Sam Bowman writes:
Hi Pfreunden,
This Friday at 12:15 in the Greenberg Room I will be giving a practice talk for CLS on a wacky but useful new approach to vowel harmony. Come lend an ear! But wait! There’s more! Since the practice talk shouldn’t go more than thirty or forty minutes, we will start the meeting with an organizational session. If you would like to give a talk this quarter (practice or otherwise), or if there is anyone you would like to invite, come by and let us know!
Two arguments for vowel harmony by trigger competition
Sam Bowman (Stanford)
I present two phenomena in front-back vowel harmony which are difficult to account for in standard theories, and argue that with some necessary elaborations, Trigger Competition (TC, Kimper, 2011) is best suited to account for both. TC is a new harmony framework based on a positive constraint (imperative) set in Serial Harmonic Grammar, and allows for agreement between non-adjacent segments. The constraint considers both the distance between trigger and target and the nature of the trigger in assigning rewards, allowing for a fairly sophisticated approach to non-participating segments.
Read the rest of this entry »
Join us in the Greenberg room today at 3:30 for the first colloquium of the quarter. This week’s edition will be a talk by Tibor Kiss (Ruhr-Universität Bochum) on “Missing determiners and obligatory adjectives”. Afterwards, stick around for a social celebrating the April 8 holiday Hana Matsuri, the Japanese Flower Festival honoring the Buddha’s birthday. We’ll munch on sushi, edamame and wasabi-flavored things. Hope to see you all at both!
Missing Determiners and Obligatory Adjectives
This talk is concerned with determinerless PPs in which an adjective becomes obligatory, as illustrated in (1).
(1) Die Verkäuferin ist von einem unbekannten Mann mit *(blutiger) Spritze bedroht worden.
the shop assistant is from a unknown man with bloody syringe threatened PASS
‘The shop assistant has been threatened by an unknown man who used a syringe containing blood.’
In order to determine grammatical factors for the omission and realization of a determiner, we have applied annotation mining (Chiarcos et al. 2008, Kiss et al. 2010) to large data sets containing the prepositions *unter* (under, below), *über* (over, above), *mit* (with), and *ohne* (without). A generalized linear mixed model (GLMM, Zuur et al. 2007) revealed that adjectival modification increases the probability of determiner omission for the first three prepositions, but lowers it for *ohne*. Read the rest of this entry »
Alexis Burgess will be giving a talk on Monday as part of the SymSys Forum. Come to the Greenberg room from 12:15-1:05 to hear about “Identity Explained Away”.
Why do we have the concept of numerical identity? Is it just a spandrel, or do we actually use it for something important? Presumably we don’t have any urgent need to track the distribution of the identity relation out there in the world. A Fregean might say the concept’s raison d’etre is just to prompt the consolidation of mental files. A Quinean might say that it helps us acquire manifestly valuable arithmetical concepts. I have another idea.
Come to the Greenberg room on Tuesday from 2 to 3:15 to join the first SMircle meeting of the quarter. Vera Gribanova will be giving a practice talk of a paper to be delivered at CLS 49, co-authored with Boris Harizanov. They’ll also use some of the meeting time to discuss plans for the rest of the quarter.
Number mismatch in Bulgarian nominal coordinate structures
Boris Harizanov (UC Santa Cruz), Vera Gribanova (Stanford University)
Bulgarian allows singular coordinated adjectives to modify a plural noun:
(1) balgarsk-o-to i grack-o pravitelstv-a
bulgarian-n.sg-the and greek-n.sg government-pl
‘the Bulgarian and Greek governments’ (two governments: a Bulgarian and a Greek one)
Two major questions arise concerning the syntax and morphology of this construction, and their interactions. First, what is the underlying structure of such examples, and how does it correspond to their particular interpretations? Second, what is the mechanism by which number features on the two adjectives may be mismatched with the noun they modify in this nominal concord language? We argue that there are two coordinated NPs each containing identical nouns, each modied by a singular adjective. The nouns escape the coordinate structure via Across-the-Board movement, resulting in the pronunciation of a single instance of the noun (cf. similar approaches to Right Node Raising, e.g. Ross 1967). In this position, it gets marked as plural in the morphology, since the nominal phrase that contains it is plural.
This account explains a number of properties of this construction, including the prohibition on certain types of mismatches in the two conjuncts, the behavior of pluralia tantum nouns, and irregular plurals, and it fares better than its predecessors (e.g. Arregi and Nevins 2013). Our proposal leverages the idea that syntactic structure feeds both concord (on the sound side) and semantic interpretation (on the meaning side), but there is no direct link between concord and meaning. In addition, it illuminates the interaction between morphology (including number marking and concord) and Across-the-Board movement.
CSLI is hosting a workshop, Perspectives on Modality, this Friday, April 12th, organized by Annie Zaenen and Cleo Condoravdi of CSLI and Valeria de Paiva of Nuance. The program can be found here, and a description of the workshop is below:
The study of modality lies at the intersection of reasoning and natural language. In linguistic theory and in logic modality has received intensive study. In NLP it has recently become relevant, as the field is moving beyond the identification of events and their participants towards making inferences about the (likelihood of) occurrence or non-occurrence of events. This workshop assembles linguists, logicians and computer scientists to assess standard and non-standard models of the meaning and inferential properties of modal expressions and to discuss the potential compatibility of seemingly disparate formal frameworks and the way their results can be integrated in NLP.
Many of our colleagues and alumni keep doing noteworthy things! Here are some of the recent ones:
- Roey Gafter received a RICSRE Graduate Dissertation Fellowship
- Kate Geenberg received a Whiting Dissertation Fellowship
- Kyu won Moon received a Geballe Dissertation Fellowship
- Bonnie Krejci was awarded a fellowship for the Linguistic Institute at University of Michigan this summer
- BA alum Joshua Falk, currently a Baggett Fellow at University of Maryland, was just awarded an NSF Graduate Research Fellowship
- PhD alum Judith Tonhauser has been awarded the Alexander von Humboldt-Forschungsstipendium für erfahrene Wissenschaftler from the Humboldt foundation
Stanford linguists have been quite busy!
Have you heard about the Great Language Change Hoax? If not, this article will enlighten you.