Zambezia (1990), XVII (i).RESEARCH REPORTAN INVESTIGATION INTO LANGUAGE ATTITUDES OFSECONDARY-SCHOOL STUDENTS IN ZIMBABWE*CYNTHIA MPARUTSA, JULIET THONDHLANA, N. CRAWHALLDepartment of Linguistics, University of ZimbabweTHE SIGNIFICANCE OF the role that attitudes play in language education,planning and development has been noted by researchers in Africa andelsewhere. To date there have been very few documented studies onlanguage attitudes in Zimbabwe. This present study was inspired byprevious work carried out by Ngara (1982), Mparutsa (1986), and Chiwomeand Thondhlana (1989). Ngara examined the attitudes to English andShona of 60 Shona-speaking students in Form III at three different schools.Hofman (1977) writing earlier, felt able to categorize language attitudesinto intrinsic and extrinsic value systems, noting the ambivalence withwhich the dominant language, English, was viewed by first language (LI)Shona speakers. But, as Ngara notes, the liberation war radicalized Africanpolitics and Zimbabweans became increasingly aware of language as asymbol of culture and nationalism. While English is still regarded as thelanguage of officialdom and education, there is an increasingly positiveattitude towards the use and the value of Shona (Ngara, 1982, 24-6).Mparutsa (1986), in an unpublished study of 100 Harare residents, foundthat, although English was seen as the language of success, a highproportion of the respondents would like to see Shona used more widely.Similarly, Chiwome and Thondhlana (1986), in their interviews withsecondary-school teachers and students, reported that, while manystudents chose to do English at 'O' level, they often felt that they couldexpress themselves better in Shona.PROCEDURE AND RESULTSThis article seeks to explore some of the contradictions in the languageattitudes of school children in Zimbabwe that were discovered in the datacollected for a study of language attitudes among secondary-school pupils.1Preliminary findings of this study are based on responses to questionnairesgiven to 100 upper-secondary-school pupils. The questionnaire was dividedinto two parts: Section A dealt primarily with language use questions and* This article is based on a paper originally given at the Socio-linguistics in AfricaConference, University of the Witwatersrand, Jan.-Feb. 1990 to be published in R. Herbert(ed.), Language and Society in Africa: The Theory and Practice ofSociolinguistics (Johannesburg,Witwatersrand Univ. Press).1 The survey and the results given in this article are intended as a pre-test for a larger-scale research project Into the question of language attitudes that will be taking place over thenext few years.8586 LANGUAGE ATTITUDES OF SECONDARY-SCHOOL STUDENTSSection B dealt with language attitudes (see Appendix 1). The questions inSection B were taken from Ngara's study so as to enable us to examinepossible shifts in attitudes from before and after Independence. The upper-secondary-school age-group was chosen for three main reasons: firstly, asthese pupils had received most of their education since Independence, itwas expected that they would have internalized some of the changes inthe curricula; secondly, the results from this age-group could be comparedwith the results given by Ngara and by Chiwome and Thondhlana; andfinally, we had relatively easy access to this age-group for an initial inquiry.The University of Zimbabwe Open Day (12 August 1989) was an excellentopportunity to reach students from both rural and urban schools from allparts of Zimbabwe. In addition, to round out our study, we were able toadminister some 20 questionnaires to students at a nearby secondaryschool. Of the 100 respondents all were secondary-school students. Themajority were 16-20 years of age. A more accurate figure on age distributionis not available as more than half the respondents did not fill in the answerto the question asking the respondent to give his or her age group.In the construction of the questionnaire we embedded one attitudinalquestion (Question 4) in Section A which was, otherwise, mostly concernedwith the language use of the respondents. It was thought that, becausethis question, which asked whether the respondent would favour a changefrom English to Shona/Ndebele as the medium of instruction in secondaryschools, was surrounded by 'factual' response-based questions, it wouldtrigger a different ideological response from that elicted by the attitudinalquestions in Section B. To see if the answers to Question 4 correlated withthe types of answers in Section B we divided the respondents into twogroups. Group A included all those who had answered 'yes' to Question 4(15 respondents), and Group B included all those who had answered 'no'(83 respondents). Two respondents did not reply to Question 4; this grouphas been been called the NR-group. For the purposes of correlating groupascription with answers to the other questions in Section A we haveincluded the NR-group in Tables IŠVII.It was hypothesized that those respondents who favoured an increasein the use of indigenous languages in school curricula would have noticeablydifferent responses to the attitudinal questions in Section B. As it wasthought that Group A respondents wouldtend to value Shona or Ndebelemore highly than Group B respondents, we attempted to correlate A and Bgroupings with the answers given in Section A and in Section B. Overall itappeared that the groupings did reflect a pattern of use (see Tables I-XIQ.However, when we attempted to correlate the groupings with the attitudinalquestions of Section B we realized that there was no appreciable differencein the responses of Group A and Group B respondents. This in itselfproved interesting. We asked ourselves why respondents would argue inQuestion 4 that English had to remain the dominant educational languagebecause of its international position, yet rally to the support of the in-digenous languages in their answers to questions in Section B. In particular,we were interested in the responses to Question 11 which asked whethera person who does not speak or write English can be considered noteducated. Of the total 83 respondents in Group B 68 said that they wouldCYNTHIA MPARUTSA, JULIET THCNDHLANA AND N. CRAWHALL 87consider unilingual indigenous-language speakers to be 'educated'. Bywhat criteria? Not their own it would seem, nor those of the educationalsystem which does not present unilingual indigenous-language educationas an option. It was at this point that we began to look at the issue ofcontradictory attitudes held by secondary-school students towards theindigenous languages.It was thought that Question 4 and Section B questions were triggeringdifferent ideological responses in the respondents. It would appear thatSection B aroused a sense of duty in most respondents, encouraging themto defend the indigenous languages against probing inquiries which mightbe construed as hostile and which could be seen to be tied, even ifantithetically, to colonial prejudices. Group A and Group B respondentsseemed to rally to the defence of the indigenous languages even if theirresponses contradicted their answers to Question 4. In contrast, Question4 triggered a response closer to what might be called a 'practical' ideologicalposition, that is, eliciting an answer referring more to concerns aboutpersonal advancement.SOME THEORETICAL CONSIDERATIONSThe following theoretical discussion is presented as an attempt to link oursocio-linguistic work with the larger body of critical linguistics. Criticallinguistics is a school of linguistics which has developed in Britain andAustralia, at Lancaster and Deakin Universities, respectively. In particular,the theoretical considerations presented here have been influenced bythe work of Coward and Ellis (1977) who developed a theoretical base forthis new direction in linguistics by bringing together two of the twentiethcentury's most important psycho-social theories, Marxism and Freudian/Lacanian psychoanalysis. The terminology used in this section should beseen in the context of both dialectical materialism and psychoanalysis.In both the construction of hypotheses and in the analysis of the datawe necessarily had preconceptions which influenced our choice of areasupon which to focus our attention. In particular, we took as given thenotion of 'contradiction' in attitude testing: when applied to languageattitudes this connotes larger ideological-cultural contradictions.Contradictions should not be regarded as being irreconcilably opposed toone another, but rather as being motivated by different socio-economicinterests. Their relationship is dialectical; the arena of this dialectic ispsycho-sociological. That is to say a person might hold two attitudesabout a language simultaneously. These two attitudes could have twodifferent historical origins. For the individual concerned these twocontradictory attitudes would have an influence on one another. Withoutthe individual discarding either attitude they would have to be moderatedin circumstances where the contradictions became explicit. The data forthis study suggest that at least two primary contradictions exist in thelanguage attitudes of the respondents: a contradiction between use andattitude, and a contradiction between attitudes towards indigenous-language development and personal advancement.LANGUAGE ATTITUDES OF SECONDARY-SCHOOL STUDENTSFor the speaker, ideological-cultural contradictions are usuallymanifested unconsciously and when challenged they are described alongthe lines of 'common sense'. It is one of the objectives of attitude testing tobring to the surface (that is, into analytic discourse) some of thecontradictions present in the speaker's mind, both conscious andunconscious. We wish to emphasize, however, that not all ideological-cultural contradictions have the same structure or historical significance.There is a sizeable body of Marxist writing on the subject of contradiction,including Mao Tse Tung [Mao Zedong]'s discussion (1965) of antagonisticand non-antagonistic contradictions, which is relevant to this discussion.From the psychoanalytic tradition we can also see that the two competingideological positions (or attitudes) that form a contradiction may beconscious elements of a particular culture, or they may be based in theunconscious.This theoretical approach may sound convoluted at first but it shouldbe noted that the contradictions that we are discussing are recognizedideological issues in Zimbabwe. The liberation war put considerableemphasis on pride in indigenous culture and language. The use of theindigenous languages was a counter-ideological instrument used againstthe colonial ideological hegemony and its linguistic instruments, Englishand Chilapalapa (Fanagalo). Yet, since Independence, English has remainedthe de facto language of power and economic advancement. School curriculareflect this contradiction in a number of ways; and, possibly, teachersthemselves reproduce this contradiction through their own attitudes andactions.ZIMBABWEAN UNGUISTIC CONTRADICTIONSFrom previous knowledge of the educational system and observation ofattitudes, we are of the opinion that first-language Shona and Ndebelestudents often prefer working in their first languages rather than in English.This view has been substantiated by Chiwome and Thondhlana (1989)during their interviews with Shona-language teachers in secondary schools.Teachers reported that students often claim to be able to expressthemselves better in their first language. Yet when it comes to choosingsubjects at 'O' and 'A' level many students choose English in preference toShona as there is a widely held belief that a certificate in an indigenouslanguage is not likely to help young people in the job market. English isconsidered by many to be the language of economic opportunity andadvancement. This is true in both rural and urban schools, as well as inhigh- and low-density suburban schools. The degree to which this attitudeprevails in these population groups is not known as yet and we hope toextend our study to look at these differences. Consequently, most first-language indigenous-language speakers find themselves in a position ofpreferring their first language for communicative purposes yet functioningin an educational, social and economic system that emphasizes theimportance, even the dominance, of English.From the data we observed that Group A and Group B have differentuse patterns which coincide with their responses to Question 4. LookingCYNTHIA MPARUTSA, JULIET THONDHLANA AND N. CRAWHALLat the answers to the questions in Section B we cannot see an appreciabledifference in attitude between members of Group A and Group B. Thereare two ways of looking at the data. On the one hand, there is a contradictionbetween Group B respondents' use of the indigenous languages and theirdefence of these same languages. (As has been suggested already, thiscontradiction may be a manifestation of an ideological contradictionbetween the hegemonic class culture and counter-hegemonic revolutionaryculture.) On the other hand, it should be noted that the respondents'support for the indigenous languages (a revolutionary-nationalist counter-ideological view) in the answers to questions in Section B is not carriedthrough in their reported use of these languages: this suggests that studentsare not consciously aware of the contradiction and do not recognize theirrole in maintaining the dominance of English (hegemonic culture). This isparticularly true of those who claim that the position of English is justifiedbecause of its international significance. The overwhelming support forthis last position raised the question of how this idea entered the intellectualworld of secondary-school students. It appeared that few, if any, hadconsidered that English is the first language of only a small percentage ofpeople and that many non-anglophone countries cope adequately usingEnglish as a second language without sacrificing their first language. Thisis a crucial point for African socio-linguistics owing to the usual dominanceof the colonial languages in post-colonial polities. It is unlikely, for instance,that such attitudes would be found among students in Asian, SouthAmerican or European countries. We were faced with the question ofwhether those whose answers were categorized as acceptance of thestatus quo had been exposed to the idea that languages are not static andthat the language situation in Zimbabwe can be changed.METHODOLOGICAL CONSIDERATIONSIn this report we attempt to look at some aspects of theory-building asthey pertain to our investigation. We are concerned with clarifying ourmodel of society and our methodological theory so as to improve ourmethodology where possible. The questions we asked in this pilot studyare questions which other researchers in the field have been asking, andthere is a clear indication that some of these questions, especially thoseon attitudes, elicited mere stereotyped responses or cliches, as is evidencedby the contradictions noted above. There is a need, therefore, to design abetter instrument that will enable us to obtain a more reliable answer tothe question of language attitudes in Zimbabwe. It is probably necessaryto realize that the problem is not that of the distinction between professedattitudes and actual behaviour. Goode and Hatt (1952) noted that verbalbehaviour is one kind of behaviour which expresses certain dimensions ofsocial reality which is as real as any other kind. Thus the fact that a groupexpresses an ideal is of great importance for behaviour even when there isnon-conformity with that ideal.It is essential to realize that what people say they do or believe andwhat they actually do or believe may be quite different. Thus themethodological problem is not a simple one. It should be noted, therefore,90 LANGUAGE .ATTITUDES OF SECONDARY-SCHOOL STUDENTSthat the penetration of stereotypes or cliches does not necessarily revealwhat people really do or what they say they will do. Either what they sayor what they do may be a facade. One needs to get behind external,evasive behaviour, of whatever variety, in order to find reliable answers.Our research goal might have to be that of checking certain types ofbehaviour against other types of behaviour through questioning andobservation. There is, perhaps, a need to replace or supplement thequestionnaire method with that of the structured interview. An interviewhas the advantage of giving the researcher an opportunity to elicit moreinformation by probing when the answers are vague, irrelevant or evencontradictory. The interviewer gives the respondent emotional supportand stimulates the expression of deeper levels of thought. A sometimesuseful aid is to be found in what has been termed the 'sieve question'. Thisis a question that sifts out those who should not be answering the questionbecause they do not possess the necessary knowledge or experience. Thesieve question may follow or precede the important question but onlyanswers from those who have the requisite knowledge will be counted.Alternatively, opinions of those with specialized experience, such asteachers, may be compared against those without such specializedexperience, such as students.Another set of devices for penetrating the cliche is the 'assuming'question and the 'adverse' question (often used for delicate matters likesex). These are similar in that they both convey to the respondent that theinterviewer already knows the facts. The advantage of this type of beginningis obvious if one is to move beyond the facts and expressions of conformity.The respondents find the discussion of their attitudes much easier becausethey sense that the interviewer is not going to be shocked. They alsoassume that the interviewer already knows what goes on so that attemptsat evasion will be useless. The common assumption behind such questionsis that the respondent will deny the statement if it is false but if it is truethen a basis for frank discussion has been laid.CONCLUSIONLanguage attitudes have a profound effect on the life and languages ofbilingual people. The goal of language attitude studies is to contribute toour understanding of which languages are positively evaluated, which arelearned, which are used, and which are preferred by bilingual people, allof which have implications for language planning. The data from our studysuggest that there is not necessarily a single set of language attitudes buta complex system of seemingly contradictory positions. These results,although unanticipated, seem to reflect the Zimbabwean socio-historicalexperience. Our pilot study has given us the impetus to embark on thisresearch on a wider scale. The task at hand Is to redesign and refine ourresearch instruments in the light of our prerf/fttnary findings.CYNTHIA MPARUTSA, JULIET THONDHLANA AND N. CRAWHALL 91ReferencesCHIWOME, E. and THONDHLANA, J. 1989 'The teaching of Shona throughthe medium of Shonaand English in high schools and at the Universityof Zimbabwe', Zambezia, XVI, 159-72.COWARD, R. and Ems, J. 1977 Language and Materialism: Developmentsin Semiology and the Theory of the Subject (London, Routledge andKeegan Paul).GOODE, W. J. and HATT, P. K. 1952 Methods in Social Research (Tokyo,McGraw-Hill).HOFMAN, J. E. 1977 'Language attitudes in Rhodesia', in J. Fishman,R. Cooper and A. E. Conrad (eds.), The Spread of English (Rowley MA,Newbury House), 276-301.MCGINLEY, K. 1987 'The future of English in Zimbabwe', World Englishes,XV, 159-64.MAO TSE TUNG [ MAO ZEDONG] 1965 'On contradiction', in Selected Worksof Mao Tse Tung: Volume 1 (Beijing, Foreign Languages Press), 311-46.MPARUTSA, C. 1986 'An Investigation into Diglossia in Harare' (Harare,Dept. of Linguistics, Univ. of Zimbabwe, mimeo.).NGARA, E. 1982 Bilingualism, Language Contact and Planning (Gweru,Mambo Press).92 LANGUAGE ATTITUDES OF SECONDARY-SCHOOL STUDENTSAPPENDIX I: QUESTIONNAIRECircle age group 12-18 19-19 20-25SECTION A1 What language do you consider to be your mother tongue?2 Which radio station do you listen to most often?12 3 43 How often do you read Kwayedza?sometimes always never4 English is largely the medium of instruction in secondary schools.Would you favour a change to Shona/Ndebele? Explain why.5 How many books in English have you read in the last two years?A lot A few None6 How many books in Shona have you read in the last two years?A lot A few None7 Do you think we should have more public signs (e.g. at hospitals,roadsigns) written in Shona? Ndebele?Yes NoSECTION B: DO YOU AGREE WITH THE FOLLOWING STATEMENTS?8 All educated Shona/Ndebele men and women should know somethingabout Shona/Ndebele language and literature.Yes No9 Shona/Ndebele and English are equally important.Yes No10 All educated Shona/Ndebele men and women should be able to speakEnglish.Yes No11 A person who does not speak and write English is not educatedYes No12 Shona/Ndebele is a useless language; there is no reason why it shouldbe taught in schools and university.Yes NoCYNTHIA MPARUTSA, JULIET THONDHLANA AND N. CRAWHALL 93APPENDIX 2: TABLESTable IWHAT LANGUAGE DO YOU CONSIDER TO BE YOUR MOTHER TONGUE?LanguageShonaNdebeleEnglishOther*TOTALResponse (%)831124100* Included in this category are speakers of Nyanja, Chewa, Lori and Tonga.Table IIWHICH RADIO STATION DO YOU LISTEN TO MOST OFTEN?Group A Group B NR GroupStation Response (%) (n = 15) (n = 83) (n = 2)Radio 1 (English language) 1 - 1 -Radio 2 (Shona/Ndebele language) 29 8 21Radio 3 (popular music) 67 5 60 2Radio 4 (education) 1 - 1No answer 2 2 - -Group A is composed of those respondents who answered 'yes' to Question 4.Group B is composed of those respondents who answered 'no' to Question 4.The NR Group is composed of those respondents who did not answer Question 4.ResponseSometimesAlwaysNeverNo answerTable IIIHOW OFTEN DO YOU READ KWAYEDZA*Response (%)6511231Group Ain = 15)645Group B586181NR Groupin'2)11' Kwayedza is a Shona-language newspaper with an Ndebele supplement.Table WENGLISH IS LARGELY THE MEDIUM OF INSTRUCTION IN SECONDARY SCHOOLS.WOULD YOU FAVOUR A CHANGE TO SHONA/NDEBELE? EXPLAIN WHY.ResponseResponse (%)Yes (Group A) 15No (Group B) 83No response (NR Group) 2Unintelligible/no response517ReasonsInternal Internationalcommunication communication8 *19 1 36 §Pro-statusquo11 *Anti-statusquoCM 1 1Irnm«CO* These reasons included: Shona is more understandable; Shona is the mother tongue of many Zimbabweans; and Shona is an official Qlanguage in Zimbabwe. OOt One respondent said that there was enough English used and that the indigenous languages should be promoted, the other said 'we hate >manose' (referrring to the practice of speaking nasally, a commonly ascribed characteristic of children educated at private schools in wealthy -<areas of the city. It is seen as a desire of some Black youths to imitate the speech patterns, and life-style, of privileged White youths). faI1 These explanations included statements that Shona and Ndebele are not sufficiently developed to take on an increased load in the school Qsystem. Certain respondents pointed out that English was well understood by all Zimbabweans. Other respondents stated that young PZimbabweans want to learn English. c/>§ English was seen as the language of international trade and communication by these respondents. °Z# These respondents thought that implementing large-scale curriculum changes was too difficult, or too late. Others said that English was «"not bad', neither was it bad to speak it.CYNTHIA MPARUTSA, JULIET THONDHLANA AND N. CRAWHALL95Table VHOW MANY BOOKS INResponseA lotA fewNoneResponse (%)80182ENGLISH HAVE YOU READTWO YEARS?Group A(n = 15)1131Group B(n = 83)67151IN THE LASTNR GroupCM 1 1Table VIResponseA lotA fewNoneHOW MANY BOOKS IN SHONA HAVE YOUIN THE LAST TWO YEARS?*Response (%)353332Group A(n = 15)843Group B{n = 83)262829READAT? Group(n=2)11* Note: The question should have read "books written in Shona/Ndebele'.Table VIIDO YOU THINK WE SHOULD HAVE MORE PUBLIC SIGNS WRITTENIN SHONA/NDEBELE?Response Response (%)Group A(n = 15)Group B(n = 83)NR Group0 = 2)YesNoNo answer71272155427*2* Of these respondents 21 were mother-tongue Shona, 4 were Ndebele speakers and 2were English speakers. It is interesting to note that 1 of the 4 speakers of other indigenouslanguages answered 'no' to this question.Table VIIISHOULD ALL EDUCATED SHONA/NDEBELE MEN AND WOMEN KNOW SOMETHING ABOUTSHONA/NDEBELE LANGUAGE AND LITERATURE?ImResponseYesGroup AGroup BTOTALNoGroup AGroup BTOTALResponse (%)15809533None/unintelligibleresponse517-ReasonsInternal Internationalcommunication communication819 342Pro-statusquo210-Anti-statusquo_1ITUDESOFSECONDAFY-SCHOOL STUDEI1Note: The NR Group is not Included in this table.Table IXARE SHONA/NDEBELE AND ENGLISH EQUALLY IMPORTANT?ResponseYesGroup AGroup BTOTALNoGroup AGroup BResponse (%)115667425TOTAL 29Did not respondGroup B 2TOTAL2None/unintelligibleresponse41214ReasonsInternal Internationalcommunication communication712172411Pro-statusquo8-Anti-statusquo-23IAMP>3A, JULIE!' THONDHLANA AND N. CFAfote. The NR Group is not included in this table.Table XResponseYesGroup AGroup BTOTALNoGroup AGroup BResponse9758458TOTAL 13Did not respondGroup A 1Group BTOTAL1SHOULD ALL EDUCATED SHONA/NDEBELE MEN AND WOMENBE ABLE TO SPEAK ENGLISH?(%) None/unintelligibleresponse21631-ReasonsInternal Internationalcommunication communication616 3313 31Pro-statusquo110--Anti-statusquo-11-LANGU/iam=1IDES OF 53ECONDARY-SCHOOL STUDENTJNote: The NR Group is not included in this table.Table XIIS A PERSON WHO DOES NOT SPEAK AND WRITE ENGLISH EDUCATED?OZReasonsResponseYesGroup AGroup BTOTALNoGroup AGroup BTOTALResponse (%)31518126779Did not respondGroup AGroup B 1None/unintelligibleresponse13414;TOTAL1InternalcommunicationInternationalcommunication71628Pro-statusquoAnti-statusquo2>=1Zo>>I>Note: The NR Group is not included In this table.8Table XIIIS SHONA/NDEBELE A USELESS LANGUAGE THAT SHOULD NOT BE TAUGHT IN SCHOOLS AND AT UNIVERSITY?ResponseYesGroup AGroup BTOTALNoGroup AGroup BTOTALResponse (%)224138194ReasonsNone/unintelligible Internal International Pro-statusresponse communication communication quo2317Note: The NR Group is not included in this table.1 18 - -19 35Anti-statusquo-210NGUAGE A3'UDES OF SECOI3JiooenTUDEN