Hegemony and language dynamics (Malay versus Kelantanese)

Source: https://pdfs.semanticscholar.org/740a/69c31f4b34a0ffe44bcc85a25a471f4ed044.pdf


Language-dialect code-switching: Kelantanese in a multilingual context


Malaysia being a multiracial and multicultural society, code-switching is an important characteristic of the overall dynamic picture of linguistic in- teraction amongst Malaysians. It is common for Malaysians to code-switch in a single situation, even within a sentence. However, most research in code-switching addresses only language code-switching. This pa- per presents an analysis and description of languagedialect code-switch- ing (specifically Standard Malay/Kelantanese Malay code-switching) be- haviour of Malay students, aged 2025, who are at present following an undergraduate programme at the University of Malaya. The central con- cern of this paper is to determine the structural as well as the functional constraints of Malay/Kelantanese code-switching in a new ethnolinguistic environment (cf. Myers-Scotton 1993). The data used in this study comes from two sources: (1) recordings of spontaneous conversations collected over a period of 3 months, and (2) interviews conducted with the infor- mants, the main purpose of which is to determine their reasons for code- switching. An attempt will be made to relate the observations and conclu- sions drawn from the examination of data to the question of language maintenance of a dialect in a multilingual community.


The present study focuses on the interpersonal aspect in interactions and looks at the code-switching (i.e. the use of two or more languages or dialects in the same conversation or utterance) behaviour of Kelantanese Malay undergraduates. It attempts to present some main findings of an analysis of code-switching carried out at the University of Malaya among Kelantanese Malay undergraduates. Based on the assumptions that code-switching data provides both social and linguistic information, such an undertaking is deemed viable.

Multilingua 22 (2003), 2140 01678507/2003/0220021Walter de Gruyter

22 Zuraidah Mohd. Don

Since Malaysia is a multilingual and multi-dialectal country, it is inevi- table that Malaysian speakers are constantly faced with the options of making meaningful language choices when interacting with people of different races or people from different dialect areas. It is assumed that these speakers have alternative linguistic means available to them when ‘constructing’ their social and cultural experiences in the world. In multi- lingual settings code-switching is a central part of bilingual/multilingual discourse (Zuraidah 2003). The choices that speakers make usually carry a message beyond the referential meaning of the utterance (Myers-Scot- ton 1994).

Malaysia is demographically multi-dialectal but functionally monolin- gual/mono-dialectal country. Wherever a language is spoken as a native language, as in the case of Malay in Malaysia, dialect differences emerge as they do in any other language spoken over a wide area. The majority of the Malays learn and use one dialect at home and proceed through an educational system that requires them to adopt a different dialect, i. e. the standard variety of Malay. When children from different dialect backgrounds attend school, the dialect that they learn at home will be in contact with the dominant language, i. e. standard Malay, a variety accepted by the speech community as the prestige dialect used in formal situations and public domains. Standard Malay is not the language vari- ety of any specific region of the country although it is said to be based on the Johore-Riau dialect of Malay.

The Kelantan dialect

The Kelantan dialect (henceforth, KD) is spoken by speakers who came from Kelantan, a state on the east coast of Malaysia sharing its borders with Trengganu, Pahang, Perak and Thailand. It is also spoken by peo- ple living in areas at the borders of Kelantan/Trengganu, Kelantan/Pa- hang and a few areas in Southern Thailand, among which are the Golok River, Narathiwat, Yala and Patani.

As regards its functional distribution, KD is the everyday spoken language of the community in Kelantan, used at home and among family and friends as well as in other speech situations which are private and informal. However, in Kelantan itself the dialect is even used as the medium of communication in public and formal settings, such as the mosque, government and private institutions. It would not be too far- fetched to say that in Kelantan the local dialect is the language variety in which all social interactions except for those in the written medium are carried out.

The strong attachment of the Kelantanese to their dialect is clearly reflected in the phrase that they often use to refer to the act of speaking

Language-dialect code-switching 23

the standard variety, i. e. kecek luwa, which means ‘speaking an outside language’. KD thus serves as one of the most powerful markers, if not the most powerful marker, of local rather than national identity. It is clear that the Kelantanese regard their dialect as a fundamental part of their regional culture and a symbol of group membership and loyalty. The fact that they refer to the non-Kelantanese as oghei luwa (‘outside people’) and to the Kelantanese as oghei kito (‘our people’) clearly shows how strongly they feel about their regional identity and their dialect that symbolises that identity. In short, the Kelantanese regard their dialect as a symbol of regional identity, a social cultural symbol of their state and a common means of communication (lingua franca) among the Kelan- tanese (Zuraidah 2003).

KD has its own set of features of pronunciation, vocabulary and syn- tax, which differ from standard Malay (henceforth SM). It has very dis- tinctive local colourings and can at times be unintelligible to speakers of other dialects. Nevertheless, these differences do not represent corrup- tions from the standard variety. Rather, they are the result of normal linguistic divergence. Thus, when Kelantanese students attend school, they will be following sets of linguistic rules which are rather different from those considered acceptable by the school and society. The linguis- tic ‘distance’ between the standard oral variety and the Kelantan dialect is great enough to cause considerable difficulty for the non-Kelantanese to understand a Kelantanese speaking their local dialect. It is this diver- gence, too, that interferes with the ability of the Kelantanese to speak or write in standard Malay (cf. Farid M. Onn and Ajib Che Kob 1993). This is especially so for those who have been brought up in Kelantan and have gone through the formal education there.

Farid M. Onn and Ajib Che Kob (1993) reported that students from the states whose dialect differed greatly from the standard Malay variety (e. g. Kelantan, Trengganu, Negri Sembilan and Kedah) recorded relatively poorer performance in the Malay paper in the Malaysian Education Certicate Examination than those who spoke SM, especially students from Johore, Selangor and the Federal Territory. They concluded that one of the main reasons that contributed to their relatively poor performance was dialect interference.

In a study of the students’ and teachers’ perception of the influence of KD on SM, it was discovered that 57.3 percent of the students from selected schools in Kelantan either used KD or a mixed code when they interacted with their teachers in the classroom (Farid M. Onn and Ajib Che Kob 1993). What was worrying was that only about 77.4 percent of the Malay language teachers used the standard variety. The rest code- switched and code-mixed.

In the same study, it was also reported that the students identified the following aspects of the language as problematic: syntax (43.96 percent),

24 Zuraidah Mohd. Don

spelling (41.67 percent), pronunciation (37.42 percent), vocabulary (31.28 percent) and affixation (23.67 percent). The teachers concurred with the students in terms of the aspects of the language that posed problems for the students, although the percentage given varied: pronun- ciation (73.52 percent), syntax (53.58 percent), vocabulary and spelling (46.42 percent) and affixation (27.41).

Consequently, when these students enter university, they will have problems communicating with their counterparts from the other states, the majority of whom speak the informal variety of SM or a dialect that differs slightly from it. Given the close proximity of the dialect to the dominant language, i. e. the standard variety of Malay and the presence of many bilingual and bidialectal speakers who use the dominant language, i. e. Malay, as is often the case in an institution of higher learning, dialect thus becomes a serious issue.

Code-switching, a language contact phenomenon

Although the research on code-switching varies in terms of research goals and issues, it shares one common feature in that it deals with language contact which frequently ‘involves face-to-face interactions among groups of speakers, at least some of whom use more than one language in a particular geographical locality’ (Thomason 2001: 3). In this paper, code-switching is defined as the use of more than one code in the course of a single discourse in a multilingual setting (see, e.g., Eastman 1992; Heller 1988). Code as used in the present study refers to two different dialects of the same language, i. e. the standard dialect and its non-standard oral dialect variety. In some places code-switching is viewed as deviant or exceptional behaviour, but in many bilingual/multi- lingual communities it is a common occurrence and should be accepted as normative linguistic behaviour (Duran 1994). In fact, in certain socie- ties like Bukavu, code-switching is the norm (Goyvaerts 1988).

One of the concerns of the present study is the social meaning of code-switching, which is dealt with at great length and in great detail by researchers like Gumperz (1982), Blom and Gumperz (1971) and Heller (1988). From their studies and those of others, we know that code- switching is not just a kind of dysfluent speech, rather the switching from one code to another is consistent both linguistically and sociolin- guistically.

In his attempt to link code-switching to group identity, Gumperz (1982: 66) introduces the notion of ‘we-code’ and ‘they-code’. According to him:

The tendency is for the ethnically specific, minority language to be regarded as the ‘we-code’ and become associated with in-group and

Language-dialect code-switching 25

informal activities, and for the majority language to serve as ‘they- code’ associated with the formal, stiffer and less formal outgroup rela- tions.

However, Auer (1991) finds the association between particular codes and identity rather difficult to maintain as there is no one-to-one mapping between the former and the latter; ‘rather such relationships are them- selves negotiated and constructed in the interaction, drawing on cultural resources located both inside and outside the interaction itself’ (Sebba and Wootton 1998: 284). For example, in the extracts of data examined while KD is the ‘we-code’ among the Kelantanese, reflecting their ‘we- ness’, the informal variety of SM is not in the strictest sense the ‘they- code’ as defined by Gumperz. Rather, it is the unmarked language of everyday interaction among Malaysian speakers in an institution of higher learning.

By examining code-switching using the cooperative maxim, Heller (1988) postulates that social relationships among interactants are estab- lished by means of a joint association between a code choice and social context. Relevant to this is the question of access to and availability of linguistic resources. As regards accessibility and availability of code, Blommaert (1992: 66) postulates that ‘a resource may be available in a society, but not accessible to all members of that society’. Being a multi- cultural community, it is natural for Malaysia to have the availability of several languages and dialects, but access to them is restricted to certain groups. Thus, when a speaker switches from one code to another, s/he automatically excludes those who do not know the code or have the same mastery of the code. This exclusion of the non-language user is clearly illustrated in the study conducted by David (2001) on the Malay- sian Sindhi community, where it is shown that the use of English by the younger members of the community excludes the older monolingual members. Although the switch to the dialect by the Kelantanese in the speech exchanges examined was not meant to exclude the other co-in- teractants, the fact that the non-Kelantanese subjects could not under- stand or speak the dialect means that they were excluded from the con- versation. In this regard, the dialect thus becomes an exclusive code ac- cessible to those who speak it.

Code-switching is highly sensitive to the social factors of language such as the locale, the interlocutor or the role relationship of the interac- tants and the discourse topic (cf. Ervin-Tripp 1964; Sankoff 1971; Giles and Powesland 1997). In describing situational code-switching, Gumperz (1982: 6061) reiterates that ‘[d]istinct varieties are employed in certain settings (such as home, school, work) that are associated with separated bounded kinds of activities (public speaking, formal negotiations, special

26 Zuraidah Mohd. Don

ceremonials, verbal games, etc.) or spoken with different categories of speakers (friends, family members, strangers, social inferiors, govern- ment officials, etc.)’. This suggests that speakers always have a reason for choosing one code instead of another, and this switching of code can be seen as a resource for indexing situationally salient aspects of context in their attempts to achieve interactional ends (Heller 1988). In short, code-switching is a way in which interactants handle the available com- municative/linguistic resources.

Some descriptions of code-switching imply a certain level of compe- tence in the languages used because speakers who code-switch use two languages/dialects simultaneously or interchangeably (Valdes-Fallis 1977). Gumperz (1982: 60), for example, regards code-switching as ex- changes that form a single unitary interactional whole.

Speakers communicate fluently, maintaining an even flow of talk. No hesitation, no pauses, changes in rhythm, pitch level or intonation contour marks the shift in code. There is nothing in the exchange as a whole to indicate that speakers don’t understand each other. Apart from the alternation itself, the passages have all the earmarks of ordi- nary conversation in a single language.

However, there are situations in which speakers may begin to speak in a language in order to accommodate or speak in a language that enjoys a higher social prestige, but because they are not fully competent, they are not able to sustain their discourse in that variety all the time. This could explain why a few of our informants were not at all at ease when speaking the standard variety and kept switching to their dialect when they could not express what they wanted to say in standard Malay.

The present study

Much research on code-switching in Malaysia has focused on languagelanguage code-switching (cf. David and Naji 2000 for Tamil; David 2001 for Sindhi; David and Nambiar 2002 for Malayalees) while work on dialectdialect code-switching is non-existent by comparison. The pre- sent study attempts to fill the gap in the existing literature by making a modest contribution to the subject of code-switching by Kelantanese Malays. The speech of educated Kelantanese Malays can be looked at in terms of two polar dialects a standard variety of Malay and a non- standard variety, i. e. the Kelantan dialect.

The study is part of the project (Zuraidah 2003) which analyses the effects of ethnolinguistic vitality on the speaking behaviour of Kelan- tanese undergraduates at the University of Malaya and is based on a

Language-dialect code-switching 27

corpus of spontaneous speech used by the subjects in both in-group and out-group interactions. It examines the social function of code-switching with a focus on the switching of codes as both a form of accommodation and alienation. What is interesting is that while the switching from KD to SM was regarded as an attempt to accommodate on the part of the Kelantanese, the switch to their own shared code KD can be looked upon as alienation in the sense that it excludes non-Kelantanese participation in the conversations.

Code-switching among the Kelantanese Malays presupposes an awareness of one’s own linguistic identity and at the same time serves to offer another language, a more neutral lingua franca, in outgroup communication to indicate a spirit of willingness to accommodate and to respect. The study will also present in brief a structural description of the Malay Kelantan code-switching looking at some linguistic variation (encompassing phonological, lexical and grammatical variants) found in the data. The examination of the discourse is interesting since the speak- ers have a choice of a whole range of intermediary strategies, which include the modification of either code and the relative use of both.


The hypothesis presented in the present study is based on the available information on the Kelantanese and their close association with their dialect and the norms which are expected to govern the situation being studied.

The first variable that may determine code selection is the co-interac- tants. It is hypothesised that the Kelantanese undergraduates would use their own dialect in in-group interaction and would switch to the stan- dard variety or an approximation of SM when interacting with non- Kelantanese. What is expected is that, when a group is made up of both Kelantanese and non-Kelantanese, the Kelantanese will use KD or a mixed code when they address their Kelantanese friends and speak the informal variety of SM when they address their non-Kelantanese friends or both.

The second variable is the place in which the interaction takes place. This relates to the question of whether setting has an effect on language choice. It is postulated that the Kelantanese students will use KD in class only when they are having a discussion with their fellow Kelantanese but will switch codes when they speak to their non-Kelantanese peers. However, when they present their discussion in class they will speak the standard variety. Due to the lack of competence in the standard variety (Zuraidah 2003), it is expected that some form of interference from the local dialect will occur particularly in terms of pronunciation.

28 Zuraidah Mohd. DonThe data

The first type of data comprises approximately 5 hours of recorded natu- ral conversations of selected first year students of the University of Ma- laya. The language used in interactions between Kelantanese is catego- rised as in-group language and that between Kelantanese and non-Kel- antanese is categorised as out-group language. A working hypothesis that guides the data collection efforts and is supported by the analysis below is that speakers of KD consider SM the appropriate code for use with out-group interactions and consider their own dialect appropriate for in-group interactions. The second type of data comprises interviews conducted with five subjects, approximately three weeks after the taping of their conversations. Among the information provided by the data includes the subjects’ linguistic preferences and competence, the reasons for their choice of code and their feelings about their interlocutors’choice of code.

Only one student in each group was informed about this research weeks before the taping. These students were employed as research assis- tants to collect the data. The other informants were informed of this after the taping, and their permission was obtained before their conver- sation was used as part of the data. The data used in this study are principally in the form of cassette recordings, supplemented by observa- tional notes made by the research assistants. All the recordings were made in a naturalistic setting.

The area from which this data was drawn is the University of Malaya, a university situated in the Klang Valley, with a multiracial and multicul- tural population. Being a multicultural community, the University pro- vides a particularly salient case of the linguistic heterogeneity and lin- guistic alternatives available to speakers not only of four easily distin- guishable languages (Malay, Chinese, Indian and English), but also dia- lectal differences especially within the Malay language.

The data extracts are numbered sequentially, for example, Extract 1, Extract 2 and so forth.1 The interlocutors in the discourse are identified by a letter (i. e. A, B, or C) and the number placed immediately before each letter (i. e. 01, 02, 03, etc.) indicates a sequential arrangement. Each text is provided with a translation given in brackets at the end of each turn. It is worth pointing out at this juncture that the translation is not a literal translation of the dialect but an approximation of the original with no change in meaning. In the speech exchanges cited below, in- stances of KD are in italics.

Analysis of data

The Kelantanese subjects exhibit some degree of code-switching in their linguistic performance, and there is more data than could possibly be

  1. 01  A:
  2. 02  B:
  3. 03  A:
  4. 04  C:
  5. 05  A:
  6. 06  B:
  7. 07  A:

Eh, mu gi mano lamo tok jupo?
(‘Where did you go? I hadn’t seen you for a long time.’)
Aku balik kela[te.
(‘I went back to Kelantan.’)
[APade pung tok nampok bate idong. (both laugh)
(‘No wonder I didn’t see you at all.’)
C arrived and immediately joined in the conversation
(0.2) Ah awal kau dua oghang sampai. Selalu lambat. Kenapa rajin ni?
(‘You too are early today. Usually you are late. How come you’re hardworking’)
Aku ingat nak buat kerja sikit. Tak jadilah dia ni ada sini. Kacau [je.
(‘I thought of doing some work. But because of him I couldn’t do anything.’)
[AEh jange saloh aku. Mu ye malah.
(‘Don’t blame me. You’re the one who’s lazy.’)
(.) Yolah aku tahu aku tok rajin maceimu, betul tak? (laugh) (1.0)
Kalau aku rajin aku tak ada sini dah.

Language-dialect code-switching 29

covered in the space allotted to this paper. On the other hand, the non- Kelantanese only used the informal variety of SM, and such behaviour is expected of them since none of them were able to converse in KD nor understand it that well.

The following is an example extracted from a conversation between a group of three male friends gathered in the classroom while waiting for their English teacher. The conversation begins with two Kelantanese speakers (A and B) conversing with each other and a friend (Speaker C), who is from a different state and cannot speak or understand the dialect, later joins them. As expected, the initial speech exchange between A and B is conducted entirely in their shared mother tongue, i. e. KD. Given the fact that they regard their dialect as closely related to their identity as people of Kelantanese descent, their code choice is expected. However, when their friend (i. e. C) appears, the code is no longer the same. They accommodate to the language of C, which is the informal variety of SM. While A seems to have no difficulty in switching from one dialect to another, B, who is not a fluent speaker of Malay, seems to be en- countering some difficulty in speaking entirely in the standard variety (see e. g. line 08). Let us now examine the exchange:

Extract 1 (KD in italics; translation in brackets; selected segments un- derlined):


30 Zuraidah Mohd. Don

08 B:

09 C:

10 A:

(‘Yes, I know I’m not as hardworking as you, am I?
If I’m hardworking I wouldn’t be here.’)
Meme aku rajeng. Eh, engkau ada bawok buku aku nok pinjan?
(‘It’s true that I’m hardworking. (looking at B) Have you got the book that I want to borrow?’)
(laugh) Apa bawok bawok!. Apalah kau ni cakap BM pun tak betul.
‘Bawaklah’ bukan ‘bawok’. (laugh) Kau kena hantar cepat tau.
(‘What “bawok” “bawok”? You can’t even speak BM prop- erly.
It’s “bawak” not “bawok”. You must return soonest possible you know.’)
(1.5) erm janji
(‘erm promise’)


A and B are talking to each other entirely in KD (see lines 0103) before C joins in the conversation addressing both of them using SM. In response to A’s banter (underlined in line 05) that he is not able to do much work because of B’s presence, B (in line 06) responds using KDEh jange saloh aku. Mu ye malah (‘Don’t blame me. You’re the one who’s lazy’). When asked why he chose KD and not SM since A spoke to him in SM, B said the response was meant for A in reply to what he said earlier about him, and since A was Kelantanese it was natural that he used KD. In line 07 (Yolah aku tahu aku tok rajin maceimu, betul tak?) A complies with B’s choice of code and switches to KD, thus invoking regional identity and in-group membership. However, he switches to SM when he utters the tag ‘betul tak?’ (underlined in line 07), which accord- ing to him was meant for C. In his own words A described his code choice as follows: Saya guna bahasa Melayu sebab saya nak C setuju dengan saya dan saya sengaja buat begitu supaya B tahu bahawa saya tujukan kepada C bukan B (‘I used SM because I wanted C to agree with me and I did that on purpose so that B knew that I was addressing C, not him’).

Likewise when B responds to A’s banter (Meme aku rajeng ‘It’s true that I’m hardworking’ in line 08) he chooses to do it in KD and switches to SM when he directs his next utterance to C ‘Eh, engkau ada bawokbuku aku nok pinjan’, ‘Have you got the book that I want to borrow?’(underlined in line 08). Because of his lack of fluency in SM, the interfer- ence of his mother tongue on SM is unavoidable. This is manifested in his pronunciation. Notice how he mispronounced some words: ‘bawa’pronounced as bawok, ‘pinjam’ as pinje.


Language-dialect code-switching 31

The majority of speakers of KD, especially those who seldom interact with speakers of other dialects, find it difficult to maintain their pronun- ciation in SM. It is often the case that the less proficient Kelantanese speakers of SM tend to mispronounce words that end with nasals ‘/m/’,‘/n/’, ‘/n/’ and the vowel ‘/a/’. Speaking ‘pure’ standard Malay seems to require special attention and effort. This is not surprising because unlike the standard variety that has only 31 phonemes (i. e. 6 vowels and 25 consonants), KD has 35 (i. e. 15 vowels and 20 consonants). As discussed earlier, although B begins his utterance in SM, he is not able to sustain it, and as a result he switches to KD. The effect of language competence on the choice of code is noted by Gardner-Chloros (1997), in which it is reported that speakers who converge using French, a variety with more social prestige, switch to Alsatian because they are ill at ease in French.

Both A and B knew that C could not understand or speak their dialect so they quickly switched codes when they addressed C. When asked why he switched codes, A’s reply was that switching aided comprehensibility (Kalau kita cakap Melayu oghang paham. Cakap Kelantan payah orang lain nak paham. Semua orang guna bahasa melayu di universiti; ‘If we speak Malay people will understand us. Others do not really understand the Kelantan dialect. Everybody uses BM in the university.’) There was a very strong awareness that they were obliged to switch to the standard variety in order to be understood. And when asked why they used the dialect when speaking to each other, both A and B said that KD was their mother tongue and they felt comfortable using it. They also said that when they spoke to each other in their shared code, they felt there was a strong regional bond between them. They were aware of the dialect sociolinguistic meanings for local identification and knew that they had to use SM for outgroup interaction and deployed the codes available to them strategically in their talk.

The following example illustrates how new information about the re- gional identity resulted in a redefinition of the exchange. The partici- pants in the following exchange are three female participants, two of whom (i.e. B and C) are Kelantanese. The two at first speak to each other in SM, but they immediately code-switch using KD (see lines 05, 07and 09) when they know of their shared regional membership.

Extract 2 (KD in italics; translation in brackets; selected segments un- derlined):

  1. 01  A: Kawan aku Siti. (‘My friend Siti.’)
  2. 02  B: Hai, Aini. Dari sastera ke?
    (‘Hei, Aini. Are you from the Arts?’)

32 Zuraidah Mohd. Don

  1. 03  C:
  2. 04  A:
  3. 05  B:
  4. 06  A:
  5. 07  B:
  6. 08  C:
  7. 09  B:
  8. 10  A:
  9. 11  B:

Tak, APM. Tapi kena ambil Bahasa Inggeris di Fakulti Sast- era.
(‘No, APM. But I’ve got to take English at the Arts Faculty.’) Siti ni dari Kelantan, orang kampong engkau.

(‘Siti is from Kelantan; she’s from your hometown.’)

Oh Mano?

(‘Oh! Which part?’)
Ah dah mulalah tu, kecek Kelatei
(‘Oh dear, now she starts talking in her Kelantan dialect.’ ) (0.1) Apa salahnya sekali sekala. Dari mano? Koto Baghu? (‘What’s wrong with that; it’s once in a while. From where? Kota Bharu?’)
Dok eh Pase Mah. You dari mano?
(‘No, Pasir Mas. Where are you from?’)
Koto Baghu. Baghu ni balek panah sunggoh. Lasong tok ujei. Kita ni dah lambat. Jom pergi nanti malu nak masuk [kelas. (‘Kota Bharu. When I went back the weather was so hot. There’s no rain at all. We’re late already. Come let’s go. It’ll be embarrassing to enter the class’)
[Yalah kalau jumpa geng Kelantan tak cakap lain dah. (‘Yeah, when you meet your fellow Kelantanese you won’t speak any other language.’)
(0.15) Sorrylah erm dah tabiat, buke gitu heh? Dio deki la tu. Bukan boleh aku kecek Kelante (All laugh)
(‘Sorry. It’s like a habit, isn’t it right? (turn to C) He’s envious of us. I can’t even speak the Kelantan dialect.’)


The above extract, which provides an interesting example of the sudden shift of code, demonstrates how code is used to construct a relationship between speakers. It is a deliberate choice made by the co-interactants to convey the fact that they are from the same state and reinforce re- gional bonds. The fact that A makes some comments when there is a switch of code suggests that switching to another code conveys a mes- sage beyond the referential content (see Heller 1988).

Like the conversation in Extract 1, the above exchange is mostly made up of intersentential code-switching (represented in italics in turns 05, 07, 08 and 09). When B knows that C is Kelantanese, he immediately switches to the dialect Oh Mano? (turn 05) asking C which part of Kelan- tan she is from. The switch to KD is not well received by A, and this can be inferred from her comment about the switch ‘Ah dah mulalah tu,kecek kelatei’ ‘Oh dear, now she starts talking in her Kelantan dialect’(underlined in turn 06), which implies her annoyance. B accommodates by responding to A’s comment in SM ‘Apa salahnya sekali sekala’


Language-dialect code-switching 33

‘What’s wrong with that, it’s once in a while’ (underlined in turn 07). When addressing C, B again switches to the preferred code, KD Dari mano? Kota Baghu? ‘From where? Kota Bharu’ (turn 07). C replied in KD Dok eh Pase Mah. You dari mano? (turn 08), and thereon the conver- sation between B and C is carried out in KD. When C responds in KD, rapport is established between the two of them.

There are only two occurrences of intrasentential code-switching in Extract 2. The first uttered by A, i. e. ‘kecek kelate’ (‘speaking the Kelan- tan dialect’ in turn 06) is an expression most often used by both the non- Kelantanese and Kelantanese alike to refer to someone speaking KD. However, each of the expressions fulfills a different function. While for the Kelantanese it signifies in-group solidarity and regional identifica- tion, for the non-Kelantanese it emphasises the Kelantanese ‘they-ness’and their ‘other-ness’. According to Blommaert (1992), this group iden- tity which is effected through the dialect makes use of an exclusive re- source alienating those who do not speak the code. Thus, when the Kel- antanese switch to their dialect, their code choice inevitably excludes those who do not know the dialect. This was later confirmed by A, who said he felt left out when his co-interactants (i. e. B and C) switched to KD. A described this feeling of alienation as follows: Saya rasa seperti dipinggirkan. Mereka seolah-olah tak ingin saya paham apa yang mereka cakap dan saya rasa seperti saya tidak dingini disit ‘I feel alienated. It seemed to me as if they didn’t want me to understand what they’re talk- ing about and I felt really unwanted there’. The problem that most non- Kelantanese face is that they find it difficult to understand the dialect, and this prevents them from participating in the conversation.

When asked why he pronounced the phrase kecek kelate in KD as earlier on in the interview, he admitted to not being able to speak KD, A said, ‘Saya sengaja sebab saya mahu B tahu yang saya kurang senang dengan dialek Kelantan yang dia selalu guna bila bercakap dengan or- ang Kelantan. Bila saya kata kecek kelate saya ingin mereka sedar yang saya rasa seperti orang asing’ (‘I did it purposely because I wanted B to know that I was uncomfortable when he switched to the dialect when talking with his fellow Kelantanese. When I used ‘kecek kelate’ I wanted them to realise that I felt like an outsider.’).

The second, an interesting intrasentential code-switching, forms part of a tag question (represented in italics in turn 11), ‘Sorrylah, dah tabiat,buke gitu heh?’ (‘It’s like a habit, isn’t it?’). Notice that this tag question is made up two codes: the first part is uttered in English and SM and the second in KD. B responds to A’s comments about his code-switching by saying that speaking KD has become a habit. The tag which follows the statement merely asks for routine confirmation of what B already believes. What is interesting here is that B begins his utterance with an

34 Zuraidah Mohd. Don

apology expressed in English (in turn 11), and this is followed by an explanation uttered in SM following which is the tag in KD.

In the interview, B said that he apologised in English because he re- garded it as a small offence; to apologise in Malay would sound too serious. The apology clearly indicates that B was very much aware that he was being ‘uncooperative’ in the sense of speaking a code that was quite unintelligible to A, thus going against the rule of polite conversa- tion (see, e.g., Heller 1988). Despite that, he still went on using the dialect to make comments about what A had said previously. When asked why he continued speaking in KD when he knew that A did not like it that much, he replied: ‘Dah biasa bercakap sesama sendiri dalam bahasa Kelantan susah nak tukar secara otomatik dalam BM. Lagipun orang Kelantan memang cakap dialek Kelantan sesama sendiri. Sudah jadi sebati dengan kami.’ (‘I’m so used to speaking to my fellow Kelantanese in the dialect. It’s very difficult to code-switch to BM automatically. Anyway we always speak to each other in the Kelantan dialect. It comes naturally to us.’)

The extract below is different from Extracts 1 and 2 in that it is a more formal and less personal activity. The participants were engaged in a group discussion discussing the topic assigned to them in a seminar room. This speech exchange is made up of a different group of students, i. e. two males and two females. Two of the participants who are Kelan- tanese (i. e. C and D) know each other well and the other two, the non- Kelantanese (i. e. A and B), are acquaintances. The following is a frag- ment of their discussion.

Extract 3 (KD in italics; translation in brackets; selected segments un- derlined):

01 A: 02 B:

03 C:

04 D:

Macam mana ni? Cuba engkau baca soalan tu.
(‘How are we going to do it? Could you read the question?’) (.) Nyatakan sumbangan Ferdinand de Saussure terhadap perkembangan linguistik abad [ke 20.
(‘Discuss the contributions made by Ferdinand de Saussure towards the development of linguistics in the 20th century.’) [AGapo? Eh sorry terlajak kecek Kelante. Apa dia aku tak paham? (All laugh)
(‘What? Eh sorry for speaking Kelantan. A slip. Say it again. I didn’t get you.’)
Gapo? (laugh) Mu selalu lagu tu. Kelate pekak. Kalu mu bente oghe semuo tok pahe, betul tak? Nasib baik kita tak pileh dia. (laugh).
(‘What? You’re always like that. Strong Kelantanese accent.


05 A:

06 B: 07 D:

08 C:

Should you present nobody would understand, right? It’s for- tunate that you’re not chosen’)
(1.0) Eh engkau oghang ni teruklah. Ini perbincangan kelas. Cakaplah bahasa standard. Cepatlah. Aku yang kena ben- tang. Tinggal 10 minit je lagi.

(‘You two are really incorrigible. This is a classroom discus- sion. Speak the standard language. Quick. I’ve got to present. We’ve got ten more minutes.’)
Cuba kita lihat nota dia ba[gi.

(‘Come let’s refer to the notes that she gave us.’)
[A Ha ni kajian bahasa dari segi sinkronik dan diakronik. Ah cepak tulihlah.
(‘Ha here. Study language from diachronic and synchronic perspectives. Hah write quickly.’)
(1.5) Eh malah aku. Tok pahei. Cuba tengok dalam nota sink- ronik dan ah diakronik?
(‘I don’t think I want to do it. I’m lost. Look at the note on synchronic and ah diachronic.’)

Language-dialect code-switching 35

C’s utterance Gapo? (‘What?’) (turn 03) was a spontaneous response to what was said earlier. This assumption is supported by his subsequent utterance, which suggests that the switch is an automatic reaction and unintentional. As postulated by Stolen (1992), code-switching may be triggered by some words in the discourse and this triggering takes place below speakers’ conscious awareness (Clyne 1980). Gapo (‘What?’) is a request for the addressee to repeat what he had said previously. It sug- gests that A did not understand what B had said. Knowing that B is not Kelantanese, he quickly apologised for the slip and reformulated it using SM. Gapo can only be understood by the Kelantanese or those who speak KD as there is no such lexical item or structure in SM. Here, B’s reformulation aims at correcting the use of the ‘wrong’ code, which is considered marked in this situation. Indeed, according to Alfozenti (1998: 185), the switching of code ‘highlights a conflict between norms of situational appropriateness and spontaneity of linguistic usage’.

Subsequently, in the next turn D, a Kelantanese, responds with a ban- ter, criticising C for his strong Kelantanese accent and says that nobody would understand him if he were to make the presentation, ‘Kalu mubente oghe semuo tok pahe, betul tak?’ (‘Should you present nobody would understand, right? It’s fortunate that you’re not chosen’ in turn 04). What is interesting is that while the first part of the tag question is in KD, the tag (underlined in turn 04) is in SM. When asked why he mixed code, D’s reply was: ‘Saya nak bezakan bahagian ayat yang dituju- kan kepada A dan bahagian yang ditujukan kepada yang lain’ (‘I wanted


36 Zuraidah Mohd. Don

to distinguish the part of the utterance that is meant for C and that for the others’). He said that he chose to produce the tag in SM because it was meant for A and B, the purpose of which was to get them to agree with the comment that he made about C’s fluency in SM. The comment itself ‘Kalu mu bente oghe semuo tok pahe’, which forms the statement of the tag question, was uttered in KD as it was directed at C.

C’s refusal Eh malah aku, tok pahei to comply with D’s directive that is uttered in KD Ah cepak tulihlah ‘Ah quick write’ (italicised in turn 07) is also in KD. When asked why he responded to D in KD, his reply was that since it was D who gave the instruction it was natural for him to respond to D in their shared code. Furthermore, he was rather annoyed with D for making fun of his competence in SM, and he expressed his annoyance in KD so that the others knew that his annoyance was towards D and not them. Then, in the next utterance he switched to SM as he did not want the others to feel that he was excluding them.


Being the standard variety and a lingua franca in an institution of higher learning, the Kelantanese subjects consider SM to be the language wel- comed by their non-Kelantanese Malay co-interactants although they speak a different dialect. When the Kelantanese addressed the non-Kel- antanese, they complied with the latter’s code by switching to the stan- dard variety of Malay, and this compliance is what Giles and Powesland (1997) refer to as speech accommodation. Speech accommodation here can be regarded as a device that the speaker uses to make himself better understood and be perceived more favourably. According to Giles and Powesland (1997: 233) code-switching in this regard can be viewed as a speaker’s ‘attempt to modify or disguise his persona in order to make it more acceptable to the person addressed’. This is clearly illustrated in Extract 1 whereby A, who is conversing in KD, to B (turns 0103) ac- commodates to the language of C (turn 05) by switching to SM.

What is apparent from the examination of the data is that the use of different codes by the Kelantanese participants signals their linguistic preferences dependent on who they are addressing and this pattern oc- curs within large conversational sequences. A common pattern consists of two sequences of switches: the speaker accommodates to the language of the non-Kelantanese and in the second he switches to his or her pre- ferred code, i. e. KD when addressing his fellow Kelantanese (e. g. turn 07 in Extract 1; turns 07 in Extract 2). On the other hand, the non- Kelantanese, who are restricted by their linguistic competence, do not switch codes. They communicate entirely in SM, the unmarked choice.

Language-dialect code-switching 37

The code-switching that takes place in the discourse of the Kelan- tanese speakers as demonstrated in the extracts above is intersentential code-switching embedded in intra-turn and inter-turn switches. There are few constituents that are smaller than a clause. When these constitu- ents are produced (e. g. bawok and nok pinjan (in Extract 1, turn 08) they are caused by the interference of KD. Except for the presence of these few words, there are no other switches of smaller constituents.

There seems to be a strict compartmentalisation or separation of codes with little intimate mixing of linguistic systems within utterances. Code- switching is usually at a syntactic clause boundary. When the mixing does occur, it is brought about by the competence of the speaker speak- ing the preferred code. The switching of code from KD to SM and vice versa is determined by who the speaker is addressing or responding to. This could be one of the reasons why the switching occurs only after the completion of an utterance. What we have is a sentence which is uttered fully in one code in response to what is said by a previous speaker, and the production of another sentence in another code when addressing another speaker. As pointed out by Eastman (1992), as a rule, speakers do not switch code midstream.


In the particular case discussed here code-switching is seen as a com- municative resource available to speakers who have access to more than one code. This idea is in line with Gumperz’s work (1982 and 1992) that views code-switching as much more than a strategy which the speaker employs at will to generate conversational inferences. He postulates that‘[c]ode-switching signals contextual information equivalent to what in monolingual settings is conveyed through prosody or other syntactic or lexical processes. It generates presuppositions, in terms of which the content of what is said is decoded.’ (1982: 98).

The code-switching data examined here is languagedialect code- switching within a particular genre (conversation, discussion) during a particular speech event in an urban setting. To the extent that frequent switching between KD and SM occurs in out-group situations regardless of setting, the notion of situational code-switching adequately accounts for all instances of code-switching by Kelantanese Malay.

Code-switching within a single turn of talk is a common characteristic activity of this group of participants. Most of the code-switching in- volves a whole utterance. There were few occurrences of small constitu- ents in utterances. What is clear from the examination of the data is that the Kelantanese speakers switch their codes to SM when addressing the

38 Zuraidah Mohd. Don

non-Kelantanese and KD when talking or responding to the Kelan- tanese. It is as if they monitor their code-switching determined by who the interactants are. Setting somehow does not determine the choice of code.

University of Malaya, Kuala Lumpur


1. The following conventions are used in the transcriptions: (1.5) length of silence in seconds
(.) micropause
[ overlapping utterances

latching (i. e. no latching between overlapping turns)A step up in pitch


Alfozenti, G. (1998). The conversational dimension in code-switching between Italian and dialect in Sicily. In Auer, P. (ed.), Code-switching in Conversation: Language, Interaction and Identity. London: Routledge, 180211.

Auer, P. (1991). Bilingualism in/as social action: A sequential approach to code- switching. In Papers for the Symposium on Code-switching in Bilingual Studies Theory: Significance and Perspectives. Strasbourg: European Science Foundation, Network on Code-switching in Bilingual Studies.

Blom, Jan-Peter and J. Gumperz (1971). Social meaning in linguistic structures: Code- switching in Norway. In Gumperz, J. and D. Hymes (eds.), Directions in Sociolin- guistics: the Ethnography of Communication. New York: Holt, Rinehart and Win- ston, 407434.

Blommaert, J. (1992). Code-switching and the exclusivity of social identities: Some data from campus Kiswahili. Journal of Multilingual and Multicultural Develop- ment 11 (1) and (2), 5781.

Clyne, M. (1980). Triggering and language processing. Canadian Journal of Psychol- ogy 36, 400406.

David, M. K. (2001). The Sindhis of Malaysia: A Sociolinguistic Study. London: Asean.

David, M. K. and Ibtisam Mahmoud Naji (2000). Do minorities have to abandon their languages? A case study of the Malaysian Tamils. International Scope 3, 2034.

David, M. K and M. Nambiar (2002). Exogamous marriages and out-migration: Lan- guage shift of Catholic Malayalees in Malaysia. In David, M. K. (ed.), Method- ological and Analytical Issues in Language Maintenance and Language Shift Studies. Frankfurt: Peter Lang, 125134.

Duran, L. (1994). Towards a better understanding of code-switching and interlan- guage in bilinguality: Implications for bilingual instruction. Journal of Educational Issues of Language Minority Students 14, 6988.

Language-dialect code-switching 39

Eastman, C. E. (1992). Code-switching as an urban language contact phenomenon.Journal of Multilingual and Multicultural Development 13, 12.

Ervin-Tripp, S. M. (1964). An analysis of the interaction of language, topic and lis- tener. In supplement to American Anthropologist 66, 86102.

Onn, Farid M. and Ajib Che Kob (1993). Persepsi murid dan guru terhadap pengaruh bahasa daerah dalam pembelanjaran bahasa Melayu di kalangan pelajar Melayu. In Onn, Farid M. and Ajib Che Kob. (eds.), Simposium Dialek: Penyelidikan dan Pendidikan. Kuala Lumpur: Dewan Bahasa dan Pustaka.

Gardner-Chloros, P. (1997). Code-switching: language selection in three Strasbourg department stores. In Coupland, N. and A. Jaworski (eds.), Sociolinguistics: A Reader and Coursebook. London: Macmillan Press, 361375.

Giles, H. and P. Powesland (1997). Accommodation theory. In Coupland, N and A. Jaworski (eds.), A Reader and Coursebook. London: Macmillan Press, 232239.

Goyvaerts, D. (1988). Indoubil: A Swahili hybrid in Bukavu. Language in Society 17, 23142.

Gumperz, J. (1982). Conversational code-switching. In Gumperz, J. (ed.), Discourse Strategies. New York: Cambridge University Press, 5999.

(1992). Contextualization revisited. In Auer, P. and A. Di Luzio (eds.), The Contextualisation of Language. Amsterdam: John Benjamins.

Hamers, J. F. and M. H. Blanc (1990). Bilinguality and Bilingualism. London: Cam- bridge University Press.

Heller, M. (1988). Code-switching: Anthropological and Sociolinguistic Perspectives. Berlin/NewYork: Mouton de Gruyter.

Husband, C. and V. S. Saifullah Khan (1982). The vitality of ethnolinguistic vitality.Journal of Multilingual and Multicultural Development 3, 193205.

Mohd, Osman Taib (1961). Pengaruh pengucapan dialek terhadap ejaan. Dewan Ba- hasa dan Pustaka 5 (8), 6375.

Myers-Scotton, C. (1988). Code-switching and type of multilingual communities. In Lowenberg, P. (ed.), Georgetown University Round Table on Languages and Lin- guistics. Washington: Georgetown University Press, 6182.

  •  (1993). Social Motivations of Code-switching. Oxford: Clarendon Press.
  •  (1994). What do speakers want? Code-switching as evidence of internationality in linguistic choices. In Silberman, P. and J. Loftin (eds.), SALSA II: Proceedings of the Second Annual Symposium about Language and Society Austin, Texas Linguis-tic Forum 34. Austin: Department of Linguistics, 7893.
    Nwoye, O. G. (1993). Code-switching as a conscious discourse strategy: Evidencefrom Igbo. Multilingua 12 (4), 365385.
    Poplack, S. (1980). ‘Sometimes I’ll start a sentence in English y termino en españ ol.’Two codes or one? Journal of Multilingual/Multicultural Development 13, 83102. Sankoff, G. (1971). Language use in multilingual societies: Some alternative ap- proaches. In Pride, J. B. and J. Holmes (eds.), Sociolinguistics. London: Penguin,89103.
    Sebba, M. and T. Wootton (1998). We, they and identity: Sequential versus identity-related explanation in code-switching. In Auer, P. (ed.), Code-switching in Conver-sation: Language, Interaction and Identity. London: Routledge, 262286.
    Stolen, M. (1992). Code-switching for humour and ethnic identity: Written Danish- American occasional songs. Journal of Multilingual and Multicultural Development13: (1/2), 215228.
    Thomasan, S. G. (2001). Language Contact: An Introduction. Edinburgh: EdinburghUniversity Press.

40 Zuraidah Mohd. Don

Valdes-Fallis, G. (1977). Code-switching among bilingual Mexican-American women: Towards an understanding of sex-related language alternation. International Jour- nal of the Sociology of Language 7, 6572.

Zuraidah, M. D. (2003). Vitality perceptions and intergroup contacts: Kelantanese in a new ethnolinguistic environment. International Journal of the Sociology of Language, 161 (in press).