Urban segregation, Group Areas and the apartheid city¶
What drove urban residential segregation and the Group Areas Act (ideology, material interests, the local state, planners), and how far does it persist?
Works engaging the debate: 15
Positions¶
- Continuity from local segregation before 1948 (Maylam, Mabin)
- Material vs sanitation-syndrome or strategic explanations
- Local state autonomy vs central imposition
- Post-apartheid 'neo-apartheid city' vs real change
Evidence base¶
The debate rests on two largely separate bodies of evidence. The first, and the larger, is archival process-tracing of state and municipal records. Mabin (1990) and Mabin (1992) reconstruct the genealogy of the Group Areas Act from Land Tenure Advisory Board files (ARG 10, 12, 144), Group Areas Board (GGR) and Department of Planning (BEP) application files, the Asiatic Land Tenure Committee personal file of D.S. van der Merwe (ALTC 5), SEPC (SEC 47) and NRDC (RNH 279) correspondence, and East London and Simonstown municipal series. They add the Thornton (UG 8/1940), Wilcocks and SEPC Report No. 5 (UG 34-1944) commissions, the Sauer Report, Hansard and author interviews with T.J.D. Fair and R.J. Davies. Maylam (1990) reads Durban Town Clerk's Files, Native Affairs Department files and Durban Native Administration Committee agenda books against 1936 and 1946 census counts and sex ratios; it reports African population rising from about 71,000 in 1936 to about 150,000 in 1949. Parnell and Mabin (1995) combines pre-Union and Transvaal local-government records (TPS/TPB), SEPC and NRDC files and planner interviews. Robinson (1994) uses Port Elizabeth Mayor's Minutes, municipal files and a handful of administrator interviews. Mabin (1986b) uses De Beers and Kimberley Central minute books, Standard Bank inspection reports, mine inspectors' reports, Cape Blue Book export series and the 1881 and 1891/1904 censuses for Kimberley. The syntheses (Maylam (1995)), (Sapire and Beall (1995)) and Crush (1984) add no data of their own and argue from these case studies.
The second body is quantitative and census-based, and it is much thinner. Christopher (1990) computes Duncan segregation indices from census enumerators' summary returns at enumeration-tract level for towns between 1911 and 1985, with Mann-Whitney and Spearman tests across regions and town sizes. Christopher (1991) aggregates the official Group Areas Proclamation Registers by period, race and province and converts them into persons per proclaimed hectare using census counts, alongside SAIRR displacement figures. On persistence, (Krugell (2014)) fits Zipf rank-size regressions to 1911 and 2011 census populations aggregated to 1911 district boundaries, and uses 1911 vehicle counts as a wealth proxy. McCarthy (2000a) and McCarthy (2000b) use nominal Gross Geographic Product by magisterial district for 1968-1994. Seekings (2010) compares 1993 household income data with NIDS wave 1 (2008) and cites Cape Town service-access data for 1996-2003.
Contested evidence¶
On continuity versus rupture, the sides read different kinds of evidence rather than the same series. Maylam (1995) and Maylam (1990) argue that segregation evolved continuously. They draw on pre-1948 municipal planning, such as Durban's 1940s zoning blueprints and the United Party-era plans for the Western Areas and District Six. Mabin (1990) and Mabin (1992) trace the Act's institutions back to the Smuts-era Land Tenure Advisory Board and the SEPC. Yet the only long quantitative series, (Christopher (1990))'s Duncan indices, shows segregation rising from 1911 but with its sharpest jump in the 1960s. Coloured medians in the Western Cape go from 64.8 in 1960 to 86.6 in 1970, and regional differences inherited from the colonial period converge only after central enforcement. The proclamation registers in Christopher (1991) likewise peak in the 1960s. Read this way, the census and register data support a discontinuity in implementation even if the policy lineage was continuous.
On motive, (Maylam (1995)) treats the 1914 Tuberculosis and 1918 Influenza commissions as evidence that the sanitation syndrome was a pretext for material interests. Parnell and Mabin (1995) reads public health and slum legislation as a primary driver that also affected poor whites, but her support is her own unpublished thesis. Mabin (1986b) attributes the Kimberley compounds to class struggle after the 1883-84 strikes. He nonetheless concedes that location segregation was driven by sanitary discourse together with capital's labour needs, and he relies on isolated employment figures (white mine employment falling from about 3,900 in 1881 to 1,210 in 1884) rather than a series. Maylam (1990) uses Durban committee papers to attribute the 1930s proclamations to police and ratepayer pressure rather than capital. Christopher (1991) infers an intent to seize valuable sites and eliminate Coloured and Indian competition from land-per-head disparities, citing 5.5 Whites per proclaimed hectare in 1970 against 44.8 Coloureds and 20.4 Asians. That inference lacks any deeds or valuation evidence.
On local autonomy, the same kind of municipal file is read in opposite directions. Maylam (1990) and Parnell and Mabin (1995) see councils as enthusiastic instigators of segregation. Mabin (1992) stresses the leverage the Board exercised over planners through National Housing Office finance and circulars, which implies central pull. On persistence, the 'neo-apartheid' and 'real change' positions rest on incommensurable units. Krugell (2014) and McCarthy (2000a) measure inter-district concentration of population and output. Seekings (2010) measures class shares and service access. None of these computes post-1994 intra-urban segregation indices, and Seekings relies on Christopher's later work and schematic maps.
Underused and missing sources¶
Several sources that would settle these disputes are cited as gaps but left unexploited. Group Areas Board deliberations, deeds registries and property valuation or sale-price data would test (Christopher (1991))'s claim about economic motive and Crush (1984)'s point that white owners profited in Mowbray. Systematic coding of Board and LTAB files across many towns would test (Mabin (1990))'s claims about planner enthusiasm and Board pressure. Municipal budgets by race, which (Maylam (1995)) says are asserted but never documented, are also missing. So are pass-law prosecution and removal series and sectoral labour-demand data for Durban, which (Maylam (1990)) lacks, and continuous mine employment, wage and compound series for Kimberley, which (Mabin (1986b)) lacks. Removal counts by town would allow the 1960s jump in Christopher (1990)'s indices to be tied to enforcement rather than to timing. For the post-apartheid question, sub-place dissimilarity indices from the 1996, 2001 and 2011 censuses are the obvious missing test of Seekings (2010). McCarthy (2000a) would need real, boundary-consistent district GGP.
Strength of key claims¶
| Basis | Share of key claims |
|---|---|
| Systematic data | 10% |
| Partial data | 54% |
| Illustrative or anecdotal | 17% |
| Asserted | 19% |
Comparative figures for all debates are on the Literature overview.
Best-evidenced works¶
Christopher (1990), Christopher (1991), Mabin (1992), Mabin (1990), Maylam (1990), Krugell (2014)
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 |
|---|---|
| Christopher (1990) | Segregation reached very high levels but total separation was unattainable; about 10% of the urban population in 1985 still lived in the 'wrong' group area. |
| Christopher (1990) | Colonial regional differences survived almost intact until the 1960s and persisted in muted form afterwards, while central legislation drove convergence. |
| Christopher (1990) | Finds no significant relationship in South Africa, in line with Taeuber and Taeuber (1965) and against Farley and Wilger (1987) for US cities. |
| Christopher (1991) | Mainly emphasises racial segregation and preserving 'White South Africa', but also infers that the government wanted to acquire valuable urban sites and permanently destroy Coloured and Asian economic competition. |
| Christopher (1991) | Sceptical: the changes were modifications that relaxed the broad sectoral design but not segregation. Group areas continue to be proclaimed and enforced, and the post-apartheid city may show only modifications of the apartheid model. |
| Crush (1984) | Rejects functionalist accounts, both racial-domination and cheap-labour variants, as reductionist and not borne out by historical scrutiny. |
| Krugell (2014) | Supports Van der Merwe and Naudé and Krugell that there are too many small towns and too few, over-large cities; the counterfactual without apartheid is treated as speculative. |
| Mabin (1986b) | Rejects Christopher, Davies and Western's view that apartheid cities are a more systematic version of the colonial city. Argues that residential segregation arose from capitalist class struggle, following Simon's call to address the political economy. |
| Mabin (1990) | Rejects the view (Kuper, Watts and Davies; Meer; Johannesburg's lawyer) that the Act was primarily about Indian property in Natal driven by Durban Council. Stresses Cape coloured segregation campaigns, Transvaal interests and planning developments. |
| Mabin (1990) | Argues the GAA was not a pure product of a grand apartheid plan; the HNP manifesto lacked any means of implementation. It built on Smuts-era institutions and planning ideas, and policy was contested (aligns with Posel, Hindson, Hyslop and Kros). |
| Mabin (1990) | Finds planners largely embraced racial zoning as a continuation of their reform agenda and were 'seduced' by group areas, while cautioning that the evidence for a model radial/buffer design guiding the LTAB is thin. |
| Mabin (1990) | Notes the 'ritual repetition' of the material explanation but says the actual influence of property, commercial and development interests on group areas policy remains unknown and needs research. |
| Mabin (1992) | Rejects the master-plan view. Argues the Act grew out of earlier segregatory measures, planning bureaucracies and pressures that were wider than the Act itself. |
| Mabin (1992) | Says Natal and Indian property restrictions were part of the story but are inadequate as the explanation. It stresses the Cape 'Coloured' segregation push and national planning thinking. |
| Mabin (1992) | Criticises these accounts for glib assumptions about the Act's background. It adds a bureaucratic and planning-professional dimension. |
| Maylam (1990) | Rejects the view that the local state is simply an extension of the central state. The Durban corporation enjoyed substantial autonomy until the late 1930s, and even afterwards it asserted its own interests. |
| Maylam (1990) | Argues for continuity. Durban's council, which was not Nationalist-controlled, was an enthusiastic proponent of segregation before and after 1948. The Nationalists added centralisation, uniformity and compulsion. |
| Maylam (1995) | Sanitation was probably a pretext or metaphor; material interests (ratepayers, merchants, white workers, capital's labour control and land needs) weighed more heavily. |
| McCarthy (2000a) | Metro areas always accelerated faster. The non-metro share was somewhat higher while apartheid decentralisation was strong, and growth shifted to metros and coasts afterwards. |
| McCarthy (2000a) | Stresses path dependence. Radical locational shifts occurred only with primary activity (mining), and the paper expects future growth to elaborate on past patterns. |
| McCarthy (2000b) | Tests this and finds metros always accelerated faster, though the non-metro share of growth varied with apartheid policy. |
| McCarthy (2000b) | Path dependence dominates, citing international experience such as China. The paper endorses 'sticking to the beaten track' with SDIs and rejects the previous government's growth-centre strategies. |
| Parnell and Mabin (1995) | Argues the debate has distorted urban research by equating urban residential segregation with the cheap-labour imperatives of territorial segregation (Maylam, Hindson). It stresses that urban slum clearance, public health and modernist planning also drove the removal of Africans. |
| Parnell and Mabin (1995) | Rejects the view that segregation simply flowed from racist national ideology. It stresses municipal councils as active gatekeepers who lobbied for extended controls, and the divergent interests of central, provincial and local government. |
| Parnell and Mabin (1995) | Following Dubow and Rich, holds that the capitalist class's power in introducing territorial segregation has been over-simplified and exaggerated. It notes that business's role in intra-urban segregation is poorly documented. |
| Robinson (1994) | Sides with Atkinson's 'moral communities' and post-colonial hybridity against her own earlier control-and-resistance framing. Concedes that administrators nonetheless enforced pass laws and surveillance. |
| Sapire and Beall (1995) | Reports a growing challenge to the urban-bias orthodoxy; cities are increasingly seen as sites of economic agglomeration, and the poor within cities lack equal access to resources. |
| Sapire and Beall (1995) | Endorses Maylam and Mabin and Parnell's critique: the literature is empirically rich but fragmented and racially compartmentalised, and it neglects integrated, comparative and 'from above' studies. |
| Sapire and Beall (1995) | Treats ethnicity as situational and historically contingent, a product of modernisation and material competition. It stresses rural and regional dimensions beyond the earlier 'retribalisation' view. |
| Sapire and Beall (1995) | Accepts Bickford-Smith's evidence that Cape Town had a special status, while noting that pockets of cross-racial contact existed elsewhere. |
| Seekings (2010) | Partly rejects it. Segregation and class inequality persist, but the neo-liberal critique ignores redistributive financing, improved services and grants, and underestimates desegregation in richer areas. |
| Seekings (2010) | Economic barriers and white racism matter, but they are insufficient. Segregation also persists among equally poor African and coloured groups, through public housing allocation to informal-settlement 'communities' and through social preference. |
| Seekings (2010) | Under apartheid caste shaped the range of class positions and class shaped the caste system. After 1994 class is increasingly decoupled from race at the top, while the lower classes remain entirely African. |