In the Marxian sense of the term Africa has ‘suffered’ not from being exploited, but from not being exploited enough; not enough capital has been invested; too few Africans have ever been employed productively enough to create relative surplus value; the reinvested surplus has been too small. (Saul and Leys 1999)
Context
Political institutions judged deficient are common media images of sub-Saharan Africa: a landscape of weak if not failed states, tribalism and civil war, and a sharp disjuncture between the formal character of whatever democratic institutions exist and how politics is conducted in practice. The result of this, it is claimed, is the material backwardness of sub-Saharan Africa: a failure to develop. These views are extraordinarily rife. They have, in turn, found their way into more academic understandings. The state and its relation to society emerge as the central problem: the argument is that given the state's weakness and the vacuity of representative institutions we should not expect anything other than developmental failure (e.g. Diamond 1988, Chazan 1994, Chabal 1998).
In this article we argue that this is to get things the wrong way round. Rather, the reason that Africa has weak, undemocratic states is subsequent rather than antecedent to failures of capitalist development; these failures in turn have to do with conditions that even the strongest of states would retreat from rectifying, i.e. the dispossession of the vast majority of the population in the sense of their separation from the land. Over most of sub-Saharan Africa, people enjoy customary rights of possession in land and so are in a position where they can produce their own means of subsistence.1 Accordingly, their desire to work for a wage is very limited. Yet, on the one hand, the historical record is clear: it is only with the creation of markets in both labour and land that capitalist development can take off. And, on the other hand, bringing about that separation would be politically fraught. In the African case, not only would it be a challenge to the prerogatives of chiefs, who have the responsibility of granting access to land to married males, but it would also be resisted by the broad mass of immediate producers for whom land is the means of physical survival, and dispossession is to be thrust into a world of great uncertainty, if not pauperisation. It is thus no accident that incipient capitalism has historically been promoted by authoritarian states such as those in South Korea, Singapore and Taiwan, which are immune from, or disregard, popular pressures.
In the first part of this paper we lay out an argument outlining why it is that the development of stable democratic regimes, along with strong states, requires capitalist forms of development, and why it is that, in turn, the latter require the separation of immediate producers from the means of production. Capitalist development produces a working class, the necessary resistance of which entails its further deepening. This has its political as well as economic moments, as a labour movement develops and pushes for representation at the level of the state, and as the state, furthermore, and in response, strengthens its capacity to act as an essential part of capital's social division of labour. The paper then turns to the African case.
First, we examine three major theories of the state in Africa, pointing out how they lack that strong material base that we believe to be essential if they are to be defensible. In two cases, those of Chabal and Daloz (1999), and Bayart (1993), the state and its relation to the economy is understood in dominantly culturalist terms. In their view, it is the cultural specificity of sub-Saharan Africa that makes the state an ineffective means of promoting development while at the same time giving it superior attractions over the private economy as a means of enrichment. In the third case, that of Mamdani (1996), the approach assigns a privileged causal role to the political: the African state is as it is – weak and undemocratic – because of structures laid down by the European empires. This is Mamdani's so-called bifurcated state. Little is said by Mamdani about the developmental consequences of this structure – whose existence over most of sub-Saharan Africa we do not contest – though there have certainly been consequences.
Second, we point out at length why it is that, pace these theories, and paraphrasing the title of one of the books in question, sub-Saharan Africa does not work; why capitalist development has difficulty taking off and, therefore, why we should not expect strong states and flourishing democracies. Our diagnosis is rooted in the general absence of the necessary preconditions for capitalist development: i.e. the separation of immediate producers from the means of production, in this case the land.
Third, we examine two cases in sub-Saharan Africa, those of South Africa and Zambia. South Africa is the case that proves the rule. With the possible exception of Botswana, it is the one instance where there has been a vigorous process of capitalist development, and the concomitant development of a fairly strong state, along with democratic institutions that are arguably stable; it is also the one country in which the separation of immediate producers from the means of production has been most complete. Zambia, on the other hand, underlines the difficulties of political development where capitalist development is relatively weak and confined to a few enclaves having to do, in particular, with mining.
1. Logics of capitalist development, the state and democracy
The connection between stable democratic government, on the one hand, and ‘development’, on the other, has been a staple of the sociology literature (for example Lipset 1959, Cutright 1963, Cutright and Wiley 1969). Typically, this has been a question of establishing a correlation between, first, some quantified index of the stability of democratic institutions and, second, some measure of development – usually gross national product per capita. Significantly, from the standpoint of the argument in this paper, ‘stable democratic government’ is the ‘dependent’ variable in these studies: the effect of ‘development’, in other words. One can score easy points by pointing to the crudity of these exercises, but the relationship is a strong one and calls out for interpretation. As Rueschemeyer et al. (1992, p. 29) conclude, after a lengthy review of this literature, ‘One massive result of these studies still stands: there is a stable positive association between social and economic development and political democracy.’
Several related claims can be made here. The first is that the conceptions of freedom and equality underpinning the modern democratic state – in both its modernity and its democratic institutions – are inseparable from the practices of generalised commodity exchange.2 Marx himself was clear about it, if anxious to underline the essentially ideological character of the particular meanings of freedom and equality instantiated:
The sphere of circulation or commodity exchange, within whose boundaries the sale and purchase of labour-power goes on, is in fact a very Eden of the innate rights of man. It is the exclusive realm of Freedom, Equality, Property and Bentham. Freedom, because both buyer and seller of a commodity, let us say of labour-power, are determined only by their own free will. They contract as free persons, who are equal before the law. Their contract is the final result in which their joint will finds a common legal expression. Equality, because each enters into relation with the other, as with a simple owner of commodities, and they exchange equivalent for equivalent. (Marx 1976 [1867], p. 280)
Marx's famous quote assumes the production of commodities with commodities: labour power is a commodity so the material practices of bourgeois freedom and equality are universalised. Even so, the proletarianisation of the population did not automatically result in democratic rights. Surplus value was initially extracted in what Marx called its absolute form, i.e. through extension of the working day. Workers' gains in the form of a shorter working day were capital's losses in the form of a reduced surplus and therefore reduced profits. As Fine and Harris make clear:
It can be too readily assumed that the establishment of wage-labour heralds the rights associated with freedom of exchange, but these are rights that have to be won in class struggle. The restriction of those rights best serves the reproduction of absolute surplus value … The (subsequent) systematic organization of the working class around a program for political and economic emancipation promised grave dangers for the bourgeoisie, unless they could be accommodated through a restructuring of capitalist social relations. (Fine and Harris 1979, pp. 113–114; see also Jessop [1990], p. 176 for a similar claim.)
Democratic rights therefore had to be fought for, and capital only yielded once it felt secure. It is here that the arguments of Göran Therborn (1977) become important. Therborn recognises the essential role of the labour movement in the struggle for democratic rights, but also the fact that democratic rights could only be won in a context of real subsumption, and the displacement of surplus value in its absolute form by surplus value in its relative form; i.e. the surplus would be increased, not by extending the length of the working day, but by increasing the productivity of the worker through the capitalist's reorganisation of the labour process. The great fear in the nineteenth century had been that the extension of the franchise would result, through punitive taxation, in the liquidation of capital. However, this was to disregard what Therborn calls ‘the elasticity and expansive capacity of capitalism … Rises in productivity make possible a simultaneous increase of both rates of exploitation and real incomes of the exploited masses’ (1977, p. 30). It is that same rise in productivity, moreover, which lays the basis for strong states with regulatory reach. So-called ‘failed’ states are universally states in which the basis for a sustained process of capitalist development has yet to be laid down and it is no accident that most of these, by common agreement, are located in Africa.
2. The case of sub-Saharan Africa
2.1. Theorising state institutions south of the Sahara
There are three major statements on the state in sub-Saharan Africa: those of Jean-François Bayart (1993), Patrick Chabal and Jean-Pascal Daloz (1999), and Mahmood Mamdani (1996). The first two are of a broadly culturalist persuasion, while Mamdani's understanding is much more statist. They are all informative, if ultimately unconvincing.
Chabal and Daloz focus on the weakness of the African state and a correlative failure of representative institutions. The immediate, though not fundamental, issue for them is a failure to structurally differentiate the state from the rest of society. Rather the state is treated as a private resource by public officials, elective and appointive alike; hence the rampant nature of corruption, nepotism, and downright fraud, as well as the diversion of resources that might otherwise go to building state institutions that would be more effective from the standpoint of economic development. Underlying this is something that they define as cultural, and specific to Africa: the pervasiveness of patrimonialism. State officials retain their positions and their legitimacy through the development of personal networks with clients and these are structured by considerations of kinship, communal affiliation, and ethnicity. They become ‘Big Men’ in return for a distribution of the largesse they are able to divert to themselves through occupying state positions.
Bayart has a similar view of the state in sub-Saharan Africa: a resource to be plundered for private purposes. At the centre of his analysis he places the same sort of patrimonialism that forms a key element in the interpretation of Chabal and Daloz. He hints, furthermore, at a similar cultural interpretation when talking about the Big Men or, in his terms, those with ‘big bellies’ – a play on the sub-title of his book, The politics of the belly: ‘A man of power who is able to amass and redistribute wealth becomes a “man of honor”. In this context material prosperity is one of the chief political virtues rather than being an object of disapproval’ (p. 242), although, as he adds, it has to be shared to be legitimate.
Mahmood Mamdani's interpretation is quite different and centres on what he calls the bifurcated state. What Mamdani offers, contra Chabal and Daloz, is a more political, even politicist, understanding than a cultural one. Under empire, the pattern of European rule was extraordinarily invariant. The colonial territory was divided into two. Indirect rule prevailed in one and direct rule in the other, hence the notion of bifurcation.
In areas subject to indirect rule, the crucial go-betweens were the chiefs who ruled according to some conception, albeit colonial, of the customary or the traditional. Here, the population was tribalised and remains so since this was the criterion according to which the colony as a territory was divided up for purposes of indirect rule and through which claims to land could be made. Tribal tenure was the rule with land allocated by the chief. In consequence, the only way in which market relations penetrated the areas of indirect rule was through migratory labour beyond their boundaries, and through the sale of agricultural products.
In the areas ruled directly by the colonial government, relations were not so much tribalised as racialised. Civil law prevailed and private property was largely in the hands of whites. Whites enjoyed rights of representation in a colonial assembly but Africans were excluded. In some cases this division corresponded to that between the urban and the rural, though in cases like Southern Rhodesia and South Africa, where there were sizeable numbers of white settlers, this distinction was less clear. Even so, Mamdani uses the urban/rural distinction to signal the jurisdictional separation brought about by the bifurcated state.
These arrangements facilitated colonial rule, and not just because indirect rule allowed economies to be made in the use of scarce European personnel. As we have seen, the bifurcated state divided the areas under indirect rule according to tribe. In addition, it introduced tensions between the urban and the rural which impeded the development of a resistance movement. On the one hand, the chiefs would support the continuation of white rule since they were its beneficiaries. Indirect rule provided them with all manner of income – court fees, forced labour and other exactions. On the other, the interests of detribalised Africans in the urban areas, and those of the more tribalised migrant workers from the rural ones, were not necessarily convergent. The migrant workers had limited interest in the struggle to improve African conditions of life in the urban areas since, as migrant workers, they had little to gain from it. African liberation movements were more often than not seen as a threat to their interests. At the same time, they targeted the chiefs as beneficiaries of the ancien régime and therefore elements to be swept away. Accordingly, the problem for the postcolonial state has been, in Mamdani's words, connecting the urban and the rural. Democratic institutions get confined to the ‘urban’ areas, while the ‘rural’, in the form of chiefs who marshal their tribal members through their power to allocate land, resists civil rule. At the same time, the chiefs try to tribalise conflict with the urban, defining the liberation movement as an ethnic one in disguise.
In sum, in both Chabal and Daloz, and Bayart3, development, or rather the lack of it, is seen as a consequence of state institutions. For Chabal and Daloz, the need to nurture networks forecloses the possibility of accumulation. Bayart is a little more guarded. While the gist of much of his argument is the parasitic nature of the state in sub-Saharan Africa, and the diversion of resources away from any serious development initiatives, he also talks about the role played by low population densities as a condition for weak state institutions. This idea, however, is misplaced. Given property relations in sub-Saharan Africa these practices make absolute sense. Where immediate producers remain in possession of the means of production, money cannot do its developmental work. As Marx emphasised, money only becomes capital where there are markets in labour and in the means of production so that workers and the necessary instruments and objects of labour can be brought together. These markets are poorly developed over most of sub-Saharan Africa.
Mamdani is a little more complex. His book bears a strange relationship to the South African case. On the one hand, it is as if that particular example, especially the internecine struggles attending the dissolution of apartheid, were a major incitement to writing the book in the first place. Yet he admits that the strengthening of the structures of the bifurcated state under apartheid, involving the conversion of the former reserves into so-called homelands, were a response to anxieties about the political implications of an African proletariat: permanently settled in cities and severed from its links to the traditional social structures of the native reserves. These are anxieties that make no sense outside of a particularly vigorous process of capitalist development. What he does not say is that the common view in South Africa, and well before apartheid, was that migratory labour was an essential precondition for the profitability of the mining industry (Wolpe 1980), and that, in turn, the native reserves, through their – albeit deteriorating – capacity to provide for the reproduction of workers' families were an essential precondition for the cheapness of migratory labour. This, of course, suggests that far from a particular state structure conditioning development, it was the other way around.
In other words, and in our view, it is the process of capitalist development that should be at the centre of our interpretation when we try to understand political development, not just in Africa, but elsewhere in the world. The obstacles to capitalist development in the subcontinent are the key to its political specificity. It is to an understanding of those obstacles that we now turn.
2.2. The weakness of capitalist development in sub-Saharan Africa
In a nutshell, and as per the quotation at the beginning of this paper:
Africa has ‘suffered’ not from being exploited, but from not being exploited enough; not enough capital has been invested; too few Africans have ever been employed productively enough to create relative surplus value; the reinvested surplus has been too small. (Saul and Leys 1999, p. 17)
Given the fact that over most of Africa immediate producers still enjoy rights of possession in the land, i.e. rights to the use of the land but not to sell or rent it out, which precludes using it as security against credit, we should not be surprised at the difficulties that the accumulation process has faced. This does not mean that there has not been considerable individualisation of land tenure and some impetus towards accumulation through the sale of agricultural products. Bernstein (2007) refers to the emergence of what he calls ‘vernacular’ land markets: rental and sale without formal legal recognition. There are land transactions in the form of gifts, exchanges and rental, as well as sales by the traditional authorities and significantly conflicts over land, regardless of the dominance of tribal forms of tenure, are in the ascendant (Berry 2002).
These conflicts are in part an expression of an accumulation process, though only in part, since the ownership of land is also seen as desirable in political environments that are often chronically uncertain. Nevertheless, accumulation is occurring and Bernstein (2007) and Peters (2004) have drawn attention to the increasing differentiation among farmers that this has entailed. However, there are clearly also limits to this. Since land cannot be offered as collateral it is hard to secure bank financing to expand farming. This may be the reason why, as Bernstein (2007) points out, ‘relative success in labour markets and salaried employment is typically key to the viability (reproduction) of petty commodity production in farming’ (Bernstein 2007, p. 18). In other words, earnings from elsewhere substitute for loan finance. Even then, conflicts over access to land, and continuing customary rights, can and will inhibit investment. Accordingly, the differentiation process to which Bernstein refers is likely to be slow. No one can be forced off the land, so the formation of a wage labour force, except in the form of migratory workers leading a semi-proletarianised existence, is extremely difficult.
This incidentally seems to be the major implication of the extensiveness of indirect rule in sub-Saharan Africa. It froze property relations in pre-capitalist forms over vast areas, limiting, as Mamdani recognised, the penetration of market relations, though he did not draw out its implications for blocking capitalist forms of development. The extensiveness of pre-capitalist property relations, and its implications for capitalist development, however, is obviously not an original claim. Others have made the same point (for example, Boone (1994), Saul and Leys (1994), Leys (1994), and Phillips (1977)). Our contribution has been to show how it sheds light on processes of capitalist state formation.
We should also add that it is in substantial contrast to those views of African development that draw on what can be called, after Brenner, neo-Smithian forms of Marxism which privilege the sphere of exchange. Dependency theory has had its advocates for the African case, as in the work of Walter Rodney (1972). By the same token we would not expect, pace some optimists, that recent waves of Chinese and Indian investment in minerals as well as the surge of investment in oil will bring lasting development outside of new enclaves. The point is that it is difficult to see how they are going to transform existing production relations.
This does not mean that the more global context has not been important, just that it has not been important in the way that dependency theorists suggest and have continued to suggest (Bienefeld 1988, Amin 2006). After all, the fact that sub-Saharan Africa is dominated by social formations of a pre-capitalist nature requires explanation, and the colonial experience is certainly part of that. As Mamdani pointed out, tribal tenure was a very significant aspect of the bifurcated state. In those areas under indirect rule, tribal chiefs had the authority to allocate land, typically to married males; private property rights were excluded.
The story of tribal tenure is a little more complicated, however. It was not just a matter of solving an administrative question where Europeans were spread very thinly on the ground. According to Anne Phillips (1977), in the years leading up to the Second World War, the creation of an African proletariat shorn of its connections to countryside and tribe was something feared by the colonial authorities. In the mind of the colonial administrator, the land question and that of detribalisation were closely connected. In this context, the semi-proletarianisation offered by migrant labour rooted in access to the land was a more attractive one.
Catherine Boone (1994) has come to similar conclusions with respect to postcolonial regimes. Anxieties over the maintenance of political stability have continually trumped the project of an autonomous process of capitalist accumulation. Rather tribal chiefs and clan leaders became absorbed into clientelistic networks of the sort emphasised by Chabal and Daloz, and Bayart, drawing on such resources as foreign aid and the indirect taxation of peasant producers through marketing boards; the latter was especially the case in West Africa. From this angle, Mamdani's thesis assumes further significance. The bifurcated state, and the subsequent incorporation of rural leaders into patrimonial networks, become means through which évolués in charge of central organs of the state can maintain control and, simultaneously, line their own pockets through their control of connections with the world outside: control of import licences, of foreign aid, of concessions to mining companies, even of emergency food aid.4 In this regard, and parenthetically, there is no problem of connecting the urban to the rural: Mamdani's ‘rural’ in the way he describes it is a godsend to the ‘urban’. But, more generally, the colonial experience casts a long shadow over sub-Saharan Africa, though not in the way envisaged by dependency theory.
2.3.1. The case of South Africa
In the context of pressing home these claims, the South African case is in many regards key. First, it is the one country of sub-Saharan Africa that has experienced a strong process of capitalist development and a concomitant proletarianisation of immediate producers. There is still some subsistence production in the former homelands, but it is relatively insignificant. For the vast majority of the population of South Africa, even allowing for the substantial informal sector, working means working for a wage. It is the one part of the subcontinent in which the claims of Chabal and Daloz, and Bayart, that class is virtually absent, are palpably untrue. Second, of all the countries of sub-Saharan Africa, South Africa is the one that can most defensibly claim to have achieved relatively stable representative institutions, and a strong state that has achieved a high level of penetration of civil society. And, despite current anxieties, it has remained relatively free from the sort of parasitism so prevalent elsewhere in Africa; the distinction between state and civil society remains a clear one.
Over sixty years ago, at the dawn of apartheid, the country was already clearly equipped with a state with a deep and far-reaching regulatory power. Much of this was, admittedly, mobilised for seemingly5 racial ends. But it was also a major presence in, among other things, white agriculture; industry through parastatals; city planning; and the creation of a modern infrastructure of railways, airports, water and sewerage provision, and seaports. All of this was made possible by a relatively successful national accumulation process, at least until the late 1970s. To this have been added democratic institutions of seeming stability. South Africans have enjoyed a non-racial franchise since 1994. South Africa is the closest that sub-Saharan Africa comes to a successful modern, stable and democratic state.
It is no accident that this has been built on a strong capitalist base. Dispossession of indigenous people has been a constant in South African history. Through superior access to firepower, and to techniques of governmentality, most of the land surface of what is now South Africa had fallen into white hands by the time it was declared an independent country in 1910. Even so, and significantly for the history of democracy in South Africa and the relative slowness with which it came about, in the areas of white possession there were large numbers of Africans who enjoyed access to the land, albeit mediated by the white landowner. These were the sharecroppers, known also as ‘squatters’, and the labour tenants; the latter provided labour services to the white landowner in exchange for access to some land. Completing the process of primitive accumulation was to be a long and painful process.
Accordingly, the expansion of gold mining that laid the basis for modern South Africa was fraught with obstacles. Money capital could be attracted from outside the country, but creating a wage labour force at the low wage rates necessary if the mining of ore with very low gold content, and at considerable depths, was to be profitable, was not easy. A fundamental element of the solution hit on was migrant labour.6 This was facilitated by taxes on Africans payable in money and, with the passage of time, increasing land hunger in those areas in which Africans still enjoyed access to land through tribal tenure. Through migrant labour, wages could be kept within the tight margins of profitability confronting the industry in South Africa by reducing the miner's reservation wage; the practice of migrant labour did the trick here since the wage demanded did not have to cover dependants who lived off the land in what were coming to be known as the native reserves.7
Nevertheless, and pace Mamdani's interpretive framework, the connections between the urban and the rural were not kept within those confines. There was always permanent urbanisation of the African in South Africa. It was in substantial part the increasing African presence in cities that led, as Mamdani correctly saw, to the attempt of apartheid governments to intensify the structures of the bifurcated state. This, however, was part of a much larger social engineering project that was eventually to be highly consequential for democratisation in the country.
The ultimate goal was to produce a wholly white South Africa, with Africans enjoying citizenship in what were to be independent homelands erected on the geographic basis of the native reserves. This in turn required a huge effort to expel them from the ‘white’ rural areas. There were still very large numbers of Africans on white farms enjoying some sort of access to the land; these were Africans that white farmers-cum-landowners could not yet dispense with. The contribution of the South African government, in what was known as ‘whitening the platteland’, was to complete the process of primitive accumulation and, eventually, the proletarianisation of the African. To reduce the need for African labour on white farms, the government granted subsidies to white farmers for a massive programme of mechanisation. This in turn produced what the apartheid authorities referred to as ‘surplus people’, who were then ‘repatriated’ to appropriate homelands-in-the-making.
The effect, however, was to turn migratory labour into the practice not of semi-proletarians who enjoyed some access to land in the reserves but into full-blown proletarians the families of whom could not survive without the continual wage work of some family members in the city. For, in the massive ‘resettlement’ of Africans into the homelands, there was to be no provision for access to land. This paved the way for the so-called ‘career’ migrant worker (Crush 1989, pp. 19–20; 1995, p. 25) involved in repeated, year-in, year-out labour contracts. This was in stark contrast to what had prevailed hitherto, where, although migrant workers were engaging in longer and longer periods of migratory labour, this labour was still discontinuous in nature. The initial effect of this was to give enormous impetus to the African labour movement and, ultimately, to the anti-apartheid movement. The number of Africans in unions soared in the late 1970s and early 1980s. Deprived of land, ‘resettled’ Africans faced a new horizon of uncertainty. At the same time, the fact that wage labour was now their permanent commitment meant that changing conditions of work assumed a new relevance, if not urgency. And while it is true that the national as opposed to the class pole of the liberation struggle came to dominate, the union movement was of immense significance. Nor was the latter restricted to just the support it gave to the township-based civic organisations, particularly in terms of work stay-aways to put pressure on the government. The union movement secured marked pay increases so that the balance of consumer power began to shift in an African direction: not of dramatic proportions but enough to give Africans firepower in the consumer boycotts which became part of the arsenal of weapons drawn on in struggles with local government in the 1980s, thus sending a message to white small business in particular that the game was up.
This is a history of dispossession, capitalist development and subsequent democratisation repeated nowhere else in sub-Saharan Africa. Over most of sub-Saharan Africa immediate producers remain confirmed in their access to the means of production, with access mediated by the tribe. According to Catherine Boone (2007), some 80% of all arable land in sub-Saharan Africa is currently held in some form of ‘customary’ or non-statutory tenure. Private property in land, which would open the way for social differentiation, technical development and the creation of a wage labour force, remains a distant dream. This does not mean that workers, however limited in numbers, are politically marginal or destined to lose. We outline below, in the case of Zambia, their significant achievements in the historical struggle for democratisation. Yet, the domination of customary access to land and the reproduction of bounded enclaves of capitalist accumulation put tight limits on class politics.
2.3.2. The potentials and limitations of the working class: the case of Zambia
Of the remaining countries of sub-Saharan Africa, Zambia perhaps best illustrates the potentials of, but also the structural limitations facing the working class as an agent of democratisation. Due to the early development of capitalist copper mining, Zambia has had a vibrant working class, if one that still remains a small minority of the immediate producers. Organised class politics led by the unions was instrumental in the Zambian struggle for independence in the early 1960s and in the popular mobilisations that led to the institution of multiparty democracy in 1991, and continues to play an important role in the country's polity. However, the structural constraints to the universalisation of capitalism in the country ensure, among other things, that class is unevenly developed, and the democratising tendencies of working class politics, despite its significant achievements, fulfilled only partially. This section develops this argument through a brief history of working class politics in Zambia.
Historically, the Mineworkers' Union of Zambia (MUZ) has been the best-organised and often the most radical element of the Zambian labour movement. Copper mining was established in the British colony of Northern Rhodesia, as Zambia was then known, at the turn of the previous century, but its large-scale expansion occurred only in the late 1920s; despite a temporary setback resulting from the Great Depression, it accelerated throughout the 1930s. In turn, the first major militant labour actions occurred in the form of the mineworkers' strike in 1935, and then again in 1940. A lot can be understood about labour politics from the comparative trajectories of these two strikes (Henderson 1973, 1975), particularly regarding the replacement of ethnicity by class as the hinge on which workers' politics operated. Very briefly, the 1935 strike was a response to increased poll taxes that every adult African was supposed to pay. After reducing the tax during the Depression to account for the downturn, the administration's decision to raise them again led African workers to strike. The action was called by ‘tribal’ elders, who were the elected leaders of tribe-based worker groupings, a system that viewed Africans through a tribal lens and was, in part, created by mine managers to divide workers along ethnic lines (Rasing 2002 , pp. 79–81). A leading part in the action was played by the Mbeni Dance Society, one of several such groups representing members from particular ethnic groups, which was responsible for conducting rituals and ceremonies in the mining towns. The strike began in Mufulira but soon spread to other towns on the Copperbelt, in part due to the Dance Society networks already present in those places. In 1935, therefore, workers' organisation had a decidedly ethnic tinge.
The 1940 strike, on the other hand, was precipitated by the perceived insufficiency of wages and therefore came to be known as the ‘ten shilling strike’. There were also several more general discontents that helped to spread the strike, including the persistence of the job colour bar, whereby many skilled positions were reserved for white workers. It is interesting to note that the leading role in the 1940 strike was played by the ‘boss-boys’, or the class of native workers picked out for special training, and which assisted European supervisors. Though relatively better off than other native workers, the boss-boys felt the effects of the colour bar most directly, as it prevented their advancement in the workplace despite the fact that they were well trained and experienced. Further, and crucially, the striking workers removed the tribal elders from their erstwhile role as spokespersons for the workers. The comparative trajectories of these two strikes suggest that class became increasingly salient in workers' actions and consciousness in the Copperbelt. Although precipitated by contingent issues, these actions set the basis for future unionisation and for representation in the state which, in the colonial context, translated into a struggle for political independence. In contrast to many other anti-colonial contexts, where labour was not a significant force and rebellions originated in the countryside, the militancy of labour unions paralysed the Zambian economy throughout the 1950s and was crucial to the freedom struggle in the country (Rakner 2003, p. 49). Workers pressed for the overthrow of the colonial state in the expectation that an indigenous state would take the claims of the labour movement more seriously, apart from redressing pressing matters, such as unfair taxes and racial discrimination. Certainly, after independence in 1964, these and other demands of the labour movement were fulfilled. The other side of the coin, however, was that there was considerable pressure on mineworkers to tone down their workplace demands and accept exploitation in the interest of the nation (Larmer 2002, p.104). Worker organisation in the new one-party state created in 1969 was limited to a single, state-recognised labour union, and the labour movement thus came into increasing confrontation with the state, demanding greater freedom.
Workers continued to be politically active throughout the Kaunda-led one-party state. They were in the forefront of the protests against the initial imposition of the International Monetary Fund (IMF)-induced structural adjustment programme (SAP) in 1986 which, among other things, withdrew food subsidies, leading to the overnight doubling of the price of mealie-meal (corn flour). They were also at the forefront of the successful campaign to end the one-party state in the late 1980s (Rakner 2003, p. 63), under the umbrella of the Movement for Multiparty Democracy (MMD).
Nevertheless, and despite their important role in the broader movements placing democratic demands on the state, overall class politics is limited by historical and structural constraints. The relative numbers of proletarianised workers in African countries are low overall because of the limited extent of capitalist expansion. Among workers, the number of unionised workers has witnessed a decline as a result of the structural adjustment programme, which led to massive retrenchments as the state cut back its involvement in the economy. Even during periods of strength, however, class politics has had an uneven geography, and this corresponds to the unevenness of capitalism in Zambia. As in many other countries in sub-Saharan Africa, one finds in Zambia a patchy geography of capitalism. The extractive industries dominate the capitalist economy but are confined to certain enclaves that are intimately connected to global markets and remain mostly independent of the political economy of the countryside. The incorporation of the latter, in turn, occurs as a reserve for labour. The development of the working class, therefore, follows the uneven geography of capital, which means that organised labour as an agent of democratisation is almost exclusively a phenomenon limited to the capitalist enclaves.
Apart from the numerical weakness imposed by this patchiness, it also reproduces the conditions constitutive of the continuing salience of ethnicity and ‘tribalism’ in politics (see Posner 2005), divisions that are used by capitalists to undermine class-based challenges to processes of exploitation. Fieldwork conducted by one of the authors in the North-Western Province in Zambia affirms this assertion. Here, a newly constructed copper mine – the largest in Africa – has set up a system of recruitment for over 1500 unskilled positions that privileges local Kaonde applicants and works through their ‘traditional chiefs’. This tribal system is legitimised by the argument that the Kaonde must benefit from the resources within their ‘homeland’, since they are a tribe that is generally considered ‘backward’ because of the historical exclusion of the region from capitalist development (Crehan 1997).
In contrast, migrants, mainly Bemba-speakers from the more urbanised Copperbelt Province, point to the ‘tribalism’ implicit in the arrangement and appeal instead to the promise of equality latent in their national citizenship. Apart from causing ethnic fissions, this system ensures that the unskilled/skilled division within the workplace is also at the same time a Kaonde/outsider divide, erecting real barriers in the practice of class politics. On the whole, while the labour movement has been a force for democratisation in Zambia, this case, and the unevenness of capitalism more broadly, point to important constraints to the full development of the working class, and with it the potential for universal emancipation.
Conclusion
Inevitably, therefore, the conclusion to this paper has to be rather bleak. There are no shortcuts to development and neither, therefore, are there shortcuts to democracy and away from state weakness. Capitalism has proven itself to be the great motor of development but, in the rather clinical terms of classical Marxism, the dispossession of immediate producers is a necessary precondition and this will generate resistance. A sweeping away of patrimonialism, if that were possible, or of the powers of the ‘traditional authorities’ ensconced in Mamdani's bifurcated state in favour of a far-reaching democratisation, just will not do the trick; if anything, it will be counter-productive, since it will simply reproduce the barriers to the necessary dispossession and to the disruptions and oppressions that are sure to follow in the wake of a subsequent capitalist development process. The recent histories of South Korea and Taiwan are object lessons in this regard: no democratic road for them.
It is a truism that Africa has been the victim of history. But this is so in more subtle and differentiated terms than might be realised. As a result of a global privileging of ideas of human rights and bourgeois democracy, laying down the foundations of capitalist development has become much more difficult. The end of the Cold War has not helped either. While, as Marx argued, freedom and equality were ideas embedded in the practices of commodity exchange from the start, it would take struggle over a lengthy period of time before they became constitutional ideals. Dispossession occurred in South Africa long before there was a United Nations or global media insisting on what seems to us the futile mission of democracy for all, and on spreading such democracy around the world, regardless of relations of production. The process was completed there but perhaps only just in time; the ‘whitening of the platteland’ and concomitant ‘resettlement’ could occur without massive rehearsing in the popular media and scrutiny from the angle of human rights. It also benefited from the support offered to the apartheid state by Western governments more concerned about Communist takeover than the sufferings of literally millions of Africans. The same applies, of course, to the cases of South Korea and Taiwan, where severe limits on democratic rights were also supported. From the African standpoint, Fukuyama's ‘end of history’ must sound like a sick joke; for them, and from the standpoint of that capitalist history that lays the foundations for bourgeois democracy, it never started.
Over most of sub-Saharan Africa, therefore, the immediate future at least of development seems to hold nothing much better than the sort of enclave development we have described in the case of Zambia, and a slow rotting away of the old pre-capitalist relations of production that prevail outside those enclaves. Modernity in the sense of the consumer delights and technical fixes of late capitalist development has penetrated deep into the African hinterlands. As monetisation extends into the hinterlands, intensified by the demands of state taxation, this might release sufficient labour power to provide the basis for some industrialisation. The absence of an agrarian revolution in the countryside, however, will be a severe limit, placing upward pressure on food prices and therefore on wages. There seems no easy way out if sub-Saharan Africa is to acquire those democratic institutions so earnestly desired by most observers, both academic and lay, even while they misconstrue their significance.
Notes on contributors
Kevin R. Cox is in the Department of Geography at The Ohio State University. His interests include the politics of capitalist development and South Africa. Rohit Negi teaches in Global Studies at the University of Illinois at Urbana-Champaign. He is interested in state-capital relations and postcolonial politics in Africa. His dissertation research focused on the politics of the latest mining boom in Zambia.