Skip to content

New Labour Frontiers for Old: Black Migrants to the South African Gold Mines, 1920-85

Literature › Works

David Yudelman; Alan Jeeves (1986). New Labour Frontiers for Old: Black Migrants to the South African Gold Mines, 1920-85. Journal of Southern African Studies, Vol. 13, No. 1. Taylor & Francis / Oxford University Press.
Type journal article
Scope Economics and economic history of apartheid
Method archival narrative
Period studied 1920–1985
Regions Witwatersrand and Orange Free State gold fields, Mozambique, Malawi, Lesotho, Southern Rhodesia, Bechuanaland, Northern Rhodesia and Nyasaland
Themes mining migrant labour, labour markets wages, business and state, origins segregation, sanctions divestment, unions strikes
Citation key yudelman1986

Argument

The authors argue that the 1970s shift in the sourcing of black gold-mine labour from foreign to South African supplies was not a state-imposed policy of 'internalisation'. It was driven mainly by the closing of foreign labour frontiers (Malawi 1974, Mozambique 1976) and the loss of the option of expanding to new ones. Mine labour policy was a ceaseless quest for stability through geographic expansion to new sources, and this let the Chamber of Mines keep the low-wage migrant system intact for decades, notably through the 1930s-40s push into the tropical north. Only when the frontiers closed, combined with a high gold price, did wages rise and stabilisation begin. The paper is a study of state-capital relations and the origins of migrants, not of the supplying areas or the migrants' experiences.

Evidence and method

Chamber of Mines and WNLA annual reports; archival files (Transvaal Archives, Botswana and Zimbabwe national archives, PRO Kew); government commission reports (Lansdown, Low Grade Mines); internal Chamber and Human Resources Laboratory reports of the 1970s; interviews with Tony Fleischer (1984); press; and an appended Chamber of Mines table of labour sources 1920-83.

Place in the debates

Findings

  • The share of South African blacks in the mine labour force moved in a long cycle: high in 1936, low in 1973, and high again by 1985. (p. 3)

    By 1936, however, 58 per cent of the black labour force originated in South Africa. With the reopening of recruitment in the north at that time the South African proportion went into a long decline and fell to 22 per cent in 1973.

  • The 1970s change was a shift to nearer, safer sources rather than nationalist-based policy; Lesotho's share grew while Malawi and Mozambique were cut. (p. 3)

    the changes occurring from the mid-1970s should be interpreted less as a nationalist- based policy of 'internalisation' than as a shift to sources closer at hand and less vulnerable to the possibility of unilateral withdrawal by a hostile government.

  • The labour system's stability depended on its ability to keep expanding to new sources; the geographical limits were reached in the 1960s and 70s. (p. 5)

    Like many empires, the mines' labour system has remained stable only while it has been able to expand. In the 1960s and early 1970s, the geographical limits to growth were finally reache&.

  • Black wages were the 'give' in the system during adversity and low wages underpinned the industry's profits; industry data on low-grade ore was manipulated to justify wage rigidity. (p. 7)

    wage rates, especially black rates, constituted the 'give' in the system during periods of adversity.

  • The tropical recruiting investment of the 1930s encouraged the Chamber to hold a hard line on wages, unionisation and the colour bar for roughly thirty years. (p. 13)

    All of these policies and prejudices might certainly have been called into question had it not been for the prospect that tropical labour would rescue the mines' traditional labour system. For thirty years at least, it did exactly that.

  • Reliance on foreign labour with infinitely elastic supply held down black mine wages; internal Chamber documents admitted cash wages had slightly fallen since 1935. (p. 16)

    showing concern that the average cash wages paid to black miners had actually fallen off slightly since 1935. The reason given for this was that most of the labour was drawn from foreign sources and the elasticity of supply had therefore been infinite.

  • The 1974 Malawi withdrawal, which followed an air crash, collapsed the Malawian contingent and was costly and disruptive for the industry. (p. 18)

    the result was the decline of the Malawi contingent at the mines from 120,000 to virtually nothing by the end of 1975.

  • Mozambican supplies fell sharply in 1976 because of new administrative procedures, not a deliberate withdrawal. (p. 18)

    in January 1976 there were 114,385 Mozambicans on the mines; by the end of December of the same year there were only 48,565.

  • Large wage rises in the mid-1970s resulted from the combination of high gold price (ability to pay) and fear of labour shortage. (p. 21)

    The steep black wage increases of the mid-1970s finally resulted from a combination.of an ability to pay, from the acuteness of the labour shortage, and the fear of even worse shortages.

  • By the mid-1980s the labour force was stabilising, with South Africans a majority and Lesotho the largest foreign source. (p. 23)

    By mid-1985 some mines had a 90 per cent rate of return, and the enrolment of novices was running at only about 15 per cent of the workforce. Recruitment from Lesotho was 23.4 per cent of the 436,000 total

Evidence ledger

Sources used

Source Kind Period How used
WNLA Annual Reports, 1902-18, 1928-39 / 1934-45 / 1970-8 / 1974-8 chamber of mines or recruiting records 1902-1978 Document fluctuations in recruiting levels by supplying area, tropical recruiting levels and the withdrawals of the 1970s.
South African Chamber of Mines, Annual Reports (CMAR) for 1929 and 1932 (Summary of Evidence before the Low Grade Ore Commission) chamber of mines or recruiting records 1929-1932 Used for the Mozambique quota response in 1928-29 and as examples of industry-crafted data presented to commissions.
Low Grade Mines Commission 1919-20, Final Report (UG 34-'20) and Chamber evidence commission or official report 1919-1920 Cited as an example of commission reports that followed the industry's low-grade argument and endorsed renewed access to tropical labour.
NRC evidence to the Mining Industry Board 1922, TAD GNLB 339, 81/22/243 state archives 1922 Illustrates industry evidence to commissions that framed low-grade mine survival as essential.
Report of Witwatersrand Mine Native Wages Commission, 1943 (UG 24-1944) [Lansdown Commission] commission or official report 1943-44 Cited for the 'pay limit' calculations and the industry's rigidity on wages and government assistance.
Union of South Africa Secretary for Mines Archive (MNW 804 2489/25) incl. Neergaard to Gemmill 21.5.1928 state archives 1928 Source for P. Neergaard's 1928 tour report on the recruiting potential of Nyasaland, Northern Rhodesia, Tanganyika and Mozambique.
Zimbabwe National Archives, CNC S1561 series (Gemmill, CNC, PM correspondence; 1939 employers conference) state archives 1936-1939 Document Gemmill's negotiations and threats with Huggins and the Southern Rhodesian government's reactions to WNLA expansion.
Public Record Office, CO 525/166/44053/3, minute by J. A. Calder 4.2.1937 state archives 1937 Shows the Rand mines' 1937 expectation of drawing 70,000 miners per year from the north, which the authors call an over-estimate.
Botswana National Archives, Secretariat series S344 (WNLA northern Bechuanaland organisation; medical commission 1947; Baring letter) state archives 1936-1947 Documents WNLA's road, depot, airstrip and medical infrastructure building in Bechuanaland.
TAD MNW 804 2489/25 and NTS 2124, NTS 2115 files (Gold Producers Committee, Chamber/Native Affairs meetings, Mines ministers' letters) state archives 1928-1936 Support the claim that WNLA committed itself by the mid-1930s to expensive extension of its northern network.
Chamber of Mines President's letter to Minister of Mines, 6 October 1961 chamber of mines or recruiting records 1961 Shows the Chamber's argument about labour shortage and the need to recruit freely abroad.
J. A. Gemmill, 'Native Labour on the Gold Mines', Seventh Commonwealth Mining and Metallurgical Conference, 1961 company records or reports 1961 Used as the basis of the Chamber's case for migrant and foreign labour and admission that the colour bar was of academic interest under migrancy.
Tony Fleischer interviews, conducted by D. Yudelman, 8 and 14 March 1984 interviews oral history 1984 Evidence for Banda's Malawi withdrawal, Anglo American's phasing out of Mozambicans, urban recruiting, and state bureaucracy positions.
J. A. Parsons, 'A Review of the Factors Affecting the Supply in the Short-Term of Black Manpower to the Gold Mining Industry', Human Resources Laboratory Research Report 51/74 company records or reports 1974 Provides the 1974 assessment of the potential for recruiting South African blacks and wage competitiveness.
Mine Labour Organisations (NRC)(WENELA), 'Investigation into the Future of Mine Labour Organisations', Interim Report, June 1974 company records or reports 1974 Fleischer's first detailed plan for the Chamber's recruiting organisations.
The Employment Bureau of Africa (Teba) Five Year Plans 1978-83 company records or reports 1974-1983 Cited for costs of the Malawi withdrawal, warnings against over-reliance on one source, and the two-tier labour plan.
Parsons and Mashaba, 'The Recruitment of Black Workers for the South African Gold Mines in the 1970s and 1980s', HRL Research Report 22/80 company records or reports 1980 Shows the Chamber's assessment that joint action by supplier countries was probably groundless.
Appendix: Principal Sources of Mine Labour, 1920-83 (Chamber of Mines data, WNLA member companies) chamber of mines or recruiting records 1920-1983 Annual table of black mine labour by source area underpinning the sourcing-trend claims.
Francis Wilson, Labour in the South African Gold Mines (Cambridge, 1972) secondary literature compilation 1889-1969 Relied on for the claim that black real wages did not rise 1889-1969 while white earnings did.
C. W. Stahl, 'Migrant Labour Supplies, Past, Present and Future' in Bohning (ed.), Black Migration to South Africa (Geneva, 1981) secondary literature compilation 1970s Used for the lack of increased South African recruits after wage rises, black unemployment and decline in agricultural employment.
Ruth First, Black Gold: The Mozambican Miner (1983) secondary literature compilation 1975-1976 Source for Mozambican miner numbers on the mines in 1975-76 (115,309; 114,385 falling to 48,565).
Financial Mail, 'Recruitment the Teba Way', Gold: A Survey, 19 July 1985 newspapers periodicals 1985 Source for mid-1985 composition of the workforce by origin and return rates.
Toronto Globe and Mail, July 30 and August 5, 1985; S.A. Mining World, Dec 1983; Optima articles (Gemmill 1952; Wilson 1984) newspapers periodicals 1952-1985 Document Botha's threat to expel migrants, Fleischer's comments on SALC, and Gemmill's northern recruiting targets.

Inference

Quantification: Descriptive tables.

Archival and documentary narrative history of state-capital relations and recruiting policy, supported by a time-series table of labour origins by region (1920-83) and a few figures on wages, gold price and recruit numbers; elite interview and internal Chamber documents are used for the 1970s.

Key claims

Claim Basis Strength
The proportion of South African blacks in the mine labour force fell from 58% in 1936 to 22% in 1973, then rose to about 60% by mid-1985, as shifts to sources closer at hand rather than a state policy of 'internalisation'. (p. 102) Chamber of Mines/WNLA data in the appendix table and mid-1985 Financial Mail figures; interpretation by authors. Systematic data
The 1974 Malawi withdrawal and the Mozambique disruption of 1975-76 were the catalytic events that forced the industry to local sourcing and higher wages. (p. 117) Fleischer interviews, Teba plans, Parsons report, Ruth First's figures and the appendix table. Partial data
The Chamber and WNLA invested heavily from the 1930s in an expensive northern 'tropical' recruiting strategy that, despite slow returns, shaped labour policy for three decades and discouraged alternatives such as stabilisation or mechanisation. (p. 112) WNLA Annual Reports, Zimbabwe and Botswana archives, MNW and NTS files; WNLA's tropical numbers (about 20,000 by 1939). Partial data
Industry data and the commissions of inquiry presented low-grade ore arguments in a self-serving way, and historians have been misled by this evidence in explaining migrant labour as an economic necessity. (p. 106) Chamber evidence to commissions (1919-20, 1922, 1930/32, 1943) examined critically; authors' reading of the documents. Illustrative or anecdotal
By 1970 both state and industry knew black wages had stagnated because of elastic foreign supply, yet the industry chose to continue low-wage foreign-dominated recruiting until the gold price rise and 1974 crisis changed ability and motivation to pay. (p. 115) Internal Chamber documents and James Gemmill's 1970 briefing, Chamber-Koornhof exchange, Wilson's data, gold price changes. Partial data
The state and the Chamber were symbiotic in the 1975 crisis, as shown in trilateral meetings with the farmers' union to open agricultural recruiting areas, though the concessions yielded little labour. (p. 120) Chamber letters and Fleischer interviews; Stahl on agricultural employment decline. Illustrative or anecdotal
By the mid-1980s the labour force was stabilising: average stay lengthened to 13.5 months from 8, with return rates of 90% on some mines and novices only about 15% of the workforce. (p. 122) Lemon (1984), Fleischer interviews, Financial Mail 1985. Partial data

The author's caveats

The authors caution that the Chamber's and commissions' data were industry-crafted and must be treated critically; they note Calder's 1937 estimate was an extravagant over-estimate and that forecasting is hazardous. Little is said about the reliability of the appendix series.

Gaps in the evidence

The argument leans heavily on Chamber of Mines and WNLA documents and on one set of interviews with a Chamber insider, so it lacks independent evidence from state files and from supplying areas or migrants themselves. Wage series by group, mine-level labour costs, and labour turnover data would better test the claims about cost and the shift from foreign to local sourcing. Several are archived in company and Chamber collections; the archival material was accessible to the authors but the quantification of wages and productivity is thin.

Replicability: Data described but not public.