ok, as is he. Among other linguists working in the most affected area of Japan, our colleague Shigeru Miyagawa reports that Jun Abe, Akira Kikuchi, Hirohisa Kiguchi and Ken Takita have all been confirmed to be ok - and that the various linguists he has heard from in the Tokyo area are all right too. Shigeru’s own family and friends are fine as well, as are the families and friends of our current students from Japan.

As Yusuke Imanishi has informed us, information about the response to the earthquake organized by MIT students is available here.

From the chapter on MIT in the Yale Daily News Insider’s Guide to Colleges 2011:

Speaker: Patrick Grosz
Title: Analyzing non-canonical V1 in a V2 language
Time: Monday, March 14, 11:30am-12:30pm
Location: 32-D461

In this talk I discuss three types of V1 constructions in German (a language that is otherwise V2 or V-final): V1-polar exclamatives and V1-optatives (cf. Scholz 1991 for a discussion of both), as well as V1-antecedents in conditionals. Given that V2 can be assumed to arise from two separate processes, V-to-C movement and (obligatory?) filling of the SpecCP position (cf. the HSK article by Holmberg 2010), I will address the following two questions for such V1 clauses: What triggers V-to-C movement, and what (if anything) occupies their null SpecCP position?

Speaker: Ev Fedorenko (MIT BCS)
Title: The power of individual subject analyses in investigating the functional architecture of the language system
Time: Tuesday 3/15 at noon
Location: 46-3189

Language is our signature human cognitive skill and one of the most studied. Yet many fundamental questions about how language is implemented in the brain remain unanswered. I have previously argued that at least part of the difficulty in making progress may be due to an almost exclusive reliance on traditional group analyses in neuroimaging studies of language. Because of anatomical differences across individual brains, such methods are bound to underestimate functional specificity and may thus obscure true functional organization. Furthermore, group-based methods make it difficult to compare results across studies and accumulate knowledge about the functional profiles of the relevant brain regions. In recent work (Fedorenko et al., 2010), we developed methods for defining key language-sensitive regions in each individual brain, enabling us to launch a research program aimed at carefully characterizing the profiles of each of these brain regions, with the ultimate goal of understanding the representations and the computations that enable us to produce and understand language. In this talk, I will first address the question of whether language-sensitive regions (including the classical Broca’s and Wernicke’s areas) also support non-linguistic processes that have been argued to share cognitive and neural machinery with language. I will show that these brain regions respond very little to a wide range of non-linguistic tasks, including arithmetic, domain-general working memory, domain-general cognitive control, and music. I will then present some preliminary findings from two other lines of work. First, I will describe a study investigating the relationship between the language system and abstract conceptual processing. And second, I will present preliminary results from ongoing experiments that are attempting to identify the precise linguistic computations conducted in each region.

Speaker: Claire Halpert
Title: Argument licensing in Zulu
Time: Thursday, March 17, 12:30-1:45pm
Location: 32-D461

Recent work on Bantu languages (e.g. Baker 2003, 2008; Carstens and Diercks to appear; Diercks to appear) has revived the notion that the Bantu family lacks abstract Case (Harford Perez 1985). In this talk, I address the issue of how arguments are licensed in Zulu by examining two types of constructions: raising constructions and constructions involving morphologically impoverished (“augmentless”) arguments. I show that Zulu has both raising-to-subject and raising-to-object out of agreeing finite CPs, which is empirically problematic for Case-driven analyses of raising, as well as for theories that link (subject) Case-licensing to agreement or EPP (e.g. Chomsky 2000; Bobaljik 2008; Boskovic 2002; Marantz 2000). While Carstens and Diercks (to appear) and Diercks (to appear) argue on the basis of similar data from other Bantu languages that Case is not active at all in Bantu, I use data on the distribution of augmentless arguments to argue that while T doesn’t assign Case in Zulu, we do find argument licensing effects associated with v. Augmentless arguments face a more limited distribution than their augmented counterparts in Zulu. I propose that the behavior of these augmentless arguments in raising-to-object and applicative constructions in particular stems from their need to receive structural case from v. Augmented arguments, in contrast, have inherent case and thus do not exhibit the same positional restrictions. Finally, I propose that we find additional evidence for v as a probe reflected in the “conjoint/disjoint” verbal morphology alternation, which is sensitive to the contents of vP (cf. Buell 2006).

WHO: Jan Anderssen
WHAT: Weak and strong readings of indefinites
WHEN: March 18, 2:00PM-3:15PM
WHERE: 32-D831

Suggested reading is Partee 1989 “Many quantifiers”, just to brush up on the issues.

There is no meeting of the Phonology Circle scheduled this week. We will resume on Tues Mar 29 (after spring break), with a talk by Sverre Johnsen.

Upcoming talks:
Mar 29: Sverre Johnsen
Apr 5: Anne-Michelle Tessier
Apr 12: Ricardo Bermudez-Otero
Apr 13, 3-5pm: WCCFL Practice Talks ***Note Special Day and Time
Apr 26: Jongho Jun
May 3: Nina Topintzi
May 10: RUMMIT practice talks

You can view the current, up-to-date version of the schedule here (click ‘agenda’ to see the schedule as a list), or subscribe via iCal here.

Speakers: Alya Asarina and Jeremy Hartman
Title: Genitive subject licensing in Uyghur subordinate clauses
Time: Monday, March 7, 11:30am-12:30pm
Location: 32-D461

In this talk we analyze embedded clauses in Uyghur, a Turkic language spoken in the Xinjiang province of China. We make three main claims. First, we claim that genitive case on the subjects of these clauses is licensed by agreement with a clause-external D head. Second, we claim that these clauses are always embedded by a head noun, though in some cases this noun is phonologically null. Third, we show that these clauses are full CPs that can be headed by an overt complementizer. Putting these claims together, we conclude that Uyghur exhibits agreement and case-assignment over a CP boundary, and thus provides a challenge for certain proposals about the locality of these relationships. We discuss several approaches to the Uyghur facts, as well as similar phenomena in other languages, within the framework of Phase Theory (Chomsky 2001 et seq.).

Speaker: Michael Kenstowicz and Filomena Sandalo
Title: Vowel Harmony and Dispersion in Brazilian Portuguese
Time: Tuesday 3/8, 5-6pm, 32-D831

Like Standard Italian, Brazilian Portuguese distinguishes seven vowels in stressed syllables. The inventory is reduced to five in unstressed syllables (through merger of open and closed mid vowels) and to just three in unstressed final syllables (through merger of mid and high vowels). Earlier phonetic studies of Major (1986) and Clegg & Fails (1992) find that the duration of vowels as well as the relative openness of the low vowel decreases in the order stressed > medial unstressed > final. Flemming (2004) points to these data as support for his dispersion theory of vowel inventories where the neutralization of contrasts results from the interplay of constraints on articulatory effort and phonetic distance. The earlier literature does not consider the behavior of pretonic syllables where the open vs. closed mid vowel contrast is also neutralized but a height harmony process occurs that disperses the mid vowels to three values as a function of the height of the following stressed syllable. The result is a phonetic inventory of seven vowels in pretonic position. Our presentation is a progress report on a study of the harmony process and its potential bearing on the dispersion theory.

Upcoming talks:
Mar 8: Michael Kenstowicz and Filomena Sandalo
Mar 15: AVAILABLE
Mar 29: Sverre Johnsen
Apr 5: Anne-Michelle Tessier
Apr 12: Ricardo Bermudez-Otero
Apr 13, 3-5pm: WCCFL Practice Talks ***Note Special Day and Time
Apr 26: Jongho Jun
May 3: Nina Topintzi
May 10: AVAILABLE (RUMMIT Practice Talks)

You can view the current, up-to-date version of the schedule here (click ‘agenda’ to see the schedule as a list), or subscribe via iCal here.

Speaker: Marc Authier (Penn State University)
Title: PF-deletion, Topicalization and French modal Ellipsis
Time: Thursday, March 10, 11am-12pm
Location: 32-D461

This paper shows that the hypothesis briefly discussed in Busquets and Denis (2001) that French displays both PF-deletion and pro-form ellipsis of sentential complements with different classes of verbs is indeed correct. It subsequently focuses on the PF-deletion type (i.e., modal ellipsis) and argues that like English VP ellipsis, this type of ellipsis should be syntactically derived by way of movement and that French elided TPs stand in a topic position. The syntactic chain thus created by Move is composed of two links: the first merge copy, which is always silent, as is the case with most types of movement, and the remerge copy, which also remains silent, being recoverable from the discourse, but sometimes allows phrasal subparts to undergo further A?-movement to Spec, FocP, a position where they are fully spelled-out at PF. The absence of intervention as well as freezing effects associated with the latter movement are then explained using the rescue by PF deletion account developed in Bo?kovi? (2011). This novel theory of French modal ellipsis is shown to correctly predict the existence of what Busquets and Denis (2001) named French Pseudo- Gapping, as well as the much more restricted distribution of modal ellipsis in untensed contexts. It also provides a partial answer to why French does not have VP ellipsis and why English does not have modal ellipsis.

Speaker: Lisa Reed (Penn State University)
Title: Modifying and strengthening the PRO hypothesis
Time: Thursday, March 10, 12:30-1:45pm
Location: 32-D461

Abstract (PDF)

Date: Friday, March 11, 2011
Time: 3:30-5:00 PM
Place: 32-141 (PLEASE NOTE THE ROOM)
Speaker: Jason Merchant, University of Chicago
Title: Predicate/argument asymmetries, Agreement, and gender

Abstract:

Building on similar results from Romance, I show that Greek animate masc/fem noun pairs fall into three classes with respect to their behavior under ellipsis: those that license mismatches in gender in either direction (epicence nouns like jatros ‘doctor’), those that don’t alternate in either direction (e.g., adhelfos/adheli ‘brother/sister’) and those whose masc form license ellipsis of a feminine but not vice versa (e.g., pianistas/pianistria ‘pianist’). I argue that this pattern can be accounted for if masc/fem can be variously specified as (in)delible, where delible features are subject to deletion under Agreement (extending von Stechow, Heim, and others), before ellipsis resolution is computed. This analysis, unlike previous underspecificational approaches, can account for the novel observation that these alternations are found only on NPs in predicate positions, and that no noun pair alternates when the target of ellipsis is in an argument position. Finally, I demonstrate how easily this account can be implemented using LF-copy as the ellipsis identity mechanism, while mainstream LF-identity/semantic identity theories find it surprisingly difficult to capture the asymmetry.