Class conflict and colour bars in the South African gold mining industry, 1910-26¶
Johnstone (1971). Class conflict and colour bars in the South African gold mining industry, 1910-26.
| Type | other |
| Scope | Economics and economic history of apartheid |
| Method | theoretical or conceptual |
| Period studied | 1910–1926 |
| Regions | Witwatersrand, Transvaal, South Africa |
| Themes | mining migrant labour, labour markets wages, unions strikes, business and state, origins segregation, politics ideology elections, comparative theory |
| Citation key | johnstone1971 |
Argument¶
The paper argues that the 1922 Rand revolt was a conflict between two white classes over the form racial discrimination should take. It was not a contest between rational economic development and racial prejudice. The mine owners' profits rested on a 'wage colour bar': a monopsonistic recruiting system plus a ceiling on African wages that kept African labour cheap. White workers defended a 'job colour bar' that reserved skilled work for whites. As a falling rate of profit pushed the Chamber of Mines to attack the job bar at the margin, conflict followed. The author criticises the liberal view, typified by Hutt, which blames the job colour bar and white labour while ignoring the capitalists' dependence on racial discrimination. He frames colour bars as instruments of class interest, using a structural, quasi-Marxist argument.
Evidence and method¶
Mainly official reports and Chamber of Mines statements: the Low Grade Mines Commission (1919-20), the Economic and Wage Commission (1925), the Mining Industry Board (1922), the Unemployment Commission (1921) and Gold Producers' Committee statements. Also archival files from the Department of Mines and Industries and secondary literature (Hutt, Doxey, Horwitz, Frank, Arrighi, Hobsbawm).
Place in the debates¶
- Liberal vs revisionist/class interpretation of the colour bar and the 1922 conflict (see The job colour bar, white labour and its erosion). Rejects the liberal thesis (Hutt, Doxey, Horwitz) that mine owners were rational modernisers attacking an irrational white-labour colour bar. Holds that the owners depended on the wage colour bar and only contested the degree of protection white workers had won.
- Cheap labour thesis / role of capital in racial discrimination (see Cheap labour thesis and the function of the reserves). Strongly supports it. Cheap, rightless African labour was the foundation of Rand profitability, and the mining capitalists actively created and operated the wage colour bar.
Findings¶
-
Ownership of the Rand gold mines was highly concentrated, which enabled the industry to act as a monopolistic, unified body. (p. 2)
The ownership and contslal of the Witwatersrand gold fields was concenkated in the hands of a few groups of capitalist oonipdes.
-
White workers' basic class interest was to defend their privileged labour-elite status through the job colour bar rather than seek solidarity with African workers. (p. 4)
thus the protection ead consolidation of their privileged s t a b as a labour Blite and as
-
Mine profits depended on cost control because of low ore grade and a fixed gold price. White wages were hard to restrain, so African wages became the target. (p. 5)
Therefore, the fkdaumta3 ama of ooet reduction had to be that of non-wte -8,
-
The Native Recruiting Corporation schedule from 1912 set a maximum average piece-work wage for Africans, with fines for companies that exceeded it. (p. 6)
wage payable to Africane on piece work t o 2/3d. per shift, with the penalty of a fine for a q coqpaay exceeding: it.
-
The 1922 strike ended in a bloody defeat of white workers, and the Chamber obtained all its goals. (p. 8)
ended with 230 dead & twice as v hjwed.
-
The mine owners did not oppose the colour bar system as such; they relied on it through the wage colour bar. (p. 11)
Far from being opposed to the colour bar system, they depended on it, thmugh the wage colour bar
-
The mining capitalists' attack on the job colour bar is compared to the European bourgeoisie's attack on feudalism, an attempt to reform the relations of production to suit their own class interests. (p. 11)
colour bar, like the attack of the Etmopean bourgeoisie agsinet feudalism and deepotic mcmarchy, waa essentially sn ettempt to
-
The conflict was fundamentally over profit realisation, though it took the specific form of a conflict over 'the colour bar'. (p. 12)
It was this profits realization problem that the conflict waa essentially about.
-
A note cites an estimate that African mine real wages were no higher, and possibly lower, in the mid-1960s than in 1911. (p. 13)
it has been eathated that the real wages of kfrican gold mine workere were no -er and were possibly even lower in the aid-19608 than in
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Statements of Evidence of the Gold Producers' Committee of the Transvaal Chamber of Mines, 1925 (Statements no. 3 and 5) | employer association records | 1925 (covering 1910s-1924) | Supplies Chamber estimates of cost sensitivity, production costs per ton, and the Chamber's own statements on the need for native-specific laws and cheap labour. |
| Report of the Economic and Wage Commission, 1925, U.G. 14/26 | commission or official report | 1925 (data 1910-1922) | Source for the observation of an absent labour-market response to shortage, and for the index of white miners' real wages 1910-1922. |
| Report of the Low Grade Mines Commission, 1919-20, U.G. 24/20 | commission or official report | 1919-20 | Source for the maximum average wage system, African and white wage trends 1912-1919, production cost rises, and African grievances about low wages. |
| Report of the Mining Industry Board, 1922, U.G. 39/22 | commission or official report | 1922 | Source for the Status Quo Agreement text, the gold price and cost squeeze, and the fall in white employment after the 1922 strike. |
| First Interim Report of the Unemployment Commission, March 1921, U.G. 16/21 | commission or official report | 1921 | Supports the estimate that returning to the normal gold price would put 28 mines out of business. |
| Memorandum of the Under Secretary of Mines and Industries to the Secretary of Mines and Industries, 13 August 1924, Dept of Mines and Industries File MM 1699/24 | state archives | 1924 | Supports the estimate of how many white employees lost their jobs after 1922 had been re-employed by 1924. |
| Report of the Native Grievances Inquiry, U.G. 37/14 | commission or official report | 1914 | Cited (with Low Grade Mines Commission and African witness evidence) to show African miners' main grievance was low wages. |
| Evidence of African mine workers and their representatives to various government commissions | commission or official report | 1910s-1920s | Backs the claim that African miners saw low wages rather than the job colour bar as their key grievance. |
| Annual Reports of the Transvaal Chamber of Mines | employer association records | c.1890s-1920s | Cited (with other items in note 5) for the average ore grade on the Rand. |
| Merriman Papers, 1908, No. 86 (letter J. X. Merriman to J. C. Smuts, 16 July 1908), South African Public Library, Cape Town | state archives | 1908 | Contemporary observation of poor-white and new-comer attitudes to Africans, used for the white workers' racial elite orientation. |
| Parliamentary Papers, Cd. 1892 (London, 1904): Further correspondence relating to the Affairs of the Transvaal and Orange River Colony; Transvaal Labour Importation Ordinance Schedule I; Farrar speech at Benoni 1903 | parliamentary debates | 1903-1904 | Documents the Chamber's initiative in securing job colour bar provisions in the 1904 ordinance, showing accommodation to the job bar. |
| Status Quo Agreement of 1918 between the Chamber of Mines and white workers (as quoted in the Mining Industry Board report) | employer association records | 1918 | Shows consolidation of the job colour bar and the terms of the later conflict. |
| Wilson, D. G., 'Gold Wage Inquiry a Must', Financial Mail, 10 May 1968 (based on Ph.D. research) | secondary literature compilation | 1911-1960s | Used for the estimate that African real wages in the mid-1960s were no higher than in 1911. |
Inference¶
Quantification: Illustrative figures.
Theoretical, class-analytic argument illustrated by selected official figures and quotations from commission reports and Chamber statements. A brief historical narrative of 1910-26 is used to reinterpret the 1922 conflict, and the author critiques the liberal interpretation by reasoning from the structure of costs and class interests.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Mine ownership was highly concentrated and centralised through the Chamber of Mines and joint recruiting bodies, giving a monopolistic control over labour conditions. (p. 2) | Historical narrative on the group system, Chamber of Mines, WNLA and NRC, supported by a quoted economist's characterisation; no data on concentration. | Illustrative or anecdotal |
| Mine profitability depended on low working costs because of low ore grade and fixed gold price, so that African wage suppression was the key cost lever; a 1s per ton cost difference meant about £1.5 million in profits. (p. 5) | Chamber of Mines calculations (Gold Producers' Committee statements), ore grade statements, Low Grade Mines Commission data. | Partial data |
| A wage colour bar (recruiting monopsony, WNLA, and the maximum average wage system) held down African wages, which rose far less than white wages and fell in real terms by 1919. (p. 6-7) | Low Grade Mines Commission and Economic and Wage Commission figures on wage rates and real wage indices; Chamber statements. | Partial data |
| The 1922 conflict was triggered by a squeeze on profits (falling gold premium, rising costs to 25/8d per ton in 1922), leading the Chamber to abandon the Status Quo Agreement; the strike was crushed and costs fell afterwards. (p. 8-9) | Mining Industry Board 1922 report, Low Grade Mines Commission, Unemployment Commission, Gold Producers' Committee statements on costs 1921-24. | Partial data |
| The mine owners were not fundamentally opposed to the job colour bar; they accommodated it (1904 Ordinance, 1911 Regulations, Status Quo Agreement) and opposed only the degree of protection after the profits crisis. (p. 11) | Parliamentary papers on the Farrar speech and the 1904 Ordinance, the 1911 Regulations, the Status Quo Agreement and the Hildick-Smith ruling, narrated rather than systematically documented. | Illustrative or anecdotal |
| Liberal accounts that portray the conflict as rational economic development versus racial prejudice misrepresent it by neglecting the wage colour bar; the conflict was over the form of racial discrimination and distribution of its benefits. (p. 11) | Critique of Hutt, Doxey, Horwitz, Hessian and commission reports, with reference to Chamber policy documents and African grievance evidence. | Asserted |
The author's caveats¶
The author states that references are kept to a minimum for reasons of space and that the paper does not offer a detailed study of any particular aspect, only a general analysis of the structure of the conflict.
Gaps in the evidence¶
The argument would be better tested with systematic Chamber of Mines and company records on wage bills, working costs, and labour complements by mine and year, and with the Chamber's internal correspondence to show whether it intended to restructure the job colour bar. Direct evidence of the Chamber's strategy in 1921-22 (minutes, executive committee papers), and of white union records, would help; most of this exists in archives. Quantitative comparison of the savings from substituting Africans for whites, and a fuller series of real wages, would also give stronger support for the profit-squeeze interpretation.
Replicability: Data poorly documented.