Origins, coherence and ideology of apartheid and Afrikaner nationalism¶
Was apartheid continuous with segregation or a rupture in 1948, a grand plan or improvised, and driven by ideology, class or Afrikaner nationalism?
Works engaging the debate: 31
Positions¶
- Continuity with segregation vs post-1948 break
- Grand design vs improvised and contested project (Posel, Evans)
- Class analysis of Afrikaner nationalism (O'Meara) vs ideological and ethnic accounts
- Purists vs pragmatists; populist white coalition vs elite design
Evidence base¶
The debate has been argued overwhelmingly from the published record of the state and of organised white politics. The most common sources are House of Assembly and Senate debates and English and Afrikaans newspapers. Next come the great policy commissions: Fagan (UG 28/1948), the HNP's Sauer Colour Question report (1947), Tomlinson (UG 61/1955), Smit, Van Eck and the Board of Trade and Industries reports. Statutes, usually as compilations, complete the core (Cornell (1960); Legassick (1974); Carter (1958); Davies et al. (1976)). Interpretive and synthetic works rest largely on secondary literature with selective official figures (Maylam (1995); Freund (2012); Posel (2011); Cobley (2007)). Theoretical treatments of apartheid's economic rationale use no primary data at all (Lundahl (1982a); Lundahl (1982b)). A smaller set of studies go into archives: Land Tenure Advisory Board, Group Areas Board and municipal files for the genealogy of the Group Areas Act (Mabin (1992)). Department of Native Affairs (NTS) files, Chief Native Commissioners' conference minutes and NBRI housing-cost surveys document the remaking of native administration (Evans (1997)). ASSOCOM and FCI Non-European Affairs files, Brakpan and WRAB municipal records, IANA proceedings and interviews trace 1960s Bantu Affairs policy-making (Posel (1990b)). Fagan and Tomlinson Commission evidence, together with SAAU, AHI, SABRA and Instituut vir Volkswelstand submissions, show intra-nationalist divisions (Posel (1987)). Further archival bases are DRC mission council minutes and the Malan collection (Giliomee (2003)), Malan's and his contemporaries' private papers (Korf (2010)), United Party and Smuts papers (White (1990)), and Sanlam, FVB and Federale Mynbou board minutes (Verhoef (2009)). The Broederbond literature rests on the 1944 Military Intelligence reports in the Hofmeyr Papers, the Botha Commission (RP 20/1965) and Bond-linked periodicals (O'Meara (1977)), or on leaked circulars and a 1977 'Professions and Ages Breakdown' (Wilkins and Strydom (1978)).
Genuinely quantitative evidence is thin and scattered. One strand is Christopher's segregation indices, computed from the 1951 census manuscript enumerators' returns: a median White index of 74.8, and a Coloured index of 49.3 against 44.3 in 1936 (Christopher (1994b)). Another is O'Meara's use of Union Statistics for Fifty Years and Spandau's real earnings index for white labour-process change and post-1948 wages (O'Meara (1978)). A third is Carter's and Korf's constituency and delimitation tables showing minority-vote victories (Carter (1958); Korf (2010)). White recalculates Afrikaans-speaking support for the UP-LP in 1948 at about 39.5% from roll data, and draws on UP minutes showing 674 of 1,696 branches moribund in 1947 (White (1990)). Charney uses Rapport polls and the 1982 Star Retail Data Library survey for the class basis of the 1982 split (Charney (1984)). A wheat thesis uses Wheat Board annual reports (1939-96) for prices, subsidies and per capita bread consumption, which rose from 23.2kg in 1948 to 47kg in 1980 (Author not named in text (MSc candidate (2012)). Census-derived urban African population growth figures (58% in the 1950s, 28% in the 1960s) underpin Posel's argument that influx control failed (Posel (2011)).
Contested evidence¶
The same commission texts are read in opposite directions. Legassick reads the Fagan and Sauer platforms side by side as substantially overlapping on influx regulation, labour bureaux and differentiation, and thus as evidence of continuity (Legassick (1974)). Posel reads the Sauer Report as an internally contradictory hybrid of 'total segregation' and 'practical' apartheid, not a blueprint (Posel (1987)). Giliomee treats it as an aggregate of grass-roots opinion, built from roughly 5,000 circulars and 500 replies, with unresolved labour contradictions (Giliomee (2003)). Mabin notes that neither the Sauer Report nor Cronje's texts contain any mechanism for compulsory residential segregation (Mabin (1992)). Davies et al. read the Fagan report instead as proof of the UP's inability to resolve agriculture's labour crisis (Davies et al. (1976)).
The Volkshandel editorial is central to O'Meara's claim that Afrikaner capital was hegemonic. Posel argues it begs the question when read against the AHI's own evidence (Posel (1987)). The Broederbond's weight turns on incompatible bodies of evidence. O'Meara uses the 1944 Military Intelligence membership breakdown to make the Bond a petty-bourgeois northern vanguard (O'Meara (1977)). Wilkins and Strydom infer government by the Bond from leaked membership overlap with the Cabinet and the civil service (Wilkins and Strydom (1978)). Giliomee, relying mainly on Stals's unpublished history and on absence of evidence, assigns the Bond little role in making the plan before 1948 (Giliomee (2003)). Posel invokes Broederbond ascent after Sharpeville from press reports and interviews (Posel (1990b)).
The 1948 election is explained from non-overlapping sources. White draws on UP organisational records and electoral rolls (White (1990)). The wheat thesis cites Hansard and NP pamphlets on bread rationing (Author not named in text (MSc candidate (2012)). Davies et al. use seat outcomes read as agricultural and white-worker defection (Davies et al. (1976)), O'Meara cites union capture via Bond front organisations (O'Meara (1978)), and Carter and Korf point to delimitation and rural weighting (Carter (1958); Korf (2010)). None tests its mechanism against the others with constituency data.
On continuity, Christopher's pre-1948 segregation indices and legislative chronologies support continuity (Christopher (1994b); Cornell (1960)). So do Maylam's synthesis of municipal case studies (Maylam (1995)) and Freund's account of parastatal persistence (Freund (2012)). Evans's NTS evidence of DNA centralisation under Verwoerd (Evans (1997)) and Posel's phase periodisation (Posel (2011); Posel (1990b)) instead locate a break, whether in administrative capacity or after 1960. In short, class accounts rest on aggregate series and organisational statements, while ideological accounts rest on church and intellectual texts. The two literatures rarely confront the same documents.
Underused and missing sources¶
Several evidence ledgers note that constituency-level returns for 1943 and 1948 exist (Heard; Weide and Weide) but have never been linked to district socio-economic data. Such a link would test the rationing, organisation, farm-labour and worker-mobilisation explanations of 1948 against one another.
Other sources are hardly used. Pass-law prosecution and influx-control statistics from the South African Police Commissioner's annual reports, and labour bureau placement records, would test claims about enforcement and failure (Cornell (1960); Evans (1997); Posel (2011)). Municipal expenditure by race has been asserted but never documented (Maylam (1995)). Agricultural census series on land and labour per farm would test the Lundahl models. Manpower Survey tapes are known only via Crankshaw.
The decisive internal records are almost entirely absent. Broederbond minutes and Bondsraad records, NP caucus and Cabinet papers, and BAD/DNA ministerial files are noted as missing by O'Meara (1977); Posel (1987); Posel (1990b); Giliomee (2003); Korf (2010); Wilkins and Strydom (1978).
For statutes it holds only a secondary legislative timeline (the project archive (see Source Archives)). Its contribution to this debate is therefore limited to the demographic denominator unless commission and electoral sources are acquired.
Strength of key claims¶
| Basis | Share of key claims |
|---|---|
| Systematic data | 10% |
| Partial data | 56% |
| Illustrative or anecdotal | 19% |
| Asserted | 15% |
Comparative figures for all debates are on the Literature overview.
Best-evidenced works¶
Mabin (1992), Evans (1997), Posel (1990b), Christopher (1994b), White (1990), O'Meara (1978)
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 |
|---|---|
| Abel (2014a) | Follows Posel: the Bantustan system tried to serve both aims by exploiting labour migration while keeping Africans separated and politically weak. |
| Abel (2014b) | Adopts Posel's framing of tension between political supremacy and the economy's need for African labour, and treats the Bantustan system as a compromise between the two. |
| Author not named in text (MSc candidate (2012) | Adds wartime and postwar bread rationing and the white bread promise to explanations that stress farmers and the Afrikaner working class (O'Meara). The evidence is acknowledged to be largely anecdotal. |
| Carter (1958) | Argues ideal apartheid works as a rationalisation and long-range goal. In practice Nationalist policy is mainly restrictive, protecting white farmers and workers, with little Reserve development behind it. |
| Carter (1958) | Both accept white supremacy. The United Party accepts partial economic integration and limited union recognition, while the Nationalists ban recognition and create state-controlled dispute machinery. |
| Charney (1984) | Rejects both as descriptive rather than explanatory. Argues the split's roots lie in class realignment within Afrikanerdom. |
| Christopher (1994b) | Stresses continuity from the colonial and Union segregationist eras. Apartheid differed mainly in the scale and ruthlessness of state intervention, and in its ideological codification. |
| Christopher (1994b) | Notes that one author called the 'Grand Design' a myth and that the National Party was ambivalent at first. It argues that Verwoerd's partition policy gave direction by the 1960s and 1970s, so the work treats state partition as the official aim by then. |
| Cobley (2007) | Reports Posel's view that the welfarist vision was 'an integral part of the genealogy of apartheid' and the editors' view that the 1940s offered several open futures. Cobley relays both without adjudicating. |
| Cobley (2007) | Cites Marks's suggestion that a failure of leadership among liberal-minded whites opened the road to the apartheid state. Cobley notes a wistful 'if only' undertone in the volume. |
| Cobley (2007) | Reports Nattrass's argument that the motives were straightforward: the need for expanded production, efficiency and minimal disruption in industry. |
| Cornell (1960) | Stresses both: a 'more or less definite policy of racial segregation' from 1909, developed after 1948 into deliberate apartheid. Many measures (pass laws, land acts, colour bars) are traced to pre-1948 statutes. |
| Davies et al. (1976) | Disputes the common assumption that the UP was liberalising. Its policy was contradictory, and wartime relaxations 'were born of its inability to act decisively'. |
| Evans (1997) | Evans rejects a neat rupture: central apartheid themes were already present in the 1930s-40s and the policy emerged unevenly through the 1950s, but 1948 was necessary to reorganising the state. He suggests 1953 had a stronger claim as a turning point than 1948 and gives weight to Verwoerd's managerial leadership. |
| Evans (1997) | Accepts Posel's evidence of inconsistency and partial success, but argues she underplays the deliberate planning, centralising and authoritarian managerial faith of state cadres. |
| Evans (1997) | Argues apartheid was installed largely through everyday civil administration, not military means; coercion was 'leached' into rent collection, space, transport and documentation, and complicity of liberals and chiefs qualifies the pure repression-and-resistance model (against Muthien and Lacey). |
| Freund (2012) | Stresses continuity in parastatals, social policy and segregationist planning from the Smuts years. The change lies in the harsher language, separate development and Afrikaner staffing. |
| Giliomee (2003) | Rejects the view of Keppel-Jones and Davenport and Saunders that apartheid was simply old segregation relabelled. Sees a break driven by party competition and the war, but within mainstream Western racial thought and social engineering. |
| Giliomee (2003) | Rejects the German/Nazi linkage as 'tenuous at best'. The Broederbond and northern intellectuals contributed little, and the Cape NP and DRC missiology were the main sources. |
| Giliomee (2003) | Gives priority to ideology, the survival of the Afrikaner volk, and justification of political domination over the cheap-labour-supply argument. He concedes that the Sauer Report focused on eliminating 'surplus' labour and channelling labour to mines, farms and industry, and that it exposed an unresolved clash between migrant labour and settled labour. |
| Giliomee (2003) | Stresses the 'idealistic' trusteeship and liberal-nationalist variant (Gerdener, van Wyk Louw), which he credits with the idea's resilience, while acknowledging that it justified denial of black rights. |
| Korf (2010) | Rejects the Booyens and Thom hagiographies and the 'dour dominee' image; it presents a humanised, warts-and-all and contradictory Malan. |
| Korf (2010) | Agrees with Posel that implementation under Malan was haphazard. The policy largely carried over the 1930s 'anti-measures' (marriage, immorality, group areas, Coloured franchise), while its developmental promise was never delivered. |
| Korf (2010) | Argues the apparent unity was a public-relations success. In reality there were internecine power struggles by region and generation (Cape versus north, Malan versus the Young Turks and the OB). |
| Korf (2010) | For Malan there was no difference between segregation and apartheid. The word marked a rhetorical and 'positive' shift from 1943, with the core continuous with Hertzog's segregationism. |
| Korf (2010) | Malan's racial policy was fluid and subordinate to Afrikaner and poor-white interests. Coloureds and Indians moved between 'ally' and 'threat', with a hardening after the Carnegie Commission and 1938. |
| Legassick (1974) | Stresses continuity and the application of the cheap forced labour system to secondary industry. Partly accepts Wolpe's view that apartheid is specific to secondary industrialization and reserve decay, but criticizes his neglect of changes in capitalist production. |
| Legassick (1974) | Differences were real but modest. Both favoured differentiation, regulation of African movement and European paramountcy. |
| Lundahl (1982a) | Argues that segregation's root cause may lie in political security fears of a white minority rather than economic rationality. The Bantustan policy preserves white political dominance, though it also has a secondary economic role in cheapening labour. |
| Lundahl (1982b) | Argues the root cause of segregation is white minority security rather than economic rationality, though the pass system also serves an economic role in lowering the supply price of labour. |
| Mabin (1992) | Stresses substantial continuity, with Smuts-era agencies (Land Tenure Advisory Board, Natural Resources Development Council) as precursors. The Herenigde Nasionale Party supplied cohesion, comprehensiveness and retroactivity. |
| Maylam (1995) | Rejects the liberal view of a sharp National Party break from a liberal United Party. Stresses long-term continuity, with the Group Areas Act as culmination of pre-1948 trends, while acknowledging apartheid's greater rigidity and central-state intervention. |
| Nielsen (1922) | Rejects innate inferiority; attributes lack of achievement to conservatism, isolation and absence of opportunity and education. |
| Nielsen (1922) | Rejects parallelism as plan-less drift and champions territorial separation as the only workable policy, endorsing the Natives Land Act and Native Affairs Act. |
| Nielsen (1922) | Combines claims of equal capacity with an argument that instinctive racial antipathy makes shared citizenship impossible, which justifies separate development. |
| Nielsen (1922) | Argues alarm over miscegenation is exaggerated and coloured people are not inherently inferior, but discourages mixing and opposes penal legislation. |
| O'Meara (1977) | Rejects both the 'cancerous secret government' and the 'service organization' views as inadequate. It is a policy-making body of petty bourgeois militants, influential but not running the state. |
| O'Meara (1977) | Argues that culture-based explanations take as given what needs explaining, and that Afrikaner nationalism must be read through class forces and class struggle. |
| O'Meara (1977) | Criticises Hexham and Moodie for staying at the level of ideas and overstating Calvinist/Gereformeerde influence. |
| O'Meara (1977) | The north was petty bourgeois and allied with poorer farmers and workers, while the Cape petty bourgeoisie was allied with wealthy commercial agriculture and finance capital (Sanlam, Santam). |
| O'Meara (1978) | Rejects conspiracy accounts (Hepple) and the presumed inherent appeal of Afrikaner nationalism; offers a class-based explanation focused on the labour process, the position of Afrikaners in industry, and petty-bourgeois organisers. |
| Posel (1987) | Rejects it: Afrikaner capital's 'practical' blueprint was neither hegemonic nor uncontested, and the purist 'total segregation' faction was larger and more powerful than O'Meara allowed. |
| Posel (1987) | Rejects it: the Sauer Report was contradictory and could not serve as a simple recipe, and 1950s policy was partly ad hoc reaction to immediate problems. |
| Posel (1990b) | Rejects the grand-design paradigm; apartheid evolved through phases, with a second phase after c.1960 produced by realignments within Afrikanerdom and contested within and beyond the state. |
| Posel (2011) | Rejects the monolithic reading; argues apartheid evolved through phases, was never implemented according to plan, and was marked by internal contradiction and 'radical provisionality'. |
| Posel (2011) | Argues the pragmatic version, backed by powerful Afrikaner capitalist interests, prevailed after 1948, aiming at cheap black labour under tight state control; only a small minority favoured economic sacrifice. |
| Posel (2011) | Locates apartheid within global high-modernist planning and welfarist statecraft, but stresses a state ill-equipped for its ambitions ('dinosaur with a massive body and small brain'). |
| Robinson (1994) | Endorses the newer consensus (Posel, Dubow, Bonner et al.) that apartheid was improvised, ineffective in parts and linked to modernising state projects, but argues these accounts still keep dualistic subjects. |
| Robinson (1994) | Argues, following Hyslop, that housing and township administration had an element of 'normality', without detaching them from the project of domination. |
| Verhoef (2009) | Rejects O'Meara's structural class analysis as ideologically driven and not based on company archives; stresses savings mobilisation, community self-help and business principles. |
| Verhoef (2009) | Argues that the Afrikaner path relied on the people's own savings and risk-bearing, not state support, and contrasts it favourably with statutory BEE transfers. |
| White (1990) | Presents Smuts as pragmatic and evolutionary (accepting African urbanisation, the Fagan Commission, 'uplift'). He relied on immigration to ease white fears, but the policy was vulnerable to National Party attack and was politically unacceptable to most white voters. |
| Wilkins and Strydom (1978) | Strongly asserts control: 'The South African Government today is the Broederbond and the Broederbond is the Government', while also quoting Chairman Viljoen that the Bond works through members rather than as an executive body. |
| Wilkins and Strydom (1978) | Treats the Bond as decisive, via the oxwagon trek, unification of Afrikaners and takeover of the Mineworkers' Union; concedes most historians do not give its role high priority. |
| Wilkins and Strydom (1978) | Presents Afrikaner economic advancement (Volkskas, Sanlam, Reddingsdaadbond, 'Afrikanerising the economy') as a core Bond aim, with Afrikaner business as the 'senior partner'. |
| Wittenberg (2003b) | The paper stresses that the fragmented structures required ever more coordinating layers and generated corruption, expense and unintended interdependence (commuting, migrant dependence). Decentralisation was fundamentally aimed at coercive control of Africans, and equitable partition was never politically feasible. |