SEGREGATION AND

APARTHEID IN
TWENTIETH-
CENTURY SOUTH
AFRICA

Edited by William Beinart

and Saul Dubow

London and New York

 First published 1995
by Routledge
11 New Fetter Lane, London EC4P 4EE

This edition published in the Taylor & Francis e-Library, 2003.

Simultaneously published in the USA and Canada

by Routledge
29 West 35th Street, New York, NY 10001

Editorial contributions © 1995 William Beinart and Saul Dubow

© 1995 Individual contributions

All rights reserved. No part of this book may be reprinted or

reproduced or utilized in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication Data

Segregation and apartheid in twentieth-century South Africa/edited
by William Beinart and Saul Dubow.
p. cm.—(Rewriting histories)
Includes bibliographical references.
1. Apartheid—South Africa. 2. South Africa—Race relations.
3. South Africa—History—1909–1961. 4. South Africa—
History—1960– I. Beinart, William. II. Dubow, Saul.
III. Series.
DT1757.S44 1995
323.1'68'0904–dc20 94–36134

ISBN 0-203-42544-8 Master e-book ISBN

ISBN 0-203-73368-1 (Adobe eReader Format)

ISBN 0-415-10356-8 (hbk) ISBN 0-415-10357-6 (pbk)
 3
CAPITALISM AND CHEAP
LABOUR POWER IN
SOUTH AFRICA:
From segregation to apartheid1*
Harold Wolpe

More than any other single work, Harold Wolpe’s discussion of the cheap
labour and reserve-subsidy thesis became central to the radical analysis of
segregation. Wolpe had been an opposition activist in South Africa at the
time of the political upheavals before and after Sharpeville (1960). He
escaped imprisonment to establish himself as a politically committed
academic in Britain. Wolpe argued that cheap labour in the South
African context was best procured through the system of migrant labour
and that key elements of segregation policy reinforced this arrangement.
The migrant labour system ensured that the mines predominantly used
the labour of adult males whose families remained in rural areas.
Capitalists were able to pay African workers meagre ‘bachelor’ wages
because the costs of both the physical and social reproduction of the
labour force were borne by their families who remained primarily
responsible for maintaining subsistence agriculture in the reserves.
Wolpe’s analysis was innovative in that it recognized the inadequacy of a
simple class analysis of South African society and attempted to theorize
the relationship between segregation, the labour market and reserves. If
this system was to survive, reserves for African people had to be
entrenched (as envisaged in the 1913 Land Act); mass urbanization, he
implied, would undermine the cheap labour supply. Wolpe’s article
continues to examine how the reserves strategy was pursued in the
middle years of the century when the capacity of these areas to provide
subsistence for their inhabitants was undermined. He suggests that the
more directive and coercive homeland policy post-1948, by which the
state intended to exclude Africans through tighter control of population
movements rather than development of the reserves, resulted from the
collapse of subsistence production in these areas.
60
 CAPITALISM AND CHEAP LABOUR POWER

INTRODUCTION
There is undoubtedly a high degree of continuity in the racist
ideological foundations of apartheid 2 and of the policy of
segregation which prevailed in the Union of South Africa prior to
the election of the Nationalist Party to power in 1948. It is,
perhaps, this continuity which accounts for the widely held view
that fundamentally apartheid is little more than segregation
under a new name. As Legassick expresses it: ‘after the Second
World War segregation was continued, its premises unchanged, as
apartheid or “separate development”’.3 According to this view,
such differences as emerged between segregation and apartheid
are largely differences of degree relating to their common
concerns—political domination, the African reserves and African
migrant labour. More particularly, the argument continues, in the
political sphere, apartheid entails a considerable increase in White
domination through the extension of the repressive powers of the
state; the Bantustana policy involves the development of limited
local government which, while falling far short of political
independence and leaving unchanged the economic and political
functions of the reserves, nevertheless, in some ways, goes beyond
the previous system in practice as well as in theory; and, in the
economic sphere apartheid ‘modernizes’ the system of cheap
migrant labour and perfects the instruments of labour coercion:

Apartheid, or separate development, has meant merely

tightening the loopholes, ironing out the informalities,
eliminating the evasions, modernizing and rationalizing the
inter-war structures of ‘segregationist’ labour control.4

While it will be necessary, at a later stage, to question this

characterization of the differences between segregation and
apartheid, it is relevant to consider at this point how the variance
between the two ‘systems’ summarized above has been
explained.
Generally, the explanations advanced account for the increased
racial oppression manifested by apartheid on the basis of the
contention that the governing Nationalist Party’s ideology is more
racist than that of its predecessors, and for the intensified political
repression by reference to the Party’s totalitarian ideology.
According to this view, the government, in pursuance of its racist

61
 HAROLD WOLPE

ideology, and even at the cost of economic rationality, introduced

a series of measures which extended racial discrimination to its
limits. The effect of this was to produce widespread opposition
which the government met, acting in pursuance of its totalitarian
ideology, by a drastic curtailment of political rights and an
elaborate system of state security. This set in train a vicious cycle
of resistance and repression which led, in due course, also to
international condemnation of, and pressure on, South Africa.
The Bantustan policy of separate development was the response
to these combined internal and external political pressures and
was designed both to divert opposition and to transfer conflict out
of the ‘white’ urban areas to the African rural ‘homelands’.5
Legassick, however, has proposed a far more complex account
of apartheid, at least in so far as the control of the labour force is
said to be the main area of change, but ultimately his explanation
is also unsatisfactory. He argues that the main components of the
policy of segregation are:
restrictions on permanent urbanization, territorial
separation of land ownership, and the use of traditional
institutions as providers of ‘social services’ and means of
social control …[and] Along with other mechanisms of
labour coercion …the system of migrant labour which
characterized South Africa’s road to industrialization.6
This system, which emerged in a period in which ‘gold’ and
‘maize’ were the dominant productive sectors of the economy,
undergoes rationalization and ‘modernization’ in the context of
an economy in which massive ‘secondary industrialization’ is
occurring. Apartheid is the attempt of the capitalist class to meet
the expanding demand for cheap African labour in the era of
industrial manufacturing capital; at the same time it is the
realization of the demand of White workers for protection against
the resulting increased competition from Black workers. The
outcome of the ‘modernization’ of segregation in the African rural
areas (the Bantustans) is to leave the ‘economic and political
functions (of the reserves)…unchanged’ and thus to preserve the
economic and social foundations of the system of cheap migrant
labour. This is complemented in the urban industrial areas by the
refinement of the mechanisms of labour coercion which
guarantees the cheapness of African labour. Legassick describes
the situation in graphic terms:
62
 CAPITALISM AND CHEAP LABOUR POWER

apartheid has meant an extension to the manufacturing

economy of the structure of the gold-mining industry. In the
towns, all remnants of African land and property ownership
have been removed, and a massive building programme of
so-called ‘locations’ or ‘townships’ means that the African
work force is housed in carefully segregated and police
controlled areas that resemble mining compounds on a large
scale. All the terms on which Africans could have the right to
reside permanently in the towns have been whittled away so
that today no African…has a right to permanent residence
except in the ‘reserves’.
The attempt to relate alterations in policy to changes in social
conditions—primarily the development of a class of
manufacturing industrialists—unquestionably represents an
advance over the simplistic view that apartheid is the result of
ideology. Intense secondary industrialization does have a
bearing on the development of apartheid but the mere fact that it
occurs does not explain why it should lead to the attempt to
extend the ‘structures of gold-mining’ to the economy as a
whole. Legassick is clearly correct in arguing that secondary
industrialization intensifies the demand for a cheap African
labour force at various levels of skill and that this is
accompanied by new problems of control for the capitalist state.
The problems of control (including the control of wage levels)
are not, however, simply or primarily a function of the demand
for labour power which is cheap, but crucially a function of the
conditions of the production and reproduction of that labour
power. It is in this respect that the crucial gap in Legassick’s
analysis appears, for by focusing largely on the development of
secondary industrialization and by assuming that the economic
and political functions of the reserves continue unchanged and,
therefore, that the migrant labour system remains what it has
always been, he fails to grasp the essential nature of the changes
which have occurred in South Africa. The analysis of these
essential changes—the virtual destruction of the pre-capitalist
mode of production of the African communities in the reserves
and, therefore, of the economic basis of cheap migrant labour
power and the consequent changes in and functions of ‘tribal’
political institutions—will constitute the subject matter of the
third and subsequent sections of the paper.

63
 HAROLD WOLPE

IDEOLOGY, POLICY AND CAPITALISM IN SOUTH

AFRICA
A few exceptions apart,7 the literature—radical, liberal and racist
alike—analyses and describes the society in terms of racial
concepts. 8 Even where the relationship between classes is
incorporated into the discussion, race is nevertheless treated as
the dominant and dynamic force.9 ‘Racial segregation’, ‘separate
development’, ‘racial discrimination’, ‘racial groups’ (African,
White, coloured and Asiatic), ‘colour bar’, ‘White ruling class’,
‘race relations’, etc., etc.—these are the concepts of the analysis of
South Africa. The predominance of these concepts can, no doubt,
be attributed to the opaqueness of racial ideology, which is
reflected, inter alia, in the formulation of laws in racial terms, in the
content of the mass media, in the policies and ideological
statements of all the political parties and organizations (both
Black, White and also mixed) and in almost the entire intellectual
product of the society.
The overwhelming importance accorded to race in these
approaches is apparent, above all, in their treatment of the
relationship between racially oriented action and ‘the economy’.
Thus, on the one hand, the content of ‘Native’ or ‘Bantu’ policy (to
use the official terms) which can be found in the legislative
programmes, government policies and commission reports both
before and after 1948, is analysed in its own terms and treated as
being concerned solely with the regulation of ‘race relations’. On
the other hand, whether the economy is conceived of in terms of
liberal economics,10 or in Marxist terms as a capitalist mode of
production,11 racial beliefs are treated as a force external to, but
productive of, distortions in the otherwise rational economic
system. In its most advanced form this leads to the ‘theory’ of the
plural society which both reflects the dominant ideology and
provides an apparently scientific corroboration of it. 12 This
approach accepts, precisely by reference to the racial or ethnic
content of the laws, policies and ideologies current in the society,
the critical salience of race to the exclusion of the mode of
production. The basic structure of the society is seen, in this and
the other analyses referred to, in the relationship between a
dominant White group and a dominated Black group.
It is of fundamental importance to stress that in this perspective
the state in South Africa comes to be treated as the instrument of

64
 CAPITALISM AND CHEAP LABOUR POWER

oppression of Whites over Blacks but (precisely because class

relationships are not normally included in the analysis) as neutral
in the relationship between classes. It in no way detracts from the
conception of the state as an instrument of White domination,
however, to insist that the South African state is also an
instrument of class rule in a specific form of capitalist society.
Indeed, while there have been, of course, variations in emphasis
and detailed policy (variations which stem, in part, from the
specific class composition of and alliances in the parties which
have ruled from time to time), nevertheless, since the
establishment of the Union of South Africa in 1910 (to go back no
further), the state has been utilized at all times to secure and
develop the capitalist mode of production.13
It is not possible in this paper to discuss in detail the historical
evidence which demonstrates this. First, the state has acted
directly through the law (e.g. Land Bank Act which provides for
subsidies and grants to White farmers), through special agencies
(for example, the Industrial Development Corporation which has
been important in the growth of, inter alia, the textile industry),
through the development of state enterprises and in other ways to
foster capitalist development.14
Second, the repressive apparatus of the state (police, army,
prisons, courts, etc.) has been used broadly in two ways. First, as the
occasion arose, to coerce workers, whether Black or White, on
behalf of or in support of employers. A small selection of the more
dramatic examples of this would include the 1914 White mine
workers’ strike, the 1922 general strike (Rand Revolt) of White
workers, the 1946 African mine workers’ strike and the 1972
Ovambo workers’ strike. Second, to enforce the laws which either
overtly guarantee the perpetuation of capitalism—laws such as the
Industrial Conciliation Act 1924, the Masters and Servants Act, the
Native Labour (Settlement of Disputes) Act 1953, the Native Labour
Regulation Act 1911, and so on, or (as in the case of most laws
affecting Africans) which covertly perform the same functions—for
example, the Natives Land Act 1913, and the Native (Abolition of
Passes and Co-ordination of Documents) Act 1952.
It is precisely from the racial terms that are employed in these
laws that their ideological function can be determined. The
enactment of laws, the express purpose of which is the regulation
of relationships between racial groups and the ordering of the
conduct of the members of legally defined racial categories, is
65
 HAROLD WOLPE

both an expression of racist ideology and a means of reinforcing

that ideology. This is so because not only do racial laws, in
common with other laws, appear as neutral to the capitalist
structure of the society by taking that structure as given, but more
importantly, like other laws but in a different way, they actively
operate to mask both the capitalist nature of the society altogether
and the consequences of their provisions for the functioning of
that system.
The history of South Africa shows the emerging dominance, first
through British imperialism, and then also through internal
capitalist development, of the capitalist mode of production. The
development of this dominant mode of production has been
inextricably linked with two other modes of production—the
African redistributive economies and the system of labour-
tenancy and crop-sharing on White farms. The most important
relationship is between capitalism and the African economies and
although it is not entirely satisfactory to do so, for reasons of space
the discussion which follows is restricted to this relationship.
These two modes of production may be briefly characterized as
follows.
(a) First, the capitalist mode of production in which (i) the
direct labourers, who do not own the means of capitalist
production, sell their labour power to the owners of the means of
production who are non-labourers, and (ii) the wages the labourer
receives for the sale of his labour power are met by only a portion
of the value of the product he actually produces, the balance being
appropriated as unpaid labour (surplus value) by the owners of
the productive means.
(b) Second, the mode of production in the areas of African
concentration (particularly, but not exclusively, the reserves) in
which (i) land is held communally by the community and worked
by social units based on kinship (the enlarged or extended family),
and (ii) the product of labour is distributed, not by exchange, but
directly by means of allocation through the kinship units in
accordance with certain rules of distribution.
This is not to argue either that other forms of production were
not developing in the interstices of the African societies or that
they were not continuously undergoing profound changes.15 On
the contrary, as will be elaborated later, the central argument of
the present paper is based on the occurrence of such
transformations. What must be stressed, however, is that in the
66
 CAPITALISM AND CHEAP LABOUR POWER

period of capitalist development (from, say, 1870) African

redistributive economies constituted the predominant mode of
rural existence for a substantial (for much of the period, a
majority), but continuously decreasing number of people.
The simultaneous existence of two modes of production within
the boundaries of a single state has given rise to the notion of the
‘dual economy’.16 As Frank and others have shown for Latin
America, however, the assumption that different modes of
production can be treated as independent of one another is
untenable.17
In South Africa, the development of capitalism has been bound
up with, first, the deterioration of the productive capacity and
then, with increasing rapidity, the destruction of the pre-capitalist
societies. In the earlier period of capitalism (approximately 1870
to the 1930s), the rate of surplus value and hence the rate of capital
accumulation depended above all upon the maintenance of the
pre-capitalist relations of production in the reserve economy
which provided a portion of the means of reproduction of the
migrant labour force. This relationship between the two modes of
production, however, is contradictory and increasingly produces
the conditions which make impossible the continuation of the pre-
capitalist relations of production in the reserves. The consequence
of this is the accelerating dissolution of these relations and the
development, within South Africa, towards a single, capitalist,
mode of production in which more and more of the African wage-
labour force (but never the whole of it) is ‘freed’ from productive
resources in the reserves. This results in important changes in the
nature of exploitation and transfers the major contradiction from
the relationship between different modes of production to the
relations of production within capitalism.
Here we arrive at the critical point of articulation between
ideology, racial political practice and the economic system.
Whereas segregation provided the political structure appropriate
to the earlier period, apartheid represents the attempt to maintain
the rate of surplus value and accumulation in the face of the
disintegration of the pre-capitalist economy. Or, to put it in
another way, apartheid, including separate development, can best
be understood as the mechanism specific to South Africa in the
period of secondary industrialization, of maintaining a high rate
of capitalist exploitation through a system which guarantees a
cheap and controlled labour force, under circumstances in which
67
 HAROLD WOLPE

the conditions of reproduction (the redistributive African

economy in the reserves) of that labour force are rapidly
disintegrating.

THE AFRICAN RESERVES—THE SOCIAL AND

ECONOMIC BASIS OF CHEAP MIGRANT LABOUR
POWER
In commenting on the conceptions of the ‘subsistence’ economy in
the dual economy thesis, Laclau stressed that:

The latter (i.e. the ‘subsistence’ economy) was presented as

completely stagnant and inferior to the former in capital,
income and rate of growth. All relations between the two
were reduced to the provision by the backward sector of an
unlimited supply of labour to the advanced sector. It has
now been repeatedly shown that this model underestimates
the degree of commercialization which is possible in rural
areas, as well as the degree of accumulation in peasant
enterprises.18

Arrighi, Bundy and others have shown that the processes of

commercialization and accumulation were, no less than in Latin
American societies, occurring in African rural economies in
Rhodesia and South Africa.19 It is none the less true that by not
later than 1920 the overwhelming economic and political power
of the capitalist sector had succeeded, whether through unequal
terms of trade or otherwise, in underdeveloping the African
economy so that it no longer presented any significant
competitive threat to White farmers. Production, in the African
reserves, of a marketable surplus became increasingly rare, finally
disappearing altogether. Unlike some other situations elsewhere,
therefore, the capitalist sector was unable to extract the (non-
existent) surplus product directly from the African pre-capitalist
sector. The relations between the two sectors were, indeed,
‘…reduced to the provision by the backward sector’ of a supply of
labour power to the capitalist sector. The peculiar feature of this
labour force is that it is migrant and temporary, returning to the
reserves in between periods of work, and retains means of
production in the African economy or has a claim on such means.
The exploitation of migrant labour power of this kind enables the
68
 CAPITALISM AND CHEAP LABOUR POWER

capitalist sector to secure an increased rate of surplus value. How

is this effected?
A number of attempts have been made to explain why it is that
Africans who are in possession of agricultural means of
production in the reserves nevertheless enter wage employment
in the capitalist sector. It is unnecessary for present purposes to
consider these explanations which, in any event, are generally
inadequate. What is relevant is the conventional
conceptualization of wage labour as the means of supplementing
deficiencies in the income derived from production in the
reserves. In this view the need to supplement income arises from
inefficient farming methods, inappropriate values, an outmoded
social system and so on which lead to underproduction—the only
relationships between the capitalist sector and the traditional
economy are territorial and through the market for western
consumer goods which capitalism introduces into the latter
economy. Underlying this conception is, thus, the dual economy
thesis in which there is no place for an analysis of the way in
which capitalism enters into, lives off and transforms the rural
African economy.
If, however, the African economy and society is treated as
standing in an ancillary relationship to the capitalist sector, then a
different analysis follows. When the migrant labourer has access
to means of subsistence, outside the capitalist sector, as he does in
South Africa, then the relationship between wages and the cost of
the production and reproduction of labour power is changed.
That is to say, capital is able to pay the worker below the cost of his
reproduction. In the first place, since in determining the level of
wages necessary for the subsistence of the migrant worker and his
family, account is taken of the fact that the family is supported, to
some extent, from the product of agricultural production in the
reserves, it becomes possible to fix wages at the level of
subsistence of the individual worker. Arrighi has shown this to be
the basis of cheap labour in Rhodesia, and Schapera has argued
this for South Africa on the basis of the following quotation from
the Chamber of Mines’ (the largest employer of migrant labour)
evidence to the Witwatersrand Native Mine Wage Commission
(21/1944):

It is clearly to the advantage of the mines that native

labourers should be encouraged to return to their homes
69
 HAROLD WOLPE

after the completion of the ordinary period of service. The

maintenance of the system under which the mines are able
to obtain unskilled labour at a rate less than ordinarily paid
in industry depends upon this, for otherwise the subsidiary
means of subsistence would disappear and the labourer
would tend to become a permanent resident upon the
Witwatersrand, with increased requirements.20
In the second place, as Meillassoux has pointed out:
The agricultural self-sustaining communities, because of
their comprehensiveness and their raison d’être are able to
fulfil the functions that capitalism prefers not to assume…
the functions of social security.21
The extended family in the reserves is able to, and does, fulfil ‘social
security’ functions necessary for the reproduction of the migrant
workforce. By caring for the very young and very old, the sick, the
migrant labourer in periods of ‘rest’, by educating the young, etc.,
the reserve families relieve the capitalist sector and its state from the
need to expend resources on these necessary functions.
The accessibility to the migrant worker of the product (and of
the ‘social services’) of the reserves depends upon the conservation,
albeit in a restructured form, of the reciprocal obligations of the
family. The interest of the capitalist sector in preserving the
relations of the African familial communities is clear—if the
network of reciprocal obligations between migrant and family
were broken neither the agricultural product nor the ‘social
services’ of the African society would be available to the worker. It
is no accident that the South African state has consistently taken
measures, including the recognition of much of African law and
custom, the recognition of and grant of powers to chiefs, the
reservation of areas of land, etc., aimed at preserving the ‘tribal’
communities.
In passing it may be noted that the pressures towards retaining
the family communities in a restructured form came also from the
migrant labour force. As Meillassoux puts it:
the capitalist system does not provide adequately for old-
age pensions, sick leave and unemployment compensations,
they have to rely on another comprehensive socio-economic
organization to fulfil these vital needs…. It follows that
the…preservation of the relations with the village and the
70
 CAPITALISM AND CHEAP LABOUR POWER

familial community is an absolute requirement for wage-

earners, and so is the maintenance of the traditional mode of
production as the only one capable of ensuring survival.

However, the preservation of the social relations of the familial

community is, quite obviously, only one aspect of the migrant
cheap labour system. The social obligation to provide subsistence
and security is only of relevance to both migrant and employer if
that obligation can actually be met from the agricultural product.
But this requires both the retention of the pre-capitalist mode of
production, at least in so far as this guarantees the allocation of
productive land to all members of the community, and the
maintenance of certain levels of production. Both of these raise
strategic problems for the capitalist sector.
The first problem relates to the tendency in capitalist
development for ownership of land to become concentrated and
the consequent development of a landless class ‘free’ of means of
production. The importance of this stems from the obvious fact
that landless families would be unable to supplement the
migrant’s wages.
The drive towards land acquisition came from sections of both
the African and White groups and the threat that this might lead
to a landless class of Africans was met by two different sets of
measures.
The Natives Land Act 27/1913 defined (or scheduled) certain
areas as African reserves and laid down that no African could
henceforth purchase or occupy land outside the reserves.
Simultaneously the Act prohibited Whites from acquiring, or
occupying, land in the reserves. It was stated in parliament, at the
time, that the purpose of the Act was to ensure the territorial
segregation of the races. This stated purpose has generally been
accepted, by politicians as well as social scientists, as a sufficient
explanation for the Act which has come to be regarded as the
cornerstone of territorial segregation. Recently, however, some
writers have argued that the Act can be interpreted as an attempt
to remedy the shortage of African labour on White farms, and to
prevent Africans utilizing communal or private capital from
repurchasing European owned land which had been acquired by
conquest.22
Be that as it may, the consequences (possibly unintended) of the
section of the Act which prohibits the purchase and occupation by
71
 HAROLD WOLPE

Whites of land in the reserves have been consistently ignored or

misconstrued. The effect of this provision was very far-reaching—
it halted the process, whether through the market or otherwise, by
which more and more land was wrested from or made
unavailable to Africans. Since the reserves (particularly with the
additions made in terms of the Native Trust and Land Act 1936)
roughly coincided with the rural areas into which Africans had
already been concentrated, the Act had the effect of stabilizing the
existing distribution of land. Liberal historians have stressed the
‘protection’ this provided against a further diminution of the land
held by Africans, but the importance of this ‘protection’ in
preventing the economic basis of migrant labour power from
being undermined through landlessness has been almost
completely overlooked.
The removal of reserve land from a market open to White
capital did not eliminate the possibility of land becoming
concentrated in the hands of a relatively small class of African
landowners. Indeed, the fact, already mentioned, that Africans
were beginning to repurchase land outside of the then de facto
‘reserves’, is proof that some Africans had the necessary resources
for land purchase. No doubt, an immediate effect of the Natives
Land Act would have been to lead these potential purchasers to
search for suitable land in the reserves, but here there were other
obstacles in their way. The Glen Grey Act of 1894 and various
other proclamations and enactments (which were to be extended
and elaborated from time to time until the 1930s) laid down the
rule of one-man-one-plot in the reserves. This rule impeded the
concentration of land.
The second strategic problem arises from the necessity to
maintain production in the reserves at a level which, while not too
low to contribute to the reproduction of migrant workers as a
class, is yet not high enough to remove the economic imperatives
of migration. While, as Arrighi has shown, there is no simple
relationship between production levels in the rural economy and
the rate of migration—social, political and other economic
conditions may affect this—none the less low levels of agricultural
and craft production constitute a necessary condition of labour
migration. This is so because both the demands of and the
economic returns from high output farming would tend to render
the population immobile. At the same time, if output is allowed to
drop too low then the reserve product becomes relatively a less
72
 CAPITALISM AND CHEAP LABOUR POWER

important element in subsistence and, unless wages are increased,

threatens the reproduction of migrant workers.
In the earlier period (roughly prior to 1930) the state in fact did
extremely little to develop or assist agriculture in the reserves.
Statistics of food and other agricultural production are extremely
sparse. Clearly, by the mid-1920s surpluses were either extremely
small or non-existent and continued to decline. Thus the Director
of Native Agriculture estimated that the income from the sale of
produce (after consumption needs had been met), for a family
unit of five in the Transkei, to be £4 per annum in the period before
1929. In the Northern Transvaal, Van der Horst points out:
The extent to which grain was purchased to supplement
domestic production resulted in there being practically no
income from the sale of farm produce for the purchase of
other food, and clothes, or for the payment of taxes and
school fees.23
Nevertheless, the evidence as a whole shows that the drop in the
level of production to this point did not yet threaten the migrant
system.
The level of production is not, however, the only point. What is
equally important is the extent to which productive activities and,
therefore, means of subsistence are distributed among all the
families in the reserves and, in particular, among the families with
which migrant wage workers are connected. This is so since the
product of the reserve economy will only be available to
contribute to the reproduction of the migrant labour force if the
migrant’s family (given that he remains in a relationship of
reciprocal obligation with them) is in fact producing means of
subsistence in the reserves. There is virtually no data specifically
on this point but it nevertheless is possible to infer from the
situation as a whole that few wage workers, up to say 1920 or so,
did not have a supplementary source of subsistence for
themselves and their families in the reserves.
Table 3.1 shows the clear predominance of Africans in the
mining industry as compared to other major sectors of the economy
in the period 1910–40. Practically all African mineworkers were
(and are) recruited through the Chamber of Mines recruiting
organizations from the reserves (and also from territories outside
South Africa—up to 50 per cent in the years covered by Table 3.1),
and returned to their homes on completion of a term of service
73
 HAROLD WOLPE

Table 3.1 African employment in mining, private industry and the South African
Railways and Harbours

* These figures include coloured and Asiatic workers as well, numbering

probably about 6,000 in each year shown.

of a year or so. From this, and from our knowledge of the general
economic situation of the areas from which mineworkers were
recruited, it can be inferred that they retained economic links with
their kin in the reserves.
The conclusion can thus be drawn that in the early period of
industrialization in South Africa (the period of gold mining) the
reserve economy provided the major portion of Africans
employed in capitalist production, at any given moment, with
supplementary subsistence and was thus a crucial condition of the
reproduction of the migrant working class. The crucial function
thus performed by the policy of segregation was to maintain the
productive capacity of the pre-capitalist economies and the social
system of the African societies in order to ensure that these
societies provided a portion of the means of reproduction of the
migrant working class.

THE CORROSION OF THE SOCIAL AND ECONOMIC

BASIS OF CHEAP MIGRANT LABOUR POWER
The production and reproduction of the migrant labour force thus
depended upon the existence of a rough equilibrium between
production, distribution and social obligation in the reserves—the
level of production in the reserves together with wages being more
or less sufficient to meet the (historically determined) subsistence
requirements of migrants and their families, while land tenure

74
 CAPITALISM AND CHEAP LABOUR POWER

and familial community relationships ensured the appropriate

distribution of the reserve product. This equilibrium was,
however, inherently fragile and subject to irresistible pressures.
Given the developed incapacity of the reserves to generate a
surplus product, the limited area of land available (fixed by the
Natives Land Act), the increasing pressure of population and,
therefore, congestion on the land, the loss, at any given time, of a
large proportion of the economically active adults to temporary
employment in the capitalist sector, the relatively backward and
inefficient farming methods, the only possibility of ensuring
appropriate levels of agricultural production is through
investment by the capitalist sector. In fact, as was pointed out
earlier, the state’s expenditure on agricultural development in the
reserves has always been extremely low, increasing only
marginally as conditions of production worsened. The immediate
consequence of all this was a rapid decline in the agricultural
product in the reserves.
By the 1920s attention was already being drawn to the
deterioration of the situation in the African areas and in 1932 the
Native Economic Commission Report (1930–2) commented at
length on the extremely low productivity of farming in the
reserves, on the increasing malnutrition and on the real danger of
the irreversible destruction of the land through soil erosion. Every
subsequent government commission dealing with the reserves
reiterated these points and drew attention to the decline in
output.24 By 1970, Gervasi summed up the situation:
There is good evidence that the standard of living of
Africans in the Reserves has actually fallen over the last two
decades. The condition of the Reserves can only be
described as one of abject poverty. There is a mass of other
evidence corroborating the income statistics. According to a
survey conducted in 1966, almost half the children born in
most Reserves were dying before the age of five. In fact
mortality of this kind is unknown in any other industrial
country. It can only mean that the vast mass of the
population in the typical Reserve is living well below the
level of subsistence most of the time.25
The conclusion which emerges is that, overall, production in the
reserves provides a declining fraction of the total subsistence of
migrant labourers.
75
 HAROLD WOLPE

The level of production, however, is not the only relevant

aspect; the way in which the product is distributed must also be
considered. Capitalist development produced further changes
which had the effect of altering the pattern of distribution so that
the diminishing agricultural product became more and more
unequally distributed and less and less available to wage
labourers. The development of classes in the reserves (or, perhaps,
strata within classes), which had already begun in the nineteenth
century, was intensified and broadened. There can be little doubt
that the processes leading to the concentration of land-holding in
the hands of a relatively restricted class have been continuous and
that, correlatively, the class of landless rural dwellers is
substantial and growing. The Native Laws Commission (1948),
among other studies already referred to, has provided some data
concerning these developments. The Commission reported, for
example, on the Ciskei:

What goes on in the Reserves? Nearly 1/3 of all families

have no arable land. The average land-holder works, what
is, under the climatic conditions obtaining in the Ciskei, a
sub-economic unit of land. He owns what is, because of its
poor quality, sub-economic numbers of stock. Above him is a
relatively small favoured class of bigger owners. It is known
that there are individuals who own 100 head of cattle and as
many as a thousand sheep. Below him are thousands who
own nothing. In Keiskama Hoek, before the drought, 29% of
all married men owned no cattle, another 33% from one to
five head.

The Tomlinson Commission (1956) [see n. 24—Eds], also reported

finding substantial numbers of landless inhabitants of reserves. In
addition the Commission provided evidence of striking
inequalities (12.7 per cent of the families earn 46.3 per cent of the
total income accrued inside the reserves) thus adding to the
picture of income-less or very low income groups. It follows from
this that a proportion of the families living in the reserves produce
either very little or, in the case of the landless and cattleless, no
means of subsistence.
Thus far I have discussed the economic changes in the reserves
which undermined, to a significant degree, the economic basis of
the migrant labour system and, by the same token a substantial
76
 CAPITALISM AND CHEAP LABOUR POWER

economic prop of cheap labour power. The essence of the

argument has been that the amount of subsistence available to the
migrant labour force and their families in the reserves has either
diminished because the overall decline in production has resulted
in a decrease in the product per capita or has virtually disappeared
because of the partial or total loss, in the case of some families, of
means of production.
This, however, is only one aspect of the process, for, in the
second place, the product of the reserves may no longer be
available to the migrant as a means of subsistence for himself and
his family in the reserves by reason of the termination of the
reciprocal social obligations of support between the migrant and
his kin in the reserve, even where the latter continues to produce
subsistence. An important condition for this change is the
permanent urbanization of a substantial number of workers. The
process of secondary industrialization and the development of
the tertiary sector of the economy provided the opportunity for
the development of, and was accompanied by, an ever-increasing,
permanently urbanized, industrial proletariat.
The first point to note is that the percentage of the African
population in the urban areas increased from 12.6 per cent in 1911 to
23.7 per cent in 1946 and by 1971 was approximately 38 per cent.
It is unnecessary to detail the growth of manufacturing—it
contributes today more than gold mining and agriculture
combined to the national product. What is more pertinent is the
data showing the changes in African urbanization and
employment in manufacturing.
In Table 3.1 figures of African employment in private industry
were set out for some years in the period 1910–40. The changes
between 1940 and 1970 are shown in Table 3.2 following:
Table 3.2 African employment in private industry

The significance of these figures derives from the fact that, in

contrast to Africans employed in mining, those employed in
secondary industry are not brought into employment (or returned
77
 HAROLD WOLPE

to the reserves) through recruiting organizations. They are, of

course, subject to the pass laws and other legal provisions
restricting their right of residence in urban areas, laws which have
become increasingly rigorous over time. Nevertheless,
employment in manufacturing coupled with residence in
‘locations’ and townships undoubtedly enabled large numbers of
African workers to settle permanently in the urban areas and in
due course to raise families there.
It is, once again, extremely difficult to calculate with any
accuracy what proportion of urban industrial workers have
become fully dependent upon wages for subsistence, that is to say,
how many are fully proletarianized.26 Not only are the statistics
incomplete and unsatisfactory but in addition very little analysis
has been made of the relationship between permanently
urbanized workers and the African societies in the reserves.27
Despite this, however, there can be no doubt that during the
period 1910–70 (and particularly in the period during and since
the Second World War) the number of Africans in the urban areas
having no relevant links with the reserves has grown steadily and
rapidly and that they today constitute a significant, if not major,
proportion of African industrial workers.

APARTHEID: THE NEW BASIS OF CHEAP LABOUR

The focus in the two previous sections has been largely on the
economic foundation of cheap migrant labour power in the
reserve economy and on the processes which have continuously
and to an ever-increasing degree undermined this foundation.
The immediate result of the decline in the productive capacity of
the pre-capitalist economies was a decrease in the agricultural
product of the reserves resulting, therefore, in a decrease of the
contribution of the reserves towards the subsistence necessary for
the reproduction of the labour force. This threatened to reduce the
rate of surplus value through pressure on wages and posed, for
capital, the problem of preventing a fall in the level of profit.
The solution, for capital, to this problem must take account of
the complementary effect of the erosion of the economic
foundations of cheap migrant labour power upon both the
African rural societies and the urbanized industrial proletariat. I
have already shown that the system of producing a cheap migrant
labour force generated rural impoverishment, while at the same
78
 CAPITALISM AND CHEAP LABOUR POWER

time it enabled extremely low wages to be paid to Africans in the

capitalist sector. But increasing rural impoverishment, since it
removes that portion of the industrial workers’ subsistence which
is produced and consumed in the reserves, also intensifies urban
poverty. This twofold effect of capitalist development tends to
generate conflict, not only about wages, but about all aspects of
urban and rural life and to bring into question the structure of the
whole society. This broadening and intensification of conflict is
met by political measures which in turn lead to an increasingly
political reaction.
This struggle began long before 1948 when the conditions
discussed above began to emerge (and control measures to be
taken), but the particularly rapid urbanization and
industrialization fostered by the Second World War sharpened
and intensified the trends we have been discussing and the
resultant conflicts. The 1940s were characterized by the variety
and extent of the industrial and political conflicts especially in the
urban, but also in the rural areas. In the period 1940–9 1,684,915
(including the massive strike of African mineworkers in 1946)
African man-hours were lost as compared with 171,088 in the
period 1930–9. Thousands of African workers participated in
squatters’ movements and bus boycotts. In 1946 the first steps
were taken towards an alliance of African, coloured and Indian
political movements and this was followed by mass political
demonstrations. Towards the end of the 1940s a new force—
militant African intellectuals—appeared on the scene. There were
militant rural struggles at Witzieshoek and in the Transkei. These
were some of the signs of the growing assault on the whole society
(and the structure of cheap labour power which underpinned it)
which confronted the capitalist state in 1948.
For English-dominated large-scale capital (particularly mining
but also sections of secondary industry), the solution both to the
problem of the level of profit and to the threat to their political
control implicit in growing African militancy was to somewhat
alter the structure of segregation in favour of Africans. Indeed, the
1948 recommendations of the Native Law Commission
(appointed in 1946 by the United Party government precisely in
response to the changing nature of African political struggle) for
an alternative mode of control of African labour which included
certain restricted reforms and modifications of the racial political-
economic structure, were accepted by the United Party as its
79
 HAROLD WOLPE

policy in the 1948 election in which it was defeated by the

Nationalist Party. The implementation, had it occurred, of that
policy might possibly have had consequences for both the
Afrikaner petit bourgeoisie and, also, the White workers which
would have led them into a collision with the state. The point is
that reforms which would have resulted in higher real wages and
improved economic conditions for Africans could only be
introduced without a corresponding fall in the rate of profit
provided they were bought at the cost of the White working
class—that is to say, either through a drop in the wages of White
workers or the employment of Africans, at lower rates of pay, in
occupations monopolized, until then, by White workers.
Historically the latter aspect has been at the centre of the conflicts
and tension between the White working class and large-scale
capital—conflicts which also reached their peak in terms of strikes
in the 1940s.
The alternative for the Afrikaner working class, resisting
competition from African workers, for the growing Afrikaner
industrial and financial capitalist class, struggling against the
dominance of English monopoly capital, and, perhaps, for a petit
bourgeoisie threatened with proletarianization by the advance of
African workers (and the Indian petit bourgeoisie), was to assert
control over the African and other non-White people by whatever
means were necessary. For the Afrikaner capitalist class, African
labour power could be maintained as cheap labour power by
repression; for the White worker, this also guaranteed their own
position as a ‘labour aristocracy’. Thus the policy of apartheid
developed as a response to this urban and rural challenge to the
system which emerged inexorably from the changed basis of
cheap labour power. What was at stake was nothing less than the
reproduction of the labour force, not in general, but in a specific
form, in the form of cheap labour power.
At the most general level, that of control of the African political
challenge, apartheid entails the removal of the limited rights
which Africans and coloureds had in the parliamentary
institutions of the White state; the revision of old and the
introduction of a whole complex of new repressive laws which
make illegal militant organized opposition (e.g. Suppression of
Communism, Unlawful Organizations and Sabotage Acts, etc.),
and the building of all-powerful agencies of control—security

80
 CAPITALISM AND CHEAP LABOUR POWER

police, Bureau of State Security, the army, and police and army
civilian reserves, etc.
In the economic sphere measures have been introduced to
prevent or contain the accumulation of pressure on the level of
wages. Most obvious in this regard is the Natives (Settlement of
Disputes) Act which makes it illegal for Africans to strike for
higher wages or improved working conditions. This, coupled
both with the fact that African trade unions are not legally
recognized and that their organization is impeded also by other
measures, has effectively prevented the emergence of an African
trade union movement capable of having any significant effect on
wages. The decline in industrial strikes since 1948 and the
tendency of real wages for Africans to fall indicates the success of
government policy.
Less obvious, but having the same purpose of controlling the
development of strong African pressure for higher wages, are the
important measures introduced by the Nationalist government
relating to African job and geographical mobility. The nature and
meaning of these measures has been obscured by the terms of the
relevant laws and the government’s policy statements to the effect
that Africans were to be regarded only as temporary migrants in
the urban areas, there only as long as they ministered to White
needs.
The pass laws and the Native Urban Areas Act 1923 which
regulated the right of residence in urban areas, were, of course,
available in 1948. The ‘modernization’ of the pass laws under the
Native (Abolition of Passes and Coordination of Documents) Acts
and the establishment of labour bureaux which serve to direct
African workers to where White employers require them has been
effected through a battery of amendments to old laws and the
introduction of new laws which give the state exceptionally wide
powers to order Africans out of one area and into another. There
are practically no legal limitations on these powers which can be
used to remove ‘excess’ Africans from areas where their labour is
not required or ‘troublesome’ Africans to outlying, isolated areas
where they will be politically harmless. All Africans are, legally,
only temporary residents in the urban areas.
In its application to the urban areas, apartheid appears
predominantly and with ever-increasing thoroughness in its
coercive form. In its application to the reserves it has undergone a
number of changes in content—culminating in the programme of
81
 HAROLD WOLPE

self-development—in which the attempt both to establish forms

of control which Africans would regard as legitimate and to
institutionalize conflict has been an increasingly important
ingredient although coercion is never absent. This policy towards
the reserves has been, whatever other purpose it may have had in
addition, centrally concerned, as in the past, with the control and
supply of a cheap labour force, but in a new form.
The idea of the total separation of the races, although an
integral element of the Nationalist Party’s programme, was not
regarded as an attainable objective by the government. The
impossibility of achieving total separation was underlined by the
Tomlinson Commission which estimated (or rather, as we now
know, grossly overestimated) that by the turn of the century, if all
its recommendations for the reconstruction of the reserves were
implemented, there would be parity of Whites and Africans in the
‘White’ areas.
Nor, in the early years of its regime, did the government accept
the possibility of the reserves becoming self-governing and
autonomous areas. In 1951 Verwoerd (then Minister of Native
Affairs) told the Upper House of Parliament that the Opposition
had tried to create the impression that:

I had announced the forming of an independent Native

State…a sort of Bantustan with its own leader…that is not
the policy of the Party. It has never been that, and no leader
has ever said it, and most certainly I have not. The Senator
wants to know whether the series of self-governing areas
will be sovereign. The answer is obvious. How could small
scattered states arise? We cannot mean that we intend by
that to cut large slices out of South Africa and turn them into
independent states.

There is, in fact, little to suggest that, in the first few years of rule,
the Nationalist Party had a fully worked-out policy in relation to
the reserves or one which differed significantly from that of earlier
governments. There are, however, two important points to be
noted.
First, the government already had clearly in mind the
establishment of an apparatus of control which would be cheap to
run and acceptable to the African people. The 1951 Bantu
Authorities Act which strengthened the political authority of the
82
 CAPITALISM AND CHEAP LABOUR POWER

(compliant) chiefs, subject to the control of the state—indirect

rule—was the first (and, at the outset, very conflictual) step in that
direction. Second, political control in the reserves was obviously
recognized to be no solution to the problem of the never-ending
enlargement of a working class totally removed from the reserves.
The rather modest proposals of the Tomlinson Commission to
spend £104 million over ten years for the reconstruction of the
reserves, to end one-man-one-plot in order to create a stable class
of farmers and a landless class of workers, and to develop the
reserves economically through White capital investment on the
borders and in the reserves themselves, were not accepted by the
government. There are probably two reasons for this rejection.
First, facts brought to light by the Commission showed that to
implement the Commission’s recommendations relating to
agricultural development would have served simply to hasten the
ongoing processes which were obviously resulting in the
formation of a class of landless rural dwellers and to intensify the
migration of workers to the urban centres resulting in a class of
workers unable to draw on the reserves for additional subsistence.
Consequently, expenditure on agricultural improvement may
have seemed pointless and even dangerous since it would
exacerbate the pressures and conflicts in the towns. Second, the
abolition of restrictions on land-holding and the assisted
development of a class of ‘kulaks’, as recommended by the
Commission, also carried with it certain possible dangers. On the
one hand, this could lead to a resurgence of African competition to
White farmers which it had been one of the purposes of the
Natives Land Act of 1913 to destroy. On the other hand, the
emergence of an economically strong class of large peasants
presented a potential political threat to White domination.
Whatever the reasons, by 1959 the government’s policy began
to change in significant respects. Without attempting to set out a
chronological record, I want to analyse the emergence after 1959
of separate development as the mode of maintaining cheap labour
in the reserves (complementing that in the urban areas) which
takes as given the changes in the African ‘tribal’ economies and
erects, under the overarching power of the capitalist state, an
institutionalized system of partial political control by Africans.
That is to say, the practice and policy of separate development
must be seen as the attempt to retain, in a modified form, the
structure of the ‘traditional’ societies, not, as in the past, for the
83
 HAROLD WOLPE

purposes of ensuring an economic supplement to the wages of the

migrant labour force, but for the purposes of reproducing and
exercising control over a cheap African industrial labour force in
or near the ‘homelands’, not by means of preserving the pre-
capitalist mode of production but by the political, social,
economic and ideological enforcement of low levels of
subsistence.
In 1959, in the parliamentary debate on the Promotion of Bantu
Self-Government Act, the Prime Minister Dr Verwoerd stated:
if it is within the capacity of the Bantu, and if those areas
which are allocated to him for his emancipation, or rather,
which are already his own, can develop into full independence,
then it will develop in this way.
(Hansard, 1959, col. 6520)
This was echoed by Vorster in 1968 (Hansard, 1968, col. 3947):
We have stated very clearly that we shall lead them to
independence.
Significantly, the ideological shift from White supremacy to self-
determination and independence was accompanied by a parallel
alteration in the ideology of race. Thus, whereas in all its essentials
Nationalist Party ideology had previously insisted upon the
biological inferiority of Africans as the justification for its racialist
policies, as the government was impelled towards the Bantustan
policy so it began to abandon certain of its previous ideological
positions. Now the stress fell upon ethnic differences and the
central notion became ‘different but equal’. In 1959 the Minister of
Bantu Affairs an Development, De Wet Wel, stated:
There is something…which binds people and that is their
spiritual treasures, the cultural treasure of a people. It is
those things which have united other nations in the world.
That is why we say that the basis of our approach is that the
Bantu, too, will be linked together by traditional and
emotional bonds, by their own language, their own culture,
their national possessions….
(Hansard, 1959, col. 6018)
More and more the term ‘race’ gives way to ‘nation’, ‘ethnic
group’, ‘volk’.
There is an obvious necessity for this ideological change since a
84
 CAPITALISM AND CHEAP LABOUR POWER

policy of ethnic political independence (for each of the eight

ethnic groups identified) was incompatible with an ideology of
racial inferiority. Nor would the latter have facilitated the attempt
to set up the complex machinery of government and
administration intended, in fact, to institutionalize relations
between the state and the reserves and to carry out certain
administrative functions necessary for economic development in
the reserves. What all this amounts to, as one writer has expressed
it, is ‘racialism without racism’.
The Transkei Constitution Act was passed in 1963 and
provided for a legislative assembly to exercise control over
finance, justice, interior, education, agriculture and forestry, and
roads and works. The Republican government retains control,
inter alia, over defence, external affairs, internal security, postal
and related services, railways, immigration, currency, banking
and customs. It need hardly be stressed that this arrangement in
no way approaches political independence. At the same time it
must not be overlooked that within limits, set both by the
Constitution and the available resources, the Transkeian
government exercises real administrative power. By this means
the South African state is able to secure the execution of certain
essential social control and administrative functions at low cost
particularly as a considerable portion of government expenditure
can be obtained through increased general taxes. Thus in 1971 the
Transkeian government’s budget was £18 million of which £3 1/2
million was obtained through taxation of Transkeian citizens.
It is, however, in the sphere of economic development that the
emerging role of the reserves can be seen most clearly. I am not
here referring to the rather minor role of the various development
corporations (Bantu Development Corporation, Xhosa
Development Corporation and so on) in fostering economic
development in the reserves. In fact, up to the present they have
largely served to assist small traders and commercial interests by
means of loans—that is, they appear to be instruments for the
nurturing of a petit bourgeoisie and have little to do with
economic growth in the reserves. Far more important is the state’s
policy of industrial decentralization.
This policy which has been the subject of government
commissions and legislation is also the concern of a Permanent
Committee for the Location of Industry. At all times the policy of
decentralization has been tied to the Bantustan policy and this
85
 HAROLD WOLPE

meant, at first, the establishment of ‘White’ industries on the

borders of the Black ‘homelands’. Between 1960 and 1968 some
£160 million was invested in industrial plant in the border areas
and approximately 100,000 Africans were employed in these
industries which were absorbing 30 per cent of Africans entering
jobs each year by 1969. By 1971 there were plans for a rapid
expansion (including car factories and chemical plants) of
industrial development in the border regions. I would suggest
that the policy of border industrial development can only be
understood if it is seen as an alternative to migration as a
mechanism for producing cheap labour power. There are three
aspects of the situation which need to be stressed.
First, neither the provisions of the Industrial Conciliation Act
nor Wages Act determinations made for other regions apply to the
border industries. This is extremely important in two respects.
Since the Industrial Conciliation Act is inapplicable, Section 77
which empowers the minister of labour to reserve certain jobs for
particular racial groups also does not apply and neither do the
provisions of industrial agreements which reserve the higher-paid
skilled jobs for White workers. This being so it becomes possible
to employ Africans in jobs which, in the ‘White’ areas, are the
exclusive preserve of White workers. The effect of this, in
conjunction with the inapplicability of wage determinations for
other areas, is that a totally different and much lower wage
structure becomes possible and has arisen.
Second, as elsewhere, African trade unions are not recognized
and the provisions of the Natives (Settlement of Disputes) Act
apply.
The third, and in some ways perhaps the most important
aspect, relates to the conditions of life of the African workers in
the border industries. Not only, as has already been indicated, is
the level of subsistence extremely low in the ‘homelands’ but in
addition there are virtually no urban areas which might tend to
increase this level. The assessment by the state, employers’
organizations and so on, of African subsistence requirements in
the reserves is much lower than in the main industrial centres.
This fact is not altered (or, at least will not be altered for a
considerable period) by the necessity of establishing townships of
some kind for the housing of workers employed in industry. It is
an interesting index of the state’s policy that a major item of
expenditure for the so-called development of the reserves has
86
 CAPITALISM AND CHEAP LABOUR POWER

been for town planning. A United Nations Report (No. 26, 1970, p.
15) stated:
Town planning has throughout been a major portion of
expenditure. Thus in 1961 a five-year development plan for
the reserves was inaugurated which projected an
expenditure of £57 million, but two-thirds of this amount was
allocated for town planning, while the next largest item—
£7.3 million—was for soil conservation.
The towns planned will be, no doubt, simple in the extreme,
supplying little in the way of the complex services and
infrastructure of the ‘White’ urban areas. Despite the state’s
expenditure all the indications are that what will be established
will be rural village slums.28
Recently, the government reversed its previous rejection of the
Tomlinson Commission’s recommendation that Whites be
allowed, under certain conditions, to invest capital in the reserves.
As in the case of the border industries various incentives are held
out to induce investment. These include ‘tax holidays’, tariff
reductions, development loans and so on. All the considerations
discussed above in relation to the border industries apply with
equal force to industrial development within the reserves. It is still
too soon to say anything about the likely level of investment
inside the reserves although some investment has already
occurred. Nevertheless, the change in policy must be seen as a
further significant step towards the establishment of an extensive
structure of cheap labour power in the reserves.

CONCLUSION
The argument in this paper shows that apartheid cannot be seen
merely as a reflection of racial ideologies and nor can it be reduced
to a simple extension of segregation.
Racial ideology in South Africa must be seen as an ideology
which sustains and reproduces capitalist relations of production.
This ideology and the political practice in which it is reflected is in
a complex, reciprocal (although asymmetrical) relationship with
changing social and economic conditions. The response of the
dominant classes to the changing conditions, mediated by these
ideologies, produces the two faces of domination—segregation
and apartheid.
87
 HAROLD WOLPE

The major contradiction of South African society between the

capitalist mode of production and African pre-capitalist
economies is giving way to a dominant contradiction within the
capitalist economy. The consequence of this is to integrate race
relations with capitalist relations of production to such a degree
that the challenge to the one becomes of necessity a challenge to
the other. Whether capitalism still has space (or time) for reform in
South Africa is an issue which must be left to another occasion.

EDITORS’ NOTE
a Bantustan: the term applied in the 1950s to areas reserved for African
occupation. Many of these had existed since the nineteenth century
and, although a small proportion of the total area of the country, they
included the heartlands of some old African chiefdoms. The
Nationalist government intended to extend and consolidate them
into ten units, pushing the total area to over 13 per cent of the land
surface. They were given a form of self-rule which was later
intended to become political independence. Bantustan, initially used
by H.F.Verwoerd, was taken up by the opposition critical of the
balkanization of the country, as a disparaging term for these mini-
and micro-states.

NOTES
* [The footnotes for this article have been altered from the original
Harvard system to endnotes for consistency. Both notes and text
have been shortened—Eds.]
1 In revising an earlier draft of this paper I have benefited from
criticisms and comments made by a number of people. I am
particularly grateful to S.Feuchtwang, R.Hallam, C.Meillassoux and
M. Legassick.
2 Although the term ‘apartheid’ has more or less given way to ‘self-
development’ in the language of the Nationalist Party, it remains the
term mostly widely used to characterize the present system in South
Africa.
3 M.Legassick, ‘South Africa: Forced Labour, Industrialization, and
Racial Differentiation’, later published in R.Harris (ed.), The Political
Economy of Africa (Boston, 1975). See also A.P.Walshe, ‘The Changing
Content of Apartheid’, Review of Politics, XXV (1963), 360; B.Bunting,
The Rise of the South African Reich (London, 1964), 305.
4 Legassick, ‘Forced Labour’.
5 M.Szeftel, ‘The Transkei: Conflict Externalization and Black
Exclusivism’, Collected Seminar Papers on the Societies of Southern
Africa in the 19th and 20th Centuries, vol. 3, Institute of
Commonwealth Studies, University of London (1972).

88
 CAPITALISM AND CHEAP LABOUR POWER

6 Legassick, ‘Forced Labour’.

7 M.Legassick, ‘Development and Underdevelopment in South
Africa’, unpublished paper, Royal Institute of International Affairs
(1971) and ‘Forced Labour’; H.Wolpe, ‘Industrialization and Race in
South Africa’, in S.Zubaida (ed.), Race and Racialism (London, 1970);
S.Trapido, ‘South Africa in a Comparative Study of
Industrialization’, Journal of Development Studies, 7, 3 (1971);
F.Johnstone, ‘White Prosperity and White Supremacy in South Africa
Today’, African Affairs, vol. 69, issue 275 (1970), 124–40.
8 H.J. and R.E.Simons, Class and Colour in South Africa 1850–1950
(London, 1969); S.van der Horst, ‘The Effects of Industrialization on
Race Relations in South Africa’, in G.Hunter (ed.), Industrialization
and Race Relations (London, 1965); P.Van den Berghe, South Africa: A
Study in Conflict (Berkeley, CA, 1967); N.J.Rhoodie, Apartheid and
Racial Partnership in South Africa (Pretoria, 1969).
9 Simons and Simons, Class and Colour, 614–15.
10 Van der Horst, ‘The Effects’; Van den Berghe, South Africa;
R.Horwitz, The Political Economy of South Africa (London, 1967).
11 Simons and Simons, Class and Colour; A.Asherson, ‘South Africa:
Race and Politics’, New Left Review, 53 (1969), 55–67. For a critique,
see Wolpe, ‘Industrialization and Race’.
12 Van den Berghe, South Africa; L.Kuper and M.G.Smith, Pluralism in
Africa (Berkeley, CA, 1969).
13 See L.Althusser, ‘Ideology and Ideological State Apparatuses’, in his
Lenin and Philosophy (London, 1971); N.Poulantzas, ‘The Problem of
the Capitalist State’, New Left Review, 58 (1969), 67–78.
14 See Horwitz, Political Economy, 355.
15 There is a serious lack of adequate material on, or analysis of African
societies. It is clear that no adequate account of the dynamics of
South African society can be arrived at without a proper history of
these societies.
16 D.Hobart Houghton, The South African Economy (London, 1964).
17 G.A.Frank, Capitalism and Underdevelopment in Latin America
(London, 1967).
18 E.Laclau, ‘Feudalism and Capitalism in Latin America’, New Left
Review, 67 (1971), 19–46.
19 G.Arrighi, ‘Labour Supplies in Historical Perspective: A Study of the
Proletarianization of the African Peasantry in Rhodesia’, Journal of
Development Studies, 6 (1970), 197–234; C.Bundy, ‘The Emergence and
Decline of a South African Peasantry’, African Affairs, vol. 71, issue
285 (1972), 369–88.
20 Arrighi, ‘Labour Supplies’; I.Schapera, Migrant Labour and Tribal Life
(London, 1947).
21 C.Meillassoux, ‘From Reproduction to Production’, Economy and
Society, 1, 1 (1972), 102.
22 F.Wilson, ‘Farming’, in M.Wilson and L.Thompson (eds), Oxford
History of South Africa, vol. 2 (Oxford, 1971); Legassick, ‘Forced
Labour’.
23 S.van der Horst, Native Labour in South Africa (London, 1971).
89
 HAROLD WOLPE

24 Union of South Africa, Report No. 9 of the Social and Economic

Planning Council, The Native Reserves and their Place in the Economy of
the Union of South Africa (U.G. [Union Government Parliamentary
Papers] 32/1946); Report of the Native Laws Commission (U.G. 28/
1948); Union of South Africa, Summary Report of the Commission for the
Socio-Economic Development of the Bantu Areas within the Union of South
Africa (U.G. 61/1955)—the Tomlinson Commission. [The Report
appears to have been published in 1955, but the document and
attached White Paper appear to have been released in 1956—Eds.]
25 S.Gervasi, Industrialization, Foreign Capital and forced Labour in South
Africa (United Nations ST/PSCA/Set A./10, 1970).
26 Hobart Houghton, South African Economy, 86–7.
27 But see P.Mayer, Townsmen or Tribesmen (London, 1962).
28 The appendix to Mayer, Townsmen, 2nd edn (1971) provides an
account of a ‘dormitory’ town.

90