Code-switching in Iranian Elementary EFL Classrooms : An Exploratory Investigation

This paper presents the results of a small-scale exploratory investigation of code-switching (CS) between English and Farsi by 4 Iranian English foreign language (EFL) teachers in elementary level EFL classrooms in a language school in Isfahan, Iran. Specifically, the present study aimed at exploring the syntactical identification of switches and the functions that the switches served. The data consist of field notes and scripts of audio-recordings of the teachers’ talk collected during classroom observations. The findings suggest that CS is a frequently applied strategy and a valuable resource for bilingual teachers in foreign language classrooms, and its judicious and skillful use can boost the quality of teaching. Moreover, it was found that EFL teachers in this study tended to use the learners’ L1 (i.e., Farsi) to serve a number of pedagogic and social functions, which contributed to better teacher-student classroom interaction. Implications may be drawn for language teacher education programs and for further research on systematic investigation into actual classroom practices.


Introduction
Code-switching (CS) is a common phenomenon of language contact, which is broadly discussed in every subfield of linguistic disciplines (Nilep, 2006).However, researchers often do not agree on a clear definition of it.It has generally come to be understood as "the alternative use by bilinguals (or multilinguals) of two or more languages in the same conversation" (Muysken, 1995, p.7) or "in the unchanged setting, often within the same utterance" (Bullock and Toribio, 2009, p.2).According to Myers-Scotton (1997), codes involved in CS do not necessarily constitute those of standard languages.In fact, she expands the scope of CS to subsume switching between languages (Azuma, 2001;Treffers-Daller, 1998;etc.),dialects (Alfonzetti 1998), styles and registers (Farris, 1992).In this paper, 'code' is used to refer to two distinct languages, English and Farsi.This paper is an attempt to explore Iranian EFL teachers' patterns of language alternation, and the functions of such alternations, between English, the target language, and Farsi, the learners' L1.As the title suggests, this research is, in essence, exploratory.Gaining insights from Glaser and Strauss (1967) grounded theory, we adopted a heuristic method of collecting and then analyzing data for emergent themes and issues to be investigated further.That is, rather than enter the field with a priori assumptions, we have attempted to capture instances of CS by observation through the lens of four case studies.

Development of classroom CS research
Classroom CS research, as opposed to social CS or general CS in no specific context, has focused on both teacher-learner interaction and the influence CS may exert on students' learning .Early studies on CS centered around bilingual education programs for minority pupils in the United States in the 1970s and early 1980s (Martin-Jones, 1995).These studies were quantitative in nature and aimed at illustrating the influence of CS in bilingual classroom communication on children's linguistic development.However, Milk (1981Milk ( , 1982) ) and Guthrie (1984) started a breakthrough line of research in the realm of CS by departing from a purely quantitative study of classroom communication.They started using audio-recordings and descriptive frameworks to focus on the ways in which teachers and students fulfill tasks with two languages.
More recent studies have focused on describing language teachers' CS patterns in foreign language classrooms (see Greggio and Gill, 2007).Duff and Polio (1990) reported observations from several foreign language classrooms, claiming that the use of target language varied from 92% to 100%.Polio and Duff (1994) investigated the functions for which language teachers employed CS.They categorized cases of CS into functions such as 'classroom administrative vocabulary', 'grammar instruction', and 'classroom management'.Kraemer (2006) studied the amount and purpose of L1 use by five language teacher assistants, and reported that a "fair amount' of L1 was used, specifically for classroom management and translation.Hobbs et al. ( 2010) compared CS behavior of native and non-native speaker teachers in Japanese language classrooms.Along the same line of research, this study aimed at exploring the types of CS patterns applied by four Iranian EFL teachers and the functions such switches served.

Types of CS
By drawing on extensive observations of several cases of CS, Sankoff and Poplack (1981) have identified three types of CS: tag-switching, inter-sentential switching and intra-sentential switching.Tag-switching, also known as extra-sentential CS (Muysken, 1995), involves the insertion of a tag or a short fixed phrase in one language into an utterance which is otherwise entirely in the other language.Inter-sentential CS constitutes a switch occurring at a clause or sentence boundary, where each clause or sentence is either in one language or the other.Finally, intra-sentential CS involves switching within the clause or sentence boundary.
The teaching of English as a foreign language has markedly increased in Iran in the last two decades.However, we still know little about what actually goes on in Iranian EFL classrooms.Specifically, classroom interaction and teachers' language choices in the EFL classroom context in Iran seem to be under-explored.That is, more evidence is needed to document the patterns of teachers' language choices and the pedagogic functions such language choices serve in different classroom environments.In an attempt to address this need, the present study draws on previous research to (a) identify the types of CS in four Iranian EFL teachers' talk in their elementary level classrooms, and (b) discover the functions that these teachers' CS served in classroom interaction.

Research context
The investigation was made in four EFL classrooms at the elementary level in an English language school in Isfahan, Iran.There were on the average 20 students who learned English for 90 minutes, three times a week in every classroom observed.The students ranged in age from 12 to 16.The school manager and the four participating teachers approved the authors' observation of the classrooms in focus.The teachers were not informed about the purpose of the observations.

Research questions
Intending to describe the CS patterns occurring in teachers' utterances and explain the functions of the switches in the EFL classrooms observed, the following two questions were addressed in this study: 1) What types of CS are there in Iranian EFL teachers' talk in their elementary level classrooms?
2) What functions does teachers' CS serve in classroom interaction?

Participants
This research was a case study of four Iranian EFL teachers (see Table 1), teaching elementary level classrooms in an English language school in Isfahan, Iran.They were all selected using convenience-case sampling; thus, they were all 'accessible, easy to contact, and well-known [to us]' (Wellington, 2000, pp. 61-62).They were all BA holders in TEFL and they had all completed the same teacher-training programs.They were similar in age (22-28 years old), and they did not differ considerably in terms of their teaching experience.It should be noted that these four teachers had taken the paper version of Test of English as a Foreign Language (TOEFL) in partial fulfillment of the requirements for getting a job in the language school and they had all scored above 580, which reveals that they were all at advanced levels of English language proficiency.Also, observations revealed that they enjoyed high communication skills.

Data collection
The first author of this paper observed the classroom teaching of the four participants a minimum of five times (see Table 1).The observer sat at the back of the class the entire 90 minutes of class time, and made comprehensive field notes (i.e.systematic and comprehensive description of all classroom events) that consisted of number of students, teaching and learning activities, language used, verbal and non-verbal interaction, etc.Moreover, having the permission of the school manager and the participating teachers, all the lessons were audio-recorded.After all the class visits were complete, the audio-recordings were carefully transcribed.

Data analysis
Following the exploratory nature of the study, transcriptions of the audio-recordings and the entries in field notes from the observations were reviewed and analyzed using data-driven methods such as ' making sense of, sifting, organizing, cataloguing, [and] selecting determining themes' (Holliday, 2006, p. 99).Since the data illustrated clear instances of 'moves' by teachers from one CS function (e.g.praising students) to another (e.g.explaining a grammatical point), the researchers opted to examine the data at the functional level.Also, analysis was qualitative in nature in that, rather than imposing pre-determined sets of categories, general categories emerged from the data after careful analysis, examination and labeling by the researchers.The resulting categories include the following: It should be noted that almost similar categories are found Kraemer's (2006) study on the use of English in German language classrooms and in a study by Hobbs et al. (2010) on CS in Japanese classrooms.However, pre-imposed categories were not used in this study and not all categories present in such studies appear in ours.Moreover, the above-mentioned categories were the most prevalent categories observed in our collected data, and certainly not the only ones.

Types of CS
Analyzing and reviewing the field notes and the transcripts of the audio-recordings revealed that the CS distribution patterns of the four teachers were quite similar.Thus, the units consisting of mixed varieties of the four teachers' switches are put together to make the comparison of the three types of switch patterns possible.As shown in Table 2, and in line with the findings reported by Qian et al. (2009), inter-sentential switching (79%) far outweighed the intra-sentential switching (18%), which was in turn more than tag-switching (3%).Inter-sentential switching involves switching at the sentence boundary or among sentences as seen in Example 1.In this example, the teacher adopted an inter-sentential CS from English to Farsi to make his students' comprehension easier.In fact, by switching to the more familiar code, teachers can pass messages more efficiently and enhance their students' comprehension (Kim, 2002;Qian et al., 2009).
(Because she is hard of hearing.) Intra-sentential CS occurs within the clause boundary, as illustrated in Example 2, where the teacher inserted an English phrase in his demonstration, which was predominantly in Farsi.
(After verbs such as 'enjoy' and 'keep', the -ing form of the verb is used.) Tag-switching involves the insertion of a tag or a short fixed phrase (i.e.discourse markers or sentence-fillers) in one code into an utterance, which is entirely in another code.In Example 3, the teacher used a tag amaa belakhare (but eventually) to serve as a discourse marker, inserted at the beginning of the second sentence.

Example 3 (Reza)
T: The boy in the story had a few mistakes.Amaa belakhare, he passed the exam.
(But eventually) As other researchers have also reported (Castellotti, 1997;Turnbull and Arnett, 2002 among others), it was revealed that such CS can render the input more salient and easier to process.
In sum, it can be claimed that the teachers in this study engaged in all the three types of CS in focus.Inter-sentential switching occurred more frequently than intra-sentential switching and tag-switching.The higher frequency of occurrence for inter-sentential CS can be attributed to teachers' intention of giving clearer instruction and eliciting more responses (Qian et al., 2009), or sustaining students' interest and encouraging their participation (Richards, 1998).

Functions of CS
Previous literature on classroom CS has revealed that teachers use CS as both a methodological and a social strategy in classroom interaction (Simon, 2001).This fact was corroborated in our study as well, and the following functions were observed to be the most prevalent functions in our data for which the four teachers applied CS.

For instructions
All the four teachers used CS as a means to highlight key learning points and important task requirements.That is, they applied CS to emphasize certain points or to direct students in correctly completing an activity.For instance, it can be seen in Example 4 that the teacher (Mehran) switched to Farsi to ensure that the students could fully understand how to do a role-play activity.The teacher seemed to be concerned that his students may not understand the directions correctly in the target language, so he resorted to students' L1 (i.e., Farsi) to help their comprehension.

Example 4 (Mehran)
T: You should play the role of a teacher asking your partner some questions.Soaalhayee ke miporsi baayad shabihe soaalhaye safheye 45 bashe.
(The questions you ask should be similar to those on page 45.) Such cases of CS to instruct students were commonly observed in the other three teachers' talk and they have been reported by other researchers as well (Hobbs et al., 2010;Qian et al., 2009;Kim and Elder, 2005).

For efficiency
CS was also commonly used by the four participants to save time and effort and to ensure the optimal efficiency of communication.Previous research (Hobbs et al., 2010;Qian et al., 2009) has indicated that teachers may wisely switch to their students' L1 when they are pressed for time, or when their efforts to define a word in the target language may not lead to optimal results.This is particularly true of Example 5 where rather than giving himself and the students a headache by trying to define the word 'insect', the teacher switched to Farsi.

Example 5 (Amir)
T: This animal usually eats insects.
S: What is the meaning of insect?

For translation
It was observed that, in line with the findings of previous studies (Guthrie, 1984;Kang, 2008;Hobbs et al., 2010, among others), all the four teachers used CS to ensure students' understanding by means of translation as in Example 6.

Example 6 (Ali)
T: Do you remember the story?What happened to the policeman?
Only two or three students responded: "He fell off the ladder." T: Yes, he fell off the ladder.Az nardebaan oftaad.
(He fell off the ladder.) In this example, the teacher immediately translated the sentence into Farsi so as to reinforce the response by those few students and to draw the attention of those who did not respond to his question.Translation, as one of the most commonly used functions of CS, was also observed frequently in all the classrooms when the teacher seemed to find it more effective than other strategies such as miming or paraphrasing.

For praise and/or encouragement
The four teachers were found to apply CS as a strategy to praise their students after successfully completing a task or for their good classroom performance.For instance, in Example 7, the teacher praised a student for his good pronunciation after the student finished reading a paragraph from his book.Apparently, the teacher resorted to the student's L1 (i.e., Farsi) to make sure that the student wouldn't miss any part of the praise, which is beneficial to classroom rapport and student-teacher relationship (Myers-Scotton, 1997;Qian et al., 2009).

Example 7 (Amir)
T: Baarikalaa! Talafozet kheili khoob bood! (Good job!You had a very good pronunciation!) Likewise, CS is often used by teachers to encourage students and boost their confidence, especially when they are faced with difficult tasks.In Example 8, the teacher turned to Farsi to encourage a student, who was standing in front of the classroom and was probably a little scared, to say more.In fact, teachers can use CS, as a strategy, to establish and reestablish rapport and to strengthen solidarity in their foreign language classrooms (Simon, 2001;Myers-Scotton, 1997).

Example 8 (Mehran)
T: What do you see in picture 1?

For correction
Our observations revealed that teachers frequently applied CS as a means to offer correction to their students or to point out their errors.In Example 9, the teacher switched to Farsi to draw a student's attention to his mistake and elicit the correct response.Hobbs et al. ( 2010) and Qian et al. (2009) have also reported CS to be a common and effective means of providing correction by foreign language teachers.

Example 9 (Reza)
T: What is the girl in the picture doing?S: She is going under the slide.
(No, she's not going under the slide; she's going down the slide.)S: She is going down the slide.
T: Yes! Very good! 4.2.6For explanation CS was applied by all the four teachers to elaborate on important points, especially while teaching the target language grammar.To avoid confusion on the part of their students, the teachers resorted to L1 and explained rules of grammar mostly in Farsi, as seen in Example 10, where the teacher was explaining how the structure 'be going to+verb' is used to refer to future plans.
(Simple form of the verb is used after 'to'.)4.2.7 For discipline CS was also used as a means of classroom management when teachers were not satisfied with their students' behavior or performance.This can be seen in Example 11, where students were doing a pair-work activity.While all the students were busy talking to their partners, one of them left his seat to go to another corner of the classroom and the teacher, who was closely observing his students, switched to Farsi to ask the student to go back to his seat.

Summary
In sum, our observations revealed that the four teachers in this study shared a lot in common with regard to their CS patterns.That is, as reported in other case studies of CS (Qian et al., 2009;Hobbs et al., 2010), the teachers did not differ significantly in their CS tendencies, either in the syntactic types or the functions of switches applied.Table 3 show the percentage of the CS functions for which the four teachers switched to students' L1 (i.e., Farsi).

Conclusion
The following conclusions were drawn from the analysis and discussion above: First, this study lends further support to previous studies reporting that CS is a readily available and frequently applied strategy for foreign language teachers in classroom interaction and classroom management (Merritt et al., 1992;Polio and Duff, 1994;Kim and Elder, 2005;Qian et al., 2009).In fact, we agree with Polio and Duff's (1994, p.315) argument that for non-native speakers, "it may be unreasonable to expect the exclusive use of the target language in the classroom" because teachers are expected to use every possible means at their disposal to fulfill their duty of educating their students and to ensure the smoothness of classroom interaction.In this regard, CS can be deemed as a precious resource for bilingual teachers in foreign language classrooms (Kang, 2008;Kim and Elder, 2005).CS, as a discourse strategy, is even more amenable to such contexts where learners and teachers have homogenous ethnic and linguistic backgrounds.Moreover, given the prevalence of large size and occasionally mixed-ability classrooms in some EFL settings (Carless, 2004), teachers would work to their students' benefit if they judiciously used both L1 and the target language.
Second, our findings are in line with those of Kim and Elder (2005) and Duff and Polio (1990) in the sense that although our teachers enjoyed high English proficiency levels, they did not maximize the target language (i.e., English) in their classroom interactions.This claim runs counter to results reported by Carless (2004) indicating that EFL teachers' language use was mainly determined by their target language proficiency.
Third, the present findings confirm that, as other researchers have reported (Mitchell, 1983;Duff and Polio, 1990;Gearon, 1998;Hobbs et al., 2010), foreign language teachers tend to use the learners' L1 more frequently to serve a certain number of pedagogic and social functions such as translation, explanation, praise, etc., which could all contribute to better teacher-student classroom interaction.
Fourth, it appears that we still do not have a clear picture of what goes on in foreign language teaching classrooms; thus, more extensive and systematic investigations into actual classroom practices seem necessary to explore, examine and understand the pedagogic practices undertaken by foreign language teachers.In fact, it is hoped that implications from studies such as ours will lead to a more in-depth understanding of practices and principles underpinning the teaching and learning of English at the local as well as the global level.A key implication of studies like this reveals that caution must be exercised before accepting every EFL policy or paradigm because as Borg (1999, p. 122) points out, suggestions about language teaching practice must be grounded in "empirical accounts of what teachers in classrooms actually do".
In a nutshell, it can be claimed that judicious and skillful use of CS can lead to better teacher-student classroom communication, boost the quality of teaching, help students' comprehension, and foster a healthier friendlier teacher-student relationship, especially for the lower levels.

Limitations and further research
This study is admittedly open to criticism and improvement in many respects.One of the limitations of the present study is the lack of evidence from the teachers' perspectives.The question of why teachers alternate between the two codes for different pedagogic functions needs to be addressed with support from other kinds of data such as interviews.Interviews would have given us a clearer picture of the teachers' motives for their switches.Another limitation of the study is its focus on only four teachers which may not permit us to generalize beyond these particular teachers.Thus, a more comprehensive investigation of the issue can make generalizations to a larger population possible.
Further research may focus on observations of CS practices of a wider range of EFL teachers at different language proficiency levels and with different language experiences to complement the present findings.Future studies may also aim at investigating the effects of CS on students' learning in EFL contexts.
directing students to complete an activity)

Table 2 .
Types of CS in the four teachers' talk

Table 3 .
The distribution of the most frequent CS functions performed by teachers