Improvising away from fixed verbal formulae in the four-tablet oracle of sangomas in contemporary Botswana


Wim van Binsbergen

  paper presented at the session ‘Het discours van de expert, orale traditie tussen formule en uitvoeringtradition between formula and ’ (‘the expert’s discourse: oral performanceAssociation of African Studies, 2003 ’), Convenor: Jan Jansen, Netherlands Conferentie perspectives in Africa studies’, 26 ‘Power, Politics and Poetry: Dutch september, Faculteit Sociale Wetenschappen, Pieter de la Courtgebouw, Wassenaarseweg 52, Leiden

e-mail to Wim van Binsbergen

to Wim van Binsbergen's website on African religion | to the Shikana portal

Wim van Binsbergen, Improvising away from fixed verbal formulae in the four-tablet oracle of sangomas in contemporary Botswana



The rich literature available (cf. van Binsbergen, in press, and references cited there) offers ample evidence for the essential formal and generic unity of such major African divination systems as Ifa, Sixteen Cowries (Nigeria, Benin), Sikidy (Madagascar), and Hakata (Southern Africa), all of which can be regarded as transformations of the system of ‘Sand Science’ (‘ilm al-raml) or ‘Sand Calligraphy’ (khatt al-raml) which spread from Abbasid Iraq all over the Islamic world, the Indian Ocean region, and Africa from the late first millennia CE onwards. However, between these various regions, which are up to a few thousand kilometres apart, considerable variation has developed in the physical details of the divination apparatus, in the accompanying interpretative catalogues by reference to which the diviner interprets the formal patterns produced by the apparatus, and in the discursive style through which diviners structure the sessions with their local clients.

            Current research on Ifa, especially such research as is conducted by African academic philosophers (e.g. Abimbola 1976, 1983, 1991; Bewaji 1992, 1994; Eze 1993; Uyanne 1994), tends to attribute to the Ifa interpretative catalogues the following characteristics, among others: the text is orally transmitted, it is yet considered to be fairly immutable, and it is seen as a timeless and constant repository of wisdom – as a philosophical corpus – whose use for divination is almost secondary and an adulteration. The corpus is extremely systematic and elaborate, and whenever it happens to be codified (by practicing diviners in West Africa, Cuba or the South-eastern USA, or by ethnographers such as Bascom) books of hundreds of pages emerge, with a neat internal structure that follows the productive mathematics of the systems 2n-based formalism. This suggests a rather one-way structure for the divinatory session as a therapeutic ritual: the eternal wisdom of the corpus is supposed to speak, through the apparatus and through the diviner’s mouth, with a timeless authority in the light of which the client’s personal, topic concerns can only humbly submit, so as to be radically transformed so as to conform to the available, time-honoured and formalised wisdom texts.

            The older literature from Southern Africa (Bartels 1903; Coertz 1931; de Jager & Seboni 1964; Eiselen 1932; Garbutt 1909; Hunt 1950; Hunt 1954; Hunt 1962; Laydevant 1933; Seboni 1949; Seboni 1949) suggests a situation very similar to this West African model: the Southern African diviner is supposed to learn, for every formal combination the divinatory apparatus yields, a specific praise formulae, so the sum of his knowledge would also be 2n such formulae (here, usually n= 6, so 64). Typically, this older literature was based on decontextualised, in vitro, ethnographic practices (seeking to tap the expert’s knowledge through formal question-and-answer interviews in a setting chosen and determined by the ethnographer), which did not convey the praxeological dynamics of real sessions. Against this background it is surprising to find, in the divinatory practices of sangomas in Botswana today, a great disregard for the conventionalised praises with fixed textual contents. Instead, each of the 16 combination which the oracle can produce, are interpreted by a highly improvising, creative and intuitive practice, in which the conventionalised associations of each combination along a number of divergent dimensions (ancestors, witchcraft, the body, social interaction, death, animals and totems, etc.) are mobilised in a kaleidoscopic fashion, guided more by the emerging conversation between diviner and client hic et nunc – and somewhat on equal terms – , than by the timeless and immutable wisdom attributed to a verbal corpus.

            Historical and sociological reasons will be advanced to interpret the amazing ethnographic contrasts between Botswana today, on the one hand, and, on the other, Southern Africa in the early twentieth century, as well as West Africa.


At the request from the chair, in the oral presentation during the panel, discussion of the above themes was combined with a more general theoretical and methodological remarks on North Atlantic researchers’ study of oral tradition and performance in contemporary Africa. Here the emphasis was on participatory fieldwork, not so much as a means of collecting cognitive information, but as an intersubjective testing ground where the researcher constantly subjects herself to the scrutiny of the host community in order to allow the latter to assess any progress the researcher in making in conceptual, linguistic, and performative cultural learning. Therefore the fact that all papers in this panel deal with a researcher’s apprenticeship towards a local African domain of expert symbolic production, is highly significant and should not be allowed to be obscured under layers of imposed, etic, objectifying scientific rationality. In the apprentice process the many aspects of culture become accessible which are not primarily expressed in words – not so much because the belong to a domain constructed as secret, but because they belong to a domain (comparable to intimacy, corporeality, sexuality) that is too central to experience and too intersubjective than that explicit verbalisation could enhance, rather than diminish, its value. These are themes elaborated in van Binsbergen (2003).

            These remarks are not intended to be dismissive of the etic approaches that dominated at least two of the other papers in the panel; instead, the message is to try and find a balance between such analytic rationality, and a celebration of the emic intersubjectivity around which the original apprenticeship revolves. To underline the importance also of an etic approach, finally an extemporised attempt is made to link what appears to be recurrent themes throughout the panel’s papers: the interplay between the creation of value, the verbal expression, and the financial reward. The female bard, the expert jembe drummer, the Malian sand diviner and the Southern African sangoma diviner all perform and represent a voice from outside the vicissitudes of the ongoing social process between humans. Those themselves involved in this ongoing social process take this outside voice to be eminently valid and inescapable – hence constructions of its legitimation in supernatural terms (the experts’ powers are attributed to ancestors or the high god, the sangomas are even incarnations of ancestors). One almost gets the impression as if the construction, through the experts’ performance, of such an outside focus of comment, exhortation and criticism, and the verbal and financial exchanges that link this focus (and the experts’ work) to the ongoing social process, is in itself much more than a comment – yea, constitutes a very major factor in the social order. Perhaps it is no accident that, of all available social roles, the local format of the apprenticeship towards the local status of expert is extended to the North Atlantic researcher, the outsider who seeks to be an insider and who by her distancing commentary in the form of ethnography also brings out essential aspects of the local social order.


Abimbola, W., 1976, Ifa: An exposition of the Ifá literary corpus, New York: Nok.

Abimbola, W., 1983, ‘Ifa as a body of knowledge and as an academic discipline’, Journal of Cultures and Ideas, 1: 1-11

Abimbola, W., 1991, ‘Poesie VI: Aus “Sechzehn große Gedichte aus Ifa”‘, in: Kimmerle, H., ed., Philosophie in Afrika: Afrikanische Philosophie: Annäherungen an einen interkulturellen Philosophiebegriff, Frankfurt am Main: Qumran, pp. 226-234; excerpt from: Abimbola, W., ed., 1975, Sixteen great poems of Ifa, no place: UNESCO

Bartels, M., 1903, ‘Die Würfelzauber südafrikanischen Völker’, Zeitschrift für Ethnologie, 35: 338-378.

Bascom, W., 1969, Ifa divination: Communication between gods and men in West Africa, Bloomington: Indiana University Press.

Bewaji, J.A.I., 1992, ‘A critical analysis of the philosophical status of Yoruba Ifa literary corpus’, in: H. Nagl-Docekal & F. Wimmers, eds., Postkoloniales Philosophieren Afrika, vol. 6, Wien: Oldenburg, p. 140-154

Bewaji, Tunde., 1994, ‘Truth and ethics in African thought: A reply to Emmanual Eze’, Quest: Philosophical Discussions, 8, 1: 76-89

Coertze, P.J., 1931, Dolosgooiery in Suid-Afrika, Stellenbosch: University of Stellenbosch, Annale van die Universiteit van Stellenbosch, jaargang ix, reeks B, afl. 2.

de Jager, E.J., & O.M. Seboni, 1964, ‘Bone divination among the Kwena of Molepolole district, Bechuanaland Protectorate’, Afrika und Übersee 48: 2-16.

Eiselen, W.M., 1931, ‘Voorwoord’, in: Coertze, P.J., 1931, Dolosgooiery in Suid-Afrika, Stellenbosch: University of Stellenbosch, Annale van die Universiteit van Stellenbosch, jaargang ix, , reeks B, afl. 2, 1 ongenummerde pagina.

Eiselen, W.M., 1932, ‘The art of divination as practiced by the Bamasemola’, Bantu Studies, 6: 1-29, 251-263.

Eze, E., 1993, ‘Truth and ethics in African thought’, Quest: philosophical Discussions, 7, 1: 4-18

Garbutt, H.W., 1909, ‘Native witchcraft and superstition in South Africa’, Journal of the Royal Anthropological Institute, 39: 530-558 + 2 pages of illustrations.

Hunt, N., 1950, ‘Some notes on witchdoctor’s bones’, NADA (Southern Rhodesia Native Affairs Department Annual), 27: 40-46.

Hunt, N., 1954, ‘Some notes on witchdoctor’s bones’, NADA (Southern Rhodesia Native Affairs Department Annual), 31: 16-23.

Hunt, N., 1954, ‘Some notes on witchdoctor’s bones’, NADA (Southern Rhodesia Native Affairs Department Annual), 31: 16-23.

Hunt, N., 1962, ‘More notes on witchdoctor’s bones’, NADA (Southern Rhodesia Native Affairs Department Annual), 39: 14-16.

Laydevant, F., 1933, ‘The praises of the divining bones among the Basutho’, Bantu Studies, 7, 4: 341-373.

Seboni, M.O.M., 1949, Maboko — Maloba le Maabane, Johannesburg: Via Afrika;

Seboni, M.O.M., 1949, Rammone wa Kgalagadi, Johannesburg: Via Afrika;

Uyanne, F.U., 1994, ‘Truth, ethics and divination in Igbo and Yoruba traditions: (A reply to Emmanual Eze)’, Quest: Philosophical Discussions, 8, 1: 91-96

van Binsbergen, W.M.J., in press (2003), Intercultural encounters: African and anthropological lessons towards a philosophy of interculturality, Berlin/Muenster: LIT.

van Binsbergen, W.M.J., in press, ‘Islam as a constitutive factor in African ‘‘traditional’’ religion’, in: Breedveld, A., van Santen, J., & van Binsbergen, W.M.J., eds., Islam and transformations in Africa, Leiden: Brill


to Wim van Binsbergen's website on African religion | to the Shikana portal


click the image to return
to the Shikanda portal
page last modified: 27-09-2003 14:27:17