Go to USC home page USC Logo USC: COLLEGE OF ARTS AND SCIENCES: DEPARTMENT OF LANGUAGES, LITERATURES AND CULTURES
UNIVERSITY OF SOUTH CAROLINA
COLLEGE OF ARTS AND SCIENCES | LINGUISTICS HOME |
WHAT IS LINGUISTICS?

PEOPLE

ACTIVITIES

COURSES

BABBLE NEWSLETTER

GRADUATE PROGRAM

UNDERGRADUATE PROGRAM

RESOURCES

RELATED SITES:

ANTHROPOLOGY

COMMUNICATION SCIENCES AND DISORDERS PROGRAM

ENGLISH

ENGLISH PROGRAM FOR INTERNATIONALS

INSTRUCTION AND TEACHER EDUCATION

LANGUAGES, LITERATURES, & CULTURES

PHILOSOPHY

PSYCHOLOGY

CAREER DEVELOPMENT PROGRAM
USC  THIS SITE

SPEAKING OF TOPIC INTRODUCTIONS IN THE LADIES AUXILIARY: A SINGLE-GENDER AND MIXED-GENDER COMPARATIVE ANALYSIS.

Author: THERESA MCGARRY
Degree: Ph.D.
Year: 2004
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Advisor Name: Ahearn-Laura-M
ISBN: 0493971254
Number of Pages: 243
Language: ENGLISH

This study examines topic introductions in the all-women and mixed-gender business meetings of an outdoor sports club in the Midwestern U.S. Previous research on gender and language has identified certain features that tend to occur with different frequency in the speech of women and men. However, gender-indexing language has also been shown to be instantiated differently in different contexts. To gain more understanding of how gender, context, and language usage interact with regard to topic introductions, I compare the introductions made by the women club members in the two meeting types and the men members in the mixed-gender context. The three-way comparison is carried out with regard to multiple aspects of topic introductions. In addition to frequencies of topic introductions and changes among the three groups, I examine types of introductions distinguished according to specific kinds of topic relevance and types of topic changes distinguished according to the interaction immediately preceding them. I also analyze the processes of topic introduction that occur in the meetings, i.e. types of interaction sequences that effect introductions. Finally, I compare the frequencies of specific elements of introductions, such as questions and new-topic pre-signals. The results of these analyses are interpreted with attention to the functions of the introduction utterances in the meetings.

 LANGUAGE, AND IDENTITY: L2 ACQUISITION IN POST-SOVIET MOLDOVA

Author: MATTHEW H. CISCEL
Degree: Ph.D.
Year: 2003
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Advisor Name: Ahearn-Laura-M
ISBN: 0493971254
Number of Pages: 281
Language: ENGLISH

Abstract: The study draws on diverse fields from second language acquisition theory and linguistic anthropology to history and language policy. It focuses on two research questions: (1) To what extent do attitudes toward languages correlate to competing notions of national and social identity within the Republic of Moldova? (2) How does the variability in these attitudes and identities affect the acquisition of second language (L2) proficiency in standard Russian, Romanian, and English? The first question is addressed using ethnographic and psychometric methods, including the matched guise technique and follow-up interviews. Survey respondents include over one hundred students of English in Moldova's capital city. The second question is tested by comparing attitude data with measures of L2 proficiency in a small subset of the survey respondents. Together with qualitative explorations of Moldova's recent history and its social milieu, the quantitative results of the surveys suggest that language attitudes and social identities create predispositions with regard to the acquisition of a particular L2. Specifically, evidence is found for a post-colonial effect that continues to maintain the status of Russian, despite policy efforts to establish a stronger role for Romanian. In addition, the role of English, as an international language associated with ideologies of progress, is argued to further complicate the dynamics of multilingualism and identity crisis in the country. The use of multiple methods and models related to linguistic and social identity creation contribute to a textured, complex presentation of socially situated L2 acquisition in Moldova, informing both language policy in the region and the often overlooked social aspects of L2 acquisition theory.
Subject Code: Anthropology-Cultural (0326); Language-Linguistics (0290); History-European (0335)
Source: VOLUME 64-01A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 198
Publications Number/Order Number: AAI3076758

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .

THE PRO-DROP PARAMETER IN SECOND LANGUAGE ACQUISITION REVISITED: A DEVELOPMENTAL ACCOUNT

Author: LARRY LAFOND
Degree: Ph.D.
Year: 2001
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Advisor Name: HOLT, D. ERIC
ISBN: 0493316973
Number of Pages: 249
Language: ENGLISH

Abstract: This dissertation applies a particular theory of language acquisition and representation, Optimality Theory (Prince and Smolensky 1993, Grimshaw 1997), and a particular learning algorithm within this theory, the Constraint Demotion Algorithm (Tesar and Smolensky 2000), to the problem of how second language acquisition of pro-drop takes place for learners whose first language does not instantiate the grammatical properties traditionally associated with pro-drop.

The overarching goal of this study is to provide an account of the developmental stages in the second language learning of three grammatical properties: null subjects, inversion, and that-trace. Although there is no lack of such accounts from earlier generative perspectives, the need remains for a comprehensive developmental account from an Optimality-theoretic perspective. This dissertation begins to address that need.

The study here is based on several empirical tests (a translation task, a pilot study, and a grammaticality judgment task) that were administered to 370 adult native English speakers studying Spanish at the University of South Carolina or the Pennsylvania State University. Each task was designed to investigate learner competencies regarding null subjects, inversion, and that-trace. A key conclusion from these studies is that the acquisition of Spanish by native speakers of English involves a reranking of universal syntactic and discoursal constraints in these languages. Specifically, this dissertation argues that acquisition of Spanish occurs through the demotion of certain syntactic constraints in the English native grammar so that these constraints are dominated by discoursal constraints in the Spanish second language grammar.

This cross-sectional study not only tracks learners through developmental stages, but it is also theory driven, because the theory of grammar used in this dissertation permits specific predictions about the interaction and relative importance of constraints in Spanish and English and, ultimately, of the acquisitional route learners take. The application of Optimality Theory to interactions between discourse and syntax in second language learning represents a new and potentially productive line of inquiry that may advance our understanding of both second language learning and grammatical theory.
Subject Code: Language-Linguistics (0290); Language-General (0679); Education-Language-and-Literature (0279)
Source: VOLUME 62-07A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 2402
Publications Number/Order Number: AAI3020955

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .


A SOCIOPRAGMATIC APPROACH TO THE USE OF META-DISCOURSE FEATURES IN EFFECTIVE NON-NATIVE AND NATIVE SPEAKER COMPOSITION WRITING

Author: EUNITA D.A. OCHOLA
Degree: Ph.D.
Year: 2001
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Advisor Name: CAROL MYERS-SCOTTON
ISBN: 0493317902
Number of Pages: 309
Language: ENGLISH

Abstract: Issues of writing revolve around effective writing. Many instructors are concerned that their students do not write effectively, while most students are concerned that they do not know how to write so they are understood. Furthermore, some non-native writers write more effectively than others, including some native-writers.

This dissertation examines factors that determine effective writing by non-native and native speakers. The study assumes that text production involves satisfying a text at three related levels: the schematic superstructures, the propositional content, and messages of intentionality. <italic>Meta-discourse features</italic> (such as <italic><underline>By this I mean</underline></italic>&hellip;) express messages of <italic>intentionality</italic> that convey attitude toward the subject matter (including interpersonal and intrapersonal involvement). The study investigates how writers use schematic superstructures and <italic> meta-discourse features</italic> in effective writing.

Meta-discourse features in 64 compositions were examined: 32 ESL compositions by Dholuo first language subjects, in their third year at Kenyatta University, Kenya, and 32 compositions by English native-speaker freshman at Midlands Technical College, Columbia, SC. Prior to the analysis, three native-speaker composition instructors independently rated the compositions as effective or ineffective.

A sociopragmatic approach based on the theory of intentionality and a rational choice model as explicated in the Markedness Model (Myers-Scotton 1993 &amp; 1998), the theory of Generalized Conversational Implicatures (Levinson 2000), the Cooperative Principle (Grice 1975), and Relevance Theory (Sperber &amp; Wilson 1986) was adopted for data analysis. The results show that both non-native and native writers conformed to the target language's schematic superstructures; they also used meta-discourse features to express messages of intentionality. A major result of this study shows that effective writers used more meta-discourse features to express messages of intentionality than ineffective writers; results were almost identical for both native and non-native speakers.

The study claims that effective writers exploit their linguistic repertoire to maximize audience awareness and indications of their own awareness to achieve maximum communicative reward, and concludes that meta-discourse features structure discourse at a higher level than propositional content. Consequently, instructors should sensitize their students to the use of meta-discourse features, and writers should view writing as an interactive enterprise between themselves and their reader and between themselves and their text.
Subject Code: Language-Linguistics (0290); Language-Rhetoric-and-Composition (0681); Education-Language-and-Literature (0279)
Source: VOLUME 62-07A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 2404
Publications Number/Order Number: AAI3020964

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



BENEATH THE SURFACE: SIGNS OF LANGUAGE ATTRITION IN IMMINGRANT CHILDREN FROM RUSSIA

Author: ELENA SCHMITT
Degree: Ph.D.
Year: 2001
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Advisor Name: CAROL MYERS-SCOTTON
ISBN: 0493236058
Number of Pages: 287
Language: ENGLISH

Abstract: This longitudinal study focuses on changes in the grammatical structures produced by Russian pre-adolescent speakers under the circumstances of language attrition in the English dominant environment. The central goals are: (1)&nbsp;to identify processes involved in language attrition; (2)&nbsp;to determine the mechanisms that underlie language attrition; and (3)&nbsp;to discuss the relative hierarchy of stages of language loss. A total of 2,182 CPs are analyzed for the presence of signs of language attrition which may be present overtly through &lsquo;classic&rsquo; codeswitching, bare form production, generalization of certain morphemes, etc. and/or covertly in the form of convergence, bare form production, reduction, simplification, and others.

All the analyses are carried out within the frame work of the Matrix Language Frame model (Myers-Scotton 1993a, 1997), the 4-M model (e.g., Myers-Scotton &amp; Jake 2000a, 2001), and the Abstract Level model (e.g., Myers-Scotton &amp; Jake 1995, 1998). The analysis of the data has shown that the amount of convergence and bare form production increases in the second data set. The levels of codeswitching remain rather stable; however, the patterns of codeswitching substantially change between the first and the second recordings. The difference between &lsquo;classic&rsquo; codeswitching, convergence, and bare form production is discussed in terms of the availability of the matrix language predicate-argument structure and morphological realization patterns that lead to the production of integrated forms. The diminished access to forms and convergence. The findings further indicate that some types of morphemes (e.g., content and early system morphemes) are lost more easily from the language than others (late system morphemes). It is also demonstrated that the underlying language at various levels depending on the stage of language attrition. Finally, it is shown that all the processes of the first language attrition involve the mechanisms of codeswitching and convergence.

Overall, an integrated account of mechanisms, processes, and stages of language attrition resulting from a close contact of English and Russian languages in the English dominant environment is proposed.
Subject Code: Language-Linguistics (0290); Sociology-Ethnic-and-Racial-Studies (0631)
Source: VOLUME 62-05A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1817
Publications Number/Order Number: AAI3013450

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



SIMPLIFIED INPUT: AN IVESTIGATION OF FOREIGNER TALK/TEACHER TALK ON COMPREHENSION AND VOCABULARY ACQUISITION

Author: RICHARD HALLETT
Degree: Ph.D.
Year: 1999
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: BRUCE PEARSON
ISBN: 0493149333
Number of Pages: 151
Language: ENGLISH

Abstract: This dissertation addresses the following research questions: (1) Does simplified input (SI) facilitate L2 comprehension? (2) Does SI facilitate L2 comprehension across learners at different levels of proficiency? and (3) If SI does facilitate L2 comprehension, do the learners retain this acquired knowledge over a period of time? In the main study, 75 ESL learners enrolled in an intensive English language program at an American university were divided into two groups. Subjects in the experimental group watched a videotaped SI lesson on specific lexical items in addition to viewing a videotape of a news segment. Subjects in the control group received no SI; rather, they simply viewed the same news segment. The study found that (1) SI does facilitate L2 comprehension and acquisition of lexical items in the oral mode, (2) SI does facilitate L2 comprehension and acquisition across learners at different levels of proficiency, (3) SI appears to be used in the same basic manner by all L2 learners, regardless of their level of proficiency, and (4) L2 learners retain the knowledge acquired from SI over a period of time.
Subject Code: Language-Linguistics (0290)
Source: VOLUME 62-02A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 551
Publications Number/Order Number: AAI3006037

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



LANGUAGE CONTACT AND COMPOSITE STRUCTURES IN NEW IRELAND, PAPUA NEW GUINEA

Author: REBECCA SUE JENKINS
Degree: Ph.D.
Year: 2000
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON
ISBN:0493149414
Number of Pages: 315
Language: ENGLISH

Abstract: This dissertation examines language contact phenomena in New Ireland Province of Papua New Guinea. It discusses a wide range of such phenomena including codeswitching (CS); borrowing; convergence; language shift, attrition, and death; and pidgin/creole formation within the theoretical framework of the related extended Matrix Language Frame, Abstract Level, and 4-M models. The concentration of languages in New Ireland, the widespread use of Tok Pisin as a lingua franca, and an educational system using English as the medium of instruction creates an environment of extensive language contact.

The primary focus of the study is a comparison of Tok Pisin with the Austronesian (AN) substrate languages, using Tigak as a typical AN language. The hypotheses tested propose that the AN substrate provides the morphosyntactic frame for Tok Pisin and that this frame is a composite Matrix Language (CML) based on those very similar Austronesian languages, that the (CML) and constituent types change as the linguistic situation changes, that morpheme type restricts the source of morphemes in a pidgin, and that the direction of influence can change as a contact language stabilizes. The results provide extensive evidence of the composite nature of the morphosyntactic frame of Tok Pisin and of its AN source and demonstrate current changes in constituent types and the (CML) structure in Tok Pisin due to renewed English influence. The results verify that content and early system morphemes may come from any language contributing to a pidgin, but that late system morphemes are restricted in predictable ways. The variability produced by L1 interference is also documented.

A secondary focus of the study is the analysis of other contact situations in northwestern New Ireland. The results show that the languages have influenced each other in the past but that the strongest current influences are those exerted by Tok Pisin and English on the indigenous languages. The influence of these two languages is producing extensive CS, borrowing, language change due to convergence to Tok Pisin and English, and language attrition.
Subject Code: Language-Linguistics (0290)
Source: VOLUME 62-02A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 552
Publications Number/Order Number: AAI3006044

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .


THE ROLE OF ABSTRACT LEXICAL STRUCTURE IN FIRST LANGUAGE ATTRITION: GERMAN IN AMERICA

Author: STEVEN GROSS
Degree: Ph.D.
Year: 2000
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON
ISBN:0599875313
Number of Pages:220
Language:ENGLISH

Abstract: in first language attrition in a language contact setting. Framed within the extended Matrix Language Frame model, this study seeks to explain the patterns of linguistic variation that lead to language change in the speech of German immigrants under long-term contact with English. To do this, a group of six elderly bilingual German immigrants who have been living in the United States for at least 40 years were interviewed. These interviews took the form of informal conversations in the participant's native German dialect and generated approximately eight hours of linguistic data.

This dissertation argues that the changes occurring in the linguistic system of the German immigrants come from the innate organization of the mental lexicon and the way in which a linguistically-encoded message is generated. Under the psycholinguistic pressures of language contact, an individual's first language will be reorganized on the basis of how and when linguistic units become available to language production.

structure: lexical-conceptual structure, predicate-argument structure, morphological recombined to from a composite structure.

The morphosyntactic features present in the immigrants' linguistic system indicate that the system has changed to the extent that a composite Matrix Language projects the grammatical frame. However, these effects are not uniform throughout the system. The data indicate that morphemes that are activated early in the production process, at the level of lexical-conceptual structure, are the most vulnerable to the effects of language attrition. Morphemes that are activated at a later stage, those that encode language-particular grammatical distinctions, are the most resistant to attrition. These findings challenge the results of a number of other language attrition studies that indicate that attrition is most evident in the early loss of distinctions such as case assignment, gender, and subject-verb agreement.
Subject Code:Language-Linguistics ( 0290) ; Language-Modern ( 0291)
Source:VOLUME 61-07A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 2681
Publications Number/Order Number:AAI9981259

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .


WHAT KIND OF PEACE IS THIS? METAPHOR IN THE U.S. PRESS COVERAGE OF ISRAELI-PALESTINIAN NEGATIONS

Author: CATHLEEN BRIDGEMAN
Degree: Ph.D.
Year: 2000
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: ANNE BEZUIDENHOUT

Subject Code: Language-Linguistics (0290); Mass-Communications (0708); Journalism (0391); Political-Science-International-Law-and-Relations (0616)
Source: VOLUME 61-04A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1376

This study examines the metaphors for peace used in the American press coverage of the Israeli-Palestinian, peace process. Metaphors are considered within the framework of conceptual metaphor theory as discussed in Lakoff and Johnson (1980) and Lakoff (1993). The study seeks to answer four research questions concerning, metaphor frequency, type, variation, and range in an effort to explore the understanding of peace inherent in the metaphors used in the press coverage and to contribute to empirical and discourse level considerations of conceptual metaphor theory. It is argued that conceptual metaphor theory offers a way to account for the range of metaphors used systematically and concepts like peace. It is also argued that considering metaphor in actual discourse indicates areas in need of further development in conceptual metaphor theory and demonstrates the need for more attention to the political and social implications of metaphor choice.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



PHONOLOGICAL PROCESSES IN ENGLISH AND KOREAN: DIRECT OPTIMALITY THEORY APPROACH

Author: CHANG-KYUM KIM
Degree: Ph.D.
Year: 1999
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: BRUCE PEARSON

Source: VOLUME 61-04A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1379

The purpose of this study is to analyze certain Korean and English phonological processes and to examine sound replacement in L2 learners using Direct Optimality Theory (DOT). DOT has its roots in Standard Optimality Theory (SOT) which in turn has its roots in Generative Phonology. It allows constraint violations and evaluation. However, this theory is different from Standard Optimality Theory in that it represents phonological forms with pure markedness and requires phonological information to be uniform in all levels.
This study based on DOT claims that prosodic information is present at all phonological levels. SOT and Generative Phonology represent every underlying form as an impossible form since underlying forms are not syllabified whereas surface forms are. However, this study examines some Korean phonological processes, and argues that prosodic information is consistent at all phonological levels.
This study also argues that DOT succeeds in being more explanatory with respect to Korean and English phonological processes such as consonant clusters, neutralization, unreleasing, and palatalization.
In addition, this study deals with sound substitution in Korean-speaking learners of English and English-speaking learners of Korean based on DOT. This study claims that the replacement of L2 sounds with L1 sounds is caused by constraint transfer of marked features from a learner's native language. Direct Optimality Theory represents a word or a segment with the constraint violability of pure markedness. Each marked feature is a violated constraint with a ranking. This violated constraint is transferred when L2 learners learn unfamiliar sounds in the target language. English-speaking learners of Korean transfer English violated constraints of marked features for Korean sounds when they learn Korean, while Korean-speaking learners of English transfer Korean violated constraints of marked features for English sounds.
This dissertation argues that Direct Optimality Theory is more plausible in explaining Korean and English phonological processes than Generative Phonology and Standard Optimality Theory. The study uses DOT to show how second language learners replace L2 sounds for L1 sounds in the early stages of learning.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



NEGOTIATING POWER IN BUSINESS MEETINGS

Author: MELANIE MOLL
Degree: Ph.D.
Year: 1999
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: LAURA M. AHEARN
Subject Code: Language-Linguistics (0290)
Source: VOLUME 61-04A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1381
Publications Number/Order Number: AAI9969511

Abstract: The purpose of this dissertation is to qualitatively examine discourse practices within business meetings held by the representatives of the South Carolina BMW Manufacturing Corporation and two of its leather suppliers, Draexlmaier Automotive of America (DAA) and Sommer Allibert Industries (Allibert). The situational requirements of the three companies involved in these meetings are such that the interactions can be seen as intrinsically competitive. In other words, each company presumably seeks to achieve the best possible outcome for itself and its members. Against the backdrop of this scenario, I examine the construction of directives, evaluations, and agency assignment as these relate to conversational styles. Although conversational styles have often been characterized as powerful or powerless depending upon the various linguistic features they contain (Brown and Levinson 1987), my data show speakers often use a style that establishes a mutually supportive, i!
nclusive team. Thus, while the situation remains competitive, speakers discover within the organized activity the local logic of collaboration (Boden 1995). Constrained by the JIT (Just-In-Time) manufacturing system, all three companies are under the same time pressure to deliver products as efficiently as possible to their respective customers. Because of the interdependence of a supplier network, being mutually supportive through the use of a collaborative, indirect style is a way to consolidate effort and achieve maximum effect. The motivation of participants' usage of this conversational style is based on the requirements of the type of interaction in which they participate, and features are thus best described as multifunctional. In this way, the characterization of a style as weak or powerful is measure in connection with the situational context of the interaction rather than being intrinsic to a particular feature or strategy.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



RELEVANCE THEORY AND THE MARKEDNESS MODEL IN SLA: COGNITIVE APPROACHES TO PRAGMATICS AND SECOND LANGUAGE ACQUISITION

Author: MARY SUE SRODA
Degree: Ph.D.
Year: 2000
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON
Subject Code: Language-Linguistics (0290)
Source: VOLUME 61-04A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1382
Publications Number/Order Number: AAI9969526

This study explicates a model of pragmatic enrichment based on Sperber and Wilson's Relevance Theory (1986, 1995, 1997). This model differs from traditional Relevance Theory in two ways: (1)&nbsp;the process of pragmatic enrichment is characterized as non-linear and modular, thus describing pragmatic enrichment not as a module but as a modular process; (2)&nbsp;a mechanism for evaluating explicature and implicature relative to community-wide standards and norms for discourse is integrated as a module in pragmatic enrichment (Myers-Scotton (1993, 1998, forthcoming). The modified model is referred to as the RT/MM.
An analysis of pragmatic failure (PF) in three naturally-occurring corpora of second language (L2) learners of English was conducted to investigate if the RT/MM can account for the different types of pragmatic failure which occur in authentic discourse. The data support an RT/MM perspective of L2 pragmatics.
What the RT/MM offers is a unified means to explain a wide range of instances of pragmatic failure. In doing so, this model makes it clear that pragmatic failure is not idiosyncratic or unanalyzable. Rather, even though one type of PF differs from another type, the RT/MM shows how all types of PF result from a breakdown of related premises on how communication operates.
The RT/MM provides a new level of explanation of particular use to researchers and teachers in the field of L2 pragmatics. That is, with the ability to account for how hearers recover ostensive inferential meaning relating both to propositions and social relationships, the RT/MM provides a more comprehensive perspective of how second language learners acquire pragmatic ability.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



THE HIDDEN DIMENSIONS OF LANGUAGE CONTACT:THE CASE OF HUNGARIAN-ENGLISH BILINGUAL CHILDREN

Author: BOLONYAI, AGNES
Degree: Ph.D.
Year: 1999
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTO
Source: DAI-A 60/07, p. 2466, Jan 2000
Descriptors: LANGUAGE, LINGUISTICS (0290); SOCIOLOGY, ETHNIC AND RACIAL
STUDIES (0631); LANGUAGE, MODERN (0291)
Descriptor Codes: 0290; 0631; 0291

This dissertation examines the effects of language contact and first language (L1) attrition in Hungarian as spoken by Hungarian-English immigrant children. It explores how and what organizational principles of the mental lexicon play a role in determining what types of surface structures occur in other language contact phenomena, such as convergence and L1 attrition in bilingual language production. Naturally-occurring bilingual speech produced by six Hungarian-English bilingual children who are growing up in the United States form the data base. The seven to nine years old children were English dominant at the time of the data collection. The data analysis examines morpheme distribution and changes in morphemic structures in terms of three levels of abstract lexical structure and how morphemes are elected in production. It is demonstrated that the children produce some Hungarian morphemes less accurately than other morphemes when speaking Hungarian. The asymmetrical distributions of preverbs and case endings are of specific interest in this study.
The main claim of the study is that what occurs in L1 attrition and convergence largely depends on how the level at which an L1 morpheme is accessed interacts with level(s) at which the competing L2 morpheme is accessed. The results show that system morphemes that are accessed late in the production are more likely to be affected by attrition first than system morphemes that are elected early. A major result of this study is that insights into the nature of preverbs in Hungarian are captured in two new principles, which explain their distribution in L1 attrition. Further, the study demonstrates that much of what appears to be &lsquo;loss&rsquo; of surface morphemes in L1 attrition is, indeed, replacement of abstract features in the L1 by corresponding features in the L2 at different levels of production. The result is the emergence of a composite Matrix Language (ML), in which levels of abstract lexical structure are &lsquo;split&rsquo; and &lsquo; recombined&rsquo; from two linguistic systems. The study concludes that the restructuring mechanisms in L1 attrition and convergence are systematic and reflect how children attempt to keep their linguistic systems internally consistent.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



STRUCTURAL CONSTRAINTS ON ARABIC/ENGLISH CODESWITCHING:TWO GENERATIONS

Author: OKASHA, MAHA
Degree: Ph.D.
Year: 1999
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON

Source: VOLUME 60-04A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 1107.
Descriptors: LANGUAGE, LINGUISTICS (0290)
Descriptor Codes: 0290

This dissertation examines the bilingual speech of two generations of Arab-Americans in the city of Columbia, South Carolina. The focus of this study is on the codeswitching (CS) patterns of these speakers. CS is a language contact phenomenon in which two or more languages are used within the same discourse unit. Although CS serves social functions such as indicating social status, prestige, and language attitudes, it is not a random linguistic behavior. Research has shown that it is subject to structural constraints that indicate when and where CS takes place in the discourse. The emphasis of this study is on these constraints as formulated in the Matrix Language Frame model by Myers-Scotton (1993a) and extended later by two submodels by Myers-Scotton and Jake (1995, 1998).
There are two sets of data for this study. In one set, I recorded the speech of twelve Arab-Americans who have been in the United States for at least ten years. In the other set of data, I recorded the speech of ten second generation Arab-Americans who were born and raised in the United States. A total of sixteen hours of recorded conversation was collected. The recording was conducted on a one-to-one basis in order to collect natural speech in a relaxed setting.
This study has two goals. One is to analyze the two data in the light of the MLF model to test its validity, and the other is to compare CS patterns of the two generations and try to account for their differences. This study reveals how incongruence between languages at an abstract level can affect the patterns of CS. It also shows how the socio-pragmatic orientation of the second generation affects the structure of their CS.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



THE LINGUISTIC SITUATION IN VALLE D'AOSTA:A STUDY ON THE FUNCTION AND THE STRUCTURE OFCODESWITCHING AND CONVERGENCE BETWEEN ITALIAN AND FRENCH

Author: SARULLO, PAOLA
Degree: Ph.D.
Year: 1998
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON

Source: VOLUME 60-02A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 408.
Descriptors: Language-Linguistics (0290); Language-Modern (0291)
Descriptor Codes: 0290, 0291

This dissertation examines the role Italian plays in the French spoken by bilingual native speakers of Valle d'Aosta, a region in northwest Italy. Valle d'Aosta presents an interesting language contact setting because two official languages with equal status (Italian and French) and a dialectal variety (Patois) co-exist in the repertoire of this speech community. However, Italian remains predominant in the region, because the region receives economic aid from the Italian government, the media are for the most part in Italian, and the medium of instruction in the schools is mainly Italian.
The corpus analyzed consists of audio-recorded naturally occurring conversations and interviews of bilingual Valdostans. The language of these data is what the bilingual native speakers themselves refer to as French. However, this French is characterized by a composite Matrix Language with characteristics mainly from French, but also from Italian and Patois. That is, what is called Valle d'Aosta French (VDA French) is not a single language, but is a composite variety containing features from both Italian and French. While the surface forms all come from French, some of their abstract lexical structure comes from Italian.
The goal of this research is to analyze the influence of Italian in regard to grammar and abstract lexical structure. In some instances, abstract structure from VDA Patois is also apparent. In addition, codeswitching that includes Italian or VDA Patois lexical items in a French grammatical frame is also studied.
A general finding, supported by quantitative analysis and, as predicted by the model of analysis, is that structures found in codeswitching and convergence cannot always be explained only in terms of surface configurations. That is, parts of abstract structures interact and are organized in principled ways, even though such utterances may appear to be the product of a disorganized use of language.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



THE ACQUISITION OF NARRATIVE SYNTAX

Author: REID, DAVID
Degree: Ph.D.
Year: 1998
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: ANNE BEZUIDENHOUT

Source: VOLUME 59-07A OF DISSERTATION ABSTRACTS INTERNATIONAL. PAGE 2476
Descriptors: Language-Linguistics (0290); Psychology-Developmental (0620)
Descriptor Codes: 0290, 0620

This dissertation examines the manner in which children aged four through eleven acquire the culturally shared structure of narrative form. Based on the categories and general narrative nomenclature described by William Labov, this study breaks down narratives into abstracts (prefaces), orientations, high points and resolutions, and examines the synchronic make-up and diachronic development of each for children in the critical period of acquisition. The data, consisting of 405 narratives collected at The Hammond School, in Columbia, South Carolina, during the 1994-1995 school year, is a departure from (and addition to) that of Labov in that it was produced by pre-teen children and, importantly, was entirely naturally occurring and unelicited.
The findings with respect to the development of prefaces showed that the preface as a multiturn structural component of narrative form is acquired early and maintained throughout the period under study. More importantly, it was observed that the internal nature of the narrative "work" being performed in the preface evolved considerably, moving from a semantically empty discourse marker to an evaluated characterizational abstract, suspending conversational turn-taking and orchestrating listener response--a developmental process "from the outside in." Similarly, the orientation sections of developing narrators appear at early ages, the nature and degree of orientation increasing as the children reach maturity, another case of function following form. The high point--the hingepin of mature narrative discourse--was found to be typically indiscernible in 4-year-old speakers; however, by age 11, the high points, or climactic moments of the narratives, were identifiable in 100% of it! he sample. This critical developmental curve was due in large part to the gradual acquisition of evaluative linguistic skills. These consisted primarily of repetition, negatives, descriptive terms, reference to internal psychological states, reported speech and more complex syntactic forms--all devices for marking and evaluating the components of the narrative. The evaluative linguistic devices both communicate the point of the narrative and cue the listeners toward the expected responses. Resolutions and codas were also found to be vital aspects of narrative development, examples showing both logical resolution of conflict and
temporal frame shift as closing mechanisms.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



INTERACTIONAL STRATEGIES AND THE ROLE OF QUESTIONS IN THE ACQUISITION OF ACADEMIC DISCOURSE (DISCOURSE ACQUISITION)

Author: LAUNSPACH, SONJA LORENE
Degree: PH.D.
Year: 1998
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: ANNE BEZUIDENHOUT

Source: Volume 5905A of Dissertations Abstracts International. Page 1547 . 253 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290 ; 0681

This dissertation examines how freshman composition students acquire the discourse of academic writing in a small group setting. The study is grounded in 2 primary frameworks, Conversation Analysis and the Situated Learning Model of Lave and Wenger (1991). Specifically this research investigates the role that questions and interactional strategies play in the acquisitional process of the students.
The data for the research consist of video tapes of 2 small writing groups led by a facilitator that met weekly for an entire semester. A detailed analysis of the transcripts of the group interactions has revealed 6 discourse functions for questions: SEQUENCE, FOCUS, DIRECTION, PEDAGOGICAL, INFORMATION and ELABORATION. In addition to the question functions found, 5 interactional strategies were identified: LEADER INTERPRETATION, RESTATEMENT OF THE ASSIGNMENT, ASPECT FOCUS, SOLICITATION OF PEER INTERPRETATION, and ADVICE GIVING.
The results of the analysis show that the use of questions and interactional strategies contribute to the acquisition process of the student. First, the group leader uses these elements to model both the discourse and writing process. Secondly, these interactional practices encourage the students to participate in the discussions, and active participation facilitates learning (Lave and Wenger 1991). Thirdly, participating in these interactional practices allows the students to mediate and construct new understandings of the different parts of the writing process through talk (Vygotsky 1987). Fourthly, the use of these strategies over the course of the semester provides a quantity of input necessary to the acquisition process.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



PENNSYLVANIA DUTCH WITH A SOUTHERN TOUCH": A THEORETICAL MODEL OF LANGUAGE CONTACT AND CHANGE (GERMAN, CODE SWITCHING, CONVERGENCE, BILINGUALISM)

Author: FULLER, JANET MCCRAY
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Major Professor: CAROL MYERS-SCOTTON

Source: Volume 5806A of Dissertations Abstracts International. Page 2182 . 229 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290

This dissertation examines the role language contact plays in inguistic change in bilingual settings. Framed within the Matrix Language Frame model, this research seeks to describe and explain the phenomena and mechanisms of contact-induced language change. To do this, three data sets are used. The first and second are corpora of interviews with twenty native German speakers who, at the time of data collection, were living in the United States and spoke fluent English. These interviews are divided into distinct corpora according to the length of time the participants had spent in the United States. In the first, the participants were students who had spent approximately a year in the United States; in the second, the participants were permanent residents who had lived in America for at least five years. The third corpus consists of eighteen interviews with Pennsylvania German speakers; these data are the main focus of the analysis. The Pennsylvania German data were collected among Mennonites in South Carolina.
Using these three corpora, a time-line of language contact phenomena can be established. The development of specific features of language contact and change can then be tracked across the data sets to contribute to our understanding of the mechanism of language change. In this analysis, English is the donor language, providing structural input into a changing German variety; German is the recipient language for changes.
The main claim of this analysis is that contact-induced language change is lexically-based; that is, structural features are brought into recipient language in connection with lexical items from the donor language. Thus lexical borrowing is the catalyst for the borrowing of grammatical morphemes, as well as linguistic material on all three levels of complex lexical structure: the lexical-conceptual level, the level of predicate-argument structure, and the level of morphological realization patterns.
The features examined in this study include: plural and participial marking; past tense marking; calques and semantic shifts; the progressive aspect construction; subordinate clause word order; noun phrase structure; the loss of reflexive markers; and nominative experiencer constructions.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



DISAGREEMENT IN JAPANESE: THREE CASE STUDIES (CONVERSATION, ARGUMENTATION, CROSS CULTURAL COMMUNICATION)

Author: DORRILL, MASAKO AMEKURA
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Major Professor: GRETA D. LITTLE

Source: Volume 5806A of Dissertations Abstracts International.
Page 2181 . 256 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290

Based on three representative case studies of disagreement in Japanese, this dissertation argues for a fundamental disagreement structure that is a critical element in the broader interactional constraints of disagreement.
The fundamental disagreement structure observed in these data sets is as follows: the next turn speaker selects his or her turn (i.e., disagreement turns tend to be optional) and also the point to be disagreed with in the prior turn speaker's utterance by referring to it explicitly or implicitly, then presenting his or her own contrasting point. Forming a contrast is obligatory, imposing the structural constraints. Thus once an explicit contrast is formed, also because of the optionality of the disagreement turn, no response to the disagreement implies a possible agreement; thus an interlocked competitive turn-taking with simultaneous talking/overlaps ensues. How to identify an argumentative point is therefore critical.
This identification is shown to be through a mechanism called tying, a term for the connection between a previous speaker's utterance and the subsequent speaker's utterance. This tying can be direct, through the use of exact repetition, thematic wa, addressee terms, etc., or indirect, through the use of paraphrase repetition, parallelism, inclusive 'and', etc.
Direct tying is found among speakers intimate with each other and indirect tying in other situations. The distinction of indirect tying strategies from direct ones lies in the avoidance of identifying the speaker being disagreed with. Direct tying strategies focus exclusively on an explicitly identified point to be argued and possibly explicit speaker identification. Two data sets illustrating indirect tying strategies show a further distinction: sharing inferential knowledge allows for the formation of a contrast without identifiable tying to the argumentative point but relying on a parallel structure of the sequential turn order in one group; in the other group, of American and Japanese businessmen, contrast is formed by barely recognizable paraphrase repetition of the argumentative points within a turn. This approach to analyzing disagreement sheds light on the systematic strategies employed in each situation, where speakers manage to deal with disagreement structural constraints while satisfying other specific sociocultural and interactional constraints.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



ACQUISITION OF DATIVE ALTERNATION IN ENGLISH BY SECOND LANGUAGE
LEARNERS (KOREAN)

Author: LEE, DONG-HAN
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: BRUCE PEARSON

Source: Volume 5803A of Dissertations Abstracts International.
Page 845 . 216 pages
Descriptors: LANGUAGE, LINGUISTICS ; LANGUAGE, MODERN
Descriptor Codes: 0290 ; 0291

In this dissertation, the effect of learning principles, the effect of first language influence, and developmental sequence of dative verbs based on semantic roles were investigated in the acquisition of English dative structures by Korean learners. Data on previous studies were reanalyzed to see the operation of learning principles based on developmental stages. The reanalysis of the previous studies showed that there were multidevelopmental sequences in the acquisition of dative structures in the target language as Hawkins 1987 claimed. In addition, three experimental tasks, a sentence construction task, grammaticality judgment test, and production elicitation task, were administered to 94 Korean learners of English at three proficiency levels to investigate the operation of learning principles (the subset principle, generalization, and preemption/loss principle), effect of L1 influence, and developmental sequence of dative verbs. The results for my studies showed that there were three developmental stages affected by the operation of three learning principles. L2 learners' initial developmental stage is introduced by the subset principle. L2 learners' second developmental stage, where learners' wrong generalization about the target grammar leading to overgeneralization errors, is explained by the generalization principle. Learners' overgeneralization errors can be reduced by the application of the preemption/loss principle in developmental stage three. L2 learners' L1 influence in SLA was also investigated. The analysis of the data showed that the influence of L1 was not prominently shown in the initial stage, rather the subset principle triggered by learners' previous knowledge can explain learners' initial acquisition of the unmarked structure in the dative structure of English. The investigation of developmental sequence of the class of dative verbs showed that recipient verbs were preferred with advance in proficiency levels.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



DEVELOPMENT OF FUNCTIONAL CATEGORIES IN CHILD KOREAN (GRAMMAR, LANGUAGE ACQUISITION)

Author: HAN, HO
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: A. RENE SCHMAUDER

Source: Volume 5803A of Dissertations Abstracts International.
Page 844 . 242 pages
Descriptors: LANGUAGE, LINGUISTICS ; PSYCHOLOGY, DEVELOPMENTAL
Descriptor Codes: 0290 ; 0633; 0620

This dissertation treats the development of functional categories in early child grammar. Based on early child English data which lack functional elements such as verb inflections, complementizers, and (in) definite articles, Radford (1990) among others claimed that functional categories do not exist in early child language prior age two but rather mature later in parallel. Early child utterances thus have the structure of small clauses without IP, CP, and DP. This is called the Small Clause Hypothesis (SCH). In this dissertation, I refute the SCH and argue that early child grammar does not lack functional categories.
After I discuss the pros and cons of the SCH in previous studies, I present child Korean data, which indicate that verb inflections occur very early, while CP and DP elements occur late. I argue that the early acquisition of verb inflections in Korean is rooted in their morphosyntactic nature: verb stems cannot stand alone. Korean children are sensitive to this morphological well-formedness condition. I further argue that the late acquisition of CP and DP elements is not due to the absence of CP and DP in early child Korean but rather is mainly due to their complex semantic/pragmatic nature. It can be thus conjectured that the developmental order of IP and CP/DP elements is governed by the grammatical properties of those elements.
To see if my arguments can be generalized, I investigate the acquisition of other agglutinative languages including Turkish, Hungarian, and Finnish, and show that inflectional elements are produced early by children learning these languages, who must be sensitive to morphological well-formedness.
The SCH adopts the Maturation Hypothesis (MH) and assumes that functional categories mature like biological organisms. I refute the MH, and argue in support of a version of the Continuity Hypothesis (CH) which states that functional categories are available from the outset, although they are not activated until children's cognitive capacity develops well enough to process functional elements. This argument for the CH is supported by the acquisition data in this dissertation and by the result of my experiment in the acquisition of unaccusatives in English.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



UNDERDETERMINED BINDING: AN HPSG BINDING THEORY AND EXPERIMENTAL STUDY OF ADULT JAPANESE LEARNERS OF ENGLISH AS A SECOND LANGUAGE (HEAD DRIVEN PHRASE STRUCTURE GRAMMAR)

Author: HAMILTON, ROBERT LEE
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Co-directors: STANLEY DUBINSKY ; A. RENE SCHMAUDER

Source: Volume 5803A of Dissertations Abstracts International.
Page 844 . 406 pages
Descriptors: LANGUAGE, LINGUISTICS ; EDUCATION, LANGUAGE AND LITERATURE ; PSYCHOLOGY, DEVELOPMENTAL
Descriptor Codes: 0290 ; 0279; 0620

In this dissertation I accomplish two main objectives. First, I develop a theory of anaphoric and personal-pronominal binding within the framework of Head-Driven Phrase Structure Grammar (HPSG) that is empirically more satisfying and internally more coherent than previous accounts, including among others the standard HPSG binding theory of Pollard & Sag (1992, 1994) and the reflexive-marking account of Reinhart & Reuland (1991, 1993). Second, with this theory of binding in view I report on a series of experiments with adult Japanese-speaking learners of English (and native English-speaking controls) designed to investigate whether adult second language (L2) learners are sensitive to the domain of Binding Condition A in English. In the main experiment, 85 L2 subjects and 85 native controls were administered two tasks, a written sentence-completion task for screening purposes and a written truth-value judgment task for measuring subjects' acceptance of nonlocal binding in English. The L2 learners, like the native controls, accepted significantly more
nonlocal binding of English reflexive anaphors when the anaphor was in a structural position exempt from Condition A (e.g. the object of certain picture-NPs) than when in a position subject to Condition A (e.g. a direct object), as determined by a logistic regression analysis of the data (z = 3.96, p $<$ 0.0001 for L2 Group; z = 6.70, p $<$ 0.0001 for Control Group). I argue that this outcome constitutes an underdetermined binding pattern for the L2 learners given certain features of the L2 English input and of the learners' native Japanese. This in turn suggests that these adult L2 learners had direct access to Binding Condition A, hence to Universal Grammar.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



(IN)DEFINITENESS AND MONOVALENT VERBS IN SPANISH (INDEFINITENESS)

Author: GRIFFIN, ELAINE HUNTER
Degree: PH.D.
Year: 1997
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: STANLEY DUBINSKY

Source: Volume 5803A of Dissertations Abstracts International.
Page 843 . 359 pages
Descriptors: LANGUAGE, LINGUISTICS ; LANGUAGE, MODERN
Descriptor Codes: 0290 ; 0291

This dissertation explores the cross-linguistic implications of syntactic and semantic relationships between nonovalent verbs and their (in)definite arguments in Spanish. Although unergative and unaccusative verbs in French and Italian are overtly set apart as separate intransitive classes by their auxiliaries and by the Definiteness Restriction (DR) imposed on postverbal arguments of unaccusatives, Spanish lacks both types of differentiation. Several syntactic tests are administered to a corpus of Spanish intransitives in order to identify a set of unaccusative verbs. Because these diagnostics point to the underlying object status of unaccusative arguments, unergative verbs invariably fail the tests.
Belletti's (1988) claim that unaccusatives can assign only inherent partitive Case to their arguments is rejected, because it cannot account for Spanish unaccusative verbs, which do not impose a DR on their postverbal arguments. An expletive-associate relationship with cross-linguistic variation of the features shared between expletives and arguments can better explain agreement phenomena and the distribution of (in)definite arguments of unaccusatives in French, Italian, English, and Spanish.
Existential haber and other stative verbs do not demonstrate the behavior of either unergatives or unaccusatives. It is claimed that haber's arguments are generated in the Spec VP position, but are licensed for inherent accusative Case. The Minimalist Program framework accounts for the V NP word order, as well as the dialectal V NP agreement variant.
The DR on the arguments of haber is attributed to its semantic content. A modification of Heim's (1982) file change semantics model, in conjunction with the type/token distinction, can adequately explain the DR phenomenon.
Finally, the restricted distribution of (-determiner) NPs is analyzed. It is found that the inability of (-determiner) NPs to carry a token index prevents them from appearing in certain positions where they do not meet agentivity requirements.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



VARIATION IN THE ACQUISITION OF MORPHEME TYPES IN THE INTERLANGUAGE OF CHINESE AND JAPANESE LEARNERS OF ENGLISH AS A SECOND LANGUAGE

Author: WEI, LONGXING
Degree: PH.D.
Year: 1996
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Major Professor: CAROL MYERS-SCOTTON

Source: Volume 5707A of Dissertations Abstracts International.
Page 3003 . 317 pages
Descriptors: LANGUAGE, LINGUISTICS ; EDUCATION, LANGUAGE AND LITERATURE
Descriptor Codes: 0290 ; 0279

This dissertation addresses developmental stages in interlanguage (IL) of adult second language acquisition (SLA). The goal of this study is to predict how IL forms are produced in adult SLA and the developmental sequence of IL grammars. It demonstrates that the principles of the Matrix Language Frame (MLF) model of Myers-Scotton (1993, 1996) of describing and explaining intrasentential codeswitching are also at work in IL development. It shows that a 'composite matrix language' structures IL and that the non-target language aspect of the makeup of this composite can be predicted beyond merely referring to 'transfer'.
The general experimental hypotheses tested are: (1) The content vs. system morpheme distinction operates in IL construction: TL content morphemes are acquired before TL system morphemes. (2) The sources of system morphemes decide the sequence of acquisition of the TL lexical structure and affect IL constructions. (3) In building the IL composite, the L1 can only contribute abstract lexical structure at the level of lexical-conceptual structure, predicate-argument structure, or morphological realization patterns.
To test the hypotheses, 17 morphosyntactic categories were selected. The specific hypotheses about morpheme acquisition order were formulated according to the categories.
60 Chinese and Japanese subjects were involved and three developmental stages were identified: pre-basic, basic, and beyond-basic. Interviews designed for data collection were conducted in the form of natural conversation. Comparisons were carried out in the same stage, between the stages with the same L1 background, and between the stages across the two L1 backgrounds.
The statistical test results support the hypotheses: Content morphemes are acquired before system morphemes, and within the morphosyntactic categories, 'indirectly-elected' system morphemes are acquired before 'structurally-assigned' system morphemes. The results also show that the acquisitional differences among the learners in the same stage and across L1 backgrounds are not significant, but those between the stages against the same L1 background are significant. This study concludes that the IL system contains elements from the L1 and the TL, the Matrix Language of the IL system is a composite, which frames the IL surface forms. Both the L1 and the TL function as Embedded Languages and their contributions to the IL are constrained in different ways. This composite more closely approximates the TL as the learner progresses. The ML vs. EL distinction has implications for IL studies, and particularly for the explanation of the variation in IL systems. The content vs. system morpheme distinction and sources of system morphemes provide a new approach to the examination of morpheme accuracy/acquisition order.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



PROSODIC ORGANIZATION AND AUDITORY MEMORY

Author: REEVES, CAROLYN H.
Degree: PH.D.
Year: 1996
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: A. RENE SCHMAUDER

Source: Volume 5707A of Dissertations Abstracts International.
Page 3000 . 241 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290 ; 0633

These studies investigated the hypothesis that a list read aloud with a recurrent stress/intonation pattern imposed on it will be recalled better than a list read in a monotone voice, especially in the recency portion of the serial position curve. They further examined whether the stress/intonation pattern improved recall by making some items more salient than others or by dividing the list into groups. The results from Experiments 1-12, which used two stress patterns (anapestic and dactylic), two list lengths (8 and 9), and three stimulus types (digits; words sampled without replacement within each list, from a pool of nine; and words sampled without replacement, within and across lists, from a pool of 900) in a suffix procedure, strongly supported the grouping hypothesis. Support for the salience hypothesis was at best ambiguous.
Several studies have investigated the effect on list recall of multiple suffixes, suffixes in which one item is repeated several times, but as yet there has been no consensus on whether a multiple suffix attached to a monotone list improves recall at the terminal list position as compared to a single suffix. Experiments 13 and 14 of this study showed that multiple suffixes read with different stress/intonation patterns interacted in a complex but systematic way with the stress-patterned lists, sometimes affecting serial positions throughout the list.
It is suggested that because stress and intonation are closely associated with linguistic functions such as word segmentation, lexical retrieval, and syntactic processing, prosodic factors such as stress and intonation may function to provide perceptual cues to linguistic structure, cues which can facilitate processing and provide an advantage in recall, especially for young language learners.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



JAPANESE/ENGLISH CODESWITCHING: THE STRUCTURE OF CODESWITCHING AS AN UNMARKED CHOICE AND ITS RELATION TO LANGUAGE PROFICIENCY

Author: KITE, YURIKO K.
Degree: PH.D.
Year: 1996
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: CAROL MYERS-SCOTTON

Source: Volume 5707A of Dissertations Abstracts International.
Page 2816 . 230 pages
Descriptors: EDUCATION, BILINGUAL AND MULTICULTURAL ; LANGUAGE, LINGUISTICS; LANGUAGE, MODERN
Descriptor Codes: 0282; 0290 ; 0291

The purpose of this paper is to investigate what is called 'codeswitching as an Unmarked Choice' as explicated in the Markedness Model (MM) (e.g. Myers-Scotton 1983, 1993b). Code-switching (CS) is defined as 'alternations of linguistic varieties within the same conversation' (Myers-Scotton, 1993b:1). For this study, one multilingual speech community was selected based on the conditions set forth by the MM to empirically test the model's predictions about the language choices and types of CS. The languages involved in CS are English and Japanese and the subjects are high school students ages 14 to 19.
CS was found to be one of the choices in these students' linguistic repertoire in this speech community for informal interactions with their peers. This choice correlated strongly with anything to do with friends, rather than setting or topic in the notion of domain. In other words, friends as interlocutors were strong predictors of CS as Unmarked choice.
When the types of CS produced in this community were analyzed, all kinds of CS, namely extra-sentential, inter-sentential and intra-sentential CS were found. This is as expected for this kind of community. To be more specific, the study showed evidence for a relationship between the type of CS and the speakers' language proficiency. Among the two claims made in the CS literature, the current study supported the claim made by the Matrix Frame Language Model (e.g. Myers-Scotton 1993b, 1995) that it takes higher level of language proficiency to produce sentence level, or inter-sentential CS.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



DISCOURSE RHYTHM IN OVERLAPPING UTTERANCES (CODESWITCHING, SOCIOLINGUISTICS)

Author: BARROWS, CATHERINE GOULD
Degree: PH.D.
Year: 1996
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Major Professor: CHARLES GOODWIN

Source: Volume 5707A of Dissertations Abstracts International.
Page 2994 . 364 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290

This dissertation investigates and argues for the existence of the discourse rhythmic frame as separate from individual utterance rhythms (Couper-Kuhlen & Auer 1988). Using a primarily conversation analysis methodology, this research focuses on the nature of discourse rhythm as opposed to individual utterance rhythms and investigates the interactional function of the discourse rhythmic frame and divergence from that frame.
Selected data which were transcribed initially using a conversation analysis approach were examined through the additional linguistic approaches of phonology and sociolinguistics. The nature of the discourse rhythmic frame is investigated through the application of prosodic phonology (Nespor & Vogel 1986) to a set of sequentially related utterances. I construct the metrical grids which indicate the rhythms of these utterances from the prosodic trees. The results of mapping the simplified forms of these final metrical grids to the discourse rhythmic frame indicate that the discourse rhythmic frame does not impose primary and secondary status on the beats encoded in the frame and the discourse rhythmic frame offers a viable account for the phenomenon of simultaneous onset following a gap in the conversation. These findings offer an alternative to the rhythmic chain analysis of Couper-Kuhlen (1993).
In addition, the sociolinguistic theory of code-switching for the purpose of gaining interactional power (Scotton 1988; Myers-Scotton 1993) is used in the investigation of onset non-synchronic overlaps. Both quantitative and qualitative analyses are used. These data indicate that there is a tendency toward rhythmicization at transition-relevance positions (Jefferson 1973; Couper-Kuhlen 1993 inter alia) and provide support for my hypothesis that rhythm shifts in overlapping utterances are self-enhancing moves toward gaining interactional power.
Finally, a conversation analysis approach to these data indicates that the discourse rhythmic frame is crucial in the coordination of the actions in assessment activities (Goodwin & Goodwin 1987) and can account for the failure to correct certain types of speech errors as well as the occurrence of some repairs and/or repair markers in the absence of overt speech errors. In addition, the divergence from the discourse rhythmic frame may be a crucial element in the delineation
of footing shift and reported speech.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



THE SYNTAX OF BANTU DOUBLE OBJECT CONSTRUCTIONS

Author: SIMANGO, SILVESTER RON
Degree: PH.D.
Year: 1995
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: STANLEY DUBINSKY

Source: Volume 5608A of Dissertations Abstracts International.
Page 3108 . 284 pages
Descriptors: LANGUAGE, LINGUISTICS
Descriptor Codes: 0290

The close connection between verbal morphology and argument structure in Bantu languages has been at the center of much linguistic debate in recent years. It has been noted that the presence of certain affixes permits the verb to take more arguments than those licensed by the verb's lexical conceptual structure. Researchers, however, disagree on whether these affixes attach to the verb in the lexicon or in the syntax. In this study it is argued that transitivizing affixes are separate predicates which attach to the verb in the syntax through predicate union. It is argued that each predicate assigns a unique thematic role and determines a specific initial grammatical relation for its argument in the clause. More importantly, the study argues that initial grammatical relations are not determined by the complex predicate, but rather, by the individual predicates in the clause. The study claims that the applicative affix represents unrelated predicates whose differences are masked by surface homophony in languages like Chinsenga and Chichewa. The different types of applicative predicates determine different grammatical relations for their arguments. A critical point made in this study is that the determination of grammatical relations by individual predicates is language-specific rather than universal. That is, although a given predicate assigns the same thematic role universally, it may be associated with different grammatical relations in different languages.
There are some constructions in Bantu dealing with possessor relations in which the verb seems to be 'transitivized' without the mediation of verbal morphology. A predicate union analysis reveals that such constructions involve the union of a verb with a nominal predicate and, more importantly, the analysis shows that there are two types of union predicates in universal grammar: one type is specified to originate in an initial stratum and terminate in a non-final stratum, the other type is specified to originate in a non-initial stratum.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



STRUCTURAL SIMILARITIES BETWEEN SIERRA LEONE KRIO AND TWO WEST AFRICAN ANGLOPHONE PIDGINS: A CASE FOR COMMON ORIGIN

Author: NJEUMA, BERNADETTE JOSSO
Degree: PH.D.
Year: 1995
Corporate Source/Institution: UNIVERSITY OF SOUTH CAROLINA ( 0202 )
Director: BRUCE L. PEARSON

Source: Volume 5608A of Dissertations Abstracts International.
Page 3106 . 181 pages
Descriptors: LANGUAGE, LINGUISTICS ; LITERATURE, AFRICAN
Descriptor Codes: 0290 ; 0316

This work is a comparative analysis of two pidgins and a creole. The representative languages of the study are Cameroon Pidgin (CP), Nigerian Pidgin (NP), and Sierra Leone Krio (SLK). Grammatical features of the two pidgins, jointly referred to as West African Anglophone Pidgins (WAAP), are analyzed in comparison to those of Sierra Leone Krio.
The motivation for this comparison initially arose from questions on whether the terms pidgins versus creole were linguistically distinctive with reference to the languages of this study. The study compares core features of the preverbal complexes of all three languages as well as a number of fundamental grammatical structures and discourse functions, to determine the possible similarities between these languages.
The findings of the study are that both pidgins and the creole reveal remarkable similarities in many areas. These findings in themselves serve as a premise for positing a theory of common origin for the languages. The proposed theory is that all three languages may be related via a Proto Pidgin (PP) which emerged in the West coast of Africa around mid 17th century. Claims to a common origin are based on the economic history of the West African coastal area following the
intensification of European trade activities in this area. These economic factors, in conjunction with a common geographical location and missionary activities, account for the emergence and spread of these three languages. Evidence for this position is presented in another comparison of features of these languages to those of some indigenous West African languages.
The document also recognizes and accounts for the unique position of SLK and its dual affiliation to both the PP and the language of the freed slaves returning to the Sierra Leone settlement in the late 18th to early 19th centuries. The final pages raise the sociological or linguistically null issue that is claimed to underlie the similarities between the languages, drawing the conclusion that the languages are grammatically and functionally parallel.

The dissertation citations and abstracts contained here are published with permission of UMI Company. Further reproduction is prohibited without permission.
Copies of the dissertations may be obtained by addressing your request to UMI Company, 300 North Zeeb Road, Ann Arbor, MI 48106-1346, USA. Telephone (734) 761-4700; Email: info@umi.com ; Web-page: http://www.umi.com .



DIALECT CODESWITCHING AMONG LOWER CLASS SOCIOECONOMIC SPEAKERS IN THE SOUTHERN UNITED STATES: A SOCIOLINGUISTIC STUDY