Classroom code-switching Code switching, Guide, Progetti e Ricerche di Linguistica Inglese

Scarica Classroom code-switching Code switching e più Guide, Progetti e Ricerche in PDF di Linguistica Inglese solo su Docsity! DOI 10.1515/applirev-2013-0009   Applied Linguistics Review 2013; 4(1): 195 – 218 Angel Lin Classroom code-switching: three decades of research Abstract: In this paper I provide a review of the historical development of differ- ent research paradigms and approaches adopted in studies on classroom code- switching. I also discuss the difficulties and problems faced by this field of studies and share some of my own critical reflections on how this field might move for- ward in the future, speaking from the position of a researcher who has been en- gaged in this area of studies for close to three decades. Keywords: classroom code-switching, classroom research paradigms, interac- tional sociolinguistics, disciplinary bilinguals, translanguaging, L1 use in L2 classrooms, bilingual classrooms Angel Lin: The University of Hong Kong. E-mail: angellin@hku.hk Introduction: Classroom Code-switching – Delimiting the Field We all seem to know what classroom code-switching is about. For example, one can easily define classroom code-switching as language alternation – the alter- nating use of more than one linguistic code in the classroom by any of the class- room participants (e.g., teacher, students, teacher aide), and this can include both code-mixing (intra-clausal/sentential alternation) and code-switching (alternation at the inter-clausal/sentential level) (Lin, 1990, 2008). However, whether we refer to it as code-mixing, switching or alternation, this “code-X” ter- minology begs the question of whether language should, in the first place, be conceptualized as discrete “codes” with stable boundaries. The term, “code”, in linguistics has been borrowed from information theory, and Alvarez-Caccamo (2001) delineates the original and derived usage of the term as follows: In information theory, a code is a mechanism to pair two sets of signals in non-ambiguous, reversible, and context-free ways. For instance, in morse code the letter “s” is always ren- dered as three dots, regardless of particular circumstances (context independence); “s” can only be rendered as three dots (non-ambiguity); and three dots are always to be understood 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 195–218  2708_4-1_09  (p. 195) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 196   Angel Lin as “s” (reversibility). . . . This “code” notion was systematically applied to speech first by information theorists (Fano) and, then, fundamentally, by Roman Jacobson. Jacobson re- framed Saussure’s langue/parole dichotomy in terms of code/message. In this model, the speech signals would match “meanings” in the linguistic “code,” equivalent here to “gram- mar.” However, Jacobson’s model is not exempt from ambiguities, loose ends, and perhaps contradictions. . . . Inferential views of communication propose that most understanding depends on the particulars of the relationship between literal contents and contexts . . . this has led to a disabling of the applicability of the “code model” to human communication. (Alvarez- Caccamo, 2001, p. 23–24) Recent years have further witnessed increasingly poststructuralist views on lan- guage, seeing language not as static “codes” with solid boundaries but rather, as fluid resources in meaning-making practices (Pennycook, 2010). These views are captured in the recent use of the terms, “code-meshing” (Canagarajah, 2011a, 2011b) and “translanguaging” (García, 2009; Creese and Blackledge, 2010), which seek to take away the “markedness” of the linguistic phenomenon that is tradi- tionally called “code-switching” and reconceptualize it as a social practice that is part and parcel of everyday social life. This plethora of terms is aptly summarized by Lewis, Jones and Baker (2012) in their analysis of the historical development of the term, translanguaging: A plethora of similar terms (e.g., metrolingualism, polylanguaging, polylingual languaging, heteroglossia, codemeshing, translingual practice, flexible bilingualism, multilanguaging, and hybrid language practices) makes this extension of translanguaging appear in need of focused explication and more precise definition. Such varied terms are competitive with translanguaging for academic usage and acceptance (Lewis et al., 2012, p. 649) Further complicating the picture is the overlapping field of studies variously known as: first language (L1) use in second and foreign language (L2) classrooms (Turnbull and Dailey-O’Cain, 2009), use of local languages in English classes (Mahboob, 2011), incorporation of L1 in foreign language teaching and learning (Brooks-Lewis, 2009), the role of the mother tongue in foreign language class- rooms (Butzkamm, 2003), student use of the mother tongue in the task-based classroom (Carless, 2007), L1 use in the L2 classroom (Edstrom, 2006), bilingual pedagogy in EFL (Forman, 2010), first language and target language in the foreign language classroom (Littlewood and Yu, 2009). And the kinds of classrooms studied can be content classrooms or language classrooms (or various hybrid in- stances lying on a continuum between these two prototypical types; see Figure 1 in Lin and Man, 2009, p. 137). (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 196–218  2708_4-1_09  (p. 196) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 197–218  2708_4-1_09  (p. 197) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   199 manner in which the bilingual teacher employed extensive switching between Spanish and English to create humour, both as a means of social control (via the creation of a sense of solidarity) and as a way to arouse students’ interest. Guthrie (1984) used similar research methods in a study of an ESL lesson attended by 11 first-grade Cantonese-American students (ranging from limited- English proficiency to fluent). Two types of lessons were analysed: reading in English with a Cantonese-English bilingual teacher, and oral language with an English monolingual teacher. Field notes and audio-recording of six hours of lessons were obtained and coded by two bilingual observers. Guthrie found that interactions of the English monolingual teacher with the limited-English- proficiency students in the oral lessons were characterized by a higher proportion of conversational acts such as ‘attention-getters’, ‘requests for action’ and ‘pro- tests’, indicating a certain lack of teacher control and a frequent loss of student attention. On the other hand, while the bilingual teacher used Cantonese (L1 of the students) very rarely (less than 7% on average) in the English reading lessons, when she did it was for a distinct reason. She told the researchers that she tried to avoid using Cantonese during these lessons and was surprised to find she has used L1 as much as she had. The functions of L1 use reported by Guthrie can be summarized as: (a) to act as a “we-code” for solidarity, (b) to clarify or check for understanding, (c) to contrast variable meanings in L1 and L2 and to anticipate likely sources of confusion for students. While the functional coding approach dominated early work, in some studies (e.g., Milk, 1981; Guthrie, 1984) preliminary use of ethnographic interviews and interactional sociolinguistic methods were incorporated, a trend which con- tinued in later work. Major Contributions Many early studies seemed to have worked with the assumption that functional categories were stable, valid categories of classroom speech and that analysts could reliably assign utterances to each category. Yet the functional coding ap- proach in early studies in fact involved a lot of sociolinguistic interpretive work on the part of the coder. This interpretive work was, however, not made explicit but taken for granted in the form of final frequency counts of L1 and L2 distri- buted across different functional categories. Later studies (e.g., Lin, 1990, 1996, 1999, 2006; Merritt et al., 1992; Adendorff, 1993; Ndayipfukamiye, 1994; Polio and Duff, 1994; Eldridge, 1996; Martin-Jones, 1995, 2001; Heller, 1999, 2001; Jacobson, 2001; Simon, 2001; Martin, 1996, 1999, 2003; Creese, 2005; Üstünel and Seedhouse, 2005) have, to varying degrees, (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 198–218  2708_4-1_09  (p. 198) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 199–218  2708_4-1_09  (p. 199) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 200   Angel Lin dispensed with a priori lists of functional categories and drawn on research approaches from interactional sociolinguistics and ethnography of communica- tion (e.g., Goffman, 1974; Gumperz, 1982; 1986); conversation analysis (Sacks, 1965/1992); interpretive research paradigms; critical social theory (Bourdieu and Passeron, 1977); and critical research paradigms to study classroom code- switching (Heller and Martin-Jones, 2001; Li and Wu, 2008; Li, 2011). Just as interactional sociolinguistics (IS) and ethnography of communication (EC) provide the most useful analytic tools for researching and understanding code-switching in different settings in society, their concepts and methods have been drawn upon in classroom studies on code-switching. For instance, the most frequently and fruitfully used ones are: code-switching as contextualization cues (Gumperz, 1984) to signal a shift in the frame or footing (Goffman, 1974) of the current interaction (e.g., see Adendorff, 1993). Frame or footing is the definition of what is happening and it is constantly being negotiated, proposed (signaled) and re-defined by the speakers engaged in interaction. Different frames or footings that are being evoked (or signaled and proposed by a speaker) involve the simul- taneous negotiation of different role-relationships and the associated sets of rights/obligations. Lin’s studies (1990, 1996), for instance, drew on these interac- tional sociolinguistic analytic concepts to analyse code-switching in Hong Kong classrooms. Below is an example from Lin’s (1996) reanalysis of Johnson’s (1985) data in Hong Kong secondary schools, using IS analytic concepts. The data pre- sentation format is as in Johnson’s: Tape-recorder counter numbers precede ut- terances; bold italics indicate originally Cantonese utterances, and only teacher’s utterances have been transcribed. Example (1) A junior secondary math teacher in Hong Kong begins his lesson in English and then breaks off and switches to Cantonese to deal with late-comers; once they are settled, he switches back to English to continue with the lesson work (“Example 1” in Johnson, 1985, p. 47): 008  Close all your text book and class work book. 012  There are some classmates not back yet. Be quick! 017 Now, any problem about the class work? Johnson (1985) analyses the Cantonese utterance as an example of an informal aside done in Cantonese. While agreeing partially with this analysis, we note, however, that if it is to mark out a mere topical digression, the teacher can well (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 200–218  2708_4-1_09  (p. 200) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 201–218  2708_4-1_09  (p. 201) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   201 have done this by means other than code-switches, e.g., intonation changes, hand-claps or pauses to bracket the aside (see example in Lin, 1990, pp. 32–36). The use of these contextualization cues (Gumperz, 1984) does not involve a viola- tion of the institutional “use-English-only” constraint which teachers in Anglo- Chinese secondary schools in Hong Kong were well aware of. It can, therefore, be argued that what is being signalled here is not only a topical aside, but also a radical break in the English pedagogic frame and an urgent change in the teacher's concerns. The switch from English to Cantonese seems to relay to his students this implicit message, “Now I'm so annoyed by these late-comers that I have to put aside all kinds of teaching, including that of English teaching, and concentrate on one single task: that of getting you to settle down quickly! And you’d better take my command seriously as I’m focused on enforcing it!” This break in the English pedagogic frame to highlight a different, urgent set of con- cerns cannot have been achieved without the teacher's switch from English (L2) to Cantonese (L1). The key, therefore, to understanding the implicit meanings signalled by code-switches lies in a recognition of the sociolinguistic fact that whenever Hong Kong Cantonese have something urgent and earnest to relay to one another, they tend to do so in their shared native language; whenever Hong Kong Cantonese speak to one another in English despite their having a common native language, it is usually because of some institutionally given reasons, for instance, to teach and learn the English language in an English immersion classroom. When teachers want to establish a less distanced and non-institutionally defined rela- tionship with their students, they will also find it necessary to switch to their shared native language, Cantonese. Similar kinds of analysis drawing on IS and EC research methods are offered in Simon’s (2001) study of code-switching in French-as-a-foreign-language class- rooms in Thailand. Teachers are seen as code-switching for a number of pur- poses, among which are those of negotiating different frames (e.g., formal, insti- tutional learning frame vs. informal friendly frame), role-relationships and identities (e.g., teacher vs. friend). Code-switching is seen as having a ‘momen- tary boundary-levelling effect’ in the classroom (Simon, 2001, 326). Whether sim- ilar effects might be achieved by code-switching in different contexts would, how- ever, seem to depend on different sociolinguistic statuses and values associated with different codes in different societies. In studies along this line, IS and EC analytical concepts and methods are drawn upon to analyse instances of classroom code-switching. The findings look remarkably similar across different sociocultural contexts. Code-switching is seen to be an additional resource in the bilingual/multilingual teacher’s commu- nicative repertoire enabling her/him to signal and negotiate different frames (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 200–218  2708_4-1_09  (p. 200) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 201–218  2708_4-1_09  (p. 201) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 204   Angel Lin (2) Language-Focus-IRF: Teacher-Initiation [L1/L2]1 Student-Response [L1/L2] Teacher-Feedback [L2], or use (2) again until Student-Response is in L2 (3) Start (2) again to focus on another linguistic aspect of the L2 response elicited in (2); or return to (1) to focus on the story again. This kind of discourse practice allows the teacher to interlock a story focus with a language focus in the reading lesson. There can be enjoyment of the story, via the use of the story-focus IRF, intertwined with a language-learning focus, via the use of the language-focus IRF. We have noted above that the teacher never starts an initiation in L2. She always starts in L1. This stands in sharp contrast with the discourse practices of Teacher C (another teacher in the study) who al- ways starts with L2 texts or questions in her initiations. It appears that by always starting in L1, Teacher D always starts from where the student is – from what the student can fully understand and is familiar with. On the other hand, by using the language-focus IRF format immediately after the story-focus IRF format, she can also push the students to move from what they are familiar with (e.g., L1 ex- pressions) to what they need to become more familiar with (e.g., L2 counterparts of the L1 expressions) (see Lin, 1999). The fine-grained sequential analysis of classroom code-switching drawing on both Sinclair and Coulthard’s (1975) IRF analytical tradition and conversation analysis (CA) continued in later work as ex- emplified in Üstünel and Seedhouse (2005)’s study on how learners displayed their alignment or misalignment with the teacher’s pedagogical focus in an EFL classroom in a Turkish university. The fine-grained discourse analytic methods were also productively used in conjunction with a stimulated recall procedure in Scott and De La Fuente (2008)’s study of the role of L1 when pairs of intermediate- level college learners of French and Spanish are engaged in consciousness rais- ing, form-focused grammar tasks. As we shall see in the next section increasingly studies are drawing on a wider range of research methods including both qualita- tive and quantitative ones. Recent Developments In this section we shall look at research that hints at a slightly different research angle and research that starts to draw on research approaches from diverse fields 1 “L1/L2” denotes “L1 or L2”. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 204–218  2708_4-1_09  (p. 204) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 205–218  2708_4-1_09  (p. 205) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   205 such as genre theories, theories of academic literacies (Setati, Adler, Reed and Bapao, 2002) and cognitive processing perspectives and experimental methodol- ogies (Macaro, 2009). Setati et al. (2002) provided a mid-term report on findings from their larger ongoing study of code-switching and other language practices in Mathematics, Science and English language classrooms in South Africa. These schools had ad- opted a small-group inquiry teaching approach and built on notions of additive bilingualism and strategic code-switching as encouraged by the authorities. While good in their intentions, this approach might have overlooked some pitfalls in two areas: 1. The indirect, student-centred, exploratory, group-work, learning-from-talk teaching approach: This is found to be done mostly in students’ L1. However, without teacher’s input on scientific content (e.g., in whole-class instruction), students may suffer from a lack of input in the English academic discourses required to talk about science topics or writing extended texts in English. 2. So, some traditional teacher-fronted whole class teaching may be needed to provide the necessary L2 academic discourses to students, especially those in rural areas. Setati et al. (2002) found that the progressive pedagogies (e.g., student-centred group work) alone did not provide the much-needed direct teaching of subject domain-specific academic discourses and English academic literacies and thus aggravated social inequalities. Setati’s et al.’s (2002) report, however, did not show much analysis of how this academic discourse can be provided or inserted into the progressive teaching approaches along with the integration of some con- ventional pedagogies. While this report seems to be work-in-progress, it does point out the importance of drawing on research tools of genre analysis of differ- ent subject-specific academic discourses in future studies of code-switching in the classroom. We shall continue the discussion of the potential contribution of genre-based pedagogies to classroom code-switching research in the final sec- tion. Let us now turn to the recent work of Macaro (2009), who has drawn on cognitive processing perspectives and experimental approaches. Macaro (2009) presented the findings of two studies on the effect of code- switching on students’ vocabulary learning. In the first study a sample of 159 Chinese learners of English, aged 16, were randomly assigned to two different conditions. The context was a reading class in which the teacher orally interacted with the whole class around two challenging English texts. There were two ses- sions, each with a different text, and the conditions were rotated with each text. In the first condition, the teacher provided a first-language equivalent of words in the text that she knew her students were unfamiliar with as determined by a (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 204–218  2708_4-1_09  (p. 204) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 205–218  2708_4-1_09  (p. 205) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 206   Angel Lin pre-test of vocabulary knowledge. In the second condition, the same teacher pro- vided learners with English definitions of the same unfamiliar words. Students in each condition were thus given different types of information about unknown words (code-switch vs. paraphrase). A third group was an intact class that acted as a control group, which was given both types of information (code-switch and paraphrase). A pre-test of receptive vocabulary showed that the target vocabulary items were all unfamiliar to the students, that there were no statistically signifi- cant differences in their vocabulary knowledge between the 3 classes, and addi- tionally the 3 classes were chosen because they did not differ in general English proficiency according to their school proficiency tests. Students were given an immediate post-test, and a delayed post-test after 2 weeks. The findings are sum- marized as follows: Text 1 (about sport): the L2 paraphrase group scored significantly higher in the immediate post-test than the other 2 groups; however, this advantage disap- peared in the delayed post-test and there were no significant differences among the 3 groups. Text 2 (about the life of Walt Disney): there were no significant differences among the 3 groups in both the immediate and delayed post-tests. Macaro concluded that there is at least “no harm” in giving L1 equivalents of words during the teaching activity around the reading texts in terms of long-term vocabulary acquisition and he further hypothesized that giving L1 vocabulary equivalents “lightens the cognitive load freeing up processing capacity to focus on the meaning of the text as a whole” (2009, p. 43). In the second study students’ responses to teachers’ code-switching (e.g., giving L1 equivalents of unfamiliar words) were tapped through a stimulated re- call procedure. The study was set in China, in two universities (one teacher in each university), and involved first year students learning English as a foreign language (EFL). The researcher videotaped sixteen 45-minute lessons of a num- ber of these EFL classes and then, immediately following the lesson, asked indi- vidual learners (n = 32) to take part in a stimulated recall session carried out in the students’ first language. Based on the students’ responses Marcaro inferred that when provided with the L1 equivalents of unfamiliar L2 words, “the amount of processing that a learner has to do is in fact increased rather than decreased”, suggesting more cognitive processing taking place, and students may have been afforded “deeper processing opportunities” than when they are provided with L2 definitions (Macaro, 2009, p. 47). (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 206–218  2708_4-1_09  (p. 206) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 207–218  2708_4-1_09  (p. 207) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   209 tively with what kinds of effect would require nothing short of concerted research efforts breaking disciplinary boundaries and drawing on a whole range of theo- retical perspectives and research methods. Scarcity of theory-driven research questions Research questions in the field tend to arise from practical classroom concerns (e.g., to uncover the good sense or rationality of the existing practices). While this is a normal source of research questions in education research, if the research literature cannot build up an expanded, diversified repertoire of theoretical frameworks that will motivate the formulation of increasingly sophisticated re- search questions, the studies would tend to be overly descriptive and repetitive (e.g., the classroom code-switching literature tends to be replete with studies describing the useful classroom functions of existing code-switching practices). Recent studies that draw on cognitive theories of vocabulary learning (e.g., Macaro, 2009) and theories of discipline-specific genres and academic literacies (Setati et al, 2002) would seem to be a welcoming development although we also need to complement these approaches with approaches from the interpretive and critical paradigms as classroom code-switching involves not only cognitive pro- cessing but also identity/ideology reproduction (or transformation). Lack of variety in the research questions and research designs There is a lack of longitudinal studies. Studies in the literature tend to be one- shot or cross-sectional. There is scarcity of studies on students’ code-switching, and also written code-switching (but see Canagarajah, 2011a, 2011b). There is a lack of studies conducted by teachers (as teacher-researchers) or students (as student-researchers) themselves on their own classroom code-switching prac- tices (but see Song and Andrews (2009) for an interesting study of four teachers’ own perspectives on their code-switching instances in their classrooms through a stimulated recall procedure; their students’ perspectives were also tapped using a similar procedure). There is also a lack of studies on the direct comparison of code-switching in the language and the content classrooms. If I might have sounded a bit too critical of the state of affairs in our field, it is because I am deeply aware of the invisible ideological grip that seems to have exercised its spell on researchers in this field. It is to a critical uncovering of this ideological grip that I shall turn to in the next section. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 208–218  2708_4-1_09  (p. 208) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 209–218  2708_4-1_09  (p. 209) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 210   Angel Lin Production and Legitimation of Knowledge on Classroom Code-switching: Invisible Ideological Shaping Forces Researchers on classroom code-switching seem to have been working “against the grain” of dominant ideologies in pedagogical theories and policies in many parts of the world for many years. When the knowledge produced by a piece of research is aligned with the dominant theories of the field, it is easier for it to be accepted and legitimated in the symbolic market of academia (e.g., widely cited and circulated in the education field). When, however, the piece of knowledge produced is not aligned with the dominant pedagogical theories of the day it cannot easily attain the status of received knowledge in the field and constantly feels the need to justify and prove itself, thus the “normalizing mission” implicit in most of the studies of classroom code-switching, as discussed above. Some promising research programmes might get interrupted or derailed. For instance, three decades ago, R. K. Johnson was already experimenting with different bilin- gual ways of presenting teaching content, both in oral and written modes and documenting the effects of different modes of presentation (bilingual vs. mono- lingual; oral vs. written) on students’ comprehension of content in Hong Kong secondary schools (Johnson, 1983; Johnson, 1985; Johnson, Chan, Lee, and Ho, 1985). Johnson and his colleagues investigated the effects of various modes of presentation and questioning (e.g., English / Chinese / bilingual texts and ques- tions, or different combinations of them). He also looked at the code-switching strategies used by experienced teachers in English medium schools. Research studies in the early and mid-1980s in Hong Kong were characterized by optimism in the possibility of developing bilingual oral and/or written strategies in English medium schools to solve the dilemma created by the overwhelming parental de- mand for an English medium education for their children and the often limited English proficiency of the majority of children to benefit from a purely English medium education. In the first study (Johnson 1983), it was found that teachers systematically code-switched between Cantonese and English for different purposes. In general, English was found to be associated with text-dependent, formal and didactic functions; whereas Cantonese was found to be associated with text-independent, informal and explanatory functions. In his conclusion, Johnson wrote: Separation of the languages is one simple, but possibly also simplistic, approach to the problems of bilingual education, and I am not convinced that there is anything intrinsically wrong with code-switching in bilingual classrooms. At the very least, the teaching strate- (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 210–218  2708_4-1_09  (p. 210) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 211–218  2708_4-1_09  (p. 211) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   211 gies identified here are capable of greater sensitivity to differences amongst learners and groups of learners than the separation approach. (Johnson, 1983, p. 282) In the second study, Johnson et al. (1985) tested for the effects of different linguis- tic modes of presentation and questioning on the subsequent comprehension test scores among 1,296 Form 3 (Grade 9) students. It was found that irrespective of the linguistic mode of presentation of the texts (on the topic of how bean curd is made), students scored higher on average when answering Chinese questions, and irrespective of the linguistic mode of questioning, students scored higher on average when the texts had been presented in the Cantonese mode or the bilin- gual mode. When asked about their preferences on the medium of instruction, less than 3% of the 1,296 students preferred English-only instruction. In the oral mode, the students were about equally split in their preference for Cantonese- only instruction or Cantonese-English bilingual code. In the written mode, over 70% of the students preferred to study with Chinese texts, although 11% would also like to have English glosses added to the Chinese text, and 32% would also like to have a corresponding English text side by side with the Chinese text. Ap- parently, the majority of students preferred a bilingual to an English-only mode of instruction. The production of knowledge along this line, however, was not aligned with the dominant theories in TESOL and applied linguistics in that era. The bilingual approach to the medium of instruction was problematized in a paper by Swain in 1986. After having visited Hong Kong and reviewed the school language situation in Hong Kong, Swain (1986) argued against the bilingual medium practices, which she described as an instance of “the mixing approach” (1986, p. 3). John- son’s programme of research came to an end in the late 1980s and his innovative and eclectic approaches to researching bilingual classroom practices (both written and spoken) were not widely circulated or known after the 1980s. Three decades have gone by and we see that the fields of applied linguistics and second language learning have broadened and embraced alternative theo- retical perspectives including sociocultural theories, critical theories, postcolo- nial theories and many scholars have changed significantly their stance towards classroom code-switching. For instance, Swain and her colleagues have pub- lished in Hong Kong a handbook entitiled “How to live a guilt-less life using Cantonese in the English Class” (Swain, Kirpatrick and Cummins, 2011). At this juncture although researchers investigating classroom code-switching still feel the need to constantly prove and justify – to legitimate classroom code-switching, we are perhaps a bit freed from the tight grip of the former times and could afford to be much more critical of our own work so as to advance our field further. Below I outline some directions for future work that might carry our goals further along. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 210–218  2708_4-1_09  (p. 210) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 211–218  2708_4-1_09  (p. 211) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 214   Angel Lin patterns and participant structures. All these require an approach that allows for try-and-see and then document and re-try another pattern and see what happens and re-design future action plans that will progressively better achieve the goals through both bilingual and other pedagogical practices. The above suggestions might sound like an ‘unholy’ eclectic approach to the lin- guistic or research methodological purist. However, to have breakthroughs in our current state of affairs in researching classroom codes-witching, we need to be both pragmatic and flexible in our research paradigms and approaches. We also need concrete designs of bilingual classroom strategies and research studies that can systematically develop these designs and show their effectiveness (with respect to the situated goals of the classroom). When we can break away from the implicit grip of the “normalizing mission” perhaps we can afford to be more critical of the research methods we have traditionally used to study classroom code-switching. For instance, without designing more of the kind of systematic longitudinal, interventionist studies that can work on further refining bilingual classroom strategies and pedagogies to achieve the goals deemed worthwhile in specific contexts, our research literature on classroom code-switching might be seen as repetitive of apologetic statements about the good sense or diverse func- tions of classroom participants’ practices and we cannot advance our knowledge of how classroom participants can do better what they are already doing with dif- ferent degrees of success or failure. Acknowledgements Parts of this paper appeared in the author’s 2008 review article. However, the cur- rent paper is a significantly re-written, expanded and updated new version. The author is grateful for the very useful comments of the editor and an anonymous reviewer on an earlier draft of this version. References Adendorff, R. 1993. Code-switching Amongst Zulu-speaking Teachers and Their Pupils: Its Functions and Implications for Teacher Education, Language and Education 7(3), 41–162. Auerbach, E. 1993. Re-examining English only in the ESL classroom. TESOL Quarterly 27(1), 9–32. Alvarez-Caccamo, C. 2001. Codes. In A. Duranti (Ed.) Key Terms in Language and Culture (pp. 23–26). Oxford: Blackwell. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 214–218  2708_4-1_09  (p. 214) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 215–218  2708_4-1_09  (p. 215) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM Classroom code-switching   215 Bourdieu, P. and Passeron, J. C. 1977. Reproduction in Education, Society and Culture. London: Sage. Brooks-Lewis, K. A. 2009. Adult learners’ perceptions of the incorporation of their L1 in foreign language teaching and learning. Applied Linguistics, 30(2), 216–235. Butzkamm, W. 2003. We only learn language once. The role of the mother tongue in FL classrooms: death of a dogma. Language Learning Journal, 28, 4–14.  Canagarajah, S. 2001. Constructing hybrid postcolonial subjects: Codeswitching in Jaffna Classrooms. In M. Heller and M. M. Jones (Eds.) Voices of Authority: Education and Linguistic Difference (pp. 193–212). Westport, CT: Ablex. Canagarajah, S. 2011a. Translanguaging in the classroom: Emerging issues for research and pedagogy. In Li, W. (Ed.) Applied Linguistics Review. Berlin/New York: Walter de Gruyter. 1–28. Canagarajah, S. 2011b. Codemeshing in academic writing: Identifying teachable strategies of translanguaging. The Modern Language Journal, 95(3), 401–417. Carless, D. 2007. Student use of the mothertongue in the task-based classroom. ELT Journal, 62(4), 331–337. Cook, V. 2001. Using the first language in the classroom. Canadian Modern Language Review 57(3), 402–423. Creese, A. 2005. Teacher Collaboration and Talk in Multilingual Classrooms. Clevedon, Multilingual Matters. Creese, A. and Blackledge, A. 2010. Translanguaging in the bilingual classroom: A pedagogy for learning and teaching? The Modern Language Journal 94(i), 103–115. Cummins, J. 2007. Rethinking monolingual instructional strategies in multilingual classrooms. Canadian Journal of Applied Linguistics, 10, 221–240. Edstrom, A. 2006. L1 use in the L2 classroom: One teacher’s self-evaluation. Canadian Modern Language Review 63(2), 275–292. Ferguson, G. 2003. Classroom code-switching in post-colonial contexts: functions, attitudes and policies. AILA Review, 16, 38–51. Flanders, N. A. 1970. Analyzing Teaching Behavior. Reading, Mass: Addison-Wesley. Forman, R. 2007. Bilingual teaching in the Thai EFL context: One teacher’s practice. TESOL in Context 16(2), 19–24. Forman, R. 2010. Ten principles of bilingual pedagogy in EFL. In A. Mahboob (Ed.) The NNEST Lens: Nonnative English Speakers in TESOL. Newcastle upon Tyne: Cambridge Scholars Press. Frohlich, M., Spada, N. and Allen, P. 1985. Differences in the Communicative Orientation of L2 Classrooms. TESOL Quarterly, 19(1), 27–57. García, O. (2009). Bilingual education in the 21st century: A global perspective. Oxford: Wiley-Blackwell. Goffman E.1974. Frame Analysis: An Essay on the Organization of Experience. New York: Harper and Row. Gumperz, J. J. 1982. Discourse Strategies. London: Cambridge University Press. Gumperz, J. J. 1984. Ethnography in Urban Communication. In J. C. P. Auer and A. Luzio (Eds.) Interpretive Sociolinguistics: Migrants, Children, Migrant Children. Tubingen: Narr. Gumperz, J. J. 1986. Interactional Sociolinguistics in the Study of Schooling, in J. C. Gumperz (Ed.) The Social Construction of Literacy. Cambridge: Cambridge University Press. Guthrie, L. F. 1984. Contrasts in Teachers’ Language Use in a Chinese-English Bilingual Classroom. In J. Handscombe, R. A. Orem and B. P. Taylor (Eds.) TESOL’ 83: The Question of Control (pp. 39–52). Washington D.C.: TESOL. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 214–218  2708_4-1_09  (p. 214) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 215–218  2708_4-1_09  (p. 215) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 216   Angel Lin Halliday, M. A. K. 1994. An Introduction to Functional Grammar (second edition). London: Edward Arnold. Haroon, H. A. 2005. Teacher Code-switching and its Functions in Mathematics and Science Lessons. Asia Pacific Journal of Language in Education 7(1), 1–25. Heap, J. L. 1985. Discourse in the Production of Classroom Knowledge: Reading Lessons. Curriculum Inquiry 15(3), 345–379. Heller, M. 1999. Linguistic Minorities and Modernity. London: Longman. Heller, M. 2001. Legitimate Language in a Multilingual School. In M. Heller and M. M. Jones (Eds.) Voices of Authority: Education and Linguistic Difference: Westport, CT: Ablex. 381–402. Heller, M. and Martin-Jones, M. 2001. Voices of Authority: Education and Linguistic Difference. Westport, Connecticut and London, Ablex Publishing. Johnson, R. K. 1983. Bilingual switching strategies: A study of the modes of teacher-talk in bilingual secondary school classrooms in Hong Kong. Language Learning and Communication 2(3), 267–283. Johnson, R.K. 1985. Report of the ELTU Study of the Oral Medium of Instruction in Anglo- Chinese Secondary School Classroom. Hong Kong: Faculty of Education, University of Hong Kong. Johnson, R. K., Chan, R. M. L., Lee, L. M. and Ho, J. C. 1985. An investigation of the effectiveness of various language modes of presentation, spoken and written in form III in Hong Kong Anglo-Chinese secondary schools. Hong Kong: Hong Kong Government, Education Department. Legarreta, D. 1977. Language choice in bilingual classrooms. Journal of Social Issues 23, 9–16. Levine, G. S. 2011. Code choice in the language classroom. Bristol: Multilingual Matters. Lewis, G., Jones, B., and Baker, C. 2012. Translanguaging: developing its conceptualisation and contextualization. Educational Research and Evaluation, 18(7), 655–670. Li, Wei, and Wu, C. J. 2008. Code-switching: Ideologies and practices. In A. He and Y. Xiao (Eds.), Chinese as a heritage language: Fostering rooted world citizenry (pp. 225–238). Honolulu: National Foreign Language Resource Centre and University of Hawai’i Press. Li, Wei. 2011. Multilinguality, Multimodality, and Multicompetence: Code- and Modeswitching by Minority Ethnic Children in Complementary Schools. The Modern Lanaguge Journal 95, 370–384. Lin, A. 1990. Teaching in Two Tongues: Language Alternation in Foreign Language Classrooms, Research Report No.3, Hong Kong, City Polytechnic of Hong Kong. Lin, A. 1996. Bilingualism or Linguistic Segregation? Symbolic Domination, Resistance and Code-Switching in Hong Kong Schools. Linguistics and Education, 8(1), 49–84. Lin, A. 1999. Doing-English-Lessons in the Reproduction or Transformation of Social Worlds? TESOL Quarterly, 33(3), 393–412. Lin, A. and Martin, P. 2005. Docolonisation, Globalisation: Language-in-Education Policy and Practice. Clevedon: Multilingual Matters. Lin, A. M. Y. 2006. Beyond linguistic purism in language-in-education policy and practice: Exploring bilingual pedagogies in a Hong Kong science classroom. Language and Education, 20(4), 287–305. Lin, A. M. Y. 2008. Code-switching in the classroom: Research paradigms and approaches. In K. A. King, and N. H. Hornberger (Eds.) Encyclopedia of language and education, 2nd Edition, Volume 10: Research methods in language and education. New York: Springer Science. 273–286. (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 216–218  2708_4-1_09  (p. 216) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (CS4)  WDG (155×230mm)  DGMetaScience     J-2708 ALR 4:1   pp. 217–218  2708_4-1_09  (p. 217) PMU:(WSL) 10/01/2013  29 January 2013 10:41 PM