Zambezia (2002), XXIX (i).THE SHONA WRITING SYSTEM: AN ANALYSIS OF ITSPROBLEMS AND POSSIBLE SOLUTIONSWISEMAN MAGWAMidlands State UniversityAbstractThis article critically examines the Shona writing system, starting with anevaluation of the early efforts by the missionaries, Doke and the ShonaLanguage Committee to establish a common writing system. The articleseeks to identify weaknesses in Shona orthography and suggests ways ofimproving it. It concludes that the present writing system needs to be revisedin order for it to be a tool for national development and that languageexperts need to work to improve the current alphabet, word division, andspelling so that the Shona language becomes a national asset in the newmillennium.INTRODUCTIONThe Shona language is spoken in five dialectal clusters, which areChiKaranga, ChiManyika, ChiNdau, ChiZezuru, and ChiKorekore.ChiKaranga is spoken in areas surrounding Masvingo town in the south,ChiManyika in areas around Mutare in the east, ChiNdau in Chipinge inthe south east, ChiKorekore in Mutoko in the north east, and Chizezuru innorth-central Zimbabwe. Outside Zimbabwe, the language is spoken inBotswana, Zambia, South Africa, and Mozambique. It should be notedthat Shona dialects are mutually intelligible and are, thus, dialects of thesame language. Indeed, 80 to 90 per cent of the vocabulary is common tothe whole area of Shona speaking communities.Despite this high degree of intelligibility in the spoken form, however,the written language is very problematic in a number of ways. For instance,there is no real standard way of writing the Shona language, as words insome dialects cannot be spelt using the current orthography and thereare numerous problems with word division and spelling. Indeed, thewriting system (alphabet, spelling and word division) cannot adequatelyrepresent the spoken form of the Shona language and is linguisticallyconstricting, making it difficult for speaker-writers to write the spokenlanguage correctly. Speakers of different dialects of ChiShona experiencedifficulties in spelling and word division because of a defective writingsystem, adopted in 1967, and known as the "standard" orthography thatdoes not cater for dialectal variations and also does not allow speaker-writers to write their language the way they speak it.2 THE SHONA WRITING SYSTEMFor example, the spelling system in use bears little resemblance tothe spoken language. An example is that, while the standard spelling ofthe verb "to eat" is <dya>, in Korekore, it is <dja>, while in Zezuru, Ndau,and Karanga, it is >djwa<; >dhla>; and >hla<, respectively. The writingsystem is, therefore, far from being standard, a situation that causesproblems to students studying Shona in the country's education system.It is clear, therefore, that the system is in need of urgent revision.HISTORICAL OVERVIEW OF THE SHONA WRITING SYSTEMEfforts to establish a common writing system can be divided into threedistinct periods, namely: the period of early missionary efforts (1890-1928); the period of Doke's New Orthography (1932-1954), and the periodof revisions (1955-2000). According to Chimhundu (1992, 97), attempts towrite ChiShona were begun in different places mostly by missionarieswhose linguistic backgrounds were Indo-European languages and whowere working independently of each other. Thus, each dialectal clusterevolved its own writing system, often, different from those of the otherdialects.Ł Concerns about the differences in the various dialectal orthographiesled to a vigorous debate at the 1928 Missionary Conference whetherefforts should be made to develop a single standard Shona writing systemor whether there should be two writing systems in Mashonaland,representing the Zezuru and Karanga dialects, respectively. After failingto resolve the issue, the Conference issued the following statement:This conference finds itself unable to decide at present between thealternative of standardising two dialects for Mashonaland viz.ChiKaranga and ChiZezuru or of standardising a unified language builton all the four existing dialects. We would respectively request thegovernment to approach the International Institute of African Languagesand Cultures with a view to obtaining a suitable expert to investigateand advise upon the matter (Doke, 1931, 5).The result was the appointment of C. M. Doke, then Professor ofBantu languages at the University of Witwatersrand, to undertake thetask, with the assistance of three priests, namely, Revs. Barnes, Burbridge,and Louw.As noted, Doke was to study the language situation in the countryand advise the government on the way forward. At the end of his tour,Doke made twelve recommendations pertaining to the alphabet, worddivision, grammatical standardisation, choice of vocabulary, and thedevelopment of literature. Doke recommended that a unified grammar bestandardised based on ChiKaranga and ChiZezuru dialects, that theconjunctive method of word division be used in Shona writing, and thatW. MAGWAthe orthography be unified based on the principles set out by theInternational Institute of African Languages and Cultures. Consequently,Shona orthography was developed on the principle of one sound, onesymbol (Doke, 1931, 83), and the following alphabet was recommended:<a, b, _, c, d, d, e, f, g, h, i, j, k, m, n, _, o, p, r, s, _, ?, t, u, v, ?, w, x, y, z, ?,?> (Fortune, 1972, 17).Professor Doke's contribution to the development of a writing systemamong the Shona was enormous. The government accepted Doke'sorthography, but many, including the Native Affairs Department, the PostOffice, and the Press, for unclear reasons, opposed it and never adoptedit.Meanwhile, there was an increasing call for the introduction of theRoman alphabet to simplify printing, reading, and writing. Consequently,an orthography committee was set up and made its recommendations ina pamphlet entitled A Guide to Standard Shona Spelling in 1955. Among itsrecommendations was that the special characters introduced by Dokeshould be abolished, while only Roman letters should be used. Therecommended alphabet was much simpler and contained only thefollowing letters and diagraphs:<a, b, ch, d, e, f, g, h, i, j, k, m, n, ny, ng, o, p, r, s, sh, sv, t, u, v, w, y, z, zh,zv> (Magwa, 1999, 18).There was no longer a distinction between <b> and <_>, >d< and <?>,>_< and <ng>, <v> and <?>. The symbols <_> and <z> were replaced bydiagraphs <sv> and <zv>, respectively. The symbol <x>, a sound found inChiKaranga and ChiNdau was removed from the alphabet.The proposed alphabet was approved by government for general useand by the whole Shona writing community. However, the revisions weredefective because, in writing, they did not reflect the distinctions made inspeaking and so could be ambiguous. Although the new orthography wasclearly retrogressive, it was received quite happily by the Shona peopleon the somewhat dubious grounds that the orthography was beingdesigned for Shona speakers who would know from the context which ofthe two phonemes the ambiguous letter stood for. Dissatisfaction withthis aspect of the 1955 orthography led to further revision in 1967.In 1967, the 1955 orthography was brought into line with that of 1932by a language committee whose membership included the Secretary forEducation and a professor of African languages at the University ofRhodesia. The committee adopted all the phonetic distinctions made byDoke but pointed out that this did not involve the addition of new orTHE SHONA WRITING SYSTEMspecial symbols to the alphabet, which remained based on the Romanalphabet. The committee's objective was to make the spelling clearerand easier to understand. The Minister of Education approved theserecommendations and the following alphabet was adopted:<a, b, bh, ch, d, dh, e, f, g, h, i, j, k, m, mh, n, nh, ny, n', o, p, r, s, sh, sv, t,u, v, vh, w, y, z, zh, zv> (Fortune, 1972, 21).This system of writing, which, unfortunately, makes it difficult forShona speaker-writers to write their language correctly, is what is still inuse today.PROBLEMS WITH THE CURRENT ORTHOGRAPHYFrom 1967 onwards, speakers of different Shona dialects wereexperiencing certain difficulties arising from the defective alphabet andthe spelling and word division system. The current orthography islinguistically constricting in a number of ways. For example, the standardalphabet does not have symbols representing the sounds [1] and [x],which are found in ChiKaranga, ChiNdau, and ChiKoreore dialects. Forexample:ChiNdau Š muhlobo (type)Š kahle (good)ChiKaranga Š pxere (youngsters)Š maxeu (sweet beer)ChiKorekore Š xumbudzi (young goats)Š xurume (men)The speakers of these dialects are, thus, compelled to write in amanner that is different from the way they speak because of theinadequacy of the standard alphabet. Under the standard spelling, writersare compelled to write as follows:MuhoroUswahumbudzimutowopwereinstead ofinstead ofinstead ofinstead ofinstead ofmuxorouxwaxumbudzimuhlobopxereAnother problem with the standard orthography is the use of thebreathy voice, which is very common in Shona dialects but which is notfully represented in the standard spelling. Apart from <bh, dh, mh, nh,vh, zh, hw>, the symbol <h> should also accompany <g> and <r> toW. MAGWA 5represent a breathy voice. In the following example, for instance, theletter >g< is pronounced differently:Ngoro (cart)Nganunu (gun)Gokora (elbow)Goridhe (gold)The present alphabet does not distinguish between the differentpronunciations of the letter <g> in the above examples and neither doesit distinguish between:<r> in roro (fruit)<r> in rori (lorry)<nd> in ndiro (plate)<nd> in ndari (beer for sale)<mb> in mbada (leopard)<mb> in mbaura (brazier).ChiNdau, in particular, has several sound combinations that are notrepresented in the present writing system. For example:[hi] Š hlobo (type)[dhl] Š kudhla (to eat)[nth] Š unthu (humanhood)[nt] Š ntunzi (fly)[th] Š thokosa (talk)[thw] Š kuthwa (to pound i.e. grain)[kh] Š khamba (leopard)The current orthography is so inadequate that people with certainNdau names find it almost impossible to write them, using the standardspelling since some Ndau sound combinations are not part of the standardalphabet. Examples are:Dhliwayo (dhlMlambo (ml)Mhlanga (mhl)In ChiManyika, the sound combination [psw] as in >pswaira< (sweep)is not represented in the alphabet and thus cannot be put in written form.The Korekore people make use of the [dj] sound combination, whichagain cannot be spelt using the current orthography, thus, instead ofwriting <tedja> (we eat), they are compelled to write <tedya>, simply6 THE SHONA WRITING SYSTEMbecause the <dj> diagraph is not part of the standard alphabet. TheKaranga, too, have certain sound combinations which have been excludedby the standard alphabet. Examples are:Pxere (child)Xarani (thread)Ibge (stone)Speakers of the Zezuru dialect also have some sound combinations,which were rejected by the standard writing system, for example:Nghanunu (gun)Mbhaura (brazier)Ndhari (beer for sale)Rhori (lorry)Ghandanga (savage)According to Fortune (1972, 25), the Shona spelling system is purelyconventional and bears very little relation to what people speak. Thewriting system does not represent what it said in most dialects butpurely provides a common spelling, i.e. kudya (to eat). The aboveconventional spelling bears no relation to the actual utterances sincethere is such variety from one dialect to another in the way these lettercombinations are pronounced. The syllable <dya> is realised as"[dja] Š Chikorekore[dhla] Š ChiNdau[hla] Š ChiKarangaAnother demerit of the current writing system is that the rules ofspelling and word division are all expressed in a foreign language ŠEnglish. Magwa (1999, ix) argues that no language can be taught effectivelyusing another language. This is a serious handicap, which militates againstthe relevancy and appropriateness of the current orthography.According to Chimhundu (1992, 84), the statement of rules thatexplains how to and how not to use the alphabet is over-elaborate,unnecessarily complex, and sometimes inconsistent. There seems to beinconsistency between the statement of rules and their elaboration forthe purposes of implementation. Rule VI, for example, reads:Reduplicated substantive stems of more than two syllables are alwaysseparated by a hyphen e.g. mangwanani ngwanani (Fortune, 1972, 50).However, reduplicated stems of less than two syllables are writtenwithout a hyphen between them, e.g. magwanangwana. Complex nominalW. MAGWA 7instructions derived from ideophones are written in two parts without ahyphen, e.g. chamusvetu musvetu.The above examples clearly show how difficult and confusing theword division rules are to the ordinary speaker-writer. One ends up notknowing when to use and when not to use the hyphen, given the fact thatthese rules are in a foreign language. The rules are too complicated formost Shona speakers. The restrictions are too many, makingimplementation of the rules very difficult. Furthermore, there is nolinguistic justification for writing the word mesomeso as one word, whilewriting kare kare as two words.The Shona spelling system restricts variations in speech andsupplementary rules have been put forward to restrict a speaker-writer'soptions. The rule makers contrived to make standard forms those optionsthey considered to be used by the majority of speakers. Thus, Class 2prefix / V / is to be used although the Manyika and Korekore speakers use/ W / e.g.vanhu (people) Š Karanga/Zezuruwanhu (people) Š Manyika/KorekoreThe Karanga speakers have problems when they are required towrite the enclitic <wo> when, in fact, they pronounce it as <vo> or whenthey are discouraged from using <sh, zh> as in <shuro, shanga, zhara,zhira> and told to use <tsuro, tsanga, nzara, nzira> (Chimhundu, 1972,85).Another weakness of the Shona spelling system is that it does notaccommodate sounds that have come into the language as adoptives ordownloads, such as:Thimomita (thermometer)Thiyori (theory),Thisisi (thesis).Although in 1982, the Ministry of Education removed theserestrictions, the prevailing attitude in the same ministry and its institutionsis that the 1967 orthography rules still apply and these may only berelaxed during examinations. In practice, all those forms that do notconform to the rules set out in 1967 are discouraged. According toChimhundu (1972,86),removal of restrictions by the ministerial directive of 1982 is privilegedinformation for use only by examiners... and that it was never intendedthat teachers should allow their students literally to write as theyspeak.8 THE SHONA WRITING SYSTEMAll these problems point to the fact that the Shona writing system isfar from adequate and that there is a need to cater for dialectal variationsin the orthography if the language is to be written correctly by all Shonaspeakers.THE WAY FORWARDIt is contended here that the Shona orthography must be expanded tocater for the needs of different Shona dialects so that people can writetheir language as they speak it. For the orthography to improve, theShona Language Committee must prescribe a new standard system ofwriting, which will permit people who speak dialects of ChiShona(ChiKaranga, ChiNdau, ChiManyika, ChiZezuru, and ChiKorekore) to writein the same way, while still allowing for variations in choice of vocabulary.The first step that needs to be taken is to revise the 1967 Shonaalphabet, which is the source of many orthography problems. Thediagraphs <bh, ch, dh, mh, nh, ny, sh, sv, vh, zh, zv> should be removedfrom the Shona alphabet. There is no literary justification to have, forexample, diagraph <sv> in the alphabet, while there is no <sw> in thealphabet. Similarly, it is not justifiable to have <bh, ch, dh, mh, nh, sh,vh, zh> in the alphabet but fail to include diagraphs <gh> and <rh> in thesame alphabet.The symbols in the 1967 alphabet should be reduced from 34 to only26 letters. All symbols of the Roman alphabet, with the exception of <q>,should make up the new alphabet. The alphabet should consist of onlysingle letters. Different dialect speakers could then combine these singleletters to form acceptable sound combinations for each particular dialect.Such an all-inclusive alphabet would be more acceptable to all Shonaspeakers because they will be able to write their language the way it isspoken. It is proposed, therefore, that the new alphabet should be asfollows:<a, b, c, d, e, f, g, h, i, j, k, I, m, n, n', o, p, r, t, u, v, w, x, y, z>In addition, a set of recommended diagraphs and trigraphs should beprovided to guide speaker-writers to spell Shona words correctly andconsistently. These diagraphs and trigraphs should be used as the basisfor building syllables, morphemes, and words in the different Shonadialects. The sound combinations to accompany the alphabet should beas follows:i) Š <bh> (bhotoro) <ch> (chikoro)<dh> (dhora) <mbh> (mbhaura)ii) [w] -iii) [V] -iv) [z] -v) [y] -vi) [1] -vii) [x] Šviii) 01 Šix) [f] -W. MAGWA<mh> (mhuru)<nh> (nhoro)<gh> (ghoridhe)<kh> (mukhumbi)<rh> (rhori)<vh> (vhiri)<th> (kuthokoza)<bw> (bwato)<dw> (-budwa)<dyw> (dywaga)<gw> (-gwabvura)<jw> (-jwanya)<mw> (mwena)<mhw> (-fumhwa)<nw> (-nwiwa)<ndw> (-kandwa)<nyw> (-menywa)<psw> (-pswaira)<sw> (-swedera)<svw> (-tesvwa)<thw> (-kuthwa)<tyw> (-tywakatira)<zhw> (zhwerere)<bv> (bveni)<mv> (mvura)<zv> (zvigaro)<dzv> (dzvinyu)<dz> (dzimba)<dy> (-dyunga)<ny> (nyama)<hl> (hlobo)<ndhl> (ndhlandhlamo)<ml> (Mlambo)<gx> (kugxoka)<dj> (tedja)<pf> (pfuma)<ndh> (ndhari)<nth> (anthu)<ngh> (nghanunu)<ph> (kuphanga)<sh> (shumba)<hw> (hwahwa)<chw) (-gochwa)<dzw) (-dzwanya)<fw> (-gofwa)<hw> (hwahwa)<kw> (-kwetera)<mbw> (mbwende)<nhw) (-denhwa)<ngw> (ngwena)<nzw> (nzwisiso)<pw> (-pwatika)<rw> (rwatata)<shw> (shwiro)<tw> (-twasudza)<tsw> (tswanda)<zvw> (-rozvwa)<zw> (-zuzwa)<vh> (-vhara)<sv> (svondo)<tsv> (-tsvaira)<nzv> (nzvidya)<nz>(nzou)<ndy> (ndyire)<ty> (tyava)<dhl> (Dhliwayo)<mhl> (Mhlanga)<px> (pxere)10 THE SHONA WRITING SYSTEMx) Is] ~ <ts> (tsamba)xi) [b] Š <mfc» (mbuya)vii) [d] Š <nd> (ndege)xiii) [g] Š <ng> (ngoma) <bg> (vobgo)These diagraphs and trigraphs will enable the writing system to caterfor variations in style and vocabulary. All the five dialects will beadequately represented by these sound combinations.It is further recommended that there should be no tone marking inthe writing system to help reduce differences found in the spoken formwhere tone patterns in actual speech may vary considerably from dialectto dialect. In addition, all the rules that govern word division and spellingshould be written in Shona to enable the ordinary speaker to interpretthem with ease as is suggested in Magwa's (1999) publication ManyorerwoEchiShona.CONCLUSIONThe article has highlighted the problems that Shona writers face with theuse of the existing orthography, namely that they are not always able towrite what they speak. It is not, however, being suggested here that thereshould be a uniform standard set for the spoken language as this wouldbe an unrealistic and undesirable goal which "has never been and neverwill be realised anywhere in the world in respect of any natural language"(Chimhundu, 1992, 87). What has been strongly advocated in the articleis the desirability and necessity of developing a common alphabet, spellingand word division system that will make it possible for writers to writewhat they speak and that this task should be undertaken as a matter ofurgency in the interests of promoting the written Shona language.ReferencesBARNES, B. H. (1932) A Vocabulary of the Dialects ofMashonaland (London,Sheldon Press).CHIMHUNDU, H. (1992) "Early missionaries and the ethnoliguistic factorduring the 'Invention of Tribalism' in Zimbabwe", Journal of AfricanHistory, 33, 87-109.Š (1992) "Standard Shona: Myth and reality?" in Crawhill, N. T. (ed.)Democratically Speaking: International Perspectives in LanguagePlanning (Salt River, National Language Project).Š (1993) "The status of African languages in Zimbabwe", Sapem(October), 57-59.W. MAGWA -, AŠ (1993) "The vernacularisation of African languages after independence"in Diogenes, 4 (i), 35-42.Š (1997) "Language standardisation without policy or planning-Zimbabwe as a case study", in Royneland, E. (ed.). Language Contactand Language Conflict (Oslo, Volde College).DOKE, C. M. (1931) "The problem of word-division in Bantu", OccasionalPaper No. 2, Department of Native Development.FORTUNE, G. (1969) "75 Years of writing in Shona", Zambezia.Š (1972) A Guide to Shona Spelling (Harare, Longman).HANNAN, M. (1959) Standard Shona Dictionary (London, College Press).KAHARI, G, P. (1990) The Rise of the Shona Novel (Gweru, Mambo Press).MAGWA, W. (1999) Manyorerwo eChishona: Bhuku Rinotsanangura MitemoYokunyora Mutauro Wedu (Gweru, Mambo Press).MUKANCANWI, K. G. (1974) "A description of Shona spelling" in AfricanLanguages/Langue Africaine, 1, 225-258.MUTOKONYI, F. R. (1981) Ziva Mutauro (Gweru, Mambo Press).