Fertility, Mortality, and Migration in the Cape Colony, 1891-1904¶
Charles Simkins; Elizabeth van Heyningen (1989). Fertility, Mortality, and Migration in the Cape Colony, 1891-1904. The International Journal of African Historical Studies, 22(1). Boston University African Studies Center.
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | mixed |
| Period studied | 1891–1904 (linked forward to c.1980) |
| Regions | Cape Colony, Cape Town, Kimberley, Port Elizabeth, Eastern Cape/Transkei |
| Themes | origins segregation, demography urbanisation, mining migrant labour, agriculture land, labour markets wages, inequality poverty unemployment, methods data sources |
| Citation key | simkins1989 |
Argument¶
The authors apply modern indirect demographic techniques to the 1891 and 1904 Cape censuses and the 1895-1903 vital registration data. They estimate fertility, mortality and migration for six social and ethnic groups. They find a fifteen-year gap in life expectancy between whites and all other groups, and infant mortality about twice as high for non-whites. They also find that the white fertility transition began around 1890, while Coloured and black fertility remained at natural levels. They tie these differentials to the economy of the mineral-revolution decade: discriminatory wages and occupations, overcrowded and segregated urban housing, migrant mining, and an agricultural crisis. They then connect the 1891-1904 magnitudes to twentieth-century trends. The result is a long-run view of South Africa's very stretched-out demographic transition.
Evidence and method¶
Cape censuses of 1891 and 1904; Cape Births and Deaths Registration reports 1895-1903; Cape Statistical Registers (wage and price medians); 1893 Labour Commission report; medical officer of health reports; later Union and Republic vital statistics and censuses; Sadie and Grobbelaar demographic projections; secondary literature on Kimberley, Herschel and Cape Town.
Place in the debates¶
- Timing and character of the demographic (fertility) transition in South Africa (see Census, race classification and demographic data). Argues the white fertility transition began c.1890, with Coloured and black fertility at natural levels. Racial inequality stretched the overall transition in a historically unprecedented way.
- Welfare and standards of living under early twentieth-century economic transformation (data-poor, controversial) (see Living standards, health and social policy). Uses demographic magnitudes as welfare indicators and finds whites were the main beneficiaries of public-health gains. Real living costs changed little across 1891-1904.
- Cape economy and the mineral revolution (Mabin on capital flight to the north; Bundy on the peasantry) (see Pre-Union colonial economy, the mineral revolution and the South African War). Broadly accepts these accounts. Sees the Cape as commercially prosperous but failing to build independent sources of wealth, and increasingly dependent on the Transvaal.
Findings¶
-
A large, uniform life-expectancy gap separated whites from all other Cape groups, with non-white infant mortality about double that of whites. (p. 18)
Notable is a fifteen year difference in life expectancy at birth between whites and all other groups among whom this magnitude varies little.
-
Net immigration accounted for a large share of white population growth in 1891-1906. (p. 10)
Net immigration can therefore be estimated at 88,236 or 39 percent of the total growth of the population.
-
White marital fertility had begun to fall around 1890, indicating the onset of a white fertility transition. (p. 12)
it does not seem unreasonable to date the onset of the white fertility transition to about 1890.
-
Urban growth among blacks and Coloureds was driven mostly by net in-migration into high-mortality towns. (p. 18)
It follows that the contribution of net immigration to the increase in urban population was 61 percent in the case of whites, 57 percent for Coloureds, and 83 percent for blacks.
-
Mortality in the principal towns for Coloured and black people was extremely high, so rural-urban migration meant greater mortality hazards. (p. 18)
Associated with this life table are an infant mortality rate of 470 per thousand and a life expectancy of 18 for males and 20 for females.
-
Kimberley's diamond compounds were deadly for black migrant workers. (p. 21)
The number of industrial accidents was abnormally high, accounting for about one-third of all deaths in Kimberley annually in the 1890s
-
The Cape farm labour force was predominantly Coloured, and farm wages were the lowest. (p. 26)
Farm labor drew much more heavily on Coloured people than on either blacks or whites, 51,551 as against 7,117 whites and 37,896 blacks
-
The 1890s agricultural crisis (rinderpest, drought, phylloxera) pushed farmers off the land and weakened the black peasantry. (p. 25)
In the Transkei, where cattle were the basic unit of wealth, it is estimated that perhaps 90 percent were destroyed.
-
Wage data show colour discrimination and occupational differentials underpinning the mortality gaps. (p. 29)
The wage data show a clear pattern of discrimination which, allied to inter-group differences in occupational distribution, demonstrate the labor market conditions behind the mortality differentials between the groups.
-
The Cape economy was commercially buoyant but dependent, and slumped after 1904. (p. 25)
But the Cape was failing to build independent sources of wealth.
-
Over the twentieth century the life-expectancy gap widened until 1946, then narrowed. (p. 34)
White life expectancy gained 19 years in the interval, while that of Coloured people gained 14 and that of black people about 10.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Census of the Cape Colony 1891, Final Report (G6-1892) | population census | 1891 | Supplies age/sex/ethnic-group distributions used as the first census in indirect estimation of fertility, mortality and migration, plus urban counts and occupations. |
| Census of the Cape Colony 1904, Final Report (G19-1905) | population census | 1904 | Second census in intercensal methods (Preston-Bennett, Coale, Preston), also source for birthplace, urban definitions and ethnic categories. |
| Reports on the Births and Deaths Registration Act, 1895-1903 (G2-1895, G75-1896, G46-1897, G73-1898, G72-1899, G60-1900, G53-1901, G71-1903, G85-1905, G96-1905) | official statistical series | 1895-1903 | Gives crude birth and death rates by year and age-specific mortality for 1895-96, from which life tables and registration completeness are derived. |
| A8-1895, Report of Births and Deaths Registration, January-March 1895 | official statistical series | Jan-Mar 1895 | Shows high child mortality in the first registration quarter (2,036 of 3,950 deaths under five). |
| G74-1896, Report of the Medical Officer of Health for 1895 | commission or official report | 1895 | Shows white urban mortality above England and Wales age-specific benchmark, motivating the sanitary reform narrative. |
| G1-1898, Final Report of the Registrar of Births and Deaths for the Year 1896 | official statistical series | 1896 | Source for the 2.1 percent allowance for illegitimate births among whites. |
| Cape of Good Hope House of Assembly Debates 1894 | parliamentary debates | 1894 | Illustrates legislators' indifference to the Births and Deaths Registration Bill. |
| Births and Deaths Registration Act No. 7 of 1894; Public Health Act 1897; Immigration Restriction Act 1902; urban locations act 1902 | legislation gazette | 1894-1902 | Establishes the administrative and legal framework for data collection, public health, and segregation. |
| Cape Town Medical Officer of Health, Report for 1902 | commission or official report | 1902 | Evidence for rising tuberculosis incidence among the poorly housed Coloured population. |
| Cape of Good Hope Statistical Register 1891, 1903, 1904 | official statistical series | 1891-1904 | Supplies municipal receipts, wool revenue, median wage rates by group and price/rent/housing information used to infer real income change and labour-market discrimination. |
| G39-1893, Report of the Labour Commission | commission or official report | 1893 | Evidence on farm, urban and skilled wages by colour and labour shortage. |
| Cape of Good Hope Civil Service List, 1893 | official statistical series | 1893 | Source for clerks' salaries in the civil service. |
| A6-1906, Report of the Select Committee on the Factory Act | commission or official report | 1906 | Evidence on sweated female labour in clothing and cigarette manufacture and wage rates. |
| G29-1899, Report of the Select Committee on the Employment of Women in the Civil Service | commission or official report | 1899 | Supports claim that women were preferred in offices because they could be paid less. |
| South African Statistics 1982 (Department of Statistics) | official statistical series | 1915-1980 | White Union vital statistics 1915-19 used to link Cape turn-of-century rates to later twentieth century. |
| Union and Republic population censuses 1911-1980 (age/sex/marital tabulations used for Princeton indices If, Im, Ig, Ih) | population census | 1904-1980 | Used to compute Hutterite-standardised fertility indices for whites, Coloureds and blacks at census dates. |
| Unpublished raw data from the 1980 census (Central Statistical Services) | population census | 1980 | Used to note the 24 percent undercount adjustment of black population in 1980. |
| Report on Black Deaths 1979, Report 07-05-02 | official statistical series | 1979 | Shows black death registration was under 50 percent complete. |
| Sadie demographic estimates 1936-1960 (Projections of the South African Population 1970-2020) | secondary literature compilation | 1936-1960 | Provides black birth, death and life expectancy estimates for linking to the 1891-1904 results. |
| Grobbelaar, Projections and Analysis of the South African Population 1980-2015 | secondary literature compilation | 1960-1980 | Provides 1960, 1970 and 1980 total fertility and mortality estimates for blacks. |
| Fertility Survey 1982, HSRC Report S-129 (van Tonder) | household or labour survey microdata | 1982 | Evidence on contraceptive use among never-married black women. |
| Simkins doctoral thesis on demographic base for black South Africans since 1960 (University of Natal, 1984) | author survey or fieldwork | 1970s | Supports claim that 1970s extra-marital fertility was about half of marital fertility. |
| Official history of the war (Amery, Times History; Grant, History of the War) | secondary literature compilation | 1899-1902 | Used to estimate Cape participation and casualties in Anglo-Boer War to argue limited demographic impact. |
| Knibbs, 1911 Census of Australia, Australian Life Tables | official statistical series | 1891-1900 | Benchmark for comparing white Cape infant mortality and life expectancy. |
| Model life and fertility tables (Coale-Demeny West; Coale and Trussell natural fertility schedules; UN Manual IV) | secondary literature compilation | n/a | Standards for fitting life tables and computing natural-fertility crude birth rates. |
New data¶
Author-constructed demographic estimates for the Cape Colony 1891-1906 by ethnic group (white, Malay, Other Coloured, Mfengu, Other Black): life tables, crude birth and death rates, infant mortality, net migration by age and urban-rural components, derived by indirect methods from the two censuses and registration reports and presented in Table 1.
Inference¶
Quantification: Statistical analysis.
Indirect demographic estimation from two censuses (Preston-Bennett, Coale, Preston methods, back-projection, stable-population and model life tables) cross-checked against vital statistics and natural-fertility benchmarks for each group; compared with Australia, then linked to later twentieth-century series; the economic and social interpretation relies on descriptive wage and price tables and secondary historiography.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| White net immigration accounted for about 39 percent of white population growth between 1891 and 1906. (p. 87) | 1891 and 1904 census birthplace tables with assumptions on equal mortality of foreign- and local-born. | Systematic data |
| The white fertility transition began around 1890, with falling crude birth rates from 39.0 to 33.5. (p. 88) | Registered births adjusted upward, back-projected cohorts and Preston estimates, compared with natural fertility benchmark and British transition dates. | Partial data |
| Non-white groups experienced life expectancy about fifteen years below whites and infant mortality about double. (p. 95) | Life tables from registration data and census-based methods for each group. | Partial data |
| Urban growth for Coloureds and blacks was driven mainly by net migration, and urban Coloured and black mortality was extremely high, so migration to town meant greater mortality hazards. (p. 95) | 1904 census urban/rural tabulations combined with assumed equal urban-rural fertility and 1895-96 and 1903 urban death rates. | Partial data |
| Fertility was natural for all groups except whites, differences in crude birth rates being explained by marital fertility. (p. 98) | Comparison of estimated crude birth rates with natural fertility rates applied to married women, with illegitimacy allowances. | Partial data |
| Wage data show a clear colour discrimination, with the real cost of living changing little between 1891 and 1904, underlying mortality differentials. (p. 106) | Median wage rates from Statistical Registers, Labour Commission evidence, price and rent information. | Partial data |
| The gap in life expectancy between whites and others widened until 1946 and narrowed thereafter; the black fertility transition is only beginning. (p. 111) | Links 1891-1904 estimates to Sadie, Grobbelaar and Union vital statistics series. | Partial data |
| Agriculture entered crisis from the mid-1890s (phylloxera, rinderpest, drought, falling wool and feather prices) while the Cape economy depended increasingly on the Transvaal. (p. 102) | Statistical Register export and revenue figures plus secondary studies (Mabin, Bundy, Scully, Wilson). | Illustrative or anecdotal |
The author's caveats¶
Age reporting likely distorted, populations not closed to migration, differential underenumeration across ages and sexes is possible, black death registration underregistered and coverage problems in area C, no linguistic identification of whites, merged Hottentot/Mixed Coloured categories, 1904 census taken under war-disrupted conditions, and little research on wages and employment with unhelpful census organization.
Gaps in the evidence¶
The economic interpretation rests on median wage and price quotations and secondary literature rather than systematic wage, income, or occupational series linked to the demographic groups. Direct tests of mortality differentials would need cause-of-death data by race and district, or individual-level census manuscripts, and sources for rural black mortality such as mission or magistrate records. Many of these registration and census documents survive in the Cape archives and could be used to check the assumptions about stability, underenumeration and equal urban-rural fertility.
Replicability: Data described but not public.