Fostering critical dialogue and debate on socio-political and economic issues

  Site map  

National Lottery Logo
Dialogue events
Memorial lectures
Writings of Harold Wolpe
About the Trust
Contact us
Harold Wolpe Memorial Trust
Unit F
Colstine Terrace
88 Belvedere Rd
PO Box 44907
Tel: +27 21 674 0361
Fax: 086 670 6772

Email: [email protected]


Wolpe H (1972) ‘Capitalism and cheap labour-power in South Africa: From segregation to apartheid’. Economy and Society, 1(4):425-456

Capitalism and cheap labour-power in South Africa: from segregation to apartheid1 Harold Wolpe


Conventionally, Apartheid is regarded as no more than an intensification of the earlier policy of Segregation and is ascribed simplistically to the - particular racial ideology of the ruling Nationalist Party.

In this article substantial differences between Apartheid and Segregation are identified and explained by reference to the changing relations of capitalist and African pre-capitalist modes of production. The supply of African migrant labour-power, at a wage below its cost of reproduction, is a function of the existence of the pre-capitalist mode. The dominant capitalist mode of production tends to dissolve the pre-capitalist mode thus threatening the conditions of reproduction of cheap migrant labour-power and thereby generating intense conflict against the system of Segregation. In these conditions Segregation gives way to Apartheid which provides the specific mechanism for maintaining labour-power cheap through the elaboration of the entire system of domination and control and the transformation of the function of the pre-capitalist societies.

I. Introduction

There is undoubtedly a high degree of continuity in the racist ideological foundations of Apartheid2 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. (See, for example, Legassick 1972; Walshe, 1963: 360; Bunting, 1964: 305.) As Legassick expresses it:

After the Second World War segregation was continued, its premises unchanged, as apartheid or ‘separate development’ (ibid : 31).

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 Bantustan 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 ‘modernises’ 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. (Legassick, 1972: 47)

While it will be necessary, at a later stage, to question this characterisation 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 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 (and is) designed both to divert opposition and to transfer conflict out of the ‘white’ urban areas to the African rural ‘homelands’ (Szeftel, 1971).

Although such explanations are open to a number of objections, it is sufficient to refer here to only one. As I have argued elsewhere in relation to South Africa (Wolpe, 1970), since a racist ideology may be actualized in a number of alternative ways, it is not sufficient merely to spell out the ideology in order to explain specific political policies which are being pursued. This is all the more obvious when different policies (Apartheid and Segregation) are said, as in the present case, to flow from the same basic ideology. Whatis necessary for an adequate explanation is a specification of the historical conjuncture between ideology, political practice and the mode of production. In Section II this point will be elaborated with specific reference to the South African context.

Legassick (1972), however, has proposed a far more complex account of Apartheid, at least insofar 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 urbanisation, 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 industrialisation (ibid : 30, 31).

This system, which emerged in a period in which ‘gold’ and ‘maize’ were the dominant productive sectors of the economy, undergoes rationalisation and ‘modernisation’ in the context of an economy in which massive ‘secondary industrialisation’ 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 realisation 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’ (Legassick, 1972: 30) and thus to preserve the economic and social foundations of the system of cheap migrant labour (ibid : 46). 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:

... 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 to-day no African ... has a right to permanent residence except in the ‘reserves’ ... (1972: 47).

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 (just as the mere fact of industrialisation of itself does not lead to the opposite result-democratisation, elimination of racial discrimination, etc. as asserted by the ‘industrialism thesis’). 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.

II. Ideology, Policy and Capitalism in South Africa

A few exceptions apart (e. g. Legassick, 1971, 1972; Wolpe, 1970, 1971; Trapido, 1971; Johnstone, 1970) the literature-radical, liberal and racist alike (e.g. Simons and Simons, 1969;.Asherson 1969; Van der Horst 1965; Van den Berghe, 1967; Rhoodie, 1969)-analyzes and describes the society in terms of racial concepts. Even where the relationship between classes is incorporated into the discussion, race is nevertheless treated as the dominant and dynamic force (e.g. Simons and Simons, 1969: 614-15). ‘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 organisations (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 1948,3 is analyzed 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 (Van der Horst, 1965; Van den Berghe, 1967; Hutt, 1964; Horwitz, 1967) or in Marxist terms as a capitalist mode of production (Simons and Simons, 1969; Asherson, 1969), racial beliefs are treated as a force external to, but productive of, distortions in the otherwise rational economic system.4 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. This approach (see e.g. Kuper and Smith, 1970; Van den Berghe, 1967) 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 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, from the conflicts between class and segments of classes and from changing socio-economic conditions), 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. Viewed from this standpoint racist ideology and policy and the State now not only appear as the means for the reproduction of segregation and racial discrimination generally, but also as what they really are, the means for the reproduction of a particular mode of production.5

It is not possible in this paper to discuss in detail the historical evidence which demonstrates this,6 but, insofar as it is necessary the point can be sufficiently established by a brief reference to different functions the State has performed since the formation of the Union of South Africa.

Firstly, 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. Horwitz (1967: 355) has summed up some of the processes as follows:

There has been a significant structural change in the South African economy in the last twenty-five years, when an increasing share of the domestic market has been secured to domestic manufacturers as a deliberate objective of state policy. The instruments employed-ranging from tariff: duty protection at higher rates, advance guarantees of tariff protection to induce large-scale, otherwise high risk investments, import controls to compel foreign manufacturers-exporters to franchise or participate in South African plants, state finance and other inducement-aids for strategic import-substitution, heavily-capitalized industries-have helped industrialize the country.

Secondly, 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 1953, and so on-or (as in the case of most laws affecting Africans) which covertly perform the same functions-for example, the Native Lands 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 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.

It follows from what has been argued in this Section that, in order to break through the mask of racial ideology, it is necessary to show how these racial prescriptions articulate with the mode of production or, more precisely, the modes of production in South Africa.

Laclau’s (1971) distinction between the concept of ‘mode of production’ and the concept of ‘economic system’ is of particular importance in this context.7 He distinguishes between the two as follows:

We understand by ‘mode of production’ an integrated complex of social productive forces and relations linked to a determinate type of the means of production. ... An ‘economic system’, on the other hand, designates the mutual relations between the different sectors of the economy, or between different productive units, whether on a regional, national or world scale. ... An economic system can include, as constitutive elements, different modes of production-provided always that we define it as a whole ... (Ibid: 33).

The economic system in South Africa has always included different modes of production and, before proceeding further with the analysis, the ‘constitutive elements’ of the system in the period with which we are concerned need to be identified.

In the period under consideration the economic system or social formation has been comprised of at least three different modes of production. The history of South Africa8 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 an 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 societie 9or that they were not continuously undergoing profound changes in their relations of production. 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 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.10

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’ (e. g. Hobart Houghton, 1964). As Frank (1967; 1969) 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.

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 1930’s), 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 the conditions of reproduction (the redistributive African economy in the Reserves) of that labour-force is rapidly disintegrating.

III. The African Reserves: The Social and Economic Basis of Cheap Migrant Labour-Power

In the article already cited Laclau (1971: 23), in commenting on the conceptions of the ‘subsistence’ economy in the dual economy thesis, 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.

Arrighi (1970), Bundy (1971) 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. It is, nonetheless 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 under-developing 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.11 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 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.12 What is relevant is the conventional conceptualization of wage-labour as the means of supplementing deficiences 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 under-production-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 (1970) has shown this to be the basis of cheap labour in Rhodesia and Schapera (1947) 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 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...

In the second place, as Meillassoux (1972: 102) has pointed out:

The agricultural self-sustaining communities, because of their comprehensiveness and their raison d’etre are able to fulfil functions that capitalism prefers not to assume ... the functions of social security.

The extended family in the Reserves is able to, and does, fulfil ‘social security’ functions necessary for the reproduction of the migrant work force. 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 portion of the product of the Reserves which is thus indirectly appropriated by the capitalist sector is represented in Figure I.

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.

Although the term ‘Apartheid’ has more or less given way to ‘self-development’ in the language of the Nationalist Party, it remains the term most widely used to characterize the present system in South Africa. In this paper I intend to use ‘Apartheid’ as the generic term to refer to the period (its policies, practices and ideology) since the Nationalist Party took office in 1948-in this sense it subsumes the policy of ‘separate development’. ‘Racial Segregation’ or ‘Segregation’ is employed throughout, unless the context indicates a contrary intention, to refer to the ideology, policies and practices prior to 1948. It need hardly be added that although 1948 is obviously a year of great importance it is not intended to suggest that in that year Apartheid replaced Segregation.

See for example:
Transvaal Local Government Commission of 1922 (Stallard Commission) T.P. 1-122
Native Land Commission (Beaumont Commission) 26-1916
Native Economic Commission 1930-1932 (U.G. 22, 1932)
Social and Economic Planning Council-Report No. 9: The Native Reserves and Their Place in the Economy of the Union of South Africa (U.G. 32/1946)
Report of the Commission for the Socio-Economic Development of the Bantu Areas Within the Union of South Africa (Summary) (U.G. 61/1955) (Tomlinson Commission)
Native Lands Act 1913
Native Trust and Lands Act 1936
Native (Urban Areas) Consolidation Act 1945 and subsequent amendments Bantu Authorities Act 1951

For a critique of this approach see Wolpe (1970).

Althusser’s (1971) essay ‘Ideology and Ideological State Apparatuses is relevant on this point. Althusser (p. 124) suggests that:
… in order to exist, every social formation must reproduce the conditions of its production at the same time as it produces, and in order to be able to produce. It must therefore reproduce:
1. the productive forces,
2. the existing relations of production.
The reproduction of both the forces (that is, labour power) and the relations of production are secured for the most part
... by the exercise of State power in the State Apparatus, on the one hand the (Repressive) State Apparatus, on the other the Ideological State Apparatus. (Althusser 1971: 141.)
See also, Poulantzos, N., ‘The Problem of the Capitalist State,’ New Left Review, November-December 1969, No. 58, p. 67.

Considerable historical material on this point can be found in Simons and Simons (1969).

For a more rigorous and somewhat different treatment see Charles Bettelheim’s Theoretical Comments in Emmanuel, A., Unequal Exchange: A Study of the Imperialism of Trade, London, New Left Review Editions, 1972.

For a good outline of this history see Legassick (1972)

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.

[Where does this endnote go in the text??] While the present paper is restricted to a consideration of African rural economy in the Reserves, to some extent African agriculturalists outside the Reserves on ‘White’ land had a similar relationship to migrant labour. Both this aspect and, as pointed out to me in a personal communication by M. Legassick, the relationship of Reserve migrants to White capitalist agriculture, would need to be incorporated in a full study.

The explanation of why this occurred rather than an intensification of trade relations adverse to the Reserves is related both to the interest of mining capital in labour rather than in the agricultural product and to the utilisation of State power by Afrikaner farmers vulnerable to competition from Africans. For an elaboration of this aspect see Legassick (1972).

Arrighi (1970) has offered the most complex analysis but in the end his explanation in terms of ‘opportunity cost’ is tautological.

Figure I

The Relative Proportion of Surplus to Necessary Labour in the Capitalist Sector where:

The Working-Class is Wholly dependent upon Wages for its Reproduction



The Working-Class derives a portion of its means of Reproduction from the Reserve Economy





(Note: This figure is not drawn to scale)
Where:    S = surplus labour time/product
N = labour time/product necessary for reproduction of labour-power
N1 = the decreased proportion of labour time/product devoted to the reproduction of labour-power by the capitalist sector where portion of the necessary means of subsistence is provided by the Reserve Economy (N2)
Sl = the increased surplus labour time/product.

For the product of African agricultural production to be indirectly available to the capitalist mode of production in this way, two conditions are necessary means of subsistence must actually be produced by the non-capitalist economy and these means must be accessible to the migrant worker and his kin in the Reserves.

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 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. (1972: 103).

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 off 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 insofar 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 Native Lands 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 (Wilson, 1971; Legassick, 1972) 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 re-purchasing European owned land which had been acquired by conquest.

Be that as it may, the consequences (possibly unintended) of the section of the Act which prohibits the purchase and occupation by 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 land-owners. Indeed, the fact, already mentioned, that Africans were beginning to re-purchase 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 Native Lands 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 (which impeded but did not halt the concentration of land ownership) applied both to areas of individual and of communal land tenure. The rule was enforced by Native Commissioners who controlled land transactions between Africans.

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 (1971) 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 effect this-nonetheless 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 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 decline13 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 & per annum in the period before 1929. In the Northern Transvaal, Van der Horst (1971: 305) 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.

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. The relevant question, therefore, is whether wage-labourers in the capitalist sector stem from families which not only have means of production but which also continue to produce means of subsistence ; or whether they originate from families already cut off from subsistence in the Reserves either through landlessness or through the deterioration of the land etc.

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.

Firstly, it is unlikely that at this stage the process of land accumulation by some Africans and the corresponding landlessness of others had proceeded very far. It seems reasonable to assume, therefore, that there was still a relatively even distribution of land-at least in the sense that there was no large group of people who were completely cut off from the land or from families still producing agricultural products.14

Secondly, an inference can be drawn from the distribution of African workers in different sectors of the capitalist economy. Table I shows this distribution for the period 1910 to 1940.

Table I shows the clear predominance of Africans in the mining industry as compared to other major sectors of the economy in the period 1910 to 1935. There is evidence that many, if not the majority, of African workers employed in private industry and by the railways and harbours were migrants. Even if this is not so, however, the overwhelming majority of African workers in the capitalist sector, nevertheless, must have been migrants. Practically all African mine-workers were (and are) recruited through the Chamber of Mines recruiting organizations from the Reserves (and also from territories outside South Africa-up to 50% in the years covered by Table I), and returned to their homes on completion of a term of service of a year or so. From this, and from our knowledge of the general economic situation of the areas from which mine-workers were recruited, it can be inferred that they retained economic (largely consumption) links with their kin in the Reserves.

Table I
African Employment in Mining, Private Industry and the South African Railways and Harbours15



Private Industry

South African Railways and Harbours





































* These figures include Coloured and Asiatic workers as well, numbering probably about 6,000 in each year shown.

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 portion of the means of reproduction of the migrant working-class.

IV. 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 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 Native 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, and the tendency (related to the traditional culture and economy) of Africans to accumulate cattle and thereby overstock the land available, the only possibility of ensuring appropriate levels of agricultural production is through investment by the capitalist sector. As the Tomlinson Commission (1956)16 is later showed, soil conservation measures, irrigation schemes, fencing, mechanisation and agricultural training require heavy capital outlay. Large scale investment, however, (unless it could be met from the resulting surplus, a situation which would itself create other intractable problems since it implies the development of an economically powerful class of African agricultural producers, a retardation of the flow of migrants or, at least, the severance of the economic link between Africans moving into urban industrial employment and the product in the Reserves) would negate the very purpose served by a migrant labour force. That is to say, the effect of large-scale investment in the Reserves would be to make cheap labour-power costly in the sense that the accumulation advantages to capitalism deriving from such labour-power would be lost or reduced if the surplus was utilized in the African rural areas. 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 (1930-32) 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.17 Report No. 9 of the Social and Economic Planning Council (1946) showed, for example, the decline in production of the staple crops-maize and kaffircorn during the period 1934 to 1939. Thus maize production dropped from 3.7 million bags in 1934 to 1.2 million in 1936 and then rose slowly to 3.0 million in 1939. Kaffir-corn likewise declined from 1.2 million bags in 1934 to 0.5 million in 1936, rising to 0.7 million in 1939.

The above-mentioned reports and numerous other studies18 bear witness to the ever-increasing total and irredeemable destruction, through soil erosion, of vast tracts of land, to the decline of production and to the impoverishment of the people to a point at which starvation, malnutrition and disease cause a high rate of death and debilitation. Gervassi (1970: 43) has summed up the situation:

The economic situation of many Africans is now quite desperate. There is good evidence that the standard of living of Africans in the Reserves has actually fallen over the last two decades. The Tomlinson Commission estimated in the mid-1950s that income per head in the Reserves was approximately R. 48 ($67). A recent estimate places per capita income in the Reserves at R. 53 ($74). Since the rise in prices has been steady and rapid, the real income of Africans living outside the cities must have fallen considerably.

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.

The conclusion which emerges is that, overall, production in the Reserves provides a declining fraction of the total subsistence of migrant-labourers.

The level of production, however, is not the only relevant aspect; the way in which the product is distributed must also be considered. The important question is whether or not the family in the Reserve is able to fulfil its obligation of producing agricultural means of subsistence, for consumption in the Reserves, to supplement the wages earned by a member of the family in the capitalist sector? It was pointed out above that in the earlier period of capitalist development, the system of land tenure and the mode of distribution of the product ensured that supplementary means of subsistence were available to the migrant labour force. However, 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.

In the first place, the development of classes in the Reserves (or, perhaps, strata within classes), which had already begun in the nineteenth century, was intensified and broadened. At least three distinct classes have emerged.

Firstly, those who own or occupy land, secondly, those who are both landless and who own no cattle, many of whom appear to live in rural locations and thirdly, landless rural dwellers who own cattle which are grazed on common land.

It is not possible, at this stage, to provide accurate information of the distribution of the Reserve population in these categories, but 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 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 too 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.

Such other figures as are available indicate that the position in Keiskama Hoek is typical of the whole Ciskei. Accepting this, ... two prime facts ... have ... emerged ... (a) That nearly 30% of families are landless. … (b) That over 60% of families own 5 or less cattle including 29% who own none.

The Tomlinson Commission (1956), also reported finding substantial numbers of landless inhabitants of Reserves. In addition the Commission provided evidence of striking inequalities (12.7% of the families earn 46.3% 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 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 urbanisation 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% in 1911 to 23.7% in 1946 and by 1971 was approximately 38%.

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 I figures of African employment in private industry were set out for some years in the period 1910-1940. The changes between 1940 and 1970 are shown in Table II following:

Table 11
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 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 (cf. Hobart Houghton, 1964: 86-7) 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. 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.19 Despite this, however, there can be no doubt that during the period 1910 to 1970 (and particularly in the period during and since World War II) 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.

V. 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 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. Clearly, the nature, form and extent of the conflicts generated by the structural conditions will depend not only upon the measures of state control but on the complex conjuncture of political ideologies and organization, trade unions, the cohesion of the dominant sector, and so on. Although these may vary, what is continuously present, it must be stressed, is the tendency for the structural conditions to generate conflicts, in one form or another, which centre on the system of cheap labour.

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 1940-49 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 1930-1939. 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 under-pinned 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 militance 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 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.20 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 that the reproduction of the labour force, not in general, but in a specific form, in the form of cheap labour-power. Within its framework Apartheid combined both institutionalizing and legitimating mechanism and, overwhelmingly, coercive measures.

It is beyond the scope of the present paper to set out in any detail the structure of coercive control erected by the Nationalist Government. In a fuller account it would be necessary to do this and to show how Apartheid, as a response to the principal contradiction between capital and cheap African labour, ramifies out and penetrates into the secondary contradictions which in turn have, to some extent, a reciprocal effect on the system. It will be sufficient at this point to refer to three aspects of the mainly coercive mechanisms of Apartheid.

It is true, as was pointed out in the Introduction, that Apartheid differs from Segregation in the degree to which it perfects the mechanisms for ordering the Non-white population. This applies particularly to the coercive apparatus of the State.

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 Organisations and Sabotage Acts, etc.), and the building of all-powerful agencies of control-security 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 1925 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.

These facts have been interpreted as meaning that the Government has elaborated and perfected the migrant labour system. Control over residence and movement is clearly one essential element of a system based on a migrant labour force, but it is not the only one. Therefore, to treat the increase in the State’s legal power to declare Africans temporary sojourners in the urban area and to move them as exigencies demand as constituting the ‘modernization’ of the system, without taking account of changes in its economic basis, is insufficient. In the present case it results in the failure to grasp the essential changes in the nature of capitalist exploitation in South Africa. It has been the main contention of this paper that in South Africa the migrant labour-force, properly speaking, did not mean merely a mobile labour force, or a labour force that could be made mobile, that is that could be directed and redirected to where it was required. Above all, a migrant labour-force is labour-force which is both mobile and which has a particular economic basis in the pre-capitalist Reserve economy. With the disappearance of that economic basis, I have argued, the problems of curtailing industrial action and of political control over Africans in the urban areas became extremely acute. That control is exercised, in part, by repressive measures including, importantly, the elaboration of the State’s power over the residence and movement of labour. That is to say, the extension of the State’s power over the residence and movement of the labour force, which adds to the State’s repressive control over it (precisely, one feature of Apartheid) is a function of the economic changes in the Reserves which generate a threat to the cheapness of labour-power.

The pressure for higher wages has also been combated by largely ideological mechanisms. Two of the most important are, on the one hand, the calculation of the minimum wage required by African workers and, on the other, Bantu Education.

In regard to the former it need only be noted that all the calculations (whether made by the South African Institute of Race Relations, industry or academic investigations) are based on the given, historically determined, ‘need’ of African workers calculated on the basis of the existing enforced low standards of living. That is why the minimum found to be ‘required’ by Africans is so many times lower than that required by Whites. In this way low wages for Africans are legitimated.

The pressure on wage levels does not arise only from the economic conditions already discussed but is intensified by the very nature of the changes in the division of labour brought about by the expansion in manufacturing. One of the effects of this expansion is a new and acknowledged demand for African semi-skilled and even skilled labour-power which brings with it higher wages (although still much lower than White wages) and provokes demands for increased wages. The employment of Africans in these occupations results also in the expansion, at the lowest levels, of African education which in the form of Bantu Education’ reproduces through its ideological content the subjection of Africans.

In its application to the urban areas, Apartheid nevertheless 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 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 over-estimated) 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.

Firstly, 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 (compliant) chiefs, subject to the control of the State-indirect rule-was the first (and, at the outset, very conflictual) step in that direction. Secondly, 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. In this the Government accepted, by implication, the contention, emphasized by the Native Laws Commission (1948), that the flow of people off the land and into the urban areas was an economic process and in 1949 the Tomlinson Commission was established to:

conduct an exhaustive enquiry into and to report on a comprehensive scheme for the rehabilitation of the Native Areas with a view to developing within them a social structure in keeping with the culture of the Native and based on effective socio-economic planning (p. xviii).

Nevertheless, the rather modest proposals of the 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. Firstly, 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. Secondly, 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 Native 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 over-arching 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 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, Cols. 6520.)

This was echoed by Vorster in 1968 (Hansard 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’. Already in 1959 the Minister of Bantu Affairs and Development, De Wet Wel stated: (Hansard 1959, Col. 6018.)

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 ...

More and more the term ‘race’ gives way to ‘nation’, ‘ethnic group’, ‘volk’. A fairly recent example appears in the Congress Report (1971), of the National Party’s ideological organization, the South African Bureau of Racial Affairs, in which it was stated that the Bureau had from the beginning proceeded upon the principle that separate development was the only basis for good relations between

... the different races as it was previously referred to, but rather different peoples (volk) as we now put it ...

There is an obvious necessity for this ideological change since a 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 196321 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 i97i the Transkeian Government’s budget was £18 million of which £3½ 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 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% 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. The point has correctly been made that to date most border industries have been established in areas close to the main industrial regions of South Africa including Johannesburg, Pretoria and Durban. This is due to the fact that in remoter border regions the State, in the main, has not provided the necessary infrastructure of transport, communications and so on. But, why decentralize to the borders in any event? One answer has been to suggest that the purpose of border development is to stem the drift of Africans into ‘White’ South Africa. The question is why? 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.

Firstly, 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.

Secondly, 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 been for town-planning. A United Nations Report (No. 26, 1970: 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 slums22 alleviated marginally, if the Transkei is typical, by the allocation of garden allotments for the purpose of the production of vegetables, etc., which, incidentally, will no doubt provide the rationale for lower wages.

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.

VI. 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.

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.




Althusser, L. (1971) Lenin and Philosophy, London, New Left Books.
Arrighi, G. (1970), ‘Labour Supplies in Historical Perspective: A study of the Proletarianization of the African Peasantry in Rhodesia’, Journal of Development Studies, pp. 197-234
Asherson, R. (1969) ‘South Africa: Race and Politics’, New Left Review 53, January-February 1955.
Bundy, C. (1971) ‘The Response of African Peasants in the Cape to Economic Changes, 1870-1910: A Study in Growth and Decay’, unpublished Seminar Paper, Institute of Commonwealth Studies, London.
Bunting, B. P. (1964) The Rise of the South African Reich, London, Penguin Books.
Clenaghen, W. (1972) ‘The State in South Africa’, unpublished Seminar Paper.
Frank, G. A. (1967) Capitalism and Underdevelopment in Latin America, London, Monthly Review Press.
Frank, G. A. (1967) Latin America: Underdevelopment or Revolution, London, Monthly Review Press.
Gervassi, S. (1970) Industrialization, Foreign Capital and Forced Labour in South Africa, United Nations, ST/PSCA/Set.A. /10.
Houghton, D. Hobart (1964) The South African Economy, London, Oxford University Press.
Horwitz, R. (1967) The Political Economy of South Africa, London, Weidenfeld and Nicholson.
Johnstone, R. (1970) ‘White Prosperity and White Supremacy in South Africa Today’, African Affairs Vol. 69, No. 275.
Kuper, L. and Smith, M. G. (1969) Pluralism in Africa, University of California Press.
Laclau, E. (1971) ‘Feudalism and Capitalism in Latin America’, New Left Review, No. 67, May-June, p.19
Legassick, M. (1971) ‘Development and Underdevelopment in South Africa’, unpublished Seminar Paper for the Southern Africa Group, The Royal Institute of International Affairs, Chatham House, London.
Legassick, M. (1972) ‘South Africa: Forced Labour, Industrialization, and Racial Differentiation’, to be published in a forthcoming volume in a series on the political economy of the Third World edited by Richard Harris.
Mayer, P. (1962) Townsmen or Tribesmen, London, Oxford University Press.
Meillassoux, C. (1972) ‘From Reproduction to Production’, Economy and Society, Vol. I. No. I, p.93
Rhoodie, N. J. (1969) Apartheid and Racial Partnership in South Africa, Pretoria, Academica.
Schapera, I. (1947) Migrant Labour and Tribal Life, London, Oxford University Press.
Szeftel, M. (1971) ‘The Transkei Conflict Externalization and Black Exclusivism’, unpublished Seminar Paper, Institute of Commonwealth Studies, London.
Simons, H. J. and R. E. (1969) Class and Colour in South Africa 1850-1950, London, Penguin African Library.
Trapido, S. (1971) ‘South Africa in a Comparative Study of Industrialization’, Journal of Development Studies, Vol.7, No. 3, 309.
Van Der Horst, S. (1971) Native Labour in South Africa, Cass
Van Der Horst, S. (1965) ‘The Effects of Industrialization on Race Relations in South Africa’, in Hunter, G. (Ed.), Industrialization and Race Relations, London, Oxford University Press.
Van Den Berghe, P. (1967) South Africa: A Study in Conflict, University of California Press.
Wilson, M. (1970) ‘The Growth of Peasant Communities’, in Wilson, M. and Thompson, L. (Eds.), The Oxford History of South Africa, Vol. II, Chapter II, London, Oxford University Press.
Wilson, F. (1971) ‘Farming’ in Wilson, M. and Thompson, L. (Eds.), The Oxford History of South Africa, Vol II, Chapter III, London, Oxford University .Press.
Walshe, A. P. (1963) ‘The Changing Content of Apartheid’, Review of Politics, Vol. XXV, 343-61
Wolpe, H. (1970) ‘Industrialization and Race in South Africa’, in Zubaida, S. (Ed.), Race and Racialism, London, Tavistock Publications.
Wolpe, H. (1971) ‘Class, Race and the Occupational Structure in South Africa’, Paper delivered to the World Sociology Congress, September 1970.


  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 most widely used to characterize the present system in South Africa. In this paper I intend to use 'Apartheid' as the generic term to refer to the period (its policies, practices and ideology) since the Nationalist Party took office in 1948-in this sense it subsumes the policy of 'separate development'. 'Racial Segregation' or 'Segregation' is employed throughout, unless the context indicates a contrary intention, to refer to the ideology, policies and practices prior to 1948. It need hardly be added that although 1948 is obviously a year of great importance it is not intended to suggest that in that year Apartheid replaced Segregation.
  3. See for example:
    Transvaal Local Government Commission of 1922 (Stallard Commission) T.P. 1-122
    Native Land Commission (Beaumont Commission) 26-1916
    Native Economic Commission 1930-1932 (U.G. 22, 1932)
    Social and Economic Planning Council-Report No. 9: The Native Reserves and Their Place in the Economy of the Union of South Africa (U.G. 32/1946)
    Report of the Commission for the Socio-Economic Development of the Bantu Areas Within the Union of South Africa (Summary) (U.G. 61/1955) (Tomlinson Commission)
    Native Lands Act 1913
    Native Trust and Lands Act 1936
    Native (Urban Areas) Consolidation Act 1945 and subsequent amendments Bantu Authorities Act 1951
  4. For a critique of this approach see Wolpe (1970).
  5. Althusser's (1971) essay 'Ideology and Ideological State Apparatuses is relevant on this point. Althusser (p. 124) suggests that:
    … in order to exist, every social formation must reproduce the conditions of its production at the same time as it produces, and in order to be able to produce. It must therefore reproduce:
    1. the productive forces,
    2. the existing relations of production.
    The reproduction of both the forces (that is, labour power) and the relations of production are secured for the most part
    ... by the exercise of State power in the State Apparatus, on the one hand the (Repressive) State Apparatus, on the other the Ideological State Apparatus. (Althusser 1971: 141.)
    See also, Poulantzos, N., 'The Problem of the Capitalist State,' New Left Review, November-December 1969, No. 58, p. 67.
  6. Considerable historical material on this point can be found in Simons and Simons (1969).
  7. For a more rigorous and somewhat different treatment see Charles Bettelheim's Theoretical Comments in Emmanuel, A., Unequal Exchange: A Study of the Imperialism of Trade, London, New Left Review Editions, 1972.
  8. For a good outline of this history see Legassick (1972)
  9. 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.
  10. [Where does this endnote go in the text??] While the present paper is restricted to a consideration of African rural economy in the Reserves, to some extent African agriculturalists outside the Reserves on 'White' land had a similar relationship to migrant labour. Both this aspect and, as pointed out to me in a personal communication by M. Legassick, the relationship of Reserve migrants to White capitalist agriculture, would need to be incorporated in a full study.
  11. The explanation of why this occurred rather than an intensification of trade relations adverse to the Reserves is related both to the interest of mining capital in labour rather than in the agricultural product and to the utilisation of State power by Afrikaner farmers vulnerable to competition from Africans. For an elaboration of this aspect see Legassick (1972).
  12. Arrighi (1970) has offered the most complex analysis but in the end his explanation in terms of 'opportunity cost' is tautological.
  13. On this point see the various Commission Reports cited in note (3).
  14. See Commission Reports cited in Note (3).
  15. Adapted from Section G, Union Statistics for Fifty Years, Pretoria, Government Printer, 1960.
  16. See Note (3) for the full reference. The Commission was appointed to enquire into the socio-economic development (sic) of the Reserves.
  17. See also, in addition, to the reports referred to in Note (3), the following:
    Native Laws Commission, (U.G. No. 28/1948)
  18. See Wilson, M. (1971) and Wilson, F. 1971) and references there cited.
  19. One of the few studies dealing with this question is Mayer, P. (1962).
  20. A number of different analyses have been made of the position of different classes in the development of Apartheid. See, for example, Legassick (I972) and Clenaghen (1972). It is not necessary in the present paper, given its concern with the central relationship between White capital and cheap African labour, to pursue this point here.
  21. Other Bantustans are in various stages of formation.