South Africa¶
Ulf Engel (1999). South Africa. Elections in Africa: A Data Handbook (eds. Dieter Nohlen, Bernard Thibaut, Michael Krennerich). Oxford University Press.
| Type | book chapter |
| Scope | Wider apartheid literature |
| Method | descriptive quantitative |
| Period studied | 1910–1998 |
| Regions | South Africa, Union of South Africa, Cape Province, Natal, Transvaal, Orange Free State, Homelands (TBVC states) |
| Themes | politics ideology elections, origins segregation, methods data sources, transition post apartheid |
| Citation key | engel1999 |
Argument¶
This data-handbook chapter gives a historical and institutional overview of South African elections from 1910 to 1994, followed by tabulated results. It shows how the franchise was extended to all Europeans while Africans, Coloureds and Indians were progressively excluded, and how the National Party, backed by rural over-representation, held power from 1948 to 1994. It also documents the tricameral Parliament of 1983, the low turnout in the Coloured and Indian houses, the rise of the Conservative Party on the white right, and the 1994 election. It is a reference compilation, not an interpretive argument, and it flags weaknesses in the official electoral statistics.
Evidence and method¶
Official Yearbook of the Republic of South Africa (1989/90), Heard (1974) and Noller (1978) for election figures, plus Government Gazettes, referendum results from The Times and S.A. Barometer, the Independent Electoral Commission's 1994 report, statutes and proclamations, and census figures for the electoral body.
Place in the debates¶
- Reliability of South African electoral statistics (see White electoral politics, delimitation and party systems). The author says documentation is weaker than expected and the official list sometimes contains obviously wrong figures, so competing sources (Heard, Noller) are cross-checked but cannot always be reconciled.
- Causes of the end of apartheid (see Why apartheid ended: reform, transition and elite bargaining). Describes the removal of apartheid as an inter-linked process of internal revolt, failed controlled reform from above and international pressure, without privileging any single factor.
Findings¶
-
The white right's rise is visible in the 1987 and 1989 House of Assembly results, where the Conservative Party took a growing share of seats and votes. (p. 10)
CP 547,559 26.8 680,131 31.5
-
The NP won power in 1948 with a minority of votes, helped by rural over-representation in constituency delimitation. (p. 1)
Supported by inequalities in the delimitation of constituencies which favored rural over urban ones, in 1948 the NP managed to gain a majority of seats on a minority of votes
-
Non-white franchise was gradually curtailed and then abolished, with the Cape Coloured common roll rights removed by 1968. (p. 3)
Representation of the Cape Coloureds was finally abolished in 1968 (Separate Representation of Voters Amendment Act, No. 50 of 1968)
-
The 1983 constitution gave Coloureds and Indians separate houses with no real power, co-opting them without sharing power. (p. 3)
held no real power within the constitutional dispensation
-
The 1983 tricameral constitution opened a period of intensified protest, repression and violence. (p. 2)
South Africa entered a circle of increased political protest, repression and violence (1983–1989)
-
The 1983 referendum on the tricameral Parliament passed among white voters with a two-thirds majority. (p. 6)
Yes 850,458 52.3 1,360,223 66.0
-
The 1994 election returned an ANC majority on a large turnout. (p. 6)
ANC 12,237,665 62.6 252 63.0
-
Turnout in the 1984 Coloured House elections was very low, indicating weak legitimacy of the tricameral system. (p. 5)
1984 HoR 2,832,705 881,984 31.1 272,854 30.9 9.6
-
The transition was dominated by bilateral NP/ANC negotiations amid violence and resistance from IFP, CP and some homeland leaders. (p. 2)
The following ‘negotiated revolution’ was characterized mainly by hegemonial bilateral NP/ANC negotiations
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| The Times, 04/11/1983 (referendum results) | newspapers periodicals | 1983 | Source of the referendum result figures (text cites it for the '1978' referendum). |
| S. A. Barometer, 27/03/1992 (1992 referendum results) | newspapers periodicals | 1992 | Source of 1992 referendum result figures. |
| Written communication by RSA Government to the author (03/02/1990) | other | 1989 | Source for the composition of Parliament in 1989. |
| South African population censuses (1911, 1936, 1946, 1951, 1960, 1980, 1985, 1991) | population census | 1911–1991 | Used as population base to compute registered voters and turnout as % of population. |
| Electoral and constitutional legislation (South Africa Act 1909; Acts of 1930, 1931, 1936, 1946, 1951, 1956, 1958, 1959, 1968, 1972, 1979, 1983) | legislation gazette | 1909–1983 | Establishes the evolution of franchise extension for Europeans and restriction of non-European rights. |
| 1993–1996 constitutions, Electoral Act 1993, IEC Acts, Proclamations and Government Notices of 1993–94 | legislation gazette | 1993–1997 | Basis for describing current electoral provisions (suffrage, candidacy, PR system, seat allocation). |
New data¶
Handbook compilation of South African national election results 1910–1994 (House of Assembly, Coloured Persons Representative Council, House of Representatives, House of Delegates, 1994 Constitutional Assembly), referendums, composition of Parliament and electorate/turnout tables, with author recalculations of party shares where official totals are inconsistent.
Inference¶
Quantification: Descriptive tables.
Reference handbook: compiles and reconciles official election returns into standardized tables, with brief historical narrative and legal description; no analytical inference beyond data cleaning and noted discrepancies.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The NP won a majority of seats in 1948 on a minority of votes, aided by rural-favouring delimitation. (p. 1) | Election return tables (Official Yearbook) and asserted delimitation effect; cites Christopher in bibliography. | Partial data |
| Franchise for non-Europeans was progressively restricted and abolished, while extended to all Europeans. (p. 2) | Chronology of named Acts (1930–1983). | Systematic data |
| In 1994 the ANC won 62.6% of valid votes and 252 of 400 seats on 86.7% turnout. (p. 7) | IEC final report (1994). | Systematic data |
| Non-white tricameral elections drew very low participation (e.g. 1984 Coloured House 30.9% of registered voters; 1989 18.1%). (p. 11) | Official Yearbook and government figures for HoR and HoD elections. | Systematic data |
| The 1994 elections were judged substantially free and fair by observers despite violence and logistical problems. (p. 2) | Observer assessments (IEC, Commonwealth, EU, UN reports listed in bibliography), not analysed in text. | Asserted |
The author's caveats¶
Author states the Official Yearbook list is sometimes obviously wrong, that Noller did not always use official figures, that differences between sources cannot be explained without reference to the underlying data, and that census figures cannot be strictly applied because of racial criteria; table footnotes flag total party votes exceeding votes cast and other inconsistencies.
Gaps in the evidence¶
The tables rely on secondary and compiled returns rather than the original Government Gazette constituency returns; a systematic reconciliation against constituency-level results would resolve the flagged discrepancies. Voter-roll and census denominators are approximate for the White population and omit the disenfranchised African majority, so participation shares say little about political exclusion. Delimitation and constituency-level data needed to test the rural-bias claim exist in archives but are not presented here.
Replicability: Data described but not public.