GeoJournal 12.4 375-386

375

© 1986 by D. Reidel Publishing Company

Rural Settlements, Mission Settlements and Rehabilitation in Transkei Lewis, Colin A., Ph.D., Prof. of Geography, University of Transkei, Umtata, Transkei Mrara, A. Z., B.A., Lecturer, University of Transkei, Umtata, Transkei Abstract: During the 19th century the predominant form of rural settlement in Transkei was that of dispersed homesteads. The form of the homesteads tended to differ from tribe to tribe, while the homesteads of various tribal groups were separated from each other by uninhabitated areas, or buffer zones. During the 19th century mission settlements were introduced to Transkei, some of which developed into towns. In the 20th century, mainly as a result of government action, homesteads have been nucleated into rural settlements and villages and much of the land has been fenced. A few peri-urban and industrial villages also exist in Transkei. Over 95 % of the population of Transkei in 1980 were still rural dwellers.

Introduction Most o f Transkei, w h i c h is in S Africa, is located b e t w e e n t h e p r o v i n c e s o f Natal a n d t h e Cape, s a n d w i c h e d b e t w e e n t h e m o u n t a i n s o f t h e D r a k e n s b e r g and t h e w a t e r s o f t h e I n d i a n Ocean. O n l y t h e Herschel area, w h i c h is s e p a r a t e f r o m t h e rest, lies N o f t h e Drakensberg. T h i s isolated region was t r a n s f e r r e d f r o m t h e Ciskei t o T r a n s k e i in 1975 a n d will n o t be c o n s i d e r e d in detail in this paper.

Population structure The provisional figures for the 1980 census of Transkei indicated a population of the order of 2300000. Over 95% of these people were rural dwellers and less than 120000 were classed as urban. The only towns of appreciable size are Umtata (the capital, with a 1980 population of circa 38000), Butterworth (circa 25000), Ezibeleni (circa 15000), Cala (circa 6000) and Umzimkulu (circa 4500). None of the other settlements exceeded a

population of 3000 in 1980. The landscapes of Transkei are thus dominated by rural settlements. Transkei exhibits many of the demographic characteristics of the Third World. Details of the 1980 census are not yet available but the 1970 census revealed high birth and death rates (40.36 per thousand and 17.0 per thousand), and a high rate of natural increase (23.4). The fertility rate (of 485.9 for women aged 10-49, calculated against the number of children under five) was twice as great in 1970 as in the Netherlands and comparable with that of Peru. Although Transkei was declared an independent state by the Republic of South Africa in 1976, the country remains heavily dependent on South Africa, especially for financial support and for employment for its people, many of whom are migratory workers. Work migration affects both sexes, but at least twice as many men as women appear to be absent from Transkei in the working age groups of 20 to 55, and are presumably employed in the Republic of South Africa. As a result much of the agricultural work and the maintenance of buildings has to be undertaken by women. Children, especially boys, still herd the livestock.

376

Gee J o u r n a l

Traditional Rural Settlements Homestead Morphology During the greater part of the 19th and 20th centuries: 'The typical settlements ... were ... not villages but individual homesteads, situated a little distance apart, and occupied by the owner and his family and dependents. Formerly some owners liked to be as isolated from other homesteads as possible, to avoid having to share the grazing for their herds', (Shaw, van Warmelo 1972) Each homestead consisted, in essence, of a series of huts. Each wife expected to have her own living hut and a store hut, and since polygamy was (and still is) practised in Transkei, the number of huts tended to reflect the number of wives belonging to the head of the homestead. In larger homesteads huts were built for the accommodation of visitors and sometimes for children and particularly for unmarried youths. The size of the homesteads depended partly on the wealth and number of wives of the owner, and partly on tribal custom. There are at least nine major tribal groups still in existence in Transkei, of whom the most important (in terms of area occupied) are Thembu, Mpondo, Mfengu, Xhosa, Sotho, Bhaca, Pondomise, Hlubi, and Bomvana (Hawkins Associates 1980). Shaw and van Warmelo (1972) state that 'the average homestead consisted, among the Xhosa and Thembu, of six to twelve huts, and in Pondoland of about twenty'. During the first half of the 20th century, as polygamy declined in importance, the homesteads tended to become smaller, so that by the 1970s 'the average number of huts is

12.4/1986

about four or five'. According to Hunter (1936) less than 16% of the married males in 'a conservative district' of Pondoland were polygamists in 1932, and that percentage has since decreased. The lay-out of the traditional homestead was also partly dependent upon tribal custom: 'the average Xhosa homestead was a semi-circle of huts roughly equidistant from each other, centring on the cattle-kraal, and if on a slope, above it. The doors faced the kraal' (Fig 2). This pattern applied especially to the Thembu, Mpondo, Fingo, and Bhaca.* The Sotho, who also exist in the mountainous uplands of Lesotho, tended to group a number of homesteads together into a hamlet. Their homesteads comprised 'a few living huts, one or more out houses, a cooking hut and store-rooms ... usually grouped together and joined by a reed fence or mud walls' (de Jager 1964). It is not certain whether Zulu-type settlements formerly existed in Transkei. Wilson (1959) and Derricourt (1974) have suggested that, prior to the nineteenth century, at least part of Transkei was characterised by 'village' settlements. Callaway (n.d., c 1938) has given a vivid description of the homestead (i.e. Great Place) of the chief in Pondoland as he remembered it. "The huts were ... dark and windowless, with only the narrowest openings to do duty for doorways. The kraal in which the cattle were sheltered for the night was large, but it simply consisted of a circular enclosure protected by rough branches of trees let into the ground. There would generally be a great many men sitting about outside the cattle kraal. Women would be in and out of the huts, looking after the brewing of beer, and the cooking of meat and grain in large iron threelegged pots. The whole place would look unkempt and disorderly. There would be the unmistakable odour of beer made from kafir corn - a large millet. There would be no fence, no garden, and, beyond it, would stretch the wide-spreading reid and countless ridges upon which the homes of the people were built".

Fig 1 Major tribal groups in Transkei. Source: Hawkins Associates (1980) and Lamla (1975)

Hut types

Ora o IState ~ ,

_~' . . . . . . . J Natal

\

"~

During the 19th century major changes occurred in the types of huts built in Transkei as beehive-shaped huts were increasingly replaced, mainly by circular but verticallywalled huts, as exist today. Ross (1824) recorded that: 'Their houses are exactly after the shape of a bee's skiphive. They are composed of a frame of osier ... covered with long grass for thatch, which is bound on the frame'. Van der Kemp (1800) described such huts as 'hemispheroidal ... of about 18 to 25 feet diameter',

cooo

j

while Steedman (1835) described the huts of the Mpondo as similar in construction and diameter, 'and from 6 to 7 feet high'. By the 1820s both square huts and circular

* vide

Hammond-Tooke (1962)

GeoJournal 12.4/1986

377

but vertically-walled huts had been introduced, and were 'first noted in the neighbourhood of mission stations' (Shaw, van Warmelo 1972). Bundy (1979) has shown that square huts became widespread in Mfengu areas in the period of peasant prosperity of the 1870s-1880s, but by the 1890s 'you will find the square houses dilapidated and the roofs falling in'. The circular but vertically-walled huts, called rondave/s, have roofs that are separate in construction from walls, unlike the old beehive huts in which walls and roof formed part of a continuum. The walls are made of wattle and daub, or of sods of turf, or of brick, or even of stone, depending upon the locality and the availability of building material. The roof is normally of thatch, resting on wooden rafters or on dome-shaped roof poles, although during the 1970s and 1980s galvanised iron has become increasingly popular in place of thatch. The exterior of the walls is usually decorated, in many cases being of a brown colour on the W side of the hut and white on the E side. In some areas beautiful patterns are drawn on the walls, sometimes with cow-dung, sometimes in mud, sometimes with paint. Many of the Sotho stone-built rondavels in the rugged area around Mount Fletcher and N of Matatiele are strikingly decorated, as shown on Fig 4. Although circular huts predominate in rural Transkei, during the 1980s square and rectangular huts have become increasingly common. Externally many of them look like simple British bungalows, but it is not unusual in coastal areas for them to have walls of wattle and daub, while mud-brick walls are common further inland. Most of these European-style dwellings have corrugated-iron roofs, rather than thatch. By the 1980s beehive-type huts had become almost extinct as permanent buildings, although an example was recorded in 1985 at M'Pakama Location adjacent to Matonana, near the border of the Magisterial Districts of Elliotdale and Mqanduli. This hut belonged to a witch

A

@0

west

cA:rfLEI

LY,, S I

I

M

A

L

I ~ down

GARDEN

XHOSATYPE. SOURCE:modified from SHAW& vanWARNELO (1972)

~,~B~,cl

GOATS

BOREArW'FHU"S'/

ENCEOE )

[]JUNIOR WIVES'HUT(S) , \ X \ r. ~ ~ S T O R E HUTS ~V~

GRANARY

/ .~" /~TIMBER,BUSHES ~ t j ~ 0 R STONE ~ ,"~"P / W A L L I N g

x~sa°n~Is~ / ~retal~ers_~s ~

'G :" gRAIN PITS C COOKINgHUT

J

huts east

ZULU TYPE. SOURCE:HALL (19~,)

C~.~

down

REED FENCE OR MUD WALL

SOTHO TYPE. SOURCE:modified from deJAGER (19641

Fig 2 Models of traditional homesteads

doctor. They are also built for temporary use, by people guarding crops, or to house abahhwetha, youths who are undergoing initiation into manhood, gbahhwetha huts were seen in 1984 near Hluleka, Umtata Mouth, and

Fig 3 Traditional (beehive, left) and modern (vertically-walled, rondavelj right) huts in Transkei. Considerable variation existed in the width of the beehive huts, especially between 3 and 6m, although rondavels also vary in width. Based partly on Shaw and van Warmelo (1972) oFra~vsork

Glass thatch

F

-3-6rn

+5m

I

~

±5m

-

-

GeoJournal 12.4/1986

378

Fig4 Decorated huts in the Sotho area of Nkadi, north of Matatiele

Coffee Bay, in the subtropical coastal region. Usually such huts are destroyed by burning at the end of the initiation ceremonies. Some evidence suggests that the introduction of rondavels took place in the south west o f Transkei, near the border with Cape Colony, where mission stations were first founded. The importance of missions as centres of diffusion was testified by Pim (1933) when he wrote 'To the Wesleyan Missionary, Jenkins, who founded the Palmerton Mission in the Lusikisiki District in 1845, is commonly attributed the credit of first teaching the Pondos to build huts with perpendicular walls'. Pim believed that the replacement of beehive huts by rondavels in Pondoland occurred between 1845-1870, which was probably a few decades later than in SW Transkei. Frescura (1981), however, believes that rondavel architecture may have diffused from the Highveldt, N of Lesotho, as a result of population migrations caused by early 19th century unrest. Walton (1948, 1956) is of the same opinion.

The Livestock Kraal The focus of the traditional homesteads described by Shaw and van Warmelo was, and still is, the livestock kraal. In 1782 Carter wrote: 'A kraal is a plot of ground within a ring fence, into which the natives of this country, every evening, drive their cattle, in order to preserve them from the attacks of the wild beasts'. During the 19th century the number o f wild beasts was drastically reduced as professional hunters, many of whom were white, entered Transkei. Elephants, for instance, were exterminated, as were lions. Leopards survive only in the most isolated areas, as in the gorge at Mkambati, which is,

in any case, a Nature Reserve. Of the predatory beasts only jackal and lynx remain bothersome in the 1980s, to such an extent that the government maintains packs of hounds and professional huntsmen in order to control them. Nevertheless, domestic livestock are still normally kraaled at night, even though there is little need to protect them from animal predators. The size and shape of livestock kraals appears to vary, depending partly on the number o f animals to be kraaled, partly on topography, partly on the whim of the builders. Circular kraals are probably most common, although many are square or rectangular, particularly in areas occupied by Thembu, Hlubi and Bhaca tribesmen. The material used to build kraals also varies, and is largely dependent upon what is locally available. In the coastal subtropical areas, where wood is relatively plentiful, wattle or brushwood kraals are common. In upland and rocky areas, and in areas where wood is in short supply, as around Nomaheya in Mfengu (Fingo) territory N of Butterworth, stone is used. The height of the kraal fence varies, depending partly on the stock to be kraaled. Goats need a higher fence than sheep and sheep need a higher fence than cattle. A 2 m high fence is adequate for goats, 1.5 m is adequate for cattle. Aloes may be planted on the surface of stone kraal walls, which adds to their effective height, or around the perimeter of the kraal (Fig 5).

Burial and Grain Storage In the past it was customary for chiefs to be buried in the cattle-kraal, whereas heads of households were buried just outside the entrance to the kraal (Shaw, van Warmelo 1972). Women were buried elsewhere in the grounds near the huts and infants may even have been buried in the floor of huts. These customs still persist, to a limited extent, although by the 1980s it had become more common for members of the family to be buried in a family plot within the general confines of the homesteads. These plots are noticeable features of the landscape. Although metal silos have been erected for grain storage in a number of localities during the 1980s, grain is still stored in pits dug into the floors of kraals, as near Tsolo (M. Ngwane, personal communication), in Pondoland, and elsewhere. Grain is also stored within the homestead in circular wickerwork graneries that resemble small silo towers, while cobs of maize are dried on wicker platforms that are raised off the ground on forked sticks.

Buffer Zones Until the latter part of the 19th century buffer zones existed between the settled areas of certain tribes. These were areas of unoccupied country, as Bransby Key de-

GeoJournal 12.4/1986

379

scribed in 1865 when he went to start a mission among the Pondomise, who lived along the middle course of the Tsitsa River. As he rode eastwards from All Saints' Mission (near Engcobo, in Thembu Country) towards the Tsitsa River, a distance o f some 100 km, he encountered 'a belt o f unoccupied country some 32 k m across waving with grass, intersected by no roads. There were only one or two little narrow footpaths, hardly visible in the long grass, worn by the few travellers who passed from tribe to tribe'. As Key's biographer remarked: 'such a belt of unoccupied country between tribe and tribe was felt to be necessary as a protection. It would have been quite impossible for the Pondomise to have built their kraals [i.e. homesteads] within measureable distance of the kraals of the Thembu. At any moment the war-cry might be shouted from hill to hill, and the cattle swept away by raiders' (Calla-

way 1911). As the population of Transkei increased it became impossible to retain these buffer zones. Additionally, in 1886 the British administrators introduced and enforced a code of criminal law upon the then Transkeian Territories which rendered such zones unnecessary, (Brownlee 1923). As a result the settlement pattern spread so that, by 191 I, Callaway could write: 'In every direction are the signs of a large population. The ridges are dotted all over with kraals, and all the lower slopes of the hills reaching right down to the streams are now under cultivation'. He also added, perhaps optimistically, 'Churches, schools, and traders', shops have sprung up in every direction'.

Mission Settlements During the 19th century the establishment of mission stations added a new element to settlement in Transkei. Mission stations usually formed nucleated settlements, in contrast to the dispersed nature of the traditional homesteads of that period. Over 40 settlements presently in existence in Transkei originated as mission settlements. The first of these was Butterworth, founded by the Methodists in 1827. While a number of these mission centres have now developed into towns (Butterworth is a case in point), many remain as essentially rural settlements. Although the earliest mission stations were founded by the Methodists, it was Anglicans who pioneered the foundation of mission hospitals, initially with St. John's Hospital (now called Sir Henry Elliot) in Umtata, then, in 1893 the leper station of Emjanyana near Engcobo and the hospital of St. Barnabas near Libode. Schools were an integral part of most mission stations, as at Baziya, founded by the Moravians in ] 863 (Disch11982). The core of mission settlements is normally a church and accommodation for those directly involved in the mis-

Fig $ A rectangular stone-built kraal, with aloes growing on the stone wails, Tsolo Magisterial District

sionary work itself. One or more schools (although now run by the government) also form part of the settlement, as may a hospital. 27 hospitals in Transkei were founded by the various churches, although all but two of the hospitals in the country were taken over by the government in 1976. Training centres, in which crafts such as weaving and basketry are taught, are part of many mission settlements. The Anglican centre of St. Cuthbert's, founded in 1881 and located near Tsolo, is one of the more quietly spectacular of the mission stations, with its Gothic style church, its bells (cast by Taylor's of Loughborough in England), its convent (aided by an Order in Wantage in England), its weaving shed, schools, and the hospital of St. Lucy. Holy Cross, located near Flagstaff in Pondoland, is at least outwardly a more bustling centre. Clarkebury (founded in 1830) and Shawbury (founded in 183942) in Thembu and Bhaca territory respectively, and founded by the Wesleyan Methodists, remain important educational centres, as are other mission settlements. It is the internal structure of these settlements, especially where they contain a major church, school (or schools), training centres, and particularly a hospital, that demarcates them from nucleated rural settlements founded as a result of government policy in the 20th century and renders them worthy of separate consideration.

F r o m T r a d i t i o n a l to Modern During the 20th century, and especially since the 1950s, isolated family homesteads have increasingly been replaced by 'clusters, if not ... actual villages'. This development, Shaw and van Warmelo (1972) believe, 'is the result of economic and population pressures, and does not stem from tradition'. They should have added that it also results from government attempts to plan the rural landscape, initially so as to control soil erosion. By the 1930s, however, it was

GeoJournal 12.4/1986

380

planned that settlements would be created to accommodate people who were in the process of proletarianization, of transformation from peasants or small farmers to migrant labourers. Both Bundy (1979) and Beinart (1982) have shown that proletarianization has continued in Transkei for over a century, and that the country has been treated by capitalists in South Africa as a major reserve of (cheap) labour. The transformation of the rural landscapes of much of Transkei, especially since the end of World War II, can only be understood with reference to government and capitalist policy within South Africa as a whole. By the 1950s the landscape of vast areas of Transkei was being transformed, partly to accommodate migrant labourers and their families: nucleated rural settlements were replacing formerly scattered homesteads, wire fences were being used to divide the land into geometrical units; plantations of rapidly growing trees were being established so as to provide firewood for the rural populace; local committees were increasingly being formed to plan and organize the rural areas, and particularly to allocate lands and control stock numbers. The landscape of Transkei in the 1980s, in spite of its picturesque appearance with its thatched huts and teams of oxen, is thus one of the most highly planned of all rural landscapes. Remnants of the traditional settlement pattern nevertheless remain embedded in the landscape, especially in isolated areas and in

regions in which the populace has resisted change, as in the Bomvana territories, around Elliotdale (Fig 6).

Combating Soil Erosion in Transkei During the second decade of the 20th century the Transkeian Territories General Council became increasingly concerned with the problem of soil erosion, especially on land occupied by Natives (the Locations). As a result a scheme was initiated in 1926 'to deal with sluit erosion' (Proceed/ngs 3). In addition to engineering works, areas that suffered from erosion were planted with trees. In 1935, for example, 19761 trees were planted under the scheme. Further measures to combat soil erosion were subsequently introduced, such as, in 1937, the introduction of contouring. This entailed the creation of bunds running parallel to the contours at intervals down slope, so as to limit runoff. As the Director of Agriculture reported in 1937, 'the contouring system is the best way of dealing with erosion if sufficient funds are available' (Proceedings4).

Contouring was the duty of the Director of Engineering, and his report for 1937 was highly significant: 'A small sheet-eroded area in the Kambi Location (Umtata) has been fenced and another larger area in the Umtentu Location (Umtata) is now in the course of being enclosed.

k t

Fig6 The extent of rehabilitation and planning of administrative areas within each Magisterial District of Transkei, 1980. Source: Hawkins Associates (I 980)

LESOTHO

NATAL

~SKEI

Percentage Extent of Rehabilitation & Planning less than 25 [ ~ 25 - 50

CAPE PROVINCE

50-75 ~,

.m

,oo

more than 75 ~

GeoJournal 12.4/1986

When the vegetation has recovered, it is proposed to permit the limited grazing of stock thereon ... When the benefits of these controlled areas are appreciated, the people will request that the scheme be extended. Where works have been carried out, the benefits are most marked, but, before complete recovery can be assured, some system of grazing control must be devised' (Proceedings4). Before the end o f 1938 the Director o f Agriculture reported: 'It is hoped to carry out an experiment in Location Development and grazing control, by which selected locations will be ring-fenced and sub-divided into several paddocks, each to have or to be provided with drinking water for stock. By this means controlled grazing and stock limitation to the carrying capacity of the land will be practiced. Fodder reserves will be obtained by hay-making, and the resting of one or more camps [i.e. paddocks] each summer will produce seeding and winter grazing' (Proceedings5). The following year the Director reported that: 'The Gcuwa location divided into three wards was selected in the Butterworth district. Fencing was commenced in the Tanga ward in May and ... the ward is now divided into 7 paddocks'

(Proceedings5 }. In the same year the Engineer reported that the introduction o f legislation was essential if soil erosion was to be

curb ed: 'The pastures in most districts are deteriorating rapidly and millions of tons of soil are being lost annually. Much of this wastage could be prevented by correct methods of cultivation'.

The 'Betterment Proclamation' and the Formation of Villages While local attempts were being initiated in Transkei to combat erosion, the problem was also debated at length elsewhere in the Union o f South Africa. As a result a Union Proclamation (Proclamation 31 o f 1939), commonly known as the 'Betterment Proclamation', was made in 1939. This Proclamation: 'to combat the fault o f overstocking', allowed for the establishment o f Betterment Schemes, under which an appointed Planning Committee for each Scheme was empowered, if necessary, to reallocate lands, residential sites and grazing areas, and to control many other matters, especially relating to livestock numbers. In the following year the Director o f Agriculture in Transkei, in addressing the General Council o f the Transkeian Territories, stated that: 'The present generation holds the land in trust for future generations' and that 'the erosion enemy is within our borders and that, unless some drastic action is taken to fight the main cause, it will only be a question of time before famine stalks the land. Poverty, ill-health and distress will abound' (Proceedings5).

381

The Union government appreciated that the only way to control soil erosion was to reduce the number o f animals on the land, and to improve the system o f arable cultivat i o n . 4 n order to do this in Transkei land would have to be reallocated in such a way that some Natives would become landless, (unless o f course they were given land outside Transkei). As W. G. G. Mears, Secretary o f the Department

o f Native Affairs, told the General Council in Umtata in 1945: 'to cope with this situation the Department has devised its proposal for the establishment of village settlements'. These settlements would cater mainly for: 'the landless but wageearning class. This proposal', said Mears, 'is a very important link between the Government's plans for the rehabilitation of the reserves and the large-scale industrial developments expected after the war' (Proceedings 7). In 1937 the Department had published a White Paper which suggested that three types of villages should be created, peri-urban, industrial, and rural. During 1940 work continued on the Betterment Scheme in Tanga ward, near Butterworth. 'The stockcarrying capacity o f the land was provisionally assessed, which resulted in the stock of the location being reduced by 106 cattle, 2 0 6 0 sheep, 20 donkeys and 73 goats' (Proceedings 6). Thus, within Transkei, was implemented the concept of carrying capacity of land. Three years later, in 1943, the Director o f Native Agriculture for Transkei reported progress on the Betterment Areas in the Butterworth area. After describing the fencing of paddocks, creation of stock watering dams, culling o f excess stock and fencing o f homestead sites, he wrote: 'Centralisation of kraal sites into villages has not yet been carried out, but is being considered. When this takes place various social and health services will need to be established to assist in the well-running of the village. At present these Betterment Areas represent better agricultural conditions through which better farming practice can be and is being performed. The ultimate aims are far reaching' (Proceedings 7). Following Nears' address to the General Council in Umtata in 1945 the Council adopted the principle of what were now called Rehabilitation Schemes, although with some reservations regarding the proposed villages. The Council was o f the opinion that peri-urban villages should be 'the responsibility of the Municipality and be located within the municipal area' (Proceedings T). Additionally, the Council did not approve o f the purpose o f creating rural villages as stated in a document published by the Native Affairs Department (A new era o f reclamation 1945): 'villages in which workers in more distant areas will maintain their families, returning home periodically'. The Council objected 'to any system which savours o f migratory labour' (Mears 1947), but did not object to the principle o f rural village formation. Industrial villages were not discussed by the Council, possibly because it was felt that, in rural Transkei, they would not be created.

GeoJourna112.4/1986

382

Proclamation 116 of 1949

In 1949 the Union government issued the Betterment Areas Proclamation (number116 of 1949) which superceded the 1939 Proclamation. Under the 1949 Proclamation the Chief Native Commissioner of the area in which the Betterment Area lay was empowered to 'assess the number of cattle units which any betterment area or a specified portion thereof is able to carry', and to 'define and demarcate ... areas for residential and arable or agricultural purposes as he may deem expedient' (Proclamation 116 of 1949). These apparently dictatorial powers were tempered by the fact that land scheduled as Native Reserves, which totalled over 96% of the 'Bantu Areas' of Transkei in 1949, could only be proclaimed as Betterment Areas 'in consultation with the inhabitants'. By the end of 1952 only 2.8 % of the land in Native Reserves in Transkei had been fully protected from soil erosion and it was estimated that 245 years from 1953 would be required 'to complete Stabilisation at Present Tempo' (U.G. 61/1955, 75). The effects of the 1949 Proclamation were assessed by the Tomlinson Commission, the influential and innovative Commission for the socio-economic development o f the Bantu areas within the Union of South Africa that published its report in 1955. The Commission believed that 'if the soil is to be saved, the necessary steps will have to be taken without delay to place the technical services in a position to plan and carry out conservation works on a large scale'. To house the populace the Commission suggested that three types of settlement should be established in the Bantu areas, namely, rural settlements, villages, and towns or cities. 'The rural settlement will properly form part of the agricultural planning, and are intended as a transitional stage between urban and rural life. Peasants or small farmers entirely dependent on agriculture will live in them, but there will also be a number of people who are not farmers but migrant labourers. They may be considered as numerous little sponges designed, in the first instance, to absorb those rural dwellers who are not destined to be farmers in the processof urbanisation. Tradition, tribal customs, family bonds, attachment to birth-places, graves of ancestors etc., make it necessary that a part of the population should first settle in the rural settlements before they become completely urbanised'. The Commission considered rural settlements to be different from true villages, believing that the latter should act as service centres for surrounding areas, as did existing villages and towns founded and mainly populated by Europeans in areas such as Transkei. Rural settlements, on the other hand, were essentially dwelling places for population, with few, if any, service functions. The Commission believed that: 'Several thousand of these rural settlements will probably come into existence in the Bantu Areas, and some of them may later develop into towns. Actual urbanisation, however, should take place through the establishment of a large number

of Bantu villages at places where circumstances are favourable to growth, and some of these will develop into towns or even cities in due course' (U.G. 61, 1955).

Rehabilitation in Transkei

The stamp of approval that the Tomlinson Commission appeared to give to Proclamation 116 was echoed in Transkei by the widespread reorganisation of land and rural settlements that followed in succeeding years. This reorganisation was referred to as rehabilitation. By 1980 about 75% of all Transkei rural locations had been rehabilitated by the government, their landscapes reorganised along planned lines. Fig 6 shows that the areas of greatest rehabilitation in Transkei are located in three belts: in the SW, adjacent to the Kei River; in central Transkei between Qumbu and Port St. John's; and in the peripheral districts of Matatiele and Herschel, where more than 75 % of the total number of locations in each Magisterial District had been rehabilitated by 1980. By contrast, less than 25% of the locations had been rehabilitated in the border regions of Bizana and Engcobo, and in the isolated coastal districts of Mqandull and Elliotdale. The influence of individual chiefs and headmen has been important in the process of rehabilitation, which Southall (1982) maintains 'met with bitter and widespread opposition'. He has argued that rehabilitation 'often represented a direct attack upon inhabitants' security', by forcing people to abandon their huts ('from the householders' perspective this entailed the abandonment of prior capital investment'), cull their livestock ('a direct attack on local wealth') and contribute labour to the rehabilitation schemes. He omits to mention that rehabilitation may have beneficial effects.

Sibangweni: A Rehabilitated Rural Settlement

Sibangweni Location, which adjoins the N boundary of the commonage of the capital city of Umtata, illustrates the. effects of rehabilitation. The rehabilitation of Sibangweni occured in 1960. Prior to that date a ribbon of development of dispersed homsteads, commonly called kraals, straggled beside the main road from Umtata to Kokstad, as shown on Fig 7. Further kraals were strungout near a footpath to Umtata, while others were dotted in haphazard manner upon the landscape. There were virtually no permanent fences in the Location. Arable land, with its irregular boundaries, was scattered along the lower valley sides, especially adjacent to streams. Only a few isolated clumps of trees existed, with no organised woodlands. The whole Location had a disorganised, haphazard, and perhaps untidy appearance, and was devoid of appreciable settlement nuclei.

GeoJournal 12.4/1986

383

Subsequent to rehabilitation, as Fig8 shows, the landscape of Sibangweni was entirely different. Each of the four wards within the Location: Lurasini, Nontswabu, Gxididi, and Lutholi, had its own nucleated residential areas. Thus four nucleated rural settlements had replaced the formerly dispersed kraals. The arable land had been organised into a series of geometrical blocks and, like the grazing camps, woodlots, and residential areas, had been protected by permanent barbed-wire fences. Each ward had its grazing lands divided into four camps, which are supposed to be grazed in rotation, with one camp being rested each year in order to improve its vegetation. Planted woodlands, mainly intended to provide firewood, had been established for each ward. The committee responsible for planning the rehabilitation of Sibangweni had, in addition to initiating the transformation of the landscape as has been described, also examined the optimum stock-carrying capacity of the Location. The committee recommended that there should be one cattle un/t for every 2 ha of grazing land in the Location. One cattle unit, in Southern African terms, is the equivalent of one ox or cow, one horse, mule or donkey, five sheep, or five goats. As was done in other locations,

the Sibangweni committee assessed the vegetation, climate, morphology and soils of the area, and recommended a ratio of cattle to sheep to goats. For Sibangweni this was 1:2:1. In other words, the carrying capacity was assessed as 3080 cattle units, which as a result of the recommended ratios, should comprise 1927 Cattle, horses and donkeys, 3854 sheep, and 1927 goats (Department of Agriculture and Forestry 1960). As in other locations, however, the human inhabitants were extremely reluctant to decrease the numbers of livestock, and by 1980 there were 5150 cattle units in the Location, 2070 more than had been recommended in 1960. Although the position of settlements in Sibangweni changed markedly between the pre- and post-rehabilitation periods, there has been little change in the form of individual dwellings. Rondavels, circular and mainly thatched huts, still pre-dominate, although a few rectangular and almost European style buildings also exist. Even the kraals, the livestock enclosures, still exist and form part of the residential areas. Animals, as in former days, are still penned in the kraals at night, as is commonly the case in Transkei. In very poor areas, however, or where there are few animals, they may be penned in the huts themselves,

Fig 7 Sibangweni Location in 1948, prior to rehabilitation. Land within the boundary of the Location that is not otherwise differentiated was used for grazing. Source: Airphotographs taken in 1948, Trigonometrical Survey, Pretoria

Fig 8 Sibangweni Location in 1980, following rehabilitation. Source: Plan No. T1400/80, Department of Agriculture and Forestry, Transkei

/' / / /

\\ \

NCAM~DLANA LOCATION

UMTA~A COM~ONAG~

O

km 5~ALE

384

as in parts of Pondoland near Lusikisiki. The residential areas in Sibangweni, as their designation implies, remain predominantly residential, with only rudimentary service facilities: a few trading stores, eating houses, primary schools, clinic and church, all intended mainly to serve

GeoJournal 12.4/1986

the inhabitants of each ward. This paucity of service functions, which was envisaged by the Tomlinson Commission, is characteristic of many rehabilitated rural settlements in Transkei.

Fig 9 Aerial photograph of a partially rehabilitated area near Lusikisiki. Contrast the nucleated and planned settlement with the dispersed homesteads of areasnot yet rehabilitated. Source: Surveyor General'sOffice, Umtata

GeoJournal 12.4/1986

Industrial and Peri-Urban Villages Industrial villages, envisaged by the 1937 White Paper, are few and far between in Transkei. Where they exist they usually house forestry or tea plantation workers, as at Emzimhle Township on the Magwa Tea Plantation, near Lusikisiki. Unlike the rural villages, these industrial villages have little or no grazing or arable land associated with them. A few peri-urban villages exist, adjacent to Umtata, but they are indistinguishable from rural villages, except that they tend to house people who travel daily to work in Umtata, in addition to their families. They are not, therefore, so associated with migratory labour as are the true rural villages.

385

appearance, and are also guarded by barbed-wire, strung from wooden posts. Only in a few remote areas, as at Poenskop, NE of Port St. John's, are the gardens and arable lands still guarded by traditional wattle fences, or by temporary barricades of brushwood, as near Umgazi, SW of Port St. John's. Instead of widespread scattered homesteads, each of picturesque rondavels huddled around the animal kraal, in almost three-quarters of the Locations settlements have been clustered to form nucleated rural settlements, usually with the huts regimented into rows that are alien to the rounded contours of the surrounding countryside. Planted woodlots, again within geometric boundaries, complete the picture of an organised and planned countryside and bear witness to the implementation of government policies in the 20th century.

Conclusion The landscapes of Transkei have been transformed by the foundation of mission stations in the 19th century and especially by the introduction of Betterment Areas and of Rehabilitation Schemes in the post 1938 period. Barbedwire fences divide the formerly unenclosed grazing lands into geometric camps. The contoured arable lands, formerly of irregular shape, have assumed ageometrical and organised

Acknowledgements The authors thank Messrs Jim Feely, and Richard Haines, Prof. T. Jones Hughes, Dr. A. Simms, Mrs. B. Miller and the anonymous referees for comments on an earlier draft of this paper; Mrs. A. Hutchings for preparing Fig 3, and Mr. P. McAinsh for cartographic assistance.

References Beinart, W.: The political economy of Pondoland 1860 to 1930. Ravan Press,Johannesburg 1982. Brownlee, F.: The Transkeian Native Territories. Lovedale Institution Press,Alice 1923. Bundy, C.: The rise and fall of the South African peasantry. Heinemann, London 1979. Callaway, G.: A shepherd of the veld. Wells Gardner, Darton, London 1911. Callaway, G.: Pioneers in Pondoland. Lovedale Institution Press, Alice n.d. but circa 1938. Carter, G.: The wreck of the Grosvenor containing a narrative of the loss of the Grosvenor, East Indiaman, wrecked on the coast of Caffraria, 1782. The Van Reibeeck Society, Cape Town 1927. Census of Transkei 1980. Unpublished office record. Statistics Division, Umtata. de ]ager, E.J.: Settlement types of the Nguni and Sotho tribes. Fort Hare Papers3, 19-30 (1964) Department of Agriculture and Forestry: Reclamation and resettlement of Sibangweni Location No. 9. Government Minute No. 64/8/2/9, Umtata. Derricourt, R.: Settlement in the Transkei and Ciskei before the Mfecane. In: Saunders,C., Derricourt, R. (eds.), Beyond the Cape Frontier. Studies in the history of the Transkei and Ciskei, pp. 39-82. Longman, London 1974.

Dischl, M.: Transkei for Christ. Umtata 1982. Frescura, F.: Rural shelter. Ravan Press,Johannesburg 1981. Hall, M.: The myth of the Zulu homestead. Africa 54, 65-79 (1984) Hammond-Tooke, W.D.: Bhaca society. Oxford University Press, Cape Town 1962. Hawkins Associates. The physical and spatial basis for Transkei's first Five Year Development Plan. Salisbury (ZimbabweRhodesia) 1980. Hunter, M.: Reaction to conquest. Oxford University Press, Cape Town 1936. Lamla, C.M.: Present day diviners (Ama-qira) in the Transkei. Unpubl. M.A. thesis, University of Fort Hare 1975. Mears, W. J. G.: A study of Native administration in the Transkeian Territories, 1894--1943. Unpubl. D. Litt. thesis, University of South Africa. Pretoria 1947. Pim, H.: A Transkei enquiry. Lovedale1933. Proceedings, (3, 4, 5, 6, 7): Proceedingsof the United Transkeian Territories General Council. Umtata. The numbers refer to the respective volumes: 3, 1935-6; 4, 1937-8; 5, 1939-40; 6, 1941-2; 7, 1942--3. Ross, (Mrs.): 'Letter', in: Long, U. (compiler), An index to authors of unofficial privately-owned manuscripts relating to the history of South Africa, 1812-1920. Lund Humphries, London (letter of 1824) 1947.

386

GeoJournal 12.4/1986

Shaw, E. M.; van Warmelo, N.J.: The material culture of the Cape Nguni Part 1 : Settlement. Annals of the South African Museum 58, 1-101 (1972). Also: Part 2: Technology. ibid. 58, 103-214 (1974). Also: Part3: Subsistence. ibid. 58, 215-445

(1981) Southall, R.: South Africa's Transkei: the political economy of an 'independent' Transkei. Heinemann, London 1982. Steedman, A.: Wanderings and adventures in the interior of southern Africa. Longman, London 1835. U.G. 61 .: Summary of the report of the Commission for the socioeconomic development of the Bantu areas within the Union of South Africa. Government Printer, Pretoria 1955. •":~::::¢::,~ ~!:: i:i:~: :::::::::::::::::::::::::

~i~:~'-'~:~::~:~:::::!~$::;~

~::::~:i::::::::

van der Kemp, J.T.: An account of the religion, customs, population, government, language, history and natural productions of Caffraria. Transactions of the London Mission Society 1, 432-468 (1800) Walton, J.: South African peasant architecture - Southern Sotho folk building. African Studies 7, 4, 139-145 (1948) Walton, J.: African village. Van Schaik, Pretoria 1956. Wilson, M.: The early history of the Transkei and Ciskei. African Studies 18, 4, 167-179 (1959)

NEW PUBLICATIONS

Migration and Settlement edited by ANDREI ROGERS FRANS WlLLEKENS GEOJOURNAL LIBRARY 1985, 425 pp. Hardbound Dfl. 160.00/£44.50/US$54.00 ISBN 90-277-2119-X

This volume represents an overview of the substantive findings and methodological advances that have emanated from the landmark comparative migration project conducted by Andrei Rogers and Frans Willekens and associates at the International Institute for Applied Systems (IIASA) over the period 1976-1982. This project constitutes the first rigorous demographic investigation of intranational redistribution patterns in a comparative national context and is widely regarded as one of the most important demographic analyses of migration and redistribution to be undertaken in decades. Contents Part 1: Introduction. Data and Accounts. Part 2. Components of Change. Mortality. Fertility. Migration. Part 3. Multiregional Analysis. Population Projections. National Case

Studies: The United Kingdom, The Soviet Union, and Canada. Migration and Urban Change. Part 4: Multiregional Mathematical Demography. A Short Course on Multiregional Mathematical Demography. Life Tables. Spatial Population Dynamics. Conclusion. Glossary of Parameters and Variables. References. Author Index. Subject Index. :~:::::::

Contributors

A. Rogers, F. Willekens, P. Rees, M. Termote, J.K. K/m, L.J. Castro, S. Soboleva, P. Korcelli, J. Ledent, and K. L. Liaw.

D. REIDEL PUBLISHING COMPANY A Member of the Kluwer Academic Publishers Group

Nii~iiiiii;i;iiii!;ii;;ii;i;i;iil;i;iiil;;i~ i;;ii;;iiiiii[;iiiii~iii;iilil;i;;!i;i~l;iili!;i

P.O. Box 17, 3300 AA Dordrecht, The Netherlands 190 Old Derby Street, Hingham MA 02043, U.S.A. Falcon House, Queen Square, Lancaster LA1 1RN,U.K.

i!i!iiiiiiiiiiiiiilililiii!il;iiiiiiiiiiii~ iiiiil;ilililililiiiiiiiiiiiiiiiiiiiiiiii