This is how the term was used in Robert Rotberg and Ali Mazrui’s collection, Protest and Power in Black Africa, which appeared in 1970. The title of the book was already significant. It substituted ‘protest’ for ‘nationalism’ as the generic name of African resistance, implying that nationalism was only one form of protest. Typically, James Fernandez, a contributor to the Rotberg and Mazrui collection, wanted the adjective ‘nationalist’ only to apply to a movement if it contributed to what he called ‘modernisation’. For this reason, he reserved judgement about the nationalist credentials of the Bwiti movement in central and northern Gabon (Fernandez, 1970). (As an aside, we can note that he worried about the correctness of its application to Mau Mau fighters too [Fernandez, 1970: 454].)
This attention to the relationship between nation and modernisation was accompanied by interest in another couplet: tribe and nation. In From Tribe to Nation in Africa, published in 1970, Ronald Cohen and John Middleton sought to debunk the notion of tribe. They call it a ‘clear-cut racist stereotype’, preferring the term ‘ethnicity’ (Cohen & Middleton, 1970: 2). The notion of ‘tribe’, over and above its growing pejorative connotations, exaggerated the fixity of African social groups, their clear-cut demarcation as separate social units, and the stability of their cultures and customs (Cohen & Middleton, 1970: 2–3). Cohen and Middleton suspected that profound social changes were happening in Africa, but the problem was that they could not be described in the existing academic and popular lexicon. This was especially true of the notion of ‘detribalisation’. The term only made sense, they suggest, ‘[i]f there is a clear-cut, empirically real and, therefore, identifiable entity called a tribe or if there is a real person whom we can label the “tribal” African.’ ‘[T]hen of course’, they continue, ‘there must be a sharp change or loss when the African man or woman does not manifest tribal qualities – hence “detribalization” became a problem for research in social policy’ (Cohen & Middleton, 1970: 2). Yet such characters were more products of colonial fiction than of serious observation of Africans and African societies. Cohen and Middleton witness and also welcome a new form of society in Africa. They call it ‘plural’, and seek a new analytical vocabulary to describe it, which they find in the notion of ‘incorporation’. At stake was the emergence of ‘plural’ societies ‘in which people from differing ethnic backgrounds are interacting to a greater or lesser degree’ (Cohen & Middleton, 1970: 9). ‘The more each group changes toward reduction of boundary maintenance with respect to the other,’ they explain, ‘the greater is the degree of incorporation’ (Cohen & Middleton, 1970: 9). The collection considers this process among the Tonga in Zambia, the Lugbara in Uganda, the Alur in southern Sudan and the non-Nyamwezi in Tanzania. It investigates marriage policy and incorporation in Ghana; in Bornu, among the Mossi; in Rwanda; in the former Transkei; and in Nigeria. What was of especial importance is the effect of independence on the incorporation process (Maquet, 1970: 201), and in particular, the phenomenon of urbanisation. On these terms, ‘a nation’ referred to a plural society.
There are two noteworthy features of this literature. In the first place, it posited the African nation as a particular type of society. Secondly, African nationalism was not treated as a question of identity: it did not matter if its members identified with state institutions or state symbols. This is a recent concern, about which more will be said shortly.
The degree to which a society was plural or industrialised was not a question of subjectivity or opinion. Note, for example, D. F. Malherbe’s Stamregister van die Suid-Afrikaanse Volk (Genealogical Register of the South African People), published in South Africa in 1966. Like the anthropologies of incorporation described above, it researched the limits of the South African nation in terms of intermarriage and genealogy. There are two things to observe. Firstly, it draws a white limit, a racial boundary, despite the apparently all-inclusive term ‘South African people’. There is palpable relief in South African Prime Minister Hendrik Verwoerd’s preface to the book: ‘That the people remained white’, he marvels, ‘in spite of exceptional circumstances, is … remarkable’ (Verwoerd, 1966: v).3 In the second place, the book defines a family limit: it is a chronicle of marriages and intermarriages, of forebears and descendants that sketches the lines of common descent of white South Africans. Therein lies its importance: it is a family tree that establishes the ancestral origins of the people. To Malherbe, membership of the South African nation is not a question of identification; it is one of genealogy. Nor was this a peculiarity of Afrikaner nationalism: if we cast our view wider than Africa, we will hear the nation discussed in similar terms.
Note this extract from the utopian history of Bhudeb Mukhopadhyay. Partha Chatterjee discusses it in the context of an Indian nationalism that defined ‘the people’ as Hindu:
Although India is the true motherland only of those who belong to the Hindu jãti and although only they have been born from her womb, the Musalmans [sic] are not unrelated to her any longer. She has held them at her breast and reared them. Musalmans are therefore her adopted children. Can there be no bonds of fraternity between two children of the same mother, one a natural child and the other adopted? There certainly can, the laws of every religion admit this. There has now been born a bond of brotherhood between Hindus and Musalmans living in India (Chatterjee, 1993: 111 & 222).4
What mattered was the limit of the polis. Yet the metaphor above does not only speak of those whom the nation excludes; more importantly, it describes the character of those whom it takes in as well: they are all family, bonded by relations of fraternal love.
What counts is the character of the state. Three broad themes emerged in this regard. Anthony Smith calls them territorialism, democracy and Pan-Africanism. He means by territorialism a concern, not so much with the limits of the nation’s territory as with who controlled the state (Smith, 1983: 54). European colonisers, he argues, were highly successful in imposing the territorial aspect of the Western state on the African demographic and political map. They were able, not only to map the African political and economic reality, but also to imprint these boundaries in the ‘psychic identity and cultural vision of the new elites’ (p. 54). As a result, the ‘shape’ and ‘face’ of the state, its bureaucratic-territorial form, was already given by the time of independence. On these terms, the African nation is discussed in relation to the form of the state as a bureaucratic-institutional constellation and the particular character of its politics (clientelist, centralised-bureaucratic, ‘spoils politics’5). Smith himself defends this state-centric approach against others. As much as Pan-Africanism (and/or Negritude) informed the nationalist imaginary, furnishing it with the metaphors and images for a vision of the postcolony, giving to it a specific culture, African nationalism was, argues Smith, ‘more firmly grounded’ and ‘prosaic’, lacking the florid poetic fancy of, say, Eastern European nationalisms (p. 57). Over and above their distinctive culture or the specific content of their programmes, African nationalisms, continues Smith, were firstly and foremostly ‘territorial nationalisms’ (p. 55). He states:
It was the colonial state that became the mould as well as the target of African nationalisms, and on them it stamped its special character and aims. It has been the special features of the colonial State – gubernatorial, territorial, bureaucratic, paternalist-educational, caste-like – that have given them its peculiar impetus and shape (p. 56).
The form of the African state has been a subject of continuous interest. The nation is considered an effect of the state itself. The state in Africa or the African state, to recall Bayart’s (1993) own ambivalence, remains one of the key prisms through which Africa is apprehended.6 It has given rise to an extensive literature on the African political economy, exemplified by journals such as the Review of African Political Economy. What matters for this literature today is the relationship between the form of the state (failed, weak, in crisis) and the features of post-independence politics: limited sovereignty; corruption; and ethnic competition for state resources, governance, and so on (Comaroff & Comaroff, 2001).7 Exemplary in this regard is the widely cited article by Chris Allen, to which we shall return shortly. For the moment, we can note that in his article ‘Understanding African politics’, Allen seeks to relate the variability of politics