Whatever else may be said of Citizen and Subject, it is a landmark publishing event in contemporary African political studies. Coming out hot on the wake of South Africa's transition to a non-racial democracy, and the continuing political paralysis in most of the rest of the continent, the book has set a new agenda that seeks to answer to the historical origins of the ongoing social and governance problems in South Africa as well as north of the Limpopo. Quite unconventional in its perspective and conclusions, the book argues that the difficulties in South Africa's racially segregated past are writ large in the rest of the continent. It therefore denies South Africa and apartheid any "exceptionalism", embracing the implausible argument that "apartheid was the generic form of colonialism". Citizen and Subject may also represent the first opportunity in a long time that an African scholar has made significant theoretical waves in African studies on both sides of the Atlantic, considering the attention the book has received in South African universities as a whole, and in the US African studies community generally. The book also presents its author in his new theoretical perspective, having all but bade farewell to materialistic interpretations of post-colonial Africa, grounded in class struggles in the neo-colonial context, and embraced the Weberian perspective of authority--what Weber called "possession of the means of administration"--as the ordering factor of social conflict. All this is most refreshing, persuasively argued, and extremely well written. But it remains to be seen, when all the reviewers have had their say, and when all the factual evidence is carefully sifted, whether the book adds value to existing knowledge of the colonial origins of Africa's political predicament and South Africa's new role in it. Despite his having enjoyed reading the book, this reviewer has some strong doubts about that.

The overarching thesis of the book which this writer read, disbelievingly over and again since acquiring his copy in the middle of last year, leaves little ambiguity in the reader's mind. The institutional framework of rule enshrined in apartheid and in all late colonialism, hinged especially on its use of "indirect" rule over the natives by local chiefs using "customary law". This in turn dichotomized African societies into "citizens" (those above the writ of customary laws, enjoying some civil liberties, and mostly white), and "subjects" (primarily peasant households in the countryside) who faced the wrath and arbitrariness of native authorities, chiefs and their retinues. The book proceeds with the assumption--often enshrined in customary law statues--that native authority was coextensive with geographic "tribal" domains. Given the multiplicity of rural native authorities, the system of indirect rule so established is referred to by Mamdani as "decentralized despotism". In Southern Africa at least, white authorities sought to transplant indirect rule into the "native" townships, and to sustain the figment of "tribal" solidarity under chiefly control in the urban setting where black migratory labor was seen as transient, and still grafted to its rural umbilical cord. Somehow, and again difficult to sustain in the light of hard historical evidence, British and South African indirect rule is equated to what Mamdani calls the French colonial policy of "association". At the nadir of the books narrative, the tragedy of African independence is represented as the continent's inability to dispense with decentralized despotism even when rural socialist revolutionary programs were attempted, as they were in Tanzania and Mozambique. With the attainment of independence, and of majority rule in South Africa, the institutional framework was " deracialized but not democratized". This left the African peasantry almost everywhere "trapped in a nonracial version of apartheid". But with one major exception: in South Africa, industrialization had brought Africans in vast numbers to the cities, and in that context, indirect rule and decentralized despotism were an urban affair. So was the opposition to the system. In short, rural protest movements north of the Limpopo are generically identical to the township rebellions in South Africa.

In these circumstances, Citizen and Subject informs us, rural and urban popular resistance to decentralized despotism inevitably took an ethnic form--"tribal" political organization, as Mamdani calls it after dispensing with the quotation marks early in the book, hoping that his readers will understand that he is no apologist for the colonial coinage or the archaic sense of the term. Modern "tribalism" writes Mamdani, signifies the contradiction of (indirect) authority and the resistance it generates. To wit then, far from being reactionary, provincial and backward looking, ethnic-based peoples' resistance (be they rural or urban) "may be emancipatory" in the move toward democratic rule in Africa. As examples of emancipatory backlash against the tyranny of the decentralized despots by the rural peasants, the author describes the long-simmering Rwenzururu uprising in Toro, Western Uganda, the 1950s Mau Mau rebellion in Kenya, postcolonial "simba" revolt in the Congo (now Zaire). All attacked despotic native authority. In South Africa's townships, the revolts of the 1970s onwards aimed at the sham black administration erected by apartheid. The book's empirical data is culled especially from Uganda, but also from Kenya, Tanzania and Zaire, in addition to the large chunks from South Africa. With the partial exception of Nigeria, West Africa and Francophone Africa have only bit roles, if they feature at all. And while the township rebellions in South Africa are themselves symptomatic of emancipatory counter-action, the counter-revolutionary behavior of some hostel dwellers is itself seen as "tribal" action motivated by autocratic native rulers filling a vacuum created by lack of modern secular trade unionism. The books calls for a critical review of the democratic potential of these popular, ethnic-based rural and urban resistance movements, without being romantic about them, as Africanists recast their analytical apparatus to understand the best way out of the current political impasse.

But is the essence of the history of African colonial rule--the seeming genesis of the problem--captured by the metaphor of decentralized despotism and its malcontents? To begin with, it may be prudent not to overemphasize the African novelty of using local rulers to buttress colonial rule for the use of native auxiliaries has been inherent in the definition of colonialism through the ages. The archetypical model of Lugard's policy of using native rulers in the British empire was India, and its political sequel after India's independence in 1947 was very different from the institutional depravity that informs much of Africa today. Indeed, notwithstanding Mamdani's impressively low figures of European officers in African colonial service (even in closely-administered colonies like Kenya), the truth is that most empires have survived on a combination of might and local administrative accomplices. Reviewing the practice in the ancient world, Machiavelli says in The Prince that there are three ways to hold newly conquered lands: " first by devastating them; next by going to live there...; thirdly by letting them keep their own laws, by exacting tribute, and setting up an oligarchy which will keep the state friendly to you". Anticipating the decision of the Victorian colonial office, Machiavelli judged the third alternative--i.e. indirect rule--as the most economical and effective. Depending on the country, African colonial rule in practice combined indirect rule, European settlement and brute force, and its variation across countries and imperial powers--British, French, Belgian, Portuguese and Italian--was more varied than is suggested by a uniform apartheid, "association", or indirect rule. Thomas Hodgkin brings this out most clearly in his classic Nationalism in Tropical Africa, and we know from detailed historical work on comparative colonialism (like that of Michael Crowder) of the substantive differences between British indirect (and sometimes direct) rule, and French-style direct rule with its complements like assimilation, French education in French, African deputies in the Parisian national assembly, replication of territorial administrative circles and prefectures, etc. In fact, association did not become policy until well after the 1944 Brazzaville conference, as a sop for the would be African nationalists during the war. It pays to remark that like under the British, there were exceptions in the French system as well: the Mossi kingdom in the then Upper Volta, and Felix Eboue's installation of the grand chefs in Central Africa come to mind. But all this reinforces the terrific diversity of colonial structures at the grassroots. Indeed, long after independence, there were regions in Africa--like Northern Chad and interior Mozambique--where it was news that the colonialists had departed. The people had never heard of their arrival.

Thus while the nexus of the colonialist and his local agent may have some overall but highly general resonance, it was hardly similar in substance in British territories themselves, let alone French, Portuguese, Italian and Belgian colonies. And it was hardly apartheid in miniature except in the most perfunctory sense: effective foreign conquest requires active local auxiliaries--the interface between foreign and local laws, between citizen and subject is implicit in the definition of colonialism. Strictly speaking, if apartheid and the politics of indirect rule are equated, then apartheid was the norm not just in Uganda, Rwanda, Burundi, and Nigeria, but also in colonial India, Northern Ireland after the seventeenth century, and Native American reservations after the heyday of the US Cavalry. And as an analytical model it has minimal predictive capability since the political consequences in these situations are so divergent.

Neither is the relationship of chief and their subjects as portrayed in Citizen and Subject wholly consistent with the practice of "decentralized despotism" under indirect rule and apartheid. In his memorable 1949 essay, "The Village Headman in British Central Africa", Max Gluckman described the Janus-faced obligations of native rulers at the lower end of the colonial hierarchy. To the extent that he was successful in his duties, the headman ( and the chief) was at once a representative of popular local causes and an enforcer of unpopular colonial directives. The history of local African rulers under British colonial rule is shot through with examples of difficulties in balancing the two, with some chiefs siding with the ruled or turning tables against the colonial order. Against the wishes of the white establishment in South Africa, Khama I of the Bamangwato was an early modernizer, introducing much-needed schools and heath programs to his people. Chiefs and local spiritual leaders founded the first nationalist party in Gabon. Chief Koinange wa Mbiyu in Central Kenya was a powerful influence in the Kikuyu independent schools movement; detained without trial in the Mau Mau years, he died in prison in 1961. The list is long that would prove Gluckman's point.

However, this is less problematic than the recurring portrait in this book of "tribe", "tribespeople","tribalism", and "customary law" as concrete categories of political behavior, with or without quotation marks. Coming in the 1990s and in Southern Africa, of all places, this is surely unforgivable. For nowhere else in Africa have these terms been as severely discredited--in the sense that Mamdani uses them--as in the urban and migrant labor culture in Southern Africa. With the publication of J.Clyde Mitchell's The Kalela Dance in 1959, and the subsequent work of urban African ethnicity at the then Rhodes-Livingstone Institute in Lusaka, it was established that township-based identities bore little resemblance to "traditional" rural "tribes", which were themselves often creatures of the vortex of social and administrative changes introduced by colonialism; a process in which African peoples were creators of their new identities, not the hapless tools of colonial exploitation some left-wing authors claim "tribal" groupings to have been. Thus although as a communal appellation, the ethnic designation of "Nyanja" may have had resonance in the Northern Rhodesia coppperbelt (with urban "Nyanja chiefs" to boot), it was irrelevant in eastern rural parts of the territory, where the "tribe" supposedly originated, and where the operational categories of identity (complete with "tribal" chiefs and "native" laws) were Ngoni, Tumbuka, Chewa and many others. In one of the most articulate renditions of this phenomenon, Crawford Young (in Politics in the Congo), described the emergence of "Bangala" identity in colonial Leopoldville--complete with its ethnic political association--and narrated the surprise of the native chief in northeastern Belgian Congo, the supposed home of the "Bangala", who himself denied any knowledge of a Bangala ethnicity. Over time, we have seen an accumulation of similar ethnographic data with reference to the Tonga, Shangaan, and Tswana (in Southern Africa) as well as the urban identities of Dyula, Yoruba, Hausa, Luhya, Fang, Ugandan Nubians, and so on.

As an old witticism from this literature, African ethnic identity ( indeed all ethnic identity worldwide) is a shifting, multilayered phenomenon that is contextually defined. To contend as Mamdani does on a "conveyer belt" of migrant tribals from rural to urban, all governed by common ethnicity and indirect rule, is simply and factually untenable, however persuasive may be the archival legal statutes underpinning this concept that he quotes. And this is when one wishes Mamdani had not shed his materialistic heritage so readily. In one of the most influential books on this resurgent phenomenon of ethnicity in the 1990s, Benedict Anderson describes new and old nationalities as "imaged communities", a term that applies no less to what Mamdani and others describe, with the best of intentions as African "tribes" and "tribespeople"; terms of course that do not apply to non-African Croats, Serbs, Basques, Chechens, Pathans, Parsees, or just plain Baluchis. Over twenty five years ago, Pierre van den Berge, author of South Africa: A Study in Conflict, appealed for the abandonment of this invidious and meaningless term (tribe) in place of more objective and universal categories. Far from being mere semantics, this is sound advice. The discourse of ethnic identity and national politics can now be heard in Britain, the Russian Republic, Sri Lanka and Australia. That "tribalism" continues to be used so casually for Africans, all objective evidence to the contrary, may say something about the minimal extent by which the perception and study of ethnic movements in Africa has changed. And it testifies to the need for international comparativism of the kind Mamdani rules out in the opening remarks of this book.

One of the most surprising things about African studies at the end of the century is the extent to which some major strands of them have revived (almost unconsciously) the analytical categories that were current in the heyday of "modernization" and "development" theories of the 1950s and 1960s. According to the conventional wisdom of that era, African societies were characterized by conflict between "tradition" and "modernity", with "modernizing elites" created under colonialism, championing the later. And of course there were dissenters who saw strong benefits in using popular traditions and beliefs as a springboard for modernization. With the disappointing results of development in the 1970s, it was argued by dependency writers that the problem lay in attempting to modernize the colonial, European-run economy with then neo-colonial African "petty" or "bureaucratic" bourgeoisie in place instead of overhauling the production system and putting "the people" in control. Now it is starkly stated that with the end of apartheid and one-party rule in the north, the system is still hostage to an indigenous ruling oligarchy installed in the past (like the old "petty" bourgeoisie) that does not incorporate the people in decision-making. Once again, the system has been "deracialized but not democratized". Hence the current efforts to build a countervailing African civil society. For all its attacks on Goran Hyden's dichotomy between the modern capitalists and the traditionalistic "economy of affection", Citizen and Subject bears all the trademarks and the dilemmas of the modernization school and its sequel, of the struggle between the old and new institutions of governance and economic life. There may be nothing wrong with that. In moments of crisis like those in Africa today, it does pay to retrace one's steps in order to chart a better way forward. If the debates sparked by this book enable us to design a clearer path for national governance in Africa, it will have served a greater purpose than its author intended.

Reviewed By Michael Chege
Director, Center for African Studies
University of Florida, Gainesville