UNIVERSITY OF THE WITWATERSRAND
 A F R I C A N S T U D I E S I N S T I T U T E
 African Studies Seminar Paper
 to be presented in RW
 4.00pm AUGUST 1989
 Title: Relocation in Tseki & Phuthaditjhaba: A Comparative Ethnography
 of Planned & Unplanned Removals in QwaQwa.
 by: Sakkie Niehaus
 No. 260
RELOCATION INTO TSEKI AND PHUTHADITJHABA: A COMPARATIVE
 ETHNOGRAPHY OF PLANNED AND UNPLANNED REMOVALS IN QWAQWA.
 Isak A. Niehaus
 Department of Social Anthropology
 University of the Witwatersrand
 The dominant theme in the recent history of Qwaqwa has been the
 relocation of thousands of people from "common" South Africa into
 the "homeland". In 1929 Ross (1930: 88) estimated that the
 population of Witzieshoek (as Qwaqwa was then known) was 8 000
 and wrote:
 In the past few years many folk who could make no
 existence on the farms outside have moved in. They are
 the old folk and the poor. This would have to be
 prohibited because Witzieshoek is already full, or
 Witzieshoek would have to be enlarged by the addition
 of nearby farms. (My translation)
 Subsequent estimates of Qwaqwa's population are: 23 860 in 1970
 (BENSO 1978: 9), at least 300 000 in 1980 (Murray 1985: 289) and
 between 400 000 and 500 000 people in 1982 (Sharp 1982: 13).
 These figures are suggestive of mass population relocations
 on a scale almost unprecedented in South African history.
 It is generally possible to distinguish three sending areas.
 Murray (1985:290) estimates that perhaps 60% of people relocated
 to Qwaqwa came off white-owned farms, particularly in the
 northern, eastern and northwestern districts of the Orange Free
 State (OFS). The remaining 40%, he suggests, came from towns such
 as Harrismith, Bethlehem, Senekal, Vereeniging, Klerksdorp and
 Bloemfontein. A relatively small number of people have also been
 relocated from other "homelands". These include about 2 000
 Basotho who came from Thaba 'Nchu in 19 74 to escape harassment by
 Bophuthatswana Authorities (Niehaus 1984: 20-22) and several
 hundred Basotho and Xhosa refugees from Herschel in the Transkei,
 fleeing the reported harshness of Matanzima's rule and resisting
 the removal of their South African citizenship (vide Streek and
 Wicksteed 1981: 50-56).
 "Forced" removals mainly affected those removed from urban
 areas. People have been evicted from townships for offences
 relating to reference books and influx control. "Old townships
 attached to a number of towns in the OFS have been deproclaimed
 and razed to the ground, in the course of which many of their
 inhabitants have been relocated to Qwaqwa" (Surplus People
 Project 1983: 159). In this way the Department of Bantu
 Administration and Development had "assisted in resettling" 56
 229 Basotho in Qwaqwa by 1976. (Murray 1985: 290) .
 For those relocated from rural areas and other "homelands" Murray
 (1985: 290) writes: "probably the majority of people have removed
 voluntarily in the sense that they were not physically removed by
GG (South African government) vehicles." Rural removals need to
 be understood against the backdrop of increasingly-mechanized
 production and a constant decline in the demand from agricultural
 labour since the 1960s. The basis on which farmers were prepared
 to employ labourers had become steadily less acceptable as they
 were increasingly denied access to agricultural land and grazing.
 In their struggle to find farmers prepared to offer them more
 favourable terms of employment people often moved rapidly between
 different farms. Being denied entrance into OFS townships,
 thousands of these people made their own way into Qwaqwa in the
 hope of acquiring migrant labour contracts. Indeed, many had paid
 large sums of money to private carriers for transport to the
 "homeland". In this context, Sharp (19 82: 14) has expressed
 caution over the interpretation of such removals as "voluntary".
 On the one hand, he argues, one does not wish to deny the forces
 shaping the lives of people. On the other hand, it is important
 to recognize the extent to which people were able to decide about
 when to move to Qwaqwa.
 This paper is particularly concerned with highlighting the
 diverse implications of relocation for different sections of
 Qwaqwa's population. Despite the conceptual difficulties involved
 the the concepts "forced" and "voluntary", it is my intention to
 show that such distinctions may indeed be valuable in
 understanding relocation processes. It is argued that "official"
 removals imply at least some prior planning by the state to cater
 for the needs of those affected. Evidence indicates that people
 subject to forced removals from OFS townships by the state have
 enjoyed favourable access to resources and amenities in the
 'homeland'. Such people are highly noticeable and taken account
 of by government and administration in the subsequent allocation
 of services in the 'homeland'. By contrast, former inhabitants
 OFS rural areas who have unofficially made their way into the
 'homeland's' closer settlements have been severely disadvantaged.
 Their removals were unplanned and no administrative provisions
 had been made to meet even their most basic needs. Subsequently
 they have been largely unnoticed in the allocation of services.
 This paper reports the results of fieldwork undertaken in the
 Apollo (plural Diapollo) housing section in the town of
 Phuthaditjhaba (from August 1984 until February 1985) and the
 closer settlement of Tseki (in April, June-July and December
 1983). The housing section was opened for settlement in March
 1971. Initially it was occupied mainly by former inhabitants of
 Harrismith who were forcefully removed to Phuthaditjhaba on SADF
 trucks when the old location was destroyed. In Mare's (1980)
 terms these present a classic case of "urban removal".
 Subsequently many former residents of townships in common OFS had
 also moved into the housing section. Of the 300 Apollo houses, 75
 (25%) were sampled. As the houses were laid out in six rows, a
 25% sample was taken in each row. In contrast, most of the
 estimated 15 000 inhabitants of Tseki were relocated from 1974
 onwards and had come from white owned farms. Hardly any had been
 directly removed by agents of the state. During the period of
 fieldwork the members of 40 households were interviewed in depth,
 and on several occasions during the year. These households, which
 occupied 38 different residential sites, were randomly selected
from some 2 700 demarcated residential sites in the closer
 settlement. Despite the small sample size in Tseki, I feel
 justified in presenting basic quantitative data where this is
 appropriate, but I am more interested in qualitative data derived
 from the study.
 My account is mainly ethnographic. It proceeds with a comparative
 account of differential experiences of relocation. Description is
 devoted to local government, employment opportunity resources and
 forms of cooperation among residents of these reception sites.
 DIFFERENTIAL EXPERIENCES OF RELOCATION
 The Schoonplaas Removals. The vast majority of households who
 occupied homes in the Apollo section of Phuthaditjhaba shortly
 after their construction in 1971 were forcibly removed from the
 old Schoonplaas location of Harrismith. With the deproclamation
 of Schoonplaas 300 homesteads were destroyed. 165 households,
 classified as Zulu, were relocated at Madadeni in Kwazulu, and
 135, classified as South Sotho were relocated in Phuthaditjhaba.
 These households (approximately 2 000 people) were moved by
 lorries from the Departments of Bantu Authorities and Defence
 (The Friend 20 March 1971).
 The Police Colonel in charge of the Schoonplaas removals was
 quoted as having said that they proceeded without "major
 incidents" and that people co-operated with the authorities. "As
 soon as they knew the exact date that they would be moved, they
 started demolition work on the houses themselves" (The Friend 20
 March 1971).
 Oral testimony by informants contradicts the notion that people
 were content to be moved. Recollections of these events elicited
 responses such as: "They should rather have killed us"; "If the
 government ever decided to move us back to Schoonplaas I'll be
 the first to go"; and "I don't want to talk. Thinking about these
 things makes me feel too heartbroken". Evidence also shows
 considerable resistance to relocation.
 Schoonplaas was far from visibly attractive. Hawkins (1982:159),
 for instance described the location as:
 A village of hovels and shacks brightened only by
 an occassional garden plot...It was never a beautiful
 place and a visitor coming to Harrismith by train
 exclaimed upon seeing it: If this is Harrismith I am
 sorry I came", but it served the needs of the Natives
 living in town.
 In retrospect, former residents were nostalgic about life in
 Schoonplaas, identifying low housing rents, the large homes and
 gardens, and the opportunity to keep cattle, as admirable aspects
 of the old location. Self-built homesteads, for which as little
 as Rl.50 site rent was paid per month in 19 70, sometimes
comprised up to eight rooms with large yards in which residents
 planted potatoes, cabbage, sorghum, spinach and mealies. In
 addition many informants kept stock. Some paid nearby farm owners
 to let stock graze on their pastures. Others kept stock in their
 yards at night and hired sheperds to herd them on a nearby
 commonage during the day. An elderly informant, recalling that he
 and his father had kept 25 cattle, jokingly said: "We ate only
 meat everyday. We should actually have changed to vegetables".
 Informants also longingly spoke of the times when, with their
 dogs and knobkerries, they hunted dassies and hares on the
 Platberg and fished in the Wilge river with rods made from the
 branches of poplar trees. It was also said that hardly any
 unemployment existed in Schoonplaas. "It only depended on whether
 you wanted to work or not."
 After residents had heard about the intentions of government to
 destroy the old location, a commitee was formed to oppose the
 proposed removals. Meetings were regularly held in a local school
 hall on Thursday afternoons and in the yard of the commitee
 leader on Saturday mornings. By all accounts these meetings drew
 enthusiastic support. One informant recalled:
 We were always told when there was a meeting. Many
 people attended and the hall was full every day...
 First of all the committee members talked and then
 the sibonda (leader)...They said that the place
 should not be destroyed because it belonged to us
 and we built the houses ourselves...Then after the
 meetings the people did African dances and sang songs.
 She recalled that the following songs were sung:
 Mahlomola a Qwaqwa. The loneliness of Qwaqwa
 E rona tjhaba santsho We the Black nation
 nomerso o re kwaetetse ka thata. are being opressed.
 A makgo ngwana Mopeli. Oh! The children of Mopeli.
 Ha Mopeli e nyane e kgolo At Mopeli's place is a small
 ntho e tsekwang and big bone of contention.
 In an attempt to prevent demolition the committee collected
 donations and paid a Ladysmith attorney to negotiate their
 demands with the representatives of local government. By November
 1970, however, it became apparent that relocation was inevitable:
 the Administration Board offices of Schoonplaas were demolished
 and water supplies disconected.
 The first police vehicles which arrived in Schoonplaas were
 stoned. The following day a large crowd gathered and attempted to
 march to the municipal offices in town to voice their discontent.
 The procession was broken up by the police, who arrested a few
 committee members and wounded two women in their legs. As a
 result the residents became demoralized: "The very same people
 who were fighting helped the police load their furniture". By
 April only a dilapidated graveyard remained. "They destroyed our
 houses, schools, churches and everything else".
Some Schoonplaas stockholders refused to board trucks. Instead
 they chose to drive their animals on foot to Qwaqwa in the hope
 that grazing lands would be obtainable at their destination. The
 journey cut through the veld and fields of farm owners and
 reportedly took an entire day. Household members, having been
 brought to Qwaqwa by truck, were assigned directly to Diapollo
 homesteads, offloaded and granted rations comprising a loaf of
 bread and a pint of milk. Initially cattle grazed south of the
 Diappolo, but stock keepers were soon forced to sell these as
 further housing was constructed in Phuthaditjhaba.
 The Diapollo was not exclusively occupied by former Schoonplaas
 residents. People removed from Harrismith arrived in the housing
 section at the same time as households who had resided in a
 shanty town northeast of Phuthaditjhaba since the 1960s. Some,
 under duress removed due to employment transfers or
 contraventions of influx control laws. Others moved to obtain
 business licences or ensure schooling for their children. Living
 conditions in the shanty town were appalling. Here, households
 had lived in one-roomed corrugated-iron shacks subject to extreme
 variations in temperature. One man recalled that "sometimes it
 was so hot that we had to sleep outside and when it was cold we
 had to use a brazier (paola)". In contrast to Schoonplaas
 residents, shanty town dwellers were delighted when informed by
 the Phuthaditjhaba housing superentendent that Apollo houses had
 been made available for them. 35% of sampled households arrived
 after 1971 to occupy homes vacated by their original tenants. The
 tenants who had left had gone either to secure more desirable
 accommodation elsewhere in Phuthaditjhaba, or had moved to closer
 settlements because they had been unable to afford housing rents
 in the townships.
 Removals from the OFS countryside All of the 40 households in
 Tseki had resided on White-owned farms in various parts of the
 OFS prior to relocation. 3 2 removed directly from farms in the
 Petrus Steyn, Viljoenskroon, Reitz, Bothaville and Clocolan
 districts. Of the remaining eight households three had come after
 a brief and unsuccessful attempts to secure a foothold in an
 urban area in the province. The remaining five had come from
 Thaba *Nchu in Bophuthatswana. The members of these households
 were part of a huge influx into the 'homeland1, beginning in the
 late 1960s. People in the study were questioned closely about the
 circumstances of their removal from farms. Their accounts
 confirmed the general trend of a decline in the demand for
 agricultural labour and in real wages on White-owned OFS farms in
 this period. Their accounts also indicate the specific factors
 through which these trends were mediated in local instances and
 in the people's own experience. The following reasons for removal
 to Qwaqwa were most frequently cited by informants:
 1) Many perceived a decline in total wages on farms in the early
 1970s, owing to the stringent enforcement of stock limitations or
 to a total ban on stock holding by workers in this period. Most
 people recalled that, at some point in the late sixties or early
 seventies, farmers had instructed their workers to sell off most
 or all of the livestock they had hitherto been allowed to. graze
on the farmers' land. Some informants were quite precise about
 the events which had led up to this, showing that they had a good
 insight into the changing situation of their employers. One man
 explained that 'the "baas" had just bought new machines for
 milking his cows, which meant that he wanted all the grazing for
 himself. He called all the workers together, and told us that we
 "kaffirs" had too many animals on his land. He said we could keep
 two cows each, and he would sell the rest for us at the auction',
 Another ex-farm worker, who had kept a herd of five cattle and
 earned about R40 per annum from the sale of calves to 'whites in
 the town of Senekal1, recalled that he was told to get rid of all
 his animals by his employer in 1973. In compensation he and his
 fellow labourers were offered a cash increase of R2 per month in
 wages. According to him many of the twenty households employed by
 the farmer left in response to this measure. He maintained that
 he had been able to buy clothes for himself and his mother from
 the sale of livestock in earlier years, and argued 'how can you
 buy clothes for R2? I stayed on the farm for another year because
 my mother did not want to leave. But then I saw we were just
 becoming poorer, and I persuaded her to leave1.
 2) Several informants noted that after the sale of farms they had
 worked on, the new farm owners told them to move off the land
 because they 'brought along their own workers from their other
 farms'. Displaced workers were aware that their previous
 employers had been bought out by wealthy land-owners with several
 farms. 'The new "baas" was a very wealthy man - he even had his
 own aeroplane, for which we made a runway near the house. He was
 also very mean; he said we were all too stupid to work for him
 and that his own workers were much better1.
 3) Old people who were nearing retirement in the early 1970s,
 complained that the farmers' treatment of the elderly had become
 more harsh in recent years. In earlier years old people had been
 able to stay on farms and use the latters1 resources until death,
 in recognition of a lifetime's service to their employers. In the
 late 1960s, however, farmers began to expel such persons. Several
 informants linked this change in attitude to the fact that
 farmers came, at about this time, to hear that Qwaqwa was
 intended as the 'homeland' for the 'Basotho1. One pensioner in
 Tseki explained what had happened to her:
 These farmers were now only satisfied if you could
 work, but if you were too ill or could not work,
 they were not. Even if you were struggling you could
 not write a letter to a person here in Qwaqwa. Then
 the farmer would say: 'No. Now you can join them. You
 may now walk to the Basotho (sic) f. If the farmer
 had not been cruel I would still be living there.
 4) The sample included people who, in response to the perceived
 decline in conditions on farms, had decided to move to Qwaqwa on
 their own account, in the hope of securing migrant labour
 contracts to the industrial centres of common South Africa.
 People had heard about the rise in wages on the mines, and in
 industry generally, in this period. They were tempted by the
 prospect of a better future for them and their children: and
 complained bitterly about the absence of school facilities on or
near most farms.
 People on the farms had also heard the grandiose claims broadcast
 by Radio Sesotho, which urged them to join 'their1 chiefs in the
 ?homeland1. One person recalled that:
 The radio said: fCome and enjoy yourself in Qwaqwa.
 All the black people must come to this homeland where
 they will find fields and lots of jobs. Your chiefs
 will look after you1.
 Given the possibility of earning higher wages in other sectors of
 employment as migrants based in Qwaqwa, it was frequently the
 younger people who had paid most attention to these inducements
 to leave the farms. On the other hand, many of their older family
 members had been most reluctant to quit farms where they
 themselves had not been threatened with eviction. The decision to
 leave a farm was frequently, it appeared, a source of domestic
 conflict within family groupings. In several instances in the
 sample this conflict had endured until the time of fieldwork.
 5) There were also cases in which households had been expelled
 after younger members had left to take up migrant labour
 contracts while their dependants and elders continued to reside
 on the farms. In these instances the responses from fanners were
 uneven, depending on their differing needs for seasonal and other
 forms of 'casual1 labour. Some farm owners were reported to have
 encouraged the sons of farm labourers to engage in labour
 migration, so that their remittances could help to support family
 members whose labour was only occasionally required by the
 farmers themselves. Others were greatly opposed to farm-based
 labour migrants. One informant mentioned that, on his farm, the
 police were frequently called in to deal with women whose
 husbands were away working: 'Sometimes these raids took place
 during the day, but mostly after midnight. The women were taken
 to the courts and fined R20 or twenty days for trespassing1. He
 continued: fIt was impossible for the women to join their men in
 the compounds. They had no option but to stay on the farm and
 risk a second fine of R40'. In this manner his own daughter-in-
 law had been arrested and fined several times, so that eventually
 the whole family had decided to move to Qwaqwa.
 6) Several people recorded their impression that, as a result of
 the developments mentioned above, there had been a general
 increase in the incidence of quarrelling between farmers and
 their workers, and amongst workers themselves, in the period in
 question. They also claimed that cruel treatment of workers by
 farmers was on the increase at this time, with the latter more
 ready than in earlier times to resort to sjamboks and other forms
 of punishment for workers who were tardy in doing their work or
 broke the various norms of behaviour established, and changed at
 their own whim, by individual employers. In this period, as the
 people saw it, the old relationships of tolerant paternalism and
 willing subservience were finally broken down, and people who
 were unable to adjust to this left the farms and made their way
 to Qwaqwa.
EARLY EXPERIENCES IN PHUTHADITJHABA AND TSEKI
 Differential processes of relocation had vast implications both
 for early experiences, and for subsequent developments, in
 Qwaqwa. In the case of "forced" removal the reception of former
 Schoonplaas residents was planned for. They were accommodated in
 state-provided houses in Phuthaditjhaba. By contrast the arrival
 of former people from the OFS countryside in 1974 was
 unanticipated. They were arbitrarily assigned to, and left to
 their own devices in Tseki. The subsequent provision of
 infrastructure and "development" has definitely favoured former
 Schoonplaas residents. In this respect they have been highly
 visible, whereas former residents of the countryside have been
 systematically ignored.
 Experiences in Phuthaditjhaba. People removed into Phuthaditjhaba
 in 1971 were accomodated in 300 semi-detached, four-roomed,
 double-storey houses. These houses were built by the Orange Vaal
 Administration Board (OVAB) starting in 1970. Of them The Star
 (16 July 1973) notes:
 They come as a surprise in the dun, winter-coloured
 valley below the Lesotho border. The little houses are
 quite attractive, painted rust and ochre and cream.
 Locally these houses soon came to be referred to as the Diapollo.
 Jokingly, the name makes explicit reference to the American space
 shuttle which landed on the moon two years prior to the
 relocation. It expresses feelings about the height and size of
 these new homes.
 Initially the Diapollo lacked sanitary facilities, having only a
 single outside stand pipe tap. In contrast to Schoonplaas homes,
 which were supplied with a bucket toilet system, residents were
 forced to dig pit-latrines in their yards. In 19 77, however,
 flush toilets were built in outside rooms previously used for
 storage purposes and wash basins were installed. The monthly
 rents for these homes, including water, amounted initially to
 R3.50. But there were various subsequent increases: to R6 in
 1975; to RIO after the home improvements of 1977; and to R16.60
 in 1981. The first of these three increases evoked much
 discontent. Residents claimed that they had not been informed of
 the increase, marched to the housing office and refused to pay
 arrears. Some people went to court, others were given enough time
 to pay.
 In Phuthaditjhaba the building of Apollo homes ceased in 1974.
 They had proved unpopular among residents; elderly people
 complained that, in the dusk, it was dangerous to climb the steps
 leading up to the bedrooms on the second storey. The numerous
 people who had arrived in the town subsequently were housed in
 4000 residential units also built by the Orange Vaal
 Administration Board. These are more spacious than the Diapollo
and of the N51/9 type found in most 'homeland1 towns, and are
 locally referred to as either of the Single (bungalow) or Double
 (semi-detached) .
 With the development of Phuthaditjhaba as a major town, residents
 of the Diapollo have benefitted greatly from a number of
 "development projects". On the initiative of the Qwaqwa
 Development Corporation (QDC) three industrial parks have been
 established on the outskirts of the town. QDC involvement in
 trade has led to the establishment of a number of tripartite
 companies, a beer hall, the "international" Hotel Qwaqwa and two
 major shopping centres. Qwaqwa government departments have been
 responsible for the erection of two teachers1 training colleges,
 an old age home, a town hall, a great number of schools, an
 ostentatious new magisterial complex, a roofed central bus
 terminus, and have installed a network of powerful mast lights
 which illuminate the town's streets at night. Construction of a
 new hospital in Phuthaditjhaba began in 1977. The lavish five-
 storey building has been planned to accommodate 450 beds and will
 be fitted with the most up-to-date medical equipment. In 1981 the
 estimates costs of the project were R18 million (The Friend 3
 March 1981). These projects have served to accentuate the
 difference between township and closer settlement residents.
 Early experiences in Tseki. People who came to Tseki were given
 no choice about where to settle. Trucks from the farms were
 instructed by the Bakwena Tribal Authorities and chiefs in other
 closer settlements to come to Tseki, because it was the area
 which was being filled at that time. Upon arrival the names of
 newcomers were registered by the local headman, or his wife who
 deputized in his absence. Thereafter they were handed a piece of
 paper which indicated the number of their particular site. Most
 informants mentioned that they had had to pay the headman or one
 of his iinduna (assistants) a sum which ranged from 50c to RIO,
 for directing them to their site. According to the headman's
 records over 2 000 households moved into Tseki in this manner
 between May and October 1974 (Rand Daily Mail 2 October 1974).
 This means, we estimate, that well over 10 000 people were
 dumped, over a brief period, in a confined area where they were
 left to fend for themselves. Those who arrived usually slept the
 first night in the truck which brought them, after which they
 were forced to sleep in the open until they had built themselves
 their first shelter. Tin shanties were constructed from materials
 brought from their previous place of residence. In the early
 days many shacks collapsed from the wind, posessions were jumbled
 up, and many lost in the confusion. Houses were later rebuilt
 with mud bricks.
 People relocated to Tseki were by no means "resettled". This was
 made clear by the headman's wife in a newpaper interview in which
 she pointed to the absence of necessary amenities in Tseki. She
 was quoted (Rand Daily Mail 2 October 1974) as saying:
 We are worried about the increasing numbers. It will
 ultimately bring slum areas into Witzieshoek due to the
 lack of sanitary, schooling and other facilities
which should be provided for the incoming families.
 I am wondering what to do. I envisage a breakdown of
 the whole system as more and more people arrive causing
 the government of Basotho Qwaqwa more embarrassment.
 People soon discovered the fraudalunce of Radio Sesotho
 broadcasts: "The radio lied. Here there was no grazing for cattle
 and people had to wait a long time for work". Informants revealed
 that numbers of people returned to the farms soon after they
 arrived. Informants commented:
 The people were so angry that they returned to the
 farms. Not to the farms which they had left, but
 to any place which would take them. Some of my
 friends left within six months after getting here. If
 Qwaqwa gained independence I would also leave, even
 though I know what it is like now on the farms.
 Many people would try to run away...Nowadays many
 people want to return (to the farms), but Mopeli
 does not want such things to happen. He would stop this
 by making some policemen stand on the Harrismith road.
 There was no clean water in Tseki in 1974 and, in the absence of
 taps, people were forced to fetch water from a few wells on the
 nearby mountain slopes. Informants recalled that people got
 rashes, and sores in their mouths, from drinking the well water,
 which rapidly became polluted.
 We heard that the water was killing people...during
 that time my sister-in-law just collapsed while walking
 around. We don't know the reason why she died. It might
 have been the water, or because we had to pay a lot for
 transport to get here and had no money to buy food with.
 Many kids died from drinking that water. Sometimes when
 you were sitting you might see three or four coffins
 passing by.
 Taps were provided late in 1975. Although the taps were few in
 number, and fetching water was still an arduous and time-
 consuming occupation in the late 1980s, people noted the
 immediate improvement in health and hygiene which resulted from
 the provision of clean water.
 People who had been used to a free and abundent supply of
 firewood on the farms, were appalled by the lack of fuel. The few
 trees in the area were rapidly consumed, and people had to go
 further and further into the mountains to fetch wood. After a
 few years small businesses selling wood and coal were established
 in the area, but in 1983 a small bundle of wood ("enough for two
 days") cost R2.50, and the price of a bag of coal ("which lasts a
 week when it is not too cold") was R4. Most households could not
 afford to spend R12-15 a week on fuel, and most women were still
 walking to the mountains to bring back wood. By the 1980s people
 had to go all they way to, and sometimes over, the Lesotho, border
 10
to find a ready source. Women banded together to make a journey,
 and armed themselves with sticks and axes, claiming that
 "dissidents" form Lesotho had attacked and raped women collecting
 wood on their own in the past. The round trip took more that
 eight hours, and on numerous occassions women were stopped by the
 headman or his iinduna as they re-entered the closer settlement,
 deprived of their wood, and fined for "creating tensions on the
 border".
 The one facility which was provided quickly in Tseki was schools.
 Four- comprising one primary, one preparatory, and two high
 schools- were established in the mid-1970s. Indeed, as Robinson
 (1983) has shown, schools popped up like mushrooms in Qwaqwa in
 the 1970s. Through this vigorous building campaign educational
 facilities were not provided simply for the children of families
 relocated to the area: SeSotho-speaking pupils whose parents
 continued to reside in common South Africa were also attracted.
 In this way schooling was used as a means to create and foster a
 link between the "homeland" and its far flung "citizens". Tseki
 High School is particularly well known for the boarding
 facilities it offered to scholars from as far afield as Soweto.
 Otherwise, by 1984 "development" in Tseki was confined to five
 poorly-stocked and expensive general dealer stores, a bottle
 store, a "village garden", medical clinic and police station.
 Unlike inhabitants from Phuthaditjhaba, commuters from the closer
 settlement continue to use bus stops without shelter from wind or
 rain. Outpatients from Tseki are also unlikely to find the new
 hospital any more accessible than they did the old one.
 EMPLOYMENT OPPORTUNITY RESOURCES
 To both Diapollo and Tseki residents the most drastic impact of
 relocation occurred in the field of employment. Here the impact
 of relocation was also far from uniform. Since the early 1970s
 Diapollo residents have enjoyed advantageous access to existing
 employment opportunity resources.
 Most wage earners relocated in the Diapollo retained their jobs
 after relocation. With the destruction of Schoonplaas employers
 in Harrismith were informed by government that their employees
 were to be relocated to Qwaqwa. For these purposes they had
 granted their employees a few day's leave. However, these wage
 earners now had to commute a distance of approximately 50kms
 between their place of residence in Qwaqwa and workplace in
 Harrismith. In response to employer demands a hostel
 accommodating 50 men was constructed in Harrismith's new African
 township. Other workers secured accommodation with relatives in
 4 2nd Hill, or at their workplaces in town (domestic workers and
 hotel employees). They travelled to Phuthaditjhaba over weekends
 only. However, most workers had become daily commuters. At the
 time of relocation Qwaqwafs Setsokotsane Bus Company had only one
 bus on the Harrismith line (Informa 1982: 50). To meet the
 increased need for transport the Bantu Investment Corporation
 (BIC) donated four new Leyland buses to Setsokotsane. Subsidized
 bus fares were also negotiated for daily commuters (The Friend 20
 11
March 1971). Nonetheless workers found it possible to commute
 only with great difficulty. The poor condition of the untarred
 Harrismith road frequently led to bus breakdowns, and buses were
 in constant need of repair. Buses were overcrowded and slow. Of
 the 21 informants who were employed in Harrismith whilst resident
 in Schoonplaas, five had lost their jobs subsequent to
 relocation. Some said that because buses were too full they had
 been unable to arrive at work regularly and were dismissed.
 Others said that they did not always have enough money for bus
 tickets. Moreover when commuters worked overtime or night shifts
 they were unable to take buses home as the service did not extend
 into late night hours. Case 3.1 below illustrates some of the
 extreme difficulties faced by commuters during this period.
 Case 3.1: The dillema of a shift worker, 1971-1975.
 From 1971 to 1975 EM worked as a manual labourer at the
 Harrismith agricultural Co-Operative. "My job was to put sacks
 under machines which filled them with mealie meal...Luckily there
 was not a lot of work to do, but they only paid me RlO per week
 and that was not enough." EM's dillema was that he had to work
 alternate shifts weekly: from 06h00 to 14h00, and from 14h00
 until 22hOO. When working the morning shift he usually woke at
 03h00, took the 04h30 bus to Harrismith and the 15h00 bus home.
 However, when he worked the evening shift he could not find a bus
 to Qwaqwa until 04h00. "I'd first sleep at work and then take the
 bus. At home there would be no time for sleep...I'd just look for
 food, buy wood and coal for my family to make a fire and then
 work in the garden before taking the eight o'clock bus back to
 Harrismith...At work we slept on empty mealie bags. But the
 manager did not want us to sleep there and we sometimes also hid
 ourselves in the waiting room at the railway station. But the
 station master told us that the waiting room was only for people
 waiting for the trains. Then we again went back to hide ourselves
 at the Co-operative and to sleep on those empty bags. When the
 manager found us he scolded us. We tried to explain the transport
 problem, but he just said: You must make use of your own means of
 transport to go to Qwaqwa if you want to work here...I spent all
 those nights without sleep... my eyes were tired and just stared
 in front of me." EM was eventually dismissed for stealing hessian
 bags at work.
 In this context people who were formerly employed in Harrismith
 increasingly took up migrant labour contracts to work in the
 major South African industrial centres. A TEBA labour recruiting
 office for the goldmines of the OFS and the Transvaal, as well as
 a district labour bureau at the magisterial office had been in
 operation in Phuthaditjhaba since the 1960s.
 For people relocated into Tseki, adjustment was a far more
 harrowing experience. For most relocation meant a complete loss
 of a wage income and a change from predominantly farm to
 industrial employment. Such conditions were exacerbated by the
 fact that there was no labour bureau in Tseki for the first two
 years after relocation. A bureau administered by the Bakwena
 Tribal Authorities was opened in the area only in 1976. Yet this
 12
bureau was seldom visited by employers. Whereas the
 Phuthaditjhaba bureau offered a total of 8 489 contracts in 1983,
 only six migrant labour contracts were attested at the closer
 settlement bureau in the first half of 1983. Evidence from Qwaqwa
 hereby bear out the findings of Greenberg and Giliomee (1983)
 that Tribal Authorities exercise a limited role in the
 recruitment of labour from the "homelands". Over half of the
 workseekers in the households we studied took more than six
 months after arrival to gain wage employment; nearly 20% of these
 workseekers stayed unemployed for more than a year. During this
 time households without access to wage remittances were forced to
 rely exclusively on savings, money gained from the sale of
 livestock before arrival, and the help of kin and neighbours.
 This is illustrated in case 3.2 below.
 Case 3.2. Passing through a period of Distress
 AM's father worked on a farm in the Bothaville district of the
 Free State in the early 1970s. The farmer gave AM permission to
 work in Welkom and to leave his wife and children on the farm.
 The farmer died in 1973. The farm was sold, and the new owner
 expelled AM, who told us that "the new farmer did not want people
 to live on the farm who did not work for him". AM said that he
 had no option, but to hire a truck and come to Qwaqwa, as this
 was the only place where he could obtain a permanent place to
 live. AM's father, who was some years off retirement, refused to
 accompany his son, and one of AM's daughters, who was 15 at the
 time, also remained on the farm because she was employed as a
 domestic worker by the farmer's wife.
 Soon after arrival in Qwaqwa misfortune struck. AM lost his job
 in Welkom owning to the unplanned period of absence from work at
 the time of relocation. He was unable to find another job for
 more than a year. His children were too young to work, and his
 wife was unable to look for work because she had to care for
 them. Their savings lasted only a short time, after which the
 household was without a cash income. "At that time we were
 struggling", AM's wife recalled. She said that in the bitter
 winter of 1974 they had been reduced to burning some of their
 furniture to keep warm. During that winter AM's youngest two
 children died, and the older ones suffered from serious
 malnutrition. AM's wife took them to the doctor in Harrismith
 once, but the doctor told them that their illness was "a wrong
 way of eating", and advised her to feed them milk, meat and fresh
 vegetables. "His advice was no good", she said "I knew what was
 wrong with them, but we had no money for these things."
 AM's first response to this crisis was to send two of his
 children back to the farm to be cared for by their grandparents,
 who fed them and sent them to the farm school. "We were lucky
 that my father stayed on the farm", AM said, "because many others
 round here had left as whole families.. When their children were
 ill there was nowhere to send them". AM and his wife also asked -
 their neighbours for assistance, and said that a great deal of
 help, in the form of money and food, had been forthcoming. "In
 those early months", AM explained, "We all stood together. Now
 this has changed: if people see their neighbours hungry today,
 13
they look the other way".
 In 1975 AM found work at a construction firm that built a college
 near Tseki, but was again unemployed for four months after the
 contract was completed. Towards the end of 1975 he left Qwaqwa
 illegally to look for work independently, and found a job at a
 plumbing concern in Welkom. Once he began to remit home, the
 physical survival of his household was no longer threatened, but
 he complained: "I did not know what I was working for...I just
 worked for the neighbours to return their money".
 AM's strategy of securing a migrant labour job by "unofficial"
 means was the option most frequently embarked upon by Tsekifs
 wage earners. Many men ignored Tseki's labour bureau, and
 travelled to the bureau in Phuthaditjhaba instead. The mother of
 an unemployed young man explained:
 If you are eager to go on contract you go to the
 Phuthaditjhaba labour bureau every day. My son was
 anxious to get work, and walked there since the
 beginning of the year, because he did not have the
 money for busfare. Only some days when he was tired
 did he not go; then he just walked down to the Tseki
 labour bureau. But that was just an excuse to get
 out of home and go to talk to the other young men;
 there are never any contracts in Tseki.
 Other workseekers bypassed both the local and district labour
 bureau and went on their own to look for work in common South
 Africa. By doing so they risked arrest under influx control
 legislation. They hoped to find an employer who would have their
 jobs regularized retrospectively by specific requisition.
 The advantageous position of Diapollo residents with regard to
 employment opportunity resources is borne out by Tables 3.1 and
 3.2 below. Table 3.1 depicts the access of households in these
 respective areas to wage incomes in the 1980s. It shows that
 Diapollo households (despite smaller sizes than those in Tseki)
 had greater access to wage earnings, 8% of Diappolo households
 were without any access to wage earnings and 13% of Tseki
 households. 52% of Diapollo households, in contrast to 31% of
 Tseki households, had access to the earnings of two or more wage
 workers.
 Table 3.2 depicts the distribution of people of working age in
 the Diapollo and Tseki samples between different places of
 employment. From the table it is evident that Tseki had a larger
 proportion of unemployed people, with women from the closer
 settlement being particularly disadvantaged: 69% were unemployed
 in contrast to 25% in the Diapollo. Whereas the periods of
 unemployment experienced by Diapollo residents were fairly short
 term, many Tseki workseekers had been unemployed for
 exceptionally long periods. At least two workseekers in the
 sample had been unemployed for nine years i.e. ever since their
 arrival in Tseki in 1974. Diapollo wage earners were employed in
 14
TABLE 3.1: NUMBER OF WAGE EARNERS PER HOUSEHOLD IN THE DIAPOLLO
 AND TSEKI SAMPLES BY SIZE OF SUCH POPULATIONS (1983-1985).
 Number
 of wage
 earners
 0
 1
 2
 3
 4
 7
 TOTALS
 Number
 of
 DIAPOLLO (1984/85)
 Household sizes
 households Range
 6
 29
 19
 13
 5
 1
 73
 ( 8%)
 (40%)
 (26%)
 (18%)
 ( 7%)
 ( 1%)
 (100%)
 2- 7
 3- 9
 2-10
 3-11
 7-11
 17
 2-17
 Mean
 4.6
 5.3
 5.2
 8.0
 9.2
 17
 6.3
 TSEKI
 Number
 of
 households
 5
 2
 8
 2
 2
 0
 38
 (13%)
 (55%)
 (21%)
 ( 5%)
 ( 5%)
 ( 0%)
 (99%)
 (1983)
 Household sizes
 Range
 1-11
 3-11
 3-11
 6-13
 9-15
 0
 1-15
 Mean
 6.2
 7.1
 7.3
 9.5
 12.0
 0
 7.1
 Notes: (1) All
 (2) Inadequate
 figures are d? jure
 data - 2 Diapollo households
 TABLE 3.2: LABOUR FORCE PARTICIPATION IN THE DIAPOLLO AND
 SAMPLES ACCORDING TO PLACES OF EMPLOYMENT (1983-1985).
 TSEKI
 EMPLOYMENT STATUS
 NOT EMPLOYED
 Attending school
 Health reasons
 Domestic reasons
 Unemployed
 TOTAL
 EMPLOYED
 Inside Qwaqwa
 In Harrismith
 As migrants
 TOTAL
 TOTAL
 DIAPOLLO (
 MALE
 16
 10
 0
 22
 48
 28
 25
 20
 73
 121
 FEMALE
 17
 5
 15
 20
 57
 28
 27
 5
 60
 117
 1984/
 TOTAL
 33
 15
 15
 42
 105
 56
 52
 25
 133
 238 (
 85)
 (%)
 (14%)
 ( 6%)
 ( 6%)
 (18%)
 (44%)
 (24%)
 (22%)
 (11%)
 (56%)
 100%)
 TSEKI (1983)
 MALE
 12
 2
 0
 12
 26
 2
 5
 36
 43
 69
 FEMALE
 16
 4
 10
 18
 48
 3
 1
 4
 8
 56
 TOTAL (%)
 28
 6
 10
 30
 74
 5
 6
 40
 51
 125
 (22%)
 ( 5%)
 ( 8%)
 (24%)
 (59%)
 ( 4%)
 ( 5%)
 (32%)
 (41%)
 (100%)
 Notes: (1) Informants of working age are regarded as older than
 15 but younger than 66 years in the case of males or 61 years in
 the case of females.
 (2) All figures are based on data of de_ jure residents in the two
 areas.
 15
a broader range of jobs: earning from less than R80 per month
 (domestic and unskilled workers) to more than R600 (salespeople
 paid on commission, high school teachers, senior bus drivers and
 clerks in Qwaqwa?s public service). 42% worked locally, 39% in
 Harrismith and 19% as migrants. By contrast only 10% of employed
 Tseki residents worked in Qwaqwa, 12% in Harrismith and as many
 as 78% as migrants.
 These differences are due to a number of factors. By virtue of
 prior residence in urban areas Diapollo residents were better
 schooled and had had considerable experience of the urban work
 setting. Skills acquired in the urban context have enabled them
 to compete successfully for relatively skilled and sought after
 jobs after relocation. By contrast the former residents of the
 OFS countryside were poorly schooled with little prior experience
 of the urban work setting.
 The relative advantage of Diapollo wage workers stretches beyond
 mere skill levels. It is also based upon their favourable
 geographic location in relation to employment opportunity
 resources in and around Qwaqwa. With the inauguration of the
 South African government's decentralization scheme in 19 80 local
 employment had become a realistic option. In terms of this scheme
 Phuthaditjhaba was officially designated a "growth point" and
 three industrial parks were constructed on the outskirts of the
 town Diapollo. By 1985, 83 industries had been established here
 offering 8 561 jobs. Of these 7 061 (82%) were held by women (QDC
 1985:1-4). As workseekers are recruited fronm factory gates on an
 almost first-come-first-serve basis it stands to reason that
 Diapollo residents have easier access to these jobs. They are far
 more likely than closer settlement residents to know where and
 when employment is offered. Very few of Tseki's wage earners
 were employed inside Qwaqwa: in fact none were employed in Tseki
 itself. Informants complained bitterly that local jobs were
 monopolized by people from Phuthaditjhaba. One woman, who had
 been unemployed for four years, looked for work only in the
 industrial parks: "We all go to the factory gates, and then maybe
 the foreman would call one of us and say, "Hey you, Come!". She
 continued, "foremen at the factories select only people they know
 - their relatives, and women from Phuthaditjhaba whom they meet
 in the shebeens."
 Since 1979 employment in Harrismith as commuters had become a
 more feasible proposition than before. In that year a new tarred
 road was completed between Harrismith and Phuthaditjhaba. The
 journey was shortened and less taxing on buses. The scale of the
 Setsokotsane Bus Service had also been considerably expanded. By
 198 4 it operated a fleet of approximately 90 buses. With
 subsequent population relocations into other residential areas in
 Qwaqwa residents of the Apollo have, however, faced increased
 competition for Harrismith jobs. Nonetheless they (particularly
 former residents of Schoonplaas) are placed at an advantage due
 to their knowledge of the town's labour market and ties of
 kinship and acquaintance with the inhabitants of 42nd Hill.
 Commuter labour has not presented itself as a significant option
 to Tsekifs residents who regard Harrismith as a "foreign" town.
 16
Another problem encountered in securing Harrismith jobs was that
 a direct bus line between Tseki and Harrismith was established
 only in 1983. Prior to this commuters had to travel to
 Phuthaditjhaba before departing for Harrismith. Commuters left
 their homes at 3hOO and seldom returned before 19h00.
 Reasons for the greater participation of Tseki residents in
 migrant labour contracts are not readily apparent. It is partly
 because men in the closer settlement had experienced farm-based
 labour migration prior to relocation. Most of Tseki's male
 workseekers had their eyes set on migrant contracts. They were
 often unwilling to take up low-paid and insecure local
 employment, reasoning that if they accepted such jobs they would
 not be available at the labour bureau at the crucial moment when
 an employer arrived with secure and better-paid migrant
 contracts. The higher participation of closer settlement men in
 the labour market as migrants could also be due to the fact that
 labour histories in Tseki were collected earlier than those in
 Phuthaditjhaba. Between 1980 and 1982 there had been a great
 demand for unskilled labour for the construction of the Sasol
 plants in the Northern OFS. Contracts were, however, available
 only for a two year period and jobs were completed upon the
 termination of such plants.
 Diapollo workseekers generally found contracts less desirable.
 They agreed that such contracts often offered higher wages than
 local employment, but pointed to other undesirable factors
 associated with labour migration such as long periods of absence
 from home and spartan living conditions in the compounds. It was
 also noted that higher earnings did not necessarily imply greater
 remittances. Labour histories showed that remittances from
 different migrant workers varied widely. Some workers sent home
 as little as R300 during the year, while others remitted amounts
 in excess of R2 000. Some remitted regularly, and others very
 occasionally. Variation in the kinds of jobs held was only one,
 and not necessarily the most important, reason for the unevenness
 of remittances. In addition to this workers themselves drew
 attention to many other factors which affected the amount and
 flow of remittances from any given worker. Accommodation while at
 work was frequently mentioned. Municipal hostels, mine hostels,
 hostels in small towns, and hostels in or near large cities
 were, the workers insisted, very different environments in which
 to live. Depending on where they lived, there were greater or
 lesser demands on their wages for their own requirements, leaving
 smaller or larger amounts to be remitted to their dependents. One
 migrant who earned R250 per month in Welkom found it possible to
 remit only R60 per fortnight. "Where I stay", he said, "this is
 not easy to do. You have to save as hard as possible. When you
 send money home you don't have enough for yourself, and there is
 usually nothing at all to put into the bank." He explained that
 at the municipal hostels where he stayed no food was provided for
 migrants, but there were kitchen facilities. "Each must provide
 for himself. There are no storage places, and very much food gets
 wasted and stolen when you are at work". Unlike the mine
 hostels, the municipal hostel was close to the township, which
 provided large scope for men to spend their wages before they
 could be sent home. There was, for instance, a beer hall right
 17
next to the hostel, which was often frequented by prostitutes
 from the township.
 SUBORDINATION AND LOCAL GOVERNMENT
 As a result of relocation people have been subject to vastly
 different structures of local government/control inside Qwaqwa.
 Whereas Phuthaditjhaba residents have maintained some degree of
 autonomy from the Qwaqwa government, closer settlements such as
 Tseki are directly subordinate to a system of Tribal Authorities
 and "hereditary" chiefs (the power base of the ruling
 Dikwankwetla Party).
 Since the 1970s the Department of Bantu Administration and
 Development assumed direct responsibility for the management of
 Phuthaditjhaba's affairs. Over and above his other duties the
 district magistrate acted in the capacity of mayor. In 1979 the
 first steps were taken to grant residents some say in local
 government. A Community Council was elected to serve the
 magistrate in an advisory capacity. The magistrate at that time
 noted that in its three year term of office only two Council
 meetings were held. Only twice did Council members approach him
 with problems experienced in the town. He said that he would have
 been surprised had anybody known of the Council's existence.
 In line with developments throughout South African townships at
 that time arrangements were made for the election of a new
 Community Council in 1982. However, after a meeting with
 nominated candidates Chief Minister Mopeli cancelled the
 election. Plans soon got underway for a Town Council to be
 elected in September 1983. The Town Council was to have had
 increased powers including control over the allocation of housing
 and the granting of business licences. It was to have become
 financially "self-sufficient with housing rents as its major
 source of revenue. On 31 September 9 candidates were elected to
 the Council with a percentage poll of 9% among 5 465 registered
 voters. Eight of the successful candidates campaigned under the
 slogan Nna Le Mmuso (I with the government). They were allegedly
 sponsored by Dikwankwetla.
 Councillors ascribed widespread disinterest among
 Phuthaditjhaba's voters to poor communication. Anouncements made
 on loud hailers in the streets had little effect because people
 were at work during the day and notices at the superentendent?s
 office were seldom read. Bank (1984: 50-55) has shown that
 teachers and civil servants chose not to vote due to political
 considerations. They saw voting as an indication of support for
 the "homeland" system. He also shows that most votes came from
 pensioners and the urban poor who sought to forge relations of
 patronage with power holders. They believed that councilors could
 upgrade dwellings, improve tha garbage removal system and take
 firm action against criminal elements.
 18
Although the Diapollo is located in the ward of Phuthaditjhaba's
 new mayor, Town Council affairs were hardly ever discussed during
 fieldwork. Some informants did not know about the elections.
 Others said that they ignore 'homelandf elections and would
 leave once Qwaqwa gained 'independence1. One man remarked that
 the "mayor is a wealthy shopowner who doesn't understand our
 problems, so how can he help us?" It was evident that up to 1985
 the existence of a Town Council had hardly any impact upon
 everyday life in Phuthaditjhaba. Informants did not perceive the
 Council as particularly "oppressive". They rather felt that this
 body had in fact done nothing. Plans to introduce creches and a
 fire station, had been impeded by a lack of financial resources.
 Local sentiments are echoed in an editorial comment of The
 Pinnacle (11 May 1984) which asks:
 What have these councillors done so far? Streets are
 as dilapidated as ever, there are no signs of recreational
 facilities, ashbins remain in the streets over long periods
 without being attended to...Is the Town Council really
 proud of this pathetic state of affairs?
 In contrast people in Tseki saw local government as oppressive
 and consistently blamed it for interfering in their lives. Tseki
 was divided into seven wards, each presided over by an iinduna
 (overseer) appointed by the headman. The headman of Tseki was
 responsible to the district chief of Bolata (a neighbouring
 closer settlement) and, through him, to the Tribal Authority in
 the main Bakwena village of Namahadi (the seat of the paramount
 chief). Tseki's office had been downgraded as part of the
 reorganization of the Tribal Authority structure in the 1950s
 (vide Chapter One). The present incumbent's title was ramotse
 (headman), whereas his father had been a morena (chief).
 Implementation of the 19 82 Administration of Authorities Act left
 him a headman, because he was a member of a relatively junior
 branch of the Mopeli clan. The newly created district chief in
 the neighbouring village was genealogically senior to him. An
 elderly iinduna, appointed by the present headman1s father,
 explained that prior to the restructuring of "Tribal" government:
 The chiefs were respected leaders who stood up to the
 (South African) government. But today the power of the
 chiefs is gone! At the pitso (assemblies) they only
 execute the commands of Dikwankwetla. I continue my job
 only because it was given to me by S's father, whom I
 respected.
 This induna had lived all his life in Tseki and understood the
 nature of the changes to the institution of chiefship since the
 195 0s. The vast majority of people in Tseki, having arrived in
 the 1970s, had no background in these matters, and saw the
 headman and his iinduna as nothing more than petty functionaries
 within the homeland bureaucracy, whose role was to intrude into
 people's lives and hand out summary punishments for
 infringements of a host of arbitrary rules. People listed a whole
 series of 'offences1 for which they were liable for punishment at
 the hands of the headmen and his iinduna. These included (i)
 19
failure to register births and deaths promptly (ii) failure to
 pay the annual setabataba (site tax) or special Development Tax
 (vide Sharp and Martiny 1984) (iii) quarrelling or fighting with
 neighbours (iv) failure to keep domestic yards tidy (v) failure
 to attend the funerals of neighbours (vi) failure to contribute
 to special levies, for example for "the headmen's daughterf s
 wedding a couple of years ago" (vii) fetching firewood (viii)
 failure to observe conservation rules by, for instance, taking
 soil from outside one's own site to replaster the walls of the
 house (ix) "wasting water" (x) hanging washing on the line
 between lOhOO and 14hO0 in the summer months, thereby causing
 "hailstorms to strike the village". People found guilty of
 disobeying these conventions were arrested by iinduna and taken
 before the headman. Fines of between R5 and R20 were imposed for
 a first offence. For failure to pay fines, or for second or more
 serious offences, people were sometimes lashed by the headman, or
 taken to the "supreme Tribal court" in Namahadi, where, as one
 informant put it, "they could be lashed by head office chiefs".
 The ultimate penalty, which as far as we could tell had never
 been imposed, was the suspension of residential rights in Tseki.
 The well known saying Morena ke_ Morena ka Batho (A chief is a
 chief by the people) was of little applicability in Tseki in
 1983. No one person who was questioned mentioned a single
 positive aspect to the functioning of the Tribal Authority
 Structure in Tseki. Indeed, when asked what the greatest problem
 in Tseki was, one woman replied: "The headmen and his iinduna.
 She continued
 I hate the headmen and his iinduna. Even if people
 fight I would just walk past and won't tell the
 iinduna. In 1974, when we arrived, the headman stayed
 in a rondavel. His wife was in charge then because
 he was just a migrant worker and was always away at the
 time. Now it seems as if he is living better from all
 these fines. He has built a house, and even a shop near
 the clinic. He also drives an El Camino truck now.
 Another woman recalled that in the summer of 19 82 she had hung
 her baby's nappies on the washing line during the day, and that
 an induna had approached her and said that it was wrong to hang
 out washing because this would cause hail.
 He then fined me RlO on the spot. I told him that it
 was only a few nappies. He replied: "You talk too much,
 pay another R5?" Because I was still stubborn he raised
 the fine another three times until it stood at R30. I
 then told him I was poor and could not pay so much. He
 said he would return when I had the money but I'm not
 going to pay him. I will just tell him I have stayed poor.
 Not all Basotho believe that washing causes hail. These
 iinduna want to make all Basotho the same.
 In fear of punishment people got up in the middle of the night
 to fetch firewood, and to dig for soil in order to make the mud
 plaster for their homes:
 20
for taking soil for plaster. These fines are to make us
 buy bricks from Mopeli's brickyard own brickyard in
 Phuthaditjhaba, where large bricks cost R80 and smaller
 ones R60 per hundred.
 At the time of fieldwork in Tseki, the "village council" scheme
 had not yet come into operation. Indications are, however, that
 most closer settlement residents would view this scheme, when
 introduced in the following year, as another imposition from
 above and as an adjunct to the arbitrary power of the headman.
 THE QUEST FOR COMMUNITY
 Insofar as cooperation and mutual aid is concerned people
 relocated into the Diapollo and Tseki shared many similar
 experiences. Informants in both areas commented on the extent to
 which people had helped each other during the initial period in
 Qwaqwa, saying "there was good spirit here", "we stood together"
 and "you never asked whether other people could pay you back".
 The difficulties which people faced were, in many ways novel:
 Qwaqwa presented a new and daunting experience of deprivation
 which was a profound shock to all and, at the outset, engendered
 a grim cameraderie.
 In both areas fieldwork revealed three basic types of mutual aid
 associations beyond the household. These were burial societies,
 stokvels (saving clubs) and phabadimo (givings to the ancestors)
 as found in most South African townships. Most working class
 households in these areas were members of burial associations.
 Members (from about 50 to more than 300) paid monthly fees of 50c
 to the association and R2 whenever a fellow member, or close
 relative of such a member, died. Such contributions were used to
 assist families of the deceased with funeral expenses. Members
 also assisted with funeral arrangements. Stokvel clubs,
 comprising between only three and six members, had a less
 permanent existence. Monthly payments of RIO was made to the
 member whose turn it was to hold a stokvel. Such money was used
 to prepare food, tea and beer for sale to neighbours. All profits
 accured in this process belonged to the holder. Stokvels were
 popular and often attended by more than 30 people. Phabadimo
 feasts attracted as much public interest as stokvels. They were
 held yearly by most households to thank the badimo for luck in
 the past and to request luck for the future. On these occassions
 wealthier households often assisted poorer households by
 contributing food, snuff and ingredients for home brewed beer.
 Despite these similarities there were also clear differences in
 patterns of coopertation and social interaction between the
 Diapollo and Tseki. These differences can be seen as resulting
 from differential experiences of relocation. The effects of
 relocation upon the formation of social networks for mutual
 support and assistance is particularly important.
 21
Among Diapollo residents, who had experienced collective removal,
 social networks were eroded rather than destroyed. Informants
 relocated into the housing section from other areas recalled that
 since their arrival people from Harrismith have formed a closely
 knit community. They said that they were resented by, and that
 they themselves resented those from Harrismith.
 From the beginning we did not get on well...We were
 happy to come here (to the Diapollo), but they were
 not...They didn't like Qwaqwa and said that Qwaqwa was
 like a farm. They thought that they were clever and
 better than us and formed their own groups. They never
 mixed with us...We were taken to be the have nots...
 We really did not get on. Those from Harrismith were
 fighters. If you had a problem with one of them they
 would all be against you...A fight between the kids
 always resulted into a fight between the adults: even
 with stones.
 By the time of fieldwork, approximately 14 years after the
 Schoonplaas removals, relations between these categories of
 people were still tense. The nucleus of networks of sociability
 in the housing section in particular, and Phuthaditjhaba in
 general, tended to comprise people relocated from the same
 township such as Harrismith, Senekal, or Bethlehem. I have heard
 some men of such networks referring to each other as "home-boys".
 There is a rather ironic similarity between the "home boy"
 concept in Phuthaditjhaba and the amakhaya networks identified
 among Xhosa speaking migrants in Cape Town and East London by
 Wilson and Mafeje (1973: 47-56) and by Mayer and Mayer (1974:
 124-134) respectively. "Home- boy" networks in Phuthaditjhaba
 facilitate social support among relocated people. This is not
 unlike the support which these links were said to have in the
 urban environment. These authors also argued that these networks
 enabled migrants to sustain contact with the reserves which was
 essential in maintaining control over land and stock to which one
 could retire. This is paralleled by "home-boy" groups in the
 Diapollo. For these people contact with Harrismith carries
 important benefits. The town has remained a significant centre of
 employment and networks can be utilized as a resource in
 abtaining accurate information about the labour market.
 A large number of home-brew-selling shebeens in the housing
 section were almost exclusively patronized by former inhabitants
 of Harrismith. They provide the venue for discussions concerning
 current labour issues and reminiscences about life in the old
 location. There seemed to be an almost conscious tendency among
 attendants to dissociate themselves from Qwaqwa at an ideological
 level. Ethnicity /as even evoked to these ends. People claimed
 that they did not actually belong in Qwaqwa as they were
 Bakgolokwe, rather than Basotho, who spoke Sekgolokwe rather than
 Sesotho proper. Many said that the stokvels and phabadimo, held
 in these shebeens were sometimes attended by current residents of
 42nd Hill, whereas former shanty-town residents were only
 cautiously invited to these.
 22
In Tseki, where households were relocated in isolation rather
 than collectively, such amakhaya- type networks are absent.
 Households in the closer settlement were far more likely to be
 isolated from those who could be related to them by ties of
 kinship and prior aquaintance by vast geographical distance. This
 severely limited the extent to which households in Tseki were
 able to mobilize relationships of kinship as a source of support
 during times of severe financial crisis. [See Sharp and Spiegel
 (1985) for a comprehensive analysis of the erosion of kinship in
 "betterment" and closer settlement areas.]
 Such isolation had also made Tseki households vulnerable to
 robbery and theft. Vulnerability was exacerbated by the fact that
 a large number of Tseki's men were absent migrants. Informants
 complained that people who returned home on the late buses were
 often robbed by itsotsi and said that Tseki was particularly
 dangerous at night. Comments included: "At night you cannot even
 walk from here to the church"; "It is better to stay at home and
 to lock the doors, because when you die it's better to die at
 home"; and " If you have only 20c in your pocket they'll steal
 it". In Tseki gangs of young men reportedly broke down the doors
 of homes at night and demanded money or domestic posessions
 (dishes, pots, cuttlery and pans) from inhabitants. In addition
 some residents recalled several incidents of kidnapping in the
 closer settlement.
 CONCLUSIONS
 The study of relocation processes from common South Africa into
 the "homelands" has been dominated by attempts to compare and
 contrast the experiences of relocation prior and after the event.
 The analysis of the Suplus People Project (1983) and Platsky and
 Walker (1985) has been particularly impressive in recording the
 devastating effects of population relocations on a national
 scale. This approach has also been characteristic of excellent
 studies of particular local instances of population removals.
 Such accounts include Western's (1980) study of influx control
 removals from Mowbray in Cape Town, Whisson's (1981) account of
 relocation at Glenmore in the Eastern Cape, as well as the
 analysis by Lodge (1982) and Pinnock (1983) on the respective
 destruction of District Six and Sophiatown.
 With this in mind, this paper has attempted to present a
 comparative ethnography on the diverse types of relocation and
 their differential implications for different sections of
 Qwaqwa's population. It is suggested that whilst a comparison of
 experiences prior and after relocation are indeed valuable, such
 a focus may well obscure significant differences among relocated
 people. Residents of the Diapollo were unanimous in their
 agreement that life in Qwaqwa compared unfavourably to life in
 the old Schoonplaas location. On the other hand, residents of
 Tseki were more ambivalent. Some said that they had preferred
 living in the OFS countryside. Others were adament that closer
 settlement life in Qwaqwa was more desirable. Yet this paper has
 shown the numerous ways in which households of the Diapollo have
 23
enjoyed favourable access to resources in the "homeland" relative
 to Tseki households.
 It is my contention that, insofar as Qwaqwa is concerned, such
 differences may partially be accounted for by the important
 distinction between removals which were officially planned
 ("forced") and those which were unplanned ("voluntary"). The
 contention that removals due to directly planned action by the
 state may imply favourable access to resources does not detract
 from the devastating impact upon the lives of those affected. It
 rather serves to highlight the desparate plight of the less
 visible and more marginilxzed sections of Qwaqwafs population who
 had been subject to unplanned removals.
 At the present it is difficult to assess the extent to which this
 trend holds true of other South African "homelands". Tentative
 support for this hypothesis may be found in the position of
 Mdantsane, whose residents were forcefully relocated with the
 destruction of Duncan Village, relative to other areas in the
 Ciskei (vide Mayer and Mayer 1974: 294-318). The experience of
 Glenmore, on the other hand, contradicts this hypothesis (vide
 Whisson 1981) . It is only through further comparative research
 that greater clarity can be gained.
 REFERENCES
 Bank, L.
 1984 Town and "closer settlement" in Qwaqwa. Unpublished B.A.
 Honours Dissertation, African Studies, University of Cape
 Town
 BENSO
 1978 Qwaqwa Economic Review. Johannesburg. Bureau for Economic
 Research and National Development
 Development Bank of Southern Africa (DBSA)
 1985 Qwaqwa Development Information. Sandton.
 Greenberg, S and Giliomee, H.
 1983 Labour Bureaucracies and the African Reserves. South
 African Labour Bulletin 8 (4).
 Hawkins, E.
 1982 The Story of Harrismith. Harrismith Rotary Club.
 Lodge, T.
 1983 Black politics in South Africa since 1945. Johannesburg:
 Ravan Press.
 Mare, G.
 1980 African population relocations in South Africa.
 Johannesburg: South African Institute of Race Relations.
 24
Murray, C.
 1985 Struggle from the margins: Rural slums in the Orange Free
 State. in Cooper, F (ed) Struggle for the City. London.
 Sage Publications.
 Mayer, P and Mayer I.
 1974 Townsmen or Tribesmen. Cape Town. Oxford University Press.
 Niehaus, I.
 1984 The problems, experiences and responses of relocated
 people: Closer settlement life in Qwaqwa. Unpublished B.
 Soc. Sc. Honours dissertation. Social Anthropology,
 University of Cape Town.
 Pinnock, D.
 1984 The Brotherhoods. Cape Town: David Philip.
 Platsky, L and Walker, C.
 1985 The Surplus People. Johannesburg: Ravan Press.
 Quinlan, T.
 1986 The tribal paradigm and ethnic nationalism: A case study
 of political structures in Qwaqwa.Transformations (2).
 Ross, Eerwaarde J.J.
 1930 Die Sending te Witzieshoek. Bloemfontein. Nasionale Pers
 Beperk.
 Sharp, J.
 1982 Relocation and the problem of survival in Qwaqwa. Social
 Dynamics 8 (2).
 Sharp, J and Martiny, J.
 1984 An overview of Qwaqwa. Carnegie Conference paper no. 64.
 Cape Town: University of Cape Town.
 Sharp, J and Spiegel, A.
 1985 Vulnerability to impoverishment in South African rural
 areas: the erosion of kinship and neighbourhood as
 social resources. Africa 55 (2)
 Streek, B and Wicksteed, R.
 1981 Render unto Kaiser. Johannesburg: Ravan Press.
 Surplus People Project
 1983 Forced Remova1s in South Africa. Volume 3. Cape Town. The
 Surplus People Project.
 Western, J.
 1980 Outcast Cape Town. Cape Town: David Philip.
 Whisson, M.
 1980 Resettlement in Glenmore. Nash, M and Charton, B. (eds)
 An Empty Table? Grahamstown: South African Council of
 Churches.
 Wilson, M. and Mafeje, A.
 1973 Langa. Cape Town. Oxford University Press.
 25