Codeswitching in Beginning Chinese Class

Material Information

Codeswitching in Beginning Chinese Class an Exploratory Investigation of Teachers' Bilingual Practices in an American University Setting
Bao, Yanmin
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (117 p.)

Thesis/Dissertation Information

Master's ( M.A.)
Degree Grantor:
University of Florida
Degree Disciplines:
Committee Chair:
Blondeau, Helene
Committee Members:
Boxer, Diana
Graduation Date:


Subjects / Keywords:
Bilingual teachers ( jstor )
Classrooms ( jstor )
English teachers ( jstor )
Foreign language learning ( jstor )
Language ( jstor )
Language teachers ( jstor )
Phrases ( jstor )
Students ( jstor )
Teachers ( jstor )
Teaching methods ( jstor )
Linguistics -- Dissertations, Academic -- UF
classroom -- code-switching -- function -- language-attitude -- language-teacher -- nationality -- negotiation -- strategy
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Linguistics thesis, M.A.


This paper presents results of a case study of teachers' bilingual practices in an American university setting; it describes and compares the codeswitching practice occurring in three bilingual instructors' speech, and also analyzes and interprets the functions of their codeswitching served in language classroom as well as the potential reasons that caused their distinctive codeswitching behaviors. Though classroom codeswitching is already not a new topic, there are few studies describing the codeswitching behaviors of Chinese language teachers in English-speaking environments, and even fewer have investigated to what extent the codeswitching strategies employed by different bilingual teachers are similar and different. Therefore, this study tries to shed light on the linguistic and pedagogic practice of bilingual teachers of Chinese through both quantitative and qualitative analysis. The data consist of nine audio-recorded beginning Chinese lessons, three class periods of each instructor, and each lasting for fifty minutes. Through audio-recordings, class-observations and field notes, even though there is a Mandarin-only requirement on the syllabus, codeswitching is still inevitable and necessary in language teaching. The three bilingual teachers all participate in the similar codeswitching functions: to accommodate students' learning needs, manage the classroom effectively and also to strengthen teacher-student relationship. In addition, by comparing the occurrence percentage of codeswitching usage as well as different codeswitching types among these teachers, it is noteworthy to find that, though they differ from each other in the percentage of classroom language, the non-native speaker teacher use more Mandarin than the two native Chinese, they all resemble each other in the codeswitching patterns and strategies. Reasons to account for their distinctive codeswitching behaviors seem to have resulted from the influence of their past language learning experiences and also their own language attitudes toward teaching. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (M.A.)--University of Florida, 2011.
Adviser: Blondeau, Helene.
Statement of Responsibility:
by Yanmin Bao.

Record Information

Source Institution:
Rights Management:
Copyright Bao, Yanmin. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
816382453 ( OCLC )
LD1780 2011 ( lcc )


This item has the following downloads:

Full Text




2 2011 Yanmin B ao


3 ACKNOWLEDGMENTS First and for emost I want to express my sincere st gratitude to my M.A. committee chair, Dr. Blondeau whose g uidance, encou ragement and patience from the initial to the final level enabled me to complete this thesis Every meeting with her to discuss my thesis pro gress the joy and enthusiasm she has for sociolinguistic research was really contagious and motivate d me to develop my research interest from an abstract idea to a practical stu dy. I am especially thankful for he r valuable suggestions to my research method ology and my qualitative analysis on classroom codeswitching functions helped me improve these sect ions a lot. I am also heartily thankful to my M.A. committee member, Dr. B oxer whose encouragement and insightful comments did provide me with excellent ideas for the thesis revision work like the suggestion of using Chi square to test the significance o f my qua ntitative result, to make my findings more convincing. My sincere thanks also go to the faculty in linguistic department at UF, for providing me with excellent linguistic study atmosphere and helping me to develop my background in this field. I o ffer my sincere gratitude to Jenna Nichols as well, for help ing me proof read my thesis. And thank Dr. Blondeau again, to introduce me this excellent proof reader. She is really a nice and kind person; and she is excellent in writing as well I also want to thank the staff at the UF libraries for their keen research assistance, the participants in my study, for their open and active participation. I am grateful for time spent with my friends, because of them, m y time at UF was enjoyable Lastly, I would li ke to thank my family for all their love and encouragement. My parents have supported me through out my life; t hey are always s upporting me with their


4 best wishes, and stood by me through the good and the bad times, which motivated me to complete my study


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 2 LITERATURE REVIEW ................................ ................................ .......................... 14 A Controversial Issue in the Teaching Field ................................ ............................ 14 Literature Supporting the Monolingual Principle ................................ ..................... 16 The Ine vitability of Codeswitching in Language Classrooms ................................ .. 18 Classroom Negotiation ................................ ................................ ............................ 22 m Interaction ............... 25 Constructing and Transmitting Knowledge ................................ ....................... 26 Managing the Classroom ................................ ................................ .................. 27 Establishing or Maintaining Solidarity between Teachers and Students .......... 28 Types of Codeswitching ................................ ................................ .......................... 30 Conclusion ................................ ................................ ................................ .............. 32 3 METHODOLOGY ................................ ................................ ................................ ... 34 Participants ................................ ................................ ................................ ............. 34 Students ................................ ................................ ................................ .................. 36 Data Collection ................................ ................................ ................................ ....... 37 Data Transcription ................................ ................................ ................................ .. 38 Identifying the Codeswitching Contexts ................................ ................................ .. 40 4 QUANTITATIVE ANALYSIS ................................ ................................ ................... 45 Data Analysis ................................ ................................ ................................ .......... 45 Results ................................ ................................ ................................ .................... 50 5 QUALITATIVE ANALYSIS ................................ ................................ ...................... 58 Analysis of the Functions of Codeswitching ................................ ............................ 58 Constructing and Transmitting Knowledge ................................ ....................... 59 Translation ................................ ................................ ................................ 59 Highlighting ................................ ................................ ................................ 61 Explanation ................................ ................................ ................................ 67


6 Managing the Classroom ................................ ................................ .................. 70 Clarification ................................ ................................ ................................ 70 Efficiency ................................ ................................ ................................ .... 74 Establishing or Maintain Solidarity between Teachers and Students ............... 77 Praise ................................ ................................ ................................ ......... 78 Commenting and assessing performance ................................ .................. 80 General Conclusion on Functions ................................ ................................ .... 84 Differences among Individual Distributions of Codeswitching ................................ 85 General Conclusion ................................ ................................ ................................ 96 6 CONCLUSION ................................ ................................ ................................ ........ 99 APPENDIX A UFIRB #2010 U 1087 INFORMED CONSENT ................................ .................... 104 B DATA TABLES ................................ ................................ ................................ ..... 106 C CHI SQUARE RESULTS ................................ ................................ ...................... 112 LIST OF REFERENCES ................................ ................................ ............................. 114 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 117


7 LIST OF TABLES Table page 3 1 General demographic information and background of participants .................... 44 4 1 Comparison of the frequency distribution of code switching strategies ............. 55 B 1 st class ................................ ................................ ................................ 107 B 2 nd class ................................ ................................ ................................ 108 B 3 rd class ................................ ................................ ................................ 108 B 4 st class ................................ ................................ ................................ 109 B 5 nd class ................................ ................................ ............................... 109 B 6 rd class ................................ ................................ ............................... 110 B 7 st class ................................ ................................ .............................. 110 B 8 nd class ................................ ................................ ............................. 111 B 9 rd class ................................ ................................ ............................. 111 C 1 Difference of codeswitching usage ................................ ................................ .. 112 C 2 Difference of three codeswitching types usage ................................ ............... 112 C 3 Difference of four levels codeswitching usage ................................ ................. 112 C 4 Difference between Sun and Emma ................................ ................................ 112 C 5 Difference between Sun and Zhao ................................ ................................ .. 113 C 6 Differe nce between Zhao and Emma ................................ .............................. 113


8 LIST OF FIGURES Figure page 4 1 The average percentage of switching to English of each instructor in their classroom tea ching ................................ ................................ ............................. 56 4 2 The average frequency percentage of three types of codeswitching of each instructor in their classroom teaching ................................ ................................ 56 4 3 Th e average frequency percentage of switching to English occurs at different levels: word, phrase, clause and sentence ................................ ......................... 57


9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requi rements for the Degree of Master of Arts CODESWITCHING IN BEGINNING CHINESE CLASS: AN EXPLORATORY UNIVERSITY SETTING By Yanmin Bao December 2011 Chair: Helen e Blondeau Major: Linguistics American university setting; it describes and compares the codeswitching practice analyzes and interprets the functions of their codeswitching served in language classroom as well as the potential reasons that caused their distinctive codeswitching behaviors. Though classroom codeswitching is already not a new topic, there are few studi es describ ing the codeswitching behaviors of Chinese language teachers in English speaking environments, and even fewer have investigated to what extent the codeswitching strategies employed by different bilingual teachers are similar and different. Theref ore, this study tries to shed light on the linguistic and pedagogic practice of bilingual teach ers of Chinese through both quantitative and qualitative analysis. The data consist of nine audio recorded beginning Chinese lessons, three class periods of each instructor, and each lasting for fifty minutes. Through audio recordings, class observations and field notes, even though there is a Mandarin only requirement on the syllabus, codeswitching is still inevitable and necessary in language teaching. The three


10 bilingual teachers all participate in the similar codeswitching functions: to accommodate teacher student relationship. In addition, by comparing the occurrence percentage o f codeswitching usage as well as different codeswitching types among these teachers, it is noteworthy to find that, though they differ from each other in the percentage of classroom language, the non native speaker teacher use more Mandarin than the two na tive Chinese, they all resemble each other in the codeswitching patterns and strategies. Reasons to account for their distinctive codeswitching behavi ors seem to have resulted from the influence of their past language learning experiences and also their o wn language attitudes toward teaching.


11 CHAPTER 1 INTRODUCTION This research is a case study of three bilingu a Chinese speakers and one American) codeswitching practice in beginning Chinese class at the University of Florida (UF) The term of c odeswitching is defined as a switching between two or more languages in a conversation or an utterance In a natural setting, codeswitching behaviors among social group members can also i mply their social identity and relationship as well as some social effects (Rampton, 1995 ; My ers Scotton, 1997; Pujolar, 2000; etc ) For studies of language acquisition and learning codeswitching is used to describe either bilingual speaker behaviors or to describe classroom practices that involve more than one language usage ( Romaine, 1989; Cenoz & Genesee 2001 ; etc ) A he classroom is a pedagogical environment where teache rs and students can also be c onceived as participants in some kind of social interaction Therefore, classroom codeswitching is sensitive to both pedagogical and soc iolinguistic interaction norms it can be used for classroom negotiation and also it ha s some social effects On the s yllabus of beginning Chinese at UF there is a Chinese Speaking Only Policy in Class as one of the main course requirements : You are required to speak only Chinese in class Strictly speaking, Mandarin refers to standard Chinese while Chinese can involve different dialects However, for Chinese classes in an English speaking setting, Mandari Thus no strict difference is made between the two terms in this current study.


12 From the course syllabus, it is clear that a monolingual principle is required. However, what actually goes on in the beginning Chinese classes? Do instructors completely apply this monolingual principle? In fact, through the observation of three class periods of each instructor, codeswitching to English is a co mmon phenomenon during their real teaching practice and the percentage of codeswitching practice used by these three instructors varies from one to another. S ince there are few studies that describe the codeswitching behavior of Chinese language teachers in English speaking environment s and even fewer studies that compare the difference i n codeswitching behaviors among bilingual teachers there is a need for documenting language practices in the classroom As a result this study sets out to document codes witching practices of bilingual teachers who taught beginning Chinese in an American university setting As outlined in the title, this study is, in essence exploratory. It will try to fill in some of those gaps by examining and comparing classroom code switching behaviors among Chinese teachers at UF Through audio recording, class observation and field notes, the teaching practice s of Chinese language teachers are captured and investigated further. Some additional information volition on codeswitching usage is collected through debriefing. The poten tial interpretations accounting for th eir disti nctive codeswitching strategie s will be collected through debriefing discussed in detail through both the quantitative and qualitative analysis. I t is hoped that this study will shed l ight on the linguistic and pedagogic practice of bilingua l teachers of Chinese through the analysis of collected data.


13 As mentioned above, in this study, I intend to describe and compare the codeswitching practice occ urring in these three and then to further analyze and interpret the functions of their codeswitching served in language teaching I will then discuss the p otential reasons that caused the difference in the use of classroom la nguage. Specifically, this study addresses the following research questions: 1. In what situations and for which functions do these three bilingual instructors employ codeswitching in beginning Chinese classes? 2. How do individual language teachers differ in t heir use of codeswitching? 3. Are there any similarities or differences in their codeswitching strategies? 4. To what extent do these teachers use codeswitching to facilitate their Chinese teaching? 5. What are the potential interpretations related to the instructo codeswitching practice s ? This study is organized as follows: i n Chapter 2, the relevant research on codeswitching is presented. Chapter 3 details the research participants and methodology. In Chapter s 4 and 5, collected data are presented, an alyzed and interpreted according to quantitative and qualitative methodologies respectively. Finally, conclusions are drawn in Chapter 6.


14 CHAPTER 2 LITERATURE REVIEW The literature concerning codeswitching in classroom teaching is reviewed in this chapte r. Different from codeswitching in a natural setting, classroom codeswitching aims to ensure the maximum effect of teach ing. However, since teachers and students in a langua ge classroom can also be conceived a s form ing a small social community it is reas onable to infer that classroom codeswitching can have social effects as well. Codeswitching and the monolingual principle are two opposite teaching methods in language classrooms, and the question of which one should be applied has been a co ntroversial i ssue for a long time I will present studies that support the monolingual principle, and then through reviewing studies that advocate appropriate L1 usage, I will demonstrate the inevitability of codeswitching In language classroom, teachers and students negotiate with each other through codeswitching to result in more language learning, and based on related studies, I will identify and present functions and types of codeswitching as observed during their real teaching practice Finally, I will d evelop my own research concerns from a different perspective based on these previous studies. codeswitching practice in an English speaking environment, I mainly choose studies which focus on the reverse: the codeswitching behaviors of bilingual teachers who teach English in Chinese speaking environment. A Controversial I ssue in the Teaching F ield W hether teachers should use native language (L1) while teaching the target language (TL ) has been a controversial issue in the teaching field for many years Some researchers oppose L1 use in a TL classroom which means they support


15 exclusive use of the target language in a monolingual foreign language classroom (Willis, 1981; Chaud ron, 1 988; Dulay Burt, & Krashen 1982; etc ). This mono lingual principle has also won approval by many schools, as illustrated the section about Student Responsibilities in the UF beginning Chinese syllabus you agree to speak Mandar These beginning language and help them attain approximately the Novice High level on the ACTFL/ETS prof iciency scale: students will build up their vocabulary and be able to read, speak, understand and even write basic sentence structures That is, s tudents are expected to carry out Mandarin communication at an elementary level through the maximum exposure to the target language in clas sroom s. On the other hand, some advocate careful an d limited use of L1 through code switching (Lin, 1990; Macaro, 1997; Carless 2008; etc ) As Grosjean (1982, 146) defines utteranc T A ctually the form of a word, a phrase, a sentence or several sentences. A number of studies of codeswitching in TL classrooms have been under taken (Raschka Sercombe, & Huang, 2009; Ching yi 2009; etc ); researchers have argued that teachers should switch practice as w quite common phenomenon of language contact in language classrooms, and


16 especially for those beginners and slow language learners, codeswitching is necessary and inevitable. Literature Supporting the Monolingual P rinciple Many researchers have promoted the monolingual principle of teaching the target language through the target language such as Willis (1981); Dulay, Burt & Krashen (1982); r training course book Teaching English Through English (1981), she advocates using English as much as possible to maximize English exposure so that learners can experience and develop their own L2 system, which is also th e core idea of her book She empha sizes that teachers should teach and learner s need to learn English through the medium of English itself. Jus t like Chaudron (1988) mentions the most efficient way to acquire L2 is through the natural development and exposure to it in meaningfu l, social i nteraction, however, L2 learners seldom have such chances to engage in a natural and authentic L2 environment. T hen it turns to be their teache to create the target language environment for students to practice as much as possible so as Moreover, Kharma and Hajjaj (1989) also strongly support this idea of using L2 exclusively I n their investigation of teacher, student and s the specific use of their native language Arabi c in an English class, the y conclude that L1 should not be used, because the goal of second language teaching is to develop l native competence T herefore teachers need to build up an authentic foreign language atmosphere by teaching English o nly through English for students to learn and practice and finally improve their English ability in such kind of L2 immersion program. Macdonald (1993) also argues that it is unnecessary for teachers to switch to mother tongue to explain what wa s just said because learners can develop


17 their own L2 system if taught entirely through the target language. If teachers keep switching between L1 and L2, it would be in fact very difficult for students to become confident in using the target language, b e cause they would think the reason why their teachers continually use codeswitching is due to their low L2 proficiency, which may impede their language learning initiative. Actually, learners can develop a L2 learning system if they receive the maximum expo sure to the target language: they can gradually grab the meaning through regular repetition of new lexical items and structures, or figure out the meaning by making connection with the related contexts, or e of voice, g esture and so on. Above all, it seems the best way to teach the target language is to establish an authentic and natural target language atmosphere for learners, that is, to make students engage in an language classrooms. These studies mainly focus on how language teachers can do better to help learners develop their L2 system. According to these researchers, teachers need to teach the target language only through this target language so as to create a n authentic atmosphere for learners to practice as much as they can since they lack such opportunities out of class. And finally, learners can acquire TL through this maximum exposure; they can develop their own TL system by experiencing it in a meaningful interaction in language classroom s However, these studies just prescribe or advise which methods language bilingual teachers should employ in classroom, but seldom describe what teachers actually do in their teaching practice. There is no doubt that the monolingual principle does help and


18 between the theory and the rea lity, in real teaching practice. In th is research, I want to document how bilingual teachers respond t o the m onolingual principle. The I nevitability of Codeswitching in Language C lassrooms As mentioned above, in many EFL (English as a Foreign Language) countries or regions, especially in an Asian EFL context, like in Korea, Hong Kong and Taiwan, schools or even M inistries of Education advocate ( TETE policy. But does this mean that all teachers would strictly stick to this monolingual principle in their practice? In fact, the reality is quite different. In a survey among Form 1 r emedial English teachers in 28 secondary school s in Hong Kong, when asked about their total amount of English and Chinese using in EFL class, it wa s reported that only 4.5% of the teachers use En glish as the only medium in practice. 47.8% teachers use more English than Chinese while the rest 47.7% preferred to use half English and half Chinese or mainly Chinese (Ho & Van Naerssen, 1986). From the data, we can see that only a few teachers carried out the TETE policy while most teachers still found it is im possible to avoid using L1 in language classroom. In addition, from an audio recording of an English language lesson of a teacher in a Hong Kong secondary school, Lin (1988) also pre sents a detailed picture of what actually happen ed in the language classro om : English Cantonese (a Chinese dialect which is widely spoken in Hong Kong and Guangdong Province in China) code switching was employed by the teacher as a communicative resource in English language class. Though the teacher successfully start ed her lesso n in English such as reducing to Cantonese when a student gave her a wrong answer for her question. With her e switched to L1 often to


19 if she also realized that her behavior was an obvious violation of TETE which the teacher herself also agreed with in a private recording. For, in the begi nning of this class, either greeting or giving instruction, the teacher only employed English and she also want ed to use TETE to organize the whole class. However, when she noticed the English only principle had caused some misunderstanding or co nfusion for students, she began to be anxious and suddenly switched to Cantonese to make sure every point wa s clear to everyone. Such switching is her immediate re action ny statement about the question whether Cantonese should be used in an EFL class or not, but instead, it describes what the language teacher s actually do in real teaching practice: even if the teacher s also agree on the monolingual principl e, they still ca nnot avoid codeswitching for a number of reasons because the real classroom teaching is more complicated. One straightforward answer to the question why the L1 is used in L2 classroom, as Lin (1988) pr oposes is the target language competence of the teach er or the students, a nd also as perceived by the teacher. For example, if understanding and learning, teachers would use codeswitching strategies instead of strictly conform to the monolingual principle. Ching yi (2009) als o illustrates this similar phenomenon in his study In Taiwan, there is policy advocated by schools and the se in the classroom. For instance, in the study of codeswitching usage in two freshman English classrooms in a uni versity in southern


20 Taiwan, based on in class observations together with field notes and audio recordings, Ching yi noticed that teachers alwa ys switched to Mandarin to bridge some cultural gap when encountering cultural issues. The textbooks used by these freshmen were all monolingual English because t he school insisted the best way to learn English wa s through English, but this seemed to be a little difficult for these beginner level students especiall y when some cultural events were introduced. When the teacher repeated the four times to ask its meaning, she got no response from students. Then if she still insisted on an ish policy, this class would pr obably be in a but receiving no response S ince there wa s no such cultural event in their L1 culture, stude nts could not build their own concrete connection for this abstra ct noun in their L1 culture So the teacher switched to Mandarin to translate it first and then asked some related questions to help students gradually understand what people do in that day. In this way, the teacher accommodate s nd also foster s their English learning. From above, the interaction between teachers and students language classes is of a highly complex nature. In spite of the fact that the monolingual principle has been advocated by most educators and educational poli cy makers, teacher s still cannot totally follow this principle in their real teaching practice. Because being a teacher is quite different from being a teaching machine, one cannot give the teacher an order like using English exclusively and expect them to operate it as a robot. Though there are many researchers who support the monolingual principle, teachers would still need to adjust their language strategies in their real practice For example, when teachers intend to promot interest as well a s their attention in language learning,


21 maintain classroom discipline or shift to a more favorable relationship with students mother tongue In a real language teaching, w hen teachers notice the m onolingual principle may result in ints out have tried to create an EO [English only] classroo m but have found they have failed to get the meaning across, leading to student in T f whether or not the suggests in fact, can only be accomplished by the usage of codeswitching strategies in L2 classrooms. As a result, codeswitching is inevitable and necessary in language classes since immersion is virtually impossible and unrealistic. Swain and Lapkin (2000) also suggest that L1 use can enable students perform a specific classrooms, as Carless (2008) suggests the use of mother tongue has both positive and negative consequences. The positive influence is that learners can use L1 as a starting point when learning a new language as well as a common communicative tool. Macaro (1997) also supports this point in two pedagogical issues peer collaboration and learner autonomy teaching exclusively through the target language is not only impractical but also deprive learners of an important tool for TL mother tongue may u sefully serve social and cognitive functions, including the


22 construction of scaffolded assistance and create through collaborative dialogue the impacts of L1 use, through interv iew s with ten teachers in a Hong Kong secondary school, Carless (2008) fin d s that teachers often feel uncomfortabl e or guilty if students overuse their mother tongue in class b ecause they think they do not fulfill their responsibilities to improve student ability and such overuse of L1 cannot aid new language learning process since they only practice the language they already mast er So what thes e researchers actually support is the careful and limited use of L1 in the form of codeswitching in L2 classrooms. In fact this is a big challenge for teachers they need to know when and where would be appropriate for codeswitching to apply, and where and when it may be pedagogically invalid. Just like Raschka, Sercombe and Huang (200 9) conclu de in their is a lazy rule since teachers do not need to think about the appropriate time and way to switch codes language learning process Classroom Negotiation Form the ab ove studies that describe what actually happened in language classroom s teachers adjust their output through codeswitching to make it comprehensib le to the TL language learners, a nd only when t input is clearly received by learners th e language t eaching can process as planned. That is, between students and teachers, mutual understanding should be achieved to ensure more language learning. In SL A research, such process whereby speakers try to achieve mutual understanding, or to produce correct TL l anguage forms, or to generate additional topic related information to keep a conversation forward is interpreted as


23 negotiation (Long, 1985, 1996; Lyster & Ranta, 1997 ; Van den Branden, 1997). The negotiation interaction between students and teachers can h elp learners receive comprehensible input and produce meaningful output as well; it is thus widely believed that negotiation fac ilitates TL language learning (Long, 1983, 1985, 1996; Pilar, Mayo, & Pica, 2000). During language practice in classrooms, teach ers and students can negotiate with each other through codeswitching, and, as observed in these previous studies ( Lin, 1988; Ching yi, 2009 etc ) codeswitching is inevitable and necessary. Based on previous studies (Lyster & Ranta, 1997; Van den Branden, 1997), there are three types of n egotiations : negotiation of meaning, negotiation of form, and negotiation of content. According to Lyster and Ranta (1997), a conversational negotiation function refers to the meaning negotiation, interlocutor s attempt to a chieve mutual understanding by resolving communication problems, which is also mentioned by Gass (1997); wh ereas a didactic function involves negotiation of form, so as to make learners produce accurate TL forms, which in fact, is more like a corrective fe edback. On the other hand, different from the two types of negotiations that aim to resolve communication or language problems, Van den Branden (1997) mentions a third type of negotiation, in which, interlocutors generate some new topic related information through elicitation to keep a conversation goin g language learner can develop their foreign language proficiency through interaction. W hen learners cannot understand their interlocutors, they may negot iate the meaning to achieve comprehension D uring the second language acquisition, comprehensible linguistic input can be gained through the speech modification and restructur ation (Long, 1985) More specifically,


24 interlocutors use strategies, which serve to avoid conversational trouble, or tac tics, which repair troubles in the discourse or a combine of the two devices to achieve mutual understanding. For example, when addressing non native speakers (NNS) native speakers (NS) would check their comprehensi on, or stress key words, or make new topics salient by using frames such as OK, So, Well and Now Such kind of negotiation between NNS and NS is just like that between teachers and students in a language classroom. Though the social interaction between st udents and tea chers in language classrooms is a little different from interlocutors outside the classroom context, as pointed by Pica (1987), teachers would answer s but they cannot provide a one to one respond rance in a limited time M eanwhile, teacher need to hear from as many students as possible to ensure instruction to proceed as planned As a result, the classroom discourse is designed more like a one wa y flow from student to teacher. Despite the unequal d istribution of participation rights between teacher and students, mutual under standing can still be achieved because both teachers and students request for clarification or confirmation also check comprehension of output. For exa mple, when teachers gave classroom instructions or introduce some new linguistic forms, they would modify their utterance to make it comprehensible language understanding into account and students interrupt to quest ion until they A nd a way to ensure the comprehensibility of utterance is through codeswi t ching. For the teachers, as being language experts as well as evaluators, they need to make their output comprehen sible to the students when negotiating the meaning, and


25 also need to provide corrective feedback to help learners to produce appropriate TL language forms when negotiating the form, and sometimes to provide topic related information to push students to pra ctice TL language more during the negotiation of context as well. For example, when teaching new grammar concept s, teachers can simply switch to L1 to explain the key points to achieve mutual understanding. For the students they can switch to th eir mother tongue to seek feedback from their teachers to see whether their TL output is meaningful. Therefore, the language classroom can be regarded a s a community o f practice, where teachers and students negotiate with each other through codeswi t ching A nd classroom codeswitching that are employed during the negotiation process should fulfill different functions nteraction Since bilingual prac tice, codeswitching used by teachers would be analyzed to investigate what functions they serve in classroom interaction so as to facilitate both language teaching and students learning T his study aims to describe classroom language practice s the func tions of switching occurs rather than directly follow different types of negot iation s In addition, it seems that codeswitching used under different negotiation types may overlap because either for negotiation of form or content, the mutual understanding should be first sustained or learners cannot produce accurate TL form or carry the conversation forward. For example, when using TL to elicit additional topic related inf ormation, teachers need to ensure that learners receive comprehensible input to carry on the


26 conversation so they may highlight some key point In such case codesw itchi ng that functions as highlighting also serves to negotiate the meanin g and the content would be categorized according to the situations where codeswitching occurs. From s ome related articles that describe classroom language practice s I have identified three main functions of Constructing and Transmitting K nowledge Constructing and transmitting knowledge codeswitching se rved in EFL classes. Teachers tend to switch to L1 to explain and emphasize the linguistic forms such as translation (use L1 to explain L2 to ensure According to Ching yi udy (2009) of codeswitching in two freshman English classrooms in a university in southern Taiwan, codeswitching i s used when tea chers and students talk about monolingual textbooks a nd to complete lessons. In class, teachers switch from English to Mandarin to explain the linguistic forms such as English lexical items, phrases, sentences and grammatical rules in order to increase the comprehensibility of the English monolingual text. In particular, teachers use much more Chinese than English when exp laining grammar rules. B ased on their previous te aching experience, teachers fin d that students under stand grammar better when teaching in that using L1 for translation is an inevitable part of second language acquisition. In addition, through classroom observations with recording of classroom interactions, and also detailed interviews with teachers and students in two classes in


27 Taiwan commercial cram schools, Rasc hka, Serco mbe and Huang (2009) noticed that in one class, when a teacher first talked about her experience during the winter vacation like the inability to lose weight in Mandarin, she switched to English to highlight a topic and then continue d in L1 by asking students w hat they had learned in previous classes. The teacher use s the codeswitching to mark a frame shift from her socializing sequence to the lesson itself. S he starts the class by sharing her experiences during th e winter vacation in Mandarin to create a non threatening environment S of class, the teacher tries to lead students to gradually feel more comfortable to use English in class but not push them to use English at th e beginning. Then she maintains this non threatening environment and skillfully moves the classroom discourse on to the next step to continue where they stop last time. In this way, student s would n o t feel too much pressure and could feel more comfortable in class Managing the C lassroom Teachers often switch to L1 when giving students explicit classroom instructions to make things clear to understand especially for some important task requirements. In addition, in pressing situations where there is a need to save time and also to make sure the optimal effect of communication is attained teachers also employ L1 to convey their points clearly in order to avoid confusion or incomprehension among students. l drill s in pairs, she switched her co or t o Cantonese. But since she uses the English comprehension checks all through in class, it is not plausible to say the reason that she encodes these checks in Cantonese is because she thinks her students could not g rasp their meaning The only


28 reason to account for this switching is that the teacher wants to make sure her instruction is totally understood by every student, so t hat the followed pair work can b e smoothly practiced In addition, allowing L1 use in L2 classrooms can be very time efficient in certain situations (Chambers, 1992; Atkinson 1993). As Qian, Tian and Wang (2009) observe and record the occurrence of codeswitching in primary English classr ooms in China they find when speakers are in pressing situations, they may switch to L1 to express their idea for efficiency and so do teachers. A 40 min lesson can be a short amount of time for teachers to complete a variety of tasks. This is part icularl y true when the bell rings but the task still remains half completed or uncompleted I f teachers still use English to negotiate the progression of classroom activities, it would be difficult to regain the ir impatient student ize class should be over. As a result, the teacher in their observation finished the task by switching to L1 and she also spoke much faster than usual. Through this codeswitching, the teacher can regain impatient room activities as expected. Establishing or Maintaining Solidarity between Teachers and S tudents The above two functions both serve to ensure the maximum effect iveness of teaching while this function is intend ed to strengthen the relationship between te achers and students which resembles negotiation of relationship in society T eachers and students in a language classroom can also been conceived as participa nts in some form of socially organized interaction, in other words, classroom is a community of p ractice Specifically, teachers and learners negotiate relationship s and identities in the social and affective classroom environment (Adendorff, 1993; Merritt Cleghorn, Abagi, & Bunyi 1992). According to Myers Scotton (199 7), codeswitching can also have social


29 effects in terms of increasing or decreasing social distance In addition, she also proposes that the motivation of codeswitching is to be more favorable to the listeners (2006) ndship and solidarity such as for praise (teachers use L1 to make sure students would not miss any part of the praise which could be beneficial to student teacher relationship), for encouragement (teachers would apply codeswitching to encourag e students es pecially when faced with difficult tasks and also for weak students), or for comment (since English sounds more distanced, teachers tactfully use L1 to express their evaluation According to Ching yi (2009), when th teachers often adopt Chinese to demonstrate their authority in the classroom since Mandarin has been recognized as an official language in Taiwan for more than a half century. But when engaged in informal conve rsation, teachers also switch to Mandarin for solidarity. More interesting ly one teacher in this study even sometimes use d Taiwanese (a language that is viewed with lower status than Mandarin in Taiwan and rarely used in academic situations) to interact w ith students when walking around the teacher said in the later interview, he thought such kind of switching to Taiwanese could show his sense of humor when interact ing with students which was a quite important guided classroom. Similar as Lin (1988) mentions in her study, when code switching from English to Cantonese, the teacher seems to send a message to her s


30 teachers can shift their role relationship from English speaking teacher to bilingual helper or friend, which is quite beneficial to cla ssroom rapport and teacher student relationship. Fr om above, teachers often us e codeswitching strategies to facilitate their teaching to avoid some potential misunderstanding or incomprehension among students. M eanwhile, they also employ codeswitching to e stablish or maintain solidarity which is quite important to classroom rapport. As a result, codeswitching in language classes, as Raschka, Ser combe and Huang (2009) conclude is not a consequence of insufficient English language competence. In fac t, their use of codeswitching is quite strategic which also i mplies their high level of teaching and communicative skills Teachers need to know where and when a codeswitching should be needed in order to achieve the effects they expect for classroom effi cient management, solidarity maintaining, clear instruction and key learning points etc. That is, codeswitching is necessary and inevitable, teachers can use codeswitching to promote classroom interaction and also ensure efficient classroom management. In addition, a suitable quantity of codeswitching such as to praise or encourag ement can also cultivate strengthen the student teacher relationship. Types of C odeswitching Based on observatio n of many cases of Englis h Spanish codeswitching in natural setting Sankoff and Poplack (1981) identify three types of CS in relation to syntax, namely tag switching, intra switching and inter sentential switching. Though classroom codeswitching is different from that in a natur al setting, codeswitching used in either of the two settings can be regarded as part of bilingual language practice s A nd classroom


31 can also be conc eived as a small social community Therefore types of codeswitching observed in the natural setting can be appl ied for the classroom codeswitching as well As in three types of codeswitching are used when analyzing classroom codeswitching. Tag switching, emblematic switching or extra sentential switching (Muysken, 1995), involves the insertion of a tag or a short fixed phrase in one language into an utterance which is totally in another language. It involves discourse markers or sentence fillers; just like the following example (2 A). E xamples from (2 A) to (2 C) that are used here are all extracted with slight modification from (2009) (2 A) zen me yang ? Student 5: But they eat. T mad, angry. OK, very good story. How about others? In this example, the teacher u sed questioning. According to Qian, Tian and Wang (2009), tag switching tends to be discursive The second type of codeswitching is i nter sentential swi tching which means a switch at a clause or sentence boundary, or among sentences, where each clause or


32 sentence is either i n one language or the other. As shown in the following example (2 B) : Teacher: OK, very good, very nice, feichang bang. xiake yihou ba ta tiezai heiban sangmian, qita tongxue keyi kan [ ou work well Sti c k them (flash cards) on the blackboard after class and other students can have a look. ] In t his example, the teacher shifted her praise from English to Chinese as a single independe nt clause due to the reason that she perceived the praise may be a little diff icult for students to unde r stand in English B y swi tching to L1 even at the short clause level, the teacher passed on her praise efficiently and as seen from the vi de o recording, the students were also very happy to hear this. Mean while, the teacher also switched the instruction to Chinese entirely s o that no confusion would arise On the contrary, the third type i ntra sentential switching refers to the switching with in the clause or sentence boundary a s shown in the following (2 C) : Teacher: nimne huide dongzuo you shenme a y ou march, clap, jump [What action can you do, like march, clap, jump ] In this exam ple, switching to Chinese occurred within the sentence bou ndar y where the teacher inserted Chi nese clause in her demonstration English words in the sentence are the target words to be learned T he teacher switched to Mandarin to ensure her utterance clear ly understood This type of codeswitching involves the greates t syntactic risk since it requires tha t speak er be quite fluent in both languages. Conclusion Above all, m ost researc h on language codeswitching has ranged from describing rget language use to make


33 connecti or different types of codeswitching being observed However, few studies (except like Macaro, 2001; Sato & Kleinsasser, 1999) compare the difference in teaching practice between bilingual language teachers in the use of codeswitching and also the followed pedagogical influence in language classrooms and even fewer consider Mandarin English switching in Chinese learning classes since mos t research have focused on ESL/EFL classrooms. As a result, this present study serves to fill in some of these gaps, first to describe what actually goes on in the Chinese class in an English speaking environment, then to investigate whether there are any similarities or difference s in the use of codeswi tching among language teachers The present study al so describe s and analyze s what situations motivate bilingual teachers to accept c odeswitching in their practice as well as the function s of their codeswit ching in classroom interaction. Finally, this study aims at providing an interpretation of the underlying reasons behind different codeswitching practices of these bilingual teachers in Chinese classes


34 CHAPTER 3 METHODOLOGY In this chapter, I will fir st discuss the demographic information and gen eral linguistic background of my participants, and then describe the data collection procedures, data encoding principles and methods used for data analysis. Participants Three instructo rs of beginning Chinese class at the University of Florida participated in this study. Table 3 1 shows some of their social characteristics ( at the end of this chapter) In addition, for the general linguistic background of these three instructors, they can all speak both Mandari n and English quite fluently, not only because their rich teaching experience or high educational background, but also their great exposure to both languages: Sun and Zhao have been in US for almost 30 years, and Emma also lived in China for a bout 11 year s She lived in Taiwan from the age of 6 weeks to 7 years and then again she studied Mandarin there for like around two and a half year. Later, s he also lived in Northeast China for about one year From the above, we can have a general idea about the demog raphic information as well as the linguistic background of these three instructors. In order to keep the identities confidential as the Institutional Review Board (IRB) of UF required, all the names of these instructors occurred in this study are pseudonyms. In addition, in the Age column, I also only list their general age range instead of explicit age, b ecause both in Asian or Western culture, to ask a woman for her explicit age is often regarded as being rude and impolite. I will also expla in later that the age factor here is not that important based on the general background of these three instructors.


35 From T able 3 1, it seems that the most noticeable different characteristic among these three instructors is their nationality, two native Ch inese speaking teachers and one non native speaker T eachers have different cultural background s also means they may have different educational background s and experiences when b eing a foreign language learner. As a result, some distinctive beliefs formed during their language learning process would have gradually developed as their own teaching philosophy, which may later influence their real teaching practice. However, in this current study, since the sample is not large enough, only two native Chinese s peakers vs. one non native, it is not that plausible to make a generalization based on this nationality factor. classroom codeswitching behaviors since their past language learning experience may more or less influence their current teaching practice. To account for this point o f view, I will explain in further detail in Chapter 5. Other factors listed in the table may not require further analysis since there are no great difference s Since the participants are all females, the gender factor will be excluded when doing data analysis. Then for the factor of age, though generally they are all 50+, specifically they are not completely in the same age range with the m aximum age difference being about 10 years. However, since these three instruct ors all have rich experience teaching Chinese for about 10 years each, they all received high er education degrees of either a M aster or a PhD, and their social network are als o quite similar because they share the same working environment in UF and live in the same city, th e factor age seems does not result in any great difference in their scope of knowledge, teaching or even social experience. This is u nlike people at 20 and


36 5 0 years old who could differ in many ways like in different social experience or networks, which would greatly influence their thinking and behaviors. For example, people at 20 years old are generally college students, even if they have been w orking, their social experience and networks could not be the same as people of 50 years old Just like a college student and his/her parent, they cannot behave the same when dealing with everything ; they may have different opinions even toward the same thing. Students s codeswitching, to see what language st rategies teachers will employ in response In because language interaction b etween teachers and students is inevitable and necessary in language classes. As a result, even though t he utterance s of these fifty si x students was also audio recorded just like these three instructors in be ginning Chinese classes, ultimately English switching will be further analyzed since the primary focus of this present study is the codeswitching behavior s of bilingual teachers. I will use different numbers like Student 1and Student 2 to indicate different students once their utterance s are adopted in codeswitching occurs during the following data analysis. In addition, t hese students are from different sections of beginning Chinese classes, that is, t hey all demonstrate their language proficiency at the beginning level. As a result, different codeswitching behaviors among these three instructors cannot be considered as di fferent responses to different language proficiency level students. Therefore, for this comparative study, factors like gender, age, educational background


37 and teaching experience as well as the language proficie ncy level of their students are controlled. The main difference in this study focuses on so far is native vs. non native speaker teachers, but the sample is not large enough to make generalizations based on this factor. Data Collection All participants and students in this present study were first a sked to sign a copy of informed consent f orm approved by the UF IRB ( Appendix A). After getting all approval of these three participants and fifty six students for this study, audio recordings, class observ ations and field notes were carried out to collect data. Three Chinese class periods of each instructor, each l asting for 50 minutes, were selected for audio recording. I sat at the back of the classroom and also made some comprehensive field notes like the language used by teachers and students, activiti es description, verbal and non verbal interaction and so on. Meanwhile, a small digital voice recorder was placed s were clear for later transcription. is always a concern. On one hand, sociolinguists hope to collect natural and spontaneous speech data just like people talk when they are not being observed. On the other hand, the use of voice recorder would inevitably cause participants to monitor the ir speech. In this study, I used a small recorder to make the overall situation as natural as possible. Teachers often ignored it gradually and conducted classes as usual since the recorder was so small and was placed at the corner of the platform. Furthe authentic language data in classrooms as usual; neither the teachers nor the students


38 were informed on the real purpose of this study. The title they saw in the copies of An exploratory investigation of language prac tic e s in beginning Chinese class in was quite general so that the participants could not know which specific point my study really focuses on. Teachers were requested to conduct the l esson as usual but they were told the original aim afterwards since this study mainly focuses on their behavior of Mandarin English switching. After data collection, I also had some casual conversation s with all three instructors to delve more deeply into their opinions about their codeswitching strategies in teaching practi ce and some other related issues, such as their past language learning experiences. However, from the data collected in language classrooms, I can only describe what actually happened i n a classroom setting behaviors but not the teacher think ing process when switching codes T here may be some other subjective factors affect ing t such as their own language attitude s However, since the subjective factors cannot be collected through either classroom observation or audio recording debriefing is thus used as an additional data collection to gather some potential interpretations codeswitching practi ce s Data Transcripti on This present study mainly compares the codeswitching behaviors among different bilingual teachers, so I only pay great attention to the point within the interaction episode where Mandarin English switching is taking place and transcribe them to get an e stimate of the distribution of the two languages. That is, the sentences where the Mandarin English sw itching occur are extracted, and the preced ing and following sentences are also extracted, when necessary, to provide a context to further analyze


39 the cod eswitching. These extracted recordings are transcribed precisely in Word documents and according to Standard English and Chinese spelling (pinyin) Since the speech samples aim to help beginning level stu d ents to learn Chinese, they are structu rally less c omplex and they are all transcribed and double checked by myself. I follow the transcription convention as Thiba ult and Vincent (1990) suggest, in that transcription should reflex exactly what the participants actually said. E ( if com pared with pre scriptive grammar) are a lso transcribed, and nothing is modified as written language. For instan ce, I use an ellipsis (three spaced dots) to indicate a short pause occurred during the utterance since the disfluency is the most remarkable feature of spoken ghter or some gestures they use in brackets because these body language phenomena also play a big role in language learning and teaching. An excerpt of a transcription is provided below (3 A) from one of these three teachers, Sun, and the translation part is in square brackets: (3 A) Sun: Dazhe, jiushi discount, dazhe. ni hui kandao dajiuzhe, dajiuzhe discount? Ten percent off, ten percent off! danshine, shi ninety percent of the o riginal price. Zaizhongguo, kandao dajiuzhe, yiweishi ninety percent off, ni haogaoxing, maile (made a surprise facial expression) [Discount, which means discount, discount. You may see ten perc ent off, ten ninety percent of the original price. In China, when you see ten percent off, you think it


40 means ninety percent off, you are so happy, and buy a lot of things. When it comes to Students: (laugh) Sun: Shi ninety percent of the original price, bushi ninety percent off. S tudent 1: Why do they use jiu instead o f jiu shi ? [Why do they use nine instead of nine ty ?] Sun: Wo xiang zhongguoren shiyong jiu [I think that Chinese prefer to use nine.] Identify ing the Codeswitching Contexts Because the purpose of this study is to analyze the codeswitching behaviors of thes e three bilingual Chinese instructors, I need ed to circumscribe the context where Chines e English switching occurs. I exclude d the contexts that did not contain the c odeswitching When these instructors produced clause or sentence totally in English during the bilin gual teachers selected the langua ge English embedded in the TL language Chinese during the same speech in their teaching practice. The alternation between the two l anguages is unambiguous. For example: (3 B) Sun: Zuotian Na, jintian hekuaide shuoyixia, zhuyao shi fanyi. Did you bring your homework? Nimen you ma ? with you. So, today we will quickly review it, mainly the translation part. Did you bring your homework? Did you bring?]


41 However, at the word or phrase level, the status is ambiguous since it could be either codeswitching or borrowing. In Chinese, borrowi ng at the phrase level is rare, and in this study, these three bilingual teachers just switched from Chinese to English at the phrase level to highlight or translate some Chinese phrases to English to arouse (3 C) Sun: Z hege women gangcai kanguo le, s hi kafeise. Jia yige se, make it clear. Suoyi ni byhui get confused with kafei, ni hede drink.] The instructor Sun switches from efers to Therefore, in the present study, I only distinguish codeswitching at word level from borrowing words. Lexical borrowing refers to the use and adaption of an individual lexical item that originates from a donor, or lexifier, language to a recipient language. word from an and Meechan (1998), borro wing words are adapted to the recipient language and thus it can be only applied to the grammar of recipient language wh ile codeswitching involves the grammar s of both languages the donor and the recipient language


42 The following example (3 D) demonstrates the codeswitching occurred at the word level, and the example (3 E) is a borrowing word: (3 D) Zhao: Hao women kan nizenme apologize. u make an apology.] Student2 : Duibuqi. In this e xample, instructor Zhao switches to English at a single word level understanding whether they knew how to make an ap ology in Chinese or not. In fa ct, she can also ask this question completely daoqian rent vocabulary, so she swi tches to the English word (3 E) Zhao: Wo xianzuo feiji, zai dache. Other than zai dache, hai youmeiyou biede? Zai [I first took an airplane, and then cal led a taxi. Other than call a taxi, are there any Student 3 : Zuo chuzuqiche. [Take a taxi.] Zhao: Dui, zai zuo chuzuqiche, dishi, zai dache, doukeyi.


43 [Right, take a taxi, a taxi, call a taxi, both OK.] In this exam ple, Zhao uses recipient language, Chinese. However, because Chinese characters are not designed to represent sound other Chinese words. I n a Chinese word, every Chinese character is a single morpheme with a distinctive meaning, and the meaning of the whole word is derived from the meaning combination of each single morpheme. D), which contains two single characters Theref Anyhow this lexical item been adapted to Chinese phonological and morphological system, which also has the same word category and us age of the Chinese word taxi] That is, it is a borrowing word from English approximate pronunciation which is now applied to Chinese grammar only. In addition, the easiest way to pinpoint whether a word is a borrowing in a given language is to see whether the word is used by a mono lingual of that language or not, because codeswitching is only used by bilingual or multilingual speakers. In Ch ina, if one says bilingual speakers would naturally relate it t o the Chinese


44 Therefore, in this study, I only further analyze the contexts where codeswitching occurred and exclude the borrowing condition. Total sentences used in class and different ty pes of codeswitching are kept for further quantitative analysis as developed in Chapter 4 Based on the transcripts of lessons, conversation analysis will be applied to ide ntify the function of the switching in the language teachi ng process and also for deeper analysis as developed in Chapter 5 That is, both quantitative and qualitative analysis will be done in this present study. In this chapter, I have discussed my partic ip ants, the inev itability o f using utterance s as input, my data collection procedures, the transcription proces s and the definition. Results and findings will be discussed in the following chapters. Table 3 1. General demo graphic information and background of participants Pseudonyms Nationality Gender Age Degree Years of teaching Chinese experience Sun Chinese Female 50+ PhD 10 years Zhao Chinese Female 50+ Master 8 years Emma American Female 50+ Master 15 years


45 CHA PTER 4 QUANTITATIVE ANALYSI S In this section, I intend to use quantitative analysis to address my second and third research questions to test if there are similarities or differences among the three Two r esearch questions will be answer ed here: How d o individual language teachers differ in their use of codeswitching? Are there any similarities or differences in their codeswitching strategies? Data Analysis First, some tables are made based on the lesson plan of e ach single class period ; sentences uttered by instructors are then counted in each step as well as the number of codeswitching occurrence s ( Appendix B) In orde r to compare different codeswitching behaviors among these three instructors, two different ways of counting codeswitching strategies are adopted here: As mentioned in chapter 2, Sankoff and Poplack (1981) identify three types of CS syntactically, that is, tag switching, intra switching and inter sentential switching. Here I use some detailed example s extracted from the data I collected for this present study to illustrate each codeswitching type. Tag switching (4 A): Sun: Zheshi yitiao long pants. Ni keyi shuo changku, right? [This is a long pair of pants. You can say it long pants, right?] In this example, the teacher inserted teachers used Chinese tag marks most of the time. I argue


46 that such k ind of tag switching is not some unconscious behaviors which is different from the tag switching observed (2009) According to their classroom observation, tag switching tends to be discursive, which is caused by teachers arelessness on their language choice In this study, if tag switching also that the occurrence of tag switching and the use of Chinese tag marks might be fifty fifty However, tag switching is much less frequent in this study, the situations wh en teachers use tag switching are often noticed when introducing some new or emphasiz ing important linguistic forms or conducting the whole class through language practice (Appendix B) I n order to get tention to these important points and also to check whether their output is comprehensible to students teachers would switch tag marks in student If teachers use Chinese tag marks, though the purpose of checking comprehension can be achieved, stude aroused since Mandarin is not Inter sentential switching (4 B): Zhao: Hao, qing ba zuoye jiaoshanglai So, everyone already turned in your homework, homework? [Ok, please turn in your h omework] In this example, the instructor Zhao switc hed to English at a sentence boundary. The first sentence is entirely in Chinese while the next following is totally English. Intra sentential switching (4 C): Zhao: Hao, nimen liangge take turn, s houhuoyu an he zenmeshuo ?


47 Students: Guke. [Customer.] e boundary. Therefore, Zh ao makes highlights the students switching requires that speakers should be quite fluent in both languages. More specifically I also categorize codeswitching to English at word, phrase, clause and sentence level s That is, codeswitching to English occurs as a single wor d, a phrase, a clause or a sentence. At word level (4 D): Emma: Haoba, wox si, laikan grammar disi. [OK, I want to explain this p grammar number four.] I n this example, Emma first asked students to look at grammar number four entirel y in Mandarin, then she repeated this instruction but switched the key word to English in ord er to make sur e what she said was clearly understood. In example (4 occurr ed at word level. At phrase level (4 E): Sun: Hao, women mingtian zuo situational responses. [Ok, we will do situational responses.]


48 Here the codeswitching occurs within sentence boundary; instructor Sun directly switched in English instead of its Chinese counte beyond her current vocabulary. By doing this, she successfully make s her utterance comprehensible to learners At clause level (4 F) and (4 G): (4 F) Sun: Gangcai women xu ele yige changduan, chang gen duan put them together, Students: Length. owing example, the clause is not totally in English but with a Chinese character inserted (4 G) Student4 : Ni..xindiannao. Oh, nide xindiannao. Emma: Ni xindiannao. Nide xindiannao, dou keyi. De bushi hen xuyao. [You new computer. Your new computer necessary.] In (4 sentential switching and I categorize it as the switching occurred at clause level, because the English part is


49 neither a single word nor a phrase, and this switching also occurs only within the clause boundary. S imilar to the switching occurring at sente nc e level, as shown in (4 B), the teacher produces the second sentence entirely in English which is also inter sentential switching. The following example (4 H) demonstrates the similar condition as (4 G) with only one difference, the switching occurred w ithin sentence boundary: (4 H) Zhao: Hao,women kaishi shangke. So, everyone already turned in nide homework, homework? homework?] In this example, after signaling the start of the cl ass, the instructor s witched her following sentence almost in English. But the whole sentence is interrupted by a nide [your] ish parts are also neither independent words nor phrases. Therefore, it is inappropriate to ca tegorize this switching at word, phrase or even clause level. As a result, the situation of codeswitching at claus e or sentence level is a little complicated because it could be either inter sentent ial switching ( the clause or sentence is entirely in Engl ish) o r intra sentential switching (most part of the clause or sentence is in English and the English part is neither a single word nor a phrase). This is s imilar to the switching at word or phrase level, because it also might be tag switching or intra swi tching. Based on the collected data, tag switching in this present study only occurs A). The


50 detailed procedures about how to count the occurrence s of switching to Engl ish ar e in Appendix B. Results Since the main purpose of this chapter is to test the different codeswitching patterns of these three instructors, I list the occurrence number and frequency percentage of different codeswitching types as well as codeswitching at d ifferent levels so as to calculate the average frequency percentage, and then I make a comparison as showed in Table 4 1 ( at the end of this chapter) Table 4 1 points to a clear difference in the Chinese use in classroom teaching among the three instructo rs. The percentage of codeswitching use of Emma is less than either Sun or Zhao: her average percentage of switching to English is 22.8 % whi le Sun is 34.9% and Zhao is 53.5 %. That is, Emma used TL for classroom language the majority of the time while Zhao used the least. As illustrated in Figure 4 1( at the end of this chapter). Since these three instructors are all female at similar age and also have many ye ars of teaching experience ( Table 3 1), then gender, age and teaching experience would not be cons idered as the potential factors that influence their difference in the use of TL classroom language. From Table 3 1, t he most noticeable difference among these three instructors is their na ti onalities: Sun and Zhao are Chinese na t ive speakers and they are both from Taiwan, Emma is an American who had furthered her Chinese learning in Taiwan. Because they are from different cultural backgrounds, they may have different past language learning experienc e, which could greatly influence their current teaching st rategies and beliefs However, to explain their different codeswitching behaviors only from this factor seems not enough O ther subjective factors such as


51 are worth investigating which can be collected thr ough debriefing and conversation with instructors I will discuss the potential reasons caused the different codeswitch ing behaviors in detail in the following chapter. M ore interesting ly despite the great difference in the percent age of TL classroom lan guage usage between these native and non native Chinese language teachers, they all share a similar codeswitching strategy. That is, they all prefer intra sentential switching over the other two types of codeswitching while the tag switching is the least u sed. In addition, th eir codeswitching to English is much more easily noticed occurring at sentence level, and then at the word level. As showed in Figure 4 2 and 4 3 ( at the end of this chapter). Figure 4 2 reveals that all these three instructors use int ra sentential switching much more than the other two types of codeswitching, inter sentential and tag switching. As Sa nkoff and Poplack (1981) define intra sentential switching is switching within the clause or sentence boundary which requires that the s peaker be fluent in both languages. As mentioned in the general demographic information and background of these three instructors in C hapter 3, t hey are all able to use both Mandarin and English quite fluently: Emma lived in China for about 11 years while Sun and Zhao have both live d in the US for almost 30 years. T here is no doubt that both their Mandarin and English are fairly fluent because of their long exposure to both langua ges. In addition, they all have rich teaching experience each teaching indivi dually for about 10 yea rs or more, which means they have a strong teaching intuition and judgment to organize their codeswitching patterns to maximally fa cilitate their teaching process. For example, their


52 tag switching is not unconscious behaviors, becaus e it is often use d when teachers introduce some new or emphasize important linguistic forms or when conducting language practices and check their comprehension T eachers know where and when a switch to En glish is necess ary and appropriate to Chinese classes. In most cases, they e clause or sentence to English ; instead, they only choose the necessary part which is the key point or something that may cause co nfusion as the subject of switching. As a result, they can construct a strong atmosphere f or learning Mandarin by using Mandarin as much as possible. Therefore, their similar frequent intra switching strategy is reasonable as the representation of thei r fl uency in both languages as well as their rich teaching experience. Figure 4 3, again, illustrates the similar codeswitching strategy employed by these three instructors. The tendency of switching to English at word, phrase, clause and sentence level s show s great similarity among these three instructors: sentence level and then word level codeswitching far exceed the rest t wo levels. Based on classroom observation I notice d that switching to English which occurs at sentence and word level often served as t he explanation function. For example, when teachers introduce d new words, review ed homework or conduct ed the whole class do some language practice, they would switch to English to explain the new lexical items or content on worksheet/handout or explain so me grammar points in order to avoid ambigu ity and confusion. They switch ed to English to find a counterpart at the word level for the new understanding, and also switch to English at sentence level to explain the grammar po ints or content that may be unclear for


53 students. I will analyze these functions in detail in the next chapter by using the qualitative analysis. In addition in order to test whether or not the difference of the three teachers codeswitching behaviors is significant, the C hi square is used to calculate the statistical significance of their codeswitching difference s (Appendix C) I first perform a simple contrast comparing the difference of classroom language usage among the three instructors (Table C 1) t he total Chi square value is 116.7, with 2 degrees of freedom, and the P value is 4.6E 26 ( )=116.7, P< .000001 ] ) which affirms the claim that the three instructors do differ in the classroom language usage in fact, their difference is great, as already shown in Figure 4 1, Emma used TL for classroom language the majority of the time while Zhao used the least. Then according to Figure 4 2 it is anticipated that the P value for teachers using the three codeswitching types should be P> .05 that is, there is no significant relation between the three ty pes of codeswitching usage am ong the three teachers since they all share a similar codeswitching strategy And again, the result also confirms this anticipation, N=646)=3.9, P=0.4] (Table C 2) teachers do not differ in the three codeswitching types However, when comparing the difference of teachers switching to English occurring at different levels, the results reveals more complexity I anticipated that the P value for the three teachers using codeswitching at different levels should be P>.05 because from Figure 4 3, their codeswitching usage tendenc y seems similar, but the result is P<.05, ] (Table C 3) which indicates that the three teachers differ in t he codeswitching usage at different levels. In fact, from Figure 4 3, we can see that al though the gener al tendency is similar, there are some clear


54 gap s at phrase and clause levels among the three teachers, especially t he noticeable gap between Zhao and the other two instructors. This is different from Figure 4 2 where their tendenc ies are almost overlap ping As a result, it would be interesting to make further comparison s between every two teachers, to test whether th ey differ or resemble in the codeswitching usage at different levels. The results of Chi square reveal that only Sun and Emma do not differ in the codeswitching usage at the four level s N=299 )=1.5, P=.7 ] (Table C 4 ) Whereas the difference of s witching to English at the four levels between Sun and Zhao, or Zhao and Emma is significant, the P value is 0.02 and 0.045 respectively (Table C 5 and Table C 6) From these results, we can infer that the differenc e of codeswitching behaviors among the th ree teachers cannot be simply attribute d to the factor nationality because Sun and Emma resemble each other while Sun and Zhao differ in the codeswitching usage exclude this factor, because the differenc e of classroom language usage among the se three instructors is quite significant, and Zhao and Emma also differ in the codeswitching usage at the four levels That is, as mentioned earlier in this section, other subjective factors should also be taken into account language attitudes toward teaching ; past language learning experien ce may have influence on instructors it is not the only factor. In this chapter, by comparing the occurrence percentage of codeswitching usage as well as different codeswitching patterns among these three instructors, I have explored some potential interpretations why the non native speaker teacher us e s more TL classroom language tha n the two native Chinese ; this may be caused by the ir


55 different pa s t language learning experience However, the results of Chi square test also lend support that other factors should also be taken into account to explain the three teachers different codeswitching behaviors. I will further analyze s uch pot ential factors in C hapter 5 In spite of the obvious difference s in the percentage of codeswitching use which has also been demonstrated by the result of Chi square these three instructors all employ similar codeswitching strategies, that is, more intra sentential switching than inter sentential and tag switching; more frequent s entence level switching and word level than phrase and clause level switching. In the following chapter, I will use qualitat ive analysis to address the final three resear ch questions, that is, to figure out the causes of the different codeswitching usage in classroom teaching among these three instructors. Table 4 1. Comparison of the frequency distribution of code switching strategies Instructor Type of CS CS to English oc curs at Total CS% tag inter intra word phrase clause sentence CS Sen. Sun 15 8.1% 66 35.7% 104 56.2% 52 28.1% 28 15.1% 34 18.4% 71 38.4% 185 529 34.9 Zhao 35 10.1% 115 33.1% 197 56.8% 118 34% 35 10.1% 38 11% 156 44.9% 347 649 53.5 Emma 5 4.4% 42 36. 8% 67 58.8% 31 27.2% 12 10.5% 24 21.1% 47 41.2% 114 501 22.8


56 Figure 4 1. The average percentage of switching to English of each instructor in their classroom teaching Figure 4 2. The average frequency percentage of three types of codeswitching of eac h instructor in their classroom teaching 0 10 20 30 40 50 60 Sun Zhao Emma Average CS% 0 10 20 30 40 50 60 70 tag inter intra Type of CS Sun Zhao Emma


57 Figure 4 3. The average frequency percentage of switching to English occurs at different levels: word, phrase, clause and sentence 0 5 10 15 20 25 30 35 40 45 50 word phrase clause sentence CS to English occurs at Sun Zhao Emma


58 CHAPTER 5 QUALITATIVE ANALYSIS In this chapter, qualitative analysis is employed to try to figu re out answers to the final three research questions: In what situations and for which functions do these three bilingual instructors employ codeswitching in beginning Chinese classes? To what extent do these teachers use codeswitching to fa cilitate their Chinese teaching? What are the potential interpretations related to the codeswitching practice s ? Analysis of the Function s of Codeswitching Through the quantitative analysis in Chapter 4, we saw that al though these thr ee bilingual teachers differ in the percentage of TL classroom language use during their Chinese teaching practice s they all resemble each other in the ir codeswitching strategy: the use of intra sentential switching far exceeds the inter sentential and ta g switching, and more frequent switching to Engl ish occurs at sentence and word level. During class observations, it is nevertheless clear that these three teachers codeswitched to ensure the maximum effect of their teaching in order to avoid potential mis understanding or incomprehension among students, and also to strengthen the relationship with their students. That is, codeswitching is used by teachers to ensure their output comprehensible to students ; teachers and students negotiate with each other to r esolve communication and language problems. Meanwh ile, codeswit ching is also applied to negotiate their relationship in classroom, to establish or maintain solidarity between teachers and students.


59 I will use the three main function categories of codesw itching that I identified in Chapter 2 as a starting point, and then define each specific function and situation where teachers employ codeswitching. This provide s me an interpretative tool to situate how teachers use codeswitching to facilitate their lang uage teaching. Constructing and Transmitting K nowledge As mentioned in Chapter 2, this function is viewed as t he most frequent in EFL classes. Teachers often switch to L1 to explain and emphasize linguistic forms or to on and so on. Like i n Ching yi linguistic forms so as to increase the comprehensibility of English monolingual text s Based on my observation this also applies in the beginnin g Chinese class es in UF. All three bilingual teachers ma inly use codeswitching for translation, highlighting and expla nation, which generally involve s intra sente ntial switching. They translate or highlight some certain key words or learning especially when introducing new lexical items and reviewing homework. When teaching new grammatical concept s, they would sometimes switch to English to make things clear er to understand if they noticed the explanation i n Chinese may be T ranslation In language classes, teachers may translate parts or the entire sentence into understanding. In the da ta under analysis, t his is the most frequent specific function used by all teachers, as shown in the following examples. (5 A) Sun: Shuaka, slide the card. Shua,


60 Example (5 A) occurred when Sun introduced a new Ch shuaka [slide the card] to students, after pronouncing the new word in Chinese, she then quickly some additional information about the meaning of the because in Chinese, it is common to find one character has more than one meaning when standing alone and combining with other characters. In fact, in this situation, translation is ine vitable because these students are all at Chinese beginning level, and thus they do t it is not that easy for teachers to find a synonym for the new wor d or even explain it in Chinese Tho ugh teachers can also use body language to co action, use they need to guess what this action refers to. Using body language thus can be viewed as an auxiliary method when explaining some new words or learning points to deepen students understanding; however, the other way to introduce new words in such a situ ation is definitely to switch it to the found in Zhao hen they noticed that certain lexical current vocabulary, a switch to English subsequently happened to translate the part they thought might be unclear, and usually t ranslation occurred within the clause or sentence boundary. Like in example (5 B), when Emma tried to explain the importa she firs lim ao [polite] to tell stu dents it would in his/her utterance. T hen she


61 realized the two synonyms were a little beyond students current understanding, so she quickly translated the main points into English to en sure all the students understood wh at she was talking about. And the the efficiency of this translation because in his sentence. Therefore, mutual understanding is successfully achieved through codeswitching. (5 B) Emm a: more polite. Student 5 The situatio n where codeswitching functioned as translation in my ob servation reaffirms what mentioned in previous studies (Harbord, 1992; Ching yi 2009 ; etc ) W hen explaining linguistic forms, like new words or phrases teachers may use L1 for translation to increase the comprehensibility of these new language items Th ough teachers do have other way s of explain ing such as using body language or paraphrasing, translation is definitely easier than these methods when considering t vocabulary and understanding H ighlighting Codeswitching is also a means appl ied by all teachers to emphasize key learning points or sometimes to shift topic s which teachers consider significant. Unlike the above sub function of translation, code s witching here is not served t o provide L1 counterparts for new L2 linguistic forms so t functions to highlight important points F or example, instructor Zhao


62 wanted to make sure her students had learned how to use Chin ese to exchange items in stores; she checked the key poi nts and highlighted them in English as shown below. (5 C) Zha o: Hao, women kan ni zenme apologiz e? Student 6 : Duibuqi. Zhao: Hao, duibuqi You want to change for another pair of shoes, another zenmesh uo? Other, other, ni xue le. Student 6 : [ Ano ther /Other Zhao: Biede xiezi, biede, xiezi, biede shuang, OK? Because me asure word, before measure word would only has the number and zhe, na, OK? Hao, or ni keyi shuo bieshuang xiezi, biede xiezi, right? Wo keyi huan biede xiezi ma? Hao, zailai, no problem zenme shuo ne? s, OK? Because measure word, before measure word would only has the number and this, that, OK? OK, or you can say another pair of shoes, other shoes, rig ht? May I change other shoes? OK next, how to say no problem?] Student 6 : Meiwenti. [No problem.]


63 Zhao : En, hen hao. [Um, very good.] In this example, Zhao first asked her student how to make an apology in Chinese, and s to English to highlight this is the main point in the question. So this student quickly respond which was the right answer to the question. If Zhao asked the ques tion totally in Chinese, though all the students had learned how to apologize in Chinese, but they were not familiar with the itself; o r for these beginning level students, if questions were composed of entire Chinese especially within words they were not that familiar with, students would feel difficult y either in listening to or answering the question. After this student answered this question, Zhao then affirmed his answer by repeating it in Mandarin, and then she provided him the situation in English that he needed to change for another pair of shoes. A in her utterance and highlighted it in Eng l step by step, she also gave this student a hint to r emind him another similar they had learned. In this way, Zhao successfully achieved her purpose of m Afterwards, she still kept switching to English to highlight points that she considered significan was ungrammatical. S he em phasized this point in English with understanding by say to make sure they remember this grammatical point. As mentioned in Chapter 4 such kind of tag The


64 three teachers do not randoml y use tag switching, because they use the Chinese tag marks the majority time, but in certain situations, like when introducing some new or emphasizing important linguistic forms or conducting language practice, teachers may highlight the tag mark in stude important learning points, and also to check whether they receive the comprehensible input or not. Moreover, to deepen their understanding, she then provided an explanation why it was un by highlighting the reason in English. Therefore, from this example, we can see that the codeswitching behavior is not random: the teacher only picks the points that she think s important to highlight in the code English Through highlighting she al so successfully gains these key point s to deepen their impression and understanding Based o n class observations, all three bilingual teachers preferred to codeswitching to highlight the key omprehension, or doing elicitation or correction. Sometimes, teachers also used codeswitching to signal or highlight a topic shift to D). (5 D) Sun: Women natian zuo haimeiyou zuowan, duima? Haiyou xie tongxue meiyou zuowan, suoyi woxiang jintian ba ta zuowan OK, wo shangge xingqiliu ne, gen wo de pengyou qu aolanduo. Aolanduo shina, zhidaoma, aolanduo? [We did the price is right student with my friend. Orlando, do you know, Orlando? ]


65 Students: Orlando? Sun: Dui. Wode pengyou ta xiang mai yige zhaoxiangji. Zhaoxingji, zhidaoma? [Right. My friend wanted to buy a camera. Camera, do you know?] Students: Camera? Sun: Dui. Tadao yige shangdian qu, nage dianli de re, ta kandao yige Sony model, ta hen xihuan. Shangdianli de ren gen tashuo, yao liangbai kuai. Danshine, woshuo, as I remembered, exactly the same mod el, if the price is right, buyaomai, taiguile, yinwei wo z ha o dao de zhiyao yibaiershi kuai. (She then looked around and pointed at something to let students guess the price.) [Right. He went to a store, the shop assistant, he saw a Sony model, he liked it very much. The shop assistant told him, it was two hundred dollars. However, I said, as because the price I found is only one hundred and twenty dollars.] In this exa mple, Sun first reminded her students that they need to continue the activity guess the price that the continue this activity, instead, she switched her code to English to highlight a topic shift in form of an interjection OK and then continued in Mandarin to tell her students something that happened last Saturday when she and her friend went to Orlando. In wo rds aolanduo [Orlando] and zhaoxiangji [camera] which they learned before, meanwhile, she helped students refresh their memories like how to say price in Chinese by giving examples liangbai kuai [two hundred dollars] and yibaiershi kuai [one hundred and twenty dollars] when talking about the price of the Sony camera. After


66 that, she shifted the topic back to the guess the price activity, and asked students to guess the price of different items she randomly picked in the classroom, like a recorder, an umbrella and so on, in order to let students be more familiar with the price expression in Chinese. The teacher uses the codeswitching to mark a frame shift from the lesson to something that occurred in her daily life and later back to lesson itself T he purpose of sharing he s to create a non threatenin g environment because she wants to make students gradually feel more comfortable instead of pushing them to practice their oral Chinese to guess pric e s directly. She ski llfully moves to signal a to pic shift, and then after providing some price expression examples she moves back to the activity itself. Generally, the situation s where this highlighting sub function served in my class observation are similar in most previous studies ( Lin 198 8; Raschka Sercombe, & Huang, 2009; etc ) Teacher use codeswitching to highlight key learning points or to ntion and deepen their understanding However, the definition and classification of codeswitching functions vary among different studies. Like i n Qian, Tian and Wa the s ituation where teachers switch codes to emphasize impor tant task requirements also as the highlighting sub function. From my class obs ervation, though codeswitching wa s also applied by these three bilingual teachers when giving some important task requirement s the main function is not to construct or transmit knowledge but is to manage the class room. In this way, teachers make their instructions clear to understand


67 which guarantee s the optimal effect of communication and the following activities being smoothly carried out and I will talk about t his clarificat ion sub function in detail in the section of Managing the Classroom In fact, t here is no single definition of codeswitching sub function s; and what I identified here follow s what I observed in classrooms. E xplanation When codeswitching is adopted to const ruct and transmit knowledge, it also serve s to provide intelligible explanations for some language learning points, b ecause explaining in L1 is much easier for students to understand especially for beginning level learners. In m y class observation, all thr ee bilingual teachers use codeswitching to make things clear to understand, because English is the code shared by everyone and much easier for students to understand especially when explaining some grammatical rules In example (5 E), when a student used a in the oral practice, Emma first prai is a very good way to (5 E) Student 7 : Emma: buyong ye keyi. It also depends on the context, kan Li xiaojie shi shui. Na wo, wo shuo de hua, Li xiaojie shuohua de shihou, wo ye yao kan Li xiaojie shi shui. Does this make sense? Mingbai le ma?


68 you do not use it It also depends o n the context, depends on who is Miss Li Like me, when I, talk to Miss Li, I also need to think who is Miss Li. Does this make sense? Do you understand?] Students: Bumingbai. (No, not quite get it.) Emma: Wo gen Li xiaojie shuohua de shihou, nin it depends, it depends yaokan Li xiaojie shi shui, who is Li xiaojie It is how I referred to her, determined by her status. yinwei wo shi laoshi. Dan ruguo Li xiaoj ie shi wo de laoshi, na jius Na wo ye buhui shuo Li xiaojie, wo hui shuo Li laoshi. Hao, mingbai le? who is Miss Li. It is how I referred to her, determined by her stat t, I because I am the teacher. But i in fact I ctor Li. Ok, do you understand? ] Students: En (nodded their heads). After praising this student, the teacher naturally introd u ced the grammatical In Chi in English. However, the use of the two charact is used to show respect when the hearer has higher social status than the speaker, At first, Emma mainly used some simple Chinese se ntences to explain this grammar point but found her students still looked confused, as they Theref ore, she then elaborated the grammatical rule and picked the most important point is determined


69 and further explained in English By doing this, the compr ehensibility of her explanation has been greatly increased since students corresponded by nodding heads in unison to show their understanding Though t his expla nation sub function seems similar with translation sub funct ion, both further interpret language learning points in L 1 to help explanation function does not offer an L1 equivalent in fact, codeswitching here is employed to offer an intelligible L1 interpretation when teachers notice the L2 explanation may be above studen as illustrated in example (5 E) Exactly as Ching yi (2009) mentions in the study, particularly, when explaining grammar rules, elaboration in L1 is inevitable experience, students would unders tand grammar rules better when they are taught in L1. As a result, based on my observation and from previous studies, codeswitching functioning With reg ard to the main function of constr ucting and transmitting knowledge, codeswitching does assist target language teaching and learning By switching codes to achieve sub functions like translation, highlighting and explanation, teachers are able to explain, emphasize linguistic forms and key learning points to help and deepen when the L2 language items or grammatical concepts are beyond their current comprehension T eachers can also draw learners ttention as well ntrated attention does play a significant role in their language learning process


70 Managing the C lassroom Teachers also use codeswitching for clarification and efficiency to manage the classroom to optimize com munica tion. For class management, teachers m ay switch codes to make their instructions clear to understand which is especially noticeable when assigning some important task requirements In this way, codeswitching fulfills the sub function clarification Meanwhile, codeswitching also serves to save time to express when in some pressing situations for efficiency C larification When organizing classroom activities, codeswitching is applied by teachers to make sure their utterances like instructions have been thoroughly and clearly receiv ed by learners From my observation, these three bilingual teachers all employ codeswitching to make their opening (signal the start of a lesson), instructions (i.e. provide students directions to complete an activity), giving class objectives, and closing (signal the close of the lesson) clearly especially for some important task requirements as illustrated in example (5 F) (5 F) Sun: Nimenhao (Several students stepped up the platform to ask a question in their homework.) [Hello everyone.] Ok, if you reading, hold on to your homework, (some students raised their homework and pointed homework. I were going to say it yesterday, but we ran out the time. Hao, dajiahao, women shangke le. [OK, hello everyone, time to start our lesson.]


71 This situation happened right a little befo re the beginning of the class. W hen the teacher entered the classroom, greeted her students and tried to start th e class as usual, several students walked to her with their homework to ask a question in reading before this class as required. After looking through quickly, Sun notice d that this Meanwhile, a t that time, the class was a little bit noisy since most students were still start the class in Mandarin as usual; she swiftly and reduced their anxiety about this homework because she said she would help them l became quiet, waiting for the teacher to start the class as usual. If Sun conveyed the same message in because they were a little anxious and the clas sroom was also a bit noisy. Chinese is not their mother tongue ; it would be hard for students to focus on Chinese instruction when they were in an anxious mood In addition, it may take much more time to make this tentative announcement clearly for each student if in Man darin. Thus, the easiest way to convey this message clearly is to announce it in English. As mentioned above, codeswitching that serves to guarantee that messages are clearly conveyed frequently occurs for important task requirements. Because teachers want to m ake sure their instructions are entirely understood by each student, in order to ensure the following activities process smoothly, they often highlig ht the key points or lish as shown in (5 G)


72 Again, before moving to the following concrete example, it may be mentioned in passing that this clarification sub function may have some similaritie s with the sub function of highlighting as I discussed in previous section. In this study, I fi r st classified codeswitching fun ctions into three main categories: constructing and transmitting knowledge, managing the classroom, and establishing or maintaining solidarity between teachers and students. Afterwards, situations where codeswitching occurred different from m y observation were sub classified and each should satisfy its main function. Though codeswitching is employed to highlig ht key task requirements, the main role it play s is not to construct and transmit knowledge but to manage the classroom Codeswitching here does serve to highlight points that teachers conside r significant but attention on linguistic forms like some grammatical concepts I n fact, it serves to make every student understand instructions clearly to guarantee teaching activities go on smoothly For this reason, in this study, highlighting important task requirements cannot be identified with highlighting key learning points as the same sub function : the former just fulfills the sub function of clarification which belong s to managing th e classroom. (5 G) Zhao: Hao, women kan dierge page, conversation maidongxi. Ni qu yige zhongguo de shangdian maidongxi, duibudui? Ni yao wen duoshao qian, shouhuoyuan hui wen ni, ni yao mai shenme, ni xiang mai shenme, mai shenme yanse de, ni xiangmai ku zi, OK? If you want to buy kuzi. Ni keyi nian ma? The instruction part.


73 store to buy things, right? You need to ask the price, the shop assistant woul d ask you, what do you want to buy, what color do you want to buy, you want to buy a pair of pants, OK? If you want to buy pants Could you read for us?] In this example, Zhao wanted students to practice how to do shopping in Ch inese and she first ga ve students some instructions for this activity. She told students to turn to page two to follow the model in the convers the context of shopping in a Chinese store, and detailed the shopping procedures, so as to give students hints about what one needs to ask, like the price, the color. She swi tched her I eem reasonable that she encoded the check in English because she thought her students could not understand duib first, and in fact she also used other Chi ma [do you through the lesson. The only plausible explanation to this code change is that the teacher wanted to guarantee her instructions were completely understood by each student, thus the fo llowing oral drill could proceed smo othly. Moreover, when she gave students the explicit instruction to limit the shopping item which was a pair of pants, she als to guarantee name and asked her to read the instruction part on page two, Zhao directly encoded and highlighted her direct ion The instruction part in English to tell the student what she expected her to do


74 clearly. Therefore, in these situations, teachers often enc ode, highlight, or repeat their instructions in English when they consider it important to do so, in order to make sure followed classroom activities could be practiced smoothly. In this study situations where codeswitching functioned to ensure teaching a ctivities go on smoothly are sub classified as clarification the teacher during oral drill practice to make sure no one missed her instruction. Such similar situation also occurred in my classroom observation as illustrated in example (5 G). T hrough switching codes messages like task requirements were thoroughly and clearly received by each student. As mentioned in previous section there is no one exclusive standard to classify codeswitching sub func tions. I n this study, all sub functions I identifie d should first satisfy its main func tion Since teachers employ a combination of encoding, highlighting or repeating their instructions in L1 code to make things clear whe n managing classroom, rather than to adopt a single method like translation or highlighting in L1 clarification is appropriat e to label this sub function here. E fficiency Codeswitching can also play a specific function for timekeepi ng, with teachers swit ch ing to L1 to manage time to express their ideas clearly for some uncompleted task requirements. This sub function of efficiency usually occurs at the near end of the lesson when teac hers notice that it may be hard to complete all the planned class tasks, then in order to ensure the optimal effect of communication and save time to clear some potential confusion or misunderstanding that may arise among studen ts, teachers frequently switch to L1 H), Emma planned to


75 let her students do situational responses first, and then she wanted to move to new words learning and also to make an announcement of the coming Chinese talent show. ecause at the end of that lesson, sh lead students through the new words. As a result, for the closing of that class, she first asked if anyone still had questions ab out the situational responses. S ubsequently, she quickly told students that they needed to leave learning the new words to the next class. But at that time, students already became impatient since it was the time to let them go, thus the class was a little noisy. In order ative language to express the announcement of the Chinese talent show. (5 H) Emma: Hai youmeiyou wenti? Wo benlai xiang jintain kaishi shang zhege shengci, dishike de shengci. Danshi, haoxiang, shijian daole. ( the class became noisy) Do you all, um got t [Any questions? I planned to introduce the new words, in Lesson Ten. But, it seems, we run out of the time.] Student 8 : Talent show? Emma: Talent show, right! Student 8 : I opened it. Emma: Hao, OK. Have you decided if um, we are goin g to do um, the song? [Good, OK.] Students: I like it.


76 Students: Yeah In fact, a fifty minute lesson can be an especially pressing situation because teachers may be hard pressed to finish their pla nned tasks, especially when at the end of the class when the tasks remained half completed. In such cases, the only practical English. For teachers, it is important t o finish their planned teaching activit ies in one class period so as no t to delay the next class plan Teachers need to make sure that their students would not fall behind the overall learning process that is, they attempt to finish all their planned clas sroom tasks in a limited time From this example, we can see that when Emma initiated in Mandarin, few students paid attention to her utterance since the class had become noisy. Subsequently, she quickly talked about the coming Chinese talent show as sche duled but in English ; and one student quickly respond ed. As she continued the topic in English, more and more students corresponded in unison to show their interests and concern. Similar examples are common in previous studies, for example, as Ching yi (20 09), Qia n, Tian and Wang (2009) observe from language classroom teaching, when i n pressing situation s teachers often use codeswitching to express their ideas about some unfinished activities, a nd usually, they also spoke faster which is the same for these three bilingual teachers in my observation. For this section, I present ed the second main function that codeswitching serves in language classroom teaching, and it can be divided into two sub function s clarification and effic iency From my class observat ion, t hrough encoding, highlighting and repeating in English, teachers send their message like task requirements to students explicitly. By


77 doing this the following teaching activities could be practiced smoothly since every learner totally understands wh at their teachers really ask for. On the other hand, when in pressing occasion s especially at the near end of t he class while the task remains half comp leted, teachers usually switch attention so as to save t ime and make sure the optimal effect of communication. Thus it can be seen that codeswitching does play an important role in classroom management. Establishing or Maintain Solidarity between Teachers and S tudents Codeswitching that occurs in language clas sroom teaching is served to ensure the maximum effec t of teaching; this is where classroom codeswitching differs from social codeswitching. However, since teac hers and students i n language classrooms can been seen as a small social community as well it is reasonable to infer tha t classroom codeswitching can also imply social effects. As mentioned in some previous studies (Adendorff, 1993; Merritt Cleghorn, Abagi, & Bunyi, 1992 ), teachers and students negotiate their relationship and identities in the soci al as well as classroom environment. Specifically, as Myers Sc otton (2006) points out codes witching is applied to decrease social distance to be more f avorable to the liste ners In my classroom observation, codeswitching is also sometimes employed to stre ngthen the relationship between teachers and students. Situations where classroom codeswitching functioned to establish or maintain solidarity between teachers and students are noticeable when teachers use codeswitching as a strategy to praise or comment a nd assess their students


78 P raise to signal friendship and solidarity since students can receive the praise clearly without missing any words, which would be beneficial for the teacher student relationship. For example, as Qian, Tian and Wang (2009) observed in EFL classrooms teachers sometimes use d L1, namel y Chinese, to praise students since English sounds more distanced. From their class video recording, it showed the s tudent who heard this L1 praise was quite please d By switching to Chinese praise is achieved. This sub function of praise is also confirmed in this study; it is not frequent, but it does exist. B ased on class observat ions, generally, all three of these bilingual teachers praised students for their good performance in class with simple English, and students also seemed quite happy when hearing th is kind of praise. The reason to account for this situation is that students are all quite familiar with these words, so there is no need to encode thi s kind of praise in English. However, when the ocabulary, teachers would definitely switch t his part to English to guarantee the maximal effect of the praise is achieved. For instance, as in example (5 I), instructor Sun praised her student for his accurate calculation and correct expression of price in Chinese. Subsequently, she hen ecause students had ed Chinese, Sun only encoded this word in her praise to ensure all the students received this message clearly. Obviously, students w ere quite pleased by hearing her praise


79 because they all laughed. This praise thus is beneficial to classroom rapport as well as to the teacher student relationship. (5 I) Sun: Qin gwen ta yigong yao fu duoshao qian? [So how much does she need to pay in total?] Student 9 : Yiban ling qikuai bamao jiufen. [One hundred and seven dollars and eighty nine cents.] Sun: Dui, ni daiduile, hen hao. Did you all get the same price? re right, very good.] (Other students all nodded their heads.) OK, hao, nimen de math henhao. [ OK, good, your math is very good.] Students: (laugh). Either from previous studies (Adendorff, 1993; Merritt Cleghorn, Abagi, & Bunyi, 1992 ; Myers Scotton, 200 6; Qian Tian, & Wang, 2009) or in this present study, besides teaching effects like constructing and transmitting knowledge or managing classroom, we see that codeswitching may also help to strengthen the teacher student relationship. Though teachers use codeswitching as a strategy for praise, this switch codes for praise. From my observation, the situation where codeswit ching served for praise occurs when teachers consider understanding, like for some certain unfamiliar words. But for those simple Chinese praise words that learners already quite familiar with, codeswitching would not be employed by these three bilingual teachers.


80 C omment ing and assessing p erformance During teaching practice, teachers also use codeswitching to express their uthority in language classrooms, which is the sub function of comme nting and assessing performance In my class observation, Mandarin is a total foreign language for students to learn English is a West Germanic language wh ereas Chinese belongs to Sino Tibetan language fa mily even languages that close to each other like French and Spanish, native speakers of either the two languages still cannot perceive them as the same Romance Language, or codeswitching would not occur between the two language speakers. Then, i t is no d oubt that language as English and Chinese that belong to different language families would definitely cause a sense of distance between the ir native l anguage speakers. Because Mandarin sound s distanced for these beginning level learners teachers then swit ch to E nglish to express their comment and assess to show their disapproval or agreement response, as illustrated in example (5 J). In fact, this specific sub function also has some overlaps with the function of construc ting and transmitting knowledge, because besides to sign is also to help students learn how to self repair their mistakes based on teachers by step comments. But compared with constructing and transmitting knowledge, commenting and assessing performance is more evident and straightforward to be noticed Therefore, I simply define the sub function of this codeswitching as commenting and assessing performance since it first


81 satisfies the main function of establishing or maintaining solidarity between teachers and students. Example (5 J) illustrates this sub function. (5 J) Zhao: Youwenti ma? Youma? [Any question? Any?] Student 10 : En, you. Zhao: Ok, Is this right or, mistakes? Students: Some mistakes. Student 10 : Zhegeshuang. [This one pair.] Zhao: F irst, two measure words, Ok? So, correct it. Student 10 : Zhao: Zheshuang xie shi heshi, anything wrong? [This pair of shoes be su itable, anything wrong?] Student 10 : Shi heshi, budui? [Be suitable, not right?] Zhao: En, henhao, henhao So how can you change, correct it? Z [Um, very good, very good. So how can you change, correct it? T his pair of


82 Student 10 : Zheng shuang xie hen heshi. [This pair of shoes is very suitable.] Zhao: Hao, henhao, donglema? Adjectives, before the adjective, you can not use shi Shi wo shi laoshi, wo shi xuesheng, right? Suoyi, adjective, [Good, very good, do you understand? Adjective, before the adjective, you can not In the previous example, this student was not sure whether it was appropriate to put two measure words before a noun or not, so he asked the teacher for help S ubsequently, Zhao commented on his sentence, and showed her disapp roval by Is this righ student rea lized his mistakes in the sentence, step by step, Zhao commented on each of his self correction s by F irst, two me Zheshuang xie shi heshi, anything wrong? [This pair of shoe s be suitable, anything and fi n En, henhao, henhao. [Um, very good, very good.] So how can you change, correct it? ; she affirmed this stu new revised sentence s by show ing her first A fterwards, she provid ed further advice in English to hel p this student to repair his other mistakes in his original sentence. Therefore, this sub function is related to teachers and students negotiat ion of power in the classroom environment A s language experts and evaluators, teachers can


83 demonstrate their au However, when the teacher encoded comments in English, it seems that they also sent Chinese teacher but ju st as a language helper, you see, mother tongue rather than the r equired classroom language In this way teachers can shift their role relationship from Chinese speaking teachers to bilingual helpers, which again, is very ben eficial to strengthen the relationship between teachers and students. As a result, for the sub function of commenting and assessing performances, it first teachers express their eval or be haviors to show either their approval or disapproval. I n return, students give response comm ents and assessments Furthermore since the classroom setting can also be conceived as a small social community, a community of practice, classroom codeswitching may also have some social effect s A s mentioned in Myers study ( 2006 ) the motivation of co deswitching is to be more favorable to the listeners, which is also proven to be tru e f or this specific sub function, similar to the sub function of praise When teachers adopt the L1 code, they just shorten the social distance with their students because they place themselves at the position of language helpers, which would make learners fe el more comfortable. That is, teachers tactfully use L1 to comment and assess perfo rmances to show their authority. M eanwhile, they successfully decrease their social distance with learners through shifting roles to language helpers.


84 In addition to maximiz ing teaching effects in language classrooms, codeswitching is served to strengthen the relationship between teachers and students as well Through s witch ing codes the maximal effect of prai se can be achieved because every word is explicitly received by le arners. In this way, students would feel pleased and satisfied with their current performances and also be confident in the following language learning Additionally, teachers also need to express their evaluation behaviors and performances to help their further language learning, so it is recommendable to employ codeswitching S ince the TL language sounds more distanced, by turning to L1, students would feel more comfortable By doing this teachers first demonstrate their authority and also t actfully m ake their comments and assess ment s more favorable to learners which is beneficial to classroom rapport and the teacher student relationship In a word, classroom codeswitching does play an important role in establishing or maintaining solidarity between teachers and students. General Conclusion on Functions I have described in the previous sections the three main functions of codeswitching observed in my data. The three bilingual teachers all participate in the three following functions: to accom the class room and also to strengthen teacher student relationships. That is, when teachers notice that some lexical items or grammatical concepts are main key points to translate, highlight, or explai n explicitly to make their utterance comprehensible to students This is often found in situations of introducing new words and grammatical rules, or when checking comprehension, elicitation and correction. This function of constructing and transmitting knowledge is the most frequent and noticeable in


85 language classes. Through switching codes, teachers can maximiz e their language teaching effort g as well. Meanwhile, all three bilingual teachers also use codeswitching to manage classrooms to facilitate their teaching. Wh en assigning class tasks or giving instructions, they may turn to L1 often because they need to make sure every student grasp s their m eaning clearly, so that the following activities can go on smoothly. In addition, in some pressing situations especially when at the near end of the class, in order to save time and ensure the optimal effect of communication, teachers prefer to switch to E nglish to express their attention. On the other hand, classr oom codeswitching can also have a social effect since it helps to establish or maintain solidarity between teache rs and students. Codeswitching is applied by teachers for praise or commenting and assessing their to miss any part of their praise since it is beneficial to strengthen teacher student codeswitching is efficient for creating a non threatening environment because L1 is more favorable to these beginning level learners which would definitely mak e them feel more comfortable As a result, codeswitching does play a significant role in language classrooms to maximize teaching effect and also to strengthen teacher student relationship. Differences among Individual Distributions of Codeswitching In th e above sections, I have defined different sub functions under the three main functions that I identified in Chapter 2. From my classroom observation s I classified situations where codeswi tching is applied into specific sub functions, and also analyzed


86 ho w the three bilingual teachers use codeswitching to facilitate their Chinese teaching in these situations. In this section, I will try to figure out the answer to the last research question: What are the potential interpretations related to the instructors codeswitching practice s ? Though these three bilingual teachers resemble each other in situations of codeswitching use and also in functions that their codeswitching served during their teaching practice, it is noticeable that, as mentioned in Chapter 4, they differ in the percentage of TL classroom language use. The non native speaker of the TL Emma uses the least codeswitching in her teaching practice. Through class observations, Em ma used much more body language than the other two native tea chers. If time permitted and she thought the points were important, Emma would try to use TL as much as possible, and she would sometimes use body language to let her students th ink on their own, like to infer what a certain lexical item refers to. T he fol lowing example (5 K) dem onstrates her teaching method. I n fact, she could simply provide her students the translation of instead of spending such a long time, to let the student himself grasp the mea ning of this verb under her action cue: sitting on a table again and again. (5 K) Emma: LiYou youmeiyou zuo gonggongqiche? [Did LiYou take bus or not?] Student 11 : LiYou meiyou qiche.


87 Emma: En, meiyou qiche? Wo meiyou wen ta youmeiyou qiche, wo shuo ta youmeiyou z gonggongqiche? (She made a posture of sitting.) d she take bus or not?] Student 11 Emma: gonggongqiche? (She sat on a table.) bus or not?] Student 11 : Meiyou zuo. Emma: gonggongqiche. Ni shuo ta meiyou qiche, she has no car, she has no vehicle, danshi wo meiyou wen ni ta youmeiyou qiche. You see the difference? Ni shuo ta meiyou qiche ni kandao le ma? You just sai d she has no car, ta youmeiyou zuo gonggongqiche? Ta meiyou zuo, ta zuole shenme, zuihou ta zuole shenme? aid she t said she has no what did she finally take?] Student 11 : WangPeng de che.


88 Emma: Ta zuole WangPeng de che, meiyou zuo gonggongqiche. Zuihou, LiYou zuo WangPeng de che. WangPeng kai c he, duibudui? WangPeng drive the car, right?] Student 11 : Dui. [Right. ] Emma: ta de che, duibudui? his car, right?] Student 11 : Dui. WangPeng called the taxi and carried LiYou ? [Right. WangPeng called the taxi and carried LiYou?] Emma : Wo meiyou shuo dache Student 11 : Yo u did say. Emma: Ta zuo, zuo (She sat on the table again.), zuo le che [She sat, sat, sat in the car.] Student 11 : Sit? Emma: Dui, ta meiyou zoulu, zuo, bushi zou. WangPeng de che. Haoba, wo leisile, rang nimen zuo, rang nimen yundong. the work, let you do exercise.] Students: (laugh) In this example, when this studen LiYou youmeiyou zuo gonggongqich he missed a key


89 which totally change d the meaning of his answer. H e intended his answer could only to lead this student to correct his mistake by providing him explanations entirely composed of simple Chinese s entences; to emphasize this verb. But with so much input, this student looked confused and did not really realize wha t his error was. Then, E mma changed her method a little; she only repeate d her question and again, at the same time, she also sat on a table. By doing this, this student noticed his mistake and revise however from the later utterance, we can see that this stu thoroughly, because when the o he mis understood the ne w combination zuo WangPeng de che as WangPeng ca lled the taxi and sat on the table again. This time, the on his own, so he said From this example, we can see that Emma lead the student to understand the on his own step by step. During their negotiation on the meaning of Emma used varies methods to help students achieve comprehension. Accord ing Emma modified her speech based on the ach response so as to achieve mutual understanding. S he simplified and repeated her question, accentuate d th rough out the whole negotiation process. Although this kind of neg o t i ation can help


90 students deepen their understanding to certain lexical items, it is quite time consuming. I also asked Emma why she seldom used codeswitching in her class teaching practice since it is an eas y way for negotiation as in t at the very beginning, she could definitely saved a lot of time since a fifty minute class could be very pressing. The explanation she gave me is that this is her habit which she developed when learning Chinese in Taiwan. When she was young, she went to Taiwan for further Chinese learning, however, all her Chinese friends around her there knew little English and their language communication tool wa s only Mandarin. As a r esult, when some communication barriers occurred, the only way for her and her Chinese frien ds to communicate wa s through body language, like different facial expressions and gestures, and later she would double check the meaning of these new Chinese words in the dictionary. She said she also liked watching Chinese movies at that time, and once she heard some unknown or unfamiliar lexical items, she would write down the pinyin (a system for transliterating Chinese ideograms into the Roman alphabet) and late r looked up in a dictionary. Therefore, she had further learned Chinese almost on her own because there was no one could help her through codeswitching in Taiwan at that time. Ac cording to Emma, she thought her p ast Chinese learning experience exerted a p rofound influence on her curren t teaching practice, so she tried t o use as much Chinese as she could H owever, on the other hand, from her teaching experience, for thes e beginning level students, it i s impossible to insist on a Mandarin only teaching poli cy. Even teachers can use body language, paraph rasing or some other methods to aid their


91 teaching practice. C onsider ing the time vocabulary, codeswitching is still inevitable, which is also the point that Emma affirmed in the conversation. But if t ime permitted as shown in (5 K) she said she would prefer to use much more TL and also body language to explain certain learning points to dee pen her students even if codeswitching is much easier to negotiate the meaning. A nd she also expected her students could derive great pleasure from learning language on their own just as she did. current language teaching attitudes have been greatly influenced by her past language learni ng experience, rather than the monolingual requ irement o n the beginning Chinese syllabus. She expects maximal use of the TL l anguage; and sometimes she would tactfully use L1 to facilitate her teaching when taking bulary or time efficiency into account. On the other hand, the percentages of codeswitching usage by the two native Chinese teachers are relatively higher especially instruc tor Zhao. The two instructors are bot h from Taiwan, where switching to Chinese frequently oc curs in EFL classrooms as described in those related articles of codeswitching in Chap ter 2 (Ching yi 2009; Raschka Sercombe, & Huang, 2009; etc ) That is, they might be quite accustomed to codeswit ching when they were English learners; and this past language experiences, to a certain degree, would also influence their current teaching methods. A s Zhao said in the later casual conversation after the data collection, she thought th e most important te aching goal wa s to clear all the potential confusion among students especially when explaining grammatical rules, which is the key point she re alized when she was a language learner and the easiest way to achieve thi s goal i s


92 switching to r tongue. In addition, based on her teaching experience, a fifty minute class is really a pressing situation since there are so many new linguistic forms that need to be taught, and more important ly teachers also need to guarantee their utterance s are cle arly understood by every student. Such situation s would become even harder when learners are at beginning levels due to their limited co mprehension to the TL language, which would definitely build communication barriers either in language teaching or class room rapport if teaching a target language only through itself. Therefore, Zhao employed codeswitching often to maximize her teaching effect to make sure she could finish all the planned teaching activities without leaving any student behind the language l earning progress Based on her past language learning experience as well as so many years of language teaching experience, she regards codeswitching as an important part in language teaching, and the percentage of TL use target language proficiency level. For these beginning level students, it would be efficient if teachers turn to L1 more. I n example (5 L), the percentage of her English use has greatly exceeded the use of Mandarin when expl xian zai [first as well as the placement of the subject, but apparently, through codeswitching, the mutual understanding is achieved (5 L) Zhao: Hao, women kan yufa xian. Di liangbaiwushiba ye or first, then, xian zai, xian zai. Hao, ni keyi nian ma?


93 eight, or pag e two hundred and fifty then. Ok, (called one coul d you read for us?] Student 12 : Xian kan dianying zai chifan, first go to the movie then eat means kandingying before chifan [First go to the movie then eat, first go to the movie then eat means seeing a movie before eating.] Zhao: Ok, first go to the movi e. How about the subject? First I go to the movie, then I eat? Where do you put the subject in this sentence? In English is that I first go to say the Chinese sentence, firs t I go to the movie, then, the subject, with the subject? Students: Zhao: Wo xian kan dianying, zai chifan You can drop the second wo right? Can you say xian wo kan dianying zai chifan? [I first go to the movie, then eat. You can dro p the second I, right? Can you say first I go to the movie, then eat? ] firs in the first most time, you would say first I go to the movie, then do something, right? Why do you say, can you say xian wo kandianying, zai wo chifan ? [ Why do you say, can you say first I go to the movie, then I eat (in Chinese) ?] Students: No.


94 Zhao: Yeah, no, never, never. W xian OK? I xian wo i wo xian [Subjec OK? I i n the homework, .] From the above example, we can see that Zhao used codeswitching to achieve mutual understanding successfully, which saved her m uch time and effort to guarantee her teaching go on smoothly as planned. In fact, from my class observations, she would quickly once she noticed what she said might be above their comprehen sion. Unlike Emma, if time allowed she would use as mu ch Mandarin as sometimes it means time consuming as shown in (5 K) it definitely deepens stimulate their zest for languag e learning as well Thus it seems the two different percentages of TL language use have different teaching understanding better or to save time and effort to ensure more time left for students to practice in the classroom More in teresting ly since Zhao and Sun have similar language learning background s in Taiwan, then their percentage of TL language use might be close if past language learning experience does play a significant role in their current teaching methods. However, Sun used more TL language than Zhao, bu t less than Emma that is, her percentage of Mandarin use is right between Zhao and Emma. So I also asked Sun the same que stion about her motivation of codeswitching usage A s far as she was concerned, based on her pa st language learning experience teaching experience, codeswitch ing is necessary and inevitable, and this response is


95 explanation. But she also added that during her teaching practice, she always kept the requirement o f Mandarin only on the syllabus in mind. Therefore, she always felt a strong conflict between the monolingual policy and what she actually did in her teaching practice and this point is not mentioned in the conversation with the other two teachers. Sun wa nted to increase the comprehensibility by codeswitching; on the other hand, she also tried to control the percenta ge of English usage Because of this distinctive language attitude toward teaching she used more Mandarin than Zhao through consciously restr i cting the use of codeswitching, but still less Mandarin than Emma As a result, from these three bilingual teachers, it seems that different pa st language learning experience and language attitudes toward teaching would form distinctive teaching beliefs which would later influence their r eal teaching practice. B ased on their teaching experience, the three bilingual teachers all admit the inevitability of codeswitching in language classrooms especi ally for beginning level student s. Therefore, co deswitching is applied in classroom s to maximize their teaching effects and to strengthen teache r student relationship as well, which may account for the phenomenon why they use code switching strategies to fulfill similar functions. However, the percentage of codesw itching usage varies from person to person. Emm a furthered her Mandarin learning in Taiwan through the TL language only with the help of body language, from which she derived great pleasure and al so deepened her comprehension in Mandarin As a result, she wanted her students to also develop the similar language learning habits as she did. Her attitude toward codeswitching is a little above neutral since it is a quite easy way to achieve successful classroom negotiation


96 but if time permitted, she still pref erred to use as much Mandarin as possible If body codeswitching would be the last option she would choose, which is quite different from instructor Zhao. In her opinion, codeswi tching is the top option to save time and effort to resolve communication or language problems based on her past language learning experience, which has also proved true from her own teaching experience since students could receive her message explicitly b y switching codes. That is, her attitude toward codeswitching is quite positive and she preferred using codeswitching to facilitate her teaching. Finally for instructor Sun, her attitude toward codeswitching is just positive not that strongly support as Z hao She had committed the Mandarin only requirement to memory which caused a strong conflict with what she had learned from her p ast language learning and her later teaching experience. So she would consciously control the L1 output in her teaching pract ice. Therefore, even though the three bilingual teachers employ codeswitching strategies to fulfill similar functions ; the percentage distribution of TL classroom langu age use still varies with each individual due to the influence of their different past l anguage learning experience and language attitude toward teaching. Therefore it seems that the teaching philosophy of these three instructors plays a big ger role as their nationality (mainly their past language learning experience) in their current teachin g practices. General Conclusion In this chapter, I have used qualitative analysis to address my final three research questions. Step by step, from class observation, I first defined several sub functions of codeswitching under the three main functions Th en I classified situations where codeswitching occurred into each specific sub function and every time, c ho se one


97 representative situation to analyze in detail, to explore to what extent teachers use codeswitching to facilitate their Chinese teaching. Sinc e there is no single standard to classify specific functions of codeswitching, and sometimes one codeswitching str ategy may have some overlaps with other different functions when analyzing deeper, then in this study, each sub function I defined is based on what I observed in these language classes, and is classified to the ma in function it first belongs to, that is, the most noticeable function it fulfills in the situation. Generally, functions of codeswitching that I defined are similar to what mentioned i n previous stud ies in Chapter 2. The three teachers all participate in the thre e main functions: they switch codes for translation, meanwhile, they use codeswitching for clarification and efficiency to ensure the optimal effect of communication so as to manage classroom effectively; in addition, classroom codeswitching is also applied for praise and commenting and assessing performances to strengthen teacher student relationship. Finally, through debriefing and conversation with the three teachers, I identified t codeswitching practice s which confi r med the anticipati on as mentioned in Chapter 4, that there must be some facto rs other than nationality (mainly the past language learning experience) that ex ert big influence on t heir current teaching practices T hough these teachers use codeswitching strategies to fulfill similar functions, they differ in the percentage of TL classroom language use. Based on the information the three teachers offered, their different percentage distributions of Mandarin usage are influence d by their different past language learning experience and language attitude toward teaching. Because


98 different pa st language learning experience would form distinctive beliefs, and their own language attitudes toward teaching would also developed with the accumulation of their teaching experience, subsequently, the two factors would r eal teaching practice greatly, which just result in the great difference in the percentage distributions of their codeswitching use.


99 CHAPTER 6 CONCLUSION Though codeswitching in l anguage classroom teaching is not a new topic, there are few studies descri b ing the codeswitching behaviors of Chinese language teachers in English speaking environ ment s and even fewer studies have investigated to what extent the codeswitching strategies employed by diffe rent bilingual teachers are similar and differ ent In this thesis I have contributed to shed ding light on such language practices i n the classroom in an American u niversity setting. According to Sankoff and Poplack (1981), there are three types of codeswitching syntactically, namely, tag switching, intra switchin g and inter swi tching To further analyze d ifferent types of co deswitching from another aspect, I also categorized switching to English codes at the word, phrase, clause and sentence level as illustrated in Chapter 4. That is, in this study, the two differ ent classification s of codeswitching types are both used. Through the quantitative analysis on the codeswitching behaviors of three bilingual teacher s who teach beginning Chinese at UF, it is noteworthy to find that, though the percentage of TL classroom l anguage varies from one instructor to another, the non native speaker teacher use more Mandar in than the two native Chinese. However it has been demonstrated that the codeswitching pattern s used by these thr ee teachers are similar That is, more intra sent ential switching than inter sentential and tag switching; more frequent sentence level switching and then word level than phrase and clause level switching as shown in Chapter 4. The similarities in their codeswitching strategies indicate that as being exp erienced bilingual teachers, they all have a strong teaching intuition and judgment to organize their code switching patterns to maximally facilitate their teaching process, namely, they know where and

PAGE 100

100 when a switching to English is necessary and efficient understanding in Chinese classes. Moreover, a s mentioned in Chapter 2 and also from my class o bservation, classroom codeswitch i ng is necessary and inevitable. The language classroom is a community of practice, where teachers and students ne gotiate with each other to resolve communication and language problems Since the use of codeswitching can easily and successfully achieve mutual understanding it is often employed by teachers t o facilitate their language tea ching So ther e is no doubt that functions of codeswitch i ng have been discussed so many times in some related p revious studies (Li n, 1988; Raschka Sercombe, & Huang, 2009 ; Ching yi 2009; etc ) I n this study, even there is a Mandarin only requirement on the syllabus i n beginning Chinese classes ; codeswitching is still applied by the three bilingual teachers. Since their codeswitching patterns are similar as illustrated in Chapter 4 it would be necessary to describe and further analyze the specific functions of their c odeswitching, to investigate whether teachers use similar codeswitching patterns to fulfill similar functions or not. Meanwhile, from the quantitative analysis, the percentage of codeswitching usage also varies from on e bilingual teacher to another. T he r esults of Chi square test also lend strong support to a significant difference of codeswitching usage among the se three instructors. However, more interesting, the statistical results of difference on t experience cannot be regarded as the only factor that influences their current teaching practices because teachers of different nationalities, which means they may have different experience as foreign language learners, resemble each other at the four

PAGE 101

101 levels of codeswitching usage O n the other hand, teachers of the same nationality exhibit difference of codeswitching at the four levels Therefore, potential interpretation s related to the differences are noteworthy to fig ure out and a detailed qualitative analysis is the next logical step to resolve these research questions A s demonstrated in Chapter 5, situations that motivate these three bilingual teachers to accept codeswitching in their teaching practice are also sim ilar Classroom codeswitching is used to maximize their teaching effects as well as to strengthen teacher student relationship, which has affirmed findings of codeswitching functions in previous studie s (Qian, Tian, & Wang 2009; Ching yi 2009 ; etc ) Tha t is, besides constructing and transmitting knowledge as well as managing classrooms, codeswitching in l anguage classroom can also have social effects to establish or maintain solidarity between teachers and students. I classified situations where codeswi tching occurred into the three main functions as a starting point, and then sub classified them into specific sub functions based on my own classroom observations. Generally, w hen teachers realize the lexical items or grammatical concepts are beyond their learning points are quite important or want to signal a topic shift codeswitching would naturally occur to fulfill these specific functions like translation, highlighting or explanation to accommodat e ve time and effort to ensure that students receive comprehensible input teachers would encode, highlight or repeat some important task requirement s or instructions for clarification, or switch co des for the unfinished tasks when in s ome pressing situations for efficiency. respon ses,

PAGE 102

102 utterance meanwhile, it serves to gai n learners grasp their explicitly. In addition, praise or advice when the meaning is thoroughly accepted by switching codes which is quite beneficial to classroom rapport and teacher student relationship. Finally, from some casual conversations and debriefing with each instructor, it seems their past language learning experiences and their own language attitude s toward teaching have greatly influenced their current t eaching practice, which result in thei r clear difference in the percentage distributions of TL classroom language use For, before they became language teachers, they all went through the language learning stage, and the learning habits they established then would gradually form ed their distin ctive beliefs either in learning or teaching. M eanwhile, with the accumulation of teaching experience, their own language attitudes toward teaching would also deve loped, and the two factors could influence their current teaching practice greatly. According to their teaching experience, they all confirmed the inevitability of classroom codeswitching, th at is, they all felt necessary to switching codes to facilitate their teaching work for these beginning level learners in time limited classes But a mong the three bilingual teachers, only t he non native speaker teacher Emma further ed her TL language learning through the TL which deepened her understanding as well as stimulated her zest for language leaning, thus she preferred to use Mandarin as much as possib le to try to help her students establish the similar learning habit s which just resulted her highest percentage of TL classroom language On the other hand because of insistence to monolingual requirement her teaching attitude towar d codeswitching is not that strong positive like Zhao, thus she often limited her

PAGE 103

103 output in English consciously which resulted her percentage of TL classroom language is lower than Zhao but relatively higher than Emma. However, as we have seen above, they use similar codeswitching patterns and fulfill similar functions as well. For further study, it would be interesting to test whether such difference s and similarities do exist or not in the codeswitching behaviors between native and non native speaker t eachers. Unfortunately, my sample was not large enough in this present study; therefore, it would be not that plausible to make a generalization based on this factor. Moreover, this study just compared the percentage of TL classroom language and the freque ncy percentage of d ifferent types of codeswitching; it might also be interesting to compare discourse on codeswitching strategies In addition, for th e potential reasons that relate information I gathered was only from some casual co nversations with these teachers; to some degree, it was not that sufficient. Other factors such as different language teaching training were not taken into account which could also be investigated for further study Finally, since classroom codeswitching is used for language teaching and interesting route to take.

PAGE 104

104 APPENDIX A UFIRB #2010 U 1087 INFORME D C ONSENT Protocol Title: An exploratory investigation of language practice in beginning Chinese class in UF Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The general purpose of this study is to investigate the conflict and accommodation in the use of languages between teachers and learners i n beginning Chinese class in UF. The primary focus is on the different behaviors of language practice in classroom. E xplore and analyze the motivations, the pedagogical value for the foreign language acquisition. What you will be asked to do in the study: Your Chinese lesson will be audio recorded. Time required: 50 minutes Risks and Benefits: There are no anticipated risks, compensation or other direct benefits to you as a participant in this interview. But participation will yield results that will he lp bilingual teachers in their teaching of Chinese. Compensation: There is no compensation. Your participation will be highly appreciated. Confidentiality: Your identity will be kept confidential to the extent provided by law. Your information will be as signed a code number. The list connecting your name to this number will be kept in a locked file in my faculty supervisor's office. When the study is completed and the data have been analyzed, the list will be destroyed. Your name will not be used in any r eport. Voluntary participation:

PAGE 105

105 Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at anytime without consequence. Whom t o contact if you have questions about the study: Yanmin Bao, MA Student, Department of Linguistics Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; p hone 392 0433. Agreement: I have read the procedure described above. I voluntarily agree to participate in the procedure and I have received a copy of this description. Participant: ____________________________ Date: _________________ Principal Investi gator: ___________________ Date:________________

PAGE 106

106 APPENDIX B DATA TABLES In the following tables, TL is the short form of target language, refers to Chinese. Number in that column refers to the occurrence number of sentence totally or mostly in Chinese in each step. CS refers to codeswitching. Since codeswitching occurred at clause or sentence level can be either viewed as inter sentential switching or intra sentential switching. In order to avoid such ambiguity, numbers without brackets in the cell belo sentences entirely in English (occurred as the inter sentential switching). On the other hand, numbers in the brackets indicate the occurrence of clause or sentences mostly in English (occurred as the intra sent ential switching). Similar to the switching at w ord or phrase level the occurrence number of tag in brackets. I will use the T able B 1 to make a detailed explanation about how to count the occurrence number of codeswitching to English in this lesson. From this table, I take the indicates that there are 81 Chinese sentences, including sentences entirely in Chinese and mostly in Chine se which have overlaps of switching to English occurring at word, sentences, in total, 11 sentences (7 + 4) uttered by the teacher have the switching to English at the word level Among which, 4 are tag switching while the rest 7 are intra with independent entire English clause as inter number of swit ching to English taking place within the clause boundary but not at word

PAGE 107

107 or phrase level That is, these 81 sentences also involve 8 sentences (3 + 5) which have a switching either at or within the except t hese 81 sentences, there are other 15 sentences (8 + 7 ) 8 are enti rely in English while the rest 7 are mostly in English, were uttered in this step. As a result, the total number of se ntences in this class activity is 96 (81+15). As a whole, in this class, the total number of sentence the teachers produced is 164 (144 sentences entirely or mostly in Chinese plus 20 sentences totally or almost in English). The total number of sentences with codeswitching is 52 (5 tag switching plus 17 inter sentential plus 3 0 intra sentential switching or 13 word level plus 9 phrase level plus 10 clause level plus 20 sentence level switching). The occurrence percentage of codeswitching to English in this lesson is 52*100/164=31.7%. Table B 1. 1 st class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 2 2 Introduce color words 33 1 2 1 1 (1) 2 35 Checking comprehension 5 2 1 2 (1) 8 activity 6 2 2 2 2 6 Elicition 10 10 Correction 4 4 Introduce new words 81 4 11 25 7 (4) 6 3 (5) 8 (7 ) 96 Closing 3 1 1 3 Totals 144 5 17 30 13 9 10 20 164 Frequency % 9.6 32. 7 57. 7 25 1 7 .3 19 2 38 .5 CS% (the occurrence of CS to English in this le sson): 31.7%

PAGE 108

108 Table B nd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 3 1 1 (1) 1 3 Vocabulary quiz 8 1 (1) 9 Changing topic to clothing items 6 2 1 (1) 7 Introduce clothing words 27 2 2 10 5 (2 ) 3 1(1 ) 1 (1) 29 Change to text reading 15 15 Checking reading 40 7 4 2 7 (2) 49 Grammar points 57 4 10 15 5(4) 4 3(4) 7(2 ) 66 Closing 5 2 1 1 5 Totals 161 7 19 35 20 10 9 2 2 18 3 Frequency% 11.5 31.1 57.4 32.8 16.4 14.8 36 CS%: 33.3 Table B rd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 5 3 3 8 21 1 2 8 2(1) 2 1(2) 1(2) 24 Listening practice 22 1 1 1 1 2 3 Evaluating mini oral 1 3 5 1 1 2 3 16 Review HW 33 2 11 14 4(2) 4 2 (3) 9 (3) 45 Review measure word 13 3 3 2 1 (1 ) 2 1 5 Activity instruction 7 3 6 3 2 1 (1) 2 9 Checking performance 3 6 2 6 3 1 1 1(2) 3 9 Closing 3 3 Totals 15 3 3 30 39 19 9 1 5 2 9 182 Frequency% 4.2 41.7 54.1 26.4 12.5 20.8 40.3 CS%: 39.6

PAGE 109

109 Tab le B st class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phr ase clause sentence Greeting 1 1 1 1(1) 3 Language practice 127 13 34 43 11(13) 11 4(3) 30(18) 175 Checking group performance 9 1 3 3 1 10 Closin g 4 3 2 1 4 Totals 141 13 36 50 29 12 7 51 192 Frequency % 13.1 36.4 50.5 29.3 12.1 7.1 5 1.5 CS%: 51.6 Table B nd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 2 2 Warming up 19 1 3 11 6(1) 1 2 1(4) 24 Introduce new words 122 11 32 61 24(1 1) 8 7(3) 25(26) 173 Classroom activity 39 3 11 8 4(3) 1 2 9 (3) 51 Closing 2 2 Totals 184 15 46 80 49 10 14 68 252 Frequency% 10.6 32.7 56.7 34.8 7.1 9.9 48.2 CS%: 56

PAGE 110

110 Table B rd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 3 1 1 3 Language practice 1 27 1 2 12 5(1) (3) 2(4) 33 Checking performance 15 2 3 2 2 (1) 16 Instruction 3 4 2 (2) 5 Checking performance 2 0 2 8 5 2 2(1) 20 Instruction 6 1 1 1 (1) 6 Language practice 2 28 2 5 10 5(2) 3 3 2(2) 32 Checking performance 36 2 7 11 4(2) 3 1(1) 6(3) 45 Speaking practice in textbook 28 2 14 17 10(2) 4 2 12(3) 43 Closing 2 2 Totals 168 7 33 67 4 0 13 17 37 205 Frequency% 6.6 30.8 62.6 37.4 12. 1 15.9 34.6 CS%: 52.2 Table B 7. Emma st class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 2 1 1 2 Situational resp onses 15 2 1 1 8 24 9 (1) 2(6 ) 16(9 ) 17 7 Closing 3 3 Totals 15 7 1 1 8 25 10 1 8 2 5 18 2 Frequency% 2.3 40.9 56.8 22.7 2.3 18.2 56.8 CS%: 24.2

PAGE 111

111 Table B 2 nd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 1 1 Vocabulary quiz 15 15 Tips for memorizing characters 7 1 3 2 1 1 8 Review text content 58 2 4 2 (2) 2 60 Instruction of language practice 5 5 Grammar points 19 2 3 4 2 (2) 1 3(1) 23 Check performance 13 1 1 1(1) 13 Closing 5 3 2 1 1 3 8 Totals 123 2 10 14 9 3 4 10 133 Frequency% 7.7 38.5 53.8 34.6 11.5 15.4 38.5 CS%: 19.5 Table B rd class Lesson Plan TL Type of CS CS to English occurs at Totals tag inter intra word phrase clause sentence Greeting 3 3 Introduce new words 44 8 14 3 4 4(4) 4 (3) 51 Review text content 38 4 6 3 2 2(1) 2 40 Change to activity 5 1 1 5 Introduce new words 32 2 1 4 2(2) 1 1(1) 34 Activity practice 35 1 3 1 1 (1) 1 36 Instruction of group work 12 12 Closing 5 5 Totals 174 2 14 28 12 8 12 12 186 Frequency% 4.5 31.8 63.7 27.3 18. 1 27.3 27.3 CS% : 23.7

PAGE 112

112 APPENDIX C CHI SQUARE R ESULTS Table C 1 Difference of code s witching usage codeswitching Mandarin Total Sun observed 185 344 529 Sun expected 203.5 325.5 Zhao observed 347 302 649 Zhao expected 249.7 399.3 Emma observed 114 387 501 Emma expected 192.8 308.2 Total 646 1 033 1679 26 ] Table C 2 Difference of three codeswitching types usage tag switching inter switching intra switching Total Sun observed 15 66 104 185 Sun expected 15.8 63.9 105.4 Zhao observed 35 115 197 347 Zhao expe cted 29.5 119.8 197.7 Emma observed 5 42 67 114 Emma expected 9.7 39.8 64.9 Total 55 223 368 646 Table C 3 Difference of four levels codeswitching usage word level phrase level clause level sentence level Total Sun ob served 52 28 34 71 185 Sun expected 57.6 21.5 27.5 78.5 Zhao observed 118 35 38 156 347 Zhao expected 108 40.3 51.6 147.2 Emma observed 31 12 24 47 114 Emma expected 35.5 13.2 16.9 48.4 Total 201 75 96 274 646 ] Tabl e C 4 Difference between Sun and Emma word level phrase level clause level sentence level Total Sun observed 52 28 34 71 185 Sun expected 51.4 24.7 35.9 73 Emma observed 31 12 24 47 114 Emma expected 31.6 15.3 22.1 45 Total 83 40 58 118 299 f 3, N=299)=1.5, P=0.7]

PAGE 113

113 Table C 5 Difference between Sun and Zhao word level phrase level clause level sentence level Total Sun observed 52 28 34 71 185 Sun expected 59.1 21.9 25 78.9 Zhao observed 118 35 38 156 347 Zhao expected 110.9 41.1 47 148. 1 Total 170 63 72 227 532 ] TableC 6 Difference between Zhao and Emma word level phrase level clause level sentence level Total Zhao observed 118 35 38 156 347 Zhao expected 112.2 35.4 46.7 152.8 Emma observed 31 12 24 47 114 Emma expected 36.8 11.6 15.3 50.2 Total 149 47 62 203 461 46 1)=8, P= .0 4 5]

PAGE 114

114 LIST OF REFERENCES Adendorff, R. (1993) Code switching among Zulu speaking teachers and their pupils: its functions and implications for teacher education Language Education, 7 (3), 141 162. Atkinson, D. ( 1993 ) Teaching monolingual classes. London: Longman. Carless, D. (2008). Student use of the mother tongue in the task based classroom. ELT Journal: English Language Teaching Journal, 62 (4), 331 338. Cenoz, J., & Genesee, F. (2001). Trends in bilingual acquisition. Amsterdam: John Benjamins. Chambers, G. (1992). Teaching in the target language. Language Learning Journal 6 (1), 66 67. Chaudron, C. (1988). Second language classrooms: Research on teach ing and learning. Cambridge: Cambridge University Press. Ching yi, T. (2009). Conflict and accommodation in classroom codeswitching in Taiwan. International Jornal of Bilingual Education & Bilingualism, 12 (2), 173 192. Den Branden, K. V. (1997). Effects Language Learning, 47 589 636. Dulay, H., Burt, M. & Krashen, S. ( 1982 ). Language t wo. New York NY : Oxford. Gass, S. M. (1997). Input, interaction, and the second language learner. Mahwah, NJ: Lawrence Elrb aum. Grosjean, F. ( 1982 ) Life with two languages: An introduction to bilingualism (pp.146) Cambridge, MA and London: Harvard University Press. Gumperz, J. J. (1982). Discourse Strategies. Cambridge : Cambridge University Press. Harbord, J. (1992) The use of the mother tongue in the classroom. ELT Journal 46 (4), 350 35 5 Haugen, E. (1956). Bilingualism in the Americas: A bibliography and research guide (pp. 40) Montgomery: University of Alabama Press. Howatt, A. P. R. ( 1984 ). A history of English l anguage t eaching. Oxford: Oxford University Press. Ho, B ., & Van Naerssen, M. ( 1986 ). Teaching English through English and through English and Chinese in Hong Kong Form 1 remedial English classrooms Educational Research Journal, 1, 28 34.

PAGE 115

115 Kharma, N. N., & Hajjaj, A. H. ( 1989 ) Use of the mother tongue in the ESL classroom. International Review of Applied Linguistics in Language Teaching, 27 (3), 223 2 35. Lin, A. M.Y. ( 1988 ) Pedag o gical and para pedagogical levels of interaction in the classroom: A social interactional approach to the analysis of the code switching behavior of a bilingual teacher in an English language lesson. Working Papers in Linguistics and Language T eaching, 11, 69 87. Lin, A. M. Y. ( 1990 ) Teaching in two tongues: Language alternatio n in foreign language classrooms. Hong Kong: City Polytechnic of Hong Kong. Long, M. H. (1983). Native speaker/non native speaker conversation and the negotiation of com prehensible input. Applied Linguistics, 4, 126 141. Long, M. H. (1985). Input and sec ond language acquisition theory. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition (pp. 377 393). Rowley, M A: Newburry House. Long, M. H. (1996). The role of the linguistics environment in second language acquisition. In W. C. Rit chie & T. K. Bhatia (Eds.) Handbook of language acquisition: Vol. 2. Second language acquisition (pp. 413 4 6 8) New York: Academic. Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms. S tudies in Second Language Acquisition 19 37 66. Macaro, E. (1997). Target language, collaborative learning, and autonomy Clevendon: Multilingual Matters. classroom: Theor ies and decision making. The Modern Language Journal, 85 (4), 531 548. Macdonald, C. (1993) Using the target language Cheltenham: Mary Glasgow. Myers Scotton, C. ( 1997 ) Code s witching. In F. Coulmas (Ed.), The handbook of s ociolinguistics (pp. 217 237) London: Blackwell. Merritt, M., Cleghorn, A., Abagi, J., & Bunyi, G. (1992). Socialising multilingualism: Determinants of codeswitching in Kenyan primary classrooms. Journal of Multilingual and Multicultural Development, 13 ( 1 2 ) 103 121. Myers Scotton C. (2006). Multiple voices: An introduction to bilingualism. Oxford: Blackwell. Pica, T. (1987). Second language acquisition, social interacti on, and the classroom. Applied Linguistics 8 (1), 3 21.

PAGE 116

11 6 Pilar, M. D., Mayo, G., & Pica, T. (2000). L2 learner interaction in a foreign language setting: Are learning needs addressed? IRAL, 38, 35 58. Poplack, S., & Meechan, M. (1998). How languages fit together in code mixing. International Journal of Bilingualism, 2 (2), 127 138. P ujolar, J. (2000). Gender, h eteroglossia and p ower. A soci olinguistic study of youth culture. Berlin: Walter de Gruyter. Qian, X., Tian G., & Wang Q ( 2009 ) Codeswitching in the primary EFL classroom in China T wo case studies. System 37 (4), 719 730. Rampton, B. (1995). Crossing: Language and ethnicity among adolescents. Lon don: Longman. Raschka, C., Sercombe, P., & Huang, C ( 2009 ) Conflicts and tensions in codeswitching in a Taiwanese EFL classroom. International Journal of Billingual Education & B ilingualism 12 ( 2 ), 157 171. Romaine, S. (1989). Bilingualism. Oxford: Basil Blackwell. Sankoff, D., & Poplack, S. (1981). A formal grammar for code switching. Papers in Linguistics : International Journal of Human Communication 14 (1), 3 46. Sato, K., & Kleinsasser, R. C. (1999). Communicative language teaching (CLT): Practical understandings. The Modern Language Journal, 83 (4), 494 517. Swain, M., & Lapkin, S. (2000). Task based second language learning: The uses of the first language. Language Teachi ng Research, 4 (3), 251 274. Thibault, P., & Vincent, D. ( 1990 ). Un corpus de fran ais parl. Montral: Recherches Sociolinguistiques Willis, J. ( 1981 ) Teaching English through English Harlow Essex : Longman.

PAGE 117

117 BIOGRAPHICAL SKETCH Yanmin Bao was b orn in Zhoushan, Zhejiang Province, China. She grew up mostly in Zhoushan, gradua ting from Zhoushan Senior High School in 200 5 In 2009, s he earned her B.A. in English from Harbin Normal University Her undergraduate thesis was awarded Excellent Graduation The sis in 2009. During her four years in college, she won Outstand ing Student Scholarship for every year. In addition, i n 2006, she was awarded Triple A student and was named the Best Athlete in annual college sports as well ; at the same year, s he was elected chairman of the students association Then in 2007, s he was awarded Excellent Student Cadre Yanmin entered the graduate program in the Linguistic D epartment at the University of Florida (UF) in 2009. She was interested in sociolinguistics, especially in codeswitching and gender and language From her TA work, as teaching Chinese at UF she found it would be interesting to analyze Mandarin English code switching behaviors in Chinese classes in an American university setting since few studies ha ve focused on this aspect, and fewer have investigate d to what extent teachers differ and resemble in the classroom codeswitching usage. She chose this as her M.A. thesis research topic and collected data from Beginning Chinese classes at UF after one ye in October 2011. Upon completion of her M.A. program, Yanmin will continue her linguistic study at UF and she also received the admission to the Ph.D program in 2011