Motivations for code-switching among Arabic-French bilinguals in Morocco

Motivations for code-switching among Arabic-French bilinguals in Morocco

Language& Communication, Vol. 3, No. 3, pp. 233-243, Printed in Great Britain. 0271-5309/83 s3.00+.00 Pergamon Press Ltd. 1983 MOTIVATIONS FOR CODE...

870KB Sizes 37 Downloads 102 Views

Language& Communication, Vol. 3, No. 3, pp. 233-243, Printed in Great Britain.

0271-5309/83 s3.00+.00 Pergamon Press Ltd.

1983

MOTIVATIONS FOR CODESWITCHING AMONG ARABIC-FRENCH BILINGUALS IN MOROCCO ABDELALI BENTAHILA Introduction Code-switching, or the alternation between two different languages within a single conversation, 1 is a common feature of the speech of those Moroccans who are fluent in both French and Arabic. It is particularly characteristic of informal conversations where all the participants are bilingual, and in such circumstances a bilingual may switch between one language and the other several times within the same utterance. The syntactic features of such code-switching have been discussed elsewhere (Bentahila, 1981); here I shall look at some of the reasons why switches occur, and the contributions they make to the speaker’s communicative purpose. That code-switching is so much used may seem remarkable in view of the fact that it is apparently very unfavourably viewed by the bilinguals themselves. In an investigation described in Bentahila (1983), 75.22% of a sample of bilinguals questioned about their attitudes to such language mixture were strongly disapproving of it, their comments suggesting that they associated it with lack of education, carelessness, affectation and lack of identity. Nor do these comments seem to be provoked merely by a desire to express what is thought to be the ‘correct’ opinion, for a more indirect investigation of bilinguals’ attitudes by means of a matched guise test produced similar results; guises involving a mixture of French and Arabic were significantly less favourably perceived than those using either purely Arabic or purely French on a variety of traits, including some related to prestige and some to solidarity. In view of the apparent disrepute in which such language mixture is held, then, it is all the more interesting to ask why it is nevertheless such a common feature of the bilingual’s everyday language. The following discussion is based on an examination of the switches occurring in seven and a half hours of conversations which were recorded without the knowledge of the participants, who were all balanced bilinguals aged between seventeen and forty. The conversations took place in a relaxed home setting, involved between two and four participants, meeting on a friendly basis, and covered a wide range of topics, from idle chat about food, weather and everyday events to more serious discussion of work, politics and education. The tone of the speakers varied from light-hearted and humorous to argumentative and dramatic. An examination of the data suggests that the distribution of switches is by no means as arbitrary as might be imagined by a casual observer. It is possible to identify a number of factors which may provoke a switch; a few examples will serve to illustrate these. Switching provoked by external factors It seems possible to attribute many switches to the influence of factors external to the speaker himself, in particular to the type of topic under discussion. That Moroccans are 233

234

ABDELALI

BENTAHILA

more accustomed to use one language than the other in discussing certain topics is shown in Bentahila (1983); and these habits are reflected in the numerous switches which result in the use of lexical items from one language rather than the other when referring to certain topics. This would often seem to be because the terms from one language are more readily available to the bilingual than those from the other; although in most cases the speaker probably knows appropriate terms in both languages, he seems to recall those in one language more immediately than those in the other. The effect of availability of vocabulary has also been observed in the code-switching of Spanish-English bilinguals (Lance, 1975) and of German-English bilinguals (Clyne, 1967). The influence of topic is illustrated in our data by the frequent switches from Arabic to French for technical terms referring to medical, educational or administrative matters, as in the following quotations. (1) (2)

(3)

minIli:t tatswwab le vaccin ba3 j3uf I reaction positive2 as soon as he has the vaccination in order to see the positive reaction des urgences dajri: n fi: h le personnel taj.iql _FlaZed nnmi the emergencies, where they put the staff who have been there for donkey’s years bhal daba f la facultefhmti tadxul tsuf le doyen for example, at the university, you understand, you go to see the dean.

This tendency may reflect the fact that French is still widely used in the education and medical sectors in Morocco. It is interesting, too, that a switch to French is something made to refer to a concept associated with Europe, such as an item of food not part of traditional Moroccan cuisine, as in (4), or a type of accommodation found in modern European-style apartment blocks, as in (5). (4)

daba had le the noir xalih Zwija now this black tea leave it for a while (5) jqullu ana iandi garconniere w hada jqdar ga-5jTla5 Swija he’ll tell him I have a one-room flat and whatever, maybe he’ll increase the rent a bit.

The data reveal a strong tendency for bilinguals to switch to French in order to indicate a number, date or time. There are numerous examples where an expression of this kind in French occurs in what is otherwise an Arabic utterance. (6)

(7) (8)

huwa qailik la: majmknz nTeji?i cinquantepour cent w hada fhmti hi Teih tqri: ban trente mille francs djal 1 kra? he said to you no, it is not possible to make a reduction of fifty per cent and whatever, do you understand? Just give him about thirty thousand francs for the rent la: xri mIa 5 quatre heures moins cinq no, he went out about five to four kunt f le neuf juillet f ifra: n I was in Ifrane on the ninth of July.

The tendency to switch to French for numbers could perhaps be related to the fact that the speakers would all have been taught their arithmetic in school exclusively through the medium of French. The use of French for dates can perhaps be explained in terms of the

CODE-SWITCHING

AMONG ARABIC-FRENCH

BILINGUALS

IN MOROCCO

235

difference between the Muslim calendar, which is little used in Morocco except in religious contexts, and the European Julian calendar, which is almost exclusively used in administration and public services. While French lexical items seem to be more available than Arabic ones in the domains noted above, for religious topics the opposite seems to be the case; not surprisingly, perhaps, there is often a switch to Arabic to refer to a religious custom. c ‘est une femme de marque qui est f Si Sk1 tm.5 tzur SSal?ii:n she is a special woman who is the sort who goes to visit saints’ tombs (10) ce n ‘est pas religieux. . . le fanaticisme, elle croit f SSalhi:n 9a.3 it’s not religious . . . fanaticism, she believes in the saints’ power, why? (9)

Arabic also seems to be favoured over French for expressing insults and swearing, so there is often a switch to Arabic for this purpose: (11) elle n ‘a qu’a etudier un peu d’histoire nlial bu:ha avant de me poser des questions she’d better study a bit of history, damn her father, before asking me questions. Also among the expressions which seem to be more available in Arabic are a number of stereotyped phrases, typically used as fillers to avoid a pause, often in what are otherwise French utterances.

(12) c’est un type qui aime commander la femme w dak Si tout en &ant moderne w dak 5 elle n ‘apas accep te he is a man who likes to command women and so on, while still being modern and so on; she didn’t accept that (13) en commencant d’abord par une &ape, puis enseigner les deux . . . enseigner bhal daba qu’est-ce qu’on fait . . . bfial daba fhmti hna bhal daba on nous a conseille de traduire la medecine starting first with one step, then teach the two . . . teach, for example, what do we do? . . . for example, do you understand? us, for example, we have been advised to translate medicine, A similar tendency has been noted by Gumperz and Hernandez-Chavez (1975) in the speech of Spanish-English bilinguals, who apparently also often switch to their first language for such idiomatic phrases. In some cases a switch may be motivated, not so much by the greater availability of a word in one language, as by the fact that the word in one language is more suited to the speaker’s purpose, because it carries particular connotations absent from the corresponding item in the other language. This seems to be the case, for example, in (14), where the Arabic word [Iaskri] carries connotations of ignorance and lack of refinement which the French soldat lacks. (14) ii va parler d un Jiaskri qu ‘est-ce que tu veux he’ll speak to a soldier, what do you expect? In other cases, it is the desire to avoid certain connotations which motivates a switch; for instance, there are a number of examples where a switch to French seems to be motivated by the desire to avoid using what would be a taboo word in Moroccan Arabic.

236

ABDELtiLI

BENTAHILA

w kajn lli msaw 1 la toilette and there are those who went to the lavatory (16) wa#ied lli Iandu la diarrhee tajSwb Swija someone who has diarrhoea can take a bit of it.

(15)

Here the switch enables the speaker to avoid embarrassment. Sometimes a switch which is motivated by the need to use one language rather than the other for some particular item is not restricted to this item alone, but extends beyond it. This is the phenomenon which Clyne (1967) labels triggering. He distinguishes between anticipational triggering, where the switch is made before the speaker reaches the particular item which motivates the switch, and consequential triggering, where the speaker, after switching to one language for a particular item, continues in this language instead of returning to the pre-switch language. Both these types of triggering can be observed in our data. For instance, in (17) the switch seems to anticipate the need to use the French term licence, which is much more readily available than its Arabic counterpart, which is rarely used by Moroccans. (17) ana wka:n lqa sskna ana je doispreparer une f(c~& me, if I find accommodation I must prepare a (d_e_gfe_e_. In (18), the expression which seems to have motivated the switch is the Arabic phrase [mula: j bih], a commonly used idiom which means ‘nobody’, but again, the switch comes before this is reached. (18) tu as verifiP si tu as les m@meschases parce qu ‘unefois tii tlqa @a:j_b_b_ih_ have you checked if you have the same things, because once you come, you’ll @d

nota_%?!!*

Consequential triggering is illustrated in (19), where the switch to French is motivated by the fact that there is no real equivalent to the French term weekend in Arabic; once he has used this French word, the speaker continues in French: (19) ana ma_Trftts'fuqa3 jii raSi:d qalk 1 weekend ___---- je ne sat3 pas quand est-ce que le weekend d ‘apres lui I don’t know when Rachid is coming; he said at the weekend, I don’t know when the weekend is according to him. In (20), the initial switch is in order to use a stereotyped Arabic phrase like those discussed earlier (see examples (12) and (13)), after which the speaker continues in Arabic. (20) je fais trois kilometres, quatre kilometres pour prendre un cafe’parce que bhal --_-. 1qadija-, djal nnukta djal ss(id kan tajskr __-__ I do three kilometres, four kilometres to have a coffee because, like the case of -_--_-__---_) that joke of the chap who used to get drunk. In cases like these, then, the source of a quite lengthy switch can be found in a particular item within this stretch of speech, which for some reason favours the use of one language rather than the other. Switching as a rhetorical device

In the examples discussed so far, the speaker is in a sense constrained to make a switch by circumstances outside his immediate control, such as the availability or connotations

CODE-SWITCHING

AMONG ARABIC-FRENCH

BILINGUALS

IN MOROCCO

237

of particular words. There are however other cases where a switch seems to be more deliberately chosen and exploited by the speaker to achieve a particular type of effect. The data provide many illustrations of how code-switching can be used as a stylistic device to produce special rhetorical effects. Firstly, there are many instances where a speaker first says something in one language and then repeats the same point in the other. (21) 1muhim swltu je lui ai demand6 the important thing is I’ve asked him, I’ve asked him (22) je ne luiplaispas, quoi, maIZbtu3 he doesn’t like me, what, he doesn’t like me. Kachru (1977), who notes the same kind of repetition in the code switching of HindiEnglish bilinguals, suggests that it may be motivated by a desire ‘to avoid vagueness or ambiguity’ (p. 111). However, this does not .seem true of many of the examples in my data, where what is repeated is something very simple which could not have been misunderstood the first time, and where the version given in the second language is no more clear or specific than that first given. Instead, the purpose seems to be to add emphasis to what is said. Rayfield (1970), observing the same kind of repetition in the code-switching of Yiddish-English bilinguals, suggests that their use of this device may be influenced by the frequency with which emphatic repetition occurs in Biblical Hebrew. One could similarly link its use by Moroccans to the repetition characteristic of Arabic literary texts; but in fact there seems no need to seek such an explanation, for it seems in fact to be a very common strategy of bilinguals, which has also been attested in Spanish-English code-switching (Gumperz, 1976; Redlinger, 1976; Timm, 1975) and in Hindi-English and Slovenian-German code-switching (Gumperz, 1976). There are also other environments where a switch serves an emphatic function. There are many cases where a speaker addressed in one language switches to the other to reply and thereby emphasises his disagreement with the previous statement, as in (23): (23) A: B: A: B:

N va voirsa m&e la: la: ma ma3iSj3uf jma:h He’s going to see his mother No, no, he’s not going to see his mother.

In (24), the second speaker repeats what the first has said, but in the other language, to emphasise his astonishment and indignation. (24) A: B: A: B:

il double sapremi&re et maintenant ils i’ont jet& sjbu: h et ca colite dkjd Id-bas trop cher He’s repeating his first year, and now they’ve thrown him out They’ve thrown him out, and it already costs too much there.

In (25), the switch to Arabic is for a humorous, sarcastic comment. (25) A: B: A: B:

aprt?s il est sac%/, il n ‘est pas 14 maa: n afterwards he’s drunk, he’s not there good for him.

230

ABDELi-iLI

BENTAHILA

Another common exploitation of code-switching is to make an interruption. In such cases the break between languages reflects the break in the interrupted speaker’s flow of speech, and the second speaker changes languages in a bid to obtain attention and gain the floor, as in thk following examples: (26) A: elle est bien sauf moi je colle un peu partout alors ce qui fait . . . B: Suf 5andi wafzed 1jacket Suf was tZi m_iak A: she is fine, except for me, I mess about here and there, so what happens is that . . . B: look, I have a jacket, see if it suits you (27) A: il y a Nouzha qui habite d Mohammedia . . . B: Skun had nzha A: celle qui a &t&bRabat A: there is Nouzha who lives in Mohammedia . . . B: who is this Nouzha? A: the one who was in Rabat (28) A: toi, en tout cas, lorsqu'ilpleuvait ... B: la: 1rfti Ta:A 1krT A: you, in any case, when it rained . . . B: no! you know, hailstones fell. In examples like (26), the switch to Arabic coincides with the introduction of an entirely new topic, and may help to draw attention to this change of subject. In (27), the interruption is a request for clarification, but the first speaker, in responding, continues to use French; the use of Arabic for the interruption alone seems to mark it as separate from the general flow of information. In (28) the interruption is in order to correct a false assumption, and here the switch, as well as serving to draw attention to the interruption, may also emphasise the element of contradiction, as it does in (23). The contrast between the two languages may also be exploited for dramatic effect in a narrative or description. A very effective use of switching is seen in (29): (29) au lieu d’acheter des antibiotiques tatSriSmaS instead of buying antibiotics, she buys candles The two parts of the utterance describe two alternative ways of reacting to someone’s illness, first the modern solution of using available medical technology, and secondly the religious solution of offering candles to a saint. The change of language here perfectly accords with this contrast of content, the modern technological solution being described in French and the traditional religious one in Arabic, and thus serves to heighten the contrast quite dramatically.4 The rhetorical effects of code-switching are even more striking in some longer stretches of speech, where the pattern creatqd by switching back and forth between the two languages plays an important part in the overall dramatic impact. The following speech, which is a character sketch, will illustrate the contribution of switching over a longer piece of discourse. (30) 1 comme firned, tu le connais, c’est le Berb&re type 2 majTlq frank. I1fait mathPmatiques, il est gPnPraliste, 3 !iandu dda: r, il a deux appartements, Sa: ri 1 ?arD,

CODE-SWITCHING

AMONG ARABIC-FRENCH

BILINGUALS

IN MOROCCO

239

mais walu. II n ‘a m2mepas la t&Wsion en couleurs; il a une Detite radio b7ial hadak. C’est le Soussi DUC et pourtan t il est c&iv&, ii a un tr& bon poste II se fait beaucoup d’argent et tout, mais makajn3 Les affairespour lui: hadik ixdma, il voudrait bien la balancer et avoir une Poicerie. Like Ahmed, you know him, he’s the typical Berber, who doesn’t spend a single penny. He does mathematics, he is a general practitioner, he has a house, he has two apartments, he’s bought some land, but-nothing. He doesn’t even have a colour television; he has a little radio like this one@ointing to one in the room). He is a real Soussi; and yet he is cultivated, he has a very good job. He makes a lot of money and everything, but there is nothing. Business for him: that job, he would like to abandon--~it and have a grocer’s shop. The main point being made by the speaker here is that, although Ahmed has plenty of money, he is reluctant to spend it and prefers to live a very simple life. After first introducing the subject, Ahmed, in French, he switches to Arabic to make his essential point, that Ahmed does not like to spend money. He proceeds to make the point, very dramatically, that Ahmed is rich. This he does by listing one by one the signs of his wealth, progressing from the fact that he is educated and well placed, through the fact that he owns his own house, and, even more remarkable, other property as well, to what in Morocco is one of the ultimate signs of financial success, the fact that he has bought some land. The dramatic effect of this tirade is greatly enhanced by the speaker’s use of switching; he first makes a point in French, then one in Arabic, another in French and a final one in Arabic, and the alternation of the two languages serves to add to the impact of each point by contrasting it with the previous one. It is interesting, too, that it is the references to his education and to his possession of modern, European-style property, in the form of apartments, which are in French, while the other points, which have no such modern connotations, are in Arabic. When this progression reaches its height, he switches again to French to introduce the contradiction with mais, and then, in contrast to the lengthy enumeration of details which preceded it, comes the single Arabic word [walul-nothing. Again the important point, Ahmed’s lack of expenditure, is made succinctly and in Arabic. The speaker then goes on to add further details exemplifying Ahmed’s meanness. He switches to Arabic to make an aside, indicating the radio; Arabic seems to be used here to keep aside separate from the mainstream of his speech, as it is in (32), to be discussed below. He returns to his description of Ahmed’s wealth and status, still in French, and then again there is a dramatic switch to Arabic to repeat once more, even more emphatically, the fact that there is no outward sign of this wealth: [makajnff. It is striking that this fundamental point is made three times in the course of the speech, each time very briefly, and each time with a switch to Arabic. Finally, he returns to French for the final illustrative details, but makes one more switch to Arabic which allows him to make a reference to Ahmed’s work which reflects Ahmed’s own disparaging view of it. This is achieved through the connotations of the word [xdma] which tends to be used more often to refer to manual work than to a profession. The effect of the description as a whole thus seems to owe much to the speaker’s judicious use of code-switching, which adds emphasis, brings out contrasts, heightens progressions, and thereby increases the dramatic impact. The data reveal that switching is also used to mark an aside or parenthetical

inter-

240

ABDELALI

BENTAHILA

polation, where a speaker temporarily abandons the general topic under discussion to make some comment on another, possibly quite unrelated one. The change of language seems to be a useful device for marking such a change of topic. This strategy is illustrated in the following examples: (3 1) f lluwl kanu ga_5ma kajbanu. Tu veux un peu de coca? at the beginning they weren’t seen at all. Do you want some Coke? (32) de toutes les facons demain je monte a la fat. thersat dak tZa: Za in any case, tomorrow I’ll go up to the university That light bulb is broken (to someone trying to put the light on). Another very common tendency is for a switch to coincide with a quotation, so that the quotation itself is in a different language from that used to introduce it. This has the effect of separating the quotation from the rest of the utterance and making it stand out from its narrative background. In examples like (33) and (34), a quotation in Arabic is introduced in French. (33) il va dire hada j5ra.f 1frma: sija he’s going to say ‘this one knows the chemist’s’ (34) je t ‘ai dit xSSk taZm1 Swija 5aqlk I told you you must be logical. In these cases, there are good reasons why it is the quotation itself which is in Arabic. In (33), the person quoted is a lowgrade soldier who would not be fluent in French and whose original words would have been in Arabic; while in (34), the quotation contains an idiomatic expression, typically used by older people to the younger generation, which has no obvious French equivalent. However, the reverse pattern is more common in the data; where a conversation is reported at some length, it often happens that, while all the quotations are in French, they are introduced by the Arabic verb [qal] ‘say’. This has the effect of leaving the introductory verbs unobtrusive in the background, while the information contained in the dialogue stands out. (35) qultlu qu’est-ce qui ne va pas qali vous savez votre billet est drole, oh la la qultlu vous savez je crois que vous &es plus drole que le billet I said to him ‘What’s the matter?’ He said to me ‘You know, your ticket is funny’. I said to him ‘You know, I think you are funnier than the ticket’ (36) qali vous &es du Maroc qali vous habitez 02 qultlu eh bien j’habite d Casablanca qali vous comptez rester combien de temps en France He said to me ‘Are you from Morocco?’ He said to me ‘Where do you live?’ I said to him ‘Well, I live in Casablanca’. He said to me ‘How long are you thinking of staying in France?’ The tendency for code-switching to be used to set off a quotation has also been noted in the speech of other types of bilingual by Gumperz (1976) and Timm (1975). Switching after hesitation There is another common use of code-switching

which emerges from an examination

CODE-SWITCHING

AMONG ARABIC-FRENCH

BILINGUALS

IN MOROCCO

241

of the data. This is the tendency to use a switch in order to resolve a hesitation. There are many instances where a speaker pauses in mid-utterance, as if uncertain how to continue, and then finally begins again in the other language. In some cases, the speaker turns to the other language simply to fill in the gap caused by his hesitation, before returning to his original language once he has regained the thread of his discourse. Thus, in examples like (37) and (38), the switches to Arabic are simply to use filler words or phrases, while the speaker reflects on how best to express his main point; in each case he goes back to French when he finally continues with his message. (37) c”est tout. . . i:h . . . walu . . . m$me . . . za3ma. . . il veut s’imposer that’s all . . . yes . . . nothing . . . even . . . supposedly , . . he wants to impose himself (38) il y a une grande diffgrence . . . Zta f la boufftance . . . ?ita f la boufftance . , . smijtu . . . il y a des chases qui . . . za5ma . . , ontplus de qualit& there is a big difference . . . even in the food . . . even in the food . . . what do you call it? . . . there are things which . . . supposedly. . . are of better quality. Other examples illustrate a rather different strategy, whereby the speaker, after seeming to experience difficulty in expressing his message in one language, pauses and then continues with the same message in the other language. (39) j.ini ce n ‘est pas. . . had3 lli fsrtiu I mean, it isn’t . . . the thing I explained to him (40) qTTa.St min. . . de Casa d Vienne I had my ticket from . . . from Casa to Vienna (41) wqa_itlif. . . lorsque je ren trais de la Tchkoslovaquie it happened to me in . + . when I was coming back from Czechoslovakia . (42) taduwwz l’enfance djalk . . . par exempie , . . ana kunt f Iagadir . . . les gens sont. . , za Ema matajlSbu5 had3 you spend your childhood . . . for example . . . I was in Agadir . . . people -are . . . supposedly, they don’t play with things like that. In some cases, as in (39), the material in the second used language simply follows directly on from what is said in the first, while in others, such as (40), part of what was said in the first language is repeated in the second, and in others the switch also accompanies a change of structure, as in (41), or a fresh start altogether, as in (42). It would appear then that once the speaker has lost the thread of his remark in one language, he may find it easier to start again in the other. Clyne (1967) observes that a code-switch may be motivated by the desire to find the easiest way of expressing something, claiming that the use of code-switching is influenced by the amount of effort or strain involved in the expression of certain ideas, in that ‘bilinguals tend to employ the forms that require the least effort (strain) or number of decisions’ (Clyne, 1967, p. 82). Some of the examples of switching after a hesitation may be motivated by this desire for avoidance of effort. Conclusion

This brief survey has illustrated some of the patterns to be detected in the Moroccan bilingual’s use of code-switching in casual conversation. It may now perhaps no longer be

242

ABDELALI

BENTAHILA

found surprising that bilinguals, despite their generally negative attitudes towards codeswitching, nevertheless make frequent use of it. For it has now become clear that codeswitching serves a variety of important functions, and may contribute in a number of ways to the facility of the bilingual’s expression and the effectiveness of his communication. In the first place, it was seen that switching may allow the bilingual to use the vocabulary which he finds most available or most appropriate to a particular topic, thereby making possible greater fluency than if he had to search.for suitable terms from within only one of his two languages. The wider choice from the two languages at his disposal may also allow him to find the word which carries a particular connotation, or to avoid using a word the connotations of which might cause him or his hearers embarrassment. Secondly, we saw that code-switching can be used as a rhetorical device to achieve a variety of special effects. It can serve to emphasise a point, to add variety or to heighten a contrast; and the speaker may use it in an attempt to gain the floor or to change the topic of discourse. Thirdly, it was seen that switching may be a strategy adopted when the speaker gets lost for words; if he finds it difficult to express himself in one language on a particular occasion, he can make a second attempt in the other language. The examples quoted above are, I hope, sufficient to illustrate the point that the bilingual’s use of code-switching is not arbitrary or meaningless; on the contrary, the strategies he employs and the effects he creates suggest that code-switching is an important part of his communicative competence.

NOTES ’ The question of how to define code-switching is discussed in Bentahila (1981), where it is shown that this is quite separate from the phenomenon of borrowing, which is a feature of the language used by monolinguals as well as bilinguals. ’ In transcribing the examples, French words are represented by their ordinary orthography, and are underlined, while phonetic transcription is used for Arabic words. 3 In examples (17)-(20). the trigger element in each case is marked by broken underlining, and the arrow indicates the direction of the trigger’s effect. 4 The tendency for Moroccan bilinguals to associate French with modernity and Arabic with tradition is demonstrated in various ways in Bentahila (1983).

REFERENCES BENTAHILA, A. (1981) Attitudinal aspects of Arabic-French bilingualism in Morocco. Ph. D. thesis, University of Wales. BENTAHILA, A. (1983) Language Attitudes among Arabic-French Bilinguals in Morocco. Multilingual Matters, Clevedon. CLYNE, M. G. (1967) Transferenceand Triggering. Martinus Nijhoff, The Hague. GUMPERZ, J. J. (1976) The sociolinguistic significance of conversational code-switching. J and Gumperz, J. J. (Eds), Papers on Language and Context (Working Paper 46), pp. California Language Behavior Research Laboratory, Berkeley. GUMPERZ, J. J. and HERNANDEZ-CHAVEZ, E. (1975) Cognitive aspects of bilingual Hernandez-Chavez, E., Cohen, A. and Beltramo, A. (Eds), El Lenguajede fos Chicanos. Linguistics, Arlington, VA.

In Cook-Gumperz, l-46. University of communication. In Center for Applied

CODE-SWITCHINGAMONG ARABIC-FRENCH BILINGUALSIN MOROCCO

243

KACHRU, B. B. (1977) Code-switching as a communicative strategy in India. In Saville-Troike, M. (Ed.), Linguistics undAnthropology. Georgetown University Round Table on Languages and Linguistics. LANCE, D. M. (1975) Spanish-English code-switching. In Hernandez-Chavez, E., Cohen, A. and Beltramo, A. (Eds.), El Lenguuje de 10s Chicanos., Center for Applied Linguistics, Arlington, VA. RAYFIELD, J. R. (1970) The Lunguugtzs of u Bilingual Community. Mouton, The Hague. REDLINGER, W. E. (1976) A description of transference and code-switching in Mexican-American English and Spanish. In Keller, G. D., Teschner, R. V. and Viera, S. (Eds), Bilingwlism in the Bicentenniuland Beyond. Bilingual Press, New York. TIMM, L. (1975). Spanish-English

code-switching: elporquey how-not-to. Romunce Philology 28,473482.