A Socio-Geographic Survey of Salisbury, RhodesiaG. KayDepartment of Geography, University College of Wales, SwanseaINTRODUCTIONSalisbury was founded in 1890. In anticipa-tion of its growth and a bold attempt to createan orderly town, a relatively extensive cadastralplan was drawn up and surveyed between 1890and 1894, and some 8150 ha were reservedas Townlands and provided a controlled zonebetween the township and surrounding farm-lands. Since then Salisbury has been the sceneof rapid growth, especially in the post-warperiod. By mid-1972 its population was 467 000and included more than two fifths of the non-African population of Rhodesia and some fiveper cent of the Africans. Salisbury's spatialgrowth is even more striking than that of itspopulation, and it has an extraordinarily lowdensity of development which gives rise tomany of its particular characteristics and pro-blems. Peri-urban development preceded com-plete infilling of the original layout and in thepost-war years building proceeded most rapidlyin areas of farmland peripheral to municipalland. This led, amongst other things, to thecreation of eight autonomous dormitory town-ships for Europeans, in addition to severalAfrican townships which, for different reasons,also were established in outer areas. In orderto bring some unity to this sprawling con-glomeration, Greater Salisbury was created on1 July 1972, with an area of 550 km2. Rapidgrowth continues, and building plans passedin Salisbury for 1970/1 and 1971/2 were valuedat Rh$41 000 000 and Rh$55OO0O00 respec-tively and on each occasion constituted morethan half the total building output in Rhodesia.Even so, building lags behind needs. For exam-ple, it is conservatively estimated that 16 000'housing units' are immediately required formarried Africans who are legally entitled,through their employment, to live in Salisbury.Thus, like most primate cities, Salisbury con-tinues to face urgent problems of growth andmust consider whether the patterns set in thepast, recent though it may be, are appropriatefor the present and future.Throughout its short history, Salisbury hashad to work out a relationship between atechnologically advanced immigrant communityand a primitive indigenous population, bothanxious for advancement and each acutelyaware of their differences. From the beginning,evident differences and pre-conceived separatistnotions formed a basis for decisions and ac-tions. Common aspirations were largely ignoredand later suppressed by the creation of new legaldifferences which confirmed and hardened theline between Black and White that might other-wise have been eroded by contact and time.Thus Salisbury is a deeply divided city and hasbeen built as such by Europeans and Europeanauthorities. This division is so marked that fewgeneralisations, if any, are possible about thecity as a whole, for what is characteristic ofEuropean areas does not hold good for Africanareas and vice versa. The former generally re-flect the privileged positions and high incomesthat most Europeans are able to command whilethe latter are products of poverty and pater-nalism. It is this deep division and concomit-ant working relationship between the twoprincipal groups of townsfolk more than any-thing else that sets Salisbury apart from the'western city' of Europe or America.This paper, which is based largely on 1969census data and personal observation of thelocal scene, provides an outline social surveyof Salisbury. It briefly examines some aspectsof the growth, composition and distribution ofpopulation within the city as a whole and inits several distinctive parts. It analyses the closerelationships between the fabric and the societyof the urban regions; the various townscapesreflect the differing circumstances of humangroups and, paradoxically, the different built-environments greatly affect the way of life aridthought of their particular inhabitants. And itshows how spatial aspects of developmentunderlie numerous problems now facing thecity and suggests how these may be alleviated.THE NON-AFRICAN POPULATIONAspects of Growth and CompositionThe growth of the non-African populationis summarized in Tables I and II which showthat it has increased five-fold in the post-warperiod. Growth outside the municipal boundarybegan to assume significant proportions duringthe Second World War, and by 1969 themunicipality accommodated only half of thetotal population, a fact which underlined theneed for a Greater Salisbury authority. Thegrowth of the several principal suburbs hasnot been evenly distributed through time andthis may be attributed to several factors in-cluding the availability of water supplies, dis-tance from the city centre and industrial areas,and the date at which each suburb attained'take-off point' in terms of infrastructure andservices. During the early post-war years Water-falls and Hatfield, both lying to the south ofthe city, and Greendale, to the east, joinedHighlands as sizeable suburbs with populationsin excess of 2000 by 1951. Under governmentsponsorship, Mabelreign mushroomed a littlelater to become the largest single suburb by1956 but otherwise most growth occurred inwell established suburbs, especially in thesouthern sector. By 1961 more distant areas,notably in the north, were growing most rapidlyand the prestigious districts of Mount Pleasant,Marlborough, and Borrowdale joined the ranksof the major suburbs. Since then most growthhas been in the northern and north-easternsuburbs which added over 5000 to their non-African population between 1961 and 1969compared with an increase of only 1300 inthe southern sector.SALISBURY'STable INON-AFRICAN1904 TO 1972POPULATION1904Š1 7251911Š3 4791921Š5 6541926Š7 3241931Š 10 4811936Š 12 8691941Š 19 7531946Š 23 1071972Š117 000*1951Š 45 9531956Š 65 7501961Š 94 5481969Š105 955* Estimate by the Central Statistical Office; all otherfigures are from census returns.VTableNON-AFRICAN POPULATION 1941-69 BY URBAN AREAS*Total Urban AreaMunicipality% in MunicipalityPrincipal 'Suburbs'Lochinvar (S)Waterfalls (S)Hatfield (S)Greendale (E)Highlands (N.E.)Borrowdale (N.E.)Mt. Pleasant (N)Marlborough (N)Mabelreign (N.W.)Other areas194119 75317 02886,5827Š,1481 236Š,117,Š.397194623 1071741575,57167161276411 992105181Š.Š914195145 95332 64571,57092 2942 4622 0883 121237349714485851195665 75038 15658,017153 4394 2373 4534 20598665119014 4982 449196194 54850 22553,018745 5016 2485 9747 4832 0983 6693 2677 0051 2041968105 955 *53 51450,518086 7246 391 46 8728 621 A2 8376 4083 6976 8292 254*See Figs. 2 and 3 for the location of the Suburbs.72VMore than 90 per cent of the non-Africanpopulation of Salisbury is comprised of personsof European birth or descent and all but asmall minority of these are from English-speaking countries. Two-thirds of them wereborn in southern Africa, including 41 per centin Rhodesia and 18,5 per cent in SouthAfrica, while British-born persons comprise25,5 per cent. The next largest overseas group,the Portuguese, comprise only two per centof the non-African population but togetherwith the Italian and Greek communities theyconstitute a significant minority which, unlikethe handful of persons from northern andcentral Europe, exhibits a tendency towardsgeographical clustering (Table III and Fig. 2).Of the Portuguese 77,1 per cent live in fourareas which accommodate only 30,9 per centof the total European population; those ofItalian birth and, to a lesser extent, those bornin Greece, are also over-represented in theseareas. Each of these areas offers relatively cheapaccommodation mostly in old houses, and all ofthem are close to the main work-places in thecity centre and industrial estates. None of them,however, assumes ghetto characteristics, andpeople from countries of southern Europe com-prise only one-fifth of all Europeans in thecentral areas of the city and in Lochinvar-Southerton; elsewhere they comprise consider-ably less than one-tenth.. r vTable IIIGEOGRAPHICAL CLUSTERS OF MINORITY GROUPS OF EUROPEANS1969AreaCity centreAvenuesEast-CentralSalisburyLochinvar-Southerton% of allEuropeans3,211,712,33,730,9%ofPortuguese22,019,915,220,077,1%ofItalians6,315,223,813,558,8%ofGreeks9,931,811,02,755,4Portuguese,Italiansand Greeks as %of allEuropeans22,08,05,718,0ŠTo what extent these clusters can be attributedto the common origins or culture of the peopleconcerned or to socio-economic characteristicsthey have in common is not readily determined.However, it seems likely that the latter groupof factors is the more important and its in-fluence is in general accordance with similarwidely recognised factors affecting the locationof low-income, immigrant minorities.The more significant minorities of non-Africans are the Coloured population whichnumbered 5136 in 1969 (4,8 per cent of allnon-Africans) and the Asians (4055 -Š 3,8 percent). Most of the former were born in Rho-desia (91,5 per cent) as were most of the Asians(67,0 per cent though a sizeable proportionof them came from India (22,0 per cent). Untilvery recently no legislation prevented Asiansand Coloureds taking up residence in any partof the 'European Areas' of Salisbury and, infact, in 1969 a few were enumerated in everysuburb; for example, there were 15 in High-lands, 25 in Mount Pleasant and 82 in Mabel-reign. Nevertheless a combination of economicsocial, and cultural ties and pressures has ledto the concentration of Asians and Colouredsinto four localities (Fig. 2) The largest group(39 per cent of the total) is in Arcadia to thesouth of the city beyond the railway and Maka-busi river. In the early post-war period thissuburb provided modest accommodation fornew white immigrants but more recently hasbeen developed as a township more or lessexclusively for Coloured townsfolk. A secondlarger cluster (29 per cent) is in the central partof the city, peripheral to the main commercialarea, where many Asians and Coloureds73AGE AND SEX COMPOSITION 1969GREATER SALISBURYAsian and Coloured PopulationGREATER SALISBURY |Europeans} GREATER SALISBURYAfricansHIGHLANDSNon-AfricansHIGHLANDSAfricansBORROWDALENon-Africans jŠ'I 65i_AVENUESNon-Africans£:'_ _ r M2O---Ł 1560-50-HARARIAfricansMARIMBA PARK-IN fi MUFAKOSEAfricansFemales ] ^j7= 7654321 01 23X74\t voccupy aged houses and flats above commercialpremises. A third sizeable group (12 per cent),mainly of Coloured persons, lives in Ardbenniewhich is part of Waterfalls lying immediatelysouth of Harari African township and near theindustrial areas. In contrast to these three,Ridgeview is comprised of high-class houses andaccommodates most of the wealthier section ofthe Asian population. In this case, the cluster-ing and relative isolation cannot be attributedso much to socio-economic factors as to internaland external political pressures on the onehand and the highly developed communityconsciousness of the Asians on the other.One of the more influential factors affectingthe way of life of the townsfolk and the faceof the town is the purchasing power of thepeople, and spatial differences in this parameterare written large into the townscape. The mostmarked contrast within Salisbury is, of course,that between the African and non-Africantownsfolk, the latter as a whole being affluentwhile the former are mostly poor, but there aremarked differences within the non-Africanareas. Census data show that in 1969 28 000non-African males and 17 120 females were infull-time paid employment. The latter figureconstitutes 43 per cent of all females over15 years of age and reflects a very high levelof women at work in Salisbury's non-Africansociety. The affluence of many European house-holds is dependent upon there being more thanone worker in the family, and both mode oflife and standard of living are much affectedby the high rate of employment amongstwomen. The average annual earnings of non-African employees in Salisbury in 1969 wasapproximately Rh$2960, giving an overallannual income per capita of Rh$1280. Thisfigure does not include the earnings of self-employed persons, some of whom constitutethe richest group of households, nor does ittake into account personal incomes from pro-perty and investments of all kinds; it is there-fore a conservative indicator of income percapita.Regional variations in earned income percapita are not only a function of differentwages and salaries but also differences in thesize and composition of households which arereflected in varying age-sex structures. Withinthe non-African population the most markeddifferences in age and sex composition arethose between the Asians and Coloureds andthe Europeans (Fig. 1). Children (42,8 per cent)are particularly numerous amongst the Asianand Coloured communities, a fact which re-flects a relatively large size of family and alsoa potential for an increase in the rate of popula-tion growth. European society has a relativelysmall number of children (27,2 per cent) and,for a 'young country', a relatively large numberof elderly, retired persons. The most strikingfeature of its composition, however, is thatpersons aged between 40 and 54 outnumberthose aged 25 to 39 giving an unhealthy im-balance to its age structure because of a relativelack of young adults. Spatial differences in age-sex structures and household size do occurwithin the city's European areas and the dearthof young adults is most noticeable in 'expensive'northern suburbs such as Highlands andBorrowdale (Fig. 1) where the mean householdsize is only 3,28 and 3,16 persons respectively.The larger households in which children aremore prominent occur in the southern suburbssuch as Waterfalls (mean household size Š3,88 persons) and Hatfield (3,50) and in theCity centre (3,53). The most striking departurefrom average, however, occurs in 'the Avenues'adjacent to the city centre where a relativelywide variety of dwellings, including a largenumber of high-rise flats and many small, oldhouses, attract many young and elderly people(Fig. 1). Consequently children constitute only11 per cent of the total population of theAvenues and the mean size of household isonly 2,56 persons; and it will be clear thatthese facts have a pronounced effect upon percapita incomes.The 1969 census data showed that the lowestlevel of annual earned income per capitaamongst non-Africans was in Arcadia, thetownship for Coloured persons, where therewas only Rh$469 per head. Four other areashad less than Rh$1000 per person in 1969;Graniteside Š RhS636; City centre Š RhS768;Lochinvar-Southerton Š RhS812; and Water-falls Š RhS893. The first-named is an in-dustrial estate with only a handful of non-African residents, and all four areas occur inthe southern part of the urban region, threeof them being south of the railway and ad-jacent to industrial areas and African townships(Fig. 2). Also it will be recalled that significantclusters of Asians and Coloureds and ofminority groups of Europeans are to be found75GREATER SALISBURY POPULATION DISTRIBUTION, 1969vy-)Ł76in these areas. The next group of areas withper capita incomes of rather more thanRhSlOOO consists of the Msasa and Working-ton-Southerton industrial areas, Hatfield andthe eastern and northern parts of the formerSalisbury municipality. Farther from the citycentre lie the somewhat richer areas of Green-dale, Salisbury West and Mabelreign with in-comes of over Rh$ 12000 per head; and themore distant northern suburbs are the mostprosperous areas with more than RhSBOOper person and rising to Rh$1434 in MountPleasant and Rh$1591 in Borrowdaie. Onearea does not fit into the general pattern, namelythe Avenues where 1969 income per capitawas Rh$1468. Clearly this part of the innerarea, unlike the City centre (Rh$768), is notdominated by poor folk though some of itssmall, older premises do house low-incomefamilies. The new, multi-storey blocks of flatswhich are increasingly common in this area ofblight and transition, are mostly occupied byrelatively well-to-do households of limited size;and property re-development in the area doeslittle or nothing to rehouse the poor of thecentral areas but provides mainly for an inflowof young and elderly persons of some substance.Aspects of Distribution and DensityThe characteristics of the non-Africanpopulation examined above, with a few notableexceptions, are not dissimilar in general termsfrom widely recurrent features of many 'west-ern towns' but the low density of developmentand consequent widespread distribution ofpopulation in Salisbury are more unusual (Fig. 2and 3). The large extent and open nature ofthe built-up area of the city as a whole isindicated in Table IV which shows the deve-loped urban land as a percentage of all land onsuccessive concentric circles centred on themain post office. Only 3 km from the citycentre one-third of the land is not under urbanuses, while the 4-km and 5-km circles largelyrun through the outer parts of the originalTownlands and both have very low levels ofdevelopment. In the inner suburbs developmentoccurs1 over some two-thirds of the land butat the 8-km circle it covers only a slightly morethan half and thereafter falls rapidly to 12per cent at the 12-km circle. Open spaceswithin a city are to be highly prized and theirreal value should be appreciated by developingTable IVDEVELOPED URBAN LAND AS APERCENTAGE OF ALL LANDDistancefrom citycentre (km)123456789c/c of circleon deve-loped land967366593966665144Distancefrom citycentre (km)101112131415161718% of circleon deve-loped land371512767684and using them as amenities. However, evenif generous provision is made for amenities asa first priority (and parts of Salisbury alreadyare fairly well equipped in that respect), it isevident that within the present urban regionthere will remain numerous areas, both largeand small, that could be developed for urbanuses. And it would seem in the best interestsof the city as a whole to fill in such spacesbefore permitting further outward growth ofa general nature.Turning to the density of occupation withinthe developed European areas of the city, Fig. 3shows that only in the Avenues and partsof the central area where multi-storey blocks offlats and old houses on small plots predominateto densities exceed 50 persons/ha; the innerring of suburbs mostly have between 10 and30 persons/ha while the greater part of theEuropean residential areas have less than 10persons/ha. These figures include the Africansresident in these areas who, on average, com-prise nearly half of the total population butoccupy a very small part of the total space.These very low densities can only be under-stood in terms of early per-urban developmentwhich occurred piecemeal by sub-division oflarge agricultural holdings into smallholdingsand extensive residential lots during times whencheap land and labour permitted and evenencouraged substantial properties in areas be-yond municipal control and rates. Planninglegislation was minimal but did include a signi-ficant rule prohibiting building on plots of lessthan an acre in areas not served by water-borne sewage disposal. This remained in forceuntil very recently and affected the greater partof the Salisbury urban region. Consequently77784kSalisbury's suburbs have mostly been laid outon the basis of one acre or more per house,and a way of life and thought has been createdwhich maintain that anything less will under-mine decent living standards. In 1969 no lessthan 75,2 per cent of the European populationlived in detached houses, and flats accom-modated only 13,2 per cent (Table V).The very low densities of population aregenerally accompanied by spacious living con-ditions for Europeans if not for their servants.Fig. 4 illustrates 100 ha of Mount Pleasantwhich includes a shopping centre and an em-bryonic 'civic centre'; had a different localitybeen selected it might have included a schoolwith its playing fields or a sports club withtennis courts, bowling greens, and licensed club-house. Within the area shown no private pro-Tab)\e VEUROPEAN ACCOMMODATION INSALISBURY 1969Type of dwelling(a) Private dwellingsDetached housesSemi-detaehed housesFlats with kitchensFlats without kitchens(b) Other dwellingsDwellings and businesspremises combinedTemporary dwellingsBoarding houses,lodging housesand hotelsInstitutions andother collectivedwellingsNo.20 3375516 56322127 672ŠŠŠ,ŠOccupantsNo.72 6151 82312 55525787 2501152492 9156 235%75,21,913,00,290,80,10,23,06,4perty is less than 0,4 ha, and the whole em-braces only 157 detached houses and 12 flatsin two blocks together with 85 swimming poolsand 23 tennis courts. The houses mostly haveample garage space for one or two cars andrather less iloorspace for the accommodationof up to four servants. The spacious groundsare tastefully developed and, except in timesof drought, are well watered. The affluenceevident on such private property, however,generally contrasts markedly with the povertyof the public sector within the suburbs andinvites transfer of investment from one tothe other. Roads often are narrow, uneven andunlit while facilities for pedestrians are appall-ing, especially at night. Developed public openspaces are few and far between; water-bornesewage disposal is the exception rather thanthe rule, and water supplies leave much to bedesired in many areas.Nowadays the cost of living in most ofSalisbury's European suburbs is high and in-creasing rapidly. Houses and grounds areexpensive to purchase and maintain. Further-more, because of the very low densities ofpopulation, large 'catchment areas' are requiredto justify the provision of even basic servicesand amenities and therefore much travelling isinvolved in every-day life. Few children arewithin walking distance of their schools; andeven casual purchases of bread or milk andthe posting of a letter usually call for cartravel. Commuting to workplaces in the citycentre or industrial areas also involves lengthyjourneys. And, of course, the low density ofdevelopment militates against the provision ofa public transport system. Consequently lifewithout a motor vehicle is extraordinarily diffi-cult and many families have a genuine andcostly need, imposed upon them by their built-environment, for two cars. The general useof cars may greatly reduce time spent in travell-ing but it adds significantly to household ex-penditure from which there is no escape except,to a very limited extent, in southern and south-eastern suburbs where rudimentary bus services-do operate. In short, Salisbury's suburbsgenerally are expensive places to live in andthey provide no place for the poor. Most youngpeople and most new immigrants are relativelypoor and if they try to live up to the inflatedstandards of the suburbs they must find itvery difficult if not impossible to make endsmeet. Alternative possibilities, however, are verylimited. Only in the central and generally lessdesirable and less acceptable areas are small,old houses at all numerous while small, newflats, which are increasingly common in theAvenues, are mostly expensive. Salisbury isurgently in need of more high-density schemes,more relatively low-cost housing, and a widervariety of housetypes that would provide moresatisfactorily for the wide range of householdsthat make up society. Meanwhile present79circumstances in the European areas generallyfail to attract and keep many young adults ofrelatively limited means; they make it neces-sary for many women to work; and they re-quire large and continually increasing incomes'for the maintenance of established standards.Intensive development, however, is necessarynot only to cater for the needs of Europeanswith modest incomes. Low density developmentalso makes very heavy demands on all dis-tribution, transportation, and reticulationsystems. Nor are problems in the public spherelimited to the suburbs themselves; many trafficproblems of the city centre arise directly fromthe heavy dependence upon personal transportwhile, as noted above, the development of asatisfactory public transport system is difficultgiven the present pattern of housing. Futuresolutions to traffic problems will become in-creasingly expensive, and the same is true offuture water supplies and sewage disposalschemes and of all public services. If the muni-cipality is to keep pace with urban growth andmake good the gross deficiencies existing inthe public sector in present suburbs it will haveto make heavier demands on private purses.The first steps in this direction are being taken,and Salisbury property and car owners shouldexpect to be made increasingly aware of theirobligations towards development of publicfacilities which are extremely expensive in lowdensity areas.Finally, it should be noted that while theEuropean population finds itself increasinglyinvolved in financial struggles to maintain itscurrent mode of life and standard of living andto improve facilities within the European areas,there are limited prospects for a voluntary andsubstantial shift in the distribution of incomeand wealth in favour of the African populationwhich, as will be seen, has a totally differentand much less privileged way of life withinthe same city.THE AFRICAN POPULATIONAspects of Growth and CompositionThe African population was first enumeratedin 1962 when it stood at 215 810 and by thesecond census in 1969 it had increased to280 090. These figures represent 69 and 73 percent respectively of the total population ofSalisbury, and official estimates for mid-1972show that the latter proportion has been main-tained. The distribution of the growth of Afri-can population amongst the several major partsof the urban region is summarised in Table VI.There has been very limited growth within theEuropean areas and current policy is to effectreduction there. Harari is the oldest townshipand its population has changed little in recentyears, new building being matched by clearanceV" hATable VISALISBURY'S AFRICAN POPULATION 1962 TO 1969Number1962(a) 'European Areas'MumcapiltyOuter suburbs(b) 'African Areas'HarariHighfieldOther principal townshipsOther AreasTotal Salisbury Urban Region*32 17040 61058 25041 95024 43018 400215 81014,918,827,019,411,38,6100.0Number196932 16045 74957 95052 56071 17020 501280 09011,516,320,618,925,47,3100,0Ł* <j*Estimated African population as at 30 June 1972: 350 000 (Central Statistical Office, October 1972). It shouldbe noted that doubts have been cast on the validity of census returns for urban areas, it being suggested thatthey understate the true position, especially in the larger townships.80of older structures. Highfield, the other long-established township, expanded considerablybut its proportion of the total African popula-tion fell slightly. Most growth, in fact, has beenaccommodated in relatively new townships suchas Mufakose (20 620 in 1969), Kambuzuma(9 510), Tafara (6 030), and St. Mary's (13 160),the last mentioned having developed on missionland from 'unauthorised' squatter-type housinginto a recognized township with an orderlysite-and-service organization.Africans are drawn into the urban milieuas labour migrants and although the rate ofcirculation has slowed down considerably andthere are many residents of long-standing, veryfew Africans have a permanent stake in Salis-bury or any legal right to live there except asemployees or, exceptionally, as approved self-employed workers in the city. Rhodesia has forlong employed immigrant African workers andaliens continue to be a significant element inthe country's population. Salisbury is no ex-ception in this respect and in 1969 46 390 or 17per cent of the African townsfolk were foreign-born. Analysis of the composition of this con-tingent gives the following percentage figures:Country ofBirthMalawiMogambiqueZambiaElsewhereMales49,124,44,01,679,1Females12,34,22,02,420,9Total61,428,66,04,0100,0i,.- 4It will be noted that Malawian Africans areparticularly numerous; and that the aliens asa whole are predominantly male Š more sothan the total African population of Salisburyof which males constitute 64,0 per cent. Theorigin of Rhodesian Africans resident in Salis-bury is not well documented, but 1969 datashow that 81,0 per cent of all Africans inSalisbury area Shona-speaking while only 1,5per cent speak Ndebele and related languages;17,5 per cent speak 'other languages' and thisproportion closely reflects that of foreign-bornAfricans. These figures suggest that immigrationfrom Matabeleland is very limited indeed, andSalisbury provides less of a melting-pot forRhodesia's major tribal groups than an inter-national meeting ground.The distribution of Africans of alien originsexhibits one feature of note; they are mostnumerous in the European areas where theyconstitute 27,5 per cent of all Africans, a factwhich reflects their high level of employmentin domestic service. As a consequence they areproportionally less numerous in the Africantownships where they vary from 9,2 per centin Highfield to 14,8" per cent in Rugare andKambuzuma.The selective effects of labour migrationgreatly affect the age and sex composition ofthe African population which differs from thatof the other racial groups both in its generalnature and in the much greater spatial differ-ences that it exhibits (Fig. 1). Taken as awhole, it has a marked imbalance between thesexes, males over 15 years of age (47,9 per cent)being more than twice as numerous as females(19,9 per cent). It also includes few elderlypersons, those over 50 comprising only 4,0 percent of the total (cf. 20,5 per cent of the Euro-pean population). Children constitute nearlyone-third of the total population, which is asizeable proportion in view of the imbalancebetween the sexes. The causes of the age-sexstructure are manifold and a full explanationwould require a detailed investigation of thesociological and economic background to Afri-can urbanization. However, mention may bemade of a few of the principal factors. Theseinclude the inability of many African workerson their present incomes to support their fullfamilies in town; secondly, many Africans areunable or reluctant to give up their economicand social holdings in rural areas which aresecured by the presence there of members oftheir household; thirdly, there is an acuteshortage in Salisbury of accommodation suit-able for family life; and, finally, governmentpolicy restricts African residence in Europeanareas to domestic employees only.Spatial differences in age and sex composi-tion may be attributed largely to regional differ-ences in one or more of these factors. Thus themost extreme distortion of the age-sex pyramidoccurs amongst Africans living in the Europeanareas such as Highlands (Fig. 1) where, inaddition to the occurrence of all the abovefactors in acute form, a significant proportionof male employees come from beyond Rho-desia's borders. The population of the largerand older African townships, and of Harari81in particular, is very similar to the generalstructure outlined above, but that of newertownships reflects the greater provision ofhouses suitable for families. And the greatestcontrast is offered by the populations of Marim-ba Park, Mufakose, Rugare and Westwoodwhich not only enjoy better than averageaccommodation but also higher incomes. Inthese areas there is a much better balancebetween the sexes and children are particularlynumerous (Fig. 1).As African men are admitted to Salisburyonly as labour migrants, a very high level ofemployment is to be expected amongst them,and 1969 data show 115 540 males to be gain-fully employed (including 2190 in self-employ-ment). This figure represents 86 per cent of allmales over the age of 15. Job-opportunities forAfrican women, however, are relatively scarceand only 10 000 or 18 per cent of those aged15 or more were able to find paid employment(cf. 17 000 or 43 per cent of the non-Africanwomen). Nevertheless, because of the markeddifferences in age-sex structures, the overallratio of breadwinners to local dependantsamongst Salisbury's Africans is very similar tothat amongst the non-Africans, the relevantfigures being 1:1,2 and 1:1,3 respectively. Afri-can annual earnings, however, are much lowerthan those of non-Africans and in 1969averaged some Rh$40Q giving an overall in-come per capita of only Rh$180 which is nomore than 14 per cent of the average figurefor non-Africans. The average does, of course,mask considerable variations in earnings. Inparticular it is somewhat depressed by thelarge number of domestic servants whoseaverage annual earnings (Rh$244) were verylow; if they were separated the average for allother Salisbury workers would be approximatelyRhS450. However, if more detailed data wereavailable it would only reveal a handful ofrelatively prosperous Africans and, as acorollary, a considerably larger number ofdesperately poor households.Aspects of Distribution and DensityThe determination of African wage-rates isa legacy from days when paternalism wasmore important than pay, and prevailing ratesin most cases are still inadequate to meet thereal costs of amenities and services, includinghousing, that are provided by the government,municipality, and employers. The distributionand density of African population in Salisburytherefore are largely determined by officialplanning policies and legislation, the only majorexception being St. Mary's Township whereinitiative originally lay with the African towns-folk and the Roman Catholic Mission stationon whose land the township was established.A similar but much smaller African settlementis to be found on land owned by the MethodistChurch at Epworth. Fig. 2 summarises the1969 distribution of Africans within GreaterSalisbury and highlights, first, the significantnumber of Africans resident within the Euro-pean areas and, secondly, the occurrence ofdiscrete African townships in the southern halfof the region where they form a partial ringalong the periphery of the built-up area.Some 28 per cent of the African populationlived within the European areas where theyconstituted some 42 per cent of the total popula-tion. The greater part of these Africans areemployed locally as domestic servants or aredependants of such employees, though it isnow illegal to house the latter on private pro-perty in European residential areas, even thoughthey be members of the employee's immediatefamily. The occurrence of Africans in suchlarge numbers in the suburbs rather than theAfrican townships is a consequence of severalfactors which fall into two groups. First, thereis a widespread and long established practiceto employ domestic servants from early morningto late evening, albeit with broken hours, andat weekends. Further, the earnings of domesticservants are lower than those of any otherindustrial group except agricultural workersand averaged only Rh$20 per month in 1969.These circumstances, together with the entirelyinadequate public transport facilities betweenAfrican and European areas, persuade bothEuropean employers and African employees tofavour the established practice of accommodat-ing domestic servants at their place of work.The weight of such factors diminishes withdistance from an African township but is nevereliminated. Furthermore the acute shortage ofhouses in most African townships largely pre-cludes the transfer of domestic workers out ofthe European areas where servants' quartersgenerally are available (Fig. 4). The secondgroup of factors affecting the occurrence ofAfricans in European areas consists of thediffering needs for servants and differingŁ »14 H-41*82X ,M U NICI PA LPOOLSHOPPING CENTREEZ3 ifea /O £3SWIMMING POOLTENNIS COURTSA ADMINISTRATIVE OFFICESB COUNCIL HALL AND LIBRARY0 100 200 M83abilities to employ them. The ability to employservants is largely a function of household in-come, and in general terms it is likely to in-crease together with the need for workers. Thelarger the plot and house, and the morenumerous the amenities such as swimmingpools, tennis courts, and stables, the greater isthe need for African workers. There are markedspatial differences in such circumstances, andthese are reflected in the different ratios be-tween Europeans and Africans from one areato another (Fig. 2). Thus, for example, in theAvenues where small houses and flats pre-dominate Africans comprise only 14 per centof the total population, and nowhere in thecentral and inner areas of the city do theyexceed one-third. In contrast, in the wealthynorthern suburbs of Mount Pleasant and High-lands they constitute 47 and 50 per cent res-pectively while in Borrowdale, where quasi-rural pursuits are relatively common, theycomprise 72 per cent of the total population.The 28 per cent of the African populationliving in the European areas are convenientlysituated in respect of their workplace but theyare ill placed in respect of commercial, social,and recreational facilities for Africans whichare singularly lacking in the European suburbs.Salisbury, in fact, for long has failed to meetor even consider the reasonable basic needs ofthis sizeable section of its society which findsitself dispersed throughout a culturally alienenvironment. Recently there has been moreconcern about aspects of this situation whichculminated in legislation restricting residence ofAfricans in European areas to registered em-ployees oniy; the families of such employeesmay be housed in African townships wherefacilities are available. This legislation doesnot meet the real issues, social and political,and insofar as it militates against normal familylife it is undesirable and immoral.The African townships have been developedexclusively to serve the needs of their inhabit-ants and are relatively well equipped withfacilities of all kinds, but they are mostlyplaced at some considerable distance from themajor workplaces and from the services andamenities of the city centre (Fig. 2). Theprincipal townships (Harari, Highfield, West-wood Š Kambuzuma Š Rugare and Mukafose-Marimba Park) accommodate 53,4 per cent ofSalisbury's African population (39 per cent ofthe total population) in the south-westernsector of the urban region along the railwayline and in company with the major industrialareas, the latter being generally the nearerto the city centre. Their location is relativelyfavourable, though Highfield is some 9 km andMufakose 13 km from the city centre. However,recent expansion of African housing in thissector, at Glen Norah, lies beyond Highfield;and the newer townships in general are moreremote. Dzivarisekwa is 13 km from the citycentre and is far removed from any industrialarea; even as a place of residence for domesticservants working in the north-western suburbsit imposes a journey of some 8 km or more.Mabvuku, Tafara, and St. Mary's, being 14,17, and 18 km respectively from the city centre,are in even less favourable locations. Suchspace-relationship greatly affect the lives ofthe African townsfolk. Each township or clusterof townships becomes a semi-autonomousappendage to the city rather than an integralpart of the whole. And the expense and hard-ship of long journeys to work and to centrallylocated facilities and functions bear heavilyupon the low-income African workers andtheir families and, indirectly, add to thefinancial liabilities of those who employ themand of the city's population as a whole.Finally, reference may be made to theeffect of the location of African townships onpopulation 'gradients' from the city centre(Fig. 3). In the N.E. sector densities declinerapidly with distance, and the gradient is only'reversed' at the 16 and 17 km marks whereit is influenced by a small part of TafaraAfrican township. The pattern in the N.W.sector is similar but the occurrence of Dziva-resekwa township at 12-13 kms from the citycentre provides a pronounced anomaly. In thesouthern half of the city, and especially in theS.W. sector which is the most densely peopled,there is no similar simple pattern, and thelocation of African townships provides a seriesof peaks extending to the very periphery of theregion.Turning to the intensity of land use withinthe African townships (Fig. 3), the highestdensity of population recorded in 1969 was744 persons, ha in a very small part of Harariwhich consisted of old buildings and a shantysettlement that have since been obliterated.Nowhere else do densities exceed 300 persons/ha, and only five small areas had densities ofi- -iK84I_/JV.Ž—^^CENlrD"DUQZI1111! uSCHOOLSHOPPINGCENTREABCPUBLIC OPEN SPACEMARKET STALLSBUS STATIONADMINISTRATIVE OFFICES.4-0100 200M"85between 200 and 300 persons/ha. In the greaterpart of Harari, Highfield and Dzivaresekwaand in a large part of Mufakose there werebetween 100 and 200 persons/ha, while mostother African areas had between 50 and 100.St. Mary's township, with 30 to 50 persons/hais relatively open in character, a reflection ofits spontaneous growth on mission land.Marimba Park and Westwood have even lowerdensities which are in keeping with their reputa-tions as high-class African residential areas.It will be evident from these figures that,although heavily populated by Salisbury'sstandards the African townships are notcrowded by international standards. However,the layout of the townships does not give theimpression of modest population densities andat first sight space may appear to be intensivelyused. This is because of the regular, oftengeometrical distribution of small, single-storeyboxlike dwellings set within individual plotswith a consequent lack of sizeable open spacesfor informal recreation, amenity and cultiva-tion (Fig. 5). Such arrangements are the bestthat unimaginative and impecunious administra-tions have been able to offer; the worst, per-haps, are the massive slabs that constitutehostels for singe males. The impression thatthe townships are heavily populated is furtherexaggerated by a real shortage of living space,the actual dwellings often being grossly over-crowded. The 1969 data show that some 97per cent of Salisbury's African population livedin 65 070 private dwellings comprised of162 130 rooms giving an average of 1,7 personsper room, this being a conservative figure be-cause of the probable understatement of thepopulation. Depressing though such conditionsmay be, it is even more depressing to realisethat given the prevailing low levels of incomessuch living conditions are beyond the means ofmost Africans and but for paternalism andsubsidies there would be shanties rather thanorderly townships.CONCLUSIONThe city is a concrete expression of pastand present human conditions and inter-relationships, and as such it is not easilychanged. In Salisbury, the physical obstaclesto change are matched by entrenched viewsof a large majority of influential Europeans whofavour retention of the prevailing way of lifeand current standards almost regardless ofcosts. Difficulties and expenses of accommodat-ting rapid growth, however, have created anincreasing awareness of the need for change ifSalisbury is to continue to grow and if it isto become a better place in which to live andwork. Problems of growth, serious though theymay be, are not the principal causes of concern;these are more deep-rooted and call for radicalrethinking about the future from and face ofSalisbury. First and foremost is the deep divi-sion based in racial discrimination and grosssocial and economic inequality between Africanand non-African townsfolk. The pursuit ofpolicies which approach apartheid is un-acceptable not only because of the injusticeinvolved but also because it is unreasonablyexpensive. Indeed, the second fundamental pro-blem is the increasing inability of the city asa corporate body and of its constituent house-holds regardless of race to meet the rising costsof the prevailing way of life that has beenestablished in the past and is now firmly builtinto the urban fabric.The excessive lateral growth, the extra-ordinarily low density of development, andthe perverse location of many low-incomegroups constitute some of the more importantaspects of the latter problem which can bealleviated by physical planning. In the Euro-pean areas lateral growth should be checked ifnot stopped in favour of in-filling and sub-division of existing lots within the built-uparea and within a fixed limit on land prices.Priority should be given wherever possible tohigh-density housing schemes which provide avariety of dwellings, mostly at low prices, andwhich particularly cater for young adults andaged poor. Such development would requiregreater involvement by the city authorities and,possibly, central government not only in termsof planning legislation but also in direct partici-pation in development. Much-needed improve-ments in the public sector within the suburbsand in the city as a whole also should receiveearly attention, and this too would inevitablycall for greater public spending. The presentfinancial resources of the municipality are tooslim for its present tasks, and well established,relatively wealthy households should be madeincreasingly aware, step by step, of their res-ponsibilities to the community as a whole inlocal suburbs and in Greater Salisbury. In1971/2 rates contributed only Rh$3 498 000 or864<* if1- 4A,.. i7,9 per cent of the municipality's income, whichwas less than African rents (Rh$4,6M or 10,4per cent) and both were less than profits fromsales of liquor in African townships Rh$6,2Mor 14,0 per cent. Increased rates and local taxesmay well reduce the personal spending powerof most non-Africans but need not lead to anyreal decline in the standard of living or qualityof life since the municipality may be betterplaced to make more effective use of the fundson their behalf than the townsfolk are as in-dividuals.On the other hand, any substantial move-ment to reduce the economic gulf betweenAfrican and non-African townsfolk is likelyto be effective only if the more affluent sectorsof society, including the relatively well-to-doAfricans, are willing to give up, or are requiredto give up some of their income and wealth, orat least forego some of their likely increases inincome, in favour of the less fortunate majority.Meanwhile, living costs per unit may be reducedby improved physical planning; as, for example,by better location of new African housingwhich should have priority in large open areasnear the heart of the city; by more intensiveuse of land and a wider variety of architecturalforms within the townships; and by provisionof low-cost site and service schemes on whichAfricans may build their own houses.Erosion of present inequalities, however,need not be considered solely in monetaryterms. It is, perhaps, more important to giveearly recognition to the African's rights to apermanent place in urban society, with privi-leges and responsibilities comparable with thoseof his European fellows, albeit for the timebeing within specified townships. Minimumwages and urban facilities should be such aswill allow a man to support his wife andfamily, defined in western or modern terms,in town and he should be encouraged if notrequired to have them live with him. Increasedstability and responsibility should be matchedwith increased participation through electedcouncils in the administration of urban affairsboth locally within the townships and in-increasingly in the city as a whole. Such stepstowards a more integrated and more egalitariansociety should be possible within the near futureand without a substantial shift in politicalpower, which is in any case unlikely, indeed,financial implications of such changes are likelyto be as serious obstacles as the political ones,but they probably deserve higher priority thanany general increase in African wage-rateswhich achieve only a superficial improvementof the African's position.Finally, special attention is required for thenumerous Africans living within the Europeanareas of Salisbury. Two alternatives seempossible; the system of residential segregationcould be extended so as to transfer all domesticservants to African townships or full provisioncould be made for domestic servants and theirfamilies within the European areas. The pre-ference of the Africans concerned is not known;but the latter course has some real advantages.It would, for example, involve some inter-digitation of African and European communi-ties which might assist in reducing the gulfbetween the races.ACKNOWLEDGEMENTI am grateful to the Research Board of the University of Rhodesia for funds which enabledthe purchase of detailed data for groups of enumeration areas in the 1969 Census. Also acknow-ledged is the assistance of the Central Statistical Office for arranging the computer extractionof these data which are not otherwise available.I -. a87Figure 1, p.74 :Figure 2, p.76 :Figure 3, p.78 :Figure 4, p.83 :Figure 5, p.85 :KEY TO FIGURESAge and sex composition, 1969.Greater Salisbury Population Distribution, 1969,Greater Salisbury Population Density, 1969.100 Hectares of Mount Pleasant.100 Hectares of Highfields African Township.