Report of the Delimitation Commission appointed under the South Africa Act, 1909 (second delimitation, after the 1911 census)¶
Laurence, P. M.; Solomon, W. H.; Fawkes, A. W. (1913). Report of the Delimitation Commission appointed under the South Africa Act, 1909 (second delimitation, after the 1911 census). The Government Printing and Stationery Office, Pretoria.
| Type | primary source |
| Scope | Wider apartheid literature |
| Method | primary document |
| Period studied | 1909–1913 |
| Regions | Union of South Africa, Cape Province, Transvaal, Natal, Orange Free State, Witwatersrand |
| Themes | origins segregation, politics ideology elections, demography urbanisation, business and state |
| Citation key | laurence1913 |
Argument¶
This is the official report of the second judicial commission that redrew constituencies for the Union House of Assembly and the Provincial Councils after the 1911 census. It applies the South Africa Act's formula, in which seats are allocated by European male adults against a fixed 1904 quota while divisions are drawn by registered voters, and the result is that the Transvaal gains nine seats and the Cape gains none. The commissioners say they tried to preserve existing electoral boundaries, and they use community of interest and administrative boundaries as criteria. They note that boundary changes by other bodies, such as fiscal divisions and municipalities, complicated the work. They state that party advantage was not a consideration, and they recommend simpler, uniform registration for European voters, equal Provincial Council and Assembly sizes, and a 20 per cent margin around the quota. The document shows the machinery of a white-only franchise at Union and the Witwatersrand's new weight.
Evidence and method¶
The text of the South Africa Act (ss. 32-34, 38-43), the 1904 and 1911 censuses of European adult males, the voters' registers of 1911-12, Australian Commonwealth electoral precedents, and representations from political associations and local residents. Appendices give schedules of every new division with its voter count and boundary descriptions, and a list of maps.
Place in the debates¶
- Equal versus protected rural representation under the Union franchise formula (see White electoral politics, delimitation and party systems). The Commission notes that two quota methods leave the Cape with 51 seats despite a larger electorate than a Transvaal quota would justify. It proposes flexibility, but only within its statutory mandate.
- Whether delimitation should weigh party or political advantage (see White electoral politics, delimitation and party systems). It says that party-political effects such as South African Party versus Labour were not considerations, and that it confined itself to the statutory criteria.
Findings¶
-
The Union quota was fixed from the 1904 census of European male adults divided by the 121 seats of the first House. (p. 4)
is obtained by dividing the number of European male adults within the Union
-
The Cape electorate grew by a third of the Transvaal's increase, yet Cape representation stayed unchanged while the Transvaal's rose sharply. (p. 6)
the Cape electorate has increased to an extent exceeding one-third of the corresponding increase in the Transvaal
-
The Commission preferred to leave existing boundaries and political groupings undisturbed. (p. 8)
The Commission were of opinion that the association of groups for political purposes should, so far as practicable, remain undisturbed
-
It suggested widening the permissible margin around the quota to 20 per cent to reduce frequent minor changes and local dissatisfaction. (p. 9)
a somewhat wider discretion say 20 per cent, above or below the quota might not be allowed to the next Commission
-
The franchise for Europeans varied by province, with residence-based adult male suffrage in the Transvaal and Orange Free State. (p. 20)
in the Transvaal and Orange Free State, so far as the European population is concerned, there is an adult male suffrage, based on residence.
-
The Transvaal gained nine seats, giving it 45 divisions. (p. 20)
The Transvaal, as has already been pointed out, is entitled to nine additional members, making the total number of electoral divisions for that Province forty-five.
-
The Witwatersrand quota was set above the rest of the Transvaal's, and its white electorate was dense but transient. (p. 20)
is approximately seven per cent, above, that of the remainder of the Province, including Pretoria, seven per cent, below the quota of the Province as a whole.
-
Barberton, a mining and rural district, had only about half the voters needed for a division, which forced linkage with neighbouring districts. (p. 22)
the number of voters in the district is only 1159, which is only about half of what is required under the Act to constitute an electoral division.
-
Partisan arguments for a seat were put to the Commission, which said it did not weigh them. (p. 24)
seeing that the whole of the rest of the rural part of the Province is solid in favour of one party, an arrangement which would probably give one country seat to the opposite party has much to recommend it
-
The Commission recognised distinct interests of eastern Transvaal gold miners (labour supply, railways) from Johannesburg's. (p. 25)
a representative of the gold mining interests in the Province other than those concentrated on the Witwatersrand.
-
Rapid Witwatersrand growth justified new constituencies, with Springs nearly doubling its voters. (p. 27)
division of Springs the number of voters is nearly double what it was at the date of the previous delimitation
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Census of 1904, European male adults (South Africa Act 1909, s.34) | population census | 1904 | Fixes the statutory Union quota (349,837 European male adults divided by 121 seats) used to allocate seats between provinces. |
| Quinquennial census of the European population, 1911 | population census | 1911 | Compared with the 1904 male-adult count to decide how many members each province is entitled to. |
| Voters' registers 1911/1912 (Cape, Transvaal, Natal, Orange Free State) | official statistical series | 1907-1912 | Supplies electorate totals and per-division voter numbers on which the new quotas, minima, maxima and the schedules of divisions are based. |
| Natal registers 1911-12 and 1912-13 | official statistical series | 1911-1913 | Used to choose the 1911 register as the basis and to check whether voter movements in 1912-13 justified changes. |
| Orange Free State registration of 1912 and the 1910 list | official statistical series | 1910-1912 | Shows that all OFS divisions stayed within the statutory limits on the recent register, so few changes were needed. |
| South Africa Act, 1909, sections 32-34, 38-43 and 71 | legislation gazette | 1909 | Supplies the quota formula, the 15 per cent margin and the mandate to delimit constituencies. |
| Report of the first Delimitation Commission (1910) | commission or official report | 1910 | Provides the baseline divisions, principles and precedent (e.g. Barberton, Cape Peninsula) that this Commission preserves or revises. |
| Cape Act No. 39 of 1906 and Proclamations 160, 162, 194, 207-222 of 1911 and 252 of 1912 (new fiscal divisions and field cornetcies) | legislation gazette | 1906-1912 | Documents the boundary changes of fiscal divisions, field cornetcies and municipalities that complicated redrawing Cape constituencies. |
| Representations and memorials from political associations and residents (Unionist Political Association; Pilgrims Rest and Sabie electors; Montagu inhabitants) | other | 1912 | Evidence on community of interest and local sentiment weighed against statistical constraints when drawing boundaries. |
| Consultations with the Department of the Interior, Justice, Lands, Surveyor-General, and Civil Commissioners and field cornets | state archives | 1912 | Official correspondence and file reconciliation used to reconcile voter numbers and boundaries after proclamations. |
| Surveyor-General technical boundary descriptions and maps (Appendices II and III) | maps spatial data | 1912-1913 | Gives technical boundary descriptions and illustrative maps for the new divisions in the Cape Peninsula, Natal, the Witwatersrand and Pretoria. |
| Tables of voters per new electoral division (Schedules A-F, Appendix II) | official statistical series | 1912-1913 | Gives the number of voters for each of 50 Cape, 17 Natal HA, 45 Transvaal and 17 OFS HA divisions, showing compliance with quota limits. |
New data¶
Commission-constructed schedules of new electoral divisions (Cape 51, Natal HA 17 and PC 25, Transvaal 45, OFS HA 17 and PC 25) giving voter numbers and boundary descriptions, compiled from the 1911/12 registers; plus a provincial quota/minimum/maximum table.
Inference¶
Quantification: Descriptive tables.
Statutory-formula application: the allocation of seats applies the Act's quota arithmetic to census and register counts, and the boundary decisions are reasoned case by case against section 40(3) criteria (community of interest, existing boundaries, communication, sparsity or density), with divisions checked against the minimum/maximum band. Comparison with the previous delimitation and with Australian practice supports the recommendations.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| The Union quota is 2891, derived from 349,837 European male adults in 1904 divided by 121 members. (p. 4) | 1904 census figures and s.34 of the Act. | Systematic data |
| The statutory method gives the Transvaal nine extra seats (45) while the Cape stays at 51, even though the Cape electorate grew far less than the Transvaal's; equal electoral areas would give the Cape 58. (p. 7) | Comparison of 1911 census-based entitlement and register increases (Cape +6,189; Transvaal +17,312). | Systematic data |
| Five of the nine new Transvaal seats go to the Witwatersrand, which has the densest and most floating European population, so the register is a poor guide to the voting power available at an election. (p. 20) | Register counts and the Commission's general observation on labour and mining fluctuations. | Partial data |
| The Barberton, Lydenburg and Carolina arrangement was left unchanged because the voters in the gold-mining areas fall short of the required number by at least 600, and changing it would mutilate two magisterial districts. (p. 23) | Register counts (Barberton 1159 voters), petitions and memorials, and the 1910 Commission's decision. | Partial data |
| Frequent changes in non-electoral boundaries (fiscal divisions, field cornetcies, municipalities) in the Cape made delimitation difficult and call for a freeze between registration and delimitation. (p. 13) | Gazetted proclamations and correspondence with departments, with examples such as Pearston, Maraisburg and Montagu. | Illustrative or anecdotal |
| Cape Peninsula growth forced the creation of an additional Peninsular division and the abolition of Jansenville. (p. 11) | Register increase in the Peninsula from 25,572 to 27,450 voters and the previous delimitation. | Systematic data |
The author's caveats¶
The Commission notes that registers are not synchronous across provinces, that Natal registers annually, so that later registers will likely push some divisions beyond the limits, and that discrepancies in voter numbers occurred between field cornetcies (e.g. seventeen or fifteen voters transferred). It also warns that the Witwatersrand register gives little guidance to actual voting power.
Gaps in the evidence¶
The report rests on voters' registers and census counts but gives no underlying registration data, tabulation by race or occupation, or population estimates beyond the European adult-male count, so the effects of the qualification rules on representation cannot be tested. Records of the petitions, memorials and political-association submissions are not reproduced and could test claims about community of interest. Departmental files on the proclamation changes (Interior, Justice, Lands) probably survive in the national archives and could document the boundary discrepancies.
Replicability: Data described but not public.