The proliferation and overlapping of spiritual beings is a feature of many religions. Saints in popular Catholicism and Vishnu's avatāras in Hinduism represent, in different ways, the possibility of spiritual beings that are simultaneously one and many, clusters or series of manifestations whose inner relationships are often not fully explained. Yoruba òrìṣà are in good company. However, the way in which multiple aspects of gods are made and maintained clearly varies from one religious repertoire to another. I suggest that it is important to look at the means or medium by which fractions of gods are established, in order to understand how the relationships between them are conceived. Yoruba òrìṣà can scarcely be apprehended without taking into account the specific textuality of the oral genres through which they are created, maintained and communicated with.

The system of divination called Ifa is among the most elaborate of African systems of divination and occupies a unique position in what is often called ‘Yoruba traditional religion’. That it is inappropriate simply to regard it as part of Yoruba traditional religion indicates the nature of our problem. For its saliency in Yoruba religion, as that has been conceived by commentators both Yoruba and non-Yoruba since the early nineteenth century, has precisely been because of its capacity to ‘ride’ social change, to detach itself from much of what Muslims and Christians call paganism, and to impose itself on the respectful attention of the modern educated.

African religious history in the nineteenth and twentieth centuries has been dominated by the rapid growth of Islam and Christianity. This has been especially true of the Senegambia region of West Africa, which has witnessed the adoption of Islam by approximately 80 per cent of the region's populace and the development of a small, but influential Christian minority. Among the Diola of the Casamance region of Senegal, Islam and Christianity have both enjoyed rapid growth. The approximately half million Diola, however, include the largest number of adherents of their traditional religion within the Senegambian region. They are sedentary rice farmers and are usually described as acephalous peoples. While Muslims and Christians have been in contact with the Diola since the fifteenth century there were few conversions during the pre-colonial era (Baum, 1986). During the colonial era Islam became the dominant religion among the Diola on the north shore of the Casamance river, and Christianity also attracted a considerable following (Mark, 1985). Among the south shore communities neither Islam nor Christianity became important until after the Second World War. Seeing the increased momentum of recent years, many observers are confident that the south shore Diola will follow the northern example and convert to Islam or Christianity. Louis Vincent Thomas, the doyen of Diola ethnographers, described Diola traditional religion as ‘a false remedy to a very real crisis; fetishism will become a temporary response that will be quickly swept away by another attempt, even larger and undoubtedly more profound: Islam and perhaps we could add, Christianity’ (Thomas, 1967: 225; translations are my own, unless otherwise stated).

In recent decades a scripturalist, anti-Sufi interpretation of Islam has made steady gains in several parts of sub-Saharan Africa. For non-reformers who are confronted with this phenomenon it is easy to consider all active reformers as emerging from the same mould, as turning for inspiration and guidance to the same religious sources, differing only in the intensity of their fervour or commitment. The task of a more scholarly approach to ‘puritan’ Islamic reform, however, is to consider how it is integrated into different social contexts, how it can be used to change or reinforce the social arrangements and institutions of particular groups. This is the general aim of the present article, which considers the social background of factionalism in the emergence of a reform movement among the Songhay of Gao.

In 1938 an African building a house in the city of Ife, the cultural capital of the Yorubas and the mythical cradle of Yoruba civilisation, came upon an extraordinary cache of ancient Nigerian bronzes. In all, at least fifteen bronzes were uncovered in 1938 in a compound only 100 yards from the palace of the Oni of Ife. These bronzes were to prove of great historical and artistic significance. Until that time only two other bronzes had been unearthed in the Yoruba area, and one of those had disappeared, leaving Nigeria only a single original and a replica. In the disposition of the priceless new finds there ensued a tale of intrigue, prevarication, outraged nationalism, and narrow-minded ethnocentricism that drew into its maelstrom the British colonial government of Nigeria, the US Consulate in Lagos, and the USA's Department of State. Although the Ife bronzes, which today reside in a handsome if small museum in the city of Ife, are not so well known as, for example, the Elgin marbles or certain other antiquities taken from the Third World, the controversy surrounding their removal from Nigeria and their eventual return was filled with the same emotion and employed the same arguments heard today over the rightful location of national cultural treasures. The Nigerian dispute is made all the more poignant in that one of the major protagonists was not a money-seeking antiquities dealer, but a young American anthropologist destined to be one of the most astute and sympathetic interpreters of Yoruba culture.