The Imposition of Pass Laws on the African Population in South Africa 1916-1984¶
Michael Savage (1986). The Imposition of Pass Laws on the African Population in South Africa 1916-1984. African Affairs, Vol. 85, No. 339. Oxford University Press on behalf of The Royal African Society.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | descriptive quantitative |
| Period studied | 1916–1984 |
| Regions | South Africa, Bantustans, Western Cape, Witwatersrand, Durban, Pretoria |
| Themes | bantustans removals influx, labour markets wages, demography urbanisation, origins segregation, methods data sources, politics ideology elections |
| Citation key | savage1986 |
Argument¶
Savage argues that pass laws and influx control have balanced an 'exclusionary' white need for political security against an 'inclusionary' need for cheap labour. Until about 1950 the thrust was to channel labour into white agriculture and industry. After 1950 it became overtly exclusionary, relocating Africans from white areas into the Bantustans. He quantifies this with prosecution series, tax and trespass prosecutions, and census-based population distribution. He shows that the fall in prosecutions after 1968 reflects new enforcement methods (Aid Centres, housing and contract-labour controls, deportation, Administration Boards), not relaxation. The system creates a contradictory dynamic: poverty is exported to the Bantustans, which pushes people back into the cities.
Evidence and method¶
Annual Reports of the Commissioner of South African Police, Official Year Books, House of Assembly Debates, censuses (corrected by Simkins), Surplus Peoples Project studies, commission reports (Stallard, Fagan, Riekert, Viljoen, President's Council), SAIRR Surveys, press reports, and Lange's income calculations.
Place in the debates¶
- Whether influx control is being relaxed or abolished in the mid-1980s (see Influx control, pass laws and African urbanisation). Rejects the view of imminent abolition. Arrests rise whenever reform is announced, and control is shifting to housing, contract labour and deportation.
- Reliability of official statistics on African population and pass law arrests (see Data reliability and econometric method). Treats the figures as flawed orders of magnitude, noting census undercounts and official manipulation of figures, but uses them to show patterns.
- Pass laws as inclusionary labour-supply tool vs exclusionary security tool (see Influx control, pass laws and African urbanisation). Holds both functions operate in tension, with the balance shifting from inclusion (to c.1950) to exclusion (after 1950).
Findings¶
-
Cumulative scale of pass law prosecutions 1916-1984. (p. 2)
During the period 1916 to 1984 over 17,745,000 Africans have been arrested or prosecuted under a battery of pass laws and influx control regulations in South Africa.
-
The share of Africans in the Bantustans rose sharply after 1950. (p. 12)
rose rapidly to 47 4 per cent in 1970 and to 52 7 per cent in 1980.
-
Prosecutions peaked in the late 1960s, tracking removals and migrant labour regulations. (p. 18)
pass law prosecutions increased rapidly, climbing from 214 thousand a year to a peak in 1967/68 of 694 thousand
-
Aid Centres explain only part of the post-1968 decline in prosecutions, which does not signal relaxed control. (p. 19)
it can be estimated that since 1972 the centres have decreased pass law prosecutions by 96,300 cases annually.
-
Control shifted from street arrests to housing and employment controls. (p. 22)
emphasises the changing nature of these controls and the way in which their enforcement is being taken off the streets and into housing and factories.
-
Post-1948 policy followed the Stallard doctrine, with Verwoerd entrenching migrant labour and meeting farmers' labour demands. (p. 15)
moved toward entrenching the migrant labour system, to addressing farmers' requirements for cheap labour and to evolving an over- all policy concerning black urban workers
-
Official policy stated that Africans were only temporary residents of white areas. (p. 16)
It is accepted government policy that the Bantu are only temporarily resident in the European areas of the Republic, for as long as they offer their labour there.
-
Urban African growth slowed and turned to net out-migration after 1960 as controls tightened. (p. 11)
In the three periods 1951-60, 196s70 and 1970-80, urban growth rates show this declining rate to be 4 5 per cent, 3 9 per cent and 2 3 per cent.
-
Stringent controls drive illegal entry to cities because the income gains are so large. (p. 26)
a person living in Lebowa and 'illegally' working in Johannesburg for nine months and then spending three months in jail would improve his, or her, income by 255 per cent
-
Homeland independence creates 'new foreigners' liable to deportation without court recourse. (p. 9)
South Africa thus has created from its citizens several millions of persons who have become 'new foreigners' liable to instant deportation and endorsement out of 'white' areas.
-
Housing shortages act as an alternative form of influx control. (p. 20)
The vast pressure on housing is preventing people from entering the cities other than to work and is trapping those who do not have work into the growing pool of people unemployed
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Annual Report of the Commissioner of South African Police (1917-1920, 1925-33, 1935-39, 1941-79/80) | official statistical series | 1916-1980 | Main series of pass law arrests/prosecutions 1916-1980, used to build Tables 1-3 and Figure 1 and to periodise enforcement. |
| Official Year Book of the Union of South Africa (Year Books 1, 5, 16, 22) | official statistical series | 1916, 1921, 1934, 1940 | Fills gaps in the police series for years in which the police reports lack pass law prosecution figures. |
| House of Assembly Debates: annual answers by the Minister of Law and Order and the Minister of Cooperation and Development | parliamentary debates | after 1980 (also 1955, 1962 answers on endorsements) | Supplies pass law prosecution figures after 1980, when police reports ceased to give them, and indicates summary endorsements. |
| Union Statistics for 50 years, Table F-4 col. 38 'Bantu Control' (Bureau of Census and Statistics, 1960) | official statistical series | 1921-1958 | Considered as an alternative conviction series but rejected as unclear in coverage and incomplete. |
| Population censuses, Department of Statistics (1911-1980) and Unit for Future Research (Stellenbosch) 1980 estimate | population census | 1911-1980 | Provides African total and urban population used to compute the share of the population prosecuted under pass laws (Table 3). |
| Simkins, Four Essays on the Past, Present and Possible Future of the Distribution of the Black Population in South Africa (SALDRU, 1983) | secondary literature compilation | 1950-1980 | Corrected census data on population distribution used to compute Table 4 and migration flows by decade. |
| Annual Reports of the Commissioner of South African Police: tax and trespass prosecutions | official statistical series | 1920-1979/80 | Collated to show African prosecutions under tax and trespass laws (Table 2). |
| Minutes of 1974 meeting in Pretoria between police and Bantu Affairs Commissioners, reported in Sunday Times, 13 July 1980 | newspapers periodicals | 1974/1980 | Evidence that officials manipulated pass law offence statistics, used to caution about data reliability. |
| Surplus Peoples Project studies and Platzky and Walker, The Surplus People (1985) | secondary literature compilation | since 1960 | Figures on 3.5 million removals and categories of removed people, supplementing pass law figures. |
| Horrell, Survey of Race Relations (SAIRR) 1955-6, 1956-7, 1978, 1982 | secondary literature compilation | 1955-1982 | Data on persons turned away, endorsed out, declared idle and undesirable, and the Viljoen Committee housing backlog. |
| Report of the Committee for Constitutional Affairs of the President's Council, 3/1985, An Urbanisation Strategy for South Africa | commission or official report | 1985 | Recommendation to abolish influx control, and its admission that arrests rose after 1978. |
| Report of the Native Laws Commission 1946-48 (Fagan Commission), U.G.28, 1948 | commission or official report | 1946-48 | Used to show the official recognition of permanent urbanisation before 1948. |
| Report of the Commission of Inquiry into Legislation Affecting the Utilization of Manpower (Riekert Commission), R.P. 32/1979 | commission or official report | 1979 | Used for policy shifts toward tying urban entry to housing and jobs and repatriation grounds. |
| Secretary for Bantu Administration and Development, General Circular No. 25, 1967 | commission or official report | 1967 | Shows official policy that Africans are temporary residents of white areas. |
| Legislation: Natives (Urban Areas) Act 1923, Native Laws Amendment Act 1937, Urban Areas Act amendment 1952, Admission of Persons to the Republic Regulation Act 1972, Bill 113 of 1982 | legislation gazette | 1923-1982 | The legislative phases of influx control are traced and used to interpret prosecution trends. |
| Smit and Booysen, Urbanisation in the Homelands (Univ. of Pretoria, 1977) | secondary literature compilation | 1960-1975 | Figures for homeland urban population growth are calculated from it. |
| Newspaper and periodical reports (Cape Times, Rand Daily Mail, Die Burger, Financial Mail, New York Times Magazine, The Star, Frontline) | newspapers periodicals | 1979-1985 | Illustrative evidence on policy statements, influx control politics, Lange's income calculations and conditions in the homelands. |
| Unspecified official arrest data for 11 major urban centres 1974-84, including Administration Board officials and women arrested | official statistical series | 1974-1984 | Used for the share of arrests in cities, by Administration Boards, by women, and city-level patterns. |
| Aid Centre information, annually supplied (1972-81) | official statistical series | 1972-1981 | Combined with prosecution figures to estimate that Aid Centres reduced prosecutions by 96,300 per year. |
New data¶
The author compiled a 1916-1984 series of pass law prosecutions from Commissioner of Police reports, Year Books and parliamentary replies (Table 1, Figure 1), plus collated tax and trespass prosecution totals (Table 2) and prosecution-to-population ratios (Table 3).
Inference¶
Quantification: Descriptive tables.
Descriptive time-series analysis of prosecution counts, set against census-based population distribution (via Simkins) and a periodisation of legislation and policy. Correlation of prosecution peaks with legislative changes and relocations, supported by commission reports, parliamentary replies and press reports; an ad hoc counterfactual estimate for the Aid Centres.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Over 17.7 million Africans were prosecuted under pass laws and influx control regulations from 1916 to 1984, with prosecutions peaking in 1967-68. (p. 185) | Police Commissioner annual reports, Year Books and Hansard answers compiled in Table 1; the author notes gaps and unreliability. | Systematic data |
| The African population has shifted markedly into the Bantustans since 1950, from about 40 to 53 per cent, while the share in rural white areas fell. (p. 191) | Simkins's corrected census distribution figures (Table 4). | Systematic data |
| Prosecution trends track legislative and administrative changes, notably the post-1948 Nationalist tightening and the mid-1960s measures. (p. 197) | Prosecution series compared with a chronology of legislation, regulations and commission reports. | Partial data |
| The post-1968 decline in prosecutions does not mean a relaxation of influx control; Aid Centres explain only part of it, and housing, contract labour and commuter settlements work as alternative controls. (p. 198) | Aid Centre figures and a simple counterfactual estimate; Viljoen Committee and SAIRR housing data; press and secondary accounts of homeland settlements. | Partial data |
| Prosecution figures understate the numbers affected, because of removals, endorsements out, deportations and tax and trespass prosecutions. (p. 187) | Platzky and Walker/SPP removal totals, parliamentary answers on endorsements, Table 2 tax and trespass figures, 1981 Cape Town deportations. | Partial data |
| Enforcement shifted from the police to Administration Board officials, and women were a growing share of those arrested, notably in the Cape Peninsula. (p. 204) | Urban arrest figures for 11 major centres 1974-84, including breakdowns by enforcer and sex. | Partial data |
| Pass law policy shifted from a labour-channelling phase to an exclusionary phase from 1950 aimed at containing Africans in the Bantustans. (p. 182) | Legislative and commission chronology (Stallard, Fagan, 1952 Act, 1967 circular). | Illustrative or anecdotal |
The author's caveats¶
Savage stresses that South African statistics are flawed orders of magnitude. The census undercounted Africans, official arrest and prosecution figures conflict, and the 1974 minutes show deliberate manipulation. Conviction series are unavailable, and the police series has gaps and ends in 1980. Administration Board arrests are not clearly included, and endorsements and admissions of guilt are not captured. Independent homelands have been excluded from official statistics since 1976.
Gaps in the evidence¶
The argument that the decline in prosecutions reflects a shift to housing- and employer-based control is asserted from a few indicators and an ad hoc Aid Centre counterfactual, with no systematic data on workplace enforcement, employer fines or endorsements. Better evidence would be archival Administration Board and Bantu Affairs Department records, court records and labour bureau registrations, which existed then but were not accessible. The census-based migration inferences depend wholly on Simkins's corrections, and no independent measure of illegal urban residence exists.
Replicability: Data described but not public.