Education, Bantu Education and skills¶
What were the aims and effects of Bantu Education and apartheid schooling and skills policy, and do school quality and returns explain labour outcomes?
Works engaging the debate: 9
Positions¶
- Social control and reproduction vs serving capital's labour needs
- Hegemony and consent vs coercion
- School quality matters (Case and Deaton) and drives low African returns
- Market vs state-coordinated skills formation
Evidence base¶
The debate rests on two largely separate evidentiary traditions. The historical and political-sociological literature on the aims of Bantu Education draws on a small canon of official documents: the Eiselen Commission (UG 53/1951), Verwoerd's 1953 House of Assembly speeches, the Bantu Education Act of 1953 and the Christian National Education manifesto (Johnson (1982)) (Giliomee (2009)) (Hyslop (1987)). Quantitative description comes mostly from secondary compilations rather than departmental series. These include Horrell's SAIRR volumes, the annual SAIRR Surveys of Race Relations, Behr and Macmillan, Malherbe, Hartshorne's enrolment and examination figures, Sadie's demographic estimates, and the per-pupil spending series reworked by Giliomee and Schlemmer and by Seekings and Nattrass (Johnson (1982)) (Giliomee (2009)). Johnson adds a 1978 stratified survey of 100 Umtata whites and Daily Dispatch reporting. Hyslop is the most archival of these works. He uses ANC, CATU and TUATA records at the level of the movement and individual cases, a Potgietersrust trial record, the Cillie Commission volumes, and township and national newspapers to trace school boards and committees from Langa to the Bantustans (Hyslop (1987)). McCulloch's review reports chapters in Kallaway's volume built on discourse analysis of the Eiselen, De Lange and Walters reports, ANC exile archives, and about 80 teacher interviews (McCulloch (2004)). The skills strand, represented by two evidence ledger entries for the HSRC volume (McGrath et al. (2004a)) (McGrath et al. (2004b)), uses a longer run of commission reports. These run from SANAC 1903-05 through the De Villiers (1948) and Botha (1953-56) commissions to Riekert, Wiehahn, Kleu, De Lange and the NTB/HSRC investigations of 1984-91. They are supplemented by selected Union Education Department annual reports, Cape and Transvaal archive files, the 1952 Keiskammahoek Rural Survey and Senior Certificate 2001 school-level entries. For the post-1994 period the volume relies on firm-level evidence: the HSRC Baseline Survey of Industrial Training, World Bank Greater Johannesburg, BMR and P-E surveys, Department of Labour levy and SETA reports, and an FET graduate tracer study.
The second tradition, which addresses whether school quality explains labour outcomes, is econometric and based on microdata. Moll estimates semi-parametric Mincerian earnings functions by race for urban workers on the 1960, 1970 and 1980 census microdata, the BMR household surveys of 1975 and 1985, the 1980 CPS, the November 1990 CSS/HSRC survey and the 1993 PSLSD. He extends these back to 1965 with weighted least squares splines on grouped BMR tables, and sets the results against pupil-teacher ratios, teacher qualifications and real per-pupil spending compiled from Bezuidenhout, Department of Statistics reports and Horrell (Moll (1996b)). Anderson et al. use the 1995 October Household Survey at the individual level, with age profiles of attainment standing in for cohorts. They also draw on the 1993 SALDRU living standards survey, the 1996 census 10 per cent sample, and the Education Atlas and School Register of Needs at the district and school level. They rely on Case and Deaton's matching of 1991 magisterial-district pupil-teacher ratios to household data (Anderson et al. (2001)). Lemon's comparative piece adds a small establishment-level survey of nine Harare secondary schools, covering levies, class sizes, teacher qualifications and GCE pass rates, alongside the South African Education Renewal Strategy and NEPI documents (Lemon (1995)).
Contested evidence¶
The sharpest interpretive conflict concerns the same few texts. Johnson reads the Eiselen terms of reference and Verwoerd's remarks as proof of a deliberate design to prepare Africans for subordinate roles, illustrated with syllabus and textbook quotations taken second-hand (Johnson (1982)). Giliomee rereads the same Eiselen passages, Verwoerd's speeches and his Stellenbosch notes as the programme of a modernising social engineer. In this reading the policy responded to the collapse of the mission system and to employer demand for literate semi-skilled labour (Giliomee (2009)). Yet the employer-demand claim is supported only by a Posel quotation and a 1967 Financial Mail supplement, with no employer records. Neither side therefore tests intent against administrative or business archives.
The spending evidence is also read selectively. Johnson stresses a white-African per-pupil ratio above 10:1 and a widening gap (Johnson (1982)). Giliomee accepts that real African per capita spending fell in the first decade but sets this against rapid enrolment growth and rising matric passes (Giliomee (2009)). Lemon cites a ratio of 3.6:1 by 1991/2 (Lemon (1995)). These figures come from different compilers, years and deflators, and are rarely reconciled.
On hegemony versus coercion, Hyslop treats the participation of some 300 boards and 4,000 committees as evidence of a real if failed consent strategy (Hyslop (1987)). The evidence on extraction, purges and Bantustan despotism, however, consists of press and TUATA cases rather than board minutes or membership data.
On school quality, Moll attributes the near-zero African primary return (about 0.02-0.034 per year, 1975-90) to deteriorating quality after 1953. He infers this from the timing of rising pupil-teacher ratios and falling real spending between 1955 and 1965. He also concedes competing explanations from the same period: rising relative supply of primary-educated workers in the 1970 and 1980 censuses, mine wage increases, and wage-board compression of differentials (Moll (1996b)). His own data do not separate these. Anderson et al. accept the Case and Deaton finding that pupil-teacher ratios matter for attainment, returns and employment. They warn, however, that African returns above grade 5-7 (15-25 per cent) may be inflated by grade repetition, which they infer only indirectly from OHS grade-for-age patterns (Anderson et al. (2001)). Lemon's Harare evidence of small pass-rate differences across very unequally resourced schools sits awkwardly with a simple resource-based quality story, though it comes from only nine schools (Lemon (1995)).
On market versus state skills formation, the HSRC volume reads the NTB/HSRC and Wiehahn documents as entrenching a voluntarist low-skills equilibrium. The post-1994 claims rest on enterprise surveys with differing samples and definitions, which the volume pools by meta-comparison rather than harmonising (McGrath et al. (2004a)) (McGrath et al. (2004b)).
Underused and missing sources¶
Almost every evidence ledger notes the absence of the Native Affairs and Department of Bantu Education files, the Eiselen Commission's oral and written evidence, and cabinet and National Party caucus records. These are what could settle the intent debate between (Johnson (1982)) and Giliomee (2009). Hyslop's account of the boards would be tested by board and committee minutes, membership lists and finances, which are not used (Hyslop (1987)). The historical skills chapters lack systematic apprenticeship-registration series by race, a full run of Union Education Department reports, and census occupational tables beyond Hurwitz's 1936 and 1951 figures (McGrath et al. (2004a)) (McGrath et al. (2004b)). Administrative records on repetition and matric pass rates by race, retrospective schooling histories and longitudinal data would test the repetition hypothesis (Anderson et al. (2001)). Test scores or cohort-specific quality measures linked to workers would let Moll's quality, supply and wage-regulation explanations be separated (Moll (1996b)). Employer and industrial-council records on demand for educated African labour are not used at all.
Strength of key claims¶
| Basis | Share of key claims |
|---|---|
| Systematic data | 11% |
| Partial data | 61% |
| Illustrative or anecdotal | 19% |
| Asserted | 9% |
Comparative figures for all debates are on the Literature overview.
Best-evidenced works¶
Moll (1996b), Anderson et al. (2001), McGrath et al. (2004b), McGrath et al. (2004a), Hyslop (1987)
Most-used primary sources¶
Holdings are matched by machine against the archive's file manifest; a dash means no match was found, not that the project certainly lacks the source.
Works and their positions¶
| Work | Position |
|---|---|
| Anderson et al. (2001) | Argues that repetition, not non-enrollment, drives the gap, citing enrollment of 97% or more for Africans to age 14 and about 0.8 grades per year of age. |
| Anderson et al. (2001) | Sides with Case and Deaton and Case and Yogo that pupil-teacher ratios strongly affect attainment, returns to education and employment, treating South African families as having had little control over school quality. |
| Anderson et al. (2001) | Finds low returns through about grade 5, then high returns of 15-20% per year above grade 7. These are higher for Africans than whites, though the authors say they should be discounted for repetition. The paper notes this is higher than Moll's earlier estimates for primary schooling. |
| Giliomee (2009) | Rejects the view that it rested on inferiority or Aryan-race ideas. Argues that the aim was 'modernising racial domination' and that the policy still perpetuated racial inequality. |
| Giliomee (2009) | Holds that the system met the 1950s demand for semi-skilled operatives with basic literacy, and that secondary expansion followed business pressure only from the mid-1960s. It also served the state's need for social control over urban youth. |
| Giliomee (2009) | Accepts some validity in the Herrenvolk comparison but prefers Hyslop's analogy with mass schooling in the USA and Britain, noting that Verwoerd's policy did not draw on National Socialism. |
| Hyslop (1987) | Builds on Christie and Collins. Argues Bantu Education had a genuine hegemonic component, an attempt to win consent, but that official racism and authoritarianism undercut it. |
| Hyslop (1987) | Argues that writing on 1976 has generally ignored the school boards' role in opposing Afrikaans as a medium of instruction from 1974. |
| Hyslop (1987) | Warns against a simple picture in which 'the people' reject the system and only a handful of 'traitors' participate. Finds significant constituencies willing to enter the board structures for varied motives. |
| Johnson (1982) | Sides with a social-control and socialisation view: education is a mechanism of stratification and legitimation, not a neutral route to mobility; also argues that economic-growth rationales for reform do not change this. |
| Johnson (1982) | Sceptical of the liberal expectation that business needs for skilled labour will erode apartheid; sees business training programmes and constructive engagement as 'neo-apartheid' adapting the 'floating colour-bar' to white needs. |
| Johnson (1982) | Argues Transkei independence was cosmetic; white schools were leased back to the Cape authorities and African education was still administered by Pretoria-seconded white officials. |
| Lemon (1995) | Sides with critics of the 1979 Act. Argues that fee-based, property-linked school categories and South Africa's models B and C replace race with class. |
| Lemon (1995) | Tentative: pass-rate differences across Harare schools were small, so resources may matter little above a threshold. Calls for more detailed research and notes resources may affect personal development. |
| Lemon (1995) | Argues South Africa's education spending is already high, so improvement must come from redistribution and renewed economic growth rather than larger budgets. |
| McCulloch (2004) | The volume, as reported by the reviewer, stresses both major convulsions and longer-term continuities. Kraak argues that structural continuities have limited real change after 1994. |
| McCulloch (2004) | Kallaway's introduction raises the question of how far it was colonialism for indigenous peoples as practised elsewhere and how unique it was. The review does not report a settled answer. |
| McCulloch (2004) | The volume emphasises the gap between policy and practice, and teachers' responses ranged from outright resistance to active acceptance. |
| McGrath et al. (2004a) | Accepts the institutionalist framework but argues the high skills thesis exaggerates the spread of high skill production and neglects intermediate skills and middle society. Proposes a hybrid, differentiated model and sees a positive low-skills job strategy (Ashton) as legitimate. |
| McGrath et al. (2004a) | Kraak argues market-led training under late apartheid reflected the triumph of ideology over evidence, and that market solutions alone are insufficient; the post-1994 levy-grant and SETA system is a state-led corrective. |
| McGrath et al. (2004b) | The volume accepts the social-institutionalist high-skills framework but argues it must be adapted: Ashton advocates a positive low-skills job-creation route before upgrading, while Kraak proposes a hybrid, differentiated model with simultaneous upskilling at low, intermediate and high levels. |
| McGrath et al. (2004b) | Argues that market solutions are not enough; late-apartheid voluntarism and employer-led training perpetuated low skills, and the NSDS levy-grant and planning regime is a necessary, if capacity-constrained, corrective. |
| McGrath et al. (2004b) | Badroodien and Paterson stress social control, 'industriousness' and rural containment of Africans over economic skills formation, rather than a purely labour-market logic. |
| McGrath et al. (2004b) | McGrath and Kraak argue policy incoherence, especially the DoE's neglect of enterprise and SMME skills, is a major block to delivery. |
| McGrath et al. (2004b) | Argues reforms from the 1970s responded to economic downturn, worker and student militancy (1973, 1976), yet were deliberately partial and left racial schooling and employer voluntarism intact. |
| Moll (1996b) | Rejects discrimination as the explanation for low primary returns, since African secondary returns were high. Points to school quality (pupil-teacher ratios, teacher qualifications, spending). |
| Moll (1996b) | Argues that Industrial Council and Wage Board minimum wages and union 'living wage' policies compressed skill differentials, lowering primary returns and probably raising African unemployment. |
| Moll (1996b) | Favours improving primary quality and access, funded by reallocating from secondary and tertiary spending and raising fees at those levels, on equity grounds. |
| Moll (1996b) | Sees the mine wage increases and local recruitment as raising wages of the uneducated and so compressing the primary-schooling differential. |