Zambezia (1990), XVII (i).ESSAY REVIEWFAMINE IN ZIMBABWEAFTER TWO MONTHS of primary research in the National Archives of Zimbabwein Harare Iliffe has written a fascinating book on the changing nature offamine in colonial and pre-colonial Zimbabwe.' Although Iliffe modestlycautions the reader as to the finality of his analysis (due to the 'briefperiod of research and an inadequate knowledge of Zimbabwe'), the bookis, nevertheless, a highly valuable and long overdue contribution to Zimba-bwean historiography.The major arguments of this attractively written and easily compre-hensible publication are: pre-colonial famines did not normally causemass starvation and were not responsible for the low population before1900. Under colonialism food shortage changed its character from faminewhich killed directly (though few people) to structural malnutrition ofthe poor.Iliffe's initial motivation to research into the history of famine wasderived from his interest in historical demography. In trying to explainwhy the population in pre-colonial Zimbabwe was so low and why it roseso fast after the onset of colonialism he examines a hypothesis suggestedin the model of demographic transition. The model posits that faminemortality was a key factor in the population history of pre-colonial andcolonial Zimbabwe and suggests that the cessation of high (pre-colonial)famine mortality brought about by colonial rule could have caused therapid population growth after 1900. But the empirical evidence that Iliffefound does not Š in his view Š support this hypothesis. He argues thatalthough famine ceased to kill directly after about the mid-1920s, faminemortality was not the main constraint for population growth in pre-colonialtimes. Thus the control of famine mortality during the colonial periodcannot explain Zimbabwe's extraordinary population growth since 1900.Hence the major question raised in Iliffe's book remains unanswered.In dealing with the main problem of historical demography outlinedabove the book has a rather unhelpful focus. Whereas the answer to thequestion rests as much with an analysis of pre-colonial as with colonialhistory only one out of the thirteen chapters discusses the pre-colonialperiod; and the analysis of pre-colonial famines in that chapter appearsproblematic and ambiguous. On the one hand Iliffe suggests that 'thepoint is not to deny that serious famine, and perhaps serious faminemortality, occurred in pre-colonial Zimbabwe; it is rather to stress thatwe have almost no evidence of it' (p. 17). But the main thrust of hisargument centres around the assumption that the lack of evidence onfamine mortality reflects a lack of widespread famine-caused starvationitself. Droughts seem to have occurred frequently before 1890, he argues,but usually only a few people died from them. This was due to the fact1 J. Iliffe, Famine in Zimbabwe 1890-1960 (Gweru. Mambo Press 1990), 142 pp., Z$ 16,50,ISBN 0-86922^59-x.101102 ESSAY REVIEWthat the agro-pastoralists of pre-colonial Zimbabwe had achieved controlof famine mortality because of ecological reasons and (more importantly)by a remarkable variety of technical and socio-economic means andstrategies. Only in association with violence (and mostly in the arid lowveld)droughts have turned into severe famines: The peoples of Zimbabwesuffered recurrent scarcity but normally prevented it from causingnumerous deaths. 'Famines that killed' appear to have occurred whenviolence intensified scarcity' (p. 111).The main body of the book presents an analysis of the nine majorcolonial famines between 1896 and 1960. Within this period Iliffe identifiesa historical process of change in the character of famines effecting theAfrican population of colonial Zimbabwe. Prior to 1922 the pre-colonial,'traditional' type of famine prevailed. The famines of 1896/7, 1903, 1912and 1916 were traditional' (with 'the leading actors from an older world'(p. 31)) in terms of causation, extent and effects and also crisis management.With the exception of the post-Chimurenga famine of 1896-7 these famineswere caused by drought and the regions least exposed to early colonialismwere the main sufferers from the failure of the rains. These faminesthreatened but did not cause deaths from starvation in great numbersowing to the continued operation of pre-colonial famine-survival strategiesand (to a small but increasing degree) to famine relief by the colonial state.It was these strategies of coping with famines which were firstlyaffected by colonial changes. Already the great drought of 1912 Š althoughit was met mainly by traditional famine management techniques Š carriedtransitional characteristics. Here government drought-relief operations,as well as indirect entitlements to food2 through the colonial economy(wage labour, cattle sales) became important elements in coping with theenvironmental stress. But it was the famine of 1922 which incorporatedmost elements of transition from pre-colonial to capitalist famines. Thisfamine was the last one which killed a significant number of people directly.But in contrast to the previous famines it was not confined to peripheralareas or to Southern Mashonaland; neither was it caused by a successionof bad harvests; it was caused by one single catastrophic season. Althoughcolonial land apportionment was still not a major causative factor, theseverity of the famine was linked to developments in the capitalist economy:as a result of the Depression following the First World War cattle pricesdropped drastically and Africans were unable to secure exchangeentitlements to food by cattle sales. In terms of crisis management therewas also an important shift. In the 1922 famine government drought reliefbecame as important (if not more important) as traditional techniques asthe main check on famine mortality. Also in terms of famine-inducedlabour migration the 1922 famine marked a turning point in a longerhistorical process. For the first time a famine caused a marked upsurge inlabour migration.By the early 1930s colonial capitalism was dominant and hadprofoundly changed the character of famines in Southern Rhodesia. Fatal2 The concept of 'food entitlements' has been developed by A. Sen. See his Poverty andFamines: An Essay on Entitlement and Deprivation (Oxford, Clarendon, 1981).W. DOPCKE 103famine gave way to non-fatal food scarcity which was now geographicallyconcentrated in the areas of most intense White settlement (Matabeleland).Mulnutrition now affected predominantly the poor and socially weakmembers of African society. The causes of the food scarcities of 1933,1942and 1947 are closely linked to the impact of land alienation, evictions, theovercrowding of the Reserves and declining agricultural per capitaproductivity. Their repercussions were also reflected in the growing socialdifferentiation in African society. Iliffe argues that 'whereas [prior to the1930s] both rich and poor had fasted and feasted, now food was regularlyavailable to those who could afford it and regularly scarce for those whocould not' (p. 79). Until the late 1950s mature colonial capitalism replacedtraditional famine control and also government famine relief almostcompletely with exchange-related indirect entitlements to food andprovided for the necessary infrastructure and markets. For a limitedhistorical period most African families earned sufficient income from cattlesales and migrant labour to buy enough food to avoid starvation but notalways Š in the case of the poor Š to eliminate malnutrition.After the end of the prosperous 1950s the history of food scarcity inZimbabwe entered a new phase. Increased social differentiation and themarginalization of the rural poor during the period of economic prosperitythat followed the Second World War led to a renewed dependence ongovernment famine relief in the disastrous drought of 1960 which wasparalleled by the onset of a long-term structural crisis of colonialism.African agriculture was now clearly undermined by the effects of landapportionment. Coupled with this 1960 saw the onset of stagnation of thecolonial economy which excluded sections of the African population fromacquiring sufficient indirect entitlements to food. The pattern of foodscarcity established in 1960 is still prevalent today. Endemic malnutritionof the rural poor and socially vulnerable and a dependence on governmentfamine relief in times of drought has replaced earlier patterns of faminemortality and famine crisis management.Iliffe presents his book to the reader as a call for further research. Hehimself repeatedly points to gaps in his factual evidence and to the resultantvagueness of his interpretations. The greatest merits of the publication lieŠ I think Š in its analysis of the changing nature of famine and foodshortage during colonial rule and thus indirectly in its contribution to amore thorough understanding of colonialism in Zimbabwe. One coulddisagree with certain elements in Iliffe's interpretation but his centralarguments on colonial famines are entirely convincing to me. 1 would, forinstance, argue that real wages did fall during the Great Depression of the1930s and that the Maize Control Board operated in a manner slightlydifferent from that described by Iliffe. I would put more emphasis onAfrican strategies to earn indirect entitlements to food and less on thecapacity of colonial capitalism to create a network of exchange andtransport. The declining food production in the African Reserves has to beseen Š in my opinion Š much more in the context of increasing holdingsand rising prices of cattle.More important are the limitations arising from the nature of thesources used by Iliffe (of which he is very well aware). When dealing with104 ESSAY REVIEWthe colonial period the book relies almost exclusively on colonial records,and consequently the administrative perspective and perception of famineunderlies the analysis. The sources to an extent determine the themesdiscussed (for instance, the bias towards writing about famine relief) andexclude central facets of famine. Social and gender-related processes(within African society) generating differential entitlement to food as wellas accompanying food shortages and famines could be dealt with muchmore thoroughly.3 Also the geographical frame of analysis (the smallestunit being the District) set by the sources appears problematic. Forinstance, recent research on Mazvihwa Communal Land in SouthernMashonaland highlights the necessity of microstudies by revealing strikingdifferences in the impact of drought on human welfare between twoadjacent ecological zones Š the sand veld and clay veld environments.4The inadequacy of the sources also affects one element in Iliffe's mainargument concerning the shift from famine mortality to malnutrition duringcolonialism (of which he is again aware). The lack of mention of malnutritionin the colonial records prior to the 1930s does not mean, of course, thatmalnutrition did not exist. Famine-caused starvation and malnutrition ofthe poor could very well have co-existed before the 1930s.But what seems most problematic to me is Iliffe's analysis of pre-colonial famines and his rejection of famine-caused mortality as a majorconstraint of pre-colonial population growth based upon it. First of all itremains unclear whether he argues for an absence of serious faminemortality or for a lack of evidence on it. His assumption that control offamine mortality was achieved except when famine merged with violenceis not totally convincing. He himself lists a number of severe pre-colonialfamines which were not all intensified by warfare and violence. Theconsultation of additional sources like the collection of oral traditionsrecorded by the National Archives of Zimbabwe (the Oral HistoryProgramme) would have extended the knowledge about this type of famine.5Also Iliffe's own analysis of early colonial famines Š especially those of1912 and the 'Ndanga famine' of 1916 Š casts doubts on his interpretationof the pre-colonial period. He argues that the famines of 1912 and 1916were of the pre-colonial type in the sense that the colonial intrusion didneither cause them nor contribute substantially to their extent. Stillowing to the severity of the droughts and the partial inadequacy oftraditional strategies Š many people depended for their survival on faminerelief provided by the colonial state. For instance Iliffe argues that duringthe 1916 Ndanga famine 'the prevention of mass starvation by the supplyof roughly 2 500 tons of grain was a remarkable administrative achievement,3 See M. Vaughan, The Story of an African Famine Gender and Famine in Twentieth<enturyMalawi (Cambridge. Cambridge Univ. Press, 1987).4 K. Wilson, 'Ecological Dynamics and Human Welfare: A Case Study of Population, Healthand Nutrition in Southern Zimbabwe' (London. Univ. of London, Ph.D. thesis, 1990).5 See for example: NatfionaJI Archives of Zimbabwe, Harare], AOH/17 ([African OralHistory] Interview with Dick Munjeri Mhako, 1 Apr and 12 Apr. 1977); AOH/14 (Interview withAaron Jacha, 14 July 1977) and AOH/17 (Interview with Mr Rakafa. 17 Aug. 1977). Furtherinformation is provided in Natl Arch., N/3/33/8 (Native Department. Chief Native CommissionerCorrespondence: Miscellaneous: History of the Mashona Tribes' Dec. 3 1903 - Jan. 21 1904)W. DOPCKE 105for on this occasion other survival mechanisms could scarcely haveprevented many hundreds or thousands of deaths' (p. 66).But these additional considerations still do not answer the basicquestion concerning a possible link between famine mortality andpopulation history. It might well be that the focus and direction of enquirytaken in the book is able to give only an inadequate answer to the question.Firstly, I think that the low population of the nineteenth century might wellbe explained in the context of the very exceptional character of thiscentury (high level of violence with tremendous direct and indirect impactson the reproductive capacity of the societies) and not as a reflection of therelationship between people and environment in pre-colonial times perse.Secondly, the search for 'apocalyptic' famines with high mortality isprobably misleading in relation to population history. Minor famines (whichwere not perceived as severe famines that killed) could very well have hada decisive impact on population growth, unnoticed by contemporaryobservers and unrecorded by oral traditions. Wilson, for instance, foundout that (minor) famines led to higher infant mortality due to lower birthweight as a result of the low nutritional status of mothers.6A more general thought can be linked to the above point. Whereaslliffe's description of entitlement to food prior to the 1930s ('both rich andpoor had fasted and feasted' (p. 79)) suggests tinges of 'merrie old Africa'there are indications Š though also very fragmentary Š that famines haddifferential effects on the population along the lines of economic andpolitical status, age and gender. The gore reMvemve (a major famine of thenineteenth century), for instance, is described in the following manner:'Mvemve was a catastrophic famine . . . Those who survived did so bymigration. Some survived because they had wealth ... Women and childrendied. Those women who were in a polygamous family were divorced.'7Ranger gives a vivid description of the impact of the great famine ofMasvaure on women and children with a quotation from an oral testimony:'Women left their areas of starvation to offer themselves as wives to menin areas of plenty. Some of these women threw their children into pools orrivers so that they would appear as unmarried women.'8 For the 1886-7famine the death of slaves is recorded in Chief Negomo's territory.9Enquiries along these lines might prove helpful in establishing a linkbetween pre-colonial famines and population history. But they necessitatea deeper knowledge of the social history of pre-colonial Zimbabwe than isprovided by Zimbabwean historiography so far. lliffe's book has openedup a whole range of extremely important questions which await furtherresearch.St Anthony's College, Oxford University W. DOPCKE6 Wilson, 'Ecological Dynamics and Human Welfare'.7 Natl Arch., AOH/17, Interview with Mr Rakafa.* T. 0. Ranger, 'Women in the Politics of Makoni District 1890-1980' (London, paperpresented to the ASAUK Conference, 1981).9 See D. N. Beach, Mapondera: Heroism and History in Northern Zimbabwe 1840-1904(Gweru, Mambo Press, 1989), 64 n.14.