Zambezia (1988), XV (i).ESSAY REVIEWA.B.S. CHIGWEDERE'S PRE-COLONIAL HISTORIESOF ZIMBABWE AND AFRICAA QUARTER OF a century ago, in 1962, the study of the pre-colonial history ofAfrica as a serious academic discipline was beginning to get under way. This wasa period of high hopes. Archives virtually untapped by historians were beginningto be used, and it was hoped that the faded documents of European imperialismcould be used to recover the history of the peoples of Africa rather than that oftheir colonizers. Oral traditions were recognized as a legitimate historical source,and researchers armed with tape-recorders were beginning to set out to recoverthe histories of peoples not recorded by observers before the nineteenth century. Itwas understood that Africa's past required a multidisciplinary approach, andspecial stress was laid on the importance of archaeology and linguistics, though itwas also hoped that such disciplines as physical anthropology, serology,palaeobotany and a host of others could be pressed into service. This was theheyday of African nationalism, and Africans and Africanists were largely unitedin the hope that Africa could be given a reliable history reaching as far back as thatof Europe, the continent with which Africa was most frequently compared. HughTrevor-Roper's opinion that this was impossible and, worse, irrelevant was oftencited only to be demolished.1For about ten years after 1962, tremendous strides were made. Articles,books, research reports and whole new journals appeared regularly, andconferences were held that linked the universities of Africa with those of theoutside world, mostly in Western Europe and North America. The giganticCambridge History of Africa and the Unesco General History of Africa werelaunched, and moved slowly towards completion. The chronology of the historyof Africa appeared to be becoming securely established, to the point where by1974 it was claimed that the traditional history of the Interlacustrine region couldbe dated with remarkable precision to as far back as the eleventh century AD.2Unfortunately, the foundations of the whole structure of African pre-colonialhistory were by no means as firmly founded as they looked. New research intoalmost every field not only cast doubt on previously established 'facts', but a muchmore critical approach towards evidence of all kinds was emerging by the early1970s. In history, documents began to be read with much greater care, withinternal and external criticism of texts receiving far more attention.3 Also inhistory, oral traditions were re-examined and found to contain a high content of1 D. P. Henige, Oral Historiography (London, Longman, 1982), 21.2 J. B. Webster (ed). Chronology. Migration and Drought in Interlacustrine Africa (London,Longman, 1979).3 B. Heintze and A. Jones (eds), 'European sources for Sub-Saharan Africa before 1900: Useand abuse', Paideuma, Mitteilungen zur Kullurkunde (1987), XXXIII.8788 CHIGWEDERE'S PRE-COLONIAL HISTORIESmyth, cliche, invention and feedback from written sources.4 Outside history,much the same process was going on: archaeological theories rose and fell,5 andlinguistic reputations were destroyed.6 The journal History in Africa, founded in1974, was devoted to the testing of evidence, often to destruction. The results ofthis new spirit of criticism were often dramatic: the Almoravid conquest of Ghanaof 1076 never happened, Ganda history before 1800 is very dubious, the Jagamay not have existed, and so forth.7 The reduction of so many 'facts' in pre-colonial African history to the status of 'disproven', 'improbable' or 'not proven'probably explains why so many university students up to doctoral level havetended to desert the pre-colonial period for the supposedly secure ground ofmodern African history,8 but in fact pre-colonial African history does survive as adiscipline. If the highest hopes of the 1960s school of historians of Africa have notbeen realized, Trevor-Roper has been proven wrong: Africa does have arecoverable past, but there are limitations on just what can be known aboutcertain aspects in some periods, and some periods in some areas remain resolutelyprehistoric.9However, the 1960s and 1980s as eras of African history have one thing incommon, at least: no one researcher stands supreme. A glance at the index of justone journal, the Journal of African History, shows how true this was and is. Thisis just as true of single African countries. To take an obvious example, my ownhistory of the Shona cited 93 other researchers active in and around Zimbabwe inthe 1960s and 1970s without whose work my own would not have beenpossible.10All the foregoing is necessary if the claims and contributions of AeneasChigwedere to the field of African history are to be seen in their proper context.Chigwedere has so far published three books on history: From Mutapa to Rhodes,4 For example, J. C. Miller (ed.), The African Past Speaks: Essays on Oral Tradition andHistory (Folkestone, Dawson; Hamden CT, Archon, 1980); D. P. Henige, The Chronology of OralTradition: Quest for a Chimera (Oxford, Clarendon, 1974).5 For example, D. W. Phillipson, The Later Prehistory of Eastern and Southern Africa(London, Heinemann, 1977), and African Archaeology (London, Cambridge Univ. Press, 1985).6 J. Vansina, 'Bantu in the crystal ball', History in Africa (1979), VI, 287-333, and (1980), VII,293-325.7 D. C. Conrad and H. J. Fisher, "The conquest that never was: Ghana and the Almoravids,1076', History in Africa (1982), IX, 21 -59, and (1983), X, 53-78; D. P. Henige,' "The Disease ofWriting": Ganda and Nyoro kinglists in a newly literate world', in Miller (ed.), The African PastSpeaks, 240-61; C. C. Wrigley, 'The kinglists of Buganda', History in Africa (1974), I, 129-39;J. C. Miller, 'Requiem for the "Jaga" ', Cahiers d'Etudes africaines (1973), XIII, 121-49; J. K.Thornton, 'A resurrection for the Jaga', Cahiers d'Etudes africaines (1978), XVIII, 223-7; J. C.Miller, 'Thanatopsis', Cahiers d'Etudes africaines (1978), XVIII, 229-31; A. Hilton, 'The Jagareconsidered', The Journal of African History (1981), XXII, 191-202.' One would think that modern historians would at least know whether there had been strikes,disturbances, risings and so forth in the Lourenco Marques of the 1940s or 1950s or not. Butapparently they did not: J. Penvenne, 'A luta continua', The International Journal of AfricanHistorical Studies (1985), XVIII, 109-38.9 For an assessment of the status of oral tradition after the wave of criticism of the 1970s, seeJ. Vansina, Oral Tradition as History (Madison, Univ. of Wisconsin Press, 1985).10 D. N. Beach, The Shona and Zimbabwe 900-1850 (Gwelo, Mambo Press, 1980), 399-412.D. N BEACH 89Birth of Bantu Africa, and The Karanga Empire.11 These should not be read orreviewed in isolation from each other, because each of the later books tends toassume a knowledge of its predecessor, while the first is clearly influenced by ideasformulated but not published until later.ŁAfrica has produced some distinguished academic historians of the pre-colonial period, writing with authority on whole peoples of the past; it has alsoproduced amateurs, working outside the academic mainstream, who apply anatural talent and common sense to the history of a small group that is easilyaccessible and whose traditions can be used as part of a much greater mosaic ofresearch. Unfortunately, Chigwedere's work belongs to neither category, and getsthe worst of both worlds, being neither of an academic standard nor content toconfine itself to targets of research appropriate to its author's circumstances.Chigwedere began his History Honours degree in 1962, and graduated at theend of 1964.12 Thus he began by being exposed to the African history of the 1960sat its most optimistic point, but in 1965 he began teaching, and, except for a briefperiod in 1970-1, lost contact with the world of professional academic history, atleast as far as effective supervision was concerned.13 He gradually rose to becomeHeadmaster of, firstly, Goromonzi and then Marondera secondary schools, andbegan publishing his work in 1980.Chigwedere's references show just how far he was out of touch with academicresearch in his field: they are very few indeed, and modern academic works ofhistory are very thinly represented. (One might assume that Chigwedere wasignorant of them, but the first chapter of The Karanga Empire suggests that hewas aware of their existence but could not bring himself to discuss them.) Apartfrom these, there is a mixture of articles from such journals as NADA (no badsource, if used critically), such obsolete works as C. G. Seligman's Races of Africa(published in 1930 but which Chigwedere seems to have thought was a recentwork) and primary and secondary sources from the nineteenthcentury such asLivingstone and Stanley. Chigwedere's own researches into oral traditions are notcited properly at all, and it is virtually impossible to deduce just whom heinterviewed, or, it seems, to examine his interview transcripts. As it is absolutelyessential that it should it be possible to check on oral sources, it follows that for agreat deal of Chigwedere's work there is no proof at all. But the defects ofChigwedere's books derive from more than just limited sources: there arefundamental problems attached to his handling of evidence. These problems arein fact what makes his work interesting, from the point of view of ananthropologist, because they amount to an attempt to reconstruct the history notjust of the Shona but of the whole of Africa according to the principles of localkinship.11 A. B. S. Chigwedere, From Mutapa to Rhodes (Salisbury, Macmillan, 1980), 168 pp.,ZJ4.31; Birth of Bantu Africa ([Harare], Books for Africa, 1982), 141 pp., ZS3.80; The KarangaEmpire (Harare, Books for Africa, 1986), 174 pp., ZS9.78.12 Information from Academic Registrar, University of Zimbabwe, Dec. 1987.13 It was not entirely Chigwedere's fault that he lost contact with academic history, asresignations and deportations from the University College in the mid-1960s made such contactdifficult. But by 1970-1, when Keith Rennie and I read an early draft of Chigwedere's work, it wasclear that if academic criticism were to be applied to it then not just points of detail but Chigwedere'sentire methodology would need revision. This Chigwedere would not accept, and he went on his wayalone.90 CHIGWEDERES PRE-COLONIAL HISTORIESThe Shona use the totem (mutupd) to identify individuals, especially inquestions of kinship and inheritance. The mutupo, inherited from the father,relates to a very limited number of animals or parts of the body, and in anydistrict, although there may be several totems represented, a few are likely to bepreponderant. Thus, in Chigwedere's home district of Hwedza, the soko (vervetmonkey), shava (eland) and moyo (heart) totems are probably the most common,or at least the most conspicuous, because they relate to the dominant lineages inthe area. Moreover, members of these lineages can trace their ancestry back to anindividual first ancestor: thus virtually all, if not all, soko/mhondizvo-vudzijenapeople from Hwedza are descended from Dendenyore of the early eighteenthcentury, and similarly the shava/museyamwa and moyo/mhondizvo groups goback to various rulers holding the titles of Mbiru and Changamire in the sameperiod. Similar situations, but with different totems and ancestors, exist indifferent parts of the country. But these genealogies can be misleading: byconcentrating on the patrilineage and omitting many brothers and uncles in thatpatrilineage, as well as omitting virtually all wives' ancestors, they give anexaggeratedly pyramidical structure and the impression of a very smallpopulation in the past. In addition, there are natural limits in time beyond whichtraditions do not go: without using documents, and by using a realisticgenerational dating system, it is difficult to establish the existence of mostdynasties any earlier than 1700, although there are some significant exceptions.In view of this and of the fact in a local situation most people of the sametotem probably are related within the last three centuries, it is not surprising thatsome earlier local historians have tried to show that all, or almost all, people of theShona area with same totem come ultimately from the same ancestor. AronMarwodzi was doing this in the 1920s,14 and the missionary Harald von Sicardtook the process even further, building elaborate structures on chance resemb-lances of names and the assumption that successive waves of people, each wavewith a distinctive totem, had occupied the country, producing a sort of layeredeffect.15 Chigwedere has been influenced by these concepts, and especially by theideas of von Sicard. Yet the idea that the same totems are necessarily connected isfundamentally unsound: the same totems are found not only among completelyunrelated non-Bantu peoples in Africa, but on other continents such as Europeand North America as well. Moreover, to prove a link between two dynasties onemust have reasonably coherent genealogies that go back to the same ancestor,proven by checkable sources. It is here, as will be shown, that Chigwedere'smethodology falls down.The best of Chigwedere's three history books is the first, From Mutapa toRhodes: although it ultimately leads to an unrealistic and unproven structurereaching back to the remote past, parts of it, though not supported by checkableevidence, do correspond to the picture given by the available evidence. Thisapplies to the post-1700 period. Chigwedere identifies most of the main dynastiesof the period; and, although I would dispute some of his linkages, the overallpattern is fairly coherent. What Chigwedere does not do, however, is to go intothe detailed history of each dynasty. This is an enormous task, but ultimately if14 D. N. Beach, 'The Rozvi in search of their past', History in Africa (1983), X, 16-18.15 For a list of von Sicard's more important articles, see Beach, The Shona and Zimbabwe, AW-D. N. BEACH 91any sense is to be made of Shona dynastic history then it must be undertaken,because it is out of these basic building blocks that a structure Š or rather, severalstructures of connected dynasties Š can be identified. It is becoming increasinglyclear that many traditions have been considerably revised in this century andbefore, and very careful assessment of the evidence is necessary before a history ofa dynasty can be arrived at.It is in the pre-1700 period that Chigwedere's reconstruction runs intoincreasing trouble as it moves back into the remote past. Firstly, he lumps togetherall water-oriented and bird totems into a single group and follows von Sicard inthe assumption that the users of these totems represent a very early 'layer' ofsettlement in this country, before about AD 1000. This simply is not supported bythe evidence. For example, by following von Sicard's misreading of a NativeDepartment note on the Matibi mbedzi dynasty of the south," he ignores theevidence that Matibi's dynasty was preceded by a nzou dynasty as recently as theeighteenth century. Similarly, the dziva Ngowa had been in the modern Chiviregion only from the eighteenth century, not the tenth, while the neighbouringshiri people of Zvishavane were even more recent immigrants. Their genealogiessimply do not go back to the remote past. But worse follows: in trying to provethat the soko Mbire, by a coincidence (?) his own group, were the 'core' group ofmost Shona dynasties, Chigwedere builds on the unreliable structure of dynastiesassembled by Donald Abraham in the early 1960s.17 Although he castigatesAbraham for exaggerating the importance of 'Mutota' and the Mutapa state, withwhich few would now disagree, Chigwedere tends to give the main periodcovered by Abraham's daunting articles (c. 1400-1800) a wide berth. Possibly thiswas because ne was unable to read the Portuguese sources that are so vital formost of that period. But he also criticizes Abraham for underestimating the lengthof Shona traditional history before the fourteenth century, when even Abraham'sambitious structure was beginning to run short of 'evidence'. In short, bymisreading the archaeological evidence and taking separate names of figures froma variety of unconnected sources, Chigwedere builds a superstructure on top ofAbraham's structure that goes back from the fourteenth century to the early ninthin about six generations (!) to arrive at a 'first ancestor' named 'Mambiri' in theEthiopia-Kenya region in about AD 800.1816 H. von Sicard, 'The origin of some of the tribes in the Belingwe Reserve, 9: The Pfumbi underMacetu and Mketi', NADA (1952), XXIX, 43. The ' Matibi' referred to, who died c.1900, was thefirst of his lineage to be appointed 'Chief by the colonial government and the first to use 'Matibi' as ahereditary title. Prior to that his ancestors used the Venda system of personal names instead ofhereditary titles, but they went back only two generations to Mafukanoro, who immigrated fromVenda.17 D. N. Beach, 'The Mutapa dynasty: A comparison of documentary and traditional evidence',History in Africa (1976), III, 1-17.11 There is hardly room in this review for a detailed discussion of the slipshod nature ofChigwedere's methods, but his treatment of this 'first ancestor' will serve as an example:Chigwedere's source is not B. J. M. Foggin, as he thinks (he could not even cite his sources correctly),but Fr J. H. Seed, 'The kinship system of a Bantu tribe', NADA (1932-3), X-XI, 10-11, 65-73,35-56. Seed was making an imaginative guess about the origins of totems, and happened to use thename of his basic unit, a boy named Philip Mambiro, as his imaginary first ancestor from whom theChinamhora soko lineage came. Chigwedere, From Mutapa to Rhodes, 3,19, took 'Mambiro' to be areal person, changing his name to 'Mambiri' to make it look more like 'Mbire'.92 CHIGWEDERE'S PRE-COLONIAL HISTORIESThe 'evidence' for this is thin where it is not non-existent. But there is onesignificant point: Chigwedere claims to have relied upon a spirit medium who issaid to have emerged in the Hwedza district in 1964. This medium was said tohave been possessed by five ancestral mhondoro spirits: Nyahuye, founder of partof the Svosve dynasty to which Chigwedere belongs; Mabwemashava, rulingcirca AD 1000, and his brother Chigwangu Rusvingo of c. 1050; Gumboreshumba,the 'founder of the Rozvi empire'; and the famous Chaminuka, father of thesecond two. (These are Chigwedere's dates and given relationships.) ButChigwedere does not name this remarkable medium. Was he by chanceChigwedere himself? I have approached Chigwedere on this point and receivedno clear answer. .//'Chigwedere was the medium, then certain questions about theorigin of 'evidence' emerge. If not, then there was a quite exceptional mediumoperating in Hwedza for sixteen years who escaped the notice of researchers.In short, From Mutapa to Rhodes moves backwards in time from the realmsof post-1700 history, which is coherent even if little evidence is given, to theremote past and unproven fantasy. Birth of Bantu Africa is, quite simply,historical balderdash. It proposes that most of Africa was originally inhabited by'Bushmen', with a small 'Hamite' population in the Nile valley, and that'commingling' brought about the 'Hottentot' and 'Negro' who then occupied therest of Africa. The specifically 'Bantu' section of the 'Negro' are said to have beguntheir migration from north-east Africa about AD 600. Readers will recognize thisas part of an obsolete and racist theory that runs back through Seligman in the1920s to the nineteenth century. The research on Africa that had taken placebefore 1962 had already killed it, but the work that has been carried out since thenhas buried it with a stake through its heart. Or so one would assume.In The Karanga Empire, Chigwedere returned to more familiar ground. Hewas also, to a certain extent, going over his older arguments in more, if not entirelyconvincing, detail. While he still clung to his ideas over north-eastern Africanorigin for the Bantu-speakers, he tried to use archaeological evidence, with onlypartial success. This may be because he saw himself as being on the defensive. Hisletters to me suggest that his as late as 1985 he genuinely believed that he haduncovered certain basic truths about the history of Zimbabwe and Africa thatwould lift him to pre-eminence in the field. The history conference at theUniversity of Zimbabwe in 1982, in which academics from Tanzania, Zambia,Malawi, Mozambique and Botswana read his work, did not respond positivelytowards it. So The Karanga Empire is in a way a return to the battlefield. Yet thefaults of the earlier works persist: based on an incomplete and inadequatecommand of the sources, it triumphantly asserts points known long before, whileat the same time trying to prove some very dubious points of historicalethnography. It remains resolutely non-academic: ethnic identity is, very nearly,all. (On one point Chigwedere must be defended: if I understand him correctly, heis as worried by the growth of 'tribalism' in modern Zimbabwe as I. All his workon different divisions of Zimbabweans is supposed to unite, not divide them, byproving long-term links.)Why, the reader might ask, have I written this much about three very amateurhistory books? After all, like all Zimbabweans, Chigwedere is supposed to be ableto believe and publish what he likes. Because I argued with him on televisionD. N. BEACH 93during thirteen hot, sweaty recording sessions in late 1984?19 No. Because, for allits faults, his work is interesting from an anthropological view point because itshows what happens when one tries to apply a local kinship concept to the wholeof Africa? Partly. Because, for all the unacademic quality of his work, he achieveda certain prominence in the Ministry of Education and the National Museumsand Monuments? Partly. But my main reason is that it has been clear for sometime that the product of the Zimbabwean school system arriving at the Universityto read history seem to be under the impression that a book is a book, and that allbooks are equal and appropriate. The idea that ongoing research has disprovedmuch of what they were taught at school, or that not all books are equally reliable,appears to fill them with horror. The origins of this attitude are hard to trace. Theymay lie with lazy teachers who fail to keep up with recent research Š or they maylie with the failure of the University to make sure that new research findingspenetrate the Ministry of Education. Chigwedere's books make a 'tale without ahead' (Soko risina musoro) if it is understood that the 'head' of history isrepresented by the academic researchers and that the 'tail' lies in the classroom.The problem is to keep the two connected.University of Zimbabwe D. N. BEACH" Only ten were broadcast, in 1985.