Delineating the nation: South African censuses 1865–2007¶

A.J. Christopher (2009). Delineating the nation: South African censuses 1865–2007. Political Geography 28. Elsevier.


Type	journal article
Scope	Wider apartheid literature
Method	archival narrative
Period studied	1865–2007
Regions	South Africa, Cape Colony, Natal, Orange Free State, Transvaal, Bantustans (Transkei, Bophuthatswana, Venda, Ciskei), South West Africa
Themes	methods data sources, politics ideology elections, bantustans removals influx, demography urbanisation, origins segregation, transition post apartheid
Citation key	`christopher2009`

Argument¶

Censuses do not merely count a population but delineate the nation in statistical terms, and in South Africa each change of regime redefined whom to enumerate, what to ask and how to publish. The result is a set of nearly forty censuses with contradictory, often exclusionary images of the population, and little of the colonial-to-postcolonial continuity found elsewhere. The author works chronologically through the census reports, linking the degree of inclusion to access to the franchise and to the state's political agenda. Apartheid-era censuses, especially 1970 and 1980, manipulated enumeration to support partition and obscure White numerical inferiority, while the post-1994 censuses restored inclusiveness but retained apartheid racial categories.

Evidence and method¶

Published census reports and related official documents for the Cape, Natal, Orange Free State, Transvaal, Union, RSA, South West Africa, Southern Rhodesia, the protectorates and the bantustans (1865–2007), plus Statistics South Africa methodological reports and secondary literature on census politics in comparative contexts (Soviet, Indian, US, Israeli).

Place in the debates¶

Continuity between colonial and post-colonial censuses (Anderson and others) and census as nation-building (see Census, race classification and demographic data). Argues that South Africa does not fit the pattern of colonial-to-post-colonial continuity. White minority rule from 1910 to 1994 broke the link, which was restored only nominally after 1994.
Census as neutral count versus instrument of state classification (Scott, Kertzer & Arel) (see Census, race classification and demographic data). Sides with the constructivist view that the state's categories shape identity. The census projected each regime's political agenda, and race labels persisted across regimes.

Findings¶

In 1980 the excision of the independent bantustans and a new citizenship question produced absurd results: Africans were a minority of 'South African citizens'. The author calls this the high point of census manipulation under apartheid. (p. 6)

only 21.6 per cent of the African population enumerated was returned as being citizens of the Republic of South Africa. Thus numerically there were more White (4.2 million) than African (3.7 million) South African citizens.
By 1991 the attempt to exclude self-governing homeland citizens was abandoned, and almost all Africans were counted as South African citizens, showing the failure of exclusion policy. (p. 6)

93.9 per cent of Africans enumerated were returned as South African citizens in 1991, an indication that government policies of exclusion had failed.
The 1970 census was restructured around partition: African areas were separated statistically as homelands and Africans classified into Bantu national units. (p. 6)

the African rural areas and urban townships, assigned to each of the nine ‘Bantu homelands’, were separated statistically from the rump South African state, which was indicated as the ‘White Areas’
The three bantustans that had opted for independence held a substantial share of the 1970 population, which the 1980 census excluded. (p. 6)

The three excluded states had contained 13.5 per cent of the total population in 1970
Census inclusion was tied to the franchise. White-only enumerations were used in 1918, 1926, 1931 and 1941 for the delimitation of parliamentary seats, which was based on the White population alone. (p. 4)

the quinquennial delimitation of constituencies in the House of Assembly was dependent upon the conduct of a population census for the apportionment of seats between the provinces.
From 1921 the census was placed on a racial basis, with separate questionnaires and reports by race group, and attention concentrated on the European population. (p. 4)

The census was placed on a racial basis with each race group asked different questions on separate questionnaires
The 1951 census underpinned apartheid classification by supplying the basis for the Population Registration Act register. (p. 5)

The 1951 census formed the basis of the scheme devised under the Population Registration Act of 1950 and the census director was entrusted with the compilation of the register
Despite social engineering, White demographic decline could not be hidden: the White share of the population fell between 1970 and 1991. (p. 6)

the White population only constituted 16.8 per cent of the population, compared with 17.5 per cent for the whole of South Africa in 1970.
Post-1994 censuses kept the apartheid four-fold race classification so that discrimination and redress policies (employment equity, BEE) could be monitored. (p. 7)

to continue to use this classiﬁcation wherever possible, since it clearly indicates the effects of discrimination of the past, and permits the monitoring of policies to alleviate discrimination
Boer republican censuses rested on racial exclusion. In the South African Republic only Whites were counted, and only a third of adult White males could vote in 1890 owing to Witwatersrand immigration. (p. 3)

As a result only a third of adult White males in the South African Republic had the right to vote in 1890.

Evidence ledger¶

Sources used¶

Source	Kind	Period	How used
Census of the Colony of the Cape of Good Hope 1865 (G20-'66) and later Cape censuses 1875, 1891 (G6-'92), 1904 (G19-1905)	population census	1865-1904	Examines questionnaire content, race and religion categories and their changing number of race groups to show the inclusive imperial approach in the Cape.
Census of the colony of Natal 1891 and 1904	population census	1891, 1904	Shows Natal's restricted 1891 enumeration (Africans and indentured Indians taken from other departments' figures) and the distinctive racial classifications.
Report on the British Bechuanaland census 1891	population census	1891	Cited as following Cape and imperial classification practice and supplying Table 1 figures.
Orange Free State census 1880 and 1890 (Verslag van den census 1890)	population census	1880, 1890	Used to show a republican census concentrating on Whites and citizens, including recording of men qualified for commando.
South African Republic (Transvaal) census 1890 (Verslag van de Volkstelling 1890)	population census	1890	Shows a Whites-only count, with birthplace and franchise data on the Witwatersrand, treating the White citizenry as the nation.
Orange River Colony census 1904 and Transvaal Colony and Swaziland census 1904	population census	1904	Shows the post-war censuses following imperial guidelines, enumerating all groups and recording religion but not language.
Law compilation: Jeppe & Kotze, De locale wetten der Zuid Afrikaansche Republiek 1849–1885	legislation gazette	1849-1885	Source for the Republic's legal rule that anyone whose great-great grandparents were not White could not be classified as White.
Census of the Union of South Africa 1911, report (UG32-1912)	population census	1911	Establishes the first Union census, with a full enumeration, a simplified three-fold race classification and no language question.
Census of the European or White races 1918 (UG1-'20), and 1926 and 1936 censuses (Parts on nationality, languages)	population census	1918-1936	Supplies bilingualism and nationality percentages for Whites (42% to 64% bilingual; 52% British and 46% South African in 1926; 98% South African in 1936).
Third Census of the Union 1921: Report (UG37-'24) and Part VIII, Non-European races (UG40-'24)	population census	1921	Shows racially separated questionnaires and reports in 1921, the new Indian group, and nationality responses by group.
Union censuses 1936, 1946, 1951 and 1960 nationality volumes (UG24-'42, UG34-'54, 1960 Coloured and Asian families)	population census	1936-1960	Documents how the Indian nationality question changed (82% Indian (South African) in 1936, national origin in 1946, omitted in 1960).
Population Census 1951, Vol. I Geographical distribution (UG42-'55)	population census	1951	Used to show the 1951 census forming the basis of the population register under the Population Registration Act of 1950.
Population Registration Act of 1950; Nationality and Flag Act of 1927; 1956 constitutional amendment; South Africa Act 1909	legislation gazette	1909-1956	Legislative context for racial classification, nationality and the franchise that shaped census design.
Southern Rhodesia, Basutoland and South West Africa censuses (1921, 1926, 1956, 1960, 1970, 1981); Namibia 1991 census	population census	1921-1991	Used to show regional coordination or divergence of census language questions and racial classifications with South Africa's.
Republic of South Africa population censuses 1970, 1980, 1985, 1991 (Reports 02-05-01, 02-80-01, 02-80-12, 02-85-01, 03-01-26, 03-01-22)	population census	1970-1991	Supplies the data on national units, territorial exclusions and citizenship (21.6% African citizens in 1980, 93.9% in 1991) used to argue census manipulation.
Independent bantustan censuses: Transkei 1985, Bophuthatswana 1985 and 1991, Ciskei 1991, Venda 1991	population census	1980-1991	Provides the citizenship and titular-language results (Table 4; 76.0% Bophuthatswana citizens, 98.3% Venda citizens).
Census 1996 and 2001 methodology and specification reports (Reports 03-01-17, 03-02-02, 03-02-43, 03-02-04) and 2007 Community Survey questionnaire	population census	1996-2007	Documents post-apartheid retention of four-fold race classification, the 'other' category, imputation and population-group definitions.
Chinese Association of South Africa v Minister of Labour (High Court, Transvaal Provincial Division, 2008)	legislation gazette	2008	Cited for the legal recognition of the Chinese community as 'Coloured' for equity programmes.
UN Special Rapporteur report, Mission to South Africa (2005)	commission or official report	2005	Quoted for the claim that official statistics do not reflect indigenous peoples and still follow apartheid typologies.

Inference¶

Quantification: Descriptive tables.

Systematic comparative reading of successive census reports, questionnaires and published tables across regimes, supported by summary tables of totals, White population, race groups and territorial exclusions, and interpreted through literature on census, identity and the state.

Key claims¶

Claim	Basis	Strength
The racial inclusiveness of enumeration tracked access to the franchise and the political priorities of each regime. (p. 108)	Comparison of census coverage and questionnaires across the Cape, Natal, republics, Union and apartheid-era censuses (Tables 1-3).	Partial data
The Union government abandoned uniform enumeration after 1911, with White-only censuses in 1918, 1926, 1931 and 1941 and race-specific questionnaires in 1921. (p. 104)	Union census reports and Table 2 listing Whites-only counts and non-uniform tables.	Systematic data
White South African identity shifted from British to South African nationality between 1921 and 1936. (p. 105)	Nationality tabulations in the 1921, 1926 and 1936 Union census reports.	Systematic data
The 1980 census, which excluded the independent bantustans and asked about citizenship, produced a count in which Whites outnumbered African South African citizens, which the author calls the high point of apartheid-era census manipulation. (p. 106)	1980 census social characteristics report (Report 02-80-12) and 1980 geographical report.	Systematic data
The exclusion strategy failed: by 1991 almost all Africans enumerated returned as South African citizens and the White share continued to decline. (p. 106)	1991 census social characteristics report and population totals in Table 3.	Systematic data
Post-apartheid censuses restored full inclusiveness yet retained the four-fold apartheid race classification, justified for monitoring discrimination. (p. 107)	Statistics South Africa reports for 1996 and 2001, the 2007 Community Survey questionnaire and equity legislation.	Partial data
Unlike other post-colonial states, South Africa lacked continuity of census practice between colonial and post-colonial states. (p. 108)	Qualitative comparison of the whole sequence of South African census reports with literature on other states.	Illustrative or anecdotal

The author's caveats¶

The author notes that race is an unscientific category and that the racial terminology and group definitions varied over time. He also notes that boundaries and the populations enumerated changed, which undermines comparability between censuses, and that nationality and citizenship responses may be unreliable owing to the desire for conformity.

Gaps in the evidence¶

The argument reads census reports as state projections but does not use archival records of census directors, enumerator instructions or parliamentary debates to show how and why decisions were made. It lacks evidence on how those enumerated responded or were classified, and on undercount, which could test the claimed effects of census manipulation. Such archives (Bureau of Census files, Hansard, Population Registration records) existed and are accessible today.

Replicability: Data public and well described.