exist only in the concrete economic, political and ideological conditions of social formations’.32 But it is doubtful that Wolpe agreed with Morris’s rejection of the articulation of modes of production; although Morris used the term ‘social formation’, he did not define it theoretically. Mafeje notes that in Wolpe’s schema, on the other hand, the concept refers to specific ‘mechanisms of social reproduction or the “laws” of motion of the economy’. By ‘mechanisms of social reproduction’, Wolpe meant a combination of modes of production. For Wolpe, ‘the distinction between the abstract concept of mode of production and the concept of the real-concrete social formation conceived as a combination of modes of production constitutes the explicit or implicit presupposition of all these articles’.33 Wolpe was referring here to chapters in the collection of essays he edited. To the extent that the distinction between the two concepts is either implicit or explicit in all the essays in the collection, Mafeje reasons that the authors had clearly ignored the view that the term ‘social formation’, even for Marx, can be both an empirical concept that refers to the object of concrete analysis and, as Étienne Balibar puts it, ‘an abstract concept replacing the ideological notion of “society” and designating the object of the science of history insofar as it is a totality of instances articulated on the basis of a determinate mode of production’.34 To this Mafeje adds that one ought to make a distinction between abstract concepts and the concrete referent that needs to be explained.
Having discussed these abstract concepts, Mafeje is keen to analyse the empirical issues they were meant to explain. Broadly speaking, Wolpe’s thesis was that, with regard to the dialectic between urban and rural, the pre-capitalist mode of production ensured (or otherwise maintained) the reproduction of migrant labour. For Morris, it was the labour migration from white farms that maintained the reproduction of the labour-power of migrants. Although farmworkers were allocated pieces of land by the white farmer through the so-called labour-tenancy, they still sought work either on other farms or in the cities. This argument, such as was presented by Morris, that farmworkers could hold more than one job, ‘is tantamount to [saying farmworkers supplement] wage with wage’,35 without giving a clear indication as to the supposed essential difference between the labour tenant on a white farm and his peasant counterpart in the rural areas who uses his labour power for his own subsistence.
This prompts the question as to the real difference between Wolpe’s and Morris’s theses of labour reproduction. For Morris, the difference between feudalist and capitalist relations of production lay in the difference between land rent and wage labour. Morris used the organising concept of ‘relations of real appropriation’ to understand this dynamic, but even so he oversimplified the social relations on white farms. The pre-capitalist South African case need not translate into feudalism by virtue merely of ground rent and landlordism. South African society was polyglot, unlike that of Russia. For Wolpe, the capitalist mode of production was linked to other modes of production, ‘the African redistributive economies and the system of labour-tenancy and crop-sharing on White farms’.36 Mafeje argues that this definition of a mode of production lacks rigour because ‘labour-tenancy’ and ‘crop-sharing’ connote two different things. Nor do they lend themselves to categories of ‘feudalism’ and ‘capitalism’. Thus, when the colonial government ensured that Africans could never rise to the level of capitalist or commercial producers, they fought for their existing pieces of land and grazing rights for their livestock, and they sent family members to the rural areas with the stock they would have accumulated on white farms.
Those who engage in this practice, particularly in the Eastern Cape, are referred to as amarhanuga, those who go around collecting value (in the form of livestock) specifically from white farms, whereas those farm wage labourers uprooted from white farms are called amaqheya. Mafeje asks whether livestock could be thought of as property (means of production) or merely instruments of production. In the South Africa of the time, what did it mean to speak of property with reference to black people? For both Morris and Wolpe, property referred to land in the agricultural economy. Wolpe incorrectly said that land in the rural areas was held communally. Mafeje argues that this is inaccurate because arable land was individually registered at the magistrate’s court, the head of the family accepting liability for annual rent. Strictly speaking, therefore, such land is owned by the state and not the individual. In this sense, peasant cultivators of the land are tenants of the state. The difference between peasants and tenants on white farms (amarhanuga) is that under the ‘quit-rent’ system, the former could pass down their registered plots under customary law. Registered plots could be inherited. This presented a problem for those theorists who talked about ‘communal land’. Any family that had a plot could hold on to it perpetually, as long as they paid rent annually. This surely was no model of communally owned land.
The incorrigibility of the ‘communally owned land’ thesis, as opposed to the redistribution of land through kinship units, persists unabated in spite of the fact that theorists such as Claude Meillassoux and Pierre-Philippe Rey have written persuasively about the ‘lineage mode of production’, to which the notion of prestige goods is critical to an understanding of the social reproduction of lineages.37 Mafeje contends that whether one is talking about white farms or the rural areas, cattle among South African peasants represent prestige goods, rather than property or means of production. As such, cattle are instrumental in lineage reproduction insofar as they facilitate, inter alia, issues of lobola or bride wealth. This invites questions about the role of livestock in subsistence farming and reproduction of labour.
In South Africa, the two do not necessarily coincide. Subsistence is met through cultivation of crops or wage labour. Livestock only entail means of lineage reproduction or are instruments of production. A related point is that what is usually owned communally in South Africa is grazing ground, not arable land as such.38
When Wolpe wrote about ‘a development of classes in the reserves’ he missed the fact that the possibility of such a process was thwarted by the Land Act of 1913, which was compounded when paramount chiefs, through the Bantu Authorities Act of 1951, were given farms as bribes by the state and thereafter bantustan government ministers helped themselves and their cronies to large portions of land. Still, this need not entail a growing land-owning class. Mafeje argues that ‘to conduct class analysis we do not have to invent classes’.39 The general lessons of this discussion are that a theory not adequately sensitive to concrete realities, however progressive, is likely to be as dangerous as its reactionary counterpart. Moreover, although Mafeje is advocating Marxism as the best answer to bourgeois social sciences, he is nevertheless willing to repudiate its categories if they do not accord with concrete realities. In saying this, I am not suggesting that Mafeje was an empiricist. Rather, I am saying that Mafeje took seriously ethnographic detail and the sociology of knowledge.
On the epistemological break and the lingering problem of alterity
Although anthropology has been criticised by a number of scholars located in different parts of the world, it has not yet dispensed with its problem of alterity – the ‘othering’ or the ‘epistemology of subjects-objects’ as Mafeje puts it. Reasons for this incorrigibility of alterity in anthropology go beyond questions of theory to speak to the sociology of knowledge. Knowledge making is as contested as politics. Mafeje says that epistemologies can be changed (though he does not specify the grounds under which such changes occur) and paradigms can be done away with. He is quick to point out that it is dangerous to assume that knowledge is a result of free inquiry. As far as Mafeje is concerned, new knowledge is usually won through struggle. In his discussion paper Studies in Imperialism, he writes: ‘The requirements of social reproduction predicate that every society sanctions only such activities as are consistent with its overall mode of existence. Intellectual enquiry is no exception to this rule.’40 Those who, like Mafeje, identified Eurocentrism in the social sciences have earned themselves labels such as angry, polemical or combative – once they have received these labels, they need not be taken seriously by the academic orthodoxy. Their ‘anger’ is managed by ignoring them. Social scientists who subscribe to the notion of value-free inquiry often miss the point that lived experiences play a crucial role in how one perceives the world and therefore how one constructs knowledge. Indeed, with characteristic eloquence,