Black Worker Conflicts on South African Gold Mines: 1973-1982¶
McNamara, J.K. (1985). Black Worker Conflicts on South African Gold Mines: 1973-1982. PhD thesis, Faculty of Arts, University of the Witwatersrand. University of the Witwatersrand.
| Type | thesis |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1889–1982 |
| Regions | Witwatersrand gold mines, Orange Free State goldfields, Carletonville, Klerksdorp, Lesotho, Malawi, Mozambique, Rhodesia/Zimbabwe, Transkei |
| Themes | mining migrant labour, unions strikes, labour markets wages, resistance social movements, business and state, politics ideology elections |
| Citation key | mcnamara1985 |
Argument¶
The thesis explains 282 recorded violent collective conflicts involving Black migrant gold miners in 1973-1982 as three distinct types: foreign-worker dependency conflicts, inter-group (mainly Sotho-Xhosa) clashes, and worker-management strikes and protests. Using a survey of all known cases plus thirteen detailed case studies analysed by a five-phase anthropological conflict model, it argues that the conflicts arose from changes in the wider Southern African political environment, from reforms that produced progressive deprivation (large early-1970s wage increases followed by smaller ones), and from strains and gaps in the mine/hostel social structure. Strikes are read as informal substitutes for absent collective bargaining. Ethnic clashes are traced to labour turnover, job competition, hostel life and feuds rather than to primordial tribalism.
Evidence and method¶
Internal mine management reports, Chamber of Mines and TEBA Liaison Division files, MLO and TEBA statistics, government (McKay Commission) evidence, Chamber annual reports, and newspapers. Also the author's own Human Resources Laboratory fieldwork: hostel participant observation, interviews, and network studies. Includes a diary of 282 conflicts and 13 case studies.
Place in the debates¶
- Solidarity vs breakdown/relative deprivation theories of collective action (see Worker consciousness, strikes and conflict on the mines and shop floor). Finds that both apply: workers lacked unions, but hostel institutions facilitated mobilisation, and a J-curve of 'progressive deprivation' after the early-1970s wage rises explains the strikes.
- Marxian causal-objective vs social-action explanations of mine conflict (see Worker consciousness, strikes and conflict on the mines and shop floor). Argues the Marxian approach gives the structural background but cannot predict when conflict erupts; perceptions of injustice (Barrington Moore) explain timing, since conflict followed improvements rather than immiseration.
- Primordial tribalism vs situational explanations of mine 'faction fights' (see Worker consciousness, strikes and conflict on the mines and shop floor). Rejects official and popular tribal-savagery explanations; attributes clashes to labour turnover, hostel tensions, job competition, induna power struggles, and political events, with ethnicity as outward form.
- Institutional vs wider socio-political explanations of industrial conflict (see Worker consciousness, strikes and conflict on the mines and shop floor). Both matter: foreign-labour conflicts reflect regional politics, but strikes originate largely in internal wage and service-condition changes in an isolated institutional setting.
Findings¶
-
Newspaper coverage drastically under-records mine conflict, which justifies reliance on internal industry sources. (p. 57)
only one third of all recorded conflicts were reported in the press
-
Foreign labour share of the mine workforce reached about 80% by 1973, as mine wages fell behind manufacturing and construction. (p. 105)
dropping steadily to 35 per cent of the wages offered by these sectors in 1970
-
Black wages rose very sharply in the early 1970s, with average monthly pay up from R17 in 1971 to R250 in 1982 (roughly fivefold in real terms), and the largest increases in 1973-75. (p. 307)
-
The mines moved to 'internalise' labour supply: the share of South African Blacks rose from 22% in 1974 to 57% by 1982. (p. 152)
the proportion of South African workers reached 53 per cent in that year and increased thereafter to 57 per cent by 1982
-
Wage-increase strikes were caused by cuts in the size of annual wage awards, which frustrated expectations built up by the earlier large rises. (p. 399)
workers' expectations had remained consistently high and the resulting perceived discrepancy in the actual capabilities of the industry to satisfy these expectations
-
The 1982 wage strikes involved ten mines and thousands of repatriations, with violence aimed at supervisors/team leaders (many Mozambican), showing an occupational cleavage overriding ethnicity. (p. 334)
Over 3000 men were repatriated and at least eight hostels were extensively damaged
-
The absence of formal bargaining machinery meant strikes functioned as informal negotiation. (p. 319)
-
Overall toll of inter-group conflict 1973-82. (p. 282)
at least 155 mineworkers were killed and close to 1300 injured
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Mine management and Group head-office conflict reports (Anglo American Gold Records Division, Anglo-Transvaal, General Mining/Gencor, Union Corporation, JCI, Rand Mines files) | company records or reports | 1973-1982 | Principal basis for reconstructing the 282 (or 262/252, as the text variously gives) conflict cases and the detailed case-study chronologies. |
| TEBA (The Employment Bureau of Africa) Liaison Division Labour Unrest Files and Diaries of Conflict (TEBALD), district and monthly reports | chamber of mines or recruiting records | 1971-1982 | Supplies incident reports, interviews with discharged workers and district reports on rural unemployment used in case studies and the conflict record. |
| Chamber of Mines central records and correspondence (COMGU unrest files; COMNL native labour files for Lesotho, Mozambique, Rhodesia, Tropicals, Malawi Air Crash; COM Blacks-Discipline) | chamber of mines or recruiting records | 1967-1982 | Documents Chamber/ministerial/Banda negotiations, Rhodesian recruitment decisions and responses to unrest. |
| Chamber of Mines Annual Reports (1st 1889 to 57th 1946, and 93rd for 1982 listed) | chamber of mines or recruiting records | 1889-1946 | Main source for the provisional historical diary of conflict 1889-1972 and for labour supply history and wage cuts. |
| Transvaal Labour Commission 1905 conclusions (via 14th CMAR for 1903) | commission or official report | 1905 | Cited to show Black farmers were not compelled to take mine wages early in the century. |
| Newspaper reports (Rand Daily Mail, The Star, Die Transvaler, Sunday Times, Rapport, Citizen, Friend, Bulawayo Chronicle, etc.) | newspapers periodicals | 1967-1983 | Supplement to internal records for case chronologies; the author shows only about a third of conflicts were reported. |
| Hansard, No. 5, Column 1969, 5 March 1975 | parliamentary debates | 1975 | Cited for parliamentary concern about the new intensity of mine riots. |
| South African Institute of Race Relations Annual Surveys of Race Relations 1973-1982 | secondary literature compilation | 1973-1982 | Used with industry files and press to compile the event diary (Annexure Four) and for Mozambique/Portugal events. |
| Chamber of Mines Statistics / MLO Position of Companies / TEBA Annual Reports and Recruited Labour statistics | official statistical series | 1971-1982 | Source for tables on labour by origin, regional distribution, novices, length of stay, returnees, desertions, discharges and local engagements. |
| Chamber of Mines Accident Statistics (Prevention of Accidents Committee) and Loss Control Survey | official statistical series | 1971-1982 | Fatality and injury rates in Table 26 and Rhodesian mine hazard comparison. |
| Department of Manpower, Report of the National Manpower Commission for 1981 | official statistical series | 1971-1981 | Registered Black unemployment (Table 23) as backdrop to labour stabilization and job scarcity. |
| Human Resources Laboratory (Chamber of Mines Research Organization) unpublished research reports and interviews | author survey or fieldwork | 1970-1983 | Source for interviews, work-routine data, assault and absenteeism analyses, job distributions, hostel surveys and Rhodesian worker interviews. |
| Author's participant-observation fieldwork in mine hostels (Eastern Transvaal 1976; Free State 1978) and network studies | author survey or fieldwork | 1976 and 1978 | Provides drinking-network data (39 of 46 groups same-language) underpinning the home-friend solidarity and ethnic form argument. |
| Investigations of specific disturbances by the author's team (1977 popularity study, 1978 Free State, 1978 Rustenburg platinum mine, 1979 West Rand/Carletonville strike) | author survey or fieldwork | 1977-1979 | Used for the Wildebeestfontein shebeen conflict and the 1979 new-mine strike case studies. |
| Sample work-record card analysis and assault analyses on Free State and platinum mines (MacArthur et al. 1978; Connoway 1980) | author survey or fieldwork | 1976-1978 | Tables on job distribution by origin and assailant status used to argue Sotho job domination and Sotho-Xhosa underground friction. |
| Secondary data on Malawi, Lesotho and Mozambique economies and migration (Coleman, Thomas, Murray, Eckert and Wykstra, Stahl, First, Harris, Webster) | secondary literature compilation | 1950s-1983 | Background on labour dependency and political change; First (1983) also supplies Mozambican miner interviews. |
| Rhodesian secondary data (Dorsey 1975, Moorcroft and McLaughlin 1982, Phimister, Van Onselen, Turner) | secondary literature compilation | 1890-1980 | Used to explain high aspirations and political heritage of Black Rhodesian recruits. |
| Horner and Kooy (1976, 1979) newspaper-based diary of mine conflicts | secondary literature compilation | 1972-1979 | Cross-reference for conflict records; criticized as relying exclusively on newspapers. |
New data¶
A database of roughly 262-282 recorded conflicts on gold mines 1973-1982, each diarized by phase (breach, mobilization, crisis, redressive mechanisms, adjustment) from mine, Group, TEBA and Chamber files plus press; subsets analysed as 88 inter-group conflicts, 143 management-worker conflicts, and thirteen detailed case studies. Also an event diary (Annexure Four) and a 1889-1972 conflict diary (Annexure Three).
Inference¶
Quantification: Descriptive tables.
Combines a survey of the coded conflict database (frequency counts, trends over time, tables of underlying factors) with in-depth case studies analysed through a five-phase Turner/Swartz conflict-process model; explanations are weighed qualitatively against competing theories (Marxian, relative deprivation/J-curve, solidarity, anthropological strain/stress and factions) rather than tested statistically, as the author explicitly rules out regression.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Contemporary inter-group clashes arose from tensions of hostel communal life, shebeens/farms, labour turnover and new recruits, job competition and induna status struggles, rather than primordial tribal feuding. (p. 260) | Table 24 frequency count of factors across 80 inter-group conflicts, case chronologies from mine/TEBA reports, tables on novices and length of stay, and network studies. | Partial data |
| The 1974 Free State Sotho-Xhosa clashes stemmed mainly from numerical and occupational domination of Lesotho nationals in better-paid jobs, with BCP politics explaining mass repatriations. (p. 224) | Job distribution sample (Table 13), assault tables (14-15), MLO district manager reports, mine inquiry and press. | Partial data |
| Black Rhodesian recruits were over-represented in conflicts with management (30 per cent of conflicts with 4 per cent of the labour force) because of their education, aspirations and allocation to unpopular mines. (p. 176) | Table 7 conflict rates per 1000 employees, Tables 4-5 and 8, TEBA reports and HRL interviews. | Partial data |
| Dependence on foreign labour exposed the industry to political change in Malawi, Lesotho and Mozambique, producing conflicts involving those nationals and pushing mines toward recruiting South Africans. (p. 133) | Chamber correspondence, TEBA and mine reports, Table 2/3 labour proportions, scholarly literature on the supplying states. | Partial data |
| Newspaper reports are a poor guide to mine conflicts: only about one third of recorded conflicts were reported in the press. (p. 41) | Comparison of press reportage with internal records (Table 1). | Systematic data |
| Absence of formal bargaining structures and poor communication led workers to use informal, spontaneous collective action and hostel/underground settings facilitated mobilization and escalation. (p. 355) | Case studies (hoisting sit-downs, death benefit strikes, 1974 Free State), Table 30, accounts of PA and committee system. | Illustrative or anecdotal |
The author's caveats¶
Author states no regression analysis was possible: ten years is too short, conflict incidents are unrepresentative because mines were not required to report them centrally, and South African economic and Black unemployment data are unreliable. He notes eyewitness accounts are biased toward security concerns, information on many cases is incomplete (Table 24 covers only 80 of 88 cases; Table 30 143 of 158), the adjustment phase is particularly incomplete, and shift losses are conservative estimates.
Gaps in the evidence¶
The case database rests on management, TEBA and Chamber files shaped by security concerns, with few direct worker voices, so worker-perception claims (injustice, solidarity) are inferred rather than measured. Individual-level wage, job and origin records for strikers versus non-strikers, held in mine personnel files, could test the job-cleavage and domination arguments systematically, and mine-level panel data on wages, labour composition and conflict (which existed within the company records) would allow a statistical test of the turnover and deprivation explanations. Corroboration from interviews with workers at the time, or later oral histories, is largely absent; BCP involvement and Russian-gang claims rest on informant and company security reports.
Replicability: Data poorly documented.