CHAPTER II LITERATURE REVIEW

CHAPTER II

LITERATURE REVIEW

                      

2.1  Review of Relevant Studies

Research on language, particularly with regard to the use of code switching research is often done by previous researchers. As a comparison, these studies provide significant guidance in the research process. Research in connection with this code switching among others is performed by Syahdan (1996) in his thesis “Sasak Indonesian Code switching” studying on Sasak Indonesia code switching in West Nusa Tenggara Province, Indonesia. This study examines the use of Sasak Varieties, Alus and jamak, by the noble (menak) and commoner (jajarkarang) people, and the use of Sasak Indonesian code switching (both inter-sentential and intra-sentential). This research is in the social context that occurred between employees of the University of Mataram, which directly has no related with this research. Another study was conducted by Fajri et al (1992) with entitled "Alih Kode Pemakaian Bahasa Indonesia oleh Dosen-dosen FKIP Universitas Mataram". This study aimed to describe about the events over the code, as well as factors cause, specifically among lecturers.

Position and the difference between the study conducted by researchers above, both of this research lies in the research object. Because of the above researchers conducting research in the social environment in their life. While in this study, which examined is code switching between English and Indonesian are conducted by lecturers of English and students in the EFL classroom.

Margana (2013) study investigated of code switching in the English learning process in Senior High Schools in Yogyakarta. This study aims of describing code switching in classroom communication from the sociolinguistic point of view. The research subjects are comprised of 12 English lecturers at senior high schools in the Province of Yogyakarta Special Territory. The data were code switching utterances from English to Indonesian and vice versa in English learning in the classroom. The data were analyzed using the qualitative descriptive method. The findings show that code switching has three functions, i.e. (a) material delivery, (b) classroom management, and (3) discourse markers. With reference to such functions, code switching in English learning can be made as far as it satisfies academic functions. As the students’ English mastery improves, the use of code switching needs to be reduced.

The result of this study  is event code switching from English into Indonesian is done by lecturers of English in English language learning in the classroom has a variety of functions that are categorized into three, namely (1) the function of delivery of content, (2) the function of management of the class, and (3 ) functions of discourse markers. Function delivery of material divided into 12 functions. These functions include (1) the function of clarification or confirmation, (2) reiteration function, (3) the function of explanation, (4) the functions of exploration, (5) elaboration function, (6) the function of checking comprehension, (7) the function of the emphasis on certain linguistic elements, (8) function makes inference, (9) develop vocabulary, (10) discussed the tasks for learners, (11) give feedback, and (12) to reflect. Classroom management functions are divided into 12 functions, the function of management of this class consists of a variety of functions, namely (1) attract the attention of students, (2) provide direction, (3) mark the turn of the topic, (4) ask for help learners, (5) overcome tension, (6) maintain discipline learners, (7) provide motivation to students, (8) an appreciation of the students, (9) gives a warning to students, (10) gives a turn to the students, (11) scold learners , and (1 2) maintaining interpersonal relationships.

The findings above implies that the event code switching from English into Indonesian in English language learning in the classroom is not moot because code switching is not done at random by the English lecturers. In other words, code switching from English into Indonesian in English language learning in the classroom needs to be disseminated to lecturers of English for control of the code based on the functions mentioned above. Along with the development of English language skills possessed by learners and lecturers of English, code switching from English into Indonesian can be minimized so that the target language acquisition can be performed optimally.

Similar to this current Margana’s (2013) study is done in the classroom, but the study was conducted in high school in Yogyakarta, meanwhile this research  was conducted at IPDN NTB. Therefore, this research can be used as a reference to this study proposed here.

Nujumuddin (2012) conducted the study of code switching at the Polytechnic of Sriwijaya. The first aim of this study was to examine the use of language in classroom instruction, in particular, the use of code switching (CS) and code-mixing (CM) and linguistic factors utterances and function using CS and CM by the lecturer. Another objective was to investigate the perception of the subjects' on the use of CS and CM. This case study collects data from four professors and their students in the English Department, Polsri. Class corpus analysis identified seven functions in speech communication faculty. The findings of this study also revealed that there are different levels of agreement and disagreement about whether lecturers should use the CS and CM in the EFL classroom.

Based on data collected from the subject questionnaire responses, this study concluded that the CS and CM which aim to improve the efficiency of the information submitted. CS and CM are communicative strategies necessary for lecturers to achieve the purpose of teaching, in particular, involve students who lack English proficiency. Qualitative and quantitative methods are used to achieve the research objectives. Qualitative data were collected in an attempt to understand the different utterances, language contact and communicative function, and the data is diverse and exhausting to classify. Questionnaire in Indonesian using Likert Scale was designed to collect quantitative data to find out from the students' perceptions about the use of CS and CM.

However, he concludes that the use of CS and CM can cause problems. One of the lecturers does not agree with the use of CS and CM with respect to access to input English students. Primary exposure to English input is through an English professor. This condition requires that lecturers use English as much as possible in the classroom and they should be aware of the excessive weakness of the first language in English class. This study shows that faculty face a dilemma in the use of CS and CM, and also shows how complex language learning in Polsri. Finally, it is suggested that first language is used in the right way so that the CS and CM will help faculty achieve their instructional goals.

Subsequent research that are relevant conducted by researchers from abroad like Liu Jingxia (2010) studied at three universities in China to find common situations code switching into Chinese and try to test and prove that the lecturer code switching to first language plays a positive role in EFL classes from universities in China. This case study focuses on the attitudes of lecturers and students reveal the patterns, functions, factors and influences into Chinese code switching in EFL classes in universities of China. This study integrates qualitative and quantitative research methods to analyze code switching to Chinese lecturers through questionnaires of lecturers and students, and recording classroom.

Researchers found that code switching to Chinese is a common phenomenon in Chinese university EFL class, and that it plays an important role in the process of learning and teaching English. This study provides a detailed description and analysis of the general situation and the positive role of code switching to Chinese in EFL classes from several universities in China. However, there are still some limitations that might give some clues for further research.

This research was equally done in the classroom at the university level, and similar with this study.

Lee (2010) studied in Malaysia. The purpose of this study was: (1) To investigate lecturers' attitudes toward code switching (2) To identify the types of code switching (3) Identify the functions of code switching in the instruction (4) To identify the frequency of code switching in the instruction (5) To report the effect over code in the implementation of the English curriculum.

The findings indicate that most lecturers have a positive attitude towards code switching. They just switched codes when needed. Lecturers also believe that over the code has an important role to facilitate learning in a second language.

Furthermore, this study shows that code switching is widely accepted by the lecturer in teaching and learning English in secondary schools in the state of Labuan, Malaysia. This study also had a perspective on the professional judgment of lecturers towards the learning situation. Lecturers are using code switching in the classroom because they believe it helps students learn the target language. The findings suggest that code switching is necessary when a situation requires the use of the mother tongue or the first language in the classroom. Therefore, it is very important to better understand when and why second language lecturers use the first language or mother tongue in the learning process.

Although, there is an absolute consensus that lecturers should use only the target language in the classroom, there seems to be a general view that it should be maximized as much as possible. This is because the main goal of English teaching is to enable learners to learn and acquire language. Future research may be needed to see the extent of code switching can be adopted. In view of educators may find it advantageous to consider the effectiveness of code switching in terms of quality and quantity.

Sana Youkhana (2011) studied with two objectives: the first objective is the researcher trying to see if the frequency of code switching is different depending on the level of difficulty of questions. This study was conducted at three different high schools and involved 24 students. Hypothesis is that the student code of conduct will be less than when answering questions basic level and will be advanced level when answering questions. Small groups were observed and they read the text and answer the eight questions.

The results showed that students perform better on the code-switch advanced level questions and less on the basic level question. They code-switched 15 times the level of the basic questions and 21 times the level-forward question. However, this is not a big difference because there are only six different cases between them. The second objective is the researcher trying to see if people can be categorized switch code. It is based on the opinion of Sert (2005) there are three categories; Equivalence, repetition and holding floor and four other categories discovered by the researchers of this study; Students code switching, Student-researcher code switching, code switching Comfort and Sorry code switching, thus, the seven categories in total. The results showed that the students do different code switches and SERT categories do exist, but they are not enough which is why researchers added four other categories.

It is a phenomenon that is often observed that language learners often switch to their native language in foreign language classes, i.e. they do code switching, but why is it so? As discussed in the introduction, code switching is a common phenomenon among students in foreign language classes. Before researchers make all the observations in this study, he had some questions to be answered and tested his hypothesis. Researchers with the overall goal of this essay are to examine when and why students in high school do code switching in the foreign language classroom. To see a little more in this goal he divides into two parts.

1. Code switching and trouble.

The first objective has to do with code switching and trouble. Results showed that students did indeed code switch in question less than the basic rate of advanced level questions. The students code switching 15 times at the primary level and 21 times at the advanced level questions. Although there are differences between the levels, the difference was not so noticeable because there are only six different switch. However, researchers are still fairly consistent results with previous studies, such as Liebscher and Dailey-O'Cain, among others, which indicates that foreign language learners sometimes can switch back to their native language when they feel find obstacles in the target language, which is what that the participants in this study do in many situations shown above.

2. Code switching and category.

The second objective of this study was carried out with the code switching and category. The results of this study do agree with this SERT category equivalence, repetition and floor-holding and my own category students code switching, the student-researcher code switching, code switching comfort and sorry of code switching. There may be other code switching categories to find, depending on what the situation is like for students when they code switching.

3. Code switching as a part of learning a foreign language

Researchers feel now have a broader understanding of what code switching in foreign language. Now he understands that truly can categorize student code switching at least six different categories. Before researchers think code switching only occurs when students feel they do not have competence in the target language, but with the help of previous studies and this study now researchers have different views on code switching. Researchers also know from previous research, such as SERT which claimed that with the help of code switching they can build a bridge from the known language, the native language of those who are not known the target language. Several previous studies have shown that code switching has a significant meaning in a foreign language class. Even if we want to avoid the native language in the foreign language classroom will always be a part of it, so why not try to profit from it?

Pei-she Weng (2010) studied code switching with aims to present the use of code switching in English class. General English classes with 36 second-year students from different departments (music, Japan, management and so on) and a lecturer in English with over 20 years of teaching experience involved in this research. The EFL classroom is the tape-recorded and then the function code switching is analyzed based on this framework Hymes (1962).

In summary, this study shows that the use of code switching and first language would facilitate the learning of a second language. Thus, first language can be a useful and important element to help L2 learners to learn a foreign language during the learning process. Total of 36 students of non-English second year lecturer in English and experienced in the course of "language training" of a University of New Taipei City involved in this research. Based on the entrance examination in Taiwan, all the students are divided into several classes (A, B, C level). This class belongs to Level B (intermediate level).

In addition, this class consists of several different developments in the university: Japanese Applied, Applied and religious music. Then taught English in this study were the lecturer Taiwanese men over 20 years in this school, which is responsible for language training course at this school. The result of the above found, overall, the findings show that most students have positive attitudes towards code switching used this EFL lecturer in the classroom. First, students like lecturers to use first language in their English classes. Second, students felt the need for lecturers to use first language in explaining the complex rules of grammar, a complex concept, and define new vocabulary. Third, students demonstrate use of first language in the classroom can help them understand complex concepts and reduce anxiety. On the other hand, from the perspective of a lecturer, it is necessary to use first language in the classroom but lecturers still need to consider the purpose of each program. First language is effective in several programs related to some complex concept.

However, if the course is related to the purpose of communication, such as oral practice, lecturers can try to reduce the use of first language and then encourage students to use the L2 in the classroom even though they will make some mistakes. But basically, code switching is a strategy that lecturers can use to help language learners in the classroom. It is suggested that the strategy of code switching in the EFL classroom is not always a weakness in language learning, but it can be regarded as a kind of strategy that is useful in learning a language.

Johansson (2013) study investigated of code switching. This study aims to determine when and why the lecturers in the high school code-switch when teaching L2 English. It also appears in what language students in a choice of different classroom situations. Interviewed five lecturers and 96 students taking English 5 and 6 responded to the questionnaire.

The results showed that lecturers generally try to code-switch as little as possible but that they do code-switch in some situations where students preferred a combination of both Swedish and English or Swedish only. Two of these situations is the instruction of grammar, where the majority of students would prefer a combination of English and Swedish, and the situation of one-to-one, where the majority of students favored Sweden. Clear majority of students (87%) would like their lecturers to make them speak English more. The results showed that the majority of code switching is done by five lecturers in this study were either well prepared and have a clear purpose.

The lecturers tend to code-switch in most situations represented in previous studies such as grammar instruction and in situations one-to-one. This suggests that lecturers experience according to what researchers have found in other lecturers teach. Four lecturers generally code-switch to clarify their teaching, while one of the lecturer, especially the switch for social reasons or because of its lack in the ability of L2. When it comes to students, they prefer a combination of Swedish and English in situations such as grammar explanations and instruction tests. In the situation of one-to-one and class discussions they prefer Sweden. In addition, they tend to want their lecturers to make them speak English more. There is a consensus between the lecturer and student views when it comes to the field of what lecturers code-switched.

A remarkable fact though, is that there is no claim to the lecturer asks the students about what language they like in different situations. Although experience is an important asset when becoming a lecturer, much can be learned by asking students what they like and take their opinions into account when planning for the use of code switching in teaching someone. Code switching in high school can indeed be further investigated. It would be interesting to know the number of unintentional code switching between English lecturers in high school and saw the extent of code switching activities they plan ahead. It will also be interesting to study the lecturers are native English speakers and see if and how they code-switch. This can be done with a combination of classroom observations and interviews.

Christoffer Jakobsson and Henrik Ryden (2010) wrote a paper on the degree of code switching in four Swedish EFL classrooms. The aim of the dissertation is to investigate when and why the code switching occurs and attitudes towards code switching between lecturers and students in four EFL classrooms in two secondary high school (Jakobsson & Ryden, 2010: 8). The conclusion from this study is that lecturers observed switching code 12 times for eight lessons observed. Five of those times when explaining another task for students who do not understand and interviews also indicate that lecturers are reluctant to use Sweden during the lesson and only do so in cases when a student requires more instructions or do not fully understand the task (Jakobsson & Ryden, 2010: 30). Another conclusion from this study is that the lecturers stated that their main reason for code switching when teaching / explaining grammar and code switching it unacceptable for lessons although they also admitted the importance of it. (Jakobsson & Ryden, 2010: 34).

All the researches show that code switching is unavoidable phenomena in the world. It will be occurred in local, national or international, formal and informal situation, social or classroom context in bilinguals or multilingual.

2.2  Theoretical Review

2.2.1 The concept of Code switching

Under this section, the definition of code switching and some related terms of code switching will be presented. It involves the definition of code switching, the difference between code switching and code mixing, the difference between code switching and borrowing, base language and embedded language, marked and unmarked code switching, and situational and metaphorical code switching.

2.2.2 Definition of code switching

The term “code switching” was first used by Haugen (1956 in Alenezi, 2006) to refer to the use of unassimilated words by a bilingual speaker from a different language. According to Haugen (1956 in Alenezi, 2006), the term code switching refers to an alternation used between two or more languages.

Many sociolinguists defined code switching in widely various ways. For example, Grosjean (1982: 145) defines code switching as the alternate use of two or more languages in the same utterance or conversation. Likewise, Valdes-Fallis (1978) defined code switching as a process of alternation between two languages while maintaining a level of proficiency in both languages. Alternately, Di Pietro (1978) defined this phenomenon as the use of two languages by participants to achieve a speech act. Poplack (1980) defined code switching as the alternation of two languages within a single discourse, word, or sentence. Additionally, Romaine (1994) mentions that code switching refers to the alternate uses of two languages in the context of bilingualism and multilingualism. Myer-Scotton (1977) described code switching as the use of two or more linguistic inputs within the same discourse. In the context of monolingualism, Romaine (1995) further states that code switching refers to the use of different communicative styles and dialect within the same languages; for example when the speaker use formal and informal style in utterance.

In addition, there are some experts argue that the term code switching can be applied to monolinguals as well as bilinguals. Hudson (1996), defines code switching as the use of different languages at different times by the same participant. Zentella (1981) similarly, argues that this term applies to monolinguals because they may switch between different styles within the same language. For example, when a native speaker of American English speaks in British accent, this is described as code switching in Zentella’s definition. Clearly, there is not a consensus among the researchers as to what, exactly, code switching is. The variation of the definition of the term code switching is due to the ambiguous definition of the word language itself. Crystal (1987: 363) defines code switching as switching between languages, however, as the definition of language is tenuous at best, perhaps it is better to say switching between varieties in addition to switching between languages”. Code switching, according to Gumperz (1982: 59) is ‘juxtaposition within the same speech at move of passages of speech belonging to two different grammatical systems or subsystems”. He simplified this, saying that code switching is alternating between two or more languages within the same interaction.

From this overview of the term code switching, it is clear that different researchers use different definitions of the word code switching. In this study, the definition of code switching offered by Grosjean (1982) and Poplack (1980 in Yletyinen, 2004: 8) will be used, as it seems to be more comprehensive. They state that code switching is a pattern of alternation between two or more language by bilinguals in the same discourse or conversation.

2.2.3 Code switching versus code mixing

In the same cases, code switching and mixed code that is used as a complementary term. In a sense code switching presented to the turn of the language between sentences, while the code mixing is language that is used to turn the two languages in a sentence (Winford 2003: 105). However, sometimes the term code change is also used when referring to switching between sentences (Lauttamus cited in Yletyinen, 2004:9). However, both code switching and code mixing can also be used as a cover term, that is, they are used for all types of turnover (Pandit as cited by Yletyinen, 2004:9). Auer (1995; 1998), in turn, uses the term code alternation to refer to code switching (Yletyinen, 2004:9).

2.2.4 Code switching versus Borrowing

Borrowing refers to the use of alternative two-level language lexicon; in this case, the word is put in one language phrases borrowed from other languages (Kovacs 2001: 63). Borrowings are also intended to introduce the item word or phrase idiomatic one language to another with different grammatical systems (Gumperz, 1982: 66).

Poplack (1980: 584-585) stated that the switch is considered a borrowing, there must be non-native put in another language sentences in three integration; phonology, morphology and syntax. Conversely, if the item is treated as a non-native code switching, they must have only one type of integration (e.g. morphological integration). Nevertheless, Myer-Scotton (1993) found that differences between code switching and borrowing, in bilingual speech is not important to analyze.

Because this study aims to explore over the code in the foreign language classroom, the theory of Myers-Scotton (1993) is adopted that does not look over the code switching and borrowing as two different processes also found no such differences become important. Therefore, this study only focused on the use of the term code switching to better describe and analyze the data. In this case, the term code switching strengths include the types of code switching, i.e. intra-sentential, inter-sentential and tag-switching.

2.2.5 Other concepts related to code switching

There are other terms relating to the code switching: base and embedded language, and marked and unmarked code switching. In line with the existence of base and embedded language in discourse, Myer-Scotton (in Yletyinen, 2004:11) explains that:

There is a matrix language which sets the structural rules and to which items from the other language, the embedded language, are borrowed. The matrix language is also known as the base language. However, some researchers deny that there is a base language (Yletyinen, 2004:11).

Despite this fact, it is not clear how to distinguish between the basic language and language embedded in speech. In this regard, Musyken (1995: 182) argues that even if there is a basic language; difficult to determine what it is, because the definition depends on whether people choose, point-oriented, psycholinguistic oriented or grammar-oriented viewpoint. Outlook oriented discourse means that the base language is the language of interaction. Outlook oriented statistics will refer to the base language is presented as spoken words and most of psycholinguistic ability to determine the basic language speakers. The grammarians find initial utterances determine the basic language, because they can be 'guided' the utterance (e.g. governing verbs).

In the classroom discourse is difficult to say which language should be referred to as the basic language and as an embedded language (Simon, 2001: 320). This is because the language of discourse can change from one task to another, for example, when studying the basic grammar can be Indonesian, but when doing oral discussion the basic language could be English. In addition, to make distinctions of the above description, Myer-Scotton (1988, 1989) distinguish between unmark and marked code switching. These terms are correlated with the social relations of the speakers in conjunction with each other. According to Myer-Scotton (1989: 334), which unmarked code switching occurs in the speech that particular code option will show '' rights and obligations are expected between participants ", that is, follow the norms of society. While a marked code switching causing moved from the expected relationship of the participants to readjust their social distance (Myer-Scotton, 1989: 334-335).

In other words, a marked code is unexpected. In foreign language classes an unmark code can be characterized conventional code, i.e. the expected one, whereas a marked code is an unexpected one. For instance, in English classroom in Indonesia, the code can be unmarked with the English language as a communicative task, however, when teaching grammar unmark code can be Indonesian. The above concepts, i.e. basic language and embedded language code switching used in research of code switching. The study recognizes the different concepts related to code switching.

However, as noted above, not all of them can be used in classroom research. This study will utilize only basic difference between the base language and the language embedded in the analysis. Both terms have relevance in EFL classes because sometimes they can help explain the situation where code switching occurs; therefore, they will be used when needed.

2.2.6 Types of code switching

According to Poplack’s  (1980), there are three types of code switching: inter-sentential switching, intra-sentential switching and tag-switching, which is also called extra-sentential switching by Milroy and Musyken (1995).

Inter-sentential switching occurs between sentences, ie when the switch occurs at the boundaries of the sentence in which each sentence in a different language (Romaine, 1995: 122). Furthermore, it can also occur between turns. This type is mentioned with the least integration of code switching happens between sentences. For instances Sometimes I’ll start a sentence in Spanish y term inó en espanol (sometimes I’ll start a sentence in Spanish and finish it in Spanish).  In other words, when the switch occurs at the level of the sentence or change, it belongs to the inter-sentential code switching.

Tag-switching, proposed by Poplack (1980), which occurs when a word or phrase from one language inserted into another language sentences. In this case, the term 'tag-switching', she also uses the term extra-sentential switching to refer to the type of switching. However, the term is intended to refer not only to the tag-switching but also for inter-sentential switching. In this way, she made a clear distinction between it and the intra-sentential switching. Meanwhile, Milroy and Mysken (1995) employs the term extra-sentential switching or emblematic switching to refer to tag-switching.

In addition, according to Dumitrescu (1993, in Ene, 2006), tag switching or emblematic switching, which is defined as switching on the tag level, and includes at least two types of tags. Tag or emblematic switching can be a single noun (for instance, high frequency, habitual speech, culture-specific term and question tag, such as, honey, well, okay, yes and alright), or, most often, short sentential formulas, and question tag (for instance, isn’t it, doesn’t she, aren’t you, are you kidding, give me a break, you know, and I mean). Of the different term used by the different experts, this study followed Poplack’s (1980) and Damitrescu term of ‘tag-switching’ instead of extra-sentential switching when talking about switches that are neither inter nor intra-sentential switches.

The third type of code switching by (Poplack, 1980) is intra-sentential switching which requires a lot of integration and is usually associated with the fluent bilinguals. She mentions that intra-sentential switching occurs in a sentence or clause and involves risks since the syntax of other languages clause inserted into the first language sentence. As a result, if the two languages are mixed in a sentence, there will also be two different grammars in play, which means that the speaker should know better grammar to generate the correct grammar in speech. This is to say that the speaker must know the grammatical system of two languages to avoid grammatical utterances.

It is argued that the use of three types of code switching indicates the speaker’s proficiency level (Poplack, 1980: 605). Adept at least two languages most used tag-switching because tags can be produced with little knowledge of the grammar of L2. Meanwhile, inter-sentential switching commonly used by more proficient bilingual speakers. And the most proficient speakers often use intra-sentential switching. In this case, the intra-sentential switching requires the highest level of bilingual capability compared with the two types of code switching because the speaker must master the grammar of both languages to produce grammatically correct clauses with different grammatical systems in a sentence.

2.2.7 Function of Code Switching in Classrooms

In this section, functional categories of code switching will be presented based on a classroom context purposed by the study of Merritt et al., (1992), Canagarajah (1995) and Yletyinen (2004). They proposed different functions of code switching classroom. These functions are derived from the results of those studies focused on investigating the EFL classroom code switching. It is mentioned in Merritt el al., (1992) that there are two functions of code switching. The first function of code switching is to bring new information to the discourse. And the second function of code switching is to translate a word in one sentence.

Another researcher Canagarajah (1995) and Yletyinen (2004) studied the function of code switching in EFL classroom. Canagarajah propose micro and macro functions code switching in classroom. Micro-functions which are further divided into two categories: classroom management and transmission of content. It is under the management functions of the classroom, consideration of how code switching facilitate lecturers and students to organize classroom interactions systematically and efficiently under surveillance (Canagarajah, 1995). “Content transmission means the fact that code switching can help in the effective communication of the lesson content and language skills which have been specified in the curriculum” (Canagarajah, 1995: 179).

The function of classroom management are opening the class, managing discipline, negotiating directions, lecturer’s commands, lecturer encouragement, requesting help, lecturer compliments, mitigation, lecturer admonitions, pleading and unofficial interactions. The functional of content transmission categories are definition, explanation, parallel translation and unofficial student collaboration negotiating cultural relevance, review.

Macro-functions dealt with socio-educational implications, which included training the students for the social and communicative life outside school, since bilingualism persists through code switching in Jaffna. The use of English in the classroom is set as formal and official, which means that Tamil language is used for extra pedagogical purposes, for example, for discussing personal matters. In other words, micro-functions dealt with issues in the classroom whereas macro functions had connection to issues outside the classroom (e.g. bilingualism and language attitudes) (Canagarajah, 1995).

While Yletyinen (2004) proposed micro-function of code switching in classroom, they are: lecturer’s explanation/clarification, requesting help, students helping each other, students self-repair, lecturer’s language slip (lapses), unknown English counterpart, checking for understanding, student clearing misunderstandings, students’ initiation, and students’ comment.

Canagarajah’s (1995: 190) study furthermore shows how English and the mother tongue, Finnish, were used in different situations. There were some general patterns in the classrooms: English was used in interactions dealing with the lesson content while Finnish was used for personal or unofficial interactions. In other words, English is only reserved for interactions that are demanded by the textbook and lesson. This is in line with finding of Merritt et al. (1992) and Yletyinen (2004) the mother tongue is the less formal language while English is used in more formal way. Moreover, Canagarajah (1995: 190) found out in his study that English was the code which symbolized informality, personal and homely features.

Regarding the function of code switching, the finding of Canagarajah (1995) divides the micro and macro functions of code switching in a very detailed way while Yletyinen only focused on micro-function. These particular functional categories are used when doing this research, because some of them can be applied to the situation in Indonesia classroom.

As mentioned, Merritt et al. (1992) use the term types while Canagarajah (1995) and Yletyinen (2004: 16) use the term functions when categorizing code switching. However, when these studies are compared, similarities between the types and functions are found. Merritt et al. (1992: 114-117) describe the types as follows: reformulation, bringing new information to the content of activity, translation or word substitution and finally code switching in interactional particles (e.g. discourse markers). First of all, the type ‘reformulation’ finds its counterpart in Canagarajah’s function of explanation, and more precisely the strategy of repetition. Both reformulation and explanation are used to say the same thing using a different language.

Secondly, bringing new information to the content of activity is the types of code switching purposed by Merritt et al., which is similar to Canagarajah’s function of explanation Canagarajah (1995: 186) defines explanation as having many strategies; repetition, reformulation, clarification and exemplification just to name a few. This is to say that explanation can also bring new information to the activity at hand, thus the similarity with what Merritt’t type. In addition, Gumperz (1982: 78 in Cogan, 2003), points out that “Explanation” is one common function code switching, to refer to the message switch from one language to another repeatedly and can be used to explain or repeat what was previously said.

Thirdly, Merritt’s term ‘translation or word substitution’ is equivalent to Canagarajah’s function of definition. The purposes of the both categories are to ensure that students know what is being said by translating a word or a few words from L2 to the mother tongue. The difference is that Merritt et al talk about translation occurring within a sentence whereas Canagarajah only stated that the mother tongue is used in the form of a single lexical items or loans to define new vocabulary.

Fourthly, Merritt et al (1992: 116) state that the forth type of code switching consists of interactional particles such as discourse markers, classroom management routines and terms of address. Canagarajah (1995: 184) talks about mitigation as one of the functions. It consists of discourse markers and tags. These two categories have similarities in both introduce discourse markers as a part of the category. However, Merritt et al. do not provide an example of this category, thus the complete comparison is impossible.

Furthermore, Merritt et al. (1992: 117) stated that the linguistic markers that indicate a shift in topic used smoothly which means that they may have a slightly different meaning to the interactional particles. As a result, the forth type of Merritt et al is not a direct match to Canagarajah’s and Yletyinen’s but this comparison has shows that researchers in two categories study discusses of the different terminology that has certain similarities.

2.2.8 Morpheme Constraint Theory

David Sankoff and Poplack Shana model of code switching is the most thorough in accounting for alternational code switching (Winford, 2003: 126-127).  In this model, code switching constraints imposed two. Free morpheme constraint specifies that code switching cannot occur between lexical stems and bound morphemes. Basically, these constraints distinguish code switching from borrowing. Generally, borrowing occur in the lexicon, while code switching occurs both at the level of syntax or utterance-level construction (Gumperz, 1982;  Poplack & Sankoff, 1984; Muysken, 1995). The equality constraint predicts that switches occur only at points where the surface structure of language coincide, or between sentence elements that are usually ordered in the same manner by each individual grammar (Winford, 2003: 126-127). For example, the sentence: "I like you porque eres simpático" ("I like you because you're good") allowed for obeying the rules of Spanish and English syntax (Sankoff & Poplack, 1981).  Cases such as noun phrases and white casablanca houses ruled out because of the grammatical combination in at least one of the languages ​​involved. Spanish noun phrase consists of a determinant, then the noun, the adjective, while the adjective comes before the noun in the English noun phrase. Casa White ruled by equality constraints for not obeying the rules of English syntax, and home blanca ruled out because it does not follow the rules of syntax Spain (Winford, 2003: 126-127).

Critics say the weakness of this model of Sankoff and Poplack. Free-morpheme and equivalence constraints are less stringent, which means there are many exceptions that occur. For example, a free morpheme constraint does not explain why switching possible between certain free morpheme. Sentence: "The students have visto la pelicula italiana" ("The students have seen the Italian movie") does not occur in Spanish-English code switching, but the free morpheme constraint seems to assume that it can be (Belazi, Rubin, Toribio, 1994). Equality constraints will also ignore the switch is happening generally in languages​​, such as Hindi postpositional phrase that is activated by the English prepositional phrases such as in the sentence: "John gave a book larakii ek ko" ("John gave a book to a girl"). Larakii ek ko expression literally translated as a girl's, so grammatical in English, but this is a phrase that occurs in English-Hindi code switching despite the requirement of equality constraints (Winford, 2003: 126-127). Sankoff and Poplack model only identifies the point at which switching is blocked, as opposed to explaining the constituents can be turned on and why (Winford, 2003: 126-127).

2.2.9 Markedness Theory

Model markedness (sociolinguistic theory) proposed by Carol Myers-Scotton is one of the indexical account of social motivations for code switching. The model states that speakers use the language of choice for indexing Rights and Obligations (RO) Sets, abstract social codes in operation between the participants in a particular interaction (Myers-Scotton, 1993). 

According to Myers-Scotton, (1993) for each communication situation there is a marked, expected RO set and marked, differential one. In choosing a speaker code markedness evaluate their potential choice, determined by social forces at work in their communities, and decide either to follow or reject the normative model. Making such choices marked with a conscious effort to set a new RO. The speaker uses a rational choice of code, as a way to build their social position in accordance with the principle of negotiations: "Choose the shape of your conversation contribution such that the index set of rights and obligations that you want to be in force between the speaker and the receiver for the current exchange". Characterized choice is often accompanied by prosodic features such as pause, or meta commentary on the switch. When the choice is not clearly marked, speakers use code switching by means of exploration to maintain social balance favored. Myers-Scotton markedness has been proposed that the model is valid for all language options, beyond the limits of code switching.

The model operates in matrix theory markedness-frame language that Myers-Scotton, production-based explanation for code switching which argues constraints on the switch at the level of the mental lexicon (as opposed to the structure of the surface). This theory states that code switching speakers alternate between Matrix Language (ML) and the Embedded Language (EL). ML is a language that is more active and more often used, which restricts the use of EL. It is common, though not required, to ML to fit the option marked in typical interactions.

2.2.10    Matrix language-frame model

Matrix Language-Frame Model Carol Myers-Scotton is the dominant model of insertional code switching (Winford, 2003: 126-167). MLF models found no Matrix Language (ML) and the Embedded Language (EL). In this case, the elements of EL inserted into the frame morfosintaktis of ML. The hypothesis of this study are as follows (Myers-Scotton 1993b: 7):

The Matrix Language Hypothesis states that the procedures in the central structure grammar in language production system that takes into account the surface structure of the Matrix Language + Embedded Language constituents (linguistic) Matrix-based procedure is only languages​​. Furthermore, the hypothesis is intended to imply that the frame-building precedes the content morpheme insertion. Matrix Language can be seen as a first language or the language of the speaker where the morpheme or word more frequently used in speech, so that the dominant language is English and the other is the Matrix Embedded language. Also, an island Matrix Language is entirely composed of constituent morphemes Matrix Language (Scotton, 1993).

According to the Blocking Hypothesis, in the Matrix Language + Embedded Language constituents, block content filter blocking Embedded Language morphemes that are not congruent with the Matrix Language respect to three levels of abstraction on subcategorization. Conformity is used in the sense that the two entities, linguistic categories in this case, are congruent if they are appropriate in terms of relevant quality.

Three levels of abstraction are: (1) Even if Language Embedded realize grammatical category given as a morpheme content, if the system is realized as a morpheme in the Matrix Language, Language Matrix block the Embedded Language morphemes content. (A morpheme content is often called "open-class" morphemes because they belong to a category that is open to the discovery of new items arbitrarily. They can make words like "smurf", "nuke", "byte", etc.., And can be said objects, verbs, adjectives, and some prepositions. A morpheme system, for example, function words and inflections, revealing the relationship between the content morpheme and not give or receive thematic roles. (2) The Matrix also blocks Embedded Language morphemes language content in this constituency if not congruent with the Matrix Language morpheme content partners in terms of theta role assignment. (3) Correspondence between content morphemes Embedded Language and Language Matrix content morpheme is realized in terms of discourse or pragmatic function.

example
(1).  life ko face kiijiye with himmat and faith in apane aap. (Code switching)  

      "Face life with courage and faith in self." (Translation) (Hindi / English)

(2). hata wengine nasikia washawekwa cell. (Code switching)

      "Even people I heard was placed [in] cell."(Translation) (Swahili/  

         English).

        (http://Wikipedia.org) accessed on August 18, 2014.

We see that the examples 1 is consistent with Hypothesis Block and content morphemes criteria system, so the prediction is that the Hindi equivalent also morpheme content. Sometimes non-conformity among peers in language and language Matrix-Embedded can be circumvented by accessing the naked form. "Cell" is a form of bare and so the role of thematic "cell" was given by -wek- verb 'put in / on'; This means that the verb is a content morpheme.

Embedded Language Island Trigger Hypothesis states that when the Embedded Language morphemes are not allowed to appear in either matrix language Hypothesis or Blocking Hypothesis, triggers inhibition of all procedures to access the Matrix Language and complement current constituents as an Embedded Language islands. Embedded Language Island consists of Embedded Language morphemes and well-formed by Embedded Language grammar, but they are included in the Matrix Language frame. Therefore, the embedded language island is under the constraints of the Matrix Language grammar.

  Example:

(1). Sikuona your barau ambayo uliipoteza. (Code switching ungrammatical)

    "I did not see your letter which you lost." (Translation).

(2). Nikamwambia anipe ruhusa niende ni-ka-check for wewe. (Code switching

        ungrammatical)

        "And I told him that he should give me permission so that I go and check

          on you." (Translation)

    Nikamwambia anipe ruhusa niende ni-ka-check for you. (Code switching

            grammatical).

           (http://Wikipedia.org) accessed on August 18, 2014.

Example 1 is not grammatical because "you" accessible, so embedded Island Trigger Hypothesis predicts language that must be followed by an English head (eg your letter) as an Embedded Language islands. The reason is that it is a possessive adjective morpheme system. We see the same thing happening in examples 2 and therefore is not grammatical. However, the correct way to finish the sentence is not "for wewe", where he switched back to Nairobi but should end up as "for you" because it should end in an Embedded Language islands.

Embedded Language Implicational Hierarchy hypothesis can be stated as two sub-hypotheses: (1) constituency is farther away from the main argument sentences, which are free it is to emerge as an Embedded Language islands. (2). It is formulated in constituent structure, the more likely to emerge as an Embedded Language islands. Other more powerful, the choice of any part of the idiomatic expression will result in an Embedded Language islands (Winford, 1993: 126).

Implications Hierarchy Embedded Language Islands: (1) Formulaic expressions and idioms (mainly due to the time and manner prepositional phrases but also as a verb phrase complement). (2) another time and way of expression (3) Quantifier expression (4) Non-quantifier, non-time noun phrase as a verb phrase complement (5) Agent (grammar) Noun phrase (6) Role Themes and case-assigners, namely the limited main verb (with full inflection)

      example
1). Le matin de bonne heure ngay joge Medina pour dem juilli. Suba tee nga

         fa  war a joge. (Code switching)

        "Early in the morning you leave Medina to go to pray. Early in the morning  

         you have to leave it." (Translation) (Wolof / French)

   2). Ulikuwa ukiongea a lot of nonsense. (Code switching)

        "You talk a lot of nonsense." (Translation) (Swahili / English).

            (http://Wikipedia.org) accessed on August 18, 2014.

We see examples 1 work because of the French Embedded language island "early in the morning" is an expression of time. Also, as a side note it is repeated in the Wolof in the second sentence. In example 2, we see the quantifier is a predictable Embedded Language islands and here to see a complementary purpose finite verbs start with the quantifier.

      2.2.11 Psychological and Pragmatic Functions of Code Switching

According to Franceschini (1998) code switching appears in the setting where two or more languages may be used by the speaker. This is clearly the case in Switzerland that has a variety of languages such as Germany, Switzerland and Italy. How to code switching appears in the EFL classroom is different than that used for the bilingual code switching in a communicative dialogue, because these speakers use code switching more regular basis because they are often members of a multicultural society and thus code switching comes more naturally to them (Valdes-Fallis, 1978).

Code switching is closely connected to speech situations and interpersonal relationships that affect them (Halmari, 2004). Code switching in EFL classroom is much more complex to deal with code switching between bilingual in a social environment. This is because the role of students in the EFL classroom is to use the target language (Simon, 2001). The most common reason students to switch to their native language for foreign language study are that the mastery of a foreign language is not the same as that of their native language or their lecturer mastery of a foreign language (Simon, 2001). The switch back to the original language gives students a natural opportunity to retreat to a safe zone when the language usage exceeds grade-level linguistic competence of learners (Simon, 2001).

Student choice code is closely related to the type of the task at hand and the needs of learners to communicate their understanding of the information presented by the lecturer in the target language. In a social environment foreign language used to convey ideas, debate and for general communication while in a foreign language EFL classroom is used to understand the target language itself. In other words, in the EFL classroom we communicate about the communication itself (Simon, 2001). In addition, students use their native language to communicate between each other and thus they get a response is understandable if other learners have the same or a different perception of the information received. All of this is done so that learners can negotiate meaning in a simple way and thus help their own learning process (Simon, 2001).

2.2.12  Classroom Instruction

By definition, classroom instruction is “live instruction on the collegiate level which allows immediate interaction between student and instructor, including lectures, direction, questioning, laboratory instruction, seminars, colloquia, independent study, interactive instructional television, and interactive computer-aided instruction” (Zirkin and Sumler, 1994:2). In other words, classroom instruction is the medium that is used by the lecturer to interact with the student in the classroom. “Most interactions in the classroom had the objectives of requesting and giving information. Some related strategies are used; these include questions, statement, explanation, or narration” (Azis, 1987: 63). Moreover, according to Cazden (1987:1):

“classroom discourse is inter-individual communication but the primary goal of education is intra-individual change. In relating inter-individual communication to intra-individual change, we talking about transformations from conversations to cognition or, as Barnes (1976) said, ‘from communication to curriculum.”

In other words, the communication in the classroom makes a major contribution to the curriculum, or the subject being taught by the lecturer. They speak and explain how the lecturer in class affect students' understanding of the lesson. The way the lecturer speaks including language s/he uses, vocabulary s/he employs, structure, gestures, and many others.

In addition, Azis (1987:59) also wrote that “the pattern to lecturer-student verbal interaction would be determined by the teaching strategies used by the lecturer in the classroom”. In IPDN, one of the strategies used by the lecturers in order to give sufficient English exposure to the students is to use English as the medium of the EFL classroom instruction.

2.2.13 Code switching in foreign language classrooms

Although code switching research is mostly associated with the field of bilingual environments and communities, code switching in the foreign language classroom is, according to Sert (2005) an extensively observed phenomenon. In their work on code switching, Milroy and Muysken (1995: 90) state that research on code switching in the classroom has been conducted for almost two decades. Simon (2001: 313) claims that there has been a development in the research of code switching in foreign language learning. The language classroom has become interesting for researchers. According to Milroy and Muysken (1995: 90) code switching in the foreign language classroom is international; there has been research on this in the United States, South America, Canada, Europe and South East Asia.

Let us now see what happens when students code-switch in the foreign language classroom. Liebscher and Dailey-O’Cain (2005: 234) suggest that foreign language learners switch back to their native language when they feel they meet obstacles in the target language conversation. Sert came up with some categories of code switching in a foreign language classroom.

The first category is called Equivalence, which occurs when the student lacks competence in the target language, such as when s/he feels that s/he is not competent enough to explain something in the target language. The student therefore instead uses lexical items from the native language. This process is, according to Sert a sort of defensive mechanism.

The second category is called Floor-holding. Here the students use native language words to fill gaps in the conversation in order to avoid breaks or open spaces in the conversation. Sert claims that this process may have a negative outcome on language learning if students continue with this type of code switching for a long period of time. They may lose the competence of fluency in a conversation.

The third category is called Reiteration. Students use this function in order to reinforce and clarify a message. Sert claims that students may repeat words and phrases in their native language because they feel they did not clarify a message in the target language but also to show the lecturer that s/he has understood the task or content in the situation. Heredia and Brown (2005: 214) state that people often do it in order to be understood better. According to Yule (2010: 194) there is one thing called Communicative competence, which means that L2 learners try to use the foreign language correctly.

Rababah (2002: 6) states that there are other strategies within communicative competence. One of them is called interlanguage communication strategies, which means that L2 learners use different types of strategies to get their message through. The students want to organize their messages quickly to avoid communication problems. Typical behavior would be: use the words of the language, muttering, repeat their original sentences and words, try to avoid certain words that they may find it difficult, repeat words and sentences, ask others for the word or phrase true, and correct themselves using self-correction as Rababah call.

Simon (2001: 314) suggests that code switching in foreign language classrooms is much more complex to scrutinize than code switching in social settings. The students in the foreign language classroom often have vague knowledge of the target language compared to bilinguals in a social setting. There is indeed a difference between code switching in educational settings and in social settings. According to Wei and Martin (2009: 117) code switching in educational settings is often seen as unsuitable and wrong, while code switching in social contexts is seen as something natural and a part of bilingual speech.

2.2.14 Lecturers’ use of code switching in the classroom

The functions of code switching in ELT classrooms, Sert (2005) argues that the lecturers’ use of code switching is not always a conscious choice, and the lecturers are therefore not always aware of the functions and outcomes of the code switching process. He serves as a switch list of topics, affective functions, and repetitive functions (Sert, 2005: 2). In the case of switching topics or language lecturer changes according to the topic being discussed. Sert indicates that this widely used when teaching grammar, that lecturers shifted from language to language. He argues that when the lecturer's attention shifted to the student's mother tongue is directed to new knowledge that is being taught. In this way lecturers build bridge (native language) which is known to the unknown (new foreign language). (Sert, 2005: 2) states that the SERT switch topic also carries affective functions that serve the purpose of expressing emotions. In this case, code switching is used by lecturers to build solidarity and intimate relationships with students. Sert shows that in this case the code switching contributes to creating an environment supporting language in the classroom (Sert, 2005: 3).

The third function of code switching is a function of the lecturer in the classroom. In this case the lecturer using code switching to transfer the necessary knowledge to students for clarity; lecturer shows the most important thing of instruction, which is given in a foreign language, the students' native language. Sert also shows that the clarification of code can have a negative impact as undesirable student behavior; students who believe that will follow the instructions in the original language will not listen to instruction in foreign languages. (Sert, 2005: 3) Conclusion Sert is that it can be suggested that the code switching in the classroom is not always blockages or deficiencies in language learning, but could be considered a useful strategy in classroom interaction. But Sert also indicates that it should be remembered that the long term, when students have interaction with native speakers of the target language, code switching may be a barrier that prevents mutual clarity (Sert, 2005: 5).

2.3   Theoretical Framework

In connection with the above description, this study uses the term code switching which is defined as the transfer from English into Indonesian or vice versa is done by an English lecturer in IPDN NTB Campus for communication in the classroom. In addition, understanding the code in this case is limited to the form of language (English and Indonesian) are transferred.

2.3.1 Types of code switching in EFL classrooms

As explained before, there are three different types of code switching: intra-sentential   code switching,    inter-sentential    code switching    and    tag-switching (Poplack 1980). In the following, the present analyzed according to these three types. This analysis serves as a starting point to the later analysis of the functions of code switching.

2.3.1.1 Inter-sentential code switching

As suggested above, inter-sentential code switching occurs between sentences or clauses, or between turns. In the type of code switching is used in school when, for example, translating or explaining something (grammar, exercise etc.).  Both  the  lecturers  and  the student use inter-sentential code switching; furthermore, this type of code switching  is  used  quite  a  lot  in  the schools. Inter-sentential code switching can also occur between turns. In such a case the previous turn is in language A and the following in language B. A change of language occurs at a clause or sentence boundary, where each clause or sentence is in one language or the other. For instance:

        If you read it again, kamu akan mengerti artinya.

2.3.1.2 Tag-switching

Tag-switching means inserting a tag in one language to an utterance that is otherwise in another language. In classrooms this means that while speaking English the lecturer or a student can insert an Indonesian tag to the utterance, or vice versa. Furthermore, a tag can be moved freely in a sentence, they do not have syntactic constraints. Tag-switching occurs in secondary school, but there are only a few instances of it there. It is mostly employed by the students. Tags and certain set phrases in one language are asserted into an utterance otherwise in another language. For example: Please deh jangan ganggu aku terus, you know?        

2.3.1.3 Intra-sentential code switching

Intra-sentential code switching occurs within a sentence. According to Poplack (1980), it requires a lot of integration and therefore it is only used by the most fluent bilinguals. However, Poplack’s view may be true of naturally occurring discourse, in classroom discourse there is intra- sentential code switching although the participants are not all fluent bilinguals. This type of switching occurred within a clause or sentence boundary. For instance:

          So, all our effort sia-sia dan tidak berarti

Figure 1 shows the different types of code switching and the degree of language mixing in a sentence. One can see that in the first, inter-sentential switching, no code switching in a sentence but two different languages in different sentences or clauses. In tag-switching, has a little switch in a sentence or clause, but this code switching usually tag. The circle in the figure shows this as two interlocking circles consist of a sentence or clause in the middle where there is little code switching. In intra-sentential switching, is the largest number of code switching.

Figure 1. The types and the degree of code switching

                                                     (Adopted from Poplack in Yletyinen 2004: 16)

2.3.2       Functions of code switching in EFL classrooms

The focus of this part explained different functional categories that found in the classroom. Base on this function, the researcher  analyze the function of code switching in the classroom detail in this study. In other words, the functions of code switching will deal with them. According to Canagarajah (in Yletyinen, 2004: 53), there are many functions of the code switching in the classroom, are as follows:

2.3.2.1  Explanation

Explanation occurs when (usually) the lecturer wants or sees a need to repeat what has been previously said in another language in order to help the students understand him/her. In an EFL classroom this explaining generally happens in first language, the mother tongue of the students, since the students are less competent in the foreign language and may need an explanation to help them understand the lesson better. Canagarajah (in Yletyinen, 2004:53), stated that there are different strategies for explaining the issues being taught; she mentions repetition, reformulation, clarification and exemplification as such strategies.

2.3.2.2  Requesting for help

When students are faced with a problem or question during the lesson, they usually resort to code switching to find an answer to their problems. One common function of code switching is requesting help. This function is employed by the students; they use it when they want to ask for help. The students request help when they do not know where they are supposed to be in the book (for example wanting to know about line numbers in a chapter), or when they do not know how to pronounce a word or when they need a word translation or when they want to ask something from another student (Yletyinen, 2004:58).

2.3.2.3 Students helping each other

The previous function dealt with student-lecturer communication, where a student requests help from the lecturer who provides an answer to the asked question. In school there are also a few instances where students help each other when doing an activity which involves the whole class. Usually this kind of code switching occurs when the lecturer asks a student something in English that (s)he cannot understand and the other students help by translating the lecturer’s question. Cook (2001) sees this as a positive way of using the mother tongue in the classroom. By translating the lecturer’s words the students ensure that the weaker student knows what is happening. It does not always have to be the lecturer who translates the question. Moreover, by letting the students help each other lecturer creates a more natural communication situation. She can be consulted when a problem occurs, but she encourages the students to solve the problem among them, the one who knows what the lecturer is saying can help the weaker student(s) by explaining the lecturer’s words (Yletyinen, 2004:61).

2.3.2.4 Self-corrections

Apart from being used to help out others, students also use code switching in self-corrections. This function of code switching is quite common. The students employ self-correction in their utterance by beginning it in English but inserting one word or a couple of words in Indonesian in the middle of the utterance. When self-correction occurs, a student is usually producing an answer to the lecturer’s question and when (s)he realizes that a mistake has occurred in the answer (s)he corrects it by inserting word and then continuing the answer, but now with a more correct answer (Yletyinen, 2004:63).

2.3.2.5 Moving from one activity to another

The previous three functions of code switching (requesting help, students helping each other and self-corrections) have been employed by the students. However, as the function of explanation showed, lecturers also use code switching. The function of moving from one activity to another is employed by lecturers and students use code switching to mark a shift in the lesson. This marking of activity shift in the lesson happens when the lecturers move from one topic to another: from discussing an exercise to teaching grammar and vice versa, from learning grammar  to  looking  at  a  chapter  and  from  giving  instructions  to  doing a listening comprehension (Yletyinen, 2004:66). 

2.3.2.6 Code switching in clearing misunderstandings

There are a couple of instances where a misunderstanding occurs during  a  lesson  and  in  order  to  clear  it  code switching  is  employed.  This function of code switching is visible in the schools. When there is a need to clear a misunderstanding it is usually the case that the lecturers has misunderstood something and the student correct them using the embedded language (Yletyinen, 2004:70). 

2.3.2.7 Not knowing the English counterpart

There are a couple of instances when a student or a lecturer inserts a word   into an otherwise English utterance. Sometimes   this code switching is triggered by the fact that the English counterpart is unknown at that moment (Yletyinen, 2004:72).

2.3.2.8 Checking for understanding

Part of the foreign language learning process is to learn new words and expressions. The material is there to help this process; the students have chapters to read and exercises to do which teaches them new vocabulary. In the teaching situation, when the class is,  for example, doing an activity in  English, the lecturer should make sure that all students know all the words in that activity. If there are new words or expressions, the lecturer can ask the students what the words mean in the first language.  In  others  words,  the  lecturer  does  not  have  to  translate everything, by asking the students she lets them participate more actively to the lesson. Code switching occurs when the word or expression is translated in first language, or when the lecturer asks about the new vocabulary in first language (Yletyinen, 2004: 75).

2.3.2.9 Unofficial interactions

According to Canagarajah (in (Yletyinen, 2004:80), interactions that are not demanded by the lesson are called unofficial interactions. In his study he discovered that it was the mother tongue that was used in these instances. He describes the unofficial interactions as cases of student to student interaction, for example in group activities where procedural matters are discussed. However, sometimes the lecturer employs this function as well, for instance, to discuss extra- pedagogical matters such as happenings in the town. All in all, this function of code switching occurs when either the students or the lecturer are talking about issues not related strictly to the lesson (Yletyinen, 2004:80).

2.3.2.10 Students’ comments

The previous category of functions dealt with unofficial communication occurring  at  the  same  time  as  the  lesson  is  going  on.  The  unofficial interaction that take place has little to do with the present lesson content. The function   of   students’   comments   differs   from   the   function   of   unofficial interactions in that the comments made by students are linked with the situation at hand. The students mainly comment on the exercises or activities, or events relating to the exercise (Yletyinen, 2004:83).

2.3.2.11 Student initiation

Those switches motivated by, for instance,  a wish to  request help or  to clear a misunderstanding.  In this section, the case  of  code switching  where  a student’s code switching from first language  to  English  is  followed  by the lecturer’s  switch  to  English as well (Yletyinen, 2004:87).

2.3.2.12 Lecturer admonitions

The name for this category comes from Canagarajah’s (1995) article dealing with the functions  of  code switching.  Canagarajah  (1995:  183)  explains  this function as one that is used when the lecturer is disappointed with the students, and (s)he uses the mother tongue to express this anger or frustration (Yletyinen, 2004:89).

2.3.2.13 Grammar translation

As was pointed out above, in both classrooms one theme in the lessons is grammar. However, as the target is to teach English grammar, there is bound to be a lot of code switching, which is conscious. There are two functions of code switching when teaching grammar: grammar translation and grammar explanation. It will separate these functions from the function of explanation and treat them as individual  functions.  The  reasons  for  this  are,  firstly,  that  the  language  is treated differently in explanations and when teaching grammar. When explanation occurs the language is a means for communication. When teaching grammar, though, the language is treated as an object. In other words, code switching in explanation takes place because someone does not understand the lecturer’s English words; in contrast, when the lecturer is teaching grammar in first language, code switching occurs because of a necessity as the examples are in English. Secondly, in grammar translation and grammar explanation both the lecturer and the students employ code switching, whereas in explanation only the lecturer employs code switching (Yletyinen, 2004:91).

2.3.2.14 Grammar explanation

Grammar explanation differs from grammar translation in that during teaching grammar  in  first language  the  lecturer  uses  English  words  which  she  does  not translate. The base language is first language, English is used because the lecturer is teaching English grammar; thus the examples are in English (Yletyinen, 2004:94).

2.3.2.15 Lapses

Some of the instances of code switching cannot be categorized into functions, instead, they could be simply treated as lapses. Lapses are instances where the lecturer is speaking first language for example but says a word or a couple words in English. These English words are spoken almost accidentally, since they are  not  required.  Lapses  occur  in  the lecturers’ speech since they are used to speaking may slip an English word sometimes (Yletyinen, 2004:95).

2.3.3       Reasons motivate lecturers code switching in EFL classroom

 In the discourse of class, code switching considered by many to be both an asset and a valuable addition. This may seem surprising given that so often, the modem approach to language teaching and learning, teachers try to create a second language classes mirror from the outside world. Code switching by each student is also considered as proof that they do not think as much as possible in the L2. Again, the idea is that L2 learners and users had to press their bilingualism. Cohen (1998) agree that the language of thought for all but the most advanced L2 learner loser is definitely L1. But many teachers attached to the belief that, given the right conditions, students in the classroom they can 'think' in L2 when performing tasks. This belief is not exclusive to monolingual teachers. So, for what purpose do codeswitch language teacher? Teachers, in the context of learning, such Macaro (2000) explains that the area where the teacher uses L1 are: (1) establish a personal relationship with learners (pastoral role that teachers take requires a high level of sophistication of discourse); (2) provide complex procedural instructions for carrying an activity; (3) control the behavior of pupils; (4). translating and checking comprehension in order to accelerate because of the pressure of time (eg, test); (5). teach grammar explicitly.

2.5 Assumptions

The assumption of this study is: Lecturers and students use code switching in their classroom conversation when teaching learning proses in IPDN NTB Campus in academic year 2014/2015.