Oral Sources on Links Between Great States: Sumanguru, Servile Lineage, the Jariso, and Kaniaga

Abstract

Qui est capable, hors Dieu, de scruter le passé? Some scholars interested in ancient Ghana and Mali dare to sift relevant oral traditions of the Western Sudan in search of historical evidence, while others express doubts that these sources can contain any information of value to historians. A period markedly affected by this question is that which saw the disintegration of Ghana and the rise of Mali in the twelfth and thirteenth centuries. Despite historians' general acknowledgement of the pitfalls accompanying the use of oral tradition as a source of information, much of what we know, or would like to think we know about this era, has been drawn from the legend of Wagadu and from the Sunjata epic. Clearly a large part of the material in these oral traditions is composed of the stuff of myth and folktale, and on the face of it the prospect of trying to glean historical information from them is not an encouraging one. But woven into the patchwork fabric of these narratives are infrequent threads bearing diminishing echoes of people and events of the distant past. Vague, inaccurate, and potentially misleading as they must be, these archaic fragments nevertheless merit whatever attention is necessary to interpret their significance, in the hope that they might yield some useful historical insights. Any pretensions to historicity in the Wagadu tradition and in the Sunjata epic may be open to question because there is so little that can be verified. While the mythical quality of some elements in the texts is obvious, there are others that could have a historical basis but cannot be independently confirmed. The material consulted here is approached with the attitude that, given the rarity of firmly documented sources, historians cannot afford to ignore the possibility that there is some information worth distilling from the oral accounts of ancient Mali and the related Soninke era that preceded it.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 32 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.