Women and Wages: Gender and the Control of Income in Farm and Bantustan Households¶
John Sharp; Andrew Spiegel (1990). Women and Wages: Gender and the Control of Income in Farm and Bantustan Households. Journal of Southern African Studies, Vol. 16, No. 3. Taylor & Francis.
| Type | journal article |
| Scope | Wider apartheid literature |
| Method | survey fieldwork |
| Period studied | 1940s–1980s |
| Regions | Matatiele district, Transkei, Qwaqwa, Orange Free State, white-owned farms in the Orange Free State, north-eastern Cape and southern Natal farms |
| Themes | agriculture land, bantustans removals influx, mining migrant labour, labour markets wages, inequality poverty unemployment |
| Citation key | sharp1990 |
Argument¶
The article compares two Bantustan areas, Matatiele (Transkei) and Qwaqwa, to show that gender domination and women's responses cannot be read straight off capitalist expansion or the migrant labour system. Local history and the availability of residual resources (arable land, livestock) shape the forms men's control over remittances and women's incomes take. In Matatiele, land gives women a legitimate sphere of local income and some autonomy. In Qwaqwa, a closer settlement of people relocated from Free State farms, total dispossession makes women's subordination starker and their defiance more open. The authors trace how farm mechanisation and the shift from payment in kind to cash wages in the 1960s-70s changed gender relations on farms and fed relocation to Qwaqwa.
Evidence and method¶
Fieldwork in 1982-4 in a Matatiele village and a Qwaqwa closer settlement, including oral accounts from older informants about 1940s-70s farm conditions and a household survey. Supplemented by secondary literature (Ferguson on Lesotho, Morris, De Klerk, Bozzoli, Stichter, Eades, Marx on the Bedouin) and unpublished dissertations.
Place in the debates¶
- Capitalism and the subordination of women (reproduction/cheap labour thesis vs. local patriarchal and historical factors) (see Cheap labour thesis and the function of the reserves). Accepts the general link but calls it 'a very blunt instrument'. Following Bozzoli, it argues that male domination also has pre-capitalist and local roots and varies with resources and histories.
- The household as a unit of analysis. Rejects the household as a unitary community of interest. It treats households as sites of conflict along gender and seniority lines, especially over remittance income.
- Causes of farm labour shedding and relocation to the Bantustans (see Farm labour: mechanisation, labour shedding, wages and labour control). Links mechanisation, population growth and changed payment forms on white farms to the removal of dependants and relocation to Qwaqwa.
Findings¶
-
Women in the Bantustans are largely shut out of wage employment beyond the reserves, so migrant men's remittances are the main income source. (p. 3)
Women are virtually excluded from wage labour beyond the Bantustans.
-
The authors criticise the generalised capitalism-subordination link as too crude for analysing regional variation. (p. 3)
the link which is posited between capitalism and women's subordination is a very blunt instrument for the specific analysis of South Africa's Bantustans,
-
Complete dispossession in Qwaqwa intensifies women's exploitation and makes their subordination more visible. (p. 8)
complete dispossession, as in Qwaqwa, intensifies the degree of women's exploitation, both by men and by the capitalist system at large;
-
On 1940s Free State farms, labour was contracted through male household heads, who controlled household income. (p. 9)
workers did not contract individually with the farmer, but as members of households with which the farmer dealt through their male heads, arranging with the latter the amount of labour required and the forms of its payment.
-
Population growth and mechanisation led farmers to cut payments in kind (land, grazing, bonus) to households. (p. 10)
With population growth and increased mechanisation from the late 1960s, farmers were, it appears, no longer prepared to carry the burden of supporting people whose labour was not needed.
-
Farmers began paying cash wages directly to women, giving them income discretion they had not had before, which men resented. (p. 10)
farmers began to pay cash wages directly to women who worked for them in the kitchens and fields.
-
Mechanisation of harvesting and threshing removed jobs previously done by women and children. (p. 11)
Mechanisation in the 1960s and 1970s related to the processes of harvesting and threshing, jobs which had previously been done by women and children
-
Farmers could not recall migrant men because farm wages could not match migrant earnings, so they wanted rid of surplus dependants, feeding relocation. (p. 11)
farmers wanted to recall these men, but were unable to do so because the wages they offered did not compare with migrant earnings.
-
Men and women evaluated the move to Qwaqwa very differently. (p. 12)
whereas virtually all of the men felt that they were better off in Qwaqwa than they had been on the farms, most of the women said that the move to the Bantustan had been an unmitigated disaster.
-
Betterment froze the distribution of arable land in Matatiele, creating a permanently landless category. (p. 14)
the implementation of a betterment scheme in the 1970s had frozen the distribution of arable land.
-
Men in the Qwaqwa closer settlement had highly insecure employment. (p. 19)
no more than 35% of the adult men in the households surveyed worked continuously in wage employment throughout 1983.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Sharp and Spiegel fieldwork in a Matatiele village (Transkei), 1982-4 | author survey or fieldwork | 1982-1984 | Supplies the evidence on gender relations, remittance use, landholding differentiation and women's income-generating activities in the Matatiele village. |
| Sharp fieldwork in a Qwaqwa closer settlement (including household survey), 1982-4 | author survey or fieldwork | 1982-1984 (survey figure refers to 1983) | Supplies the evidence on gender conflict over shebeening, male control of remittances, and employment histories of men in the closer settlement. |
| Oral accounts of Qwaqwa informants on Orange Free State white-owned farms, 1940s-1970s | interviews oral history | 1940s-1970s (recalled) | Reconstructs changing labour tenancy and payment forms on farms and the gendered effects of mechanisation and the shift to cash wages before relocation. |
| Survey in the Qwaqwa closer settlement (household survey of adult male employment in 1983) | author survey or fieldwork | 1983 | Provides the single quantitative figure that no more than 35% of adult men in surveyed households worked continuously in 1983, and the men/women contrast on relocation. |
| Oral accounts of Matatiele informants on north-eastern Cape and southern Natal farm employment | interviews oral history | early 20th century to 1950s (recalled) | Supports the claim that farm labour reinforced existing gender divisions and livestock payment persisted until the 1950s. |
| J. Ferguson, 'The Bovine Mystique' (Man 20, 1985), Lesotho migrant remittance and cattle study | secondary literature compilation | 1980s | Used as comparative evidence that migrant men's dual use of remittances creates conflict, a pattern the authors find in Matatiele. |
| L. J. Bank, 'Traders and taximen in Qwaqwa' (MA dissertation, University of Cape Town, 1987) | secondary literature compilation | 1970s-1980s | Source for restrictions on African trading in Qwaqwa and the displacement of local traders into closer settlements in the early 1980s. |
| J. Segar, 'Food and health care in a betterment village' (BA Hons dissertation, UCT, 1982) | secondary literature compilation | c.1982 | Cited for landless women working for very low wages in kind on others' fields or in shebeens. |
| E. Marx, 'Labour Migrants with a Secure Base: Bedouin of South Sinai' (1987) | secondary literature compilation | 1980s | Used in the conclusion as a comparative case supporting the argument that local resources shape women's autonomy in migrant-labour systems. |
| Sharp and Spiegel earlier published work on Matatiele and Qwaqwa (Africa 55, 1985; Eades ed. 1987; Collected Seminar Papers 37, 1988) | secondary literature compilation | 1982-1988 | The authors' earlier papers on vulnerability to impoverishment, remittance interruptions and relocation underpin the comparison and are cited for the field data. |
New data¶
Author-collected ethnographic fieldwork 1982-4 in one Matatiele village (Spiegel) and one Qwaqwa closer settlement (Sharp), including household surveys and informants' oral histories; sample sizes are not reported.
Inference¶
Quantification: Illustrative figures.
Qualitative paired ethnographic comparison of two Bantustan sites, using informants' accounts, historical reconstruction from oral memory and the authors' earlier work, with comparative parallels from the literature (Lesotho, Sinai, Third World shanty towns). Causal claims rest on contrasts in land access and relocation history rather than on statistical tests.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Differences in access to residual agricultural resources (arable land, livestock) and relocation history make gender relations in Matatiele differ from Qwaqwa despite common dependence on migrant remittances. (p. 533) | Paired comparison of the authors' own fieldwork in a Matatiele village and a Qwaqwa closer settlement, 1982-4. | Partial data |
| On Free State farms men controlled household income as household heads, but from the mid-1950s the loss of land, grazing and bonuses, mechanisation and direct cash wages to women undermined patriarchal authority and prompted relocation to Qwaqwa. (p. 535) | Recollections of older Qwaqwa informants, men's and women's accounts; supported by secondary studies of mechanisation (De Klerk, Morris). | Illustrative or anecdotal |
| Qwaqwa men opposed wives' shebeening and hawking, while women defied them, because remittances were uncertain and women had more discretion over locally earned income. (p. 530) | Interviews with married men and women in the Qwaqwa closer settlement. | Illustrative or anecdotal |
| Differential labour-market access and the betterment scheme produced material differentiation among Matatiele households, shaping tension between migrant husbands and wives over remittances. (p. 539) | Matatiele village fieldwork and the author's earlier published work on the village (Spiegel 1982, 1986). | Partial data |
| Qwaqwa closer-settlement men held low-paid, insecure contract jobs, and the lack of any land or livestock increased men's control over a household's few resources, including through enforced presence at home. (p. 544) | Household survey of employment in 1983 (35% continuous employment figure) plus fieldwork observation. | Partial data |
| Most Qwaqwa women judged the move to the Bantustan a disaster while men felt better off than on the farms. (p. 537) | Authors' survey in the closer settlement; no figures reported. | Asserted |
| Unmarried young women's attitudes to marriage differ: positive in Matatiele, hostile in Qwaqwa, because of the models of wifehood available. (p. 548) | Fieldwork observations of young women in both sites, supported by cited dissertations (Niehaus, van der Vliet). | Illustrative or anecdotal |
The author's caveats¶
The authors note that neither area is unique and that comparative work is scarce; they say each of them did field research in only one area. They do not state sample sizes or sampling methods, and they use hedged language ("it appears", "it seems") for the historical reconstruction from memory.
Gaps in the evidence¶
The argument rests on two single-site ethnographies with no reported sample sizes, so representativeness is untested. Statistics on remittance amounts, income by gender and employment histories, plus farm-labour records or Bantustan administrative files on relocation, would let the claims be tested. Some such records (Transkei and Qwaqwa administrative files, labour-bureau registers, official census data) existed at the time; the farm-era history can now only be checked against archives and later oral histories.
Replicability: Data poorly documented.