In writing the history of the Gbaya-speaking peoples, a group of some half-million population inhabiting east-central Cameroon and the western third of the Central African Republic, we are seldom able to recount the careers of famous individuals, and genealogical charters which link present leaders with long-dead heroes are uncommon. For the most part, in the pre-colonial and early colonial periods, the Gbaya were not organized politically at a level greater than very localised, clan-based groupings, and political leadership was more a matter of competition for short-lived political powers among locally influential men than of formally institutionalised chieftaincy (see Burnham, 1980a: 19 et seq.). However, in some cases during the late pre-colonial period, particularly in the westernmost parts of the Gbaya region, certain leaders began to consolidate their positions as a result of political and economic relations with the neighbouring Fulbe states of Adamawa and their associated Hausa and Kanuri traders (Burnham, 1980b). The names of men like Dogo Lokoti, Bafio and Mbartua (Bertoua) then occasionally figure in the oral histories and colonial documents, and we obtain glimpses of their personalities and careers. It is all the more striking, therefore, when we encounter a Gbaya figure who stands out clearly in the historical record. Without a doubt, Karnu, the inspiration behind the movement that has been called the Karnu Rebellion or the ‘War of the Hoe Handle’ in the Gbaya language, is the most widely known name in Gbaya history.

This article reassesses the accepted model of pre-colonial Tonga (Zambian) society. In a series of books and articles Elizabeth Colson has established an anthropological framework in which to analyse Tonga social and political activities. The basic parameters of that framework would appear to be: ‘Until the beginnings of the colonial period, approximately seventy years ago the largest named territorial unit among the Tonga was the small neighbourhood community.’ Ritual offices existed within the neighbourhood but political office was embryonic or non-existent until the British Government recognized headmen and chiefs and later developed a local council with an appointed civil service. (Colson, 1980b: 35). Social life is characterized as ‘anarchical’ (Colson, 1970a: 87). Politically the Tonga were said to be ‘stateless’ (Colson, 1970a: 207) and ‘amorphous’ (Colson, 1970b: 36). In a paper given in 1968 Colson analyses the concept of tribe, and states that the Tonga only recognized themselves as Tonga when young nationalist schoolboys tried to give them self-awareness. Thus, the Tonga in Colson's view are a creation of British bureaucracy and of a burgeoning nationalism. (Colson, 1968: 202; cf. also Colson, 1970b: 36). More recently the same basic position is reiterated, though the possibility of a change of viewpoint is left open; ‘Even the idea of a social unit of all Tonga is a recent creation and is still likely to be invoked principally in the national political arena, though the continued importance of the shrine of Monze may have political overtones of which I am unaware.’ (Colson, 1977: 137).

Le terme Minyanka est aujourd'hui officiellement consacré mais il n'est probablement qu'une déformation, due aux premiers auteurs européens, de Myanka, d'ailleurs rarement utilisé par les intéressés. Ceux que tout le monde appellent aujourd'hui les Minyanka se donnent eux-mêmes le nom de Bamana, dont la signification populaire est ‘ceux qui refusent le maître’. Par maître, il faut entendre aussi bien Allah, le dieu suprême des musulmans que le maître politique. Bamana désigne le fond de population non islamisée et demeurée fidèle à ses traditions religieuses. L'expression bamana djomo s'applique à la langue minyanka, par opposition au bambara, appelé blo djomo, la ‘langue des esclaves’, parce qu'elle était apprise par tous ceux qui étaient emmenés en captivité par les armées du royaume bambara de Ségou.

In studying class formation in rural Africa, many subtleties of this process are initially elusive, even when the process itself appears to be quite straightforward. The elusiveness of reality frequently results, at least in part, from the assumptions made by the researcher. So, for example, in studying one area of freehold tenure in Zimbabwe, in which ‘squatting’ is officially disapproved and where the land is legally inherited by a single heir, the informal allocation to others of rights to arable land for cultivation has obvious importance in the process of class formation. Less obvious, however (at least to me at the time), was the significance of grazing usufruct.

Johan Pottier. ‘Defunct labour reserve? Mambwe villages in the post-migration economy’, Africa 53(2), p. 8. Below is a version of the original map included in this article correcting errors in the key and location of the Mbala State Ranch.