South Africa: Two economies - or one system of superexploitation

By Patrick Bond
[The following is the introduction to ``Transcending two economies – renewed debates in South African political economy'', a special issue of Africanus, Journal of Development Studies (Vol. 37 No. 2 2007, ISSN 0304-615x). It is republished with permission.The full issue is available for free download at http://www.nu.ac.za/ccs/files/africanus_1.pdf ]

[The first economy] is the modern industrial, mining, agricultural, financial, and services sector of our economy that, everyday, becomes ever more integrated in the global economy. Many of the major interventions made by our government over the years have sought to address this ``first world economy'', to ensure that it develops in the right direction, at the right pace. It is clear that this sector of our economy has responded and continues to respond very well to all these interventions ... The successes we have scored with regard to the ``first world economy'' also give us the possibility to attend to the problems posed by the ``third world economy'', which exists side by side with the modern ``first world economy'' ... Of central and strategic importance is the fact that they are structurally disconnected from our country's ``first world economy''.

-- Thabo Mbeki, ANC Today, 2003.


From day to day it ... becomes clearer that the relations of production in which the bourgeoisie moves do not have a simple, uniform character but rather a dual one; that in the same relations in which wealth is produced, poverty is produced also; that in the same relations in which there is a development of the forces of production, there is also the development of a repressive force; that these relations produce bourgeois wealth, i.e. the wealth of the bourgeois class, only by continually annihilating the wealth of the individual members of this class and by producing an ever growing proletariat.

-- Karl Marx, Capital, 1867.


In the pages [http://www.nu.ac.za/ccs/files/africanus_1.pdf] that follow, a group of South Africa's leading political economists tackle [South African President] Thabo Mbeki’s ``two economies'' thesis, the framework most popularly invoked for contemporary poverty policy in South Africa. In short, poverty can be beaten if sturdy (market-focused) ladders are found between the second and first economy, which unfortunately at present are ``structurally disconnected''.

On at least two earlier occasions [in South Africa], a critical mass of university-based intellectuals gathered in various publications to contest ideas of this sort: the mid-1970s when radicals fought liberals over the relationship between race and class; and the early 1990s when the South African version of the Regulation School was established. Both contributions were flawed, we will see. Since then, there has been a growing sense of the need to revisit and reconstruct old frameworks, in part because of the tremendous upsurge in popular social struggles associated with new types of exploitation. Political economists are late in responding to the challenge, but may now have established the necessary historical, theoretical and applied framework.

To work this out properly, our contributors are all aware, requires a renewed commitment to the underlying intellectual challenge posed systematically in Harold Wolpe’s (1972, 1980) early work (Wolpe's works are online at http://www.wolpetrust.org.za/main.php?include=wolpetexts.html&menu=_menus/dialogue.html ), on the way modes of production were ``articulated'' so as to link cities, mines, plantations and Bantustans. That challenge is now amplified by President Thabo Mbeki’s neo-modernisation argumentation and practice, dressed up as it often is in egalitarian, redistributive garb, reminding us of Wolpe’s own political commitments –- but then departing fundamentally from those. Hence, if the idea of ``talk left, walk right'' (Bond 2006) accurately describes the way South African elites challenge what they term ``global apartheid'', we ask whether the same process is evident domestically in the very conceptualisation of political economy.

The contributors to this particular volume come together as part of a wide-ranging effort to reinvigorate South African political economic theory and analysis. They gathered at the University of KwaZulu-Natal’s Centre for Civil Society (CCS) on 28 February 2006 for a Colloquium on Economy, Society and Nature which included a tribute to Harold Wolpe on the tenth anniversary of his passing.1 In addition to Wolpe, two other great political economists active until 2005 in regional debates -– Guy Mhone and Jose Negrao –- were also feted on 1 March at the Colloquium, as recorded in The Review of African Political Economy (March 2007) by Thandika

Mkandawire (for his cousin Mhone) and by Negrao's widow Sabina Asselle and Joe Hanlon. (Other articles in that Roape drawn from the Colloquium are Gillian Hart's work on articulations reprinted below –- for which we are grateful to the Roape editors for permission –- and analysis of resistance strategies by CCS masters student Prishani Naidoo and by one of our Wits University associates, Salim Vally.) Mhone and Negrao offered original critiques of systemic uneven development in other Southern African settings, a necessary task we have therefore not attempted to replicate in this volume.

Mhone's critique of what he called capitalist ``enclavity'' received special treatment in a recent booklet devoted to his memory, launched in Nairobi at the January 2007 Mhone Memorial Seminar of the International Development Economics Associates (with which Guy was closely associated) by Mkandawire and Indian economist Jayati Ghosh (Bond 2007b). Of extraordinary merit is the inspiring review of Mhone's ideas by Adebayo Olukoshi, contextualised within broader economic development theory. Most important, perhaps, a summary of the prolific contributions by Mhone and Negrao would have to stress the way their own praxis -– the often frictional rubbing of radical ideas against establishment power politics -– contributed enormously to the production of knowledge (Bond 2007a). Their deaths in 2005 represented a tragic loss to political economy and to all who knew and worked with them.

Wolpe, Mhone and Negrao taught us to assess the way social challenges are reflected in the system's actions and reactions. Hence a further rubbing of ideas occurred at the Colloquium when from 2–4 March, CCS joined the Rosa Luxemburg Foundation in celebrating her masterpiece of political economy, The accumulation of capital. For Luxemburg, writing in 1913, the problem of imperialism itself followed the very logic of capital ``superexploiting'' the non-capitalist terrain of the Third World, and South Africa was a classic case. In the edited collection that resulted, The accumulation of capital in Southern Africa, several chapters addressed the historical and contemporary South African economy, including Jeff Guy’s treatment of Luxemburg’s source material; a debunking of two economies by Caroline Skinner and Imraan Valodia; critiques of commodified state services by Greg Ruiters and the Black Economic Empowerment strategy by Leonard Gentle; and activist responses by S’bu Zikode, Salim Vally and Trevor Ngwane (Bond, Chitonge and Hopfmann 2007).

What sets this volume apart are the specifically South African analytical focus, the orientation to practical problems, and the varieties of overlapping critiques of Mbeki’s new dualist analysis and strategy. The pages that follow are grouped into two sets of essays – first, diagnosis and then, policy/politics – which help us to consider economic development in terms of either ``two economies'' or instead, ``superexploitation''.

For Marxists, the idea of superexploitation is often captured by ``permanent primitive accumulation'', in which the initial capitalist strategy of dispossessing non-capitalist spheres -– most famously in land enclosures which forced peasants into a proletarianisation process –- becomes permanent. Superexploitation is a way to understand South Africa’s history of extremely biased accumulation, combining capitalism and non-capitalist sites of work, of life and of nature. This process of ``uneven and combined development'' can be identified in history as integral to the ``original'' moment of capital accumulation considered by Marx as ``primitive'', i.e., in the initial stages when the new mode of production was gathering momentum not on the basis of exploitation at the point of production –- the main point of Das Kapital -– but rather in the superexploitative relations between market and non-market activities.

David Harvey has termed this broader process ``accumulation by dispossession'' (in contrast to accumulation by exploitation at the point of production), and locates it as the dynamic behind the ``new imperialism'', in the spirit of Luxemburg’s early 20th century argument. The capitalist system's recourse to systematic looting has, argues Harvey, emerged time and again during accumulation crises, including at present. He traces the most substantial economic problems in ``overaccumulation'' which set in at the global scale during the 1970s, and which has only been displaced, not resolved, since (Harvey 2003).

That was also the point at which Wolpe himself located modern South Africa's historically high profitability in an articulation of two modes of production, capitalist and pre-capitalist. But the argument arose within a vibrant intellectual context that meant it was neither the first nor last word in understanding social relations. We can consider, next, some other key strands in political-economic analysis prior to the 1970s revival of neo-Marxist political economy.

Precedents for articulation/two-economices debate

Hidden traditions of debate about the South African economy are worth revisiting briefly by way of introduction to the theme of superexploitation. We might begin with uneven and combined development, a concept introduced in Leon Trotsky's 1906 requium for the prior year's failed revolution.2 It was then that Trotsky interrogated the ``backward'' stage of Russia's development, and as a result insisted that ``permanent revolution'' was the category best equipped to deal with a hybrid economy that included both capitalist and pre-capitalist social relations.

Although Johannesburg and Cape Town activist intellectuals soon debated the merits of permanent revolution -– in contrast to the stageist theory popular in the Communist Party [of South Africa] from the late 1920s -– it is striking that the many variants of Trotskyist organising never boasted a major analytical statement that advanced our knowledge of uneven development as an interrelated class, racial, gendered, ecological, geographical and geopolitical process.3 Apart from several subsequent political and trade union oriented works by Trotskyists including Baruch Hirson, Neville Alexander and Alex Callinicos, perhaps the most powerful short political-economic statement in the tradition of permanent revolution was the 1934 ``Draft Thesis on the Native Question'' issued by the Workers Party of South Africa, authored by its main ideologue, Yudel Berlak (1993a:106–114):

The main characteristic of the South African economy as it is today is the exceptionally low level of wages of the unskilled and semi-skilled workers … Because of this intense exploitation of the black workers, the exploitation of the white workers is comparatively much less vigorous … As South Africa is predominantly an agrarian country, the bulk of the population is to be found on the land, engaged in agriculture … But the special characteristic of this peasantry is that it is a landless peasantry … There is no native bourgeoisie in South Africa, and no native bourgeois democratic national movement of any importance in existence … The pauperisation of the natives, the pauperisation of the small white farmers, the Native Problem and the poor white problem, not only hamper but bar the way for the development of the country. There is no future for South Africa, there is no place for industrial development and growth, until the internal need is studied and supplied, the level of internal consumption raised, and the whole internal market systematically developed … It must be made clear to the workers and intelligentsia of South Africa that the Native Problem, the agrarian problem, is their problem, that the liberation of the native is their liberation.

From these insights there followed critiques of the white chauvinist [South African] Labour Party and the Communist Party's ``Native Republic'' political strategy, and a call for a ``South African October''. One reply from the International Secretariat of the Trotskyist movement -– by Ruth Fischer (1993:114–119) in 1935 -– was that ``the point of first importance is smite English imperialism; all other questions are subordinate questions''.4 In a later position paper, Berlak (1993b:157) also addressed the question, ``How does the grip and rule of imperialism manifest itself?'' He answered:

[d]irectly through the finance and money market of the City of London and through their representatives, the banks, indirectly through the Chamber of Mines, Chamber of Industry, and Chamber of Commerce … The clearest manifestation of the rule of finance capital was in the matter of the Gold Standard. Although the majority of the country and the parliament were for remaining on the Gold Standard, the pressure of the City of London and the Chamber of Mines forced South Africa off gold and subjected the ``independent'' South African Pound to the British Pound… The greater this dependence on the gold industry, the more powerful is British finance capital that forced the country off gold; it was British finance capital that forced the coalition upon those who up that time had been deadly enemies (remember Smuts and Herzog); and it is British finance capital that is now forcing fusion. Setting aside the aims of both parties in Fusion [the name given the ruling party from 1934, combining English mining and smaller Afrikaner-dominated agricultural interests], such aims as higher and more stable profits, a settlement of the Native Question, that is, a permanent settlement securing the supply of cheap and yet cheaper labour, eliminating even the existing competition, we can see that by Fusion finance capital will strengthen its position.

Combining these two strands of argument -– pro-land reform and democratisation with anti-imperialism –- another Trotskyist theorist, Moshe Noah Averbach (1993:131–133), provided a unified description (if not theory) of South African social relations in 1936, which is worth citing at length:

It is necessary to grasp that the landlessness of the Africans in particular has flowed from the imperialist policy of creating a migratory African proletariat kept in readiness in vast reservoirs of labour –- the reserves – driven out of these reserves by landlessness, starvation and the poll tax, and controlled in the cities by means of compounds, pass laws, etc. In short, the land question cannot be separated from the question of the way in which imperialism built up by a supply of cheap African labour. Here the land question is not only the problem of fighting against landlordism, but furthermore a problem of fighting imperialism, with its strongholds in the cities. Just as the rural African, in most cases [sic – meaning men], is also a city worker for part of his life, so the land problem is tied up with the problem of the anti-imperialist fight which has its bastions in the big cities of South Africa …

In order to have at hand a ready source of controllable cheap labour, imperialism has deliberately prevented the development of an African peasantry, for such a peasantry would live off the land, would reduce the number of human beasts of burden to be exploited in the mines, factories and on the farms, and slow down or threaten to stop the migration of cheap labour from town and farm to the reserves and back again. Imperialism has uprooted the African tribalist, expropriated the African small farmer, prevented their growth into peasants, extended their landlessness, and kept them in a state of permanent flux between the slave conditions in the cities and starvations conditions on the reserves –- in short, imperialism has created the land question as part and parcel of its mechanism of depriving the non-Europeans of their rights, of their land, of opportunities –- part of its mechanism of the colour bar and segregation and race persecution …

The African was expropriated by sword and fire. Near the end of this process, the imperialists began to industrialise the country and to employ masses of cheap labour on the Natal plantations, in the diamond mines, the gold mines, in the industries connected with these mines, and at the big ports. They used the reserves where the expropriated Africans had been driven as real reserves -– as reservoirs of cheap labour … In the cities the bourgeoisie built up an elaborate system of compounds, passes, and regulations to control the migratory labour from the reserves. To prevent the formation of a stable, hereditary urban proletariat which would become used to the traditional methods of organisation and struggle – trade union and political -– of the city working classes all over the world, the imperialist bourgeoisie segregated the Africans from each other tribally or otherwise, and from city political life by means of compounds, and allowed a drift back to the reserves after some time of slavery in the towns.

Among urban black African workers, intellectual and political figures, there were exceptional speakers in the revolutionary tradition -– for example C.B.I. Dladla, Dan Koza, Isaac Bongani Tabata, T.W. Thibedi –- whose arguments have only sporadically been recorded, but which contributed to powerful critiques of the Stalinist strategy. However, rather than dwell upon analysis, Trostkyists used their energies to organise and forge links with labour movements in the major centres and establish small political parties, though this work ebbed and flowed depending upon the cycles of repression and nationalist hegemony over the broader progressive movement.

It is useful to compare these insights to those of the Left within the South African Communist Party (SACP), especially the theory of ''colonialism of a special type'' (CST). Drafted by Mick Harmel, CST was officially adopted by the SACP during the early 1960s, and represents an internal version of dependency theory. According to the SACP's (1989) most widely circulated analysis, the

South African capitalist state did not emerge as a result of an internal popular anti-feudal revolution. It was imposed from above and from without. From its birth through to the present, South African capitalism has depended heavily on the imperialist centres … It was within a colonial setting that the emerging South African capitalist class entrenched and extended the racially exclusive system to increase its opportunities for profit. The racial division of labour, the battery of racist laws and political exclusiveness guaranteed this … In all essential respects, the colonial status of the black majority has remained in place. Therefore we characterise our society as colonialism of a special type.

From this analysis, it was logical for the SACP to defend a vision of a revolution first of national liberation from internal colonialism, followed by the socialist struggle against capitalism. The SACP's late intellectual leader Joe Slovo also suggested that the there is no ``Chinese Wall'' separating the stages, and that ``dominant ingredients of later stages must already have begun to mature within the womb of the earlier stage''. From this approach, there are still regular calls from SACP intellectuals to hasten the ``National Democratic Revolution'' and for the state to take on a more ``developmental'' character (in other words, interventionist in the East Asian sense), without yet identifying what socialist content is appropriate.

But new generations of radical South African scholars added several other branches of Marxian analysis: Wolpe's articulations of modes of production argument during the early 1970s; neo-Poulantzian ``fractions-of-capital'' analysis during the late 1970s; the concept of ``racial capitalism'' during the early 1980s; the Social History school of the 1980s; and French regulation theory (and ``racial Fordism'') during the late 1980s. It is worth devoting attention to Wolpe's ideas, in view of the current concern about relationships that tie capitalism to non-capitalist social and environmental conditions.

Wolpe's articulation and its aftermath

As the CST framework came under repeated questioning from Left intellectuals (e.g. Callinicos 1988), it was not long before, during the early 1970s, SACP intellectual Wolpe (1972, 1980) applied the idea of articulation of modes of production to argue that South African capitalism required and nurtured the superexploitation that was available by harnessing colonial-style dominance of pre-capitalist modes of production. Explicit evidence for the Wolpe thesis came from the Chamber of Mines, which in official testimony to a government commission in 1944, admitted that a subsidy drawn from pre-capitalist society was crucial to the maintenance their profits: ``The ability of the mines to maintain their native labour force by means of tribal natives from the reserves at rates of pay which are adequate for this migratory class of native, but inadequate in practice for the detribalised urban native, is a fundamental factor of the economy of the gold mining industry''.

Wolpe argued that as capitalism increasingly dominated pre-capitalist society, employers necessarily exploited the cheap labour pool ever more thoroughly, hence the formalisation of migrant labour through formal apartheid Bantustans. Although more and more workers began living permanently in cities near manufacturing jobs, there was still a large supply of migrant labour. From 1948 through the 1970s, 3.5 million people were forcibly removed onto the reserves, which could simply not handle the environmental demands placed on them. From the standpoint of capitalist managers, it was better that the families starve, rather than move to the workplace where the wage-earner would require –- and demand and perhaps win -– higher pay to support their women and children.

Simply put, migrant workers were to be subsidised by their kin in rural areas, such that wages could be kept at ultra-low levels. What Wolpe did not express was how gendered the process became. The migrant ``tribal natives'' did not, when they were young, live under a system that required companies to pay their parents enough to cover school fees, or pay taxes for government schools to teach workers' children. When sick or disabled, those workers were often shipped back to their rural homes until ready to work again. When the worker was ready to retire, the employer typically left him a pittance, such as a cheap watch, not a pension that allowed the elderly to survive in dignity. From youth through to illness to old age, capitalists were let off the hook. The subsidy covering child-rearing, recuperation and old age was provided by rural African women. The central lesson from this crucial aspect of apartheid was that capitalism systematically looted the Bantustan areas, especially women, which supplied such a large proportion of workers. The economic functionality of apartheid was, for Wolpe, a logical and necessary outcome of the post-war development of South African capitalism.5

In Wolpe’s seminal contribution, ``Capitalism and cheap labour-power in South Africa: From segregation to apartheid'', first published in Economy and society, in 1972, he draws an explicit link between race and class:

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 ...

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. 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). As André Gunder Frank and others have shown for Latin America, however, the assumption that different modes of production can be treated as independent of one another is untenable.

In South Africa, the development of capitalism has been bound up with, first, the deterioration of the productive capacity and then, with increasing rapidity, the destruction of the pre-capitalist societies. In the earlier period of capitalism (approximately 1870 to the 1930s), the rate of surplus value and hence the rate of capital accumulation depended above all upon the maintenance of the pre-capitalist relations of production in the Reserve economy which provided a portion of the means of reproduction of the migrant labour force. This relationship between the two modes of production, however, is contradictory and increasingly produces the conditions which make impossible the continuation of the precapitalist 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 industrialisation, 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.

Wolpe's argument has been attacked from various directions, as Ari Sitas, Gillian Hart and David Masondo discuss below. As subsequent research and theoretical arguments showed, there was ample room for contesting Wolpe's chronology and understanding of the dynamics of capitalism.6 But we will return, often, in the pages below, to the power of his original insight into superexploitative processes that structurally connected the economies of mines, fields and factories to the Bantustans.

Meanwhile, international trends in historical materialism -– especially the success of Althusserian and Poulantzian structuralism -– were by the mid-1970s having a major impact on South African political economy. There emerged a fascination with which ``fractions of capital'' controlled the state at particular moments of political change.

Although the various fractions became increasingly blurred by the 1960s as South Africa's big mining finance houses diversified into manufacturing, several leading neo-Marxist researchers identified prior distinctions between capitals in terms of their sector of production (mining, manufacturing or agricultural), their location within the circulation of capital (industrial, financial, commercial, landed), or their ``nationality''(Afrikaner, English-speaking, foreign) (Davies 1979, Davies, Kaplan, Morris and O’Meara 1976, and Kaplan 1976).

The historical argument of the fractions of capital school can be summarised as follows. In 1910 the Union of South Africa was founded, and led by a South African Party representing an alliance of foreign-oriented mining capital and more prosperous capitalist agriculture, and also commercial capital and the incipient industrial bourgeoisie. The National Party/Labour Party ``Pact'' government followed in 1924, combining the interests of small white landowners, local capitalists (especially in manufacturing) and racist workers. Pact policies also supported agricultural capital.

A decade later, in 1934, the United Party was a fusion of the National Party and South African Party, with mining interests increasingly favoured. Mining ties to industrial capital also strengthened at this stage, but agricultural capital eventually deserted the party to support the Herenigde Nasionale Party. The Labour Party and Dominion Party (representing large sugar farmers and petty bourgeoisie) joined a coalition government in 1939. By 1948 a coalition of the Herenigde Nasionale Party and Afrikaner Party won the election on the broad support of capitalist agriculture, non-monopoly industrial and financial capital, the white petty bourgeoisie and white labour. Renamed the Nationalist Party in 1951, it codified the existing set of measures of racial segregation known as apartheid. In the late 1970s Nationalist Party leaders became increasingly ambivalent about their social base in white rural and labour constituencies (who have largely moved to the Conservative Party), while instead supporting reform policies originally advanced by big capital’s Progressive Federal Party (now the Democratic Alliance) (Davies, O’Meara and Dlamini 1986).

Before long, however, the Poulantzian analysis itself came under sharp attack, especially by Simon Clarke (1978). For whereas focusing on fractions of capital highlighted questions of power, the costs of this single-minded focus were excessive: the capital accumulation process was downplayed, capital-labour conflicts dismissed, and thus a sense of necessity and contingency in the development of the social and economic formation diminished.7 (There has been little or no subsequent work in the Poulantzian tradition.)

With an upsurge in protest beginning with the Durban labour movement emerging in 1973, and with the economic slowdown beginning around 1974, radicals' attention turned away from aspects of stability and control, to instability and crisis. The theory of ``racial capitalism'' was invoked to link the political and the economic. At the economic level, Marxist explanations of crisis often fall into three different camps: underconsumption, profit squeeze, and overaccumulation. First, insufficient demand for goods -– or underconsumption -– was described by John Saul and Stephen Gelb in 1981 as follows:

From the late 1960s, the growing saturation of the white consumer market limited not only sales but also the ability of the manufacturing industry to benefit from economies of scale. Since an expansion of the black consumer market was not then contemplated, this made more urgent the state's often reiterated, yet difficult to realise, call for an increase in manufacturing exports.

The implications are morally satisfying: give black consumers more income, they'll consume more, and that will spur the economy. At first blush, this sounds logical both as a way to explain economic stagnation and as a possible solution. But does it really get to the root of the way capitalism works? Marx commented that it is ``sheer tautology to say that crises are caused by the scarcity of effective consumption''. Moreover, notes Charles Meth (1991), ``underconsumption, while always lurking, is not a theory of crisis, nor a permanent hindrance to capital accumulation''.

Tellingly, perhaps, by the early 1990s Gelb (1991) came to reject the underconsumption argument, and predicted that if a post-apartheid government simply attempted a general redistribution of income from rich to poor, it ``would lead quickly to supply bottlenecks and higher levels of inflation eroding any real gains. Furthermore, it would simply exacerbate one of the underlying causes of the crisis, which originated on the supply side of the South African economy''. In other words, boosting the purchasing power of low-income blacks might fuel the economy for a short while, but would do nothing to solve the underlying problems.

A second explanation for the crisis focused on labour. As white privilege became even more extreme during the 1960s -– after black political resistance was decimated -– African labour came under much more intense pressure. On top of newfound labour militancy beginning in 1973, Saul and Gelb also identified the shortage of skilled labour as a crucial weakness created by the apartheid system's colour bar and bantu education policies. These shortages became acute by the early 1970s. The argument here is that economic crisis is rooted in resistance by (and problems of) workers, which led to a ``squeeze'' on corporate profits and a slowdown in growth.

This argument is central to the question of the labour movement's post-apartheid wage demands. The danger of this line of argument should be evident. By the late 1980s, Gelb (1987) had come to the position that rising wages were a key culprit -– an ``originating cause'' -– of the crisis. The mid-1970s' wage demands stemmed from rising inflation (not the other way around) and labour's increased clout in the wake of the 1973 Durban and Johannesburg strikes. Mining industry wages were also possible following the 1971 rise in the gold price. A second round of wage increases in the 1980s coincided with the massive consolidation of trade union organisation.

This explanation for the economic slump is distasteful for many radicals. As Meth put it, there is a ``slight awkwardness about holding an analytical position which may be used to justify an attack on workers''. While class struggle certainly affects profitability –- both on the shopfloor and at the negotiating table over how corporate income is distributed between profits and wages –- such struggles cannot be the essential reasons for capitalism falling into a long-term structural crisis. Normally, in fact, wages fall during a crisis. If worker demands are seen to be a cause of economic crisis, there is a simple capitalist solution: ``wage restraint''.

Gelb (1991) later argued the ``need to lower costs and improve productivity in the existing productive sectors, especially mining and manufacturing. This will require increased investment in new technologies, and/or lower wages at least in real terms. In other words, capital in these sectors needs to be strengthened, that is, profitability restored''.

In contrast, overaccumulation theory, the third approach to capitalist crisis, better explains the need felt during the 1980s for capital to end its support of apartheid, and the need for workers to gain higher wages so as to overcome inherited structural economic bias (against basic needs consumption). But what all such theories required was the confidence to generalise, something that some intellectuals felt increasingly uncomfortable about during the 1980s.

The fractions and racial capitalism perspectives were most harshly criticised by a new school of South African social history which prided itself for looking at society and economy not from the top (state and capital), but from the very lowest levels of the voiceless majority.8 This was directly related to the unsatisfactory character of the various macro-theories that had come before. As Ben Fine and Zav Rustomjee (1996:22) put it in their own very valuable contribution to the debate,

[s]uch fashions have subsequently given way to the new social history, which has drawn its inspiration from the work of E.P. Thompson, in which rural struggles have figured prominently as case studies. Here there is considerable antipathy to reductionism, functionalism, structuralism and many other-isms. Instead is offered rich empirical detail and an emphasis on the many-sided making of history in terms both of the different aspects of struggle and participants in it. But there is precious little explicit theory.

Indeed, no matter how rich and interesting the particularities of the social history case studies proved, they added up to very little that could be generalised for the purpose of answering the larger questions of capitalist development (Morris 1988). The broader theoretical discourse about race and class in South Africa seemed to peak in the 1970s, and with rigorous detailed probing underway in the 1980s in the context of the search for specificity, research into the nature of the mode of production tailed off markedly.

Conveniently, in the late 1980s, the larger questions were again placed on the agenda. It was a time when South Africa's capitalist class demanded, perhaps for the first time, an end to formal apartheid. The reasons for this are closely related to economic stagnation and financial crisis, but what was disconcerting was how dramatically this shook many Marxist theorists who, earlier, so profoundly rejected the liberal thesis that apartheid and capitalism were incompatible. As Gelb (1987) put it, radicals must ``develop a substantial and consistent analysis of capital accumulation which preserves their view of the earlier relationship between apartheid and capitalism, explains the transformation from long run apartheid boom to economic crisis and then analyses the crisis itself''. To that end Gelb introduced ``regulation theory'' to dissect the relative stability of South African capitalism from 1948 through the early 1970s. A new wave of international scholarship had just emerged from France under the rubric of regulation, seeking to explain how post-war ``Fordist'' economies faltered (Jessop 1990).

Capitalism does, of course, manage to generate fairly long periods of growth before its internal contradictions become overwhelming. Describing how capitalism could stabilise itself over a period of several decades was the task that the founder of regulation theory, Michel Aglietta (1979), set for himself in his seminal study of United States economic history. The label for the stability that came of this articulation, in honour of a phrase coined by the Italian Marxist Antonio Gramsci, was ``Fordism'', signifying the symbiotic relationship between mass production and mass consumption (the product of Henry Ford’s assembly line and $5/day wages). As a fully-fledged ``regime of accumulation'', Fordism relied upon intensive kinds of production in which capital ``deepened'', and production became capital intensive with high productivity. There also emerged under Fordism a wide range of social and political institutions. Those that were most important to the American version –- which then served to spread the Fordist regime throughout the advanced capitalist world -– were the Bretton Woods agreement (which stabilised the world financial system under the power of the US dollar); a social contract between big business, big government and big trade unions (which also involved the McCarthyite purge of communists); and a limited but real welfare state (which supported consumption).

South African ``racial Fordism'', as Gelb (1987, 1991) termed it, captures the post-war combination of formal apartheid with industrialisation based on import-substitution: ``As with Fordism in the advanced countries, accumulation in South Africa during this period involved the linking of the extension of mass production with the extension of mass consumption, but in a manner that was restricted on both sides of the equation, as is very familiar''. The expensive imported machinery was paid for by a relatively stable flow of foreign currency provided by mineral exports. Although political turmoil disturbed the economic boom in 1960, growth was relatively secure for at least two decades after apartheid was introduced, and this qualifies as the longest uninterrupted period of prosperity that the country's entire white population had ever had. Even short-term business cycle downturns helped correct imbalances in the system, says Gelb, in a ``reproductive'' rather than destructive way.

But white mass consumption only goes so far -– an entire industrialised economy with South Africa's aspirations could not build on so small a base. Because ``the size of the internal market is the main barrier to this type of accumulation'', ANC economists Maria Ramos and Fuad Cassim (1989) wrote, ``the Fordist model within a domestic economy must be described in terms of the conditions of its interaction with the world economy''.

South Africa's location on the periphery of the world economy gives it certain peculiar characteristics, which Ramos and Cassim call ``peripheral Fordism''. Thus, contrary to Gelb's analysis of racial Fordist regulation, the mass production-mass consumption link occurred between the global and economy and South Africa, and not primarily within South Africa, although at first, ``peripheral Fordism began by producing for a middle class both at home and abroad'', Ramos and Cassim acknowledged. ``To survive, the international) Fordist regime had to relocate to a country where high rates of exploitation existed'', which explains the post-war manufacturing boom in places like South Africa and Brazil. ``But though foreign capital has been crucial in underpinning South Africa's growth, South Africa has been unable to penetrate the world manufacturing market, in particular against the competition of more skilled and better utilisation of labour elsewhere.''

With the crisis in racial Fordism largely understood as a breakdown in the institutional apparatus that regulated capitalist instability (witnessed in the form of 1970s strikes and social unrest, the import of international inflation, and the oscillating gold price), the key task for the regulationists –- whether relying upon internal or international causality -– then became how to stitch together a new set of ``post-Fordist'' institutions and assist in the process of ``kick-starting'' capitalist growth. Wage restraint, productivity quid pro quos, social contracts and even Taiwan-style export-orientation have been advocated by Gelb and progressive economists (many connected to the Economic Trends Group ``Industrial Strategy Project'' of Cosatu [the Congress of Southa African Trade Unions]) who gained inspiration from the post-Fordist discourse.

But here the politics of regulation theory emerged into full view. As four social scientists who work in this tradition -– Avril Joffe, David Kaplan, Raphael Kaplinsky and David Lewis (1993) -– put it in the final paragraph of a key paper, what is required is ``to identify a structured form in which these strategic discussions can be pursued across the spectrum of industrial activity without at the same time becoming swamped in a wider agenda of class conflict''. Controversies over trade union policy –- particularly social contracts and shopfloor flexible specialisation -– began to emerge (Brenner and Glick 1991; Callinicos 1992).9 Meanwhile, as with earlier approaches to neo-Marxist analysis, an historical materialist critique of regulation theory became widespread internationally.

What is most remarkable about this search for theoretical explanation of race and class, of the trajectory of political economy, of macro-micro relations and of gender (O’Meara 1996), among other themes, is that at the point such theory may have been most useful to those engaged in everyday struggle against capitalism during the rise of neoliberalism, in the early 1990s, it evaporated. In the most important 1990s contribution, Fine and Rustomjee (1996) intervened with the idea of ``Minerals Energy Complex’ at the core of South African political economy, but did not elaborate this into a broader position on the evolution of social relations.

The theoretical exercises were, perhaps, so flawed in parts, that it became distinctly unfashionable to theorise about political economy. As Fine and Rustomjee (1996:21) appropriately cautioned,

the relationship between abstract theory and empirical application is not unique to the study of South Africa. But the virulent form taken by its racism within the bounds of a predominantly capitalist economy has cast considerable doubt on the simple expedient of examining South Africa's development in terms of hypotheses derived from ready-made analytical frameworks.

Revisiting the analytical frameworks

Without a substantive challenge from the Left up to that point, Mbeki’s own ready-made analytical framework of two economies –- so reminiscent of modernisation theory, as David Masondo points out in his article -– became the prevailing discourse in government circles by 2005. André du Toit and David Nevens, set out the implications for analysis and policy:

Proponents of second economy discourse do not deny that links exist between the economic mainstream and poor and marginalised people and regions -– indeed, given the abundant evidence of a long history of incorporation and integration, they would be hard-pressed to do so. Rather, they seem to argue either that where disconnection exists, it is intrinsically disadvantageous; or that disadvantage, where it exists, must be due to some form of disconnection. So although the existence of connections is not denied, ``second economy'' discourse predisposes policymakers to seek a ``better integration'' -– which is almost universally understood do be a tighter, closer one -– and leaves untouched the underlying assumption that this larger system into which people need to be integrated will necessary function to their advantage.

This collection picks up the debate over Mbeki's premise in several ways: historically, theoretically, empirically and politically. First, superexploitation is addressed in historical terms by Michael Perelman with reference to the Game Laws of England. This is followed by Ari Sitas's consideration of various feedback effects from Wolpe's legacy, including an agenda for further work.

Next, Gillian Hart compares the South African situation with global debates drawing in ideas of Stuart Hall, Ernesto Laclau and Chantal Mouffe, and Antonio Gramsci, in part to explain how capitalism and nationalism have intertwined to produce the conditions in which a figure like Jacob Zuma flourishes. Masondo then reviews how vulgar Marxism and modernisation theory worked initially in conflict, but ultimately in harmony, to distract attention from more durable race-class debates.

Charles Meth dissects the flawed, biased statistics we are compelled to deal with when considering poverty and inequality. Martin Legassick provides an overview of the overaccumulation crisis evident -– albeit displaced -– within the ``first economy''. Finally, to round off the initial section of this volume on history, theory and empirical information, Du Toit and Neves report on their nuanced research in two settings -– Mount Frere in the Eastern Cape and Khayelitsha in Cape Town -– in which the two economies thesis falters, along with notions of small ``business'' in a context of survivalism.

The second group of articles focuses on politics and policy. Isobel Frye begins by unpacking the hot rhetoric of the two economies. Whether under contemporary circumstances, South Africa can become a genuine ``developmental state'' -– or something intermediate or even just half-hearted –- is addressed in two contributions by Bill Freund and Devan Pillay.

Moving to detailed case studies of the mechanisms of superexploitation, I first provide a specific example of how two economies analysis justifies microcredit interventions within the main macroeconomic programme (the Accelerated and Shared Growth Initiative for South Africa), notwithstanding severe problems already evident in this strategy. Another important state intervention is the expansion of public works, which Melanie Samson debunks as a two-economy solution that also does ``depoliticising'' work. Likewise, the ongoing way in which social reproduction by women subsidises capitalism is illustrated in the case of home-based care, by Nina Hunter. At the workplace, the introduction of ubuntu/botho within the context of racially determined skills shortages is addressed by Simon Mapadimeng, as a warning of the limits of Mbeki's strategy. Ashwin Desai compares a new and surprisingly neoliberal ``Taylorism'' with Mbeki’s worldview, finding them both wanting. Finally, Margaret Legum focuses on the search for a revitalised South African political economy.

As these articles were being prepared for publication, another strong statement of the need to transcend ``two economies'' analysis and practice came from the Congress of SA Trade Unions (2007). One apparent task in March 2007 was to identify the difference between increasingly radical ANC rhetoric regarding forthcoming state intervention to address the two economies problem, on the one hand, and on the other, residual structural forces which prevent working people from benefiting from ruling party patronage:

The Central Executive Committee unanimously observed that the political debate is shifting to the left … [But] Cosatu will not exaggerate gains or drop its guard. The shifts are not qualitative and certainly do not signal a consensus on a new developmental path based on fundamental restructuring of the economy through an active industrial strategy that will lead to massive absorption of the unemployed into the labour market. The shifts are never unambiguous and are full of contradictions. They happen within a conservative economic climate in which it is constantly repeated that the economy is doing well and that we have turned the corner. Some are still saying that the Growth, Employment and Redistribution programme was right all along and that we are now reaping its fruits, with talk about 500,000 new jobs being used to promote glowing optimism. Yet the reality for the working class and the poor is extraordinary levels of unemployment, ravaging poverty and growing inequalities. The 500,000 jobs per annum, whilst representing small progress, in the context of the massive job destruction in the first decade, will not help us realise the modest goal of halving unemployment by 2014. Moreover, most of the jobs are of poor quality, created in sectors renowned for high levels of casualisation and subcontracting such as construction, retail and the hospitality industry. Moreover Cosatu is concerned that these shifts are managed outside of the Alliance and the ANC, but through a team of economists from Harvard and the President’s Investment Council.

Cosatu's leaders summed up the dilemma our own writers face when hearing of R400 billion in forthcoming infrastructure spending, increased access to free basic services, delivery of anti-retroviral medicines [to AIDS sufferers] and the like: ``the question remains: is this a cynical move to occupy a Left position that has been occupied by Cosatu during this year of the ANC conference or is this a genuine shift to the Left?'' One might point out that the state's major resource transfers will benefit the construction companies and other powerbrokers responsible for the Gautrain, Lesotho Highlands Water Project dams, soccer stadia, arms deals, nuclear plants, Coega industrial subsidies and the like. Further evidence for the ``cynical move'' is that ``developmental state'' rhetoric is articulated, still,

within a conservative climate which repeats the notion that the economy is doing well and that our society has turned the corner. This conservative convergence threatens to overshadow and downplay the inequitable nature of the current growth path. As such, there is an ideological battle to maintain a message that it is big business and not the poor who are the major beneficiaries of economic reforms … While [the ANC] remains a broad church, there is a hegemonic conservative bloc, which has attempted to move the movement to the centre left. This is manifest in the hollowing of the internal organisation, poor internal democracy and the locus of the centre of power in the state, changes in the key cadres driving the ANC, and on economic policy. Another key concern is the tendency to use the ANC as an instrument of individual accumulation, whether via the state or through business connections. However the environment is not static and is subject to intense contestation. Therefore the principal task of the working class is to recapture the ANC as a progressive and radical liberation movement …

We may have access to political office but the nature of our democracy is not far-reaching in political and economic terms. Wealth and power still resides with a white minority and for many blacks, opportunities may have opened but are tempered by the glass ceiling at the workplace, unemployment, poverty and inequality. Hence there is still a need for a broader multi-class alliance led by the working class to prosecute the National Democratic Revolution, noting that for the working class the ultimate goal is to achieve socialism.

While entirely sympathetic, our ambition in producing this set of articles is more modest: to critique the mid-2000s turn to the ``two economies'' metaphor by the ruling party's leadership, in part by looking at the theoretical problems it raises, and in part by critically considering practical implications in policy and politics that follow from this framework. While this compels an analysis of South African capitalism that stresses superexploitative relationships within a single system, it is now past time for poor and working people, anti-racists, women, environmentalists and all others arrayed against capitalism to take the further step of forging unity, and ultimately generating a programme to transcend ``two economies'', to deconstruct ``developmental state'' rhetoric by exposing underlying processes of poverty generation and positing workable radical reforms, and to achieve a thorough-going liberation.

Biblography

Aglietta, M. 1979. A theory of capitalist regulation. London: Verso.

Alexander, R.J. 1991. International Trotskyism 1929–1985: A documented analysis of the movement. Durham: Duke University Press.

Averbach, M.N. 1993 [1936]. ‘A comment on Trotsky’s letter to South Africa’. Revolutionary History Vol. 4 (4).

Bond, P. 2006. Talk left walk right: South Africa’s frustrated global reforms. Pietermaritzburg: University of KwaZulu-Natal Press.

____. 2007a. ‘Primitive accumulation, enclavity, rural marginalisation and articulation’. Review of African Political Economy Vol. 111.

____ (ed). 2007b. Beyond enclavity in African economies: The enduring work of Guy Mhone. Johannesburg: Open Society Initiative of Southern Africa, Lilongwe: ActionAid, New Delhi: International Development Economics Associates, Nairobi: University of Nairobi Institute of Development Studies, and Durban: Centre for Civil Society.

____, H. Chitonge and A. Hopfmann (eds). 2007. The accumulation of capital in Southern Africa. Berlin: Rosa Luxemburg Stiftung and Durban: Centre for Civil Society.

____ and A. Desai 2006. ‘Explaining uneven and combined development in South Africa’. In Permanent revolution: Results and prospects 100 years on, edited by B. Dunn. London: Pluto.

Brenner, R. and M. Glick. 1991. ‘The regulation approach: Theory and history’. New Left Review, July–August.

Burlak, Y. 1993a [1934]. ‘Draft thesis on the native question’. Revolutionary History, Vol. 4 (4).

____. 1993b [1936]. ‘Draft thesis on the war question’. Revolutionary History, Vol. 4 (4).

Callinicos, A. 1988. South Africa: Between reform and revolution. London: Bookmarks.

____. 1992. Between apartheid and capitalism. London: Bookmarks.

Clarke, S. 1978. ‘Capital, fractions of capital and the state: “Neo-Marxist” analysis of the South African state’. Capital and Class Vol. 5.

Congress of SA Trade Unions. 2007. Cosatu CEC Statement. Johannesburg, 1 March.

Davies, R. 1979. Capital, state and white labour in South Africa, 1900–1960. Atlantic Highlands: Humanities.

____, R.D. Kaplan, M. Morris and D. O’Meara 1976. ‘Class struggle and the periodisation of the state in South Africa’. Review of African Political Economy Vol. 7.

____, D. O’Meara and S. Dlamini 1986. The struggle for South Africa, Vol. I. London: Zed Press.

Drew, A. 1996. South Africa’s radical tradition: A documentary history. Cape Town: Bunchu Books.

Fine, B. and Z. Rustomjee. 1996. The political economy of South Africa. Johannesburg: University of the Witwatersrand Press.

Fine, R. and D. Davis. 1990. Beyond apartheid: Labour and liberation in South Africa. London: Pluto.

Fischer, R. 1993 [1935]. ‘Concerning some tactical questions in South Africa’. Revolutionary History, Vol. 4(4).

Gelb, S. 1987. ‘Making sense of the crisis’. Transformation Vol. 5.

____ (ed). 1991. South Africa’s economic crisis. Cape Town: David Philip.

Harsch, E. 1990. South Africa: White rule, black revolt. New York: Monad.

Hirson, B. 1990. Yours for the union: Class and community struggles in South Africa 1930–1947. London: Zed Books.

Harvey, D. 2003. The new imperialism. Oxford: Oxford University Press.

Jessop, B. 1990. ‘Regulation theory: Retrospect and prospect’. Economy and Society, Vol. 19.

Joffe, A., D. Kaplan, R. Kaplinsky and D. Lewis 1993. ‘Meeting the global challenge: A framework for industrial revival in South Africa’. Paper presented to the IDASA conference on South Africa’s International Economic Relations in the 1990s, 27–30 April.

Kaplan, D. 1976. ‘The politics of industrial protection in South Africa, 1910–1939’. Journal of Southern African Studies.

Legassick, M. 1974. ‘South Africa: Capital accumulation and violence’. Economy and Society, Vol. 3.

Marx, K. 1967. Capital, Vol. 1. New York: International Publishers.

Mbeki, T. 2003. ‘Steps to end the two nations divide’. ANC Today, 3, 33, 22 August.

Meth, C. 1991. ‘Productivity and the economic crisis in South Africa: A Marxist view’. Unpublished manuscript. Durban: University of Natal.

Mokgatle, N. 1971. Autobiography of an unknown South African. Berkeley: University of California Press.

O’Meara, D. 1996. Forty lost years. Bloomington: Indiana University Press.

Ramos, M. and F. Cassim 1989. ‘Racial Fordism’. Unpublished paper presented to economic conference. Lausanne.

Roux, E. and W. Roux. 1972. Rebel Pity: The life of Eddie Roux. Harmondsworth: Penguin.

Saul, J. and S. Gelb. 1981. The crisis in South Africa. New York: Monthly Review.

South African Communist Party 1989. The path to power. London.

Van der Pijl, K. 1984. The making of an Atlantic ruling class. London: Verso.

Van Onselen, C. 1993. ‘The reconstruction of a rural life from oral testimony: Critical notes on the methodology employed in the study of a South African sharecropper’. Journal of Peasant Studies, Vol. 20(3).

____. 1996. The seed is mine: The life of Kas Maine, a South African sharecropper, 1894–1985. Cape Town: David Philip.

Williams, M. 1975. An analysis of South African capitalism: Neo-Ricardianism or Marxism? Conference of socialist economists’ bulletin, Vol. 4(1). Wolpe, H. 1972. ‘Capitalism and cheap labor power’. Economy and Society, Vol. 1.

____ (ed). 1980. The articulations of modes of production. London: Routledge and Kegan Paul.

Endnotes

1 ___. 1988. Race, class and the apartheid state. Paris, UNESCO. Contributors to this part of the CCS Colloquium on Economy, Society and Nature, on 28 February 2006, included Ann-Marie Wolpe (Wolpe Trust), Michael Perelman (CalState), Ari Sitas (UKZN), Caroline Skinner (UKZN), Imraan Valodia (UKZN), Renato Palmi (UKZN), Sthembiso Bhengu (UKZN), Nina Hunter (UKZN), David Hemson (HSRC), Simon Mapadimeng (UKZN), Isobel Frye (Naledi), Charles Meth (UKZN), Devan Pillay (Wits), Mark Butler (groundWork), Lubna Nadvi (UKZN), Richard Ballard (UKZN), Bill Freund (UKZN), Ashwin Desai (UKZN), Vishnu Padayachee (UKZN), Margaret Legum (SANE), David Masondo (Wits), and Martin Legassick (UWC). That evening, Ike’s Books and CCS hosted a memorable launch of two new books: Amanda Alexander's edited collection of CCS Wolpe Lectures, Articulations, and Dennis Brutus’ political memoire and poetry collection Poetry and Protest (Chicago, Haymarket Press and UKZN Press). Those paying tribute to Brutus included the inspiring radical poets Vonani wa ka Bila (Timbila Collective), Mphutlane Bofelo and Pinky Magwaza (Jubilee SA). From Wolpe Trust, we thank Ann-Marie Wolpe, Peta Wolpe and Tracy Bailey for generous support.

2 This section is expanded upon in Bond and Desai 2006.

3 See the special issue of the journal Revolutionary History entitled Colour and Class: The Origins of South African Trotskyism, 4, 4, Spring 1993 and the invaluable work of Allison Drew (1996). A variety of other books document partial experiences in the Trotskyist political tradition, e.g. Alexander 1991, Fine and Davis 1990, Harsch 1990, Hirson 1990, Mokgatle 1971, and Roux and Roux 1972.

4 Trotsky’s (1971:248-255) own reply included warnings against ``polemical exaggerations arising from the struggle with the pernicious national policy of the Stalinists''.

5 In a subsequent book, Wolpe (1988) backtracked substantially from the earlier position that apartheid was necessary to capitalist development, and he came to suggest that central aspects of their mutual evolution were contingent.

6 The first, and perhaps most powerful critique (focusing on the tenuous link between the Bantustans and cheap labour), was by Michael Williams (1975), while Martin Legassick’s (1974) work on the increasing capital intensity of manufacturing offered a different and in many ways more fertile direction of inquiry.

7 The most sophisticated use of Poulantzian ideas not as fractions but as circuits of capital can be found in Van der Pijl (1984).

8 The greatest work in this tradition is Van Onselen (1996), with theoretical arguments registered in Van Onselen (1993).

9 This is best documented in issues of the South African Labour Bulletin.

[Patrick Bond directs the Centre for Civil Society at the University of
KwaZulu-Natal: http://www.ukzn.ac.za/ccs ]