In Chinese, a large percentage of the vocabulary consists of compound words where exemplars in the same taxonomic category often share the same head noun (Chen & Chen, 2006; Chow, McBride-Chang, Cheung, & Chow, 2008). This structural characteristic may facilitate an early understanding of the noun taxonomy. The current study aims to investigate taxonomic development in a group of Mandarin-English speaking children in the United States. A contrast association task (i.e., "A dog is not a ____") and a category association task (i.e., "A dog is a kind of ____") were used to elicit responses from different levels of the taxonomic hierarchy (e.g., coordinates, superordinates). Participants were 25 bilingual children aged 3 to 8 and 25 English monolingual age matches. It was predicted that the bilingual group would produce more task-specific taxonomic responses (i.e., coordinates in contrast association; superordinates in category association) than their monolingual counterparts. The results, however, were somewhat opposite to this prediction. Monolinguals were found to, in general, perform better in the category association tasks and the two groups performed similarly in the contrast association task. When English vocabulary size was taken into consideration, there was no statistically significant difference between the monolingual and bilingual children on the category association task. Factors which possibly explain such a difference between the two language groups in the two tasks, or the absence of a significant difference when vocabulary size was incorporated as a covariate are discussed. / text

Thesis advisor: C. Patrick Proctor / Traditionally strict language separation policies in dual language bilingual education (DLBE) programs reflect parallel monolingualism and have been criticized as failing to recognize the sociolinguistic realities of bilingual students (García & Lin, 2017). To leverage bilingual learners’ full linguistic repertoires as resources, this study explored how Sánchez, García, and Solorza’s (2018) translanguaging allocation policy could be strategically and purposefully designed in a third grade Mandarin/English DLBE classroom where the majority of the students were English-dominant speakers. Taking the form of participatory design research (Bang & Vossoughi, 2016), I (as a researcher) and a Mandarin teacher worked together to co-design translanguaging documentation, translanguaging rings, and translanguaging transformation spaces across different content areas – Chinese Language Arts, Science, and Social Studies. During the process, we also engaged in equitable forms of dialogue and listening to openly discuss, negotiate, and develop our translanguaging co-stance in iterative ways. Data collection included classroom and design meeting recordings, observational field notes, and teacher and students’ artifacts and interviews throughout the school year of 2018-19. Inductive and deductive coding were adopted for data analysis. Findings revealed that translanguaging pedagogies took many shapes based on contextual factors, such as the different pedagogical purposes and curricular demands across content areas. Students were able to develop deeper content understandings, build cross-linguistic connections, and develop their bi/multilingual identities and critical consciousness in those flexible bilingual spaces. Findings also demonstrated that the ideological (re)negotiation between the researcher and the teacher was a bumpy and discursive journey, replete with tensions, confusions, and difficult conversations. Overall, it was a balancing act to create translanguaging spaces while maintaining the language-minoritized (Mandarin) space and privileging students’ use of Mandarin given the societal dominance of English. This study provides implications for new theoretical and pedagogical understandings of translanguaging, and suggests that researcher-teacher collaboration provides a promising way to generate evidence-based, practitioner-informed, and context-appropriate knowledge for DLBE curricular and pedagogical improvements. / Thesis (PhD) — Boston College, 2020. / Submitted to: Boston College. Lynch School of Education. / Discipline: Teacher Education, Special Education, Curriculum and Instruction.

L’alternance codique (Code-switching, CS) est l’un des comportements naturels les plus courants chez les bilingues. Les linguistes ont exploré les contraintes derrière l’alternance codique (CS) pour expliquer ce comportement. Au cours des dernières décennies, la recherche a plutôt été axée sur les contraintes syntaxiques et ce n’est que récemment que les contraintes prosodiques ont commencé à attirer l’attention des linguistes. Puisque la paire de langues choisie est moins étudiée dans le domaine de recherche sur la CS, les études sur la CS mandarin-anglais sont limitées en ce qui concerne les deux contraintes. Ainsi, cette étude explore à la fois les contraintes prosodiques et les schémas syntaxiques de cette paire de langues grâce à une base de données naturelle sur l’alternance codique. Prosodiquement, l’étude applique une approche fondée sur l’information (information-based approach) et utilise une unité fondamentale, l’unité d’intonation (Intonation Unit, IU), pour mener l’analyse. Le résultat de 10,6 % d’IU bilingue (BIU) se révèle fiable et offre des preuves solides que l’alternance codique a tendance à avoir lieu aux frontières de l’IU chez les bilingues. Les résultats soutiennent le travail précurseur de Shenk (2006) à partir d’une paire de langues inexplorée (mandarin-anglais). De plus, cette étude développe des solutions au problème de subjectivité et au problème d’adéquation de la base de données afin de renforcer la fiabilité des résultats. D’un point de vue syntaxique, l’étude examine les schémas syntaxiques aux points de CS de la paire de langues mandarin-anglais en utilisant des données recueillies auprès d’une communauté bilingue rarement étudiée. Un schéma syntaxique spécifique à cette paire de langues a été observé en fonction des résultats, mais l’étude suggère que ce schéma ait perturbé les résultats finaux. L’étude comporte une analyse avec les résultats de l’aspect prosodique et de l’aspect syntaxique. Lorsque les résultats divergents sont éliminés, on peut observer un résultat plus solide qui soutient davantage l’argument de la contrainte prosodique. / Code-switching (CS) is one of the most common natural behaviors among bilinguals. Linguists have been exploring the constraints behind CS to explain this behaviour, and while syntactic constraints have been the focus for decades, prosodic constraints were only studied more in depth recently. As a less common language pair in CS research, studies on Mandarin-English CS are limited for both constraints. Thus, this study explores the prosodic constraints and syntactic patterns of this language pair with a natural CS database. Prosodically, this study applies the information-based approach and its fundamental unit, Intonation Unit (IU), to conduct the analysis. The result of 10.6% bilingual IU (BIU) proves to be reliable and offers solid evidence that bilinguals tend to code-switch at IU boundaries. This supports the pioneer work of Shenk (2006) from the unexplored Mandarin-English language pair. In addition to this, the study develops solutions to deal with the subjectivity problem and the database appropriateness problem in this approach to strengthen the validity of the results. Syntactically, this study investigates the syntactic patterns at switching points on the Mandarin-English language pair using data collected from a rarely investigated bilingual community. Based on the results, a syntactic pattern specific to this language pair was observed and this study suggests it disrupted the final results. This study conducts an analysis with the results of both the prosodic aspect and the syntactic aspect. When the interfering results are eliminated, a more solid outcome can be observed which provides greater support to the prosodic constraint argument.

Research examining bilinguals who stutter (BWS) is limited; in particular there are few studies that have considered examining features of speech motor control in BWS. The present study was designed to examine features of speech motor control in bilingual speakers of Mandarin and English. Speech motor control was examined through the acoustic analysis of speaking rate, voice onset time (VOT) and stuttering adaptation. Participants ranged from age between 9 and 27 years. Upon completion of a language dominance questionnaire, two BWS participants were found to be English dominant and three were Mandarin dominant. Each BWS participant was matched to age/sex matched control participants (BWNS). Results for the BWS participants found more stuttering in the less dominant language based on a measure of percentage of syllables stuttered. All of the BWS participants demonstrated stuttering adaptation and there was no significant difference in the amount of adaptation for Mandarin and English. There was no difference found between BWS and BWNS for speaking rate and VOT. In spite of the similarity between BWS and BWNS, speaking rate in Mandarin appeared to be faster compared to English. These findings suggest that speech motor control in BWS and BWNS are similar and current application of these findings to the clinical setting is discussed.