In a previous article published in this journal analysed one East African folktale in terms of its sociological significance. In this article I present a second, similar tale which deals with the same characters and which provides additional illustration of some of the points I tried to make in my earlier work.

This folktale is in Chikaguru, the language spoken by the Kaguru, a matrilineal Bantu people living in northern Kilosa and Mpwapwa Districts. The Kaguru have cultural traditions similar to those of other matrilineal peoples in this region of Tanganyika, such as the Luguru, Ngulu, Zigua, Sagara, and others. They are organized into about ioo non-localized, exogamous matri-clans (ikungugo or lukolo), each of which is divided into many matrilineages which cannot actually trace common descent. Traditionally, political rights over land and rights to perform certain ceremonies of rainmaking, purification, and fertility descend to men through women. The traditional system has been greatly disrupted and changed by colonial rule, missionary activity, and the introduction of a market economy, but this has little relevance to the significance of the material discussed in this paper, which is concerned primarily with the traditional system.

Despite great linguistic and cultural diversity, sustained political relations among the many different groups of the Jos-Bauchi Plateau are a notably regular feature of this area of Northern Nigeria. That these relations are often expressed in a ritual context is an observation frequently made in the literature on the Plateau Pagans. My intention here is to specify some of the regular ways in which ritual paraphernalia may be manipulated for a variety of secular purposes (often political) among communities of such very different kinds as those of the Plateau mountain people and neighbouring plainsmen. These manipulative techniques appear to be important in the analysis of such widely different phenomena as the adoption of Islam, technological diffusion, the spread of art styles, and, more generally, the successful persistence, through time, of relations between societies of very different levels of complexity and organization. My data are drawn from field experience among Barawa mountain settlements and the Bankalawa–Jarawa plains communities of Dass Independent District, Bauchi Province, on the east-south-eastern slopes of the Plateau.

I do not know what events led up to the sorcery hunt in Savelugu, but in Nanton the sorcerer-finders came on the invitation of the Nantonlana. His invitation was given because, it is said, much sorcery was about. The only certain sign of it that I know is that sorcerer's fire was seen at night. Though I have seen this phenomenon three times I do not know what it is. It travels at thirty to forty feet above the ground at considerable speed, flashing on and off, sometimes as bright as a bicycle lamp. It may be a large firefly, though it does not resemble any firefly I have seen close to. In Konkomba country I thought (though I never saw it there) that it might be an effect of marsh gas. Here in Nanton I do not know what it may be.

The languages dealt with in this paper are Bush ‘A’, ‘B’, ‘C’, and ‘D’, Kwadi, Hottentot, and about 20 Bantu language groups, comprising more than 50 distinct dialects. It is concerned with pre-Bantu history and the Bush, Kwadi, and Hottentot languages, but material on Bantu is included for the following reasons: (a) The information relevant to a discussion of the peopling of Southern Africa by Bantu-speaking peoples is scattered in the available literature or is not available at all, and, (b) Bantu traditional lore has something to say on the subject of pre-Bantu indigenous peoples of Southern Africa, and there must therefore be some evaluation of the relationship of modern and early Bantu languages and an attempt must be made to define their recent and early traditional language areas.