Zambezia (1.990), XVII (i).ZIMBABWEAN DEMOGRAPHY:EARLY COLONIAL DATAD. N. BEACHDepartment of History, University of ZimbabweAT THE SEMINAR on African Historical Demography held in Edinburgh in1981, Ross noted the lack of historical studies on the population ofZimbabwe.1 He was perfectly correct: in spite of the importance of thesubject, historians studying Zimbabwe have paid very little attention todemography. There is a rich literature on the subject that comes from thegeographers, summed up and added to by the paper by Zinyama andWhitlow,2 but on the whole the geographers have tended to concentrateupon the recent past and especially that period since the first real censusof 1962. Among the historians, the study of land apportionment by Palmerfrequently referred to population figures but usually at national levels;Cobbing used estimates from the end of the nineteenth century to calculatethe Ndebele population in the country; and Mtetwa made considerableuse of population figures in his study of the large Duma region in thesouthern part of the country.3 However, population studies perse have beenneglected, at least for the early colonial period, with the exception of thework of Johnson, who used data from the early colonial period to showhow the population was variously underestimated throughout the colonialperiod until 1969.4 Even so, Johnson's work, though valuable, was basedon an incomplete use of the available material.In 1984 I produced a paper on the question of the pre-colonialpopulation of Zimbabwe.5 One part of this paper showed that attempts toproject the pre-colonial population back from the apparent increase in the' R. Ross, 'Record of discussion', in C. Fyfe and D. McMaster (eds.). African HistoricalDemography: Volume //(Edinburgh, Univ. of Edinburgh, Centre of African Studies, 1981), 9.2 L. Zinyama and R. Whitlow, 'Changing patterns of population distribution in Zimbabwe1,Geojournal (1986), XIII, 365-84.3 R. H. Palmer, Land and Racial Domination in Rhodesia (London, Heinemann, 1978); J. R.D. Cobbing, "The Ndebele under the Khumalos 1820-1896' (Lancaster, Univ. of Lancaster,Ph.D. thesis, 1975); R. M. G. Mtetwa, "The "Political" and Economic History of the Duma Peopleof South-eastern Rhodesia from the Early Eighteenth Century to 1945' (Salisbury, Univ. ofRhodesia, D.Phil, thesis, 1976).* R. W. M. Johnson, 'African population estimates Š myth or reality?' Rhodesian Journalof Economics (1969), III, 5-16. P. Mosley, The Settler Economies (Cambridge, Cambridge Univ.Press, 1983), argues from population figures but in discussing basic population data (pp. 111-13) does not take the analysis much further than Johnson.5 D. N. Beach, 'Towards a Population History of Precolonial Zimbabwe' (Harare, Univ. olZimbabwe, Dept. of History, Seminar Paper 59, 1984).3132 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAearly colonial period were futile, given the available data. However, I wasable to show that there was a basic correlation between the distribution ofpopulation in 1911 and that in the Early Iron Age and the first part of theLater Iron Age, data on the latter coming from the early work of Sinclairand Lundmark.6 Even so, it also became clear that much more work wasneeded on the population of Zimbabwe in the early colonial period. Inshort, a start would have to be made on the historical demography of thecountry.Since Zimbabwean historical demography is in its infancy, and since Ihave had no previous experience in the discipline, it seems to me that bestcontribution I can make is to begin at the beginning. In other words, I ampresenting the basic data collected so far, commenting upon the factorsthat would have affected its accuracy and making only the most tentativeassessments. It is hoped that, at best, this will lay the foundations forfuture work; at worst, it will be yet another set of extremely unreliableestimates for a region that was previously blank on the demographer's map.I have collected the population figures supplied by the SouthernRhodesian Native Commissioners of what were, by 1922, thirty-two Districts,for a period of twenty-three years from 1900 to 1922. As will be seen, theyhave some very serious defects, but they are virtually the only figuresavailable. In addition, I have concentrated upon the rural population, thisis not only because they constituted the vast majority of the populationbut because the calculation of population figures for urban and miningcentres poses special problems, primarily those associated with therelatively high mobility and varied origins of town and mine workers.7THE SOUTHERN RHODESIAN NATIVE DEPARTMENTThis government organization was formed in 1894, primarily for thecollection of tax and the recruitment of labour. It was headed, for allpractical purposes, by the Chief Native Commissioners (CNCs) at Salisburyand Bulawayo until 1913 when the CNC at Salisbury took responsibility forthe whole country. Under the CNC, Native Commissioners (NCs) tookresponsibility for the African population in each District, though thisapplied mainly to the people outside towns, mines, and, in some cases,6 P. Sinclair and H. Lundmark, 'A spatial analysis of archaeological sites from Zimbabwe',in M. Hall elal. (eds), Frontiers: Southern African Archaeology Today (Cambridge, CambridgeMonographs in African Archaeology 10, BAR International Series 207, 1984), 1-9.7 Thus in thinly-populated Districts, such as Hartley and Gwanda, the mining populationif added to the rural figures would seriously distort them. While mining populations fluctuatedconsiderably according to the state of the mining market, many miners from the north were,in fact, in transit on their way to South Africa. See C. van Onselen, Chibaro: African Mine Labourin Southern Rhodesia (London, Pluto, 1976).D. N. BEACH 33White-owned farms. In these areas general responsibility for the peoplewas divided between the NCs, the municipal, civil and police authoritiesand the owners of mines, farms and houses. As far as the populationfigures were concerned, much depended on the individual NC as to whetherhe counted people in places in which his responsibility was divided andalso as to the accuracy of his figures. Some NCs had Assistant NativeCommissioners (ANCs), some of whom were given separate sub-Districtsand tendered separate reports giving population figures for the sub-District;on the whole, however, this division of responsibility was a laterdevelopment. Responsible to the NC and ANC were Native Messengers(NMs), perhaps eight to ten per District, and it was mainly through thesemen that the NC was able to learn anything of what went on in the District.On special occasions, such as during a census, the NC could call upon thepolice. In short, there was a very small force for the 'control' of the people.The Native Department's allocation of Districts had a strong historicalcontinuity, and this was particularly true in Mashonaland. Districts werecreated to serve each major White farming or mining centre or to controlareas with a large population and, in Mashonaland, once Districts hadbeen allocated on this basis (by 1900) there were virtually no changes upto 1923 Š except by minor boundary alterations Š regardless of thechanges in population of the District. However, in Matabeleland therewere many more changes, partly because many new Districts were createdin the aftermath of 1896-7 rising which were later amalgamated, andpartly because, for a long time (in the 1900s), Bulawayo, the biggest townin the country, had no District of its own until the large Bubi District wasdivided into three smaller Districts. The Matabeleland Districts also variedconsiderably in size and population.The bigger the District, the more difficult the NCs found it to count thepopulation; after the first few years of administration virtually every villagewas known but it was still necessary to track down individuals who triedto avoid notice. Conversely, it was probably difficult for adults in smallDistricts to escape notice for very long, and this can be seen in the casesof Umzingwani, Chilimanzi, Goromonzi, Marandellas and Mazoe, wherethe apparent rate of increase decreased earlier than in large Districts.(Towards the end of the period under discussion NCs began to use motorcars: in 1920 NC Ndanga reported that he was now able to visit thesouthern part of his District more often. He was, of course, confined to thefew roads, whereas foot and horse patrols had been more extensive andtherefore more difficult for the people to avoid. Thus, the NCs begandevelop a tendency to rely even more heavily upon their NativeMessengers.) However, in every District an apparent increase in populationthat was in fact due to more villages and individuals being located by theNC was gradually replaced by figures that reflected a real increase.34 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAMuch depended upon the NC himself. Some, like Edwards of Mrewa,were in office for the entire period, and others, like Forrestall of Chibi and.1 W. Posselt of Charter, for nearly the whole period. Other Districts had aregular turnover of NCs. This did not necessarily affect the populationiigures one way or the other, though a long-serving NC had more experience,if be chose to use it. Posselt showed both extremes in his early years inoffice: a new broom' in 1903, he made a fairly meticulous census and thenrelaxed for years, adding to his totals by round thousands until he wasforced to supply more-accurate figures. A lazy NC might be tempted tokeep his estimates low, because this would mean less work in the collectionof tax. On the other hand, intelligent guesswork might not have beenfar wrong.The role of the CNCs in the collection of population figures waserratic. Unless specific requests were made for certain kinds of informationŠ such as the number of able-bodied men or the number of people livingŁ) i cert.iin kinds of land Š each NC was left very much to his own devicess u, iae data he supplied, and in some cases even District totals were»-nutf.'d for years without apparent censure. When the CNC did demand aŁ ertain conformity, as when a calculation of the population by multiplying! hf- number of adult male taxpayers by 3,5 was decreed in 1913, there wasnot necessarily instant obedience: NC Inyanga's 'rebellion' against the rulelasted for four years. At this stage of research, it is difficult to say whetherthe general failure of the CNCs of Salisbury or Bulawayo to enforceabsolutely uniform methods of assessing the population in the Districts isa blessing or a curse. On the one hand, certain categories of informationarc missing from some Districts and for some periods, but on the otherband some NCs were able to make useful observations that were notaffected by preconceived ideas from higher authority.CHANGES IN METHODS OF CALCULATIONThe first 'census' was taken in June 1895, barely nine months after thefoundation of the Native Department. Under the circumstances it is notsurprising that figures for each District were very low (see Table 0. thoughin some areas they have their uses in calculating the relative wealth ofdynasties in terms of livestock. Some NCs counted huts and multiplied bythree to arrive at the population figures for their Districts, others seem tohave counted people and huts separately, but what was significant wasthat from then until 1903 Š as and when a count could be made bothbefore and after the 1896-7 risings Š people were taxed according to thenumbers of huts that they owned: until 1901 the tax was 10s. per man perhut, and after 1901 it was 10s. per man per hut and 10s. for each wife afterthe first. This procedure was adopted because huts were easily countedD. N. BEACH 35and people were not, but the result was a decrease in the number oioccupied huts as the bachelors, at least, shared the huts and she taxburden. The 1904 tax ordinance changed the taxation system to simply S.\per adult male and 10s. per second and additional wives."In 1903 tax registers were already beginning to be compiled, and theywere updated from then on. This was a slow but sure method as far a*taxation was concerned, for, although men might put off being registeredfor a while after they became adults, eventually it would prove difficult forthem to continue to claim to be under age. Unfortunately, the tax systemmade it impossible to calculate population figures from tax receipts alone,though NC Lomagundi tried to do this in 1909. Firstly, there were arrearsof tax going back over two years or more, and secondly, a £5 tax paymentmight mean five people (all bachelors) or 10 people (five monogamistsand their wives, or a polygamist and nine wives). As Tables IIŠXXXIII show,the individual NCs had no common system of calculating the populationfrom the tax registers until after 1913. Some NCs raised the figures on thetax registers by a factor of three, some by four, some by 3,5, and some bymethods that were not purely mathematical. The CNC's order in 1913 thatall NCs should raise their population figures by a factor of 3,5 provokedsome criticism from NCs who knew that this was giving too small a figurefor their Districts. Perhaps the best-informed critic was NC Edwards ofMrewa. He had been in the District since 1895 and was married to thedaughter of Chief Mangwende, so he knew something of local familystructures. Until 1914 he had calculated his figures by noting the numberof adults due to pay tax and then taking samples of the number of childrenin certain villages to get an idea of the under-age population. He knew thatby using the 3,5 factor he would be under-counting the children, andconsequently in 1915 his total population dropped from 26 236 to 21 578,although his figures for adults had gone up. This change in the method ofcalculation was responsible for several apparent drops in populationwhich took place in the years after 1913 in different Districts (e.g. Belingwe,Sebungwe, Selukwe, Bulilima-Mangwe, Hartley, Charter, Inyanga. Ndanga-Bikita, and Gutu). Although it seems certain that in many cases the numberof children was underestimated it is not clear how great that underestimatewas, especially as the ratio of adults to children could vary from region toregion.Another factor that affected population estimates was the failure tonote men who who had gone to work out of the District and w-re not onthe tax register. NC Makoni was only one of many who realized this afterthe 1921 census. However, while full figures for such men compiled <utheir places of employment are lacking at this stage, it is highiv uniik."* Johnson, 'African population estimates'.36 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAthat this factor was ever as significant in Southern Rhodesia as it was toprove in Mozambique and countries north of the Zambezi. In 1910-11,when the local mining industry was employing more local labour than atany other time during the first few decades of the century, between 13 000and 14 000 local men were working on the mines, with another 1 700 ormore on the Rand mines.9 If we postulate 6 000 more working in Salisburyand Bulawayo at that time and make a fairly generous estimate for thenumber of men working in the smaller towns and places such as Kimberley,there might have been 25 000 men out of a total population of around715 000 working out of their Districts in 1911.10 This proportion was to risesharply in the aext decades as the towns absorbed more local labour,"but it seems certain that the great majority of these workers were workingfor wages away from home precisely because they were on the tax regis-ters and had to pay tax.In short, it seems likely that between 1913 and at least 1923 the sectorof the population that was most greatly underestimated was that of children.FAMILY STRUCTURES AND THE COLONIAL DATAAs can be seen from Tables IIŠXXXIII, the data on the family that are givenin the District reports are thin and erratic. The following are the mainpoints that emerge:Infant mortalityWhenever infant mortality was mentioned it was reckoned at 50 per centof births (NC Inyanga 1915; NC Gutu 1915; NC Ndanga 1915; NC Sebungwe1907, 1912). NC Bulilima-Mangwe, noting that 95 men and 59 women haddied in 1916 Š a healthy year Š and that this gave his District a death rateof 3,66 people per 1 000, thought that the death rate of infants was fourtimes this rate. (NC Gutu also noted in 1915 that women told him that theyregarded six births as an average number).ChildrenBecause NCs were aware of the high rate of infant mortality, the number ofchildren in a District was always underestimated, and although many NCspointed out that without a registry of births the population would alwaysbe wrongly estimated none of them was prepared to undertake such a taskwith such a high death-rate among infants. No NC ever claimed to have9 Van Onselen, Chibaro, 96, 101.10 T_ Yoshikuni, "The Origins and Development of the Salisbury Municipal Location: AStudy of Municipal Control of African Workers in Colonial Harare 1892-C.1923' (Harare, Univ.of Zimbabwe, Dept. of History, Henderson Seminar Paper 61, 1984).11 TO. Ranger, The African Voice in Southern Rhodesia 1898-1930 (London, Heinemann1970), 139.D. N. BEACH 37accurate figures for the number of children in his District, though somelike Edwards, probably came nearer to accuracy than most. The categoryof 'child' meant any male under 14 or 15 years of age, any female under 14or 15 years not already married and, in most cases, any widows notinherited by their husband's brothers under levirate marriage.MenThis category was often divided into bachelors, monogamists andpolygamists, but it is very rarely shown how wives were distributed withinthe latter category. After 1903 marriages were supposed to be registeredand in 1904 NC Matobo recorded that out of 228 marriages in his District inthat year, 118 were with a first wife, 69 with a second, 29 with a third, 6with a fourth, 4 with a fifth, 1 with a sixth, and 1 with a seventh.Married womenAs noted above, this category could include girls under 14 or 15 years ofage and many widows. Between 1904 and 1921 girls seem to have begun tomarry later.DeathsDeaths were supposed to be recorded, but very often more male thanfemale deaths were recorded. This was because, once a man or a taxablewife was on the tax books, it was very much in the interest of the family toget them off the books as socn as they died. The causes of death wereremarkably consistent over the whole period: when NCs gave a detailedbreakdown of deaths, as opposed to simple totals, they gave long lists ofevery conceivable cause and the numbers involved, usually ending upwith many single-figure categories such as suicide, snakebite, syphilis,lightning-strikes and so forth. The main killer was pneumonia, in thewinter months, followed by dysentery in the hot dry months just beforethe summer rains. Malaria was usually the third main cause of death, wellbehind the others, and was often noted as being fatal only if 'complications'set in. Smallpox was endemic but not usually fatal: NC Ndanga reckoned in1903 that only about 6-7 per cent of smallpox cases died, and after the bigvaccination campaign of 1914 still fewer died. The 'Vera' or 'Spanish'influenza epidemic of 1918-19 was the only really serious epidemic in thisperiod, and its effects are discussed in Table XXXIV.NATIONAL FIGURESGiven all the variations and errors in the methods of calculating thepopulation noted above and in the Tables, it is clear that it is not possibleto obtain very accurate figures for the total African rural population of thecountry for the period under review. This is particularly true for the38 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAperiod before 1913 because the ostensible increase in the population wasillusory owing to the fact that villages and individuals were being recordedfor the first time as the administration get to know the countryside betterand that exceptionally erratic methods of estimation were in use in someDistricts. After about 1913 the population figures in the individual Districts,when graphed, tend to show 'flatter' curves; major losses and gains ofpopulation in each District can usually be attributed to factors mentionedin the annua^reports, such as changes in District boundaries or immigrationfrom other colonies. In short, it looks as though we are getting a distortedreflection of reality. Just how distorted the reflection was remains tobe seen.If all NCs had conformed at once to raising the figures by the factor of3,5 decreed in 1913, there would at least be a common basis for discussion.But, although the populations were being calculated on this basis intwenty-three Districts by 1920, in another nine Districts (Marandellas,Makoni, Umtali, Melsetter, Victoria, Umzingwani, lnsiza, Gwelo and Selukwe)either the ratio was not used or there is no clear evidence one way or theother for the period between 1920 and 1922. The lack of 'Total Indigenous'figures from South Mazoe for the years 1916Š19 means that, for the purposesof calculating a national population figure, figures for the years 1920-2 arethe least inaccurate available. In other words, if we make the assumptionthat the population of the nine recalcitrant Districts was being estimatedon a basis of the raising factor of 3,5 or near to it, then the national total forAfricans in rural areas (discounting aliens) would have been:1920 1921 1922734 305 771 421 788 957At this point it is possible to try to allow for permutations of error.Looking at the 114 cases in those Districts in which it is possible tocalculate from the figures of taxpaying adult males what the 'TotalIndigenous' figure for each District would have been if the NC had used the3,5 raising factor, we find that in only 12 cases is the 'Total Indigenous'figure given (and based on whatever calculations the NC saw fit to use)more than 20 per cent greater than that which would have come from useof the 3,5 factor. Even in Mrewa, where the NC was, as noted above, usinga fairly realistic method of estimation, the difference rose from only 8 percent to 20 per cent between 1904 and 1912. Thus if 20 per cent is added tothe 'national' figures for 1920-2 arrived at by the 3,5 factor, we get:1920 1921 1922881 166 925 705 946 748D. N. BEACH 39At this point, however, we are reaching the limits of inferencesobtainable from these data as far as national totals are concerned. If, asthe evidence seems to suggest, the real deficiencies of the data in the1913-22 period lie in the under-counting of women and children, thenofficial records are not likely to supply the missing people.DISCUSSIONIn spite of their inadequacies, the data do provide some very interestingpoints for discussion. The main discussion is to be found in my chapter'First steps in the demographic history of Zimbabwe: The colonial periodfrom 1895 to 1922', in Demography from Scanty Evidence}2 However, the mainpoints of the investigation can be summarized here. After summarizingthe text of this article, I discuss the 1920 population distribution map (alsosupplied here, see Fig. 1) explaining that the distribution of populationrepresented by the symbols for 5 000 and 500-1 000 people is based on theentire documentation for the late pre-colonial and early colonial periods.The national distribution of population thus obtained shows a very markeddifference between areas east and west of the main watershed. This is inturn calls into question the usefulness in historical cases of conventionalgeographical divisions of the environment into 'highveld, middleveld andlowveld' areas: clearly the pre-colonial population preferred the 'GreatCrescent' of eroded country east of the watershed to the flatter country tothe west, and the difference between these two environments seems tohave been more important than that between 'highveld' and 'middleveld',although the 'lowveld' remains as an environment that attracted fewpeople. Seen in this light, it is easy to deduce that the very varied economicenvironments and defensive positions of the 'Great Crescent' were themain attraction.Archaeological studies confirm that this spatial preference also existedin the period 200-1300, while historical evidence (and the map itself)disproves the myth of widespread population movement as a result ofNdebele raids in the nineteenth century, except for a few cases in Wankie,Hartley and western Charter. Depopulation in Melsetter as a result of the12 The reason for this is that the original paper upon which this article is based waspresented at the 'Conference/Seminar on the Analysis of Census Data from Colonial CentralAfrica', held at the University of Wisconsin, Milwaukee from 18-22 August 1986. When the timecame to prepare the conference papers for publication in Demography from Scanty Evidence:Central Africa in the Colonial Era (ed. B. Fetter, Boulder, L. Rienner, 1990), it became clear thata complete rewriting was necessary. In order to develop the discussion it became necessaryto remove all of the basic data except the graphs and map. Yet, ultimately, the chapter entitled'First steps In the demographic history of Zimbabwe: The colonial period from 1895 to 1922',in Demography from Scanty Evidence, pp. 47-59 must stand or fall on the data. Consequently,with the agreement of the two editors, the data and the discussion are being publishedseparately.40 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAGaza Nguni emigration of 1889 remains a possiblity. With virtually noexport slave trade and little import of slaves, the pre-colonial picture isone of stability, but the 1300-1750 period is less well known.Apart from the temporary effects of war and famine in the southwestof the country in the 1890s, and immigrations into the north-east in 1908and 1917-18 because of disturbed conditions in Mozambique, there waslittle movement of population before 1923. Indeed, apart from centralMatabeleland, Inyanga and Melsetter, where purchase of land by Europeansdrove large numbers of people into less favourable environments, most ofthe 'Reserves' created by 1920 were in the 'Great Crescent', though largeparts of that area had also been appropriated by White settlers. The 1920population distribution map adds a new dimension to discussion of theland problem.I conclude the chapter by discussing the available evidence on thenature of the population. Firstly, the population was clearly not undergoingthe kind of decline associated with the coming of colonial rule in territoriesfarther north. Indeed, given the general healthiness of the environmentand the fact (obtained from other studies)13 that crop failures do not seemto have led to many deaths, a second point emerges: it is difficult to seewhy the population was as low as it evidently was. A possible reasoncould well be the limited amount of land actually available to the people inpre-colonial conditions Š in other words, the amount of land which wasclose enough to defensible strongholds which could also be cultivated orgrazed. However, if the people were, in fact, deliberately limiting theirnumbers it is not yet clear how this was being done. The chapter is notcalled 'First steps in the demographic history of Zimbabwe' for nothing." J. IliHe, Famine in Zimbabwe 1890-1960 (Gweru, Mambo Press, 1990).D. N. BEACH41Figure 1: ZIMBABWE (SOUTHERN RHODESIA) POPULATION 1920mmŁ 5000Ł SOO - 1000171 HL., Ł u"ŁŁ wwsssr-42ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAFigure 2: POPULATION ESTIMATES: SALISBURY CIRCLE GNNER)Ł DISTRICT BOUNDARY ALTERATIONŁ CHANGE IN POPULATION CALCULATION METHODŁ DISTRICT IMMIGRATION/EMIGRATONK ABNORMAL NUMBERS OF DEATHS(EXCLUDING THE 1918-1919 SPANISH INFLUENZA)D. N. BEACH43Figure 3: POPULATION ESTIMATES: SALISBURY CIRCLE (OUTER)Ł DISTRICT BOUNDARY AI.TK.RATU INŁ CHANGK IN COPULATION I AH I IAIIO.N Mt.THOI)A DISTRICT IMMIGRATION EMIGRATONX ABNORMAL Nl'MBKKS OF DKATHS(EXCLI'DING THfc 191»-l!t|9 SPANISH INH.UKNZA)44ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAFigure 4: POPULATION ESTIMATES: VICTORIA CIRCLEŁ DISTRICT BOUNDARY ALTERATIONŁ CHANGE IN POPULATION CALCULATION METHODA DISTRICT IMMIGRATION/EMIGRATONII ABNORMAL NUMBERS OF DEATHS(EXCLUDING THE 1918-1919 SPANISH INFLUENZA)D. N. BEACH45Figure 5: POPULATION ESTIMATES: UMTAU CIRCLEŁ DISTRICT BOUNDARY ALTERATIONŁ CHANGE IN POPULATION CALCULATION METHODA DISTRICT IMMIGRATION/EMIGRATON* ABNORMAL NUMBERS OF DEATHS(EXCLUDING THE 1918-1919 SPANISH INFLUENZA)46ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATAFigure 6: POPULATION ESTIMATES: MATABELELAND CIRCLE (INNER)Ł DISTRICT BOUNDARY ALTERATIONŁ CHANGE IN POPULATION CALCUIATION MFTHOOA DISTRICT IMMIGRATION/EMIGRATONK ABNORMAL NUMBERS OF DEATHS(EXCLUDING THE 1918-1919 SPANISH INFLUENZA)D. N. BEACH47Figure 7: POPULATION ESTIMATES: MATABELELAND CIRCLE (OUTER)Ł DISTRICT BOUNDARY ALTERATIONŁ CHANGE IN POPULATION CALCULATION METHODŁ DISTRICT IMMIGRATION/EMIGRATON* ABNORMAL NUMBERS OF DEATHS(EXCLUDING THE 1918-1919 SPANISH INFLUENZA)i»J0 I48 ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATANotes on sourcesThroughout these tables and in the main body of the article, unless it isotherwise stated, the figures for each district and each year come fromthe appropriate NC's Annual Report, except in those cases where a reporthas not yet been located but the figure for that year is given in the nextyear's report. These reports are found in the National Archives, Zimbabwe,in files N/9/1/1-25 (Native Department, Chief Native Commissioner: Reports:Annual: 1895-1922) for Mashonaland and for Southern Rhodesia after1912, and NB/6/1/1-12 (Native Department, Chief Native Commissioner,Matabeleland: Reports: Annual: 1897-1912) for Matabeleland up to 1912.Exceptions are:1895: F/4/1/1 (Statist: Reports: Interim: 1895), Secretary to NativeDepartment, Salisbury, to Statist, 21 Sept. 1895, and [Chief] NativeCommissioner Matabeleland to Statist, 30 July 1895.1896: Great Britain, British South Africa Company's Territories: Report bySirR. E. R. Martin ... [C.8547] (H.C. 1897, lxii, 561).1897: NB/1/1/1 (Native Department, Chief Native Commissioner Matabele-land: In Letters: General: 17 July - 28 Oct. 1897), Acting CNCMatabeleland to Statist, 30 Sept. 1897, and NB/6/2/1 (NativeDepartment, Chief Native Commissioner Matabeleland: Reports:Half-yearly: Apr .-Sept. 1897), Acting CNC Matabeleland to Statist,20 Oct. 1897.1904C: (Mashonaland) N/3/3/5 (Native Department, Chief NativeCommisioner: Correspondence: Census and Statistics: NativeCensus 1904, 2 Jan. - 27 May 1904), Original Returns from alldistricts.(Matabeleland) schedule in 1911C below. Note: for both 1904C and1911C Bulawayo municipal figures are included in Umzingwani.1910E: (Mashonaland) N/3/3/2 (Native Department, Chief NativeCommisioner: Correspondence: Census and Statistics: Distributionof Population in Mashonaland, 5-14 Oct. 1910), CNC to Secretaryto Administrator, 14 Oct. 1910, enclosing 'Return of NativePopulation in Mashonaland on 30 September 1910'.1911C: (Matabeleland) Southern Rhodesia, Report of the Director of Census. . . 1911. . . (Sess. Pap. A7, 1912).1921C: N/3/3/8 (Native Department, Chief Native Commisioner:Correspondence: Census and Statistics: Native Census, 1921, 23Jan. 1920- 4 Aug. 1921), Acting CNC to Director of Census, 5 AUR1921.The figures given in Tables II to XXXII are those supplied by the NCs,with no attempt made to correct their arithmetic or to perform obviousadditions and subtractions. The 'Notes' are a paraphrase or summary ofthe NCs own explanations of his figures, where any were supplied, and the'Comments' are my own.D. N. BEACH49Abbreviations used in the tablesA/B: able-bodied males between 14 and 40AL: alienated landAllMarr: all married menAliens: aliensBach: unmarried males over 14ChilF: female childrenChilM: male childrenDomAl: domiciled alienFlotAl: floating alien (i.e. alien temporarily resident in the district)MarrWo: married womenMono: married men with one wifePEA: refugees from MozambiquePoly: married men with more than one wifeRes: reservesSingWo: unmarried women over 14TM: towns and minesTotAdult: all adultsTotChil: all childrenTotF: all adult femalesTotlnd: total 'indigenous' African populationTotM: all adult malesUAL: unalienated landWid: widows50ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATATable IPOPULATION ESTIMATES FROM 1895-7 RECORDSDistrictBelingweBelingwe (Godhlwayo)BubiBulawayoBulawayo (Shiloh)BulilimaBulilima-MangweCharterGamboGwandaGweloHartleyInsizaLomagundiMakoniMaJemeMangweMarandeHasMatoboMatobo-MawabeniMaveniMawabeniMazoeMrewaMtokoNyatiSalisburySelukweTuliUmtafiUrrcingwaniVictoria18953 976 (adults)40005 320 (adults)17 07216199 (adults)3 989 (adults)540061502000014 0564 524 (2772 adults)24 24884203 961 (adults)12 4112010799329 82018967120841170003000789735 0005 97193002000018 000485645005000500012 050350020005000189824 45160006 602 (town)5 209 (adults)160986 70064688000845020 00013 500200023 8005 961189930000952030000142094350350022 395fmm M°'1 Th?e llgUTes were omitted from the following tables because ol the lack of figuresplace during"/?!Ž rtod' *"* ***'"*'* °' '** compllcated bound"y changes that tookTable IIGOROMONZI-SALISBURYftar19001901190219031904C1904190519061907190119091910E19101911C19111912191319141915191619171918191919201921C19211922teenPolyTomCNFTotCM Taint3201580446022196249119641678168419662 736268626922 70545324 410440944253696DcmAIFlotAIM.684067156 73067629557118361266413290153271509515 9871606315689157151588016 15116 5331712017662186001960518 70419043199601665819446188371912819 7402502743283102 06123730270003503 2563 2553 4493 5344 0565 87114 8771500016 76918 704160301614715 02515022156541600024743309291520673068306236333570326334903215656192111 19132003200353440565871250UMX: IMO-1 calculated x i per hm. 1M3: calculated (?) at 3,2 per hut. 19M: only slight increase because of emigration. 1907: says that there is a slight increase because of immigration from Mazoe(bul see figure.). I(IW» village count, used ux registers, estimated toul population. 1914: claims there was a careful census. 1916: population down by 2891 because of reduced estimate by ANC SalisburyTown. 1922: claims figure is down on 1921 because previously floating aliens were included.Cimrntm apart from a bit of UAL up 10 1910 (137 recorded for that year only) the District consisted of reserves, alienated land, small mines and SalisburyTown. Published census figures for 1904.1911and 1921 give 19 346,28 576 and 18 837. respectively, so obviously it depended very much on the NC of the day whether the town and urban figures were included, and whether domiciled alien, floating alienor indigenous town and mine workers were included. As can be seen, frequently the NC assumed that the town and mine figures he hid represented aliens.Table UISOUTH MAZOE-MAZOEYmr 8*c* Mono Poly tUMvr TolM tterttfo ChM CMF TolChil ToUnl ftxMI FtoMI MmtM.TU19001901190219031904C19041905(90S190719081909WOE19101911C191119121913191419151917191619191930I921C 1 1361921IK214402267 3151 34286000879393871052811104110001118017010116801220031840945005500698823723605*52 73763699280105271044512018124001290018001 8501770230035004 275523635004 7004500452552642150021*00650266106 3926142657067037200561871007500320340316110091852232422662310214920501784172417864 6275300540050004 700450045255264V,wi IM» figure from 4.47 per hue IM2-J incrc»e from imnif ration. 1*11: people me counted for census. 1*11 census liken? 1*17: indigenous emif ruing lo other Diilricu. bul domicled alienngyfe 19 btcaaaerfcey mutt rm pay u« after 3 moMhi'residence mtteaaofn. Hit births exceed deaths by IS: 1000. 1 0(10 more men. women and children registered. 1*1*: as for 1918. 1*21: domiciledand floating aliem total 19 800 men. 2 400 women tnd children From the census, prenoui repom were undemlimalei. ti U» adult male taxpayer! gives too low a figure. 1*22: all aliens total 19 000 menand 2 400 women and children.Comment the indigenous figure Tor 1915-14 ii probably due lo the introduction ol the 3.5 figure. As inGoromorui, the NC was never compietety consistent in recording the alien population.Table IVMREWABtctt Mono PolyTotU UarrWo TotAdu* mm CNF T«Ow) TotM Aliens Res AL-tTM19001901190219031904C1904190519061907190619091910E191019111911C1912191319141915191619171918191919201921C192119221350174118822159240431613 191313429582 4962 49636203 79113081356139514711460140914021 1231 1231 1141052262528692 93929853 3593 5493 5373 7423 77139024 0014 02740684 16056555 7084 343448546204 70855415 6555 75760586 093624160806 0876003532653265340541511363122661100011-120004932 133003373 3 735 15753 17 78395131804170831908820 400212162185222065229742386524 81126 2362157822 717P2 9332323622 96023 73023 94023 5682362199121324 811262362157822 693229032323622 91722 516234422349099263643324321412613103m5Notts 1910: took average per hut in different villages 1901: same method. 1903: tax register now complete. Lait year had 3 000 married men, but no census. Children calculated on average per village.1911: as nearly accurate as possible with the method adopted. 1915: reduction by 4 641 because 3,5 figure used. 1916: thinks the increase was more than this. 1917: underestimate because of extra tax.1921: emigration to other Districts, still calculated on 3,5, would be 3 000 more if real census taken.Comment the pre-1915 figures are probably baited on taxpayers plus NC's own method of calculating children.Table VMARANDELLASYŠr Bach MonoToFCWfTact* Taw190019011902190319O4C190519061907190619091910E191019111911C1912191319141915191619171911191919201921C 29M192119222 357 3 669 3 5126SS4173315368972391446615 53913 2491343614 79513461135(1146871535915141155001803216023166161923919 7842044021 89422144231052290026 22626221272562736414652611106264350350600170631713916606162451676019 5742100721301221142290024 766257452626416147414]17376412416616516066156439019501460151316005614762645672Nous lttt-3 detailed breakdown by duefdom. widi hats. Ratio of people lo hats isnearb«aemoate4. l*i^l«*M*U.m4:iiicitasepardydBeioand worked il out on Ml basis. Mil: used cemai figwes. 1«J»-1: 'lane' ««(people lo hats is near bM iiev«ro^. INJkl«»,l»U.W1OTpy' manben. up » «e»«il itaaiaaoW «ort«n o« Ł* lob««» fanw Med baiiiiiiiiii«ion. l»J»iook«> from vittafesD. N. BEACH86-Si eeŁ8 <s a * ;2§8§5gg£ SI99VRXKKK* 9*8 8i a Ł ijiIillPi!!iO,|ijjl|f J1I!ifŁ8 «liftTable VIIHARTLEYYt* BtchTernamCNF TofCh Tottnd OowV flb*(Mlm190019011MB11031904C1904190519061907190819091909E191019111911C1912191319141915191619171918191919201921C 1 191 10271921192211SO1967182711243401823467150006182689$790064896900190009630982685006547106298613881689746806937686949252IS 54493662000315826203 74140294 613556349064482533567530004000600066001000010000702692106743915291007936105216323662487928600915782296547694977112601043012791192234010511345191011440277711973658262086477 511779361838071521366827189Notes: 19M: increase because of immigration from Mazoe and Salisbury. 1915: decrease because of new compilation (3,5). 1917: decrease because people moving to other districts for better wages.1920: decrease became mines closing and 739 people shifted to Charter.Comment: I910E and 1921 figures obviously include aliens on mines for 'indigenous'. From 1913 there is confusion of the figures because aliens working on farms are combined with indigenous ondie same category of land.Table VIIILOMAGUNDI8acnft*jUMvr ToUM UmWo ToUFcumTata* Totin)flatUALm190019011KB19031904C1904190519N190719011M(1910E1910191IC1911191219131914191$19111917191)191919201921C19211922230003584700051205000140004891425355433424493715198586230532652119807204322615529955320633340027 6002780030671270552800031445336372934731670313233052231460324664151934 0293500060411131278100022541170351760076007666720096491133712000150001575016720175007480910094002140021577168001916516802160001300012788132391200050092010001192339222341578216420222*602408257035001156150105310001113127810005001000150015002000Nous: IM* 23 OOO includn 4 000 on the Zambezi. 1*M: includes Kanyemba nib-District. Increase due to re-immigration from NER. Estimate based on 3 children per married woman, regarded aslow. mtwhere numbers actually counted. 3J per £1 collected is right. Married men up 122 Ł 3.7 per cent, porygamisls up 119 - 11.2 per cent, married women up 121 - 2.4 per cent. More potygamists.but fewer have more than 2-3 wives compared with preview yean. HI* refen lo 'usual automatic estimate' but is lot-counting entire villages, hopes u have reliable census in future. 1919: N 3/3/7 NCSinoialoSN 3 life. 1919. lest counting in Siac*a sub-division sl»^females than males: adult males 2 713. adult females 2503. male children 2 753. female children 2 873 for Smoii division. Situation similar for entire district in 1911.Comment: very irregular figures to 1914. subsequent drop probably conversion to 3,5 system.58ZIMBABWEAN DEMOGRAPHY: EARLY COLONIAL DATA§ 88Se^ co Ł-S8SSJ 88Table XMTOKOfear art Mm Pot, AMUrn Tom Marrtto CNM CNF ToCt* Talnl KA Aim Fits UAL TV19001901190?19031904C 3102 4394 5 348 3 530 37641904190519061907 1516 3 229 54781906 1646 3406 58221909 1724 3479 60341910E1910 1996 3457 6 0591911 1987 3720 62901912 2016 3909 61771913 1308 3 537 5 7571914 2695 1560 4297 69001915191619171918191919201921C 3485 3707 1963 884019211922Nous: Mil: taken from an average of villages. 1903: taken from population per chief. 19M: took 1905 and allowed 1,5 percent increase. 1968: increase includes 5-6 OOOfrom PEA. 1909: more peopleretuniifig to District. 1911: at census, estimated 1 933 away at work.Cemment: the main .big increase was from PEA people from 1917. eventually domiciled.847793601048896109 7201034314 56214 5001414517 154170561758617 84718 70020 234216852165021 124217172244924 81124 47224 5632517525 5852611632 2003187532 577336343363562518 35250591632116113611750524814 44414 77615 77620 5252175622 42411 1411131616 375113501087811211117244966Table XI gINYANGAV«r Bat* Horn Prjr> AMtrn Jem UmWo ToMdUt CMM CMF WO* Tortnd PEA nts UAi. AL1900 12 4001901 133701MB 128871903 233041904C 3611 4649 5672 2940 3145 16406190419061906 9505 2466 2127 146201907 9600 6300 161001906 10000 6600 166001909 600 1146 2192? 4140 5532 9672 7326 170001910E 176001910 177X 1100 5700 102001911 178001911 C1912 4303 176001913 170001914 170001915 163661916 185721917 166741916 17539 52661919 224631920 229641921C 2946 2566 912 5032 11753 234991921 240611922 23546 4866 7869 10793Notts 19M-l:al4perhut. 19t3: 1902 lax was £2470 IOJ. for 12 887. so 4941 paid 10s, so adult males494l s 38 per cent, seems unnatural, many women and children, district healthy, sol assumemale taxpayers are 25 per cent and raise by 4. Includes 1 084 Katerere people included for the first time, real increase 9 333 because underestimated last year. 1906: many people fled lo PEA to avoid tax.19t*4140maietokensand2l92extra female tokens issued, SO4384 married women arc in polygamy (?). 4140 men minus 2192-1948, estimate I 148 aremarriedso800bachelon. Children underestimated.1912: don't think population is increasing so fast, so give same figures as I underestimated before. 1913: at 3,5x male taxpayers would have 15 960 but this is low so 1 give 17 000,10 per cent down on lastyear. 1»14: can't get figures down to 3.5, would be 16 170, so estimate 17 000.1913: took 1914 and 'natural increase'. Infant mortality 50-60 per cent. 1917: using 3,5.1918: refugees from PEA. 1»1»: PEApeople now counted. 1*21: k»l 492 lo Rusape, and 50 families returned to PEA, but 50 individuals in from PEA. 1912: SNUmulirwtes mat NC made error in 1921, increase was 462, not 1 097. Some PEArefugees in Mtoko moved into Inyanga.Commrnl: the NC's habit of atimaling in round numbers before conforming lo the 3.5 ratio in 1917, artificially suppressing what he thought the real figures should be, gives a very flat' profile up lothen. Immediately afterwards the PEA people added considerably to the population.1300150025002697272422722362345441073959100035003200346435194105432583656096855713000120001130012207123291029710832106441078111565Table XIIMAKONI-RUSAPEYm Bteh Mm Poly m»m TotM Hum ami CNF 7MCM ToUnd flw UAL AL19001901190219031904C 3397 3920 4727 1394? 4064190419051806190719061909 2033 3733 5415 4546 4734191061910 2091 3605 5375 4490 49431911C1911 2091 3805 5375 4490 49431912 2254 3339 5220 9391191319141915191619171916191919201921C 2700 4100 1025 6693 136761*211)22Ncui: IN* bm-popuiatioji ratio between 3.5 and 3.9.1M* compleie ceraui in November. I»l* centra in December. ItM: many immifranu from Inyangi and eliewhere. IUI: ceniui ihowed thain away « work had been undereMimaled. 231 irannfemd from Umuli lo Makoni.18-2000018-200002000017502196002150022000204632100020 7042070420 7042201122239225502322624 02424 770262702767026934263922696014 632162011674118691192161892419324231502350223990350300314400206250587245035220293329694020420046004500456050656570100120150150Table XIIIUMTAUyear BxhPolyTotM UarrWoCMFTmChi ToUMfl*9(Ml19001901190219031904C110419051906 18221907190619091910£ino1911C19111912 22301113 240019141915191619171919191919201921C 31061921192234944 79722052465524422661257 352323101270 3560367037505 76655955415565058003266 53648016796755081002383715766174241850014 3531706019995218482262523S0023500184501910020050222562357923970223902338723765355502415324759500838350035004000425043204000103407200710069007485775513372124301318013680139621446220001500110095070151098596590620630650670102*513500109001050011756118441000093709587945595419627152713001700200030323500INous.MM: ratio of huts lo people it 3.»7. Mt* immifnlioii from PEA. MM: children's fiiurci piesiei. rest liromUxiegulet.20Sadalailown, deaths and emifmian 10 PEA. 1911: fi|uie><ki<rabeam actual com undertaken. MM: 231 moved to Rusape.Camnti*: dan m» rnrtsde Umttti mfcan or detailed mine flinren.D. N. BEACH63a;Siis s sCO r~. CT> .Š ofŠ C>(vi s is mTable XVGUTUYearBKtiTow MarrtttiToFemuTOO* TotlH)flasUALAL19001901190219031904 C1904190619061907190619091910E19101911C19111912191319141915191619171918191919201921C192119222556285748474754726769548000835560638830958049033142130784411692622 750Notes 19M: 7) males, 19I* bonder shift with Victoria. 1913: 'carefully considered circumstances' led to figure. 1914: thinks last year overestimated, so allowed only slight in1915: reduction became of new counting method, but 440 people formerly in Chilimanzi included. 1917: has no real idea, the only deaths known are those on the tax register.29660315003250033400403504060037052378003810038300301703294534 94535945360503284732 7193281936078vstimato535 700312503142522810249062592626 6012666724 9092647526786d, so allowed only slight ui15504100412544604871561358135837315028914491235027502 750290031663406353135464 78834533801less than 1 per centTable XVICHILIMANZlYea/ Bach Mono Poly MUmTolU MarrtVo TolF TotAduHChilMChilF TotChil TotM AlmsflesUALTU1306127619001901190219031904C1904190519061907190819091910E19101911C19111912191319141915191619171918191919201921C 2912 1 9681921192230007500-8000700TMOOO1609161417232438252116972 57236152 3992 8202 863491850555 2545 2755 5403 4583 5433 7093 5533 63156288 8379 09494969 6341021910333106669 25093121000010 7501100011500130001641913 5561400014 50016 37016 89214 3331721317 69218 3891846219 39019 53219 873259642938170611631 165102618367 640 3 65010 6751197912 2021130812 88013 48013 00013 30013 50095010003 00038304 3914 4813 5654 2214 2124 4394 1622 89025964285017061 1631 165102612 6732 700 1 836CD13 53710690 199141568 12 703 4 508 2 703 1563N,M\ IW 2 Kb huts. Many (led because uf IXW riMtip state. !t«l 2 42t huts No census. IW»: 500 people on Central Estates transferred from Gwelo. 1910: increase because part of Victitransferred 1912: thinks VK) were absent at work in 1411 No census, registers only j!ive male taxpayers and married women, children a rough estimate. 1913: people move in from Victoria because of Priv\t*afions Odinance I»M Falcon Mine increases populationTable XVIINDANGA (INCLUDING BIKITA TO 1909)YearBach Mono Pay AKUarr TotU MarrWoYM ChllM ClulF TtHChil TollndUAL AL190019011902190319O4C1904190519061907190619091910E191019111911C1912191319141915191619171918191919201921C1921192210949 14 238 14 9936000015473 16305 6100940204 2406954950894971200125042013000961524335 5482512815162670006800071200720004717347029438303834931976314412948630624295782791228567317382792735056300522687326146239442515424 0662339826 753254428507766045%4703495049604960396244102166265637507592592510552552575315Notes: 1901:13 768 huts. 1903: rapid increase noted, smallpox and dysentery not so deadly as thought. 1906: increase of 2 500, fewer polygamists and married women because polygamists not reportingextra wivestoNMs, and married women struck off register. 1907: married women up by 2 892 and polygamists by 392.1908: ditto. 539 for married women and married men up by 602.18166 A/B men registeredlogo to work. 1909: fewermarried men and women recorded because overestimated before. 1910: believes last year'scount excessive, working on 3,5 now (NBŠsee comments). Bikita total no longer included.1912: 28 villages moved to Bikita, so only 206 up, but 174 more polygamists. 1913: reduced because of drought and famine Š deaths, fewer births, registration hampered. 1914: reduction because part ofthe District transferred to Victoria. 1915: reduction because now calculating from uxpaying males, infant mortality 60 per cent. 1916: 1 450 transferred to Bikila.Comment.this, the mosi populous district before 1910, suffered a repealed series of reductions. From 1910 the NC was calculating 3,5x male taxpayers and taxable women, anddidsountil 1915. Famine,border shifts and the vera' accounted for most of the rest.NIIm>TJ-<m>3o>5Table XVIIIBIKITArnr Bach Mono Pdy AtUair MvrWb MM TotCHI Taint Res UAL AL TM1910 1650 3350 4463 600 11749 220121911C1911 2078 3074 4242 650 11936 219801912191319141915191619171918191919201921C 2302 3593 777 5522 1529519211922O1910: Marts nn .W male taxpayers and taxable wives. 1916: transfers from Ndanga.tfnl although assessed as a separate District from lQl0, it was in fact run by the NC Ndanga. and his comments on general factors apply throughout.251932580521297230512521724 51023 56925 39327 4892812628 79412 6611306711 178122961462517 796188302210722 5161253212 738101191075510 5926 7144115201322064242506257020015001 500Table XIXVICTORIAPolyChilMTaCM TaintAliens19001901190219031904C1904190519061907190819091910E191019111911C1912191319141915191619171918Tgi919201921C192119229563 116707 3978 0896 30094151514538 00036 71939 000Notes: 1906: based on \9O4 census. 1910. reduction because of border changes with Chilimanzi and Gutu. 1911: actual count at census using .notched sticks carried by NMs. 1914: increase because oftransfer of people from Ndanga. 1921: real birthrate not known.im>43 00037 00032 30031000315003155235460359743637436 50036 00037 0003718435'143893240 2921525014 20014 300225692400024 20024 40024 40024 80018500194602034681008 750825017504044404004004006884707772358100760080529 450962096409 7009700980098001036010674850950950170019502194200020002000200020352 0378301<m[^ORozr~DATATable XXCHIBIBachAmurTolU ManWc CNIktChilFTolChil TotlridResUAL19001901190219031904C1904190519061907190819091910E191019111911C1912191319141915191619171918191919201921C1921192266167 0125678610028281 128820013160203882540628 78030900372003900041 0004100038 225294003143230205310003110032 0003300033 600310943360729 7303210232714336004 075550pzCD|26 99026 725278991279002865029 520260502543727446260782651429882727233326002 7002800500033453 5973 8243 7001454753933160065068022502312256422002500Notes 1901: 2 429 huts. 1906: increase in spite of very high infant mortality because of more births. 1907: claims that increase is 3 700.1911: says that the actual increase between 1904 and 1911, is3 371, using the tax register. 1913; decrease is due to emigration to other Districts, and famine killing old people and reducing births. 1916: some immigration from Victoria, but emigration to PEA and Transvaal.1919 now using 3.5* male taxpayers. 1922: Chihi and Nuanetsi sub-divisions have 22 872 and 9 842, respectively.Comment the adoption of the 3,5 ratio in 1919 and the disappearance of figures in round ihousands was due iothe retirement ot NC Forrestall, who had ruled the district since 1897.8Table XXIBULILIMA-MANGWEYear Bachrowloans QomAJFlaAi190019011902190319041904C19051906190719081909191019111911C19121913191419151916191719181919192019211921C499219224 590209736*2937 4793261132 57542 5004300043 50044 500394853666736 77848 86746 97442 55443149418404200042 23442 715426494330056552632346532 57224 62825 200282342800031 162309313175935 7183715337 75013931 7741806503354002 76124501495830721600105351026597721560615 57485759 713856054204326430085 38415600 4 200Notes: 1900-2: no census, but up because of immigration from Bechuanaland? 1904: counted half of Chief Mazwi's villages and made an estimated for the rest. 1908: emigration to BP 1909: a fewemigrated to BP. 1910:4 600 due to be transferred to the new Bulawayo and Nyamandlovu districts. 1913: only a small increase because of the border change with Nyamandlovu. 1914: 12 (WO up. but nota real increase. NC doubled previous estimates and made a more accurate count while ta>-collecting Many more live outside the reserve than »as thought M 075 vaccinations Many awav at »ork 1915:many deaths from malaria. I9U: used 5.5 ratio Š last year was 4? 1921: the census was very useful, and have allowed for inn increase and decrease since thenTable XXIIMATOBOVaar Bach Mono Pur/ UarrWo TotCM ToUnd Aim Res UAL AL TU190019011902190319041904C190519061907190619091910 22000 7887 1648 12215 2501911C -"" -1911237462465925 560250672690026300266002200021433225001763617114170131666717000173501770018500193801938019 600187500250300250250270788766506 761745674607600900091809800100501029016488431200161815481700850950320035003560122159954895076897600770075007620560056305800250189203250246200250250150200220CD1912 17636 6650 843 9954 189 >191319141915191619171918191919201921C 2605 2156 788 3901 993019211922Notts: 19M: many removals from farms, and people transferred to Tuli. 1907: has been a steady exodus since the 18% rising, people moving from hills to lower country. 1909: population has risen.but the exodus continues Š cleannces from farms and young men going to the mines. 1910: two chiefs and their people gone to Belingwe and Gwanda. Previous population slightly overestimated. 1912: onefifth of the district transferred to Gwanda and Umzingwani, also emigration. 1913: using the 3,5 ratio 3 709 gone to Gwanda and Umzingwani. 1914-15: decrease due to removal from farms.Commeru: the District started off the twentieth century with an abnormally high population because so many people had moved inert during the 18% rising. But early reports in the 1900s stale that twothirds of the population were local Nyubi, not Ndebele.Table XXIIIUMZINGWANIYur BKh Horn Pot) MarrMto TaCH Totlnd Aim BB UAL AL m190019011902 N1903 21904C 6123 >1905 31906 jrj1907 >1906 8000 o1909 8500 m1910 6100 560 100 6000 560 Q1911C 12513 O1911 6300 51912 8135 105 30 74O0 600 'V1913 6 512 154 35 7994 329 -<1914 8000 100 30 7596 274 m1915 7926 140 38 7435 315 >1916 7817 100 50 7 436 229 r~1917 6386 Š Š1918 86401919 88851920 8536 _1921C 1064 1038 164 1446 4319 8071 >1921 8647 300 80 56 6011 500 Q1922 9000Notts: from 1902 lo 1907 pin of Insiza. 1912: increase because part of Matobo and [nsiza transferred. 1913: increase because of immigration. 1914: decrease because of moves to other Districts andthe closure of mines. 1915: decrease because of moves to other Districts, though aliens up by 41, because many domiciled aliens from other Districts now registered here. 1920: non-taxpaying aliens nowincluded, but 300 moved to other Districts and the Bushtick mine closed.Comment: 191IC includes Bulawayo Town.10515410014010010080909080903035303850501685031568074007994759674357436766179288245805060118430600329274315229375464500365500400Table XXIVINSIZAYear Bach Mono Poly A/B Tom MarrWo TotChil Totlnd Mem fas UAL AL m190019011902 2915 15 8881903 2 683 16827I904C 171«1904 2992 170951905 176001906 2285 668 5425 185001907 2168 691 6 970 18 8001908 12820 4151909 136611910 15000 300 2000 12100 6001911C 164811911 145001912 131381913 134971914 12 3961915 13 776 2731916 140611917 14 5001918 14 5851919 14 9161920 IS 1971921C 1751 1729 489 2809 5898 126761921 150001922 17000Ni>m I9t2: increase, umuued. because of addilion of Umzingwani 5 296 hull 19M: very healthy year INI: lome immigration, but Umzingwani icparaled. 1*12: tost 633 uxpaying male! toGwcIoand other Districts. Retlofdistrictupby 244 because youth* now taxable. 1913: decreateof I 111 became count taken with care. Last year was merely eslimale. but tax is up to population il up. 1917: stillemigration from the District, but more immigration, ra there is a rise with natural increase Šemigration 112 over immigration, but iota! increase 429.1921: 127 families emigrated, mostly loBelingwe.1 000752828941101199410141160120002000340025943272398640734139406840004500835983468 7038092828284S2870486697569800073869496710421134100011000130022312500to M an %fr mtm At raw 11900190119021903I9O4C1904 2190 3X6IMS190619071906 7900 130319061910 52631911C1911in:191319141915 5237191619171916191919201921C2525 2646 51019211922Notts for 1910 and 1912 Ike followun e«n fi|uro are |IVCTReservem w1910 1 342 8SS1912 2 MO 2 400mm4 5194 150<-*2*356500Table XXVBUBITaf MM* am of itot r«wSIM256242*77723 5664916 1623 2377224 600I15S0 250002600027 00027 5003444 10323 111462016622120225001(330196691960019445196732160714 556 24(6126 30027 400UALm M rA m440 283 823 3 3003S2 350 1 000 1 174MM146161732106310X6501(00Łto942010 246H20S12 OX12 72412730SS367136963610236ALM2 2731026IML60540557556517SIX151440500500<A6 US3 000*102169(465550536257O0556511(661322913S6213 741If*7»Ł300TJ210631 000sso1(001 00015001921*sSfr2zObi5cStAhr--<co>-Ł1*6)1: Bubi and Bttlawiyo-jkilok Dnlricti nerfed n rorm Bria IN2<:new NC cilcutalei 25« 524 on 6 4*7 h»rj. io«i| 4 per hM «^«5 o<.! u brfort I*W: deadi rate IIJ per I 000 l»l» adaa deadlrue 3.1 per I 000. 1*12: immi|raucn from other Dnmcn 1*1) now uun| 3J 1*1*: increaae I 006 includmf Ml iliem 1*1*: lasl year'i fifin an undemMraar'1 Nr» NC Mtgesa no ml1926; I 000 men, women and chddren on nunes. 1*21: census led lomore accurate countC 11922: alien population now coumed in localComment two bif decreases, osx caused by separation of Nyamandlovij and Bulawayo in 1910. die other by die adoption of die 3.5 ratioTable XXXINYAMANDLOVUTan tferttti To/nut Too* Tata1910 11291 29 329 2(311911C19111912191319141915191619171918 M880 5 800 980 4900 200 P1919 12000 5 BOO 1000 5000 200 Z1920 12200 3480 KM 7 160 400 <J>i9jic IN in m Ł"" T.» .,,» fn19211922Notes: I9lt: created out of But* and Btttihma-Maogwc. but 616 waiting lo|o to Bub*. 1915 had malaria-1917. enugralion lo odier dumcti 19IS lou UDfrain lau ytm 4m to arflsaua alo other Districts 19Z2: ANC Goaai (nts more aenrate fijufr lor reserve.CiaiM protabl) tfancd by i«saif M rano32813*5822302 3752 285601761927122115001199412150122201220012 260118801200012200131701280013121283112252275250027002900584058005600348036003488105077080090010001 17098010001 1601 IXŁ300900086498550633079005250490050007 16077X790332530"300300400200200400400490Table XXVIIBULAWAYOBac* Mtm Poly Mvfflb TotCKI ToUntmlDomAI DomAI FUAIUALAL19101911C19111912191319141915191619171918191919201921C1921192242964170616416496586656665600541052456240120006S933448746573661200150034463232353840455050151662904 29142503 79638023500330042504260430043504400280021731574182517601870190019407540259331002950Notes: 191ft created out of Bubi, Bulilima-Mangwe and Umzingwani. 1911-12: no change, floating aliens unknown. 1914: Bulawayo municipal vaccination was 3 661. People moving to reserves inother districts. 1916: decrease because of moves to other Districts, fewer aliens. 1917: same as 1916, Old Nick mine closed. 1918: increase because more aliens now paying tax, and dependents of aliens nowcounted. 1919:4 000 indigenous from other Districts in Bulawayo, 1 500 floating aliens. 1920-21: inciease because more floating aliens now domiciled. 1922: decrease because domiciled aliens have leftComment: the usual urban confusion.Table XXVIIIWANKIEYen Bach Mono Poly Mtrttto TotCNI Tottnd Miens Res UAL AL TM190019011902190319O4C 564019041906190619071908190919101911C1911 CD1912 5536 5000 176 360 m191319141915191619171918191919201921C 1146 1004 555 2468 632619211922* Includes aliens? ' Figure includes 228 indigenous and 621 aliens.1 Figure includes 900 indigenous and 2 000 aliens. Ł Figure includes 950 indigenous and 2 000 aliens.Horn IM4: immigration from NWR. W»7:500 immignntt from South Africa. M deaths. 12,7 rwl OM, up by 1,65 on lajtvear, so Delation is 4 251^and other districts, so 200 up. 1»1* 19H estimate not far outŠtan drive. Reserves Commission and vscciilstion campaign kdlocoum of 5856. l»H:eitinm1ng «l3J?l»17:ic»neininiigrMili from Sebuiifwcnow settled, very hi|h infant mortality. IMS: increase due to domiciled aliens, 427 deaths, 300 of them influenza. 1*1* big increase because Sebungwe people now counted, and 2 000 floating aliens nowcourted. 1921: census very thorough, but floating aliens not counted.42524700490055666663553657X55004 7504670'51205375925694271149911520120955003933562000200011701335158616322900304552255000500052404150410030502970651565625195545511317615026025020022024025S28331563595621'3605503563503706808302900'2950'267Table XXIXSEBUNGWE-GOKWEYea Badi Mono fUy Am Mwrttb TofCn* Totlnl Mm Res UAL AL19001901 128921902 2980 17976 *.1903 19035 31904C 21211 §51904 21206 >1905 I190* m1907 >1906 24500 Z.1909 ^1910 £1911C 25 560 Q1911 25600 jj1912 25670 3390 21980 500 >1913 16834 16 242 592 J1914 26309 25359 950 <1915 19837 19167 670 m1916 19941 19362 579 >1917 19944 2803 16963 178 |51918 20300 19 3157 16 919 224 -<1919 16 511 21 3227 13 091 244 O1920 17 379 37 3391 13 695 256 p1921C 1791 1745 1222 4742 7603 17283 O1921 18056 38 3538 14260 220 ?1922 18186 50 3565 14401 200 j£ggivisk^16,06 per I 000, becuue more people found, people who fled lo NWR returned 1964: up by 2 171, mostly in Sebungwe, because of more complete census and more people back from NWR. 1907: 50 percent infant morlalily, some back from NWR. 19M: immigration and emigration balance. 1910: immigration from NWR. 1912: stable population because of very high infant mortality, women lose 50 per cent.1913: decrease because last year calculated on erroneous basis. 1914:162 people transferred from Gwelo. 1915: decrease because of new computation method. 1917: more accurate figures for AL from owners.400 emigrated, but natural increase the same? 1918:4 100 in this count will be transferred to Wankie. Population stable? 'Despite diligent coaxing and multiplication of every adult male on the tax registerby 3,5, the population stubbomly refuses to budge one way or the other by more than a hundred or two annually. So that, prolific as the local native tribes are and exposed to no other adverse influences thanquite a normal death rate, and a small exodus into other Districts, yet. in our process of computaion we always arrive at the same, or practically the same, monotonous total. The inference is obvious.* 1919:4 100 gone to Wankie, «o down by 3 808. no the increase is 292. 1921: most careful checking fails to alter the result arrived at by the 3,5 ratio.D. N. BEACH79— sŁS Wsj> <o W K i7> o o ^ CM 04 <S in-- 'J-tototor-r^cococo CT>O>o O u^ uS C\J S *D i^ 10 o ^* inf~i *^*i f^^ r*^, f*^ tf\ tfi co o^ ^o co too LO i/^ c5 o O u^ ^ o f^^ ^^ ^^^ 'Š *' '* f^ ^^ to <^i ^^ Ł_'» irt io r f> <Ši ^*iO CM C\J io O ^- CNJ ^ O T^- CT} ^J *Š COCO "^ "^ ^ ^* ^ 1^ CO ^- QO O^ CO C^ *ŠailIg I80ZIMBABWEAN DEMOGRAPHY: EARLY COLONiAL DATAtx!3O1§§ II8 8Table XXXIIBELINGWEYm Bacn Mono PrVy Afl ManrtVb TolCM Tol/nd Aliens Res UAL190019011902 4 210 331831903 33-340001904C 34 3351904190519C71909'9'0 JUJU J M JU / 00 H U I 3 £ 19U [Š,1911C 37 9601911 Z19121913 25 600 465 21235 1045 3 230 >1914 33 535 460 27155 2 250 4130 X1915191619171918191919201921C 4500 3906 1295 6613 1430019211922Notei 1902: many left for other Districts, and Mphephu's people gone back toTransvaal. 1904:1 50Oup, but many left District, infant mortality 30per I 000.1913: "Based on formercalculation otherwisethan 3,5 to each male uxpaying unit' discrepancy 11 254 down. 1914: says is 2 505 down on 1913 due to error in calculation in previous years. 1915: due to error in previous years, 3 000 down. 1916: 2 932down due lo miscalculation in previous years. 1917: errors of past years now corrected. Deaths up among registered lax unit, don't know why, no epidemic.1922: 776 fewer aliens on mines, 934 transferredto Gwanda. but immigration balances Ihis.Ctmmtnl NC Belingwe Menu lo have had serious problems with calculation, 1911-17.36 30736 97337 96037 56525 60033 53530 60727 5852808228 74629 30930 55830 614312603108366851147446546038847855683015502 5002 6742 07230 768316102123527155254052289322 92523 51024 0392507826110270054015189510452 250340731302 2102 3002 25023402110166521904 0603 2304130179515622 9472 9363 020314030402413Table XXXIIIGWANDA-TULIYtar Bach MonoPdyTtXU Ma/rtVoTrtfTotlrt DomMU DomAlW DomMCM AtensUAL19001901190219031904C19041905190619071906190919101911C19111912191319141918191619171918191919201921C2188192119227 763699096317090120016625674 022420042364 34044884 5974 7084 550312541714 4994 5364 6294 735486038433316568460016047622164666 73678748036656516 6601837114 07714 7001481915 200156(9161951642515902148821632916 2221786060050078890084708010027355060150011001100699711595918106087528331363783685211 1105 271727376657 7887 86983457780618870427 755662231906806742771547 40278207 9608862940172037 38110136250015001 1001 1006997115959181060875250 (Indigenous)250 (Indigenous)Notes: 1903: decrease because Mphephu's people going home. 1904: immigration from other Districts, some gone to Transvaal. 1906:500 up. 1907:1 200 aliens come to work. 1908:600 up, immigrationfrom Transvaal. 1909: up 600.1910: big increase because mine population included, also Chief Ngundu's people in from Matobo. 1914: youths now taxed? 1915:401 aliens left, 371 natural increase. 1916:women and children on mines now counted as aliens because they come from BP. 1918: unexplained drop in indigenous women. 1920: decrease this year only natives on tax register and ihose aliens on CNCform No. 2 counted. Il the two were joined, would have more than in 1919.1921: stopped counting aliens on mines. 1922: breakdown between Gwanda and Mtetengwe sub-divisions is 12 121 lo 5 939. Increasebecause part of Belingwe transferred to Gwanda, and immigration.|DistrictGoromonziMazoeMrewaMarSndellasCharterHartleyLomagundiDarwinMtokoInyangaMakoniUmtaliMelsetterGutuChilimanziNdanga-BikitaVictoriaChibiBulilima-MangweMatoboUmzingwaniInsizaBubiNyamandlovuBulawayoWankieSebungweGweloSelukweBelingweGwandaD. N.BEACHTable XXXIVNUMBER OF RURAL DEATHS FROM THELRural5005248002914062233287294415 *7300 ?48937915(55391 111108343272400 ?SPANISH"deaths reported inTM70614976887300?340114134272NC Staff got the flu300380 ?3216114716564INFLUENZAi 1918Total1 325752131310529789379749515427580VERA- ORDeaths reported in1918 casualties146277906 ?3453835254211 04216825294697726962287222110924442649831919Total3661037857538651 09241215829937156160135330315322* This figure includes deaths in both the rural areas and towns and mines.The 'Vera' struck the country in October 1918 and in the rural areas many deaths in 1918 werenot reported until 1919. A very provisional figure, based on the above data, for the number of deathsfrom influenza in the rural areas is 16 836, but this leaves out three Districts and deaths of womenand children which were never reported and deaths of people from the Districts who died on minesand in towns.Even so, of the nine Districts that reported a reduction in the size of population in 1918-19, fourdid so partly or entirely because of emigration, the closure of mines or the adoption of the 3,5 ratio.However, many NCs reported a reduced increase because of the influenza epidemic. Post-1922figures might show the real losses in children through reduced adult registration and m«rriage.