USC-GSIL

Student Workshop

May 9, 2005

presented by

Graduate Students in Linguistics

·                  Call for Papers

·                  Invited Speakers

·                  Program

·                  Abstracts

·                  Organizers - Contact us

Call for Papers:

 
 
For the fifth consecutive year, GSIL is hosting a Student Workshop on May 9, 2005. The GSIL Workshop was initiated four years ago, and is held at the end of each academic year.

In this day-long workshop students have a chance to present recent term papers, screening papers, qualifying papers, dissertations, and other work in progress to other students and the faculty in the department. Both undergraduate and graduate students are encouraged to present. Don't be shy! :) We want to know what you are working on!!! This will be an excellent opportunity to talk about your work, get feedback and develop presentation skills.

Papers from all areas of linguistics are welcome! We hope to have as many areas represented as possible!

Students who are interested in presenting should reply to this email stating the presentation's area AS SOON AS POSSIBLE.
A title and short abstract will be due May 2, 2005.

Information regarding this year's Student Workshop will be available at:

http://www-scf.usc.edu/~rebekaca/05StuWork.html

Please let us know if you have questions. We look forward to hearing from you soon!!!

Invited Speakers:

·      Laura Gonnerman

Lehigh University

“Form and meaning in morphology: Evidence from processing, acquisition, and  imaging”

·      Carolina González  

University of California, Los Angeles

“On morpho-phonological alternations” (see abstract below)

Program:

Abstracts:

On morpho-phonological alternations

Carolina González

University of California, Los Angeles

This paper explores some of the issues that arise in connection with a number of phonological and morpho-phonological alternations in Panoan languages. These alternations are reportedly sensitive to syllable count; however, recent analyses have proposed that they are conditioned metrically.

Phonological and morpho-phonological alternations in Panoan are connected in various ways. Both kinds alternate according to the metrical structure of the language, with some allomorphs/allophones occurring in weak syllables, and some in strong and unfooted syllables. They involve similar processes, mainly epenthesis, deletion and metathesis. Additionally, consonantal alternations involve the same segments: glottals, nasals, and voiceless stops. There are also some important differences. While phonological alternations typically involve consonants, morpho-phonological ones might involve vowels, consonants, or full syllables.

The in-depth study of Panoan alternations will bear on our understanding of the morphology-phonology interface and the representation and analysis of alternations. Some of the questions that need to be answered include the following:

1.                  What is the relationship between phonological and morpho-phonological alternations in general, and in Panoan languages more specifically? Are all of these cases caused by the interaction of conflicting phonological forces?

2.                  How are morpho-phonemic alternations to be analyzed? Is morpheme suppletion adequate for the analysis of all cases? Should one form be considered as basic, with rules or constraints selecting or deriving the attested allomorph in the right environment?

3.                  What contributes to the development or permanence of a morpho-phonemic alternation? Is the phonological makeup of these morphemes responsible for these alternations?

4.                  How did these alternations arise? Why are they connected to foot-sensitive positions with no direct dependence on stress?

This talk will discuss the points above, providing a preliminary investigation of the status of morphological alternations in the grammar and of the connection of these alternations with the phonology in Panoan.

Rewriting Syntax Trees to Improve Syntax-Based Machine Translation Quality

Bryant Huang

A syntax-based machine translation (MT) system utilizes syntactic information to automatically translate text from one natural language to another. For translation involving English, the Penn Treebank is a common  resource for such information. However, the tagging and bracketing decisions in this corpus were not specifically designed for MT and thus can yield unfavorable translations. In my work, I address these deficiencies by rewriting the treebank data -- through relabeling and restructuring -- to make it more suitable for MT and improve translation quality as a result. This describes a work in progress.

Variation in Rioplatense Spanish /s/ + stop Sequences

Michal Temkin Martínez

An experiment was conducted to study the variation in the production of stops following variants of /s/. Taken into account were word or phrase position as well as speech style. Tokens were analyzed for observed voicing and the presence or absence of a stop closure. Another variable which will be discussed is the coda /s/ variant. An Articulatory Phonology account is presented as an explanation for the mechanisms causing the variation of the stop, as well as presenting a relationship between the /s/ variant and the following stop.

The role of speech rate in dynamical linguistic changes

Alexsandro R. Meireles & Plínio A. Barbosa

Lafape/IEL/Unicamp

            This work, which is couched on the Articulatory Phonology theoretical framework, deals with the possible role of speech rate on diachronic changes from proparoxytones to paroxytones. On this kind of change, there is deletion of the medial post-tonic vowel of the proparoxytones. Our results suggest that speech rate can explain this historical process of linguistic change, since the medial post-tonic vowel reduces more, without deletions, than the final post-tonic vowel from normal to fast rate. These results were confirmed by Friedman's Anova. One-Way Anova also indicated that the duration of the medial post-tonic vowel is significantly smaller than that of the final post-tonic vowel. On the other hand, words such as "fôlego" (breath) and "sábado" (Saturday) reduce less their post-tonic segments in comparison with words such as "abóbora" (pumpkin). This finding, associated to Brazilian Portuguese phonotatic restrictions can explain why forms such as [folg U] e [sabd U] are not frequently found on this language. Besides, linguistic changes influenced by speech rate act according to dialect and gender. On this article, speakers from the Mineiro dialect (from Minas Gerais state) (rate: 7.5 syllables/sec.) reduced the medial post-tonic vowel more than speakers from the Paulista dialect (from São Paulo state) (rate: 6.4 syllables/sec.) and male speakers (rate: 5.8 syllables/sec.) reduced the medial pos-tonic vowel more than female speakers (rate: 5.2 syllables/sec.). These results were also confirmed by One-Way Anova.

‘It’s like talking in style’: Heritage speakers’ use of discourse markers in Spanish

Ana Sánchez-Muñoz

My work examines the style and register variation in heritage language speakers of Spanish. In particular, the use of discourse markers in informal and academic oral registers is examined in order to identify their distribution and variation across situations of use. Of particular interest in my study is the analysis of the marker como ‘like’, which in the colloquial oral registers of bilingual speakers seems to be spreading in the same direction as the marker ‘like’ in American English. The results of this study show that the choice of discourse markers, their distribution and relative frequency varies across registers.

Organizers:

o             Rebeka Campos

o             Michal Temkin Martínez

Logistics:

o             Andrea McColl

o             Stephen Tobin