in Africa should be historicised and contextualised and not reduced to psychological variables like tribalism or the purportedly innate hatred between ethnic groups. It remains the case, of course, that for societies to be considered societies they ought to have some degree of coherence and stability. However, there are no societies without internal divisions and frictions. The issue, for Magubane, is to explain these frictions in depth and contextually (that is, finding their root causes). Magubane maintains that, properly understood, present-day conflicts stem from colonial and imperial rule. The administrative personnel may have changed, but the economic and institutional structures remain.
The problem with pluralist anthropology is to isolate ethnic conflicts and other social features in space and in time.42 For Magubane, the pluralists were reluctant to situate problems in Africa in the wider context of the colonial situation or as an extension of the capitalist metropole. To the extent that African societies were brought together through arbitrary colonial borders, they were robbed of the opportunity to develop organic institutions that would foster unity and solidarity. The notion of pluralism failed to explain the role of governments in denying societies the opportunity to foster organic unity. Because some pluralists considered tribalism to be the source of conflict, they assumed that African societies will always be ridden by conflicts since, in their view, tribalism was the state of nature in Africa. In many respects, the concept of pluralism as was used did not take into account economic and social analysis of Africa – what it did do was to brush aside core issues and make conflict and tribalism seem natural. Ultimately, Magubane observes, this led to the view that these conflicts would sort themselves out or die a natural death.
Magubane further contends that the use of such concepts as tribalism and pluralism in explaining conflicts in Africa was a case of stereotypes prevailing over reality; to understand the true nature of these concepts one had to consider colonial maladministration and neocolonialism. Parochial loyalties existed and at times manifested themselves in ethnic terms, but such loyalties were typically based on ‘perceived material interests by those who exploit them’.43 To the extent that pluralists invoked history, it was only to invoke prejudices, many of which were devoid of analysis of present-day problems in Africa. Pluralists simply appealed to notions of an African as a tribesman in an essentially primitive state. The focus – even when aided by empirical research – was on epiphenomena, not the core socio-historical and structural realities. Epistemologically, pluralist anthropologists, inspired by John Furnivall, misread his argument.44 Magubane argues that ‘despite the limitations of the concept of pluralism as used by Furnivall, among the recent pluralists the concept becomes not only a distortion of the social realities but a despairing philosophy … Pluralism, as used in this sense, covers such disparate social and economic historical formations that it loses validity.’45
Magubane’s central critique of pluralism is that it merely described a multiplicity of ethnic groups within a particular nation state, yet it said very little about the relationship between the said groups – save when they were in conflict. Pluralism on the part of anthropologists only meant ‘multi’ and was never qualified or accompanied by reference to concrete historical situations.
On ethnic groups, ethnic divisions and ethnicity
Mafeje considered that the terms ‘tribalism’ and ‘ethnicity’, typically deployed interchangeably by social scientists, were used as ‘things in themselves’; the terms are ‘illusory and need to be deconstructed and replaced by radical or transcendent thought-categories’.46 In an article on the use of tribalism in Africa, Peter Ekeh notes that ‘while it now appears that the term “ethnic group” has replaced the disparaged concept of “tribe” in African scholarship, there is no clear statement about the relationship between the two – whether, especially, there has been transition from one to the other and whether there is persistent relevance in the previous analysis of tribes for our understanding of ethnic groups in modern Africa’.47 Ethnicity was the successor to tribalism in part because it was considered less offensive. Mafeje claims that the two concepts have the same ideological connotations. The advantage the term ‘ethnicity’ has over the term ‘tribalism’ is that Africans have no objection to its usage, but although ‘ethnicity’ has gained currency among African scholars, this, Mafeje realises, does not explain why it is correlated with the crisis of state power in Africa and elsewhere. He contends that ethnicity may not be what it is presumed to be – as far as he is concerned the term is a metaphor. Although in his classic text The Theory and Ethnography of African Social Formations Mafeje denies the existence of tribalism but not of tribes, in the essay ‘Multi-Party Democracy and Ethnic Divisions in Africa’, while conceding that the idea of ethnicity is a pervasive problem in Africa, he denies that it is attributable to the existence of ethnic groups: ‘In our interrogation, while acknowledging the fact that “ethnicity” has become a pervasive problem in Africa, we will try to dispel the supposition that it is attributable to the existence of a multiplicity of natural units of affiliation called “ethnic groups” within African countries.’ 48
Mafeje returns to classical sociology on the distinction between a social group and a social category. A social group is characterised by necessary patterns of social interaction (a lineage, an association, a religious sect) whereas a social category, although characterised by a common identity, has no necessary or regular patterns of interaction. He believes that the same was true of the so-called ethnic groups, members of the same race, sex or faith. It might come as a surprise, he surmises, but the same is also true of Africa’s political elites. That they are dominant does not mean that they are necessarily a coherent whole, or homogeneous. They are a category consisting of different social factions and in this sense they are a loose category, yet for me there is reason to believe that this would apply to any social group. I believe that it is shared characteristics that make them a group, not face-to-face interaction or homogeneity. For Mafeje, it is continued internecine conflicts among the elites that usually give rise to labels such as ‘tribalism’ or ‘ethnicity’. My view is that Mafeje’s argument limits the problem to the political elite and misses the possibility that contestations and tensions that give rise to jingoism exist at the level of ordinary citizens. Controversially, Mafeje argues that members of the African elite are too loosely organised and their interests too personalised to constitute a class in itself and for itself. The social category of elite is not the same thing as class.
In support of the foregoing claim, Mafeje contends that ‘historically, it is unimaginable that members of a hegemonic class would engage in unbridled mutual extermination and preside over the destruction of their supreme instrument of social control, the state, as has become the order of the day in Africa’.49 If that is the case, it is not clear how one should classify the African elite. Mafeje misses the target in this regard. What makes an elite an elite is its relations to other levels in society – social distance – and its relative size. Sub-divisions and intra-group conflicts are inherent in any social group. I am inclined to think that a distinction has to be made between contestation over control of the state between factions of the elite and situations where the legitimacy of the state and political society itself is at stake. Coherence is not an essential element in the characterisation of a group as a cultural, economic or political elite.
What matters, as was always the case with Mafeje, is to subject concepts to critical scrutiny. No theory or concept is taken for granted in his work. The question thus is whether or not such internecine struggles are necessarily a result of multi-ethnicity in African countries. Mafeje denies such an assumption and argues that the existence of ethnic groups or ethnic existence does not necessarily entail ethnicity. In his own words, ‘existence is not necessarily limited to systems of social classification’.50
Just as he makes a distinction between social groups and social categories, Mafeje makes a distinction between categorical and structural relations. To the extent that ‘systems of social classification are notional and taken for granted by their bearers they are passive and non-binding whereas socially structured relations are not only binding but are also purposeful and dynamic’.51 People from different backgrounds or sociocultural identities can live together in peace without discriminating against one another or exhibiting ethnicity. But ‘in times of structural conflict not between whole categories