Class Struggle and Migrant Labor in South African Gold Mines¶
David Massey (1983). Class Struggle and Migrant Labor in South African Gold Mines. Canadian Journal of African Studies / Revue Canadienne des Études Africaines, Vol. 17, No. 3 (Special Issue: South Africa). Canadian Association of African Studies.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1886–1982 |
| Regions | South Africa, Witwatersrand gold mines, Mozambique, Malawi, Lesotho, Botswana, Transkei, Bophuthatswana, Namibia |
| Themes | mining migrant labour, labour markets wages, unions strikes, resistance social movements, bantustans removals influx, origins segregation, comparative theory |
| Citation key | massey1983 |
Argument¶
Massey argues that oscillating migrant labour served white capital by widening the labour pool, creating a reserve army that could be shed in recession, shifting the cost of reproducing labour onto the rural reserves ("secondary exploitation"), and hampering worker organisation. He rejects the idea of a fixed "traditional wage" set at single-worker subsistence: wages reflect the balance of class power. In the 1970s the foreign supply became politically unreliable (Malawi, Mozambique, Lesotho) and miners more militant, so mine owners and the state raised wages sharply and "internalised" recruitment. He contends this will not produce a quiescent workforce, because the reserves' capacity to subsidise accumulation is eroding, and he predicts continued unrest and harsher repression.
Evidence and method¶
Wilson's wage series and WNLA/TEBA annual reports (1921-1977), Chamber of Mines statistics, Parsons's Chamber of Mines Human Resources Laboratory study, the confidential 1975-76 Inter-Departmental Committee report on mine riots, the Tomlinson Commission, Anglo American internal reports, the author's own records of 116 blacklisted Botswana miners (1974-75), interviews with 123 Botswana miners (1977) and an ex-miner, plus press reports and SA Labour Bulletin articles.
Place in the debates¶
- Cheap labour thesis / role of the reserves in subsidising accumulation (neo-Marxist articulation of modes of production) (see Cheap labour thesis and the function of the reserves). Supports the neo-Marxist view that reserve production subsidises capitalist wages (secondary exploitation), but argues that wage levels are set by class struggle rather than the reproduction cost of labour.
- Whether wages in southern Africa were pegged to a customary single-worker subsistence ('traditional wage', e.g. Arrighi) (see Cheap labour thesis and the function of the reserves). Rejects it: employers "paid as little as possible", and a 'traditional wage' was only a rationalisation; wages were determined by the state of class struggle.
- Causes of the 1970s mine unrest and of the shift away from foreign labour (see Worker consciousness, strikes and conflict on the mines and shop floor). Attributes it to both political change in supplier states and rising worker militancy and consciousness, not to outside agitators (the riot commission found none).
- Functionality of apartheid/separate development for capital (revisionist vs liberal) (see Cheap labour thesis and the function of the reserves). Revisionist: the economic basis of separate development is to make all black workers 'foreign' migrants in order to sustain low wages and control.
Findings¶
-
Employers' pooling of recruitment from 1896 created a monopsony over African labour in order to hold wages down. (p. 3)
We could not pool the supply, so we pooled the demand. The employers agreed to divide the supply among themselves.3
-
Real black mine wages were stagnant for six decades (1911-1971). (p. 5)
What is remarkable is that through most of this period, from 1911 to 1971, real wages on the mines remained virtually stationary.
-
Foreign labour's share rose in the boom of 1965-72 even as supplier states became politically risky, showing a choice for cheap foreign labour over higher domestic wages. (p. 13)
between 1965 and 1972 - a period of rapid economic growth in South Africa - the amount of foreign labor in the mines was allowed to increase from sixty-five percent to an all-time high of eighty percent.
-
Malawi's 1974 withdrawal of labour was a major shock to the mines at a time of high gold prices. (p. 13)
The subsequent loss of 100 000 workers in a single year, at a time when the gold price was approaching $200 an ounce for the first time in history, cost the South African mining companies millions
-
Contract migrancy and compounds served as political control, because turnover undermined organisation. (p. 9)
The major political advantage of this massive contract labor system to the mining companies is that workers, because they are constantly coming and going, find it very difficult to organize themselves.
-
Reserves subsidise capitalist accumulation by bearing the costs of reproducing and caring for labour. (p. 7)
In this way the reserve areas are subsidizing the accumulation process in the capitalist sector.
-
The riot commission recommended raising mine wages and cutting foreign labour, and real wages rose steeply after 1972. (p. 14)
Between 1972 and 1975, real wages in the mines increased by over 250 percent.
-
Internalisation of the labour supply succeeded quickly, with the Bantustans as the main source. (p. 14)
By December 1977 the goal of fifty percent local labor had been reached; it was quickly surpassed. By 1980 fifty-eight percent of the black workers in the mines were from inside South Africa.
-
After repeal of the Masters and Servants laws in 1974, desertion became a lever for workers. (p. 18)
since the repeal of the Masters and Servants laws in 1974, men have increasingly exercised their right simply to walk off the job.
-
Desertion rates rose sharply from 1975, reaching 11.2% of the workforce by 1977. (p. 18)
1975 7.4 28 456 1976 10.8 44 150 1977 11.2 49 286
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Witwatersrand Native Labour Association (WNLA) Annual Reports (1921-1977 series on foreign share of black miners) | chamber of mines or recruiting records | 1921-1977 | Supplies the series on the percent of black miners from foreign sources plotted in Figure 1, used to show foreign labor as an expendable reserve. |
| Francis Wilson, Labour in the South African Gold Mines (1972) wage data, with WNLA Annual Reports 1971-1977 | secondary literature compilation | 1911-1971 | Provides the annual wage and real earnings index for black gold miners (Table 1), showing stationary real wages for sixty years. |
| Tomlinson Commission report (Commission for the Socio-Economic Development of the Bantu Areas within the Union of South Africa), 1955 | commission or official report | 1955 | Quoted to show the South African state recognised the role of foreign migrant labor in keeping wages low. |
| Transvaal Local Government Commission, 1921 | commission or official report | 1921 | Quoted for the official attitude that Africans be admitted to urban areas only as temporary labor, motivating institutionalized migrancy. |
| Transvaal Labour Commission testimony of Percy Fitzpatrick, February 1904 (via Webster, SALB 1974) | commission or official report | 1904 | Quoted as mine-owner testimony that the Native Labour Association pooled demand to hold wages down (monopsony). |
| South African Institute of Race Relations, Survey of Race Relations in South Africa 1977 (wage and subsistence level data) | secondary literature compilation | 1977 | Supplies the 'Household Effective Level' of R216 for an African family, against which miners' 1977 wages are compared to show secondary exploitation. |
| South African Chamber of Mines, Statistical Tables, 1977 | official statistical series | 1977 | Source for the 1977 average cash and kind wage of black miners (R143/month). |
| Mine Labour Organisations' Records, Mafeking (blacklisting records, Aug 1974-Feb 1975) | chamber of mines or recruiting records | August 1974 - February 1975 | Author's tabulation of 116 blacklisted Tswana miners' official offences to argue about a third were sacked for political reasons. |
| Merle Lipton, 'South Africa: Two Agricultures?' (SALDRU Farm Labour in South Africa, 1978) and WNLA Annual Reports 1960, 1970 | secondary literature compilation | 1950-1972 | Sectoral average black wages (Table 3) showing mining and agriculture have lower wages than manufacturing and construction. |
| WNLA Annual Reports 1971-1977 (annual wage and real earnings index 1972-1977) | chamber of mines or recruiting records | 1972-1977 | Table 4 shows the sharp rise in black miners' real wages after 1972, linked to internalization of labor supply. |
| J.A. Parsons, 'How Attractive is Gold Mining to Black Mineworkers?' Chamber of Mines Human Resources Laboratory, 1977 | company records or reports | 1940-1977 | Supplies the series for Figure 2 relating mining wages relative to manufacturing to the share of South African-born miners. |
| Republic of South Africa, Report of the Inter-Departmental Committee of Inquiry into Riots on Mines in the Republic of South Africa, 1976 (unpublished) | commission or official report | 1975-1976 | Source for the commission's findings and secret recommendations to cut foreign labor, raise wages, and improve control, plus the worker consciousness conclusion. |
| Anglo American Corp., Industrial Relations Dept., The Perceptions and Behaviour Patterns of Black Miners on a Group Gold Mine (1976) | company records or reports | 1976 | Cited for Transkeian miners' expressed solidarity with Soweto students, showing internalized labor is not necessarily docile. |
| Author's interviews with 123 Botswana mineworkers, 1977 | author survey or fieldwork | 1977 | Interview sample supporting the claim that miners attribute disturbances to low wages and see whites and owners as main beneficiaries. |
| Field interview with Amos Kgamanyane Pilane, Mochudi, Botswana, 11 November 1977 | interviews oral history | 1977 (describing c. 1900) | Oral testimony illustrating the compound inspection and detention system on the diamond mines. |
| Mine Labour Organisations files (recruitment figures, Kimberley and Pretoria areas) | chamber of mines or recruiting records | 1974-1976 | Recruitment totals used to doubt the 1978 announcement that urban recruiting stations had failed. |
| WNLA Annual Reports 1972-1977 (desertion rates) | chamber of mines or recruiting records | 1972-1977 | Table 5 shows desertion rising from 2.7% to 11.2% after repeal of the Masters and Servants laws. |
| WNLA Annual Report 1976 (turnover rate) | chamber of mines or recruiting records | 1976 | Cited for the 137 percent annual turnover rate of black gold miners. |
| Newspapers and periodicals (Financial Mail, Rand Daily Mail, Africa News) | newspapers periodicals | 1977-1982 | Used for events like the ZAPU recruitment revelation, the failure of urban recruiting, the East Rand Proprietary cutback and the 1982 strikes. |
| South African Institute of Race Relations, Survey of Race Relations in South Africa, 1979 | official statistical series | 1979 | Source for the April 1979 Elandsrand disturbance and dismissals. |
| Beinart, 'South Africa's Internal Supply of Labour with Special Reference to the Transkei and Bophuthatswana' (UN/ECA, 1978) | secondary literature compilation | 1974-1977 | Cited, with WNLA and Mine Labour Organisations records, for the cutback of Mozambican workers and growth of Bantustan supply. |
New data¶
Author tabulated 116 blacklisting records of Botswana miners (Aug 1974-Feb 1975) from Mine Labour Organisations records at Mafeking, and conducted interviews with a sample of 123 Botswana mineworkers in 1977 plus other field interviews.
Inference¶
Quantification: Descriptive tables.
Marxist-informed historical-structural argument supported by descriptive time series and tables of wages, foreign labor share and desertion rates, official commission quotations, and small-scale interview and archival samples; correlation is shown graphically rather than tested statistically.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Real wages of black gold miners remained virtually stationary from 1911 to 1971, while foreign workers were used as an expendable reserve. (p. 432) | Table 1 (Wilson, WNLA reports) and Figure 1 (WNLA annual reports). | Systematic data |
| Migrant labor lets capital shift the costs of reproducing the workforce to the reserves ('secondary exploitation'), though its extent cannot be measured. (p. 435) | Theoretical argument plus comparison of 1977 mine wages with the SAIRR Household Effective Level. | Illustrative or anecdotal |
| Wages are set by the state of class struggle rather than by a customary single-worker subsistence wage. (p. 436) | Argued from the literature on workers' resistance, with no systematic wage-setting data. | Asserted |
| Nearly a third of blacklisted Botswana miners were sacked for offences that were political, i.e. threatening mine power relations. (p. 437) | Author's tabulation of 116 Mine Labour Organisations blacklist records (Table 2), Mafeking. | Partial data |
| After the 1975 riots commission, mines raised real wages sharply and reduced the foreign labor share from about 80 percent to 42 percent by 1980, closely tracking relative wages. (p. 441) | WNLA wage index (Table 4), Parsons/Chamber of Mines series (Figure 2), and the unpublished riot commission report. | Systematic data |
| Political change in supplier states (Malawi, Lesotho, Mozambique) and worker militancy in the 1970s undermined the reliability of foreign labor. (p. 440) | Event narrative drawn from news reports, commission report and WNLA records. | Partial data |
| Desertion rates soared after the 1974 repeal of the Masters and Servants laws, giving miners leverage. (p. 445) | Chamber of Mines/WNLA desertion figures (Table 5) and mine-level press reports. | Systematic data |
| Internalization of labor supply will not produce a less militant workforce, and repression is likely to intensify. (p. 446) | Illustrative cases (Elandsrand 1979, 1982 strikes), the Anglo American 1976 survey and Botswana miner interviews. | Illustrative or anecdotal |
The author's caveats¶
Acknowledges that the extent of secondary exploitation cannot be quantified and that, from outside, it is difficult to judge how far struggles in the mines continue, since many incidents go unreported. Notes that the riot commission report is unpublished and that recruitment figures may conflict with official statements.
Gaps in the evidence¶
The central claim about reserve subsidization lacks any measurement of reserve household production, remittances or reproduction costs, which was estimable from rural surveys and official statistics then. Claims about militancy rest on a small Botswana-origin sample and press reports and would be better tested with mine-level strike, desertion and recruitment data across mines and origin areas, partly in Chamber of Mines and WNLA archives. Causal links among wages, labor supply and politics are asserted from graphs and not tested econometrically.
Replicability: Data described but not public.