Indiana University

Previous month Previous week Next week Next month
See by year See by month See by week See Today Search Jump to month
Events for the week :
March 31, 2013 - April 06, 2013
Sunday
March 31
Monday
April 01
  • Inter-annotator agreement for dependency annotation of learner language

    Time: 01:00pm - 02:00pm 

    Place: Memorial Hall 401

     

    Marwa Ragheb and Markus Dickinson

    We report on a study of inter-annotator agreement (IAA) for a dependency annotation scheme designed for learner English. Reliably annotated learner corpora are a necessary step for the development of POS tagging and parsing of learner language. In our study, three student annotators marked several layers of annotation over different levels of learner texts, and they are able to obtain generally high agreement, especially after discussing the disagreements among themselves, without researcher intervention, illustrating the feasibility of the scheme. We pinpoint some of the problems in obtaining full agreement, including annotation scheme inclarities for learner innovations, interface design issues, and difficult syntactic constructions. In the process, we also develop ways to calculate agreements for sets of dependencies.

     

    In category: Computational linguistics

     

Tuesday
April 02
Wednesday
April 03
  • Discourse uses of subjects in Spanish: an account of VSO and SVO structures

    Time: 02:15pm - 03:15pm 

    Place: Ballantine Hall 004

     

    Chris Davidson (dissertation defense)

    Spanish is commonly characterized as a Subject, Verb, Object (SVO) language yet it also allows for variation of the placement of the subject in VSO and VOS structures. The SVO and VSO structures are controversial because there is not a consensus on what the communicational demands of the subject are when in the preverbal or postverbal positions. Alexiadou and Anagnotopoulou (1998), Ordóñez (2000), and Zubizarreta (1999) propose that VSO sentences are pragmatically neutral allowing the sentence to be used in broad focus contexts while SVO sentences are similar to CLLD structures and are constrained to those contexts where the subject can be interpreted as a topic. Contreras (1991), López (2009), and Suñer (1982) argue that the use of VSO sentences is restricted in specific contexts and cannot be used in broad focus contexts. Belletti (2004) further argues that while it is true that preverbal subjects in Spanish can be used to communicate topicality, they can also be used to communicate informational focus which suggests that SVO sentences are not constrained to a particular context.

    López (2009) proposes a system to account for CLLD, CLRD, Focus Fronting, and Rheme sentences using the features [+a]naphoric and [+c]ontrastive. These features are assigned at the phases proposed by Chomsky (2001) with [+a] being assigned to constituents moved to Spec,vP and [+c] being assigned to constituents moved to Spec,FinP. Using López’s system as a point of departure, this research will look at how SVO and VSO sentences are used in mini dialogues that have been designed to elicit constituents that communicate the features [±a] and [±c]. The assumption being that VSO sentences will be due to subjects being marked [+a] forcing the subject to remain in a postverbal position because of the communicational demands of the discourse.

     

    In category: Morphosyntax and semantics

     

Thursday
April 04
Friday
April 05
  • The Interaction of Information Structure and Syntactic Representation in Chinese

    Time: 01:15pm - 03:15pm 

    Place: Memorial Hall Room 317A

     

    Yuyin Hsu (Dissertation defense)

    This dissertation concerns the interaction of syntax and information structure in Mandarin Chinese and puts the theoretical assumption of parallelism between clauses and noun phrases to the test. It examines and validates the information structural status of the object phrases preposed to clause-internal positions. I argue that Rizzi’s (1997) “fine structure of the left periphery” proposed for Indo-European languages can be extended from the analysis of phrases that appear at the clause periphery to the analysis of phrases appearing in the sentence-internal domain in Chinese. This proposal is attested through the examination of the interaction between the preposed objects and other constructions, such as wh-questions, the ba-construction, modals, and the cleft-construction. Moreover, I argue that the so-called verb-copying structure and the fixed ordering of Topic and Focus elements in Chinese are better accounted for under the proposed analysis.

    The study also attempts to clarify the internal syntactic structure and its semantic effects of nominal expressions in Mandarin Chinese comparing them with those in other languages (Giusti 1996 and Aboh 2004). It is argued that the occurrence of measure words in Chinese represents the notion of “unit” that defines and quantifies over the noun phrase, independent of projections of demonstratives and nouns. Finally, I argue that nominals in Chinese are parallel to clauses in encoding the information structural properties to the elements appearing at the left periphery and the internal domain. This proposal accounts for the non-canonical distribution of adjectives, ellipses of various sizes of nominal phrases, and the phenomenon of internal reordering of words within nominal expressions. The dissertation introduces a wide range of new data which makes it possible to reevaluate previous proposals, and to propose a new perspective to examine the parallelisms between clauses and noun phrases in regard to the interaction between syntax and information structure.

     

    In category: Morphosyntax and semantics

     

Saturday
April 06
  • Challenges to designing morphological transducers for Turkic languages

    Time: 08:30am - 10:20am 

    Place: Woodburn Hall 106

     

    Jonathan Washington

    Morphological transducers are computational tools that map linguistic forms to morphological analyses. They have a variety of uses, ranging from their use as spell-checkers to the integral role they play in rule-based machine translation systems. This paper introduces a general academic audience to morphological transducers – how they work, how they are used, and how they are developed. More crucially, it presents some of the challenges faced in the development of morphological transducers for Turkic languages. Solutions to problems I have encountered in the development of a number of Turkic-language transducers are covered. Specifically, the talk covers the implementation of various issues related to the morphology, phonology, and orthography of these languages. While solutions to these problems are approached through knowledge of (and hence reflect in many ways) the linguistic systems employed in the languages, a functioning implementation is rarely as straightforward as what might be presented by a descriptive grammar of the language. Because of how explicit the solutions must be, a morphological transducer can be thought of as the ultimate description of a language’s morphotactics and morphophonology.

    Part of the 20th Annual Association of Central Eurasian Students Conference.
    For further details, see http://www.iub.edu/~aces/

     

    In category: Computational linguistics

     

  • Queer slang and demographic change in Early Republican Istanbul

    Time: 10:30am - 12:20pm 

    Place: Woodburn Hall 106

     

    Nicholas Kontovas

    This presentation will examine the variety of Turkish slang known as Lubunca, used primarily among gay men and trans women in Istanbul. By analyzing Lubunca in terms of both its vocabulary and its structure, I will demonstrate that it contains a large number elements derived from languages spoken by non-Muslim populations who lived in certain areas of Istanbul during the last years of the Ottoman Empire and the early years of the Turkish Republic. By comparing the current state of Lubunca to earlier forms of slang employed by proto-Queer populations in the same spaces, I show how the sharp division characterized by the use of non-Muslim language-derived elements in Lubunca coincides with a major shift in the social capital of non-Muslims within Ottoman and (later) Turkish society. With the rise of Islamo-nationalism and Turkish ethno-nationalism beginning on the eve of the twentieth century, economic opportunities for non-Muslims in Istanbul declined sharply. These ideological trends – combined with restrictions founded in Ottoman Islamic law – produced a market for sex work wherein there were over ten times the number of registered non-Muslim brothels during the Imperial Period than there were registered Muslim brothels. As laws intended on discouraging Turks from mixing with non-Turks developed, all of the registered non-Muslim brothels gradually shut down; however, Turkish society had come to fetishize the non-Muslim woman prostitute, and though the supply decreased the demand remained high. Often times in order to fill this gap, Christian and Jewish women were tracked from other parts of the Empire and surrounding territories. It is in their capacity as unregistered sex workers that these women seem to have come into contact with the gay male and trans female population of the city, who have never been legally capable of engaging in registered prostitution in either the Ottoman Empire or Turkey. By breaking down loanwords from non-Muslim languages in Lubunca into semantic categories, I conclude that it was in this environment that the most defining aspects of Lubunca develop. I also explore the possible social mechanisms whereby the Istanbulite Queer population would find it socially advantageous to adopt and perpetuate the use of Lubunca, as well as those social mechanisms which may have led to a recent decline it the use of Lubunca among younger generations of Turkish Queers.

    Part of the 20th Annual Association of Central Eurasian Students Conference.
    For further details, see http://www.iub.edu/~aces/

     

    In category: Sociolinguistics and pragmatics

     




JEvents v3.0.9 Stable   Copyright © 2006-2013