Code-Switching by Chinese English-as-a-Second-Language Students in Computer-Mediated Communication: A Multiple Case Study Yanping Cui B.A. Hebei University, 1985 M.A. Hebei University, 2000 Thesis Submitted In Partial Fulfillment O f The Requirements For The Degree Of Master o f Education In Curriculum and Instruction The University O f Northern British Columbia July 2006 © Yanping Cui, 2006 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 W ellington Street Ottawa ON K1A 0N4 Canada 395, rue W ellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-28417-9 Our file Notre reference ISBN: 978-0-494-28417-9 NOTICE: The author has granted a non­ exclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or non­ commercial purposes, in microform, paper, electronic and/or any other formats. AVIS: L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'lnternet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. i*i Canada Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A PP R O V A L Name: Yanping Cui Degree: M aster o f Education Thesis Title: Code-Switching by Chinese English-as-a-Second-Language Students in Computer-M ediated Comm unication: A M ultiple Case Study Examining Committee: Chair: DrTRobert Tait Dean o f Graduate Studies U n iv ersity o f Northern British Colum bia Supervi&ftTDr. Judith Lapadat Professef, Education Program Mfversity o f Northern British Colum bia Comm ittee Member: Dr. Cofin Chasteauneuf Assistant Professor, Education Program U niversity o f Northern British Colum bia Committee Member: Dr. Han Li Associate Professor, Psychology Program U niversity o f Northern British Colum bia E te r n a l Examiner: Dr. M argaret A nderson Professor, First Nations Studies Program University o f N orthern British Colum bia Date Approved: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Abstract This study investigated the contexts of code-switching by Chinese ESL students who differed in age and study level and who lived in a small city in British Columbia. The aim o f the study was to examine the code-switching strategies that Chinese ESL students employ when engaged in computer-mediated communication, compare the similarities and differences o f these ESL students’ code-switching strategies, and explore the relevance of code-switching to English literacy development. Participants in this study comprised three Chinese ESL students who attended elementary school, college, and university respectively, and one parent. Interviews were conducted with each o f these participants. In addition, each student was videotaped, and electronic artifacts were collected from each o f them. Findings o f the study show that code-switching was deeply embedded in its contexts, and it functioned not only as a mechanism to compensate for deficient language, but also as a strategy to fulfill social, psychological or discursive purposes. Some o f the code-switching strategies were affected by age and English proficiency. Analysis o f the data also indicated that code-switching was beneficial to the English literacy development of these ESL students to varying degrees. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS Abstract..........................................................................................................................................ii Table o f Contents........................................................................................................................ iii List o f Tables...............................................................................................................................vi List o f Figures.............................................................................................................................vii Acknowledgements...................................................................................................................viii CHAPTER ONE: INTRODUCTION........................................................................................ 1 Overview.................................................................................................................................1 Establishing the Context o f the Study................................................................................ 4 Purpose and Significance o f the Study..............................................................................10 Summary.............................................................................................................................. 13 CHAPTER TWO: THEORETICAL FRAMEWORK & LITERATURE REVIEW 14 Theoretical framework........................................................................................................14 Second language acquisition theories........................................................................ 14 Interactionist Perspectives................................................................................16 Comprehensible Input Hypothesis...............................................................17 Comprehensible Output Hypothesis............................................................ 19 Conversation Analysis.......................................................................................21 Sociocultural Perspectives............................................................................... 23 Summary.............................................................................................................................. 30 Literature Review................................................................................................................ 31 ESL Students’ Home Literacy Practices........................................................................... 32 Computer-Mediated Communication and Its Application in Language Learning 36 Computer-Mediated Communication and Second Language Learning....................... 37 Equalization o f participation.....................................................................................37 Quality o f language production..................................................................................37 Linguistic skills.............................................................................................................38 Social dynamics........................................................................................................... 39 Identity construction...................................................................................................40 Code-switching.................................................................................................................... 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in Sociolinguistics............................................................................ 44 Code-switching in Second Language Acquisition.................................................... 48 Code-switching in Computer-mediated Communication.........................................51 Research Questions............................................................................................................ 54 Summary.............................................................................................................................. 55 CHAPTER THREE: RESEARCH DESIGN AND METHODS..........................................56 Introduction......................................................................................................................... 56 Research Design.................................................................................................................. 56 Participants.......................................................................................................................... 58 Data Collection....................................................................................................................62 Procedure..................................................................................................................... 64 Researcher Role and Reflexivity.......................................................................................67 Ethical Considerations....................................................................................................... 70 Overview o f Data Analysis Approach..............................................................................72 Data analysis software............................................................................................... 73 Steps in data analysis..................................................................................................73 Summary.............................................................................................................................. 77 CHAPTER FOUR: RESEARCH FINDINGS........................................................................79 Participants’ Profiles...........................................................................................................79 Amy................................................................................................................................79 Baillie............................................................................................................................81 Cynthia......................................................................................................................... 82 Emma............................................................................................................................ 84 Data Analysis.......................................................................................................................85 Writing the Chinese Language on Computers................................................................ 94 Means of Code-switching.................................................................................................. 96 Chinese characters......................................................................................................97 Pinyin............................................................................................................................99 Graphic......................................................................................................................100 Contexts o f code-switching.............................................................................................. 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V Lexical gap................................................................................................................. 103 Structural gap.............................................................................................................114 Identity construction.................................................................................................. 122 Effective communication...........................................................................................128 Rhetorical effect.........................................................................................................131 Topic change............................................................................................................... 137 Informal situation...................................................................................................... 140 English Language Development..................................................................................... 143 Vocabulary................................................................................................................. 143 Writing......................................................................................................................... 143 Speaking......................................................................................................................144 Listening Comprehension......................................................................................... 144 Grammar.....................................................................................................................145 Comparison o f Computer-related Activities...................................................................146 Comparison o f Code-switching Strategies..................................................................... 147 Summary........................................................................................................................... 150 CHAPTER FIVE: CONCLUSIONS...................................................................................... 152 Contributions o f the study................................................................................................ 154 Implications for Practice.................................................................................................. 155 Integration o f Computer-mediated Activities into Classroom Teaching.............155 Understanding o f Code-switching by ESL Teachers............................................. 158 Design o f Collaborative Online Learning Tasks....................................................160 Limitations o f the study.................................................................................................... 166 Suggestions for Further Research................................................................................... 168 Final Remarks....................................................................................................................170 REFERENCES.......................................................................................................................... 172 APPENDIX A: INTERVIEW QUESTIONS........................................................................ 188 APPENDIX B: LETTERS TO PARTICIPANTS..................................................................190 APPENDIX C: CONSENT FORMS..................................................................................... 194 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES 1. Coding Criteria o f Code-switching................................................................................. 76 2. Code Categories.................................................................................................................90 3. Contexts o f Code-switching............................................................................................. 91 4. Similarities and Differences in Code-switching............................................................92 5. Social Functions o f Code-switching................................................................................113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES 1. Process o f Data Analysis..................................................................................................74 2. Code-switching and English Language Development.................................................. 94 3. La MonaLibean................................................................................................................ 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Acknowledgements I would like to thank my supervisor, Dr. Judith Lapadat, for her academic and financial support, and knowledgeable and constructive advice. I appreciated her patience and dedication to this study. I would also like to thank my committee members, Dr. Colin Chateauneuf and Dr. Han Li, for their thoughtful questions and suggestions. Thank you also to Dr. Margaret Anderson for serving as a knowledgeable and experienced external examiner. Her feedback was necessary and appreciated. Finally, I would like to acknowledge the faithful support o f my family members towards the completion o f this degree. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 1 Chapter One: Introduction Overview Information and communication technologies have exerted a significant impact upon every aspect o f people’s lives. These technologies have the potential to change the way children learn and teachers teach. The widespread use o f computer technologies within the broader society has fundamentally altered the nature o f children’s literacy development (New London Group, 1996; Shetzer & Warschauer, 2000; Warschauer, 1998a). According to the New London Group (1996), traditional literacy or “mere literacy” has been centered on print literacy only; that is, the reading and writing o f print texts. However, with the introduction into and use o f new technologies and visual texts in people’s literacy practices at home and at school, students encounter, use and interpret multimodal texts and are becoming proficient in multiple kinds o f literacies (Lapadat, 2003). Jewitt and Kress (2003) describe multimodal texts as hybrid combinations o f a variety o f modes, including linguistic, visual, gestural, spatial and audio. Multimodal literacy is, therefore, defined as literacy that encompasses all the various modalities ways in which meaning can be created and communicated in the world today (Jewitt & Kress, 2003). Brown, Lapadat, McGregor, and Thielmann (2006) expand this definition o f multimodal literacy to include the ability to use “the multiplicity o f written and multimodal genres and digital technologies to interact within a variety o f communities o f practice both online and offline to accomplish a range o f work-related, educational, personal, recreational, and social purposes” (p. 12). The New London Group coined the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC term “multiliteracies” to refer to the different literacies of today that, in addition to print, take into account the representational resources (visual, audio, hypertext, etc.) afforded by computers and multimedia technologies. Snyder (2001) argues that print-based reading and writing is now only part of what people have to learn to be literate, and argues that literacy needs to be conceived within a broad social order, what she has called “a new communication order” (p. 117). This new order takes into account the literacy practices associated with screen-based technologies. Within a society packed with technological changes, children’s lives have changed, and their literacy practices have changed also. Thus, it is important to study the impact o f the use o f technology upon children’s literacy development. Lapadat (2003) notes in her research proposal1 that young people are leading the users’ edge o f technological change, and any new technological innovation is quickly taken up by children and youth, and the companies that market to them. However, many traditional social institutions like schools change slowly, and a mismatch exists between children’s actual literacy practices at home and classroom practices (see also Prensky, 2001). Furthermore, as many countries such as Canada become more culturally and linguistically diversified, so too are the countries’ school classrooms. Classrooms today are different from those ten years ago, and the nature o f literacy has changed in the information age. Educators are finding that traditional instructional approaches are ineffective in meeting the academic requirements o f many students, especially those from diverse cultures. For these reasons, it is necessary to look at children’s literacy 1 This present study is part o f a program funded by a Social Science and Humanities Research Council o f Canada Standard Research Grant #410-2004-1647, awarded to Judith C. Lapadat, Principal Investigator. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2 Code-switching in CMC practices at home associated with technologies in order to provide guidance for and insight into what pedagogical innovations should be made in order to bridge the gap between children’s actual literacy practices in the home and classroom practices. It is especially important to examine English-as-a-Second-Language (ESL) students’ home literacy practices because ESL students joining schools in English-speaking countries with inadequate knowledge o f English are confronted with the challenge o f learning English in order to integrate socially and academically in their new schools and communities. This study examines Chinese ESL students’ home literacy practices with a focus on computer-mediated communication (CMC), and the linguistic strategies these students employ when they have difficulty expressing themselves in English while engaged in CMC. Ultimately a comparison is made o f the contexts in which the Chinese ESL students o f different ages and at different study levels employ various linguistic strategies to cope with linguistic difficulties. By examining and comparing students’ use o f CMC at home and their linguistic strategies, I seek to obtain information that will help educators to accommodate ESL students’ learning using multimodal/multiliterate approaches. The following section establishes the context for this study by reviewing the emergence of multiliteracy, by presenting the mismatch between classroom practices and students’ home literacy practices, and by identifying the lack of research to date in Asian ESL students’ literacy practices. In particular, I review the limitations o f the studies that focused on ESL students’ home literacy practices. This brief review points to the rationale for and significance o f this study. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 Code-switching in CMC Establishing the Context o f the Study There is a mismatch between classroom literacy practices and children’s literacy practices in their lived lives. As discussed by Lapadat (2003), global social and technological changes have required education to change accordingly, whereas the funding needed for these educational changes is limited. Teachers and administrators face community demands to do more with less on the one hand and a back-to-basics backlash on the other. Although research findings reveal that inquiry-oriented processes for children’s construction o f knowledge are effective, policy makers attach importance to standards and accountability, and classroom practitioners are caught in the middle (Wells, 2001). Lapadat points out that in Canada, the teaching force is aging, with few young teachers. As well, limited professional development opportunities result in the teachers having less exposure to new ideas (see also Lankshear, Snyder, & Green, 2000). Inadequate political and economic support for teachers’ professional development and for changing the organizational structures o f schools to accommodate social-technological change are factors that pose obstacles to pedagogical innovations (Brown, Hartwick & Lapadat, 2005; Lankshear et al, 2000; Lapadat, 2003). As a result, many classrooms o f today—the way they are equipped and organized, and the curriculum and instructional approaches used— look relatively similar to the classrooms o f thirty years ago: one teacher for many students, desks in rows, textbooks, timetables, and the same familiar subjects (Lapadat, 2003). Bringing new technologies such as computers into schools has been a slow expensive process, and putting the technical, financial, curricular, and epistemological Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 Code-switching in CMC 5 supports in place to use the computers in meaningful ways has been an even greater struggle (Soderman, Gregory & O ’Neill, 1999; Lapadat, 2003). According to a Canadian national survey o f French-as-Second-Language (FSL) teachers conducted by Turbull and Lawrence (2002), a majority o f the FSL teachers (58%) at all levels across Canada are open to using computers in FSL classrooms. Nevertheless, o f those respondents reporting computer use, 61% used computers in a computer lab separate from their classrooms. A sizeable number o f teachers surveyed (41%) have never used computer technology in their classes. They cite lack o f access as well as a lack of training and knowledge about how to integrate computer technology into their teaching as the two principal reasons for this inexperience in using computer technology in their classrooms. In a similar study by Lam (2000b), second language (L2) teachers who did not use computers cited a lack o f knowledge as the primary reason. More specifically, many L2 teachers did not know how using computers could relate to language teaching. Others suggested a lack o f access or lack o f confidence as reasons for avoiding computer use. On the other hand, students, in the context of an electronic age with its affordances o f multimedia communication technologies, have taken up tools beyond the school setting to develop literacy, such as instant messaging, chat rooms, emails, and interactive web games via the Internet (Cruickshank, 2004; Lam, 2000a, 2004; Lapadat, 2002). They weave meanings they take from the multimodal representations into their own understanding and interactions within the world (Dyson, 2001; Lapadat, 2002). Prensky (2000) asserted that “Our students have changed radically. Today’s students are no longer the people our educational system was designed to teach” (p.l). Today’s Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC teenagers are described as digital natives, that is: “ ‘native speakers’ of the digital language o f computers, video games and the Internet” (p. 1), as opposed to teachers who are digital immigrants, “who speak an outdated language (that o f the pre-digital age), [and] are struggling to teach a population that speaks an entirely new language” (Prensky, 2000, p. 2). The gap therefore needs to be bridged between research findings, educational policy, classroom practices, and children’s lived lives (Lapadat, 2003; Wells, 2001). Wells (1986) explored children’s language and literacy development at home and school through a ten-year longitudinal study in Britain, and his work has profoundly influenced our current understandings about literacy development. However, the research was conducted prior to children’s widespread use o f videogames, the Internet, and other computer technologies in their daily lives, and children’s changing literacies are just beginning to be documented (Hull & Schultz, 2001). In particular, there is a dearth of research focused on literacy and technology among L2 learners in Canadian schools. There is a wealth of research on second language acquisition (SLA), and a separate body of work on CMC. Nevertheless, the examination o f CMC as used in SLA is limited in scope as well as in the number of actual studies conducted. O f the studies that have examined the use o f CMC in SLA, most have focused on CMC and its classroom application at the college level, and relatively little attention has been paid to L2 learners’ home literacy practices through CMC. Only a small number o f studies have examined L2 learners’ literacy practices at home. For example, Cruickshank (2004) explored the literacy practices in the home o f four Arabic-speaking teenagers and found Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 Code-switching in CMC that their literacy practices in multilingual contexts were carried out by means o f new technologies such as email, chat room, messaging, and the like. Lam (2000a, 2004) examined Chinese immigrant youths’ use o f the Internet at home achieving social identities. Studies such as these describe the social and cultural applications connected with technology-mediated literacy practices. In addition to the research mentioned above, Li (2004) suggested, in her study o f two Chinese elementary school students’ home literacy practices, that the computer is a significant part of the students’ after-school learning. She did not investigate, however, how they use computers to assist them in English language learning and what they learned from the use o f computers. By focusing on the ESL students’ general home literacy practices, these studies failed to take into account ESL students’ specific linguistic behaviours and what was actually acquired linguistically through these practices, especially through CMC. W hat’s more, the studies neglected the examination o f the implied social and cultural meanings related to the linguistic behaviour as factors influencing ESL students’ English language development. There is an emerging research literature in CMC that has examined different benefits brought by CMC to English literacy and language acquisition o f different second-language learners (Anderson-Inman & Homey, 1998; Labbo, Murray & Phillips 1995-96; McKenna, Reinking, Labbo & Kieffer, 1999; Gamer & Gillingham, 1998). These studies have addressed the impact o f affective, linguistic, social, and cultural factors on SLA. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 Code-switching in CMC 8 O f central interest, for the purpose o f my study, is the research into code-switching. Code refers to any distinct language or dialect (Boztepe, 2003). Gumperz (1982) defines code-switching (CS) as “the juxtaposition within the same speech exchange o f passages o f speech belonging to two different grammatical systems or subsystems” (p.59). That is, CS refers to a) alternating between two different styles of speech or dialects of the same language within the same conversation, as in the case of monolinguals switching from formal to informal speech, or switching from one dialect to another, and b) alternating between two or more languages within the same conversation, as in the case o f bilinguals or multilinguals switching from one language to another (Boztepe, 2003; Huerta-Macias & Quintero, 1992). Kotter (2003), by exploring the online chatting of intermediate L2 learners, found that code-switching promoted L2 literacy development. Hudson (1996) views code-switching as “the inevitable consequence o f bilingualism” (p. 51). However, code-switching has long been stigmatized in education (Lara, 1989; Ramirez & Milk, 1986). When bilingual children mix languages, educators question their academic abilities and potentials, and are more likely to place them in remedial classes (Lara, 1989). Recent research, however, has suggested that code-switching may also be viewed as an additional resource for bilingual speakers to achieve particular conversational purposes and support their learning needs (Flood, Lapp, Tinajero, & Hurley, 1996; Laliberty & Berzins, 2000). It may be that bilingual students use code-switching strategically for language acquisition. At the least, code-switching behavior provides a window into the state o f students’ linguistic knowledge and learning Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC processes. Nevertheless, this linguistic behavior among bilingual children has been the least systematically studied (Romaine, 1995) o f all the research on the effect o f mixed languages on bilingual children’s language development. As suggested in the literature (Au, 1995; Luke, 1995), students’ home literacy practices have an influence on their school success. Numerous studies have indicated that many students from ethnic minority groups are low-achieving in school, and this behaviour has been explained as due to being less motivated or disadvantaged. On the other hand, ESL students from some Asian countries such as China and South Korea are often high achievers in Canadian schools (Li, 2004). Even so, a large percentage of Asian English language learners are behind their native English speaking age/peer groups in their English language proficiency (Ashworth, 2001). Research studies that clearly and systematically describe Chinese ESL students’ home literacy practices involving the use o f technology are scarce. Specifically, few researchers (Lam, 2004) have examined Chinese ESL students’ use o f CMC at home, described their code-switching during CMC, or identified the sociocultural contexts and purposes o f code switching in CMC. I have located no studies that have compared the linguistic strategies employed by ESL students who differ in age and study levels in a CMC environment at home. Therefore, in my study, I will examine Chinese ESL students’ use of code-switching in their home literacy practices while using computer-mediated communication. In order to infer the social meanings and the relevance that code-switching has for their English language development, I will identify the contexts Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 Code-switching in CMC 10 in which code-switching occurs, and compare this linguistic behaviour o f Chinese ESL students o f different ages and at different study levels. Rationale fo r and significance o f the present study This study is important because an exploration o f the use o f technologies in the home literacy practices of Chinese ESL students is particularly needed and timely for several reasons. With the number o f Chinese immigrants to North America growing, the Chinese have become the single largest group o f Asians in Canada and the United States (Li, 2004). Chinese learners are often stereotyped as high achievers and therefore are overlooked in literacy research. In fact, however, Chinese students reportedly score higher on standardized tests than other minority groups such as Latinos (Li, 2004) in many districts with high Chinese concentrations (e.g., Vancouver, British Columbia), Chinese students’ English proficiency is relatively low compared to that of native-speakers o f English (Ashworth, 2001). Furthermore, over the last decade, a large number o f Chinese high school graduates have been enrolling in Canadian post-secondary educational institutions such as colleges to take ESL courses before pursuing their undergraduate degrees at Canadian universities. However, little information is available concerning what technologies this group o f students uses at home to assist them with their English language development. In addition, the writing system of the Chinese language and the English language have little in common. The Chinese language has a non-alphabetic writing system, which is unique in that written Chinese is composed o f ideographic Chinese characters, where one or two characters are equivalent to a word or an idea. In contrast, the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 11 graphemic symbols (letters) in the alphabetic writing system of English represent phonemes or speech sounds in spoken language. These systems o f writing have little in common. When Chinese ESL students develop their literacy with the aid of technologies, how and for what purposes do they code-switch when they have trouble expressing themselves in English? An understanding and a comparison of Chinese ESL students’ use o f technologies at home, with a focus on code-switching in particular, will contribute to the literacy literature theoretically as well as pedagogically. Theoretically, an examination of code-switching between written Chinese and written English will provide insights into the functional models of literacy these students have across two very different writing systems as well as the different affordances of each writing systems for accomplishing specific social purposes. Through this study, I also aim to bring insights into the process o f second language acquisition by describing the patterns of code-switching in computer-mediated communication among Chinese ESL students and the factors that influence their linguistic choices. An examination of code-switching behaviour among these students will likely uncover the linguistic, social and psychological factors involved in SLA. This research, therefore, has the potential to make a contribution to theories of second langauge acquisition. Findings from this research will have a practical application. The findings will add to our understanding o f how to tap into and capitalize on ESL students’ linguistic repertoire so that teachers can maximize students’ learning. The results from the study will open a window to ESL educators who are endeavoring pedagogically to accommodate Chinese ESL students as well as students from diverse cultures more Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 12 broadly. In particular, these findings may promote a change in perspective about the value o f developing bilingual students’ code-switching in language acquisition, given that in the past, code-switching has been viewed unfavourably by the mainstream society. ESL educators can obtain a better understanding o f how these students use both languages and learn through both languages, and can benefit from this information which illuminates a way to assess students’ learning and progress in language acquisition. With more students coming from different linguistic backgrounds, teachers are more likely to face students demonstrating code-switching behaviors. With such knowledge, teachers can become more sensitive to the process of acquiring a second language as an evolving process, and consequently design and implement more appropriate instructional strategies. With approriate strategies, English-as-a-Second-Language students who have access to the findings o f studies o f home literacy practices will realize that their linguistic behavior and its social and cultural meanings are related to their English literacy development, hence obtaining an understanding o f their English language learning, and building up their confidence. With the potential instructional support from the teachers informed by the study, developing bilingual students will be allowed to use their linguistic and cultural resources, which Laliberty and Berzins (2000) assert supports these students’ learning needs. Building on the students’ understanding o f their native languages can increase their chance for academic success (Flood et al., 1996). For me as a second language teacher as well as a researcher, I will gain from the results o f the study more theoretical knowledge of code-switching and a pragmatic Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 13 understanding o f its educational implications. Findings from this study will inform interested researchers, including me, o f future research areas. Summary This chapter has established this study’s perspective. The overview presents the general circumstances—the impact o f the use o f technology upon children and youths’ literacy development, and the gap between children’s actual literacy practices at home and classroom practices. Students’ changing literacies in an electronic age need to be documented in order for the home-classroom gap to be bridged. In particular, studies focused on literacy and technology, especially CMC, among ESL students in Canadian schools are scarce. Although a few studies (Cruickshank, 2004; Lam, 2000a, 2004) have examined general home literacy practices employing technology o f ESL students in the United States (U.S.) and Australia, they have failed to explore what was specifically acquired linguistically and socioculturally through these practices. Chinese ESL students, a group that is increasing in number in Canada, require special attention for a variety of reasons. They are high achievers in schools. However, their English proficiency is low compared to that o f their native English-speaking peers. Little information is available about Chinese ESL students’ home literacy practices regarding the use of the computer. A study o f this group o f student can begin to make up for this scarcity in the literature. Finally, the significance o f the study in both practical and theoretical realms is outlined. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 14 Chapter Two Literature Review Theoretical Framework Second language acquisition theories. Many theories regarding second language acquisition (SLA) have been proposed, but there has been little consensus on any single SLA theory. Second language acquisition theories have basically centered around either a nurture versus nature distinction or on a empiricism versus nativism distinction. The premise o f empiricism is that knowledge originates in the environment, but is understood through our interaction between the environment and our reasoning or senses. In contrast, nativism holds that at least some knowledge is not acquired through interaction with the environment, yet is genetically transmitted and innate. In other words, some theoreticians have based their theories on environmental factors while others believe innate factors determine the acquisition o f a second language. In the field o f SLA, empiricists are most associated in modem times with the psychologists, Jean Piaget, B. F. Skinner, and Lev Vygotsky, whereas nativists are most associated with the linguists Noam Chomsky and Stephen Krashen. It is worth noting that neither empiricists nor nativists reject entirely the theories o f the other. Only the weight empiricists and nativists ascribe to the environmental and innate factors varies, either relatively little or relatively more. For example, in the empiricist paradigm, environmental factors are believed to be more dominant in second language acquisition. In the nativist paradigm, Chomsky (1975) advanced the well-known universal grammar theory in which he defined universal grammar as the system o f abstract principles and rules that are shared by all languages, and viewed language as an innate, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 15 rule-governed system (Toohey, 2000). Rule refers to “knowledge that a given linguistic feature is used in a particular context with a particular function” (Ellis, 1997, p. 13). Chomsky’s theory, however, only accounted for how language is acquired through a system o f grammatical rules and constraints without taking into account communication, the primary function of language, thereby ignoring how sociocultural, socio-historical, and sociopolitical perspectives shape learners, and learning process (Watson-Gegeo, 2004). Whereas Chomsky’s theory focused on the constraints of grammar formation and the underlying grammatical rules common to all human languages, Krashen (1985) argued that language acquisition does not require extensive use of conscious grammatical rules and proposed a model, based on a psychological perspective, to account for the actual mechanisms involved in second language acquisition (Gass & Selinker, 2001). Mechanisms are defined as “devices that specify how cognitive functions operate on input” to drive development from one stage to the next (Long, 1990, p. 654). Krashen’s Monitor Model has become an influential theory in SLA. This model consists o f five hypotheses; (a) the Acquisition-Learning Hypothesis, (b) the Natural Order Hypothesis, (c) the Monitor Hypothesis, (d) the Input Hypothesis, and (e) the Affective Filter Hypothesis. The best known among the five hypotheses is Krashen’s Input Hypothesis (Warschauer, 1997). Krashen postulates that SLA almost wholly depends on the amount o f comprehensible input, that is, one-way input in the second language. Comprehensible input is referred to as i+ l, in which i represents learners’ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 16 current language level, and 1 represents the knowledge level that is beyond learners’ current knowledge. While Krashen (1994) maintains that comprehensible input is all that is needed for language acquisition to happen, others take an interactionist position that acknowledges the role of two-way communication. Interactionist Perspectives. Long (1985) and Pica (1994) emphasize the importance o f interaction in language learning and the necessity for learners to have access to meaningful and comprehensible input for interlanguage development to occur. Input refers to the language to which the learner is exposed (Gass & Selinker, 1994). Interlanguage is the type o f language produced by second- and foreign- language learners who are in the process o f learning a language. Selinker (1972) coined this term to refer to the systematic knowledge of the target language independent o f both the learner’s first language (LI ) and L2. Hammers and Blanc (1990) suggest that between the choice o f the first language or the second language, there exists a range of intermediary strategies that include the modification o f either language and the relative use o f both. By a gradual process o f trial and error and hypothesis testing, learners slowly succeed in establishing close approximations to the system of the target language. When learners engage in meaningful activities with their interlocutors, they are compelled to negotiate fo r meaning in order to arrive at a mutual understanding for comprehensible input, test hypotheses related to their developing interlanguage system and have access to feedback related to their output (Long, 1985; Swain, 1985; Swain & Lapkin, 1995). Output refers to language production, either spoken or written (Swain, 1985, 1993), or the productive use o f language (Gass, 1997). To negotiate meaning, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 17 interlocutors must “express and clarify their intentions, thoughts, opinions, etc., in a way which permits them to arrive at a mutual understanding”(Lightbrown & Spada, 1999, p. 122). Negotiation o f meaning is also defined by Pica (1994) as the “modification and restructuring that occurs when learners and their interlocutors anticipate, perceive, or experience difficulties in message comprehensibility” (p.495). Gass (1997) suggests that learners must first focus on their own linguistic deficiencies vis-a-vis the target language— “noticing the gap" (p. 4). In other words, L2 learners must develop their own metalinguistic awareness in order to stimulate a change in their interlanguage (Schmidt, 1990). In sum, interactionist perspectives assert that interaction results in negotiation of meaning, which is acknowledged as a facilitator o f learning (Gass, 1997). Negotiation draws learners’ attention to erroneous or inappropriate forms of their utterances, and creates a situation in which learners receive feedback directly and indirectly from their interlocutors and consequently produce output, thereby facilitating second language acquisition (Gass, 1997; Long, 1985; Pica,1994; Swain, 1985).The following sections will explain the different aspects of interactionist perspectives. Comprehensible Input Hypothesis. Long (1985) builds upon and extends the Input Hypothesis proposed by Krashen (1985). Long claims that comprehensible input is necessary yet not sufficient, and believes in the role o f interaction in generating comprehensible input. According to Long, input shaped through interaction contributes directly and powerfully to acquisition, and what makes input comprehensible is the negotiation o f meaning or the modification of the interaction when a comprehension problem occurs. Modified interaction is partly accomplished through the conversational Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 18 repair moves o f negotiation o f meaning, including linguistic simplification as well as conversational modifications (Ariza & Hancock, 2003) such as clarification, confirmation checks, self repetition, recasts, explicit corrections (Jepson, 2005) or code-switching (Kotter, 2003). The resultant interactionally modified input that learners are exposed to and the way in which other speakers interact in conversations with learners is a crucial element in the language acquisition process (Lightbown & Spada, 1993), and optimizes SLA (Gass, 1997), because in the course o f modified interaction or negotiation o f meaning, learners are provided opportunities to focus on the linguistic forms o f their utterances, making linguistic forms more salient and hence creating a readiness for learning (Gass, 1997; Pica, Holliday, Lewis & Morgenthaler, 1989). The study by Pica et al. (1989) revealed that when the L2 learners show difficulty in understanding, the native speakers provide linguistic modifications in the input such as repeating and reformulating their original utterances for the learners. The learners concurrently make reformulations o f their own interlanguage utterance based on the feedback that they receive. Nevertheless, Varonis and Gass (1985) note that leamer-leamer interaction between two non-native speakers (NNS-NNS) results in more negotiation o f meaning than does an interaction between a native speaker and non-native speaker (NS-NNS), because NNS feel freer with one another to indicate non-comprehension and negotiate for meaning. What is worth noting is that when learners modify an original utterance for greater comprehensibility, they adjust its form, or in other words, they are pushed to form comprehensible output. The attention to language forms and structures is Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 19 hypothesized as conducive to interlanguage development (Pica et al. 1989). Given that Long’s comprehensible input hypothesis puts emphasis on input through interaction, Swain (1995) claims that comprehensible input is inadequate in facilitating second language acquisition, and proposes her comprehensible output hypothesis. Comprehensible Output Hypothesis. The output hypothesis proposes that language acquisition may occur through producing language (Swain, 1985, 1993). Swain emphasizes the crucial role o f comprehensible output or language production, either spoken or written, in L2 development. She argues that apart from comprehensible input, output is also crucial. The output, Swain (1985) hypothesizes, serves four primary functions in SLA, that is to, (a) enhance fluency, (b) create awareness o f language knowledge gaps, (c) provide opportunities to experiment with language forms and structures, and (d) obtain feedback from others about language use. She argues that “ [production/output]... may force the learner to move from semantic processing to syntactic processing” (p. 249). Gass (1997) suggested that no serious speculation was placed on the idea that output or language use could be part o f the learning mechanism prior to Swain's (1985) important paper, in which she introduced the notion of comprehensible or pushed output; that is, “learners are pushed or stretched in their production as a necessary part o f making themselves understood. In so doing, they might modify a previous utterance or they might try forms that they hadn't used before.” (p.249). According to Ellis (1990), “production will aid acquisition only when the learner is pushed. Opportunities to speak may not in themselves be sufficient” (p.l 18). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 20 Schmidt (1990) highlights the importance of noticing the linguistic gap. Noticing the gap or apperception occurs when learners consciously focus on their own linguistic deficiency or problematic aspects in the target language (Gass, 1997). When learners experience communication breakdowns, they are forced to attend to the L2 forms before modifying their output to make it more comprehensible. Learners’ focus on the linguistic features o f the target language and meaningful use o f their linguistic resources raises learners’ awareness o f their linguistic deficiency, which is a prerequisite for the restructuring of learners’ linguistic knowledge (Gass & Varonis, 1994; Swain & Lapkin, 1995). Noticing a problem pushes learners into a more syntactic-processing mode (Swain & Lapkin, 1995) that will help them to intemilize new forms (Pica et al., 1989) and improve the accuracy o f their existing grammatical knowledge (Nobuyoshi & Ellis, 1993). This facilitation o f SLA, however, does not seem to be acquired by comprehensible input alone (Swain, 1995). Comprehensible output assists learners in conveying meanings while providing linguistic challenges. In producing the L2, learners will become aware of a linguistic problem brought to their attention by feedback provided to them by the interlocutor that can be used to revise their interlanguage (Ellis, 1997). Noticing a problem pushes learners to modify their output. In so doing, learners may sometimes be forced into a more syntactic-processing mode than might occur in comprehension (Swain & Lapkin, 1995). The Interaction Hypotheses stress the importance o f negotiation o f meaning in facilitating second language development. Negotiation of meaning arises when Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 21 interlocutors seek to prevent a communicative breakdown from occurring or to remedy a breakdown that has already occurred (Ellis, 1999). In other words, negotiation of meaning results in interactional modifications that help learners to understand each other. Language acquisition is enhanced in the sense that input is made more comprehensible through input modification, feedback is provided, output is elicited, and learners are forced to focus on certain aspects o f their speech (Long, 1996; Swain, 1985). According to the interactionist approach, conversational interaction forms the basis for language development (Long, 1996). This is similar to conversation analysis (CA) in that the central premise o f CA is that CA focuses on analyzing talk-in-interaction. Conversation Analysis. Conversation analysis is defined as the study o f “the social organization of ‘conversation,’ or ‘talk-in-interaction,’ by a detailed inspection o f tape recordings and transcripts” (ten Have, 1997, p. 1). It represents both a theoretical stance and a methodology (Lapadat, 1999; Psathas, 1995) to analyzing naturalistic talk (Lapadat, 1999). The CA method reflects conversational analysts' theories about the nature of conversation and how the conversation can be empirically studied to reveal participants' strategies for expressing themselves, reaching understandings, repairing misunderstandings, and so forth (Lapadat, 1999). The central concern o f CA is with the nature o f turn-taking in talk-in-interaction (Hutchby & Wooffitt, 1998), such as: (a) how is naturally occurring talk organized, (b) how do interlocutors achieve orderly turn-taking, and (c) what are the systematic resources used by interlocutors in the achievement? According to Markee (2000), CA seeks to uncover sequential organization, turn-taking, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 22 and repair o f talk-in-interaction, thereby making it most suitable for analysis o f linguistic details. The detailed linguistic approach to analysis in CA, and the focus o f interest (i.e., conversation and how it is achieved and maintained), is similar to my focus o f interest in this study, especially when it comes to conversational repair. I have chosen, however, not to ground my study theoretically in the CA approach for lack o f a good fit. First, CA rejects preformulated theoretical categories and testing hypothesis, because the interest o f CA is in discovering “the structures, the machinery, the organized practices, the formal procedures, the ways in which order is produced” (Psathas, 1995, p. 3). Therefore, CA approach is a heuristic-inductive, hypothesis-generating approach, and questions are formulated through the analysis o f the data. I have, however, already formulated the research questions in my mind based on personal experiences and the literature search, and therefore have provided a theoretical framework. Second, CA focuses on sequential organization, turn-taking and repair o f the talk-in-action. Many of the rules and procedures in CA do not transfer well to the CMC environment. For example, online chat turns are not sequential in that the absence of simultaneous feedback may result in discontinuity and/or overlap within turn sequences. As a result, the communication in online chat is disrupted and discontinuous: many different topics can be carried out simultaneously, often by the same speaker. A speaker can receive multiple responses to different previous turns and use the same turn to post simultaneously several messages in response to different topics. Given the underpinnings o f conversation analysis, CA has not been much used in the literature o f CMC. The interactionist approach to SLA, however, has been applied to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 23 computer-mediated communication as a theoretical framework, in that negotiation of meaning frequently occurs in CMC between non-native-speakers (NNS), and between native-speakers (NS) and non-native-speakers (Blake, 2000; Pelletieri, 2000; Smith, 2003; Toyoda & Harrison, 2002). Blake proposes strong possibilities for using CMC in language learning because with the CMC medium, L2 learners heighten their metalinguistic awareness o f where they are in their own L2 vocabulary development and where they still need to go in order to gain more target-like lexical control. Doing tasks in a CMC environment, then, generates apperceived input, which can subsequently be used to modify and improve their vocabulary (p. 131). By examining the comprehensible input and output hypotheses discussed above, it can be seen that Krashen’s model regards language as a natural phenomenon, and emphasizes individual characteristics and cognitive processes over the wider socio-historical contexts that shape language acquisition. Although Long and Swain’s models stress the joint contribution o f the linguistic environment and learners’ internal mechanism in language acquisition, social and cultural contexts that inextricably shape language are overlooked. A more encompassing conceptual framework is necessary— the sociocultural perspective, advanced from the psycholinguistic insights o f Vygotsky and based on the nurturist ideas. Sociocultural perspectives on language learning provide a complementary position that considers language learners in direct relation to their social and cultural surroundings where participants engage in activities co-constructed with other individuals (Lantolf & Pavlenko, 1995). Sociocultural Perspectives. Socio-cultural perspectives of learning became popularized in North America in the 1990s, and have become ubiquitous in SLA Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 24 research ever since Lantolf and Appel (1994) brought together the work of scholars who were attempting to extend Vygotsky's theory to second language research. Sociocultural perspectives can be traced back to the writings of Vygotsky, a Russian cognitive psychologist. Vygotsky's theoretical framework is based on the assumption that human activities take place in cultural contexts and are mediated by language and other symbol systems. Mediation is the mechanism through which external, socio-cultural activities are transformed into internal, mental functioning. Mediation is the instrument of cognitive change (Kozulin, 1990). In describing sociocultural influences upon human mental activity, Vygotsky (1981) argued: Any function in the child’s cultural development appears.. .on two planes. First it appears on the social plane, and then on the psychological plane. First it appears between people as an interpsychological category, and then within the child as an intrapsychological category, (p. 163) Neither the interpsychological nor the intrapsychological function exists as a separate entity; they are closely connected (Wertsch, 1991a). Throughout the mental developmental process, language is viewed as the primary symbolic mediating tool between social interaction and the development o f higher order mental processes (Mitchell & Myles, 1998). The tools and signs to which one is exposed, therefore, influence or mediate new patterns of thought and mental functioning (Wertsch, 1991a). Tools can be both material and psychological in mediating human activities, and therefore human learning (Chandler-Olcott & Mahar, 2004). Vygotsky (1934/1978) distinguishes between material tools that work on the object and psychological tools that mediate the relationship to the environment, action and thought. Material tools involve Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 25 textbooks, notebooks, computers, and so forth. Sign systems, such as language, diagrams, arithmetic, formulae, or the behavior o f another human being in social interaction, are psychological tools for human activities. Human beings use these tools to mediate relationships with themselves, others, objects and the world. In this way, Vygotsky (1934/1978) linked society to mind through mediation, and viewed the acquisition o f language as a sociocultural phenomenon, linking the social with the cognitive. According to Wertsch (1991a, 1991b), sociocultural competence and learning, or what he called new mental functions and patterns o f thought, are gained from the mediational assistance o f tools, signs and scaffolding when it is offered within one’s zone o f proximal development (ZPD). Vygotsky (1934/1978) defined the ZPD as “the distance between the actual developmental level as determined by independent problem solving and the level o f potential development as determined through problem solving under adult guidance or in collaboration with more capable peers” (p.86). This idea emphasizes that an individual develops higher cognitive levels when the gaps in his/her thinking and problem solving are supported by adults, peers, or more capable others. This support is called scaffolding, the graduated assistance provided to an individual by more capable others. Through graduated assistance, the individual can internalize external knowledge and eventually reach the stage o f consciously controlling the knowledge in order to accomplish learning tasks (Wood, Bruner & Ross, 1976). The social environment supports development in such a way that what can be done collaboratively now will be accomplished independently at a later date. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 26 Wells (1999) points out that the ZPD is formed not just with an individual learner, but rather in the interaction between the learner, coparticipants, and available tools during involvement in a common activity. The joint activity requires the novice to construct with the expert a shared perspective or intersubjectivity (Rommetveit,1985; cited in Anton & DiCammilla, 1998). According to Wertsch (1985), intersubjectivity is achieved when “interlocutors share some aspect of their situation definitions” (p.59). In other words, learners must share some kind of communicative contexts within the L2 learning environment. Darhower (2002) notes that the context may be shared if interlocutors have similar background knowledge about one topic. It is this common communicative context that enables learners to cooperate and interact (Vygotsky, 1962). Hence, it is critical for an individual to participate actively in the learning process for learning to occur. Sociocultural perspectives emphasize the interdependence o f social and individual processes in the co-construction o f knowledge. Sociocultural theorists postulate that language is both a social and cognitive development phenomenon rather than a private series o f activities that occurs solely in the head. Through social interactions, children come to internalize and gain performance competence in the socioculturally defined context (Vygotsky, 1934/1978). Sociocultural theorists assert that social and cultural contexts are central to the process o f identity formation, and identity formation is an ongoing process o f learning and change. According to Wenger (1998), Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 27 building an identity consists o f negotiating the meanings of our experience of membership in social communities. The concept o f identity serves as a pivot between the social and the individual, so that each can be talked about in terms of the other (p. 145). W enger’s (1998) social theory o f learning combines the concepts o f active engagement and identity formation as mutually inclusive within social communities (Brown et al., 2006; Lapadat, 2003). Meanwhile, Wenger holds that using a tool when engaged in an activity changes the nature of that activity, and participating in the changed activity thus changes the members of the community (Chandler-Olcott & Mahar, 2003). Activity theory, which has its roots in Vygotsky, also emphasizes the role played by tools, both material and psychological, in mediating human learning. Tool use affects a number o f components within a community o f practice, defined by Wenger as a community “created over time by the sustained pursuit of a shared enterprise” (p.45). The components include the purpose for participating in the activity; the rules for participating; identities participants define and construct through the activity, and so forth. Consequently, tools are used to “enact and maintain ‘socially situated identities’ ” within the joint activity (Chandler-Olcott & Mahar, 2003, p. 362). Sociocultural perspectives have been gaining increasing influence in SLA research in that they complement the inadequacies of nativist and mechanistic theories o f SLA and attempt to erase boundaries between language learning and language use and between learning and developmental processes. The sociocultural approach to SLA is now applied to the field o f CMC. CMC involves a mediational process that is simultaneously undertaken through material tools (computer/software), and psychological tools (language/texts) (Freire, 1994). As far as second language learning is Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 28 concerned, computers mediate L2 learners’ literacy practices on the one hand and language mediates their higher mental functioning, resulting in new learning on the other. In a CMC scenario where communication is basically undertaken through language, L2 learners are engaged in social interaction with their language partners, either native-speakers or non-native speakers, via email, online chat, or audio-video conferencing. Through the interaction, a group can construct knowledge together, “thus linking reflection and interaction” (Warschauer, 1997, p. 437). Computer-mediated communication such as online discussions and asynchronous conferencing is viewed as a desirable mode o f communication and believed to heighten interactivity between learners (Herring, 1999; Lapadat, 2002; Nguyen & Kellogg, 2005). In an online environment where a visual record can be obtained in the form o f text, negotiation of meaning is facilitated and collective conceptualization o f ideas is enhanced (Herring, 1999; Lapadat, 2002; Nguyen & Kellogg, 2005). During online interaction, learners collaboratively establish coherence o f their communication by conversational repair moves (Herring, 1999). Computer-mediated communication has been theorized as a crucial means to realize interactivity in online teaching and learning environments and promote a “learner-centered interactive approach,” and to lead to the awareness o f the sociocultural nature o f the target language and the development of situated cognition (Romiszowski & Mason, 1996; Salaberry, 2001). Computer-mediated communication as a mediating tool supports evolving social and cultural communities, in which identity construction takes place within the context o f practice or joint activity (Brown et al., 2006; Chandler-Olcott & Mahar, 2003; Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 29 Nguyen, & Kellogg, 2005). Spears and Lea (1992) argue that in networked learning environments social identity stems from one’s presentation o f identity as part o f a group or in assuming a social role within the interaction. They further assert that in an environment where there is no visual feedback, one is more self-aware than in face-to-face settings. In a similar vein, Yates (1997) views online identity construction as something that occurs through discourse, and claims that all texts carry markers that can identify their authors in different ways. Hodgson (2002) suggests that an examination o f the use o f language in networked learning environments can reveal how the process o f identity-construction works through dialogue. Similarly, Beach (2000) points out that activity-theory-influenced perspectives can guide research that examines how readers, viewers, and composers o f media texts “construct their identities within the competing contexts o f their own immediate, local activity, as well as within the larger context o f a virtual media world” (p. 13). My understanding and analysis of this study reflects both the interactionist and sociocultural perspectives on SLA. It is not sufficient to choose one o f the two approaches to SLA. The interactionist approach allows me to pay particular attention to the nature o f the interactions L2 learners typically engage in. It focuses on investigating, for example, the role o f negotiation for meaning in the contexts o f both NS-NNS and NNS-NNS conversations, and the role of comprehensible input and output in SLA. Using this approach to explore the development or the use o f specific linguistic features o f L2,1 can address the issue o f L2 learning. The sociocultural approach, on the other hand, enables me to address the issue o f SL use; that is, the interactive, collaborative, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 30 and socially situated nature of L2 use and learning. These two approaches complement each other in that both the linguistic and the sociocultural features o f SLA are investigated, hence holistically unveiling the process o f language learning and use. I attempt to apply the two theories to different components of analysis in turn. The possibility also exists that some components in the data could be explained by one theory only. The integration o f both approaches thus does not create theoretical conflicts that can affect my approach to analysis and interpretation of the data. Instead, the integration increases the likelihood of understanding a language phenomenon through different lenses given the complexity of language learning and use. Summary The comprehensible input hypothesis claims that negotiation o f meaning occurs when interlocutors interact in an attempt to avoid or repair communicative breakdowns, and it is facilitative o f SLA. During negotiation for meaning, input is modified and made comprehensible. The comprehensible output hypothesis argues that comprehensible input is not sufficient for second language learning to occur, and therefore stresses the importance o f output in promoting SLA. In the L2 production, learners can notice their linguistic problems which in turn force learners to modify their output. The interactionist approach has been applied to CMC as a theoretical framework for the reason that learning activities in CMC environments promote negotiation of meaning which is believed to enhance SLA. While the interactionist approach attaches importance to the interaction between the learners’ cognitive mechanisms and the linguistic environment in facilitating SLA, it has failed in examining social and cultural contexts by which Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 31 language is bound. Sociocultural perspectives of language learning provide a complementary position for interaction hypotheses. Sociocultural perspectives view language as the primary symbolic mediating tool between social interaction and the development o f higher order mental process. Language acquisition occurs when learners co-construct knowledge with their peers or more capable others through mediating assistance o f tools, signs and human scalffolding. While the learner and his/her coparticipants engage in joint activities, they construct through mediation not only a shared perspective, or intersubjectivity, but also an identity within the community o f practice. In CMC contexts, L2 learners actively participate in learning activities and assume their social roles within the interaction. My study is grounded within both interactionist and sociocultural perspectives. Literature Review This section reviews the relationship between ESL students’ home literacy practices and academic success, CMC and its application in language learning, studies of CMC and second language learning, and studies of CS. The first area of the literature review highlights the importance o f integration o f ESL students’ home literacy practices into school literacy. The computer, as an integral part o f ESL students’ after-school activities, plays a significant role in their literacy development. CMC has started to attract researchers’ and educators’ attention to the application o f CMC to second language teaching and learning. Therefore, there is an emerging literature in studies of CMC. Code-switching, a widely observed behaviour among ESL students, appears not only in their daily conversation and classroom interaction, but also when they Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 32 communicate in English in online environments. A review o f the literature in CS is important in establishing the backdrop o f this study. ESL students ’ home literacy practices Two major theories have been suggested as possible explanations for schools’ failure to ensure academic success for students from minority backgrounds (Caimey & Ruge, 1998). The first is the theory o f cultural discontinuity (Heath, 1983), and the second is the theory o f structural inequality (Au, 1993). The theory o f cultural discontinuity argues that cultural mismatches, including differences in language, literacy beliefs, and interactional patterns between teachers and students result in difficulties in communication and interaction in the classroom, and consequently school failure (Gee, 1996; Heath 1983; Jacob & Jordan, 1993; Karanja, 2005; Ogbu, 1982). The home culture of students from minority groups is often described as traditional and static (Heath & McLaughlin, 1993). These cultural mismatches negatively affect the literacy learning o f students whose home culture does not reflect that o f the school. Some researchers have identified the difference between the language and literacy of school and that o f homes and communities as a significant factor in the achievement or non-achievement o f students at school (Foster, 1992; Heath, 1983; Scribner & Cole, 1981). As familiarity with school literacy is a mark o f school success, students from minority cultural backgrounds have to learn a different set o f literacy conventions and often experience difficulties with schooling (Lopez, 1999; McCarthey, 1997). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 33 The theory o f structural inequality looks beyond mismatches between the culture o f the home and the school (Caimey & Ruge, 1998). It argues that minority students' failure at school reflects structural inequalities in the broader social, political and economic spheres (Ogbu, 1987; Au, 1993). This theory takes into account the power relationships between groups, and holds that schools function to maintain the status quo. Au (1995) attributed minority students’ educational disadvantage to “oppressive social structures that create vast inequalities in power and opportunity favouring the dominant group” (p. 87). In a similar vein, Luke (1995) argued that schools “naturalise particular interactional patterns and textual practices in ways that systematically exclude those students from economically marginal and culturally different backgrounds” (p. 16). In the literature, many studies attribute minority students’ low achievement to their parents, and a lack of reading materials in the home (Blackledge, 2000; Delgado-Gaitan, 1990; Vincent, 1996). However, other studies have come to a different conclusion. For example, Saxena (1994) and Taylor (1997) observed that a school usually defines “books” as novels in English read for enjoyment, thus excluding the range o f reading materials in various languages in the home. With regard to minority students’ culture, Heath and McLaughlin, (1993) noted that schools often define students and their families according to their ethnic background, which is viewed as static and traditional. However, ethnic minority groups are undergoing dramatic social and cultural changes (Cruickshank, 2004), and “tangled cultural experiences” (Clifford, 1997, p. 136) typify migrant families in modem society (Inda & Rosaldo, 2002). Caimey and Ruge’s (1998) study on Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 34 minority students indicated that there is no single or simple answer to the question of how matches and mismatches in home and school literacy practices impact on students’ school success. In their study, the most academically successful students were those whose family literacy practices reproduced school literacy practices, whereas the dominant home literacy practices o f those less academically successful students failed in reproducing school literacy practices. In my view, whether family environments, languages, or the children themselves contribute to minority students’ academic success is a debatable issue. The implicit assumption of Caimey and Ruge’s (1998) study that school literacy is the best or most complete view of literacy needs questioning. Street (1995) raises the concern that non-school literacies are viewed as inferior and should be remediated for by enhanced schooling. I would argue that this view has become obsolete in an increasingly diversified and technologically advancing society. To ensure minority students’ academic success, school literacy should extend beyond its restricted experiences and attempt to respond to the cultural and linguistic diversity o f its community, especially in the information age when literacies are changing and evolving into multiliteracies. Studies have indicated that students’ literacy practices in multilingual contexts are dynamic and develop rapidly in response to technological and cultural changes (Cruickshank, 2004; Lam, 2000a, 2004). For example, in Cruickshank’s examination o f the literacy practices of four Arabic-speaking teenagers, he found that the four teenagers were taking an active role in a shift to technology-mediated literacy and in the mediation o f both Arabic and English Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 35 literacy. He observed that these teenagers had taken up new technologies such as email, chat room, messaging, and so forth for very specific purposes, such as to create a network o f friendship that was immediate but linked across space and time. Lam’s studies (2000a, 2004) of Chinese immigrant youth show how they come to occupy new social positions and identities by acquiring and appropriating new discourses in online environments. In her studies, Lam has considered not only how social contexts shape language use in online environment, but also how online communication shapes social contexts and the participants’ identity formation. If schools are to be adjusted to meet the needs o f students from diverse social, cultural and language backgrounds and to take advantage o f the increasingly electronic media for literate action and interaction, then a necessary starting point is to develop an understanding o f students’ home literacy practices. In particular, researchers need to examine home literacy practices using CMC, an increasingly popular communication and language learning form in modem society. Accordingly, school literacy practices can build on, rather than replace, students’ home literacy practices. Computer-mediated communication and its application in language learning Traditionally, computer-assisted language learning is associated with self-contained programmed applications such as drills, simulations, tutorials, instructional games and tests. With the advent o f new technologies, computers have been connected to one another in either local or global networks, and the emphasis has shifted to the role o f computers in facilitating or mediating human communication. Computers have fundamentally altered the way people communicate and have become a significant Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 36 element in the teaching and learning environment. Warschauer (1998a) highlights the incorporation of computer technology in language learning, and asserts that computer technology is no longer just a possible tool but an essential new medium o f language and literacy practices alongside face-to-face communication and the printed page. One aspect o f computer use is computer-mediated communication (CMC). Computer-mediated communication refers to the use o f online, interpersonal communication via the Internet or specific local area network (LAN) software programs (Warschauer, 1997). It means that students can communicate with each other, using a computer as the means o f communication. Interactive CMC forms are divided into two main types— synchronous communication and asynchronous communication (Lapadat, 2002; Levy, 1997). Synchronous communication takes place as a rapid or real time exchange, for example, text or voice online chat, MSN messenger, whereas asynchronous communication is protracted or has a delayed message system, free of time constraints (e.g., e-mail) (Warschauer & Healey 1998). Both communication forms underscore learning as a collaborative act that happens in a social and political context, with students and teacher working jointly in the networked interaction (Lapadat, 2002; Ortega, 1997). As the role o f computers has shifted from a “computer as tutor” to a resource and tool for learning, asynchronous and synchronous CMC have created a major shift in how teachers and students think about teaching and learning. As noted by Warschauer and Meskill (2000), language teachers throughout the US have been employing CMC in language classrooms in various forms, and this is particularly tme in higher education Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 37 where students and teachers have greater access to computer laboratories and Internet accounts than in K-12 schools. Meanwhile students have been using CMC forms, such as MSN, online chat and email as literacy practices at home. There is an increasing amount of information regarding the practices o f CMC used in language learning environments. Computer-mediated communication and second language learning Equalization o f participation. A dramatic increase has been reported in participation in CMC among postsecondary L2 learners in formal second language programs. The findings of the studies by Beauvois (1992), Kelm (1992), and Kern (1995) illustrate a greater degree of student participation in CMC contexts than in face-to-face (F2F) contexts and higher percentage o f student talk versus teacher talk. Kelm and Beauvois report increases in the participation patterns o f shy, low-motivated and unsuccessful students, who were perceived by their instructors as less willing to participate in teacher-led discussions. Similarly, a study by Kem reveals that electronic discussions engender a radical change in the proportion o f student versus teacher language production— students have more turns and produce more words and sentences in electronic discussions than in oral discussions. Quality o f language production. Studies o f online language interaction show that the language produced in CMC contexts by students in second language acquisition programs is qualitatively better and more coherent and cohesive than the language produced in face-to-face (F2F) classroom communication. Chun (1994) reports that students o f German tend to produce and initiate language that covers a wide range o f Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 38 communicative and discourse functions. Kem (1995) points out that the overall students’ language production in her French language class reflects a higher level o f sophistication than in oral discussions. Warschauer (1996) conducted a similar study and found that his students o f German produce language that was lexically and syntactically more complex. Lapadat’s (2002) study on online written interaction reports that written interaction in asynchronous CMC contexts scaffolds more sophisticated language use, because as participants strive to formulate their ideas and thoughts in writing by the use o f a written medium, they tend to use critical and higher order thinking processes such as analysis, synthesis, and evaluation. Although her study is conducted in first language contexts, its findings also apply to second language contexts in the sense that interactivity o f asynchronous CMC allows all users, including L2 learners, to take their time, reflect, consider their audience’s perspectives, and utilize higher order thinking skills (Lapadat, 2002). Linguistic skills. Substantial evidence has been presented in support o f the role of synchronous and asynchronous CMC in facilitating the acquisition of linguistic skills in a second language. A study by Warschauer (1999) suggests that computer-assisted discussion enables students to notice the input better from others’ messages and integrate that input into their own messages, therefore increasing opportunities for learning o f new linguistic elements together, such as collocations and common phrases. In a similar study (Blake, 2000) o f networked discussions using the synchronous chat programme, it was reported that CMC stimulated L2 learners to negotiate meaning, which generated language modifications and a focus on linguistic forms as L2 learners exchange Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 39 information and repair communication breakdowns (see also Kitade, 2000). Lexical confusions triggered most of the negotiations between students, as reported by Blake, rather than morphological or syntactical ones. A study by Varonis and Gass (1985) shows that in L2 contexts, leamer-leamer interaction involving Non-Native Speakers (NNS)-NNS dyadic interactions result in more meaning negotiating exchanges than Native Speakers (NS)-NNS interaction. They argued that NNS feel freer with one another to indicate non-comprehension and negotiate for meaning. A case study by St. John and Cash (1995) o f a high-intermediate learner o f German reveals that the learner systematically studied the vocabulary and phrases in his email exchange with a native German speaker and subsequently incorporated the new vocabulary and the phrases into his own writing. Wang (1998) compared dialogue journals written with paper and pencil by one group o f ESL students and transmitted over e-mail by a second group. She found that the email group communicated more frequently by asking and responding to more questions, and used a greater variety o f language functions than did the paper-and-pencil group. The more frequent communication generated by the email group suggests that CMC affords greater interactivity. These studies have shown that CMC promotes negotiation o f meaning from which comprehensible input and modified output result. Negotiation o f meaning has been identified as a crucial element in facilitating SLA (Gass, 1997; Long, 1985; Pica, 1994; Swain, 1985). Social dynamics. Researchers have reported that language learners express a positive attitude towards computer use when they engage in language learning tasks. According to some researchers (Kem,1995; see also Warschauer, 1996), CMC can boost Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 40 motivation for using the target language and reduce anxiety over classroom participation. Means and Olson’s study (1997) also indicated an increase in motivation levels when students were using computers for educational tasks. Various accounts have shown that, as compared to F2F interaction, CMC provides a more equitable platform with regard to the quantity and quality of participation across gender, socioeconomic status, and age (Beauvois, 1992; Kelm, 1992; Kem, 1995; Sullivan & Pratt, 1996; Waschauer, 1996; Waschauer & Meskill, 2000). The equalizing effect of CMC generates a less threatening forum in L2 discussion that is conducive to language learning (Waschauer, 1996). Sproull and Kiesler (1991) note that Internet communities create interaction based on shared interest rather than physical proximity, thus allowing members o f these communities to be simultaneously linked and buffered from one another. Consequently, many o f the social norms that restrict interaction in the classroom are either absent or greatly reduced, and students feel less shy and anxious, and are more willing to experiment with linguistic forms (Kelm, 1992; Kem, 1995). In a similar study, Ma (1996) reported that on the Internet, Asian students were more direct and open than they would be in F2F contexts, in which loss of face would occur if they demonstrated inappropriate social or linguistic behavior. The students Ma studied also exhibited a reduced level o f anxiety in online as compared to F2F contexts (see also Sullivan and Pratt, 1996). Identity construction. Language is an important way o f expressing one’s identity and group membership. Beebe (1981) views transfer in language learning as an assertion o f cultural identity, in which learners may choose to accommodate the speech o f their interlocutors, to conform to norms, to negotiate identity via linguistic choices to express Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 41 themselves, or to perform a variety of these actions. The interrelatedness o f SLA and identity construction has been documented in empirical studies o f classroom, natural, and CMC language learning settings (Kern, Ware & Warschauer, 2004; Lam, 2004; Lapadat, 2003; Nguyen & Kellogg, 2005; Norton, 1997; Norton & Toohey, 2001; Pavlenko & Lantolf, 2000; Spears & Lea, 1992; Toohey, 2000). There is a consensus among researchers that the process o f learning a new language is also the process o f identity construction, transformation and negotiation. For example, Toohey (2000) conducted a study on young children from minority language backgrounds in mainstream English-medium classrooms and found that their capabilities for learning English were greatly affected by their identity construction. Shotter (1993) asserts in his study that identity construction occurs within and through dialogue. He argues that through using certain kinds o f languages, people construct different social relationships as well as a sense o f their own identity. Spears and Lea (1992) argue that in networked learning environments one’s social identity is presented by being part o f a group or assuming a social role within the interaction. They further point out that one is more self-aware in networked learning environments than in F2F situations where no visual feedback is available. The idea behind their statement is that one develops a stronger social identity online than that developed in F2F settings. Nguyen and Kellogg (2005) studied gay rights and homosexuality via electronic bulletin board postings in a content-based English for Speakers o f Other Languages (ESOL) class and analyzed how students constructed their identities and demonstrated that language socialization was facilitated by the affordances of computer technology. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 42 Computer-mediated communication, which offers the use o f the written mode o f online bulletin board postings, enables the class members to focus strategically on specific word usage and expressions and to negotiate meaning socially (see also Herrings, 1999). The students collectively constructed an understanding o f key concepts such as discrimination, normal or abnormal. The process o f achieving this understanding was displayed visually in the form o f text, which helped students to “perceive a contrast between their values and others’” (Nguyen & Kellogg, 2005, p.20), and therefore become socialized into the second language and culture. Nguyen and Kellogg’s study indicated that when learning a second language, students may, through linguistic choices, conform to norms, negotiate identities, or express themselves in unique ways. The process o f constructing new identities in the CMC environment involves students’ dynamic collaborative meaning negotiations with others within a specific community of practice. Similarly, Hodgson (2002) proposed that an analysis of people’s different ways o f expression can examine the processes through which social relationships are constructed and people acquire a sense o f their own identity. She suggests that we need to examine closely the use o f language in networked learning environments so as to uncover the process o f identity construction through dialogue. Code-switching Code-switching (CS) is a widely observed phenomenon, especially in multilingual and multicultural communities. A variety o f terms have been employed in the literature to describe the specific occurrences o f this type o f linguistic behavior—code-switching, code-mixing, code-alteration, code-changing, borrowing, and so forth. Among these Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 43 terms, two pairs need clarification—code-switching and code-mixing, and code-switching and borrowing. According to Hudson (1996), there is a distinction between code-switching and code-mixing. Code-switching occurs when people, particularly bilinguals or multilinguals, intentionally switch from one language or dialect to another due to factors such as situation, subject, feeling, and sense o f belonging. This switching occurs at the sentence level, or what Sridhar and Sridhar (1980) have called the inter-sentential level. On the other hand, code-mixing occurs when people unintentionally incorporate small units (words or short phrases) from one language or dialect to another (Hudson, 1996). Code-mixing occurs on the word or phrase level, or what Sridhar and Sridhar (1980) have called the intra-sentential level. However, the boundary between the two terms sometimes is blurred, given the fact that the general degree o f language differentiation or the speaker’s intent needs to be taken into account, and more often than not, it is difficult to assess accurately a speaker’s intent (Albrecht, 2004). Unlike code-switching and code-mixing which mix languages in speech, borrowing, as explained by Hudson (1996), mixes language systems themselves, “because an item is ‘borrowed’ from one language to become part o f the other language” (p.55). For example, the French phrase, fait accompli, and the Spanish word, macho, have been borrowed into the English language. According to Hudson, borrowed words or loan words are those words that have a history: ordinary people can associate borrowed words with any other language. Myers-Scotton (1988) distinguishes code-switching and borrowing according to the role o f the social content o f a word: Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 44 code-switching involves switched words which carry social significance, while borrowing does not. It is impossible, however, to classify every occurrence o f an inserted item as code-switching or borrowing (Albrecht, 2004). Considering that these terms have not been standardized in the literature (Boztepe, 2003; Huerta-Macias & Quintero, 1992; Romaine, 1995), and my intent is to describe holistically the alternate use o f two languages which occurred among Chinese ESL students during their home literacy practices in CMC, I will use the term code-switching to describe any alternation between English and Chinese. According to Crystal (1987), the reasons for the switching are three-fold: first, when a speaker has difficulty expressing him/herself in one language, he/she may switch to the other to compensate for the deficiency; second, switching commonly occurs when an individual wishes to express solidarity with a particular social group. Rapport is established between the speaker and the listener when the listener responds with a similar switch; and third, the speaker wishes to convey his/her attitude towards the listener. These notions suggest the code-switching may be used as a socio-linguistic tool by bilingual speakers through either compensating for language deficiency or constructing identity. Code-switching in sociolinguistics. Two distinct linguistic approaches have been applied to the study of CS: structural and sociolinguistic. The structural approach to CS focuses on grammatical aspects o f CS in attempts to identify syntactic and morphosyntactic constraints on CS. Morphosyntax refers to the set o f rules that govern linguistic units whose properties are defined by both morphological and syntactic criteria. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 45 In contrast, the sociolinguistic approach, consistent with the view o f CS as primarily a discourse phenomenon, is mostly concerned with how social meanings are created in CS and what specific discourse functions CS serves (Boztepe, 2003). Auer (1984) points out that while grammatical constraints on CS may be necessary conditions for some instances o f code-switching, they are not sufficient to describe the reason for or effect of a particular code-switching. Consequently, the structural approach fails to answer the basic questions o f why code-switching occurs. Studies of CS have been carried out extensively in bilingual contexts from sociolinguistic perspectives. Blom and Gumperz (1972), whose approach has been the most influential in defining code-switching and establishing code-switching as a focus o f research, identified two different types o f code switching: situational and metaphorical. Situational switching occurs when speakers associate a certain code with social roles and situations. For instance, a teacher may deliver a formal lecture in the standard dialect, but switch to the local dialect when he/she wants to encourage an open discussion. On the other hand, metaphorical switching is triggered by changes in topic within a single social setting. For example, Blom and Gumperz (1972) observed that while the use o f a non-standard dialect occurred in greetings and inquires about family affairs among the local people in a small town in Norway, the standard dialect occurred in conversations about business transactions. They argue that aside from purely linguistic factors, a bilingual speaker’s code choice is affected by the participant, the setting, and the topic. A bilingual speaker makes code choice according to “a particular constellations o f people, gathered in particular settings during a particular span of time” (p. 423) for a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 46 certain activity. Later acknowledging that it is generally difficult for analysts to identify particular language choices as situational or metaphorical, Gumperz (1982) adopted another terminology: conversational code-switching. It was, however, described and defined largely in terms o f metaphorical switching. Myers-Scotton (1983) developed Blom and Gumperz’s situational and metaphorical dichotomy, and presented a Markedness Model which attempted to define, in general terms, the psycho-social factors that motivate a bilingual speaker to choose one language over another for a particular utterance or stretch of discourse. According to Myers-Scotton, marked refers to the unexpected use o f a code over another, whereas unmarked refers to the conventional use o f a code. The social purpose o f conversation is the negotiation o f a set o f rights and obligations between speakers and addressee, which she calls a “rights-and-obligations” set (RO set) (p.l 17). Within this model, language choice indexes an identity, and is associated with particular social roles. Therefore, the code used in CS has socio-psychological associations, making it indexical o f a RO set. By speaking a particular language, a speaker signals his/her understanding o f the current situation and particularly his/her relevant role within the context. By speaking more than one language, speakers may initiate negotiation over relevant social roles. According to this model, speakers understand the indexical value o f each language in their repertoire. For each situation, there is an expected or unmarked language, indexing the appropriate values for participant relations. A speaker generally chooses the unmarked language, since it expresses relationship between speaker, hearer, and setting. However, speakers analyze the potential risks and benefits of all other potential choices, and may make a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 47 marked choice on the basis o f this calculatioa Studies by Gumperz (1982) and Myers-Scotton (1983) have been echoed by many other researchers (Bourdieu, 1977; Genesee & Bourhis, 1988; Gibbons, 1987) who argue that code choices are made as a result o f both social factors and those related to the immediate situation. Studies have been conducted of the discourse functions of CS, and several o f these researchers have reported that CS functions primarily as a symbol o f group identity and solidarity among members o f the speech community (Beebe, 1981; Gal, 1978; Milroy, 1987). Gumperz (1982) identified six basic discourse functions that CS serves in conversation: (a) quotations, (b) addressee specification, (c) interjections, (d) reiteration, (e) message qualification, and (f) personalization versus objectivization. According to Gumperz, quotations occur when someone else’s utterance is used either as a direct quotation or as reported speech. In addressee specification, the speaker switches code to direct the message to one particular person among several addressees present in the immediate environment. Interjections refer to sentence fillers, such as the English filler you know inserted in an otherwise completely Spanish utterance. Reiteration occurs when the speaker repeats a message in the other code to clarify what is said. Message qualification is defined as an elaboration o f the preceding utterance in the other code. Personalization versus objectivization refers to the degree o f the speaker involvement in a message; for example, in the case o f giving one’s statement more authority in a dispute through CS. Gumperz’s study suggests that the functions o f code-switching range from conveying intentional meaning to signaling the social identities o f the speakers involved. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 48 Numerous studies of CS have examined the social meaning and discourse functions that CS serves from a sociolinguistic perspective. They have analyzed the many and diverse ways in which CS and society entwine, or how CS serves social purposes and is shaped by the social nature o f human beings. However, most o f these influential studies have investigated this linguistic behavior with people who were fluent in both languages, or people who alternated between two dialects, but not with learners who were in the process o f acquiring a second language. A common feature o f these studies is that they focused on “expert” code-switching by bilinguals, which is smooth, and viewed as an advanced skill (Ramaine, 1995). Sociolinguistic studies consider code-switching to be one o f the linguistic means bilinguals exploit in negotiations of their social relations. Studies are scant that focus on ESL students whose second language level does not enable them to communicate fluently. Furthermore, I would argue that the sociolinguistic approach has failed to discover what the speaker accomplishes both linguistically and socioculturally through CS, or how CS is related to the development o f L2. Despite the numerous studies o f CS through structural and sociolinguistic perspectives, studies o f CS are limited in the SLA area, especially in the area o f language learning in a CMC environment. Code-switching in second language acquisition. In studying code-switching behavior in the second language learning contexts, researchers o f recent studies (e.g., Anton & DiCamilla, 1998; Brooks, 1992; Swain & Lapkin, 2000; Wells, 1998) do not employ the term, “code-switching”. The focus o f their studies has been on LI use in the L2 classroom. LI use in the L2 classroom has long been a controversial issue between Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 49 both researchers in SLA and foreign language (FL) classroom teachers (Thoms, Liao & Szustak, 2005). Until recently, employing LI to carry out a learning task in the L2 classroom was viewed as counter-productive. Previous research tended to indicate that overuse o f LI deprives learners o f input in the target language (Carroll, 1975; Chaudron, 1988; Wong-Fillmore, 1985), with the implication that the more learners are exposed to input in the target language, the more they will learn, thus denying the role of LI in L2 acquisition. As claimed by Dickson (1992) and Gass (1988), however, mere exposure to input in the target language cannot ensure intake, which is defined by Ellis (1997) as “the portion o f the input that learners attend to and take into short-term memory”. Swain (1985, 1993) asserts that modification o f the input, as well as either or both interaction and the output, may determine whether or not input becomes intake. Therefore, Turnbull (2001) suggests that input in the target language may become intake more readily if teachers resort to LI judiciously to “catalyze the intake process in some way” (p. 531). Some recent studies by Brooks (1992) and other researchers (Anton & DiCamilla, 1998; Brooks, 1992; Brooks & Donato, 1994; Cook, 2001; Darhower, 2002; Femandez-Garcia, & Martinez-Arbelaiz, 2002; Wells, 1998) have examined the role o f LI in the L2 classroom among second language learners whose L2 proficiency was relatively low. These studies have thrown light upon the benefits o f the use o f LI in the L2 learning environment. For example, by examining the LI use o f eight pairs of Grade 11 Spanish learners engaged in an information-gap activity, Brooks and Donato (1994) report that LI use is a “normal psycholinguistic process that facilitates L2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 50 production and allows learners both to initiate and sustain verbal interaction with one another” (p. 268). Furthermore, Platt and Brooks (1994) noted that students use their first language when they engage in a difficult task involving problem solving. According to Platt and Brook, L2 learners resort to their LI when they talk to themselves (private speech); when they define a learning task for themselves (situation definition); and when they comment on their own language production (metatalk). In a similar vein, Anton and DiCamilla (1998) highlighted the critical functions of LI in the L2 learning process by examining students who participate in a collaborative writing task in the L2 environment. Their findings indicated that LI is used as a mediating tool to facilitate scaffolding; to achieve intersubjectivity with their partners during collaborative pair work; and to externalize one’s inner speech during cognitively difficult tasks. In their study, Anton and DiCamilla demonstrated that L2 learners resort to their LI not only to mediate the cognitive process used in problem-solving tasks, but also to access L2 linguistic forms. To provide empirical evidence o f the advantages o f LI in the L2 classroom, Swain and Lapkin (2000) investigated the use o f LI by two groups o f Grade 8 French immersion students as they completed a dictogloss or jigsaw task. The results o f the study show that LI use has three major functions: “(a) moving the task along, (b) focusing attention, and (c) interpersonal interaction” (pp. 257-8), and that LI is used most frequently for task management purposes. Swain and Lapkin analyzed the relationship between the amount of LI use and the quality o f students’ writing, as well as the variability in task performance across student pairs. Significant negative Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 51 correlations were found between the percentage o f LI use and the quality o f the performance in the jigsaw task group. However, no significant correlations were found in the dictogloss task group. The explanation for this result, according to Swain and Lapkin, is that the dictogloss students needed to use their first language more to understand the story than did the jigsaw students who had a series o f pictures to interpret. Swain and Lapkin concluded that judicious use o f LI can support L2 learning in the sense that it should neither be prohibited nor actively be encouraged as it may substitute for, rather than support L2 learning. The conclusion Swain and Lapkin have drawn from their empirical study is significant in understanding the role that LI plays in L2 learning tasks. However, these studies are small in number, and restricted to classroom settings where L2 learners share a first language and are situated in their home countries learning a second/foreign language. Unlike those L2 learners in the studies discussed above, English-as-second-language students who are situated outside o f their countries o f origin, such as ESL students in Canada, are confronted with not only linguistic but also psychological and cultural challenges that have an impact upon their second language acquisition. However, a few studies have been conducted that systematically investigate the function o f LI among ESL learners with relatively low L2 proficiency in natural L2 learning contexts beyond the classroom, such as in CMC environments, and therefore, importance should be attached to such studies. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 52 Studies o f code-switching in computer-mediated communication. Because new technologies such as CMC are increasingly being used in L2 learning environments, research in code-switching in CMC has been rather limited and needs to be expanded,. The limited information available about L2 learners’ code-switching within the CMC context suggests that learners employ this linguistic strategy not only to compensate for language deficiency for communicative purposes but also to create an ethnic identity. Kotter (2003) investigated how German learners of English and American learners of German exploited their bilingual knowledge during the online MOO (Multi-User Object-Oriented)-based tandem exchanges. MOO is a text-based online environment that is similar to a chatroom. This study examined how the participating language learners code-switched and how their metalinguistic skills were enhanced in the online tandem between NNS and NS. The study showed that these learners were engaged in negotiation o f meaning via MOO to avoid communication breakdowns. One of the ways was code-switching, which language learners employed to negotiate meaning and scaffold their partners’ task in order to establish and sustain mutual understanding. By code-switching, the learners also attempted to prompt their partners to revert to the established code o f a conversation or converge on their choice of code to receive additional input in their L2 and sometimes to get further opportunities to practice their own target language skills. As a result o f negotiating meaning and code-switching, the learners’ metalinguistic abilities were developed. Nevertheless, it still remains unknown how L2 learners code-switch in natural or informal electronic language learning Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 53 environments between NNS and NNS, the specific linguistic skills that would be facilitated, and the circumstances under which code-switching would occur. Lam (2004) conducted a study based on a language socialization framework to examine the home literacy practices o f two teenage Chinese immigrants in the United States. She studied their use o f English on the internet, with a focus on their social and discursive practices in a Cantonese/English bilingual chat room. Lam argues that the Internet chat room provided a context for language socialization in which a mixed-code variety o f English was adopted and developed among the focal youth and their peers across the world. This variety o f English includes varying degrees o f code-switching from predominately English to predominately Cantonese. Drawing on the linguistic and discursive conventions of Cantonese, code-switching not only helped the girls and their peers develop “a sense o f fluency in English” (p. 59), but also distinguished them from both monolingual English speakers and monolingual Cantonese speakers, thus constructing their collective ethnic identity as bilingual speakers o f English and Cantonese. The resulting new identity, as reported, has “influenced their relation to the English language in the US” (p.59). Lam has contributed to the literature by addressing explicitly Chinese ESL students’ construction o f identity on the Internet through code-switching. Her study also touches upon the issue o f the mixed-code variety o f English, specifically code-switching, and its relevance to English language development. However, this line of research could be further supported by evidence from chat room exchanges exhibiting what language Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 54 skills are developed, and tracing the contexts in which the code-switching occurs, and identifying the communicative purposes o f the code-switching. Research Questions In this study based on interactionist and sociocultural perspectives, and motivated by my personal experience as a bilingual speaking Mandarin Chinese as the first language and English as the second, I examine Chinese ESL students’ code-switching behaviour which occurred when they were engaged in home literacy practices in CMC contexts, compare the code-switching strategies employed by the Chinese ESL students, and infer the relatedness code-switching had for the students’ English language development. This study is useful in that it contributes to second language acquisition theories, enriches literature in CMC, provides guidance for ESL educators on how to innovate their instruction in order to meet ESL students’ needs, and helps ESL students build up their confidence in learning English. The research questions are formulated as follows: 1. How do Chinese ESL students code-switch across languages and modalities, according to their communicative purposes, when engaged in computer-mediated communication? 2. What are the similarities and differences in computer-mediated communication among Chinese ESL students who differ in age and level o f study, particularly the way they code-switch? 3. What is the relevance o f code-switching to the participants’ English language development? Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 55 Summary There is no conclusion in the literature as to what contributes to minority students’ academic failure. Research findings indicate that home literacy practices are one o f the factors that affect minority students’ school success. Computers have become an important part in the home literacy practices o f students from ethnic groups. Computer-mediated communication has been incorporated in language teaching and learning, mostly at the college level. Numerous studies show that CMC has brought about a myriad o f benefits to students in terms o f language learning. Code-switching, a widely observed linguistic behavior among bilingual students, occurs in CMC. Although CS has been studied extensively through structural and sociolinguistic approaches, the studies o f CS in SLA, especially in CMC, are rather limited. The available studies o f CS in CMC demonstrate that the code-switching strategies bilingual students employ serve linguistic purposes as well as social functions. Therefore, this study examined the code-switching strategies Chinese ESL students employed while engaged in CMC at home, identified the contexts in which code-switching came about, and looked at the relevance o f code-switching to English language development. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 56 Chapter Three: Research Design and Methods Introduction I used a qualitative research approach in this study. I employed a multiple case study design in order to compare the ways Chinese English-as-a-Second-Language (ESL) students who differ in age and study level code-switch in Computer-Mediated Communication (CMC). My sample of 4 participants included three Chinese ESL students and the parent o f the student who was under the age of 1 8 .1 gathered data through interviews using open-ended interview questions and probes to pursue important leads. Finally, I utilized an inductive thematic analysis approach for data analysis through a qualitative data analysis package NVivo, and this approach enabled me to discover overarching themes that emerged from the individual data and from across the participants. Each o f these aspects of the method is addressed in detail below. Research Design I chose to use a qualitative multiple case study approach in my study for a variety o f reasons. First, Lincoln and Guba (1985) argue that qualitative methods are more suitable to dealing with multiple realities and more sensitive to “the many mutually shaping influences and value patterns that may be encountered” (p. 40). Code-switching (CS) in CMC is a complex process in which a variety of factors may contribute to the occurrence of CS. The factors may involve linguistic, social and cultural elements and interact in unique ways for different individuals or within different contexts (Crystal, 1987; Lam, 2004). Gaining in-depth prospective information o f a complicated phenomenon, in this case, code-switching, may be Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC difficult to express quantitatively (Bodgan & Biklen, 2003; Strauss & Corbin, 1990). Second, a quantitative approach could oversimplify the problem and fail to capture the complexity o f the issue under investigation and required making too many a priori assumptions about each variable that may not reflect the events accurately (Miles & Huberman, 1984). Because my study was aimed at exploring, describing, and understanding the variables that may have a role to play in code-switching in CMC, I took a qualitative approach which facilitates descriptive and explorative study (Bodgan & Biklen, 2003). This approach allowed an investigation o f participants’ code-switching in CMC. From this exploratory study, I made sense o f the participants’ linguistic behavior in terms of the meanings they brought to their environments. Third, a qualitative approach enabled me, as a researcher, to act as an active learner, reporting a detailed descriptive account from the participants’ view rather than as an “expert” who passes judgment on participants (Cresswell, 1994). Within the qualitative paradigm, I adopted a multiple case study method in the present study. Yin (1989) contends that a case study method is preferable to examine a contemporary phenomenon in its real-life context, especially when the unit o f study cannot be separated from its context. Miller (1997) indicates that the case study method has the potential to account for language learning as discourse acquisition, and the real-life settings under which L2 learners operate thus become important in capturing the multi-dimensional factors that influence second language acquisition (SLA). As a linguistic behavior occurring in the course o f home literacy practices in CMC (Lam, 2004), code-switching is complex and highly contextual. Rather than attempt to study Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57 Code-switching in CMC these phenomena—code-switching, and real-life settings or context— in CMC at home separately, the present study captured the rich contexts o f code-switching in CMC and delved into the factors influencing this linguistic behavior. O f the various approaches to case studies, a multiple case study design was suitable for this study, because I not only investigated a contemporary phenomenon within its natural setting through holistic inquiry but also compared the three participants in an attempt to “build a general explanation that fits each case o f the individual cases, even though the cases may vary in their details” (Yin, 1989, p. 108). In so doing, the potential for generalizing beyond a particular case is increased, and an interpretation based on evidence from several cases can be fuller than that based on a single case (Miles & Huberman, 1984; Merriam, 1988). Generalization o f findings from a multiple case study is aimed toward some broader theory (Yin, 1989), reassuring the researcher that “the events and processes in one well-described setting are not wholly idiosyncratic” (Miles & Huberman, 1984). Participants Sampling is crucial for data analysis in a qualitative study, and “must be theoretically driven—whether the theory is prespecified or emerges as you go” (Miles & Huberman, 1994, p. 29). In my study, I employed a purposeful sampling strategy, a strategy that enabled me to seek information-rich cases that can be studied in depth (Patton, 1990). O f the various types of purposeful sampling, I selected maximum variation sampling suggested by Patton (1990; see also Miles & Huberman, 1994). Patton argues that maximum variation sampling can yield detailed descriptions o f each Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58 Code-switching in CMC 59 case, in addition to identifying the central themes or principal outcomes that cut across cases. Therefore, I deliberately searched for variations to see whether main patterns would still hold and to increase confidence in conclusions (Miles & Huberman, 1994). The study took place in a mid-sized northern Canadian city, to which I gave the code name “Northemlight”. My selection criteria were as follows. First, they should be students who differ in age and level of studies. Second, they should be recent arrivals in Canada within last year. Third, they should come from mainland China, and speak Mandarin Chinese as their first language. I set these selection criteria for the prospective participants for a number o f reasons. First, according to Blom and Gumperz (1972), language proficiency is one o f the three factors found to affect code-switching. Participants at different study levels are supposed to differ in English proficiency due to the length o f their English study and the exposure to this language. English proficiency affects the purpose behind code-switching. Second, the participants, as recent arrivals in Canada, have not completely adapted to their studies, or to the culture and life here. In particular, their English proficiency was limited at the time o f the study, and therefore, does not enable them to communicate in written English with ease. As a result, code-switching would be more likely to occur and to occur more frequently (over the course o f use of technologies at home) than would be otherwise, and code-switching would be more observable and comparable. Third, Mandarin Chinese spoken in mainland China is different from that spoken in Taiwan, and from Cantonese, one of the Chinese dialects spoken in Hong Kong. The Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 60 Chinese expressions used in mainland China and Taiwan and Hong Kong differ. Furthermore, simplified Chinese characters are used in mainland China, whereas the traditional complex forms o f Chinese characters are used in Taiwan and Hong Kong. I am from mainland China, speak Mandarin Chinese, and am most familiar with the simplified Chinese characters. Therefore, as a researcher, I can most productively analyze and interpret simplified Chinese characters and the Mandarin language. Lastly, although Hong Kong and Taiwan are parts of China, and Mandarin is their official language, they differ in terms o f values and customs due to their long separation from mainland China and differ in their political systems. For all o f these reasons, I limited my sample to Mandarin-speaking recent arrivals from mainland China. Therefore, despite my attempt to achieve variation in sampling, I was limited in my choice of participants, as I required Mandarin-speaking participants who were recent arrivals from mainland China and differed in age and levels o f study, and the population fitting this description in Northemlight is small. My sample was, in part, a sample o f convenience, but within the constraints, I was striving to achieve maximum variation. Three Chinese ESL students (Amy, Baillie, and Cynthia) who met my selection criteria and one parent (Emma) participated in the study. All names are pseudonyms. Amy, 13 years o f age, was in elementary school; Baillie, 20 years o f age, was in college; Cynthia, 25 years o f age, was in university. Emma, the parent, is an immigrant, and has been self-employed since her arrival in Canada. Selected participants in this study reflected a range o f educational institutions and levels o f study in the community, ranging from elementary to post secondary. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 61 Upon approval of this study by the University o f Northern British Columbia (UNBC) Research Ethics Board, I embarked on the recruitment o f the participants. Through a potluck dinner organized by the local Chinese community, o f which I am a member, I became acquainted with two immigrant families that had arrived in Northemlight last year. Each family had a child attending elementary school. W ith the parents o f these two children, I discussed home computer use and the parents’ attitudes towards computers. As a result o f the discussion, I invited a parent who was supportive of her child’s use o f computer at home, and her child’s participation in my study. As well, I spoke to a few women who recently arrived in the city and who were either taking ESL courses at the local community college or pursuing their undergraduate/graduate studies at the university. I also became acquainted with them at a party given by the local Chinese community. By speaking with them, I found that some o f these female students used the computer to communicate with their friends or classmates only in Chinese, and some communicated with friends or classmates sometimes in Chinese and sometimes in English. I had a further talk with those students who communicated on the internet in both Chinese and English and asked them whether they had encountered difficulty in expressing themselves entirely in English. Then I focused on those who articulated their difficulty. From these students, I invited one from the college and one student from the university to participate, both o f whom showed interest in my study, and with whom I felt comfortable talking with. Ideally, I would have liked to include a high school student in my study as well, but was not able to locate s suitable participant who met my criteria. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 62 Data Collection I used a variety o f qualitative data gathering techniques in this study. The particular methods selected for this research included (a) audio-taping interviews with the participants, (b) video-taping the participants’ actual use o f computers, and (c) collecting documents such as electronic artifacts. Yin (1993) suggests that multiple sources o f evidence can provide multiple measures of the same phenomenon. Findings from a multiple case study through triangulating multiple data sources are likely to be more convincing and accurate (Yin, 1993). Interviews. Although a variety of different data collection methods may be adopted across disciplines that employ the case study method, semi-structured interviews are popular because they address the topic at hand and draw from the informant’s emerging world view (Merriam, 1988). Guided by depth interviewing techniques recommended by Patton (1987), I used open-ended interview questions to pursue important leads and probed particular details through follow-up questions (see Appendix A for interview questions). Depth interviewing refers to “asking open-ended questions, listening to and recording the answers, and then following up with additional relevant questions” (Patton, 1987, p. 108). Open-ended questions (for example, “How do you feel about the use o f computer at home?”) solicit the interviewee’s unstructured answers, and yield in-depth responses about his/her experiences, perceptions, opinions, feelings, and knowledge (Patton, 1987). In contrast, closed-ended questions (for example, “Which o f these books do you prefer, fiction or non-fiction?”) restrict the interviewee’s responses by supplying alternative answers and requiring him/her to choose from a fixed set of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 63 possible responses. Interview questions were formulated based upon the following sources o f information: (a) the research questions for this study, (b) the literature review conducted for this study and (c) my own knowledge and experience as a teacher and L2 learner. As well, I am the parent of a teenager girl who code-switched between English and Mandarin when using CMC, and this personal knowledge helped me to frame pertinent questions. The open-ended interview questions enabled me to obtain as much information as possible from the participants, as well as to direct and regulate each participant’s responses in the sense that data are complete from each participant on the topics addressed in the interview (Patton, 1987). Video-taping. To corroborate the data gathered from the interviews, I also collected information through videotaping observations, and analyzing electronic artifacts (Yin, 1993). I observed and video-taped the participants’ computer use, specifically when they chatted online, to record what was going on in their online chatting. Electronic artifacts. The participants provided me with the electronic transcripts o f their online chatting either in chat rooms or on MSN, and other electronic artifacts of computer products, such as emails, graphics, and web pages. These electronic artifacts were their personal social communication. These documents provided detailed empirical evidence o f code-switching to clarify or corroborate the participants’ descriptions and explanations from the interviews. These supplementary data enabled me to generate a clearer understanding o f code-switching in CMC. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 64 Procedure. I met with all o f the participants separately and articulated my responsibilities and expectations o f them, and what they could expect o f me. I discussed the explanatory information about the study and also provided this information in the form o f written letters both in English and Chinese (see Appendix B) and the consent forms both in English and Chinese (see Appendix C) with all the participants prior to their signing the consent forms. My oral explanations were provided in Mandarin. The parent o f the participant from elementary school signed the consent form to allow her child to participate, and also agreed to participate herself. As well, I offered to answer any questions or address any concerns they may have had concerning the study. After the participants had given informed consent for their participation, I determined the student participants’ English proficiency by administering the Canadian Language Benchmark (CLB) Assessment which is used primarily for placement in language training by the Immigrant and Multicultural Services Society in Northemlight. This test is not a pretest in pretest-posttest comparisons used in a true experimental design. The purpose of pretest-posttest comparisons is to assess a pedagogical or technological intervention by detecting differences in learning outcomes between two points in time - before and after the intervention. The present study did not involve any intervention and did not intend to compare any learning outcomes quantitatively. I administered the test so as to identify the developmental levels o f the participants’ resources in the English language, and to give a clear indication of how participants with different resources in English employ code-switching strategies in different social contexts. The assessment measures three areas o f language ability: a) speaking/listening, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 65 b) reading, and c) writing. Given the nature o f the present study, only reading and writing were assessed. According to the Canadian Language Benchmarks, there are three stages, namely, basic proficiency; intermediate proficiency; advanced proficiency, with benchmarks 1-4 at each stage. The test results show that the participant from the elementary school was somewhere between levels 3 and 4 at Stage 1 (basic high proficiency) regarding English reading and writing. The English reading and writing level o f the participant from the college was somewhere between levels 5 and 6 at Stage 2 (low intermediate proficiency). The participant from the university reached the level of 7 or 8 at Stage 2 (high intermediate proficiency). I administered the test in order to help me interpret the purposes o f the code-switching strategies among the three participants. Upon the completion o f the test, I scheduled the interviews and video-taping. Appointments for interviews and video-taping were made at the participants’ convenience, and the parent was involved in the scheduling o f her child’s interviews and video-taping. The participants and I share a first language. I therefore conducted the interviews in Mandarin Chinese, which enabled the participants to express themselves fully and accurately. Each participant (the three students and the parent) was interviewed individually twice, with each o f the two sessions lasting approximately 30 minutes. The rationale for having two sessions is threefold. First, I was able to acquire an in-depth account o f the contexts o f their code-switching in CMC without making them tired by interviewing them for one long session. Second, it was possible for me to clarify some questions or pose additional important questions during the second session after Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC conducting preliminary data analysis o f the first interview. Third, the second session offered me a chance to check information with the participants to ensure validity of the initial data and findings. The video-taping for each o f the student participants was carried out after the second interview, and lasted about 30 minutes. I conducted the video-taping in the home of each o f the participants, with the video camera oriented at the screen at times, and at other times oriented simultaneously at the screen, the participant and the keyboard. After I completed the entire data collection for the first case, I proceeded with the second and then the third. Data collection was done case by case, because doing more than one case at a time may lead to confusion (Bogdan & Biklen, 2003). The interviews o f the four participants were tape recorded for transcription. Immediately following each interview, I transcribed the interview in Chinese and then translated it into English. I also wrote post-session memos to capture non-verbal information from the interviews (Yin, 1994), and my reflections about the interview sessions. The students provided me with some o f their electronic artifacts, such as electronic copies o f emails, MSN chat, or graphics after the second interview, and the electronic transcripts of their online chat after the video-taping. These field notes and electronic artifacts played an instrumental role in generating some follow-up questions with particular participants to obtain additional information and also in establishing a context for interpreting and making sense out o f the interview (Patton, 1987). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66 Code-switching in CMC 67 Researcher Role and Reflexivity Bogdan and Biklen (2003) recommend that researchers “lighten up” their personal opinions, and argue that it is neither possible nor desirable for researchers to enter the field o f study as “a clean slate” (p. 34) no matter how much they try to divorce their research from their past experiences, status, beliefs, and values. Researchers should aim to become more reflective and conscious o f how their past experiences, status, beliefs and values shape and enrich their research rather than trying to eliminate their previous experiences (Bogdan & Biklen, 2003). From the outset and throughout the subsequent stages of the study, I positioned myself as a researcher who had an informed theoretical framework and was responsible for data collection and analysis, as a teacher who had had experience working with students studying English as a foreign language in my country of origin, and as a learner who wished to gain knowledge about how Chinese ESL students employ code-switching with the aid o f CMC. At the same time, I recognized that I entered the field o f study also as a part-time instructor at the university, as a mother of a teenage girl, as a member of the local Chinese community, and as a full-time graduate student. I realized that the participants might have been aware o f my multiple roles, which could affect the establishment o f trust with the participants. The self-awareness o f my previous experiences and multiple statuses enabled me to proceed cautiously into the field of study. Van Lier (1988) points out that the researcher brings ideas and experiences he/she has gained over the years to the study, which provides some common ground between Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 68 the research and the setting. I was aware that I possessed an informed conceptual framework established during the literature search and shaped by my personal experiences. As well, I am a bilingual Chinese person/woman, speaking Mandarin Chinese as my first language and English as the second. I experienced code-switching when I was at an early stage o f English literacy development, and am still employing code-switching strategies in interaction with the local Chinese people and in some social functions. I taught English at a university in my home country for quite a number of years, and I have noticed the way in which my students code-switched while answering questions in English and working on compositions. As the mother o f a teenage girl, I have observed that my daughter uses the computer at home as an aid to English studies, and that she switches languages while engaged in activities on the Internet. As a current user of computer technology, I communicate by means o f the computer with my Chinese friends home and abroad, and code-switch under some circumstances when I send emails. I am conscious that the code-switching strategies I employed when I was at an early stage o f English literacy development differ considerably from the ones I adopt now. I am also cognizant that the way my students code-switched (at a time when computer technologies were not popularized in my country o f origin) might differ, to a certain degree, from the way current bilingual students code-switch with the aid o f today’s technologies. The conceptual framework, my personal and professional experiences have influenced the way I have conceptualized this research, the formulation o f the research and interview questions. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 69 Bogdan and Biklen (2003) suggest that cultural identity sometimes facilitates fieldwork, and sometimes complicates it when researchers are studying people within their same ethnic group or when they are studying people in another culture. The participants and I share the same cultural background and speak the same first language, and therefore, I had an in-depth understanding of the participants’ customs and values. Additionally, I am pursuing my studies as an international student at a Canadian university. My status as a part-time instructor at the university might cause the two participants from post-secondary institutions to keep a distance from me, and therefore, after I selected them as the participants o f my study, I endeavored to approach them as a graduate student by doing things together with them, such as inviting them for dinner and sharing my previous and current experiences as a student with them. All o f my endeavors led to my identification with the two participants without much difficulty, and consequently their good cooperation with my research study. The parent participant completed her graduate studies in the United States and is approximately the same age as I am. Our similar personal and educational backgrounds brought us together quickly. The friendship I established with the parent also brought the teenaged participant closer to me. Besides, being a mother myself, I have experiences o f approaching a teenager in a “proper” way. My multiple statuses facilitated communication and rapport with the participants and the parent. On the other hand, the self-awareness o f my multiple roles and the realization of the subjectivity I may bring to the study made me constantly confront my own opinions with the data, as suggested by Bogdan and Biklen (2003). I therefore remained Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 70 constantly aware o f my contribution to the construction o f meaning at all stages and in all aspects of my research, due to my role o f being both an insider and outsider (Bogdan & Biklen, 2003). I reflected throughout the research process upon the ways in which my involvement, such as my own values, experiences, interests and cultural identity, shaped the research and how the research may have affected me and possibly changed me as a researcher. I always thought about the assumptions I made over the course o f my study, and interrogated the subjective beliefs, thought, attitude and opinions I brought to the study. When the data conflicted with my general opinions, I tried to be open to the argument, and to shape my thinking by the empirical world I was exploring rather than being defensive o f my opinions (Bogdan & Biklen, 2003). I also kept a reflective journal to record my thoughts that arose throughout the research process. Ethical Considerations Being an ethical and responsible researcher is very important. From the time when I decided to conduct the present study and started to select participants to the time when I collected data and wrote a report o f findings, I considered ethical issues, and incorporated them into my practice. The first ethical consideration was to secure the informed consent o f all the participants and from the parent of the participant under the age o f 19. To ensure that the selected participants fully understood the nature of the study, and what would be expected o f them prior to consenting to participate, I met with each o f the participants and the parent, and explained to them the purpose o f my study. I informed them verbally and in writing that their participation was voluntary, that there was no obligation on their Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 71 part to participate, and that they had the right to withdraw at any time. The participants also were informed that if anyone o f them chose to withdraw, any data collected on him/her would not be used in my study. In order for every participant and the parent to understand fully the nature o f my study and the voluntary nature of their participation, I prepared letters to participants and the parent, and consent forms both in English and in Chinese (see Appendix B). My next ethical consideration was to ensure and maintain the participants’ anonymity and confidentiality of the personal information that they provided. The local Chinese community in Northemlight is relatively small, and as a person from mainland China myself, I know many members o f the local community. The selection o f the participants was not random, but consisted of individuals of my immediate acquaintance. Such being the case, I assured the participants in writing and verbally that confidentiality o f such information as the schools they were attending would be maintained through the use o f code names in the transcripts and in reports o f the findings. I made it clear to them that I would avoid using any information that would allude to the identification of an individual, and that I would be the only person to have access to the original audio and video-taped information they provided. Once I entered the data into a computerized password protected database, I would keep the tapes in a locked file for a maximum of two years and then they would be destroyed. The participants also were informed that the final report o f the findings from this study would be available in the form o f a thesis at the university library for anyone who wished to access the findings (see Appendix B). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 72 I assured them that this study would adhere to the required ethical rules and guidelines, and that approval from the Research Ethics Board at UNBC had been obtained. Overview o f Data Analysis Approach. Data analysis is an ongoing cyclical process integrated with all phases of qualitative research (Merriam, 1988; Miles & Huberman, 1994). Data collection should be conducted simultaneously with analysis from the first time a qualitative researcher gets to the field, thus easing the potentially huge and overwhelming data analysis task (Miles & Huberman, 1994). As the data were collected and analyzed, I had the opportunity to examine the existing data to determine whether new data needed to be gathered in order to clarify, confirm, or fill in gaps that may exist in the current data. Preliminary data analysis drove an ongoing data collection in the sense that it allowed me to reshape my perspective and revise data collection techniques to collect richer, more relevant data (Miles & Huberman, 1994). I began data analysis following the first interview with a participant. Simultaneous data collection and analysis offered me an opportunity to collect more data from her and subsequently from the other participants, to clarify and confirm my emerging understandings and to uncover ideas that required further investigation. The four participants in my study were interviewed individually for two sessions each. I analyzed the data from the first session with a view to presenting to the participants ideas that emerged from the preliminary data analysis for member checking, before moving on to collect additional data in the second session. Member checking involves taking data and interpretations back to the participants from whom the data and interpretations were Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 73 derived and checking with them the plausibility of the results (Merriam, 1988). Doing so can ensure the validity of qualitative research (Merriam, 1988). After analyzing data from the second interview and videotaping, and conducting a preliminary analysis o f the artifacts provided by the participants, such as MSN chat transcripts, emails, webpages, or graphics, I made a follow-up inquiry via telephone about specific issues. For example, I posed questions such as “Why did you code-switch here?”, or “Why did you use the Chinese interjection to code-switch?” to obtain clarification or confirmation. Preliminary data analysis and my follow-up inquiry enabled me not only to present to the participants my interpretations o f their data for member checking but also to reduce the likelihood of misinterpretations in my subsequent data analysis. Data analysis software. I used the qualitative data analysis (QDA) software package QSR NUD*IST Vivo (v2.0), commonly referred to NVivo (Richards, 1999), for data analysis and management in this study. The features and functions o f the software allowed me to make multiple passes at my data as I examined, labeled, and re-labeled code categories during the analysis. NVivo provided an avenue o f writing and storing ideas that emerged during data analysis process. It also presented me with tools for searching, coding, sorting, merging, linking my data, and conceptualizing my findings. A description and discussion o f the software functions that I used in my data analysis is provided in the Data analysis section o f the Research Findings of Chapter IV. Steps in data analysis. It is central to qualitative research to understand and apply appropriate methods o f data analysis. In this study, I applied a mixture o f both inductive and deductive approaches, with an inductive thematic analysis as the major method, so Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC as to discover overarching themes that emerged from each individual participant’s data and across all the participants’ data (Miles & Huberman, 1994). The process o f data analysis is shown in Figure 1. Data collection phase 1 Preliminary analysis Member check Coding idea units A priori categories Follow-up inquiry Data collection phase 2 Developing categories Pattern coding across participants in each category Pattern coding across categories Overarching themes Means o f code-switching English language development Contexts of code-switching Visual representation of findings Figure 1. Process o f Data Analysis. For each participant, in order to make sense out o f the data, I first performed a preliminary analysis o f the data o f the first interview before conducting the second interview with the participant. After the preliminary analysis of the artifacts collected Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74 Code-switching in CMC 75 from the student participant after the second interview and the video-taping, I made a follow-up inquiry via telephone. During this process, I underlined significant parts o f the transcription and made marginal notes to avoid losing initial impressions. Another essential component of data analysis in this step was development o f the coding criterion to operationalize the identification o f code-switching instances. In this study, I defined code-switching as alternating between English and Mandarin Chinese either at intrasentential level or at intersentential level. This definition played an instrumental role in my decision-making about the inclusion and exclusion o f instances o f code-switching. All o f the electronic artifacts the participants provided were their personal online communications solely in English, except for the segments where code-switching occurred. I thus set the coding criterion as shown in Table 1. First, I included instances o f the use o f Chinese notional words, Chinese interjections, such as aiya, eng,and en, modal particles (tone-indicating particles, such as ba, la, ne, and so forth) used at the end o f an English sentence, and the use o f a Chinese sentence, as code-switching. For example, “What is chuanqiT’’ Chuanqi, written in Pinyin, or Romanized Chinese, means a legend. Another example is : “aiya, it is time to watch TV”. Aiya, a Chinese interjection, indicates surprise. In “impossible b d \ ba, a Chinese modal particle, indicates that the speaker does not fully believe the preceding statement made by her interlocutor. In “Guess” “bu shuo jiu suan le'\ bu shuo jiu suan le, a Chinese sentence in Pinyin, means that it is okay if you don’t want to tell me. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 76 Table 1. Coding Criteria o f Code-switching Coding Criteria o f Code-switching_________________Examples____________ Inclusion of instances of code-switching Use o f Chinese notional words What is chuanqfl(legend)1 Use o f Chinese interjections aiya, it is time to watch TV. (aiya indicates surprise.) Use o f Chinese modal particles Impossible ba. (ba indicates that the speaker does not fully believe the preceding statement made by her interlocutor.) Use o f numbers which are homophones 88 (meaning “bye-bye”) with some Chinese words Use o f a Chinese sentence bu shuo jiu suan le (It is ok if you don’t want to tell me.) Exclusion of instances of code-switching Use o f Chinese onomatopes which are haha, I just finished mine, (haha also used in the same way as in the imitates the sound o f laughter.) English language Use o f numbers which are homophones Wait 4 me. (Wait for me.) with some English words Use o f emoticons My finger still h u r ts lf . Note 1. Words or sentences in the parentheses are my translations or explanations. Second, I also included the use o f numbers that are homophones with some Chinese words. For example, 88 is pronounced [ba ba], and was used to mean bye-bye. Third, numbers— which are homophones with some English words—were excluded from code-switching when used within an English sentence. For example, “wait 4 me” means “wait for me”, but does not involve switching between languages. Fourth, instances o f use o f Chinese onomatopes (words which imitate the sound of a thing or an action) which have their English equivalents, such as haha, or hehe, were excluded. Examples include “I haven’t finished my assignment”, or “haha, I just finished mine”. Haha, written in Pinyin or Romanized Chinese, is an onomatopoeic Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 77 word which imitates the sound of laughter in the Chinese language. Haha is also used in the English language with the same function as in the Chinese language. Fifth, instances o f use o f emoticons are excluded, because they are a universal means to express one’s emotions and attitudes in a CMC environment. An example is “my finger still hurts H ”. In further analysis of the field notes, interview transcripts, and the artifacts, I proceeded through three analytical steps suggested by Lincoln & Guba (1985) —unitizing textual data into idea units, coding categories, and integrating categories. In outlining the different steps, I define each step, describe explicitly how I worked with the data at each step, and identify the NVivo (Richards, 1999) functions that proved helpful in handling and analyzing the data. Summary A qualitative multiple case study approach was most suitable for my study for a number o f reasons. Code-switching behaviour, deeply embedded in its contexts, can be affected by linguistic, social, and cultural factors. A qualitative approach allowed an in-depth examination of the complexity o f code-switching and o f the factors that influence code-switching. A multiple case study design enabled a comparison o f the three cases in order to build a general pattern that could explain the code-switching behaviour o f each case. Similar findings based on several cases could be more convincing than from a single case, thereby increasing the potential for generalization beyond a particular case. I employed a maximum variation sampling strategy in selecting the participants in this study. The participants involved one parent, and three students Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 78 who differed in age and attended educational institutions ranging from elementary to post-secondary. All o f them were recent arrivals in Canada within last year, come from mainland China, and speak Mandarin Chinese as their first language. I gathered data from multiple sources, including audio-taping o f interviews with the participants, video-taping o f the participants’ actual use o f the computer, and collection o f electronic artifacts such as transcripts of online chatting, emails, graphics and webpages. Findings from a multiple case study through triangulating multiple data sources tend to be more convincing. In my study, I positioned myself as a researcher, a teacher, and a learner. At the same time, I was aware that I assumed many other roles, such as the mother o f a teenager, and a member o f the local Chinese community. My multiple roles, my self-awareness o f these roles, and my personal experiences made me reflect constantly on my contributions to the construction o f meaning throughout the research. I applied a mixture of both inductive and deductive approaches in my study. Data analysis was conducted through two phases. I used the qualitative data analysis software package, commonly referred to as NVivo, to analyze the data through three analytical steps—unitizing textual data into idea units, coding categories, and integrating categories. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 79 Chapter Four: Research Findings In this chapter, I delineate the profiles o f the participants, discuss findings by going through the specific steps adopted for data analysis, and provide interpretations o f the participants’ code-switching behaviour through microanalysis o f the excerpts drawn from the electronic artifacts. I also make a comparison o f the code-switching strategies Chinese ESL students employed when engaged in computer-mediated communication. Participants ’ Profiles Amy. Amy is a female student from mainland China. She was 13 years of age and in grade 6 when I interviewed her. She had been in Canada for approximately 5 months, and had been receiving ESL support from the school she attended. She was the only Chinese girl in her school. Her English learning commenced during a one-year stay in the United States when she was in Grade 3, and her mother was pursuing her graduate studies there. She returned to China, and finished her Grade 4 and Grade 5 in her home country. She immigrated to Canada together with her mother when she was in Grade 6. Since her arrival and the commencement o f her school life in Canada, she told me that she felt that she had an enormous amount o f free time to herself due to little homework, as opposed to that in China. As soon as school is over each day, she has time at home to pursue her own interests. What makes her happy is that she feels she is starting to pick up English. By talking with her for the first time at a potluck dinner, I learned that after she returned to China from the United States, she continued learning English by taking an English course offered in her school, and by reading some simple English books, with Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 80 the encouragement o f her mother. She enjoys painting Chinese ink and wash, playing the flute, and working with the computer, doing a number of things on it. The computer has become an important part o f her life since she arrived in Canada. She said her parents did not allow her to play with the computer while in China, because of the huge amount o f homework assigned and the tough competition among the students. The only time available to her to work with the computer was when she had a computer class in school. However, because life in Canada is more relaxed in the sense that she did not have as heavy a load o f homework as she did in China, she can now spend more time doing what interests her. Her suppressed interest in the computer was able to flourish. Yet her mother limited her to two hours a day on the computer. On the basis o f her background, I perceived Amy to be a suitable candidate for my study. The initial conversation and subsequent interviews with her portrayed Amy as vocal, lively, and keen to learn new things, especially as related to computers. These characteristics were evident in her remarks describing her typical activities after school in Canada: After school, I watch TV, do some painting or play the flute. I play computer games, download music, do emails, chat on the Internet, look for graphics, or learn to design my own webpage. The activities on the computer are exciting. When I am at school, all I meet are my classmates in Canada. When I get home, I can email and chat with my Chinese friends. I mainly chat with my former Chinese classmates or relatives who have immigrated to Canada or the US. It makes me feel good. With a computer, life is not boring any more. On the internet, I communicate with them sometimes in Chinese and sometimes in English. Sometimes, she chatted on MSN (a popular Internet chat line) with one o f her cousins who immigrated to Canada and was attending a university. Sometimes she Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 81 joined a chat room in China on the Internet called the Wang Wang English chat room. Her mother learned from others that this chat room was popular among the English learners in China, and was a good place to practice English. Initially Amy did not join the chat room for fear that her English was not good enough, but soon decided to try it out as a way to learn English. She had been participating in the chat room for nearly two months when I interviewed her. She indicated in the interview that she switched to Chinese when she had difficulty expressing herself in English in an email or when engaged in an English conversation online. Baillie. Baillie is a female student from mainland China. She was 19 years old and was attending college when I interviewed her. She came to Canada unaccompanied as an international student, and had been taking ESL courses at the college for eleven months. In China, she began to learn English in Grade 7, and continued to take it until the end of high school. Although she studied English for 6 years, she said she still felt weak at listening, speaking, writing, and grammar, due to little exposure to authentic English in China. She said that her reading is better than her other skills, and that her vocabulary needs to be enlarged. She intends to pursue undergraduate study in business, and wishes to study at a university after satisfying the ESL entry requirements. She has a personal computer, and much o f her communication with friends is through the Internet. After school, she usually reads the news on the Internet, and chats online with her classmates in China and in Northemlight. She chats on MSN more than doing email. She has a Chinese webpage instead of an English one. Baillie is soft-spoken and quiet, but was willing to provide information relevant to my study. When asked what language she uses Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 82 while chatting online or sending emails, and about her opinions about use o f the computer, she said: I chat or do email sometimes in Chinese, sometimes in English. I did my webpage in Chinese only. I use English when my classmates use English with me first. I love the activities on the Internet. They facilitate communication with my classmates. She recounted that she has little trouble conveying some simple ideas in English while engaged in chatting online. When it comes to expressing complicated ideas, difficulties occur, and she tends to switch to Chinese. Cynthia. Cynthia is a female student from mainland China. Cynthia was 25 years o f age and had been in Canada for 7 months, and was pursuing her graduate studies at university when I interviewed her. She came to Canada as an international graduate student, and has been taking graduate courses in her major field ever since. She started to learn English in Grade 7 in China, and had a 10-year history of learning English. After passing the National College Entrance Examination held every year in China, she was admitted to a science program by a Chinese university, and was conferred the degree of Bachelor o f Science after four years o f study at this university. Soon after her graduation from university, she located a job in a company, and prepared in her spare time for the Test o f English as a Foreign Language (TOEFL), as she realized that pursuing graduate studies in an English-speaking country would give her an edge in selection o f future careers. Greatly motivated, she worked hard, and received a TOEFL score o f 598, which met the English language requirement of the university she attends. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 83 After more than two years’ working experience in China, she came to Canada and switched from working to being a full-time graduate student. She said that, since her arrival in Canada, she sensed that the English she learned in China was insufficient to enable her to communicate fluently with others, verbally or in writing. She also said that her English language insufficiency posed challenges for her academic studies in the English language used at the Canadian university she attends. Therefore, she often asks her Canadian classmates or the Learning Skills Center at the university to help her with her writing. She has a personal computer, which has been both a great help with her studies and an instrumental means o f communication with her family, friends and classmates. She described herself as a person who enjoys making friends and sharing things with them. She impressed me as sociable, articulate, bright, and eager to learn. Her description o f her typical after-school activity showed that chatting on line had been an indispensable means o f communication in her life. I usually work on my assignments and chat on the Internet after school. I like the chatting software very much. It has brought convenience to people, especially to us students. It is pretty expensive for me to communicate with my families and friends in China via phone. Chatting on MSN is also a kind of fun added to my boring life, a way to kill time and have fun. She also sends emails to her Chinese friends in Chinese and Canadian friends in English. In response to a question concerning the languages she may use to chat, she remarked that she chatted mostly with her Chinese friends and classmates, sometimes in Chinese, and sometimes in English. She indicated that she had difficulty expressing herself in English time and again when talking in English. When this happened, she would Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 84 alternate English and Chinese. It is a common practice for her to use the two languages alternately. Emma. Emma is Amy’s mother. She immigrated to Canada together with her daughter. She was an engineer in China, and set up her own company there. She registered and established a corporation soon after she arrived in Canada. She pursued her graduate studies, and obtained a Master’s degree in chemistry in the United States. It was during that time that her daughter visited and stayed with her for one year. Upon completion o f her studies, she and Amy returned to China. Two years later, they came to Canada, as independent immigrants. Her ambition was to expand her business internationally, and more importantly, to seek a better education for Amy in Canada. Like most Chinese parents, she attaches great importance to her daughter’s education. She perceives that education in Canada is better, in the sense that it provides a relaxed environment which allows students to develop their interests, and ultimately realize their potential. In her opinion, Amy’s improvement in English is a top priority. As a result, she is very supportive o f activities which she feels could facilitate English literacy development. She encourages Amy to communicate as much as possible with her classmates at school, to write English emails to her classmates and friends in China, and to design an English webpage. She considers these online activities, if carried out properly, beneficial to developing Amy’s English proficiency. She said, “Doing emails and chatting in English gives Amy an opportunity to practice her writing. Chatting in English can foster in her a way o f thinking quickly in English. This influence is immeasurable.” At the same time, she was worried about the safety of the chat room in Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 85 that the chat room was usually full o f strangers. Therefore, she constantly reminded Amy that the purpose o f chatting there was to practice English, and taught Amy some self-protective measures. I interviewed Emma in order to obtain a detailed account of Amy’s computer use as well as her attitude towards computer use. Data Analysis Data analysis in the present study underwent three steps. Findings emerged during the data analysis processes, and I simultaneously provided interpretations while developing and explicating the analysis processes. The first step in the analysis involved a careful reading of each transcript and artifact, and identifying units o f data. Units of data are defined as pieces o f fieldnotes, transcripts, or documents that fall under a particular topic (Bogdan & Biklen, 2003). According to Lincoln and Guba (1985), what is taken as a unit has two characteristics. First, a unit is heuristic in the sense that it enables the researcher to have some understanding or to take some action. Second, it is the smallest piece o f information about something that is interpretable in the absence of any additional information (Lincoln & Guba, 1985). These units o f data each conveyed a particular meaning, and were then assigned a preliminary coding category. The data units were informative in the sense that they enlightened me on some specific issues relevant to my study. The units that I identified for coding varied from a few words or phrases to a single sentence or several sentences, to paragraphs (Bogdan & Biklen, 2003; Lincoln & Guba, 1985; Miles & Huberman, 1994). The coding categories I created each represented a particular topic Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 86 as presented by the data. I also wrote down thoughts and ideas that sketched out relationships I noticed (Miles & Huberman, 1994). For the actual NVivo analysis, I started by creating a project in which to store data, thoughts, key concepts, and links between them. I then imported all the document files o f verbatim interview transcripts, processed and saved as rich text format (rtf), into the project. I did not import the electronic artifacts provided by the student participants. Instead, I manually coded them for the reason that the NVivo software cannot display the Chinese characters in the artifacts, thus rendering computer coding impossible. I manually conducted a microanalysis o f verbal interactions which contained code-switching in the artifacts, such as chat room or MSN data, emails, webpages or graphic data. During the NVivo coding process, I used the coder window instead o f other coding options because I could easily and quickly change or delete codes as needed. I highlighted the text segment that I wanted to code, and then entered the desired code name in the space provided. Then I saved the code at a node. Because my purpose at this stage was to break down data into meaningful units, the codes lacked structure and all the nodes were, therefore, saved as free nodes (non-hierarchical categories). I proceeded with the process o f highlighting and coding until I had coded all units o f data in one transcript before moving on to the next. Because the coder window displays the node names o f the already created nodes, text segments from other transcripts that qualified to be coded at an existing node were simply selected and coded at the appropriate nodes. New free nodes were created to code text segments from subsequent transcripts that Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 87 expressed new ideas. This process generated a long list o f free nodes which I believed adequately covered my data. Step two o f the data analysis involved sorting the initially coded generic idea units into categories. This process utilized a priori categories, for example, effective communication, lexical gap, playful use o f language, and identity construction. As well, I developed new categories that were grounded in the data. In this way, I operationalized code-switching. I created some code categories as prespecified or a priori categories (Miles & Huberman, 1994) from theoretical frameworks, the literature, and the research questions. I assigned them to the units o f data that fell under the particular topic represented by the coding category (Miles & Huberman, 1994; Dogdan & Biklen, 2003). By so doing, I was able to discover the usefulness of the a priori categories I had created, and modify them. These a priori codes featured extensively in the participants’ interviews and artifacts, making them important components o f themes that emerged from the data. At the same time I developed new descriptive categories, which highlighted the similarities and differences among units o f data. Data units that were similar in meaning were grouped together under one category and different ones formed the basis for a different category. The development of coding categories is, therefore, an iterative cycle where the analyst re-examines the categories over and over until all data units have been coded under some category (Miles & Huberman, 1994). It was also a common practice in my analysis to sort one segment o f text simultaneously into several different categories, as suggested by Bogdan and Biklen (2003). It was possible for units o f data Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 88 to fall under multiple codes and for the coding segments to overlap each other. For example, in “aiya, I forgot”, aiya is a Chinese interjection, indicating here that the speaker suddenly remembered that she forgot something. This code-switching fell simultaneously under the categories o f lexical gap and effective communication. In order to examine and assess similarities and differences among idea units from all the transcripts, I used NodeLink in NVivo. This feature is a powerful tool that automatically links all the nodes and data units coded at them to all the transcripts in the project. Through Node Explorer, I obtained access to NodeLink, which linked me to the tool, Browse Node. This linkage enabled me to browse each node to see what codes were common across documents. Thus I carried out a node-by-node comparison across participants’ data to identify the similarities and differences o f idea units coded at them. I assigned segments that conveyed similar ideas under an a priori category or a newly developed category. I also merged codes that I felt conveyed the same ideas but were grouped under different categories. After re-examining and re-organizing these categories, I developed 14 categories, each of which represented a particular topic. These categories were: Chinese Character, Pinyin, Graphic, Lexical Gap, Identity Construction, Effective Communication, Topic Change, Rhetorical Effect, Informal Situation, Vocabulary, Listening Comprehension, Speaking, Writing, and Grammar. I used the DataBite link feature in NVivo to annotate any thoughts that I had about particular data units. Most of these annotations were interpretive or inferential in nature, making them useful in further data analysis. As well, I created a memo in which I wrote a description o f each category to assist me in maintaining a clear distinction between Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 89 categories, and thoughts and ideas that emerged about particular categories as I created them. These thoughts and ideas were revisited as the analysis progressed. At this stage, using the same code categories, I manually coded the artifacts provided by each student participant, and wrote my thoughts and ideas about the text segments on the margin beside them. Through coding the artifacts, I discovered a frequently occurring topic, which was not mentioned by any of the participants in the interviews, thereby developing one new important category, namely, Structural Gap. I added this category to the free nodes in NVivo for the purpose o f eventually developing visual representations o f the findings. Consequently the total number o f code categories amounted to 15 as shown in Table 2. In order to draw a clear picture of the contexts o f code-switching, I will provide an example from the data of each context of code-switching, as shown in Table 3. Step three o f the data analysis involved integrating code categories by determining conceptual relationships within and among the assorted categories (Lincoln & Guba, 1985). To establish the relationships, I reviewed each category and its text segment, and then conducted a cross-category comparison, in an attempt to identify which categories applied to only a single participant and which ones that ran across two or all three o f the student participants. Such a comparison is in line with the purpose o f the study—to examine the contexts o f code-switching, and compare the similarities and differences among the three participants in code-switching. For example, the category Identity Construction ran across all the student participants, whereas the category Graphic existed only in the transcripts for one participant. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC Table 2. Code Categories Categories_______________________________ Descriptions Chinese characters use o f Chinese words to code-switch Pinyin use o f Romanized Chinese to code-switch Graphic pictures with English and Chinese characters on them Lexical gap code-switching due to an interlocutor’s inability to use an English word Structural gap code-switching due to an interlocutor’s inability to construct a sentence in English Identity construction code-switching to construct an identity Effective code-switching due to a need to express an idea communication accurately Topic change code-switching due to the change of a topic Rhetorical effect code-switching for fun or to create humour Informal situation code-switching in informal situations Vocabulary the participant’s perceived increase in English vocabulary due to code-switching Listening the participant’s perceived improvement in English comprehension listening comprehension Speaking the participant’s perceived improvement in the ability to speak English Writing the participant’s perceived improvement in the ability write in English Grammar the participant’s perceived improvement in English grammar Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90 Code-switching in CMC 91 Table 3. Contexts o f Code-switching Contexts_____________________ Examples______________________________ Lexical gap When the plane was landing, I saw the city’s quanmao (a bird’s-eye view)1. Structural gap Ok. Fang ni yi ma. (I’ll let you off this time.) Identity construction A: My nose need to smell the food ma B: ok de, but how. (ma and de are Chinese modal particles used to indicate the tones of the utterances.) Effective communication A: You miss Chinese food. B: aiya, it is time to watch TV. I am leaving, (aiya indicates surprise.) Rhetorical effect u made me laugh 4 le. (The Chinese pronunciation of 4 has the same meaning o f death. The sentence means you made me laugh to death) Topic change A: help, help!!!! dC=(k2-klC)dt, t=0 Bt C=C0, k l fa k2 M m f/M B: OMG. I don’t know. (The second part o f A ’s utterance means that when dC=(k2-klC)dt, t=0, C=C0 and k l and k2 are constant. Solve the differential equation.) Informal situation This context was only discussed in the ______________________________ interview.______________________________ Note 1. Words or sentences in the parentheses are my translations or explanations. An examination of all the categories enabled me to find patterns among them. All the identified relevant categories were then grouped together by pattern codes according to their similarities. The pattern codes are superordinate categories o f code-switching acts, or overarching themes. The same procedure was implemented in the analysis o f the artifacts. As a result, themes were developed both from the research questions and from the accounts o f the participants. The pattern coding yielded three overarching themes, namely: Means o f code-switching, Contexts o f code-switching, and English language development. I made a within-case and cross-case analysis o f the coded transcripts and Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 92 artifacts to identify the similarities and differences among the student participants in code-switching, which is shown in Table 4. Table 4. Similarities and Differences in Code-switching Amy(13)‘ Baillie(20) Cynthia(25) Means of CS Chinese Character V V V Pinyin V V V Graphic V Contexts of CS Lexical gap V V V Structural gap V V V Identity construction V V V Effective communication V V Topic change V Rhetorical effect V V Informal situation V English language development Vocabulary V V V Listening comprehension V Speaking V V V Writing V V V Grammar V Note I. The numbers in parentheses after the three names indicate the age o f the participants. W ith all the free codes displayed in the node explorer, I created top-level tree nodes under the titles o f the three overarching themes to store the various code categories. I selected a free node with a particular code category, such as lexical gap, and dragged it to the corresponding tree node, such as contexts o f code-switching, as a c h ild , th en d e le te d th e free c o d e from th e fre e node area. I continued with this process until almost all the free nodes were dragged to the tree nodes into which they fit. Those free nodes that could not fit into any of the three nodes, however, were retained for further consideration. By the end of this process, I had a total of 3 nodes, 15 sub-tree Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 93 nodes, and 4 free nodes. O f these tree nodes, the three themes were the superordinate (parent) nodes, and 15 code categories formed the subordinate (child) nodes. The 4 free nodes dealt with the participants’ attitude towards use o f the computer, and included Attitudes toward Computer Use, Advantages o f Using Computers, Disadvantages o f Using Computers, and Activities on the Internet. They were used to describe the participants’ profiles. In order to explore relationships and speculate about the findings o f the study, I used the Modeler to model the patterns to assist with visualizing a conceptual framework. With Model Explorer open, I established the name o f the model by selecting Properties, and giving a name to it. Then through Add Node, I selected the three tree nodes respectively. After speculating about the relationships among the three nodes, I used Add Link to show the connection among the three nodes. Figure 2 shows the visual representation o f the findings o f the study. It indicates the three overarching themes emerging from the data, each with its subordinate code categories. The first theme, means o f code-switching, involves the three ways the participants employed to code-switch. The second theme, contexts o f code-switching, represents the seven contexts in which code-switching occurred. The first and second themes are related to the third theme, English language development, which reflects the participants’ perceived improvement in five aspects o f English skills due to code-switching between English and Chinese. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 94 gra im a vocabula writing speaking listening com prehension angi developm ent C o n te x v o f code-switching M eans of codfFswitching lexical gap C hinese characti structural g a p ^ 1 informal situation rhetorical effect ^ p o p i c ch an g e graphic Pinyin identity construction effective com m unication Figure 2. Code-switching and English Language Development NVivo eased the task of coding and identifying relationships among categories. The linkage to units of data and text segments of all participants’ data presented the extent to which code categories were shared by participants. Codes and categories were well organized and easily retrieved for further analysis. Their re-organization, if required, was rendered fast and less laborious. Writing the Chinese Language on Computers The three overarching themes that emerged from the student participants’ data—means o f code-switching, contexts o f code-switching, and English language Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 95 development, represent the means and contexts o f Chinese ESL students’ code-switching in CMC as well as their perceptions of the ways in which code-switching help them improve their English. In the following section, I will first present a brief description o f the Chinese language, in particular, its writing and pronunciation systems, in order to help readers understand the participants’ choice o f means o f CS, and then describe and explain in detail each o f the themes, using quotes from the student participants’ interviews, and excerpts from their artifacts, with a view to illustrate the factors that affected Chinese ESL students’ code-switching in CMC, and its relevance to their English language development. The Chinese written language has no alphabet, and consists o f a logographic writing system, which employs a large number of symbols, known as characters, to represent individual words or morphemes. One or two characters correspond roughly to one word. For example, there is one Chinese character, tK, meaning water, and two Chinese characters, 5515, meaning English. These symbols, in turn, are given a pronunciation, written as Pinyin, but the pronunciation varies from dialect to dialect while the meaning is constant and Pinyin used is the same in all dialects except in Cantonese. Pinyin is the international standard romanization scheme for Chinese characters, which is merely a tool for learning pronunciation, not a replacement for characters. There are four tones in Pinyin, and one Pinyin usually represents many Chinese characters. For example, han with the fourth tone can represent Chinese characters which may mean sweat, drought, brave, Chinese, and so forth. Thus you have Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 96 to learn the Chinese characters together with the correct tones to master the written language. When it comes to typing Chinese characters using standard keyboards, the Pinyin input method is used. However, when users type Pinyin on the computer, they do not type the four tone markers, because typing the markers requires a new font, which many computers do not have, and it is complicated and time-comsuming to put tone markers above Pinyin. When one toneless Pinyin is typed, a menu o f all the Chinese characters which share the same Pinyin, but with the four tones, will show up, and the user needs to select from this long menu the characters he/she intends, based on the context. Therefore, it takes more time to type Chinese characters than Pinyin. However, characters can convey an idea accurately and the meanings they express are usually clear at a glance. When Chinese people communicate with each other on the Internet, they may prefer Pinyin, if they express themselves using short sentences. They may prefer Chinese characters, if they need to express themselves using long sentences. This is because the interlocutors have to spend much more time figuring out the meanings o f a long sentence written in toneless Pinyin than in Chinese characters. A long sentence written in Pinyin without tone markers can lead to confusion (see the example in the section o f Chinese characters). Means o f Code-switching This theme indicated the different ways Chinese ESL students code-switched from English to Chinese in CMC. These means o f code-switching included inserting Chinese Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 97 characters, typing segments of text in Pinyin, and employing graphics with Chinese characters written on them. Chinese characters. The student participants’ data, including the transcripts and artifacts, showed that they all used Chinese characters as a way o f code-switching to a varying extent. Their artifacts show that they inserted Chinese words or phrases into English sentences at the intra-sentential level. Some entire sentences were written in Chinese at the inter-sentential level. Amy’s artifacts indicated that Chinese characters only occurred within the sentences that were part of the graphics that she produced (see Figure 3.), and Pinyin was the major means o f CS for her. This was consistent with her remarks in the interviews. She said, “It is much less troublesome for me to type Pinyin. So I prefer to use Pinyin to replace those English words or phrases I don’t know”. Asked during the follow-up inquiry why she incorporated characters in her graphic, she replied, I don’t know how to express some words and phrases, so I used Chinese to replace them. The mixture o f English and characters makes the picture look funnier, and I can express myself in a more vivid way with the Chinese characters. From her remarks, it could be inferred that Amy viewed Chinese characters as a more effective way o f expressing herself compared with Pinyin, yet Pinyin was more expedient in many situations. Baillie, on the other hand, preferred Chinese characters to Pinyin. This is how she explained her preference, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 98 I would rather use Chinese characters, because I will sound shallow if I use Pinyin. Chinese characters look sophisticated and can fully express my feelings. I just feel better using characters. Cynthia indicated in the interviews that she would rather use Chinese characters if she needed to express herself with a whole sentence she found hard to say in English or if she wanted to express an idea accurately. The reason for using Chinese characters, as she explained, was that a long sentence in Pinyin very likely results in misunderstanding due to the nature o f the Chinese language. Her artifacts confirmed her remarks in the interviews. For example, the Chinese character code-switching, 7^$}[£], means that “I steamed rice flour pork last night”, in which §£ corresponds to I, last night, HIT to steamed, and to rice flour pork. Wo zuo wan zheng le mi fen rou is the Pinyin representation o f the Chinese sentence corresponds to $t, zuo wan to to in which wo zheng le to M T , and mi fen rou to :/£$)$]. In this sentence in toneless Pinyin, the key word which can cause less accuracy in conveying ideas is zheng, which means steaming if with the first tone, and means cooking in colloquial Northern Mandarin Chinese if with the third tone. Therefore, the sentence in Pinyin can denote the meaning “I steamed rice flour pork last night” or “I cooked rice flour pork last night”. Cynthia also mentioned that if she and the chatting partner had a discussion about a topic which required using lots o f specific terms, such as cooking terms, the whole conversation would be switched from English to Chinese. There was a consensus among the three participants, whether expressed explicitly or implicitly, that ideographic Chinese characters are expressive in that they transmit messages accurately and vividly. Baillie’s comment that Chinese characters can make a Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 99 person look sophisticated indicates that she employed Chinese characters as a tool to position herself and represent who she was in that moment, thereby constructing her socially situated identity. Pinyin. Both Amy and Cynthia indicated in their interviews that they used Pinyin as a means o f code-switching, and their artifacts confirmed their remarks. They both considered that the quick speed o f typing Pinyin as contributing to this choice. When asked about the reason for choosing Pinyin, Amy related, “Using Pinyin is easy and efficient” Cynthia also attested to this in her words, “Personally, I use Pinyin.. ..I cannot type characters fast.” For example, in the sentence “If you put potato in mian tiao, you must be crazy”, mian tiao, meaning noodles, is written in Pinyin,. Baillie claimed in the interviews that she only used Chinese characters for code-switching. When I examined her artifacts, I found that she did adopt Pinyin to type some Chinese modal particles. She said that when chatting in English, all Chinese chatters used Pinyin if they wanted to express their tones transmitted by Chinese modal particles, with almost no exception. This situation was similar to the one discussed by Gumperz (1982) in that there are discrepancies between speakers’ descriptions o f their own bilingual usage and tape-recordings of their informal talk with the occurrences of code-switching. The discrepancy between the interview and the artifact further confirmed the claim by Yin (1994) that data from multiple sources are likely to be more accurate. Bailie’s artifacts revealed that she drew on Chinese characters to replace nouns, verbs, phrases, or sentences, but used Pinyin to replace the Chinese modal particles (see Identity Construction in later section for further information about modal Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 100 particles). A possible explanation was that she personally viewed the replacement by characters o f nouns, verbs, or sentences as code-switching, whereas use o f Chinese modal particles was not considered to be code-switching. The follow-up inquiry via telephone suggested that typing Chinese modal particles in Pinyin was already a habit for these chatters, and therefore, she subconsciously conformed to this habit. Graphics. Amy was the only participant among the three who used graphics as a means o f code-switching. She tended to search websites for funny graphics and wrote English sentences which contained some Chinese characters or Pinyin onto the graphic image in order to create a special effect on communication. Then she sent these graphics to her classmates or friends. One o f her graphics (see Figure 3.), entitled La Monolibean, was a transformed portrait o f Mona Lisa, on which Mona Lisa was holding a toy bear, with her face replaced by Mr. Beans’ funny smiling face. The words written on it read: “U didn’t reply my l / r f . Well my ftflr for u is Make u laugh— U. ” (You didn’t reply to my message. Well, my reward for you is to make you laugh to death.) Amy wanted to chat with one o f her former classmates and friend in China at QQ, a Chinese chatting site, but had failed to find her there. As a result, she left her friend a message at QQ, in the hope that her friend would reply to this message. Not having received any response from her friend, she sent this graphic to her. This novel means enabled her to convey exactly and visually what she wished to express. As she explained, “This graphic can help me express my feelings more vividly”. In mediating her activities, she employed both material tools such as computer, and psychological tools such as language, to achieve what she desired. The uniqueness in her means o f Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 101 code-switching, among the three student participants, may be attributed to the fact that she enjoys painting, she has more free time available to search for such graphics, or she has a lively personality and imagination corresponding to her age. Another explanation is that she, as the youngest participant, is growing up as a digital native (Prensky, 2000) and becoming proficient in multimodal literacy. Figure 3. La MonaLibean Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 102 Figure 3 and Excerpt 5 in the later section also present the use o f electronic communication conventions (e.g., the acronym OMG for Oh my god, u for you and ur for your). In real-time synchronous communication (e.g., chatting online), participants send messages that are brief; informal with regard to the conventions o f spelling, grammar, and punctuation; and utilize short-cuts (Lapadat, 2002). The pressure to be brief and rapid results in the decreased premium on using standard conventions of writing and the prevalence o f typographical errors (Lapadat, 2002). Such replacement of, for example, you by u, pertains to the kind of code-switching between two different styles o f speech o f the same language (Gumperz, 1982), as in this case o f switching between formal and informal English. This loosening o f the conventions o f written English in a sense gives permission for easy code-switching between languages, and the ready use o f homophones (see Rhetoric Effect for a definition and more information). No matter what means they employed to code-switch, the process o f their code-switching reflects that negotiation o f meaning is going on in order to avoid a breakdown in communicative and to achieve particular communication aims. For example, Amy’s communication intent in the example above was more than just avoiding communication breakdown. Rather, she was conveying her eagerness to talk humorously and multimodally. Meanwhile during negotiation, social alignment is formed, and identity constructed among the participants o f online communication. Therefore, the participants utilized different Chinese writing systems to code-switch for different communicative and social purposes. The following section, from interactionist and sociocultural perspectives, will illustrate with excerpts from the participants’ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 103 electronic artifacts how they responded to input from their partners and modified their output during negotiation of meaning. I also discuss what social factors are involved in this process. Contexts o f Code-switching This theme represents the various contexts in which Chinese ESL students resorted to their native language to resolve problems or tackle issues emerging during communication in English. These situations that elicited code-switching not only stemmed from linguistic difficulties, but also reflected social and cultural aspects of communication and language acquisition. In many ways, this theme most reveals the breadth and complexity o f code-switching. Lexical gap. In the analysis of the participants’ chat online, it was clear that code-switching triggered negotiation of meaning, in which scaffolded help was provided in the Zone o f Proximal Development (ZPD). Without exception, all three participants acknowledged in the interviews that they used their LI when confronted with an English lexical challenge (Crystal, 1987). Their chat records demonstrated that they switched to Chinese when they had trouble with English words or phrases. Amy, Baillie, and Cynthia were new arrivals to Canada, and their English vocabulary was not sufficient to enable them to express themselves freely. It is understandable that lexical difficulties arose during communication online, especially when they had to produce a quick response to the input from their partners in a chat online. The following excerpts illustrate how code-switching functioned as a mediating device for the novice to enlist help, and a more capable person to provide scaffolding. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 104 Excerpt 1 is a conversation between lulu (Amy’s nickname) and xueqi, Amy’s cousin who was a new immigrant and is pursuing her undergraduate studies in Canada. In this excerpt, lulu and xueqi were talking about a new movie. Excerpt 1. (from Amy’s MSN chat) x u e q i - C o n v e rsa tio n Fie £dit Actions Toots Help j.I Irivite itiftd Fites 'Video t ' rVoico Activities . 45/ GSties" T o ri' lu!u s a y s : w hat' s n a n i ya chuanqi lulu s a y s : I wanna p la y th e game x ia n j i a n q i x i a zhuarr«w x u e q i s&yy. such o l d game x u e q i so y s- e new movie lulu so y s: a new movie, C hinese o r E n g lish ? What yan yuan i n it*! xueqi says' about a le g e n d x u e q i so y s: e n g lis h iuiti s a y s : what i s a legend? xueqi s a y s ' i d on' t know th o s e a c to r s xueqi say s: chuan qi lulu s&vs: J3 A O- - #1. A * \ s s t rrr s s a o e r e c c e d a t 7 22 PM on i t x a y Vi i*i h Iu m i1 . M t i * M i u p 4. ji r ... I s' . . . . . . . . . . . - 1 -s» i a a m n ia i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC xueqi 105 C o n v e rs a tio n File- Edit Actions Tools 11 * •• i • . Tt., nueq» < ^ o x iz h a ^ q lW lis» » £ n '^ .-cr '> ft chuan g i lulu s a y s : o h ,. . . . i\i- ‘ i se e -'S^ lulu s a y s : I don’ t l i k e legend lulu s a y s : D’ u know J e n n i f e r A niston xueqi sa y s. t h i s movie i aw m atching i s a l i t t l e x i a rest xueqi s a y s $mm. y e s j l i k e h e r v e r y much lulu s a y s : i s t h a t a g u i s t o ry W lulu s a y s : I th in k she and A n g ilin a i s qing d i ^ P xueqi sa y s. no • * j u s t m is e r a b le * s b d ie d xueqi sa y s' y e s th e y are A $ 5 V o w (.Sip jV /ir h - - S p* * * $5) * %■<& In the first turn, (in this study I used turn to refer to one line in the conversation in order for readers to match conveniently the discussion with the chat shown on the screen), lulu asked in English mixed with Pinyin, the Romanized Chinese, “what’s na ni ya chuangqi?” (what is Na Ni Ya legend?). In the second turn, lulu asked about playing the computer game entitled “xian jian qi xia zhuan” (Story o f Seven Chivalrous Men with Immortal Swords). In the first two turns, lulu overtly addressed the problem of accessing the lexical items non-existent in her English vocabulary repertoire to express her ideas. As a result, she mediated her interaction by switching to Pinyin, to make herself understood. Switching to Chinese may heighten lulu’s metalinguistic awareness o f her L2 vocabulary development (Blake, 2000), and push her to pay more attention to the word in the target language (Long, 1996; Swain 1995) as long as it was contributed by the chat partner. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 106 It seemed that xueqi did not know the English name of the game either, and offered a simple response “such old game” to tell lulu that she was not interested in playing the game with her. Then in xueqi’s next turn, she proceeded with the topic about the movie, Na Ni Ya legend, lulu initiated by saying “a new movie”. Legend was not a new word to her. So, in her third turn, she built upon the topic of legends, using the word, legend. In lulu’s third turn, she wondered about who were the actors in the movie by saying “what yan yuan (actors) in it”. Lulu’s forth turn signaled that she still had difficulty understanding the word legend, so she started negotiating by enlisting help (“what is a legend”) in order to clarify the meaning o f the unknown word. Xueqi, in her fifth turn, produced her output by providing the English word, actor, in response to lulu’s question “what yan yuan (actors) in it”. Then in her sixth turn, xueqi provided linguistic modifications in the input by explaining legend in Romanized Chinese “chuan qi” to make her message comprehensible. In so doing, xueqi deployed her LI as a mediating tool to provide scaffolding to lulu possibly because xueqi herself was experiencing difficulty in paraphrasing legend in English. It is also possible that xueqi did not want to take the trouble to relate legend in English, so she reverted to her LI for the sake o f expedience. Lulu’s utterance “o h .. .1 see” signaled that mutual understanding was established collaboratively. Language input and output were provided during the turns o f the utterances. To keep the conversation going, lulu was pushed by the feedback she received from xueqi to form a comprehensible output “I don’t like legend”. Xueqi’s output in turn had become lulu’s input which pushed lulu to form this output. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 107 What is worth noticing is xueqi’s utterance on the second screen: “This movie I am watching is a little xia ren”, in which xueqi reverted to her LI xia ren (scary). The possible reason for xueqi’s code-switching was that being aware of lulu’s limited vocabulary, xueqi was mediating the interaction through the use of LI to accommodate lulu’s language competence. In this case, it seems that code-switching was employed as a contextualization strategy by xueqi for discourse purposes (Gumperz, 1982). Understanding xueqi’s utterance, lulu raised a question “is that a gui story” (is that a ghost story?), and then continued the topic about Jennifer Aniston by stating “I think she and Angilina is qing d r (I think she and Angilina are rivals in love). In these two utterances o f lulu’s, Pinyin was exploited as a mediating tool for lulu to convey her ideas. The exchanges between lulu and xueqi revealed that plenty o f negotiation of meaning and mediation by code-switching occurred, through which learning seems to be taking place in lulu’s ZPD (Vygotsgy, 1934/1978). The linguistic modifications in this communicative interaction, triggered mainly by lexical confusion, appear to facilitate second language acquisition (Smith, 2004; Tudini, 2003). This facilitation of language acquisition was also realized through scaffolding provided by a more capable person, xueqi. Learning may take place during the interaction and negotiation of meaning between lulu and xueqi. On the other hand, the interaction and negotiation o f meaning occurred in a social context where lulu and xueqi used code-switching strategies to comment on films and cultural legends, as a result o f which communication became meaningful. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 108 The following excerpt is a conversation between Cynthia and one o f her former classmates in China. The first part provides the background of the conversation, and the second part is on the screens The exchanges in Excerpt 2 were between Sissy-painful, Cynthia’s nickname and love World Cup, a former classmate of hers, a male English interpreter in China. The excerpt started with Sissy-painful talking about her left hand which she cut while cooking. Excerpt 2. (from Cynthia’s MSN chat) Sissy-painful says: © I cut my finger the day before yesterday. Sissy-painful says: © a n d a little serious. Sissy-painful says: - 'very painful. love World Cup says: the cutting ocurred during cooking? Sissy-painful says: © Y es. It was bleeding all the time. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC * lo v e W o rld C u p C o n v e r s a ti o n xb Lj r . # It A * W 109 «■. mri* 1 i ’* ' J A ‘ to r, trf-n - Ti> io v e W orld L u p <\iaox>2>'s~519t5l ^ ^ ' n a i c-«i> love W orld Clip sa y s. >* S issy-painful sa y s. I P it stopped S issy-painful sa y s: %-'no. I lost a p ie c e o fz h i jia S issy-painful savs^ U s o v s t y painful love W orld Cup s a y s. w ell th e bottom lane i s i t i s n o t on th e r ig h t hand S issy-painful sa y s. Curtis the left hand iove W orld Cup sa y s: you always sa ke c a r e le s s m ista kes S issy-painful sa y s. W *I felt a great pain when Ixiao yon Sissy-pairiful s a y s w d id I S issy-painful sa y s. ^llwhen love W orld O jp sa y s: I ....... ^ Lavalife has more single men than anyone else in Canada h JC %< (fcf-& J 0 T.-vifce 5 < n i:lF -« :i V i-Jeo Vwe A tfv fV v ^ O s- i m * “ ^ h h h ' 1 iuvwvsoUo v-up says. when we are sin o u r i n high school Sissy-painful caysw p a in fu l m em oiy Sissy-painful sa y s. IPl alw ays try to forget the bad things h appened in m y life love W orld Cup s a y s ' none ' .. j o i c y memory * .. the fu n i s so pure Sissy-painful s a y s. love W orld Cup says- th e s e memories donot appeal to your heart? Sissy-painful s ay s. '^ s o m e tim e s S issy-painful s a y s ' iH b u t I alw ays like to look forward iove W orld Cup s a y s. I am n o s ta lg ic *• yeah love W orld Cup says- i ca st back o ld tim es wore o fte n than y o u do . liy o u didth) na My display picture. C k k to change the l.«i-nmMur *ii.fiapmNrtraaa .com i~ » / >.-.■■.■■■.. "1 I just finished one assignment""" softough says: a little tired Wm*m says: oh. I tiave assignm ent too. f i S t a : * * says it will due soon, sofiough says. being astu d en t is no easy f t t e * * saysyes a « * i » : softough says: sometimes I w ant to go back kindergarden *m *#l sa/s: you are dreaming ba softough soys I am serious Mi . Sm wim siijijtm M e'f. mmmZo %fm%js, im ^ & n m M o m-mm* w$m%mm&m, b h im , r a ttwft, (2 5 0 -960-5889) (2 50-615-5578) W k ° ® # B W rn JE , m\. im m m tm i • fe lfiii# (2 5 0 -9 6 0 -5 8 2 0 ) ( officeofresearch@;unbc.ca) » mmw Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 194 Appendix C: Consent Forms Parent’s Consent Form I________________________ agree to participate, and give my permission for my child___________________ to participate in the study o f Chinese English-as-a-Second-Language (ESL) students’ code-switching while engaged in Computer-mediated Communication at home, and to take the test administered by Yanping Cui to determine the English proficiency o f my child. The information in the attached letter regarding the nature of the study and participation responsibilities was (choose one): Read by myself □ read to me □ translated for me □ I have understood the information and my signature below indicates my approval and my permission for my child to participate in the study and for these data to be used towards the completion o f Yanping Cui’s thesis report for a Master’s degree in education, as described in the attached letter. Signature_________________________ Date___________________ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC 195 Adult Participant’s Consent Form I__________________________ agree to participate in the study o f Chinese English-as-a-Second-Language (ESL) students’ code-switching while engaged in Computer-mediated Communication at home, and to take the test administered by Yanping Cui to determine my English proficiency. The information in the attached letter regarding the nature o f the study and participation responsibilities was (choose one): Read by m yself □ read to me □ translated for me □ I have understood the information and my signature below indicates my approval to participate in the study and for these data to be used towards the completion o f Yanping Cui’s thesis report for a Master’s degree in education, as described in the attached letter. Signature__________________________ Date___________________ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Code-switching in CMC H f tC 196 |W];a^ _____________ mmimmwxm m s^tfi 3^B$T&£HfiD SiJA^a^Wn MAmm£$tMo & £ _____________________ 083_______________________ '^'kf.iJ.SA ®______________ m n&cLmmmn u m m s w * m ^ m ' i m & m ' V i m tea* ^lAmm^mn ±ife3to & £ _____________________ BM_______________________ Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.