SOURCE: Bulman, Stephen. “The Buffalo-Woman Tale: Political Imperatives and Narrative Constraints in the Sunjata Epic.” In Discourse and Its Disguises: The Interpretation of African Oral Texts, pp. 171-207. Birmingham, Ala.: Centre of Western African Studies, 1989.

[In the following essay, Bulman analyzes how the storyteller achieves new meaning by reworking the hunter-stranger motif in the Sunjata.]

The Sunjata epic has achieved considerable prominence among the oral art compositions of West Africa since Niane's publication of a prose reconstruction in French in 1960. Other versions more faithful to the style of jeli (griot or bard) have since appeared in some numbers, most notably a rendition by Kèlè Mònsòn Jabatè—which has been published in three different forms (Diabaté 1970 and 1975; Moser 1984)—three recitations from the Gambia (Innes 1974) and two from Kita (Johnson 1979 and 1986). Interpretative work upon the epic (primarily as folklore) and its context has also appeared in Bird 1972, 1977, Bird and Kendal 1980, and Johnson 1978, to name a few, and a comparative survey of variants has been written as a thesis (Belcher 1985). The epic has also served as an important—if somewhat mistreated—historical source for scholars since Delafosse's Haut-Sénégal-Niger in 1912 (reprinted as Delafosse 1972); and it is still relied upon in the standard histories of medieval Mali today (e.g. Levtzion 1980). It has been clear since the earliest days of this two-fold study of the epic as art and as history that the work was an amalgam of indigenous folkloric motifs, Wandersagen, and historical reminiscence, conflated into a narrative that reflects Malinke political, social and cultural realities and ideals. But if the epic as a source has provided richly for different fields of research, little effort has been given to analysing the ways in which its various functions have interacted with each other, and with the raw material employed to create them.

In this paper I propose to examine one early section of the epic, concerning the time before the birth of the hero, in terms of the relationship between the motifs, clichés or themes used—I make no strict typological differentiation—in the composition of the narratives, and the categories of meaning which are created through their adoption. I suggest that, in this part of the epic, popular themes from indigenous hunter practice and folklore have been chosen in conjunction with a ‘wandering’ story pattern, also relating to hunters, together creating a narrative which concerns itself with the origins of social and political institutions—specifically, imperial Mali—while retaining the wandering theme's theoretical sociogonic stance.

1. THE TALE

The Buffalo-woman tale—I use the more neutral term ‘tale’ in preference to ‘legend’, ‘myth’ or ‘folktale’, terms which tend to constrict the narrative to only one of its aspects—is an optional introduction to the Sunjata epic of the Malinke, and is also told as a separate story. This study is based on fifteen variants of the tale dating from about the early 1900s to the 1980s.¹ Some are précis of traditional recitations, others prose reconstructions, while others still are full transcriptions of a performance, appearing in linear form. While there is variation in the outward form of the tale, and considerable differences over details, the basic structure of the story has remained constant over time.

The Buffalo-woman tale describes events which led up to the birth of the epic's hero, recounting how his parents came to be married. It tells of a ravaging buffalo that destroys crops and kills men, resisting all attempts of hunters to defeat it. Usually set in the kingdom of Do, the story goes that, upon the death of the king, his daughter, a member of the Kòndè (otherwise Kòntè or Kònè) clan, was offended by the behaviour of her brothers who confiscated her inheritance or who excluded her from a ritual feast. She vowed to avenge herself and went to live in a solitary hut on the edge of (or sometimes in) the bush. From here, in the form of a buffalo, the woman made raids upon the village, destroying its crops and attacking its menfolk. The hunters of Do were unable to overcome the beast, many of them being killed in the attempt, and the desperate king—the buffalo-woman's brother or nephew, although he did not know the animal's identity—made the community's situation known further afield. Two brothers, hunters, and members of the Traore (Tarawere) clan, hearing that the king would offer the hunter who successfully killed the beast a significant gift in reward—a bride, wealth, or half of the kingdom—determined to try their luck.

Before leaving Manding—their country of origin—for Do, they consulted soothsayers or marabouts as to the way in which to approach the task, and were told on reaching their destination to befriend an old woman who lived alone by the side of the road. The two hunters duly met this woman—she was the buffalo-woman herself, in her human form—and offered to perform domestic tasks for her, such as collecting wood, gathering food or cooking. These offers she bluntly refused, or else refused to acknowledge, but after much persuasion she allowed the brothers to help her. Following on from this meeting the hunters helped the old woman, sometimes lodging with her, and always being kind and generous towards her, despite her irascibility.

Meanwhile they were having no success attempting to hunt the ferocious buffalo and were on the point of giving up their quest when the old woman—impressed by the hunters' kindness towards her—admitted her dual identity and gave them the information necessary to kill herself in her animal form. She made one proviso, that upon her death they should choose from among the young women of Do offered to them one particular girl, disfigured by warts, boils, or a hunchback, called Sogolon. She was the niece of the old woman and daughter of the king. Armed with the old woman's instructions the two hunters encountered the buffalo and were able to kill it, after a long and dangerous chase, and employing magical means (usually using a spindle rather than a conventional arrow). At the celebrations in the village next day the brothers chose the ugly Sogolon as their bride, to the crowd's bemusement, and took her with them back to Manding. On the way, the elder brother tried to consummate the marriage, according to some accounts, but was fiercely repulsed by Sogolon, who became animal-like. At Manding they therefore offered the woman to the king, Nare Famaghan Kònatè, who had been informed through a prophecy that he must marry an ugly but powerful stranger if he were to have a successful heir. At first, according to some variants of the story, he had as little success as the hunters in coupling with his new wife, as she became animal-like again, and he succeeded only by raping her. Sogolon conceived a child—Sunjata—and after the night of his conception became a model wife and later mother, patiently enduring rejection from her husband and humiliation from her co-wife, until, in exile with her son, she died.

2. HUNTER FOLKLORE AND PRACTICE

The Buffalo-woman tale is to some extent a story about the aims and achievements of hunters. One approach to interpreting the tale is therefore to examine it in relation to hunting among the Malinke. There is a specific type of folklore among the Malinke traditionally associated with hunter societies, and there are certain attitudes towards society which have been claimed to be the views of hunters (Cissé 1964; Cashion 1984: 265ff., 309ff.). Charles Bird has seen this folklore as the root of the Sunjata epic itself (Bird 1972: 291-2), while Youssouf Cissé believes hunter folklore to have retained traditional conceptions by resisting Muslim learning (Cissé 1964: 218). This folklore, the concepts it embodies, and the practices of hunters might therefore be expected to illuminate our interpretation of the Buffalo-woman tale.

Studies in traditional hunting practices among the Malinke refer to an institution which shows marked similarities to the position in which we first find the hunters of the Buffalo-woman tale. One source tells of a journey of exploration undertaken by the initiate hunter, often with a companion, to communities far from his own, with the aim of acquainting himself with the wider world, practising his trade, and learning from those hunters he meets. The itinerant hunter would lodge with a related family in the village he comes to or, we are informed, he might stay with an old woman or ‘grandmother’ of the village (Cissé 1964: 184-5).

A second source, which calls the institution the Dali-masigi, describes it as a ‘hunter's adventure’, translating literally as ‘temporary camp’. According to this source, it is not restricted to initiate hunters or to members of hunters' societies at all, although its main aim is hunting, and the seeking of fame, wealth and knowledge. The adventurer travels in like manner to the initiate hunter, and earns his keep at a village through selling game he catches, protecting livestock from wild animals, and sharing his valuable herbal knowledge with the community (Cashion 1984: 240-3). It is unlikely that a Malinke audience could comprehend the Buffalo-woman tale, with its journey by the Traore brothers, sojourn with the old woman, and hunting of the buffalo, except in the light of these practices.

Furthermore, the hunters' relationships with the old woman and Sogolon reflect traditional Malinke hunter stereotypes of women. These find expression in what has been called—by Cissé (1964: 176)—the hunters' ‘constitution’: the myth of Sanin and Kontron. Authorities differ as to the relationship between these two characters. Some sources say that the couple were both wife and husband, sister and brother (e.g. Sidibe n.d.:50), others that they were mother and son (Cissé 1964: 177-8). According to Cissé, Kontron was born parthenogenically, and as an adult remained very close to his mother, to the exclusion of relationships with other women. His followers, those that he initiated into the secrets of the chase, were the ancestors of today's hunters. Deep respect for and devotion to the mother is a feature of ideal behaviour among Malinke hunters: the mother (or maternal grandmother, if she is still alive) receives the best part of a hunter's game, according to the traditional division of the kill; and it is from the maternal side that the hunter believes he receives his strength (Cashion 1984: 243-7, 105, 264n, 344). In contrast, wives are viewed very differently. It is true that the hunter husband is supposed to have no secrets from his wife (ibid.: 315), but in fact women as sexual beings are perceived with a certain amount of suspicion and apprehension. The hunter is warned against too much interest in sex, as this will distract him from his pursuit of glory, and sexual intercourse before the hunt is believed to impede the chances of successful hunting in the bush (ibid.: 345-6; Bird 1974: 104-5).

These differential views of women are manifest in Malinke hunters' tales, providing us with a clue to the interpretation of the half-animal half-woman figure of the Buffalo-woman tale. Some of these hunters' tales record metamorphoses of human beings into animals—for example Cekura and Kumba in the epic of Kambili (Bird 1974: 95; cp. also the Gow hunter tales in Dupuis-Yakouba 1911: 21-88)—while others tell of animals who transformed into humans. One story in particular, variants of which have been recorded in many places, and which tells of an animal—buffalo or elephant—who became a woman in order to deceive a hunter, is worth examining in some detail (see Sidibe n.d.: 52-9; Thoyer-Rozat 1978: 51-8; Jackson 1982: 213-6; Meyer 1987: 25-9).

The story goes that a man was so successful in his hunting that he killed nearly all of the animals in the bush. Apprehensive about their future, the remaining animals determine a plan to discover the hunter's secret powers—those that enable him to kill so much with impunity. A female antelope, elephant or buffalo (depending on the variant) volunteers to carry out the mission, and transforms herself into a young woman. In this shape she approaches the village and becomes the hunter's lover or wife. In such a position she is able to ask the hunter to reveal the secret of his success to her. In one variant (Sidibe n.d.) we are informed that her victim has told her all but one of the objects into which he can transform himself when, in the nick of time, his mother interrupts him. Nevertheless, the young buffalo-woman departs and, armed with her knowledge, the next day the animals prepare a trap for the hunter. They are able to locate him even when his shape changes but, because of the mother's foresight, he escapes as his last transformation. In another variant, the buffalo-woman discovers that his secret is a pair of ‘hunting glasses’ which enables him to see in the dark; she steals these, and the hunter is only saved by the help (in this telling) of his wife, but usually his mother, who manages to get them back from the animals (Jackson 1982: 230).

Michael Jackson, the anthropologist who recorded this latter variant of the hunters' tale from among the Mande-speaking Kuranko of highland Guinea, offers an incisive and useful interpretation of the story. Concerning the buffalo-woman of the tale, he writes that ‘The image of the buffalo-woman gains its dramatic strength from two facts; first, the buffalo … is, according to Kuranko hunters, the most formidable, clever, and dangerous of all animals; and second, the greatest supernatural danger for hunters is believed to be an unfaithful wife or an adulterous liaison.’ Concerning the hunter: ‘… the hunter-hero, who in many ways embodies the ideal attributes of Kuranko manhood, is shown to be vulnerable to sexual love.’ And concerning the mother who saves him: ‘The mother, symbolizing maternal love, helps him regain his power, and thus the whole narrative takes the form of an allegory in which the male order is maintained and perpetuated only through the constancy and perspicacity of the mother.’ Finally, Jackson notes how the tale correlates the hunter's excessive killing of animals to his failure to prevent his own seduction (i.e. his excessive love), while contrasting this sexual excess and its sad consequences to maternal love and its rewards (Jackson 1982: 230-2).

So we can say in identifying the motifs employed to build up the buffalo-woman tale that, on one level, various elements of idealised hunter practice, their folklore and ‘worldview’ are embedded in the text. Specifically, that the journey the hunters make, and the stance they adopt of protecting the village from outside threats, corresponds to established paradigms of hunter narratives. Likewise, the attitude of respect and generosity that the tale's hunters display towards the old woman reflects the accepted ethos of hunters; while the lack of success they experience with Sogolon conforms to the commonplace distrust of women as sexual beings found in hunter folklore in general. Finally, that a half-human half-animal figure, often construed to be a threat to hunters and/or society, is found with some frequency in their folklore, and—in this one hunter tale that we have examined at least—is able to dupe the hunter-hero, revealing his vulnerability and need for support.

Further meanings accrue to the notion of human-animal transformation in the Malinke context. As can be seen from Sidibe's variant of the hunters' tale analysed above, some hunters are believed to possess powers of metamorphosis into various objects or creatures. A belief in transformation of this type, to evade pursuit, is common in narratives in this part of West Africa² including some variants of the epic of Sunjata, in which Sumanguru is attributed with powers of transformation (Innes 1974: 77; Cissé and Kamissoko 1975: 395). This is a local variant of the widespread theme encapsulated in motif D671 Transformation flight.

What of the transformatory power of the Buffalo-woman and her niece, Sogolon? It might be linked to the Malinke notion of the tana of a clan. That which is tana to a person is incompatible with him or her and is tabooed. Each clan has a tana which is usually an animal that the members of the clan must not harm in any way. The interdictions are explained by saying that a special relationship exists between the clan and the particular species, which is often couched in a story that details how a person of the clan became indebted to an animal of the species. So, for example, members of the Sidibe clan will not harm the jibirile bird, for it is said that one of these birds once saved a Sidibe clan member from death by directing him to water during a drought (Camara 1976: 26-7). A belief is held that certain members of the clan can transform themselves into their tana, particularly upon death, but also—if they possess the requisite magical resources—at other times (ibid.: 27).

But there is no evidence that buffaloes are tana to the Konde clan, either in the form of an explanatory story or a reference in the buffalo-woman tale. Other explanations must be sought to account for the relation of the old woman to the buffalo. Pierre Smith notes how the Malinke hold that two distinct species of animal may be related to a clan, one of which represents the nyama (life-force) and engendering power of the clan, and the other which represents the deceased ancestors and is taboo. Transformation into the latter is of course said to occur upon death, but the form of the other animal, that representing nyama, can (it is claimed by some) be assumed during life. Smith notes how some traditions say that Sunjata Keita became a hippopotamus upon death, and that this is the Keita tana (cp. versions of the epic in Arnaud 1912: 172; Sidibe 1959: 47) but that a Keita can assume the form of a lion while still living, particularly if angered or aroused in some way (Smith 1981: 482; cp. Meyer 1987: 74; Johnson 1986: 153). Such a belief may more adequately account for the old woman's transformation into a buffalo upon the confiscation of her inheritance, and the similar transformation of Sogolon when threatened by the unwanted attentions of the hunters and Nare Famaghan.

3. THE HUNTER-STRANGER MOTIF

In addition to the hunter-related folklore themes identified above, it can be suggested that a Wandersage, also concerning hunters, underlies the Buffalo-woman tale.

If the tale is viewed from the perspective of the two hunters it can be told in the following terms: Two hunters enter a kingdom in response to a call for help; they destroy a wild beast which was threatening the populace and in return are allowed to marry the king's daughter. From this angle, the Buffalo-woman tale clearly shares many common features with those stories known collectively as the Hero myth and Dragon-slaying myth of Indo-European and Middle Eastern traditions. Lord Raglan and Jan de Vries, in their comparative surveys of the motifs involved in the Hero myth, both record that the typical hero of the pattern fights and defeats a dragon or monster before marrying the heir to the throne; another common feature is that the hero comes from outside the society in which he performs his deeds—that he is a stranger.³

Variations on this pattern are also found in Africa, where the resultant tale is often employed to account for the origin of a ruling clan. For example, the story in the Tarikh al-Sudan accounting for the early dynasty of Songhai which, it claims, originated from two destitute brothers from Yemen, one of whom killed the sacred fish of the inhabitants of Kukiya, and was then proclaimed ruler (al-Sa'di 1964: 4-8). In other cases of the African incidence of this motif the stranger is identified as a hunter. The Nyoro story of Rukidi tells of an uncouth and wild hunter who comes from outside to marry and establish the chiefdom; reference is made to an earlier killing by the same hunter of a strange creature, half-monkey and half-lion (Wrigley 1973: 219-35). The Shambaa story of Mbegha follows similar lines: Mbegha was an itinerant boar hunter who, through his profession, gained a position in his adoptive community and who became king upon killing a dangerous lion (Feierman 1974: 43-4). Igala legends of origin also centre around a hunter-stranger who marries into the group, having first shared his game with it, and whose descendants became rulers (Boston 1964: 116-26). Finally, the Bashu tale of Muhiyi, part of the Bashu chiefdom origin tale, records how a wandering hunter gave meat to the members of a sedentary community and then married into the group (Packard 1980: 160).

Comparing the stories, at one end of the spectrum a stranger, who is a hunter, in return for his services as a provider of meat and protection, gains a wife, while at the other end a stranger rids society of a grave threat and is rewarded by marriage into the ruling family. In both cases the basis of the motif is one of an exchange between indigenous and stranger resulting in a synthesis. The Buffalo-woman tale operates in the same manner, with the hunters exchanging their skill and bravery—in their destruction of the buffalo—for marriage to the daughter of the king.

What is the basis of the powerful attraction that this pattern appears to exert? Why is it that the ruling line is said in these and many other cases to be formed by the joining of a hunting stranger and an indigenous woman? Some commentators have seen in the motif a reference to an earlier historical reality: under a system of matrilineal descent and intermarriage between matrilocal ruling houses, prospective male monarchs would have by definition been foreigners.⁴ The possibility of this situation cannot be denied, but it is not necessary to posit a historical stage of this type in every community in which the motif appears. Explanations for the widespread use of the hunter-stranger motif should concentrate rather on the pattern's immanent categories. These, I suggest, hold the key to why the motif, historical or not, retains its prominent position in so much folklore and myth.

Note the contrast between the descriptions of the stranger and those of his indigenous bride. While the woman is described as the daughter of the king, the outsider is in the Songhai story a destitute and vulgar prince, in the Nyoro tale a wild and uncouth hunter, an outcast hunter of boars to the Shambaa, and simply a hunter in the Igala, Bashu and Malinke tales. Except for the Songhai prince—whose status is itself belied by his appearance—therefore, all the other strangers are hunters and lack hereditary status. J. S. Boston has noted how hunting in Igala society is reckoned as a status-achieved profession, in which all men, regardless of rank, can excel according to their skill and courage (Boston 1964: 123). The position is analogous in Malinke society. As Gerald Cashion has remarked:

Whereas in Malinke society in general, ethnicity, age, noble birth, freeman status, occupational specialization, descendancy from former slaves, or circumcision are all hereditary determinants of status or political power, none of these distinctions apply within the hunter's society.

(1984: 102)

A first contrast of achieved/prescribed can therefore be set up between the hunter-stranger as status-achieved, and the indigenous woman whose status is hereditary.

Hunting is also a profession that exists, by definition, on the edge of society, as the hunter moves from village to bush.⁵ To succeed in his task, the hunter must comprehend and to an extent partake in the milieu of his quarry—the bush, nature—as well as the life of the village. Add to this general conception the particular description of the stranger-king Rukidi of the Nyoro story, who was ‘wild and uncouth’, and of Mbegha, who was an outcast from his original community, and the Songhai princes, who were destitute and dressed in animal skins, and this prompts a second contrast between hunter-stranger and indigenous woman of bush/village, nature/culture. We see that the stranger's externality does not merely signify his foreignness from that particular society, but rather it implies that he is foreign to society and culture in the abstract.

Turning to the interaction between hunter and society, the two most extreme variants we postulated for the motif suggest either that the hunter saved society from a monstrous threat which therefore entitled him to marriage, or that a supply of meat to the community similarly entitled him to marriage within it. In those cases where an animal or monster is killed what the stranger achieves is a neutralisation of a threat external to society, that is, a threat from nature. In the other examples, the situation is structurally similar, for the hunter gives of the natural world insofar as he shares his meat. The hunter-stranger is able to fulfill this task of mediation because of his closeness to the natural. In the same way, the gift from society to the stranger provides the hunter with two things he cannot gain in his semi-natural state: marriage, and a position of hierarchical dominance. It may thus be said that the interaction brought about by the hunter-stranger motif partakes of a two-fold socialising or reframing of society. The hunter is socialised, his achievement subsumed within the structure of society, through his encounter with the community and his marriage, while society itself is redefined by the hunter's removal of the external, natural threat which surrounded it. A new society, involving elements of the natural and of the hierarchical is created in the synthesis.

One might say that the hunter is the classic Lévi-Straussian mediator between opposites, for in overcoming the natural threat, yet partaking at the same time in a particular natural constitution, he subsumes something of that natural threat within himself and thereby facilitates the integration of the natural element into the new society.

Marshall Sahlins has suggested that the significance of the hunter-stranger (or, for him, stranger-king) motif is that it separates state and society, conceiving of an original configuration in which authority existed apart from the social group, and that power was added to society through a fearful but necessary interaction with the stranger, who contained state power within himself. To quote: ‘Power and nature are alike as what is beyond and apart from the norms of culture’, and, ‘Kingship makes its appearance from outside the society’ (Shalins 1987: 76, 73). The opposition state/society carries within it some of the connotations of the male/female opposition; he writes:

The immigrant sovereign is a ferocious male: virile young warrior and penetrator from outside … The indigenous people are, at the initial moment, ‘the side of the woman’. They are associated with the powers of earth and underworld …

(Sahlins 1987: 90).⁶

The hunter-stranger motif, then, pictures the reformation of society with an internalised authority by means of a semi-natural stranger who, by destroying the external threat (which was in fact authority or power, exercised beyond the control of society), internalises that power (through himself as mediator) in his marriage into the cultural hierarchy, the land possessors, represented as woman.

A second series of themes beyond those specifically associated with Malinke hunter folklore have now been identified in the Buffalo-woman tale. The Traore brothers, it is suggested, correspond to the mediating stranger who comes from outside the community, while their destruction of the buffalo corresponds to the combat-myth element found in many versions of the hunter-stranger myth. This series of themes carries with it, it has been proposed, its own significant sociogonic meaning relating to the origin of the state, the formation of which is attributed to the double inheritance of the foreign hunter's skill and sublimation of the natural, and the autochthonous woman's legitimate right and power over the land.

Assuming that the connections I have suggested are correct, nevertheless, on four significant points at the end of the Buffalo-woman tale the story departs from the pattern of the hunter-stranger motif identified above. To begin with, it should be noticed that neither of the two hunters actually becomes Sunjata's father, this role being reserved instead for Nare Famaghan; rather, they were warded off by Sogolon's animal aspect. Secondly, Nare Famaghan himself, the actual father of Sunjata, is no ‘natural stranger’ as the motif suggests, but a static, hereditary figure. And thirdly, Sogolon, the supposed representative in the tale of the autochthonous woman, the symbol of culture, embodying the concept of hereditary authority, is in fact a wild half-woman half-animal in the mould of her aunt, the buffalo-woman. Finally, according to some accounts, Sogolon does not accept the king's advances any more than she did those of the hunters, putting up strong, though futile, opposition to her spouse, by recourse to her animal transformation.

These four points of alteration are, I suggest, purposeful and, rather than invalidating our thesis, point to further levels of meaning in the tale—specifically, to its political and social dimensions.

4. THE POLITICAL DIMENSION OF THE TALE

Taking the first variation from the hunter-stranger motif described above, it can be noted that the Traore hunters, who win Sogolon as a bride because of their defeat of the buffalo, do not in fact marry her or at least, if they do, don't consummate the marriage. Much ambiguity pertains to the relations between Sogolon and the hunters, and this is perhaps recognised by some versions of the Sunjata epic which, later in the story, attribute the hero's failure to pass the sigi contest—a ritualised trial of strength undertaken to prove legitimacy as a horon (noble)—to his mother's intimacy with the Traore brothers before Sunjata's birth, with the faint implication that the hero may have been illegitimate (see, for example, the version of the epic in Conrad 1981: 707). But this alteration in the hunter-stranger theme has, I believe, great significance for the specific purpose of the Buffalo-woman tale as an explanation for the political makeup of the Mali empire.

Over two-thirds of the examined variants place the story at Do.⁷ Do is not of course a fictional setting but was an ancient canton or chiefdom of the Malinke. It can be identified with the ‘great kingdom’ of ‘Daw’ that al-Bakri placed beyond Ghana in the eleventh century (Levtzion and Hopkins 1981: 82). And of course Manding, given in the tale as the home of the hunter and of Nare Famaghan, Sogolon's eventual husband, is equally historical, and is probably to be identified with al-Bakrī's Malal, which he refers to in conjunction with Daw. In some Malinke oral traditions, apart from the epic itself, Do is spoken of together with Kri (or Kiri). As D. T. Niane reports:

There is a verbal refrain: Do ni Kiri / Dodugu tan nifla …
which means “Do and Kiri, country of twelve towns …”

(Niane 1984: 129)

Charles Monteil stated that the origin of Mali was to be found in the two cantons of Do and Kri. He believed that Kri was the more ancient of the two, and was later superseded by Do (Monteil 1929: 344-47). Kri is not directly mentioned in the Buffalo-woman tale, but it is implied through the figures of the Traore brothers, the hunters of the story. The Traore clan, some oral traditions claim, was the leading clan of ancient Kri (Niane 1984: 127; Monteil 1929: 344). Some traditions link the clan of Kòndè to ancient Do (Niane 1984: 127) while over half of the versions of the tale are in agreement in seeing the clan to which Sogolon, the old woman/buffalo-woman and the king of Do belonged as the Kòndè.⁸ In the same way, the leading clan of Manding is given in oral traditions as being (before the Keita) the Kònatè (Monteil 1929: 345). Some versions of the tale link Nare Famaghan to the Konate⁹ and the Buffalo-woman tale nearly always links the hero and his father to Manding.¹⁰

In this way, Traore, Kòndè and Kònatè, three of the leading clans among the pre-imperial Malinke, are referred to in the Buffalo-woman tale, along with three major pre-imperial chiefdoms or states—Kri, Do and Manding. The Buffalo-woman tale, in that it describes the process by which Sunjata, the first emperor of the united Malinke state, came to be born, can therefore be read as an origin tale for the empire. By reference to the early states of the Malinke people, and their roles in accomplishing the first resolution of the story—that is to say, Sunjata's birth—the tale implies that each of the states was an important contributant to Mali's formation; and by recording the clans of the participants in the drama, the tale similarly emphasises each clan's significance in that genesis.

In some senses each clan can also be seen to share in the formation of the royal clan of the Keita. The origin of this name is not easy to determine. While Sunjata's ancestors are generally referred to as the Keita, there is almost no reference in the genealogy to the name Keita before Sunjata or his father's generation in the sixteen variants I have examined.¹¹ The relationship between Keita and Kònatè is also difficult to determine. Niane's version suggests Kònatè is the junior name for Keita (1965: 78), and certainly Sunjata is often called Kònatè as a youth (e.g. Frobenius 1913: 461; Johnson 1979: 80). Some sources say that the name Keita derives from ke, ‘inheritance’ and ta, ‘to take’.¹² Two of these sources directly link this ‘taking of inheritance’ to Sunjata's own usurpation of his elder brother's claim to the throne of Manding, suggesting thereby that the name was not used before Sunjata's own time. Insofar as the name Keita is deemed to originate in the hero's time, then, the three clans Traore, Kòndè and Kònatè might claim, via the Buffalo-woman tale, to have each played a part in the imperial clan's genesis.

So, in addition to being an origin tale in the abstract, speculating on the nature of power and authority, the Buffalo-woman tale gives a specific political origin for Mali. The empire is seen as a unified expression of Malinke states; the Keita as the unity of leading Malinke clans. I suggest that it is on this level of the narrative, in which specific political realities are encountered—albeit in an idealised form—that the cause is to be sought for some of the divergence from the hunter-stranger motif—essentially, that which ‘prevented’ the marriage of Sogolon to the hunters, and determined that her husband—the father of Sunjata—should instead be Nare Famaghan. The reasons have to do with the system of succession practised in Mali, and the motif's incompatibility with this.

As was noted above, the aetiological pattern of the hunter-stranger motif is explicable in terms of a descent of kingship through the sister or daughter, in which husbands are sought externally. Now Mali's system of succession is not certain in all cases, but was clearly not exclusively matrilineal. In such circumstances the motif's implications become problematic, for it assumes the right to rule to be passed through the woman, who is seen to embody prescribed status, while the husband is the outsider. However, we cannot rule out the possibility of some successions in imperial Mali through the female side, for some evidence of matrilineality does exist. Since Levtzion's discovery of the error in de Slane's translation of Ibn Khaldun, referring to Abu Bakr as a ‘descendant of Sunjata's sister’ rather than the correct ‘son of Sunjata's brother’ (Levtzion 1963: 346) most scholars have assumed Mali's succession to have been wholly patrilineal. Nawal Moccos Bell, however, has pointed to a case of matrilineal descent involving Abu Bakr I's right to rule, and has supported this case with other points which lead her to believe that imperial Mali may have had a ‘flexible’ system of succession, if not a bilateral one (Bell 1972: 227-34).

Such a system which, depending on various factors, might trace eligibility to rule through the male or female lines could account for the actual situation presented in the Buffalo-woman tale. Here we find that both Sogolon and Nare Famaghan are of royal descent, and both in a way represent prescribed status, culture and society. Were Sunjata the offspring of the Traore, presented by the epic as itinerant hunters with no overt royal connections, Sunjata's right to rule would find itself supported only by his mother's connections to the royal house of Do. I suggest it is by ensuring that both his parents have links to undoubtedly royal families that the traditionists can make certain that none can doubt Sunjata's right to rule, whether it be traced through the female or male side.

Of course it can be countered immediately that the Traore are traditionally associated with the rulership of Kri and so have an equally strong claim to a part in Sunjata's parentage. In the tale, though, they are presented simply as hunters, and this royal link to Kri is merely implied from other sources. But the epic might be said to accept something of this royal link, and this might account for the ambiguity (noted above) surrounding the relationship between Sogolon and the Traore brothers. One might suggest that the tale seeks to link the brothers in Sunjata's birth, although stopping short of according them a direct fathering role, but implicating them and their connections with the ancient rulership of Kri in the process of Sunjata's creation. If a parallel could be given it might be the way in which St. Matthew's gospel recites Joseph's genealogy, affirming his descent from king David, and then proceeds to claim that Joseph played no part in the creation of Jesus, but that the Messiah was conceived by the Holy Spirit.

5. THE SOCIAL DIMENSION OF THE TALE

Several variants of the Buffalo-woman tale mention the beginning of senankuya between Traore and Kòndè in connection with the hunters, linking it to these hunters' choice of Sogolon.¹³ Other authorities confirm that the clans of Traore and Kòndè are senankun,¹⁴ which is to say that certain special modes of behaviour operate between them, certain actions are prescribed, while some relationships are prohibited (Camara 1976: 32-47). The institution of senankuya (the joking relationship) forms bonds between Malinke clans in different regions and between the Malinke and their neighbours such as the Fulani and Bamana. Related clans must perform certain services (for example at marriage and funeral ceremonies); they may disregard the normal forms of politeness; they may not intermarry. Tales accounting for the origin of particular senankun relationships in terms of a personal story are often found. For example, one story from the Fulani of Senegal tells of a man of the Diakhabi clan who prepared a meal for a prospective son-in-law from the Kaba clan, and who asked the guest to choose between the rice and his daughter; the Kaba man chose the food and his clan henceforth was forbidden to intermarry with the Diakhabi (Smith 1987: 474).

Examined in these terms, the ‘non-marriage’ of the Traore hunters to Sogolon Kòndè can be viewed as an explanatory tale of origin for the senankun relationship between the two clans. So ridiculous and insulting is the choice of an ugly hunchback from among all the eligible daughters of the Kòndè that future Traore forfeit their right to marriage with the Kòndè clan. Several of the variants of the tale do indeed make this point. Niane's version hints at it, noting that the hunters left Do ‘pursued by the mockery of the Kondés’ (1965: 9), while Frobenius' account (which has Sogolon as a Diarra) goes: ‘Thus, for the first time, the Diarra and the Traore reviled each other and since then have done the like until our days’ (1913: 457).¹⁵ The hunter's failure to consummate the marriage acts, in this context, as a justification or verification for the senankun relation between Traore and Kòndè.

On the other side, the marriage of Sogolon Kòndè to Nare Famaghan Kònatè, father of Sunjata Keita, is emblematic of a further aspect of marriage convention among the Malinke. Some versions of the epic record that, after his victory over Sumanguru and establishment of Mali, the emperor Sunjata decreed that all Keita of Manding should take their wives from the Kòndè of Do (to quote Niane's account): ‘in memory of the fruitful marriage between Nare Maghan and Sogolon’ (1965: 78).¹⁶ It has been noted that the conjunction of senankun-based prohibition of marriage between clans and positive rules concerning marriage obtaining between other clans can lead to a system of generalised exchange (Camara 1976: 32-3, 46). So far we have seen that, in some versions the epic enjoins marriage between Kòndè and Keita, while banning it between Kòndè and Traore. Further to this, in roughly two-thirds of the versions of the tale examined we can note that, in exchange for giving Sogolon to the king of Manding, this one offers the Traore brothers his sister or daughter. Perhaps—although we can supply no supporting evidence—this can be seen as justification for preferred exchange between Traore and Keita.

6. REINSTATEMENT AND REFORMATION OF THE MOTIF

Returning to the issue of Sogolon's marriage to Nare Famaghan, and the significance of variations from the hunter-stranger theme, it should be noted that, in spite of the sanction that this marriage receives later in the epic, according to some accounts, the king of Manding is at first no more successful with Sogolon than either of the hunters had been. When approached by her husband Sogolon is said to stretch out to an enormous size, to grow feathers (Humblot 1951: 112), hair (Niane 1965: 11), or spines (Camara Laye 1980: 95, 62), and to assume a far greater strength than that of her would be partner. Eventually, Nare Famaghan overcomes his bride's resistance through a ruse, but why should Sogolon be described in this way as an animal-like creature in the image of her aunt, the buffalo-woman? Certainly her animal aspect helps the progress of the tale when she employs it earlier on in the story to ward off the Traore brothers when they would consummate the marriage. Also, transformatory powers are traditionally associated with nyama, which the Malinke child obtains through his mother, hence Sogolon's animal aspect can be read as a demonstration of Sunjata's powerful inheritance (Johnson 1978: 14). But another purpose, I suggest, has to do with the basic hunter-stranger motif on which the tale is formed, and in fact acts to reinstate—but also reform—the pattern that was upset when Nare Famaghan became Sunjata's father instead of either of the hunters.

It was proposed above that the essence of the hunter-stranger motif lay in its joining of a series of oppositions around the terms achieved/prescribed, stranger/indigene, nature/culture, and so on; and that it sought to institute authority as a mediation between natural power and cultural or political order. This it achieved by the marriage between an indigenous woman of high birth and a hunter who mediates nature from outside society, so that the resulting dynasty should partake equally of cultural and natural elements.

But by marrying Sogolon, the indigenous symbol, and part of the ruling family of Do, to Nare Famaghan of Manding, someone similarly indigenous and ‘cultural’, to whom the tale generally ascribes no feats of prowess or other signs of natural strength, the final equation in the Buffalo-woman tale would have lacked the natural portion: Sunjata would have merely been the product of a marriage between two representatives of culture. To avoid this, I believe, the elements that marked out the buffalo-woman—animal strength and a disregard for social norms—are reapplied to Sogolon, her niece. In this way, at the expense of the prowess of the hunters—who overcame the buffalo only to find themselves presented with another—Sogolon is able to bring to the marriage the natural element that the reworked formula of the hunter-stranger motif had lost. By these means is Sunjata presented as an archetypal hero figure, the product of both the cultural order and of natural prowess, and the fundamental message of the hunter-stranger motif is reinstated.

Furthermore, this equation is closer to the realities of gender differentiation as perceived by the Malinke. Sogolon is seen to share the same characteristics of the feared animal-woman of the hunters' tale which was analysed earlier. Then it was noted how wives are generally regarded with suspicion, and that sex is cautioned against for the successful hunter. As Sory Camara has remarked, Malinke society is totally male controlled, with women fulfilling subordinate domestic roles. Official village, cult and society positions are an exclusive male preserve; marriage is patrilocal and controlled by men; tasks are clearly demarcated, and character traits gender differentiated (men are associated with anger—but not emotionality—and war, women with passivity and sensitivity: Camara 1976: 48-57). Moreover, women are thought not to be able to keep secrets, to be led astray by their desires, and to be basically untrustworthy. Men are warned to be on their guard against women. All this can be compared with what Jackson writes of the Malinke-related Kuranko. He brings out the implied position among the Malinke when he states that, for the Kuranko, there is a paradoxical contrast ‘between a dogma of male control and an “unofficial” admission of the actual importance of female influence’ (Jackson 1982: 199-200). Women are ‘popularly thought to be untrustworthy, temperamental, weak-willed, refractory, and capricious’ (ibid.: 201). He notes that women can either be seen as ‘actively malevolent’ or ‘passive victims of their own unrestrained emotionality’ (ibid.: 202); and that—in contrast, we might say, to male-controlled marriage—women are thought of as the seducers in extra-marital affairs.

From this evidence it is possible to state that, in general, women are thought of by the Malinke to be dangerous to men, and that this danger comes from those elements within them—emotionality, capriciousness, untrustworthiness, etc.—that are uncontrolled, unsocialised—natural. In contrast, all the official extra-domestic life of the village or state is exclusively the preserve of men—culture is male.

Such an association of culture with male and nature with female is clearly contrary to the sense of the hunter-stranger motif as we described it; and it appears to us that, at the same time as the motif has altered to fit the political necessities of imperial Mali, the male/female and nature/culture oppositions have come into line with those generally perceived among the Malinke. Sogolon is still given to be a cultural indigene, the offspring of the leading clan of Do, but her characteristics, like her aunt before her, are far from those associated with a cultural emblem, and she is in fact also a symbol of nature.

CONCLUSION

Two interconnected series of themes have been identified as the basis of the tale. The first, comprising the hunter-stranger motif, widely dispersed in West Africa and beyond, consists of the journey by a ‘wild’ stranger to a community in which he establishes a dynasty by defeating a natural threat and marrying into the leading family. The second series of themes is derived from Malinke hunter folklore and consists of the idealised behaviour of hunters in travelling upon adventures, protecting villages and hunting dangerous game; and of the transforming animal-woman who is their natural enemy. In addition, the stock hunter motifs form the basis for the way hunters relate to the old woman.

These two levels of ‘input’ have been mapped one upon the other and produce, to my mind, three levels of ‘output’ or readings. The first of these I term the sociogonic level of meaning. As I have tried to demonstrate above, the hunter-stranger motif carries within it a model outlining the development of society; society is seen to be achieved through the mediation of a natural, external, male element and a cultural, internal, female element; the resulting synthesis sanctions natural power and strength by endowing it with prescribed authority. The second level of meaning is concerned with the formation of imperial Mali. Here, the three leading pre-imperial states and clans are linked in the action of the tale: Traore hunters perform the travelling and fighting role of the hunter-stranger, but their prize, the Kòndè woman Sogolon, they hand over to a Kònatè or Manding, Nare Famaghan. The goal to which they all work—the birth of Sunjata—is the symbol of Mali's formation and initial growth, and thus can all three leading pre-imperial chiefdoms and families claim a part in the building of empire. This imperial level has been seen to interact with and transform the sociogonic level: new elements distort and reform the equation, as the natural male stranger and cultural female indigene duo is replaced by an ambivalent natural-cultural female stranger and cultural male indigene. The reasons suggested were to ensure that Sunjata had an incontestable right to the throne and to involve all leading pre-imperial clans and chiefdoms in the creation of empire. Also, this reworked sociogonic level expresses Malinke associations of female with danger and nature, which the original formulation ignored. Furthermore, senankuya between Traore and Konde and marriage arrangements between the Keita, Kòndè and Traore are grounded within the narrative.

At this point some theoretical suggestions about the way the Buffalo-woman tale ‘works’ and how it was formed can be proposed. Two procedures which may have been employed to mould the hunter-stranger motif into the Buffalo-woman tale might be posited. The first is a process of culturisation. Elements of Malinke hunter folklore—ideas concerning hunter adventures, metamorphosis, women as mothers and wives, and hunters' attitudes towards them—have made of the cross-cultural hunter-stranger motif a specifically Malinke tale. Also, when the imperial output level interacts with the sociogonic level to give a new sociogonic equation, that which is produced integrates easily with the views on gender usually found in Malinke culture. This kind of culturisation of widespread themes must have been a key element in the creation of a regional tradition such as the Buffalo-woman tale.

The second procedure operates on the level of narrative and employs the audience's accepted knowledge of the basic motifs involved in the tale. Juxtaposed to this implicit knowledge are the realities of the tale, in which expectations, we might say, are continually disappointed. This procedure can be illustrated through an examination of the role of the Traore brothers. They begin the story as its apparent heroes: they answer the challenge of the buffalo and pit their strength against nature in the accepted way. Gradually, however, this role is undermined: they fail to catch the buffalo except through its own agreement, they must take as reward a semi-wild woman who then rejects their advances, and their assumed role as dynasty-engenderers is finally usurped by Nare Famaghan.

The narrative appears to play on the difference between the audience's expectations, given their knowledge of the story genre, and the actualities of the tale's plot. Traditional motifs and plot patterns found in Malinke folklore might be seen as acting as a measure by which meanings in the Buffalo-woman tale are gauged. New meaning is created through the taking of traditional forms, which carry implicit messages, and transforming them according to the new tale's uses through already-given cultural and narrative norms. It is through such transformations of known patterns and norms that the tale is able to convey messages concerning social and imperial origin. In the same way, the cross-cultural Wandersage of the hunter-stranger acts, for us, as a template against which to measure the concrete statements based upon this theme found in the tale. A clearer understanding of the tale is, therefore, only gained through both a knowledge of the cross-cultural themes and culture-specific patterns, and a careful attempt to unravel the routes by which these have been woven together to form the given tale.

Notes

1. See Zeltner 1913: 1-8; Frobenius 1913: 449-57; Humblot 1951: 111-2; Niane 1965: 4-12; Pageard 1961: 53-4; Camara Laye 1980: 35-64; Diabaté 1970: 19-29; Johnson 1979: 39-77; Johnson 1986: 109-29; Camara 1976: 254-66; Tiemoko Kone 1970: 22-342; Conrad 1981: 673-701; Courlander 1978: 84-8, 108; Konaré Ba 1983: 18-20; Cissé and Kamissoko 1988: 45-95.

2. For example, see for the Bamana Dumestre 1979: 255-69, and for the Gow, Dupuis-Yakouba 1911.

3. Raglan 1936: 180; de Vries 1963: 215f.; see also Fontenrose 1959: 9f. for themes in the dragon-slaying archetype.

4. See Graves 1955, I: 13; Neumann 1970: 139; cp. Cashion 1984: 3, 7, 243.

5. To comprehend the significance of these terms for the Malinke see Johnson 1978: 95-7; cp. Jackson 1982: 9-10 for the Kuranko.

6. Cp. Jackson on the Kuranko: ‘The creation of community depends not merely on people behaving well. It is not the outcome of a slavish application of rules or a passive adherence to routine. The creation of a viable community depends upon vital forces which are “wild” or random and which must be fetched from outside the domain of rules and roles’ (1982: 20-1).

7. Humblot 1951: 111; Niane 1965: 6; Camara Laye 1980: 35; Diabaté 1970: 21; Johnson 1986: 110; Camara 1976: 262; Courlander 1978: 84; Konaré Ba 1983: 18. Sankaran is given in some accounts instead of Do (Zeltner 1913: 1; Frobenius 1913: 451; Tiemoko Kone 1970: 42; Conrad 1981: 672); both Do and Sankaran are given in two accounts (Johnson 1979: 39, 40; Cissé and Kamissoko 1988: 57, n.35). Sankaran is often linked to the Kòndè clan, as is Do, while some traditions speak of a migration from Do to Sankaran (Niane 1974: 64; Leynaud and Cissé 1978: 26).

8. Niane 1965: 9; Camara Laye 1980: 54; Diabaté 1970: 23; Johnson 1986: 109-10; Johnson 1979: 39; Camara 1976: 257; Konaré Ba 1983: 18; Cissé and Kamissoko 1988: 46. Two accounts place the action at Sankaran, while keeping the ruling family as Kòndè (Zeltner 1913: 1; Tiemoko Kone 1970: 42).

9. Innes 1974: 147; Cissé and Kamissoko 1988: 54; Konaré Ba 1983: 18; Diabaté 1970:6.

10. Zeltner 1913: 7; Frobenius 1913: 457; Camara 1976: 266; Johnson 1986: 128; Diabaté 1970: 20; Camara Laye 1980: 65; Johnson 1979: 74; Konaré Ba 1983: 19; Cissé and Kamissoko 1988: 71; Tiemoko Kone 1970: 282; or to Mali (Niane 1965: 4).

11. The one exception I have found is in Zeltner (1913: 38) where Sunjata's first recorded ancestor is referred to as Bilali Bou Hamama Keita.

12. Johnson 1986: 182; Jackson 1979: 103; Cissé and Kamissoko 1975: 441; Conrad 1981: 705.

13. Frobenius 1913: 457 (here the family is Diarra, not Kòntè); Niane 1965: 88, n.17 (translator's note); Diabaté 1970: 24; Cissé and Kamissoko 1988: 69-71.

14. Pageard 1958: 128; Camara 1976: 36-7.

15. Cp. Diabaté 1975: 24 and Cissé and Kamissoko 1988: 71, where Youssouf Cissé writes: ‘on trouve les clans Tarawélé [Traore] et Koné unis comme les doigts de la main; ils sont alliés par les femmes, alliés par le sang et alliés militaires. De nos jours encore, ils sont, du Burkina-Faso au Sénégal, et du Mali à la côte d'Ivoire, des Sénènkoun’ (n.48).

16. See also Diabaté 1975: 25, n.38, where Massa Makan Diabaté remarks: ‘Les Konte sont considérés comme les oncles maternels de Sunjata et les beau-parents des Keta’; he goes on to say that ‘La coutume voulait que toute femme prise à la guerre par un Keta revienne de droit à la famille Konte’ (ibid.).

Thanks are due to Karin Barber, who read an earlier draft, and particularly to Paulo Farias, with whom many of the ideas in the paper were discussed at length. For the final form I am alone responsible.

Bibliography

Al-Sa'di 1964, Tarikh al-Sudan, trans. O. Houdas. Paris: Adrien-Maisonneuve.

Arnaud, Robert 1912, ‘La Singulière légende des Soninkés. Traditions orales sur le royaume de Koumbi et sur divers autres royaumes Soudanais’, L'Islam et la politique musulmane francaise en Afrique occidentale francaise. Paris: Comité de l'Afrique Francaise (144-86).

Belcher, Stephen Patterson 1985, Stability and Change: praise-poetry and narrative traditions in the epics of Mali. Ph.D. dissertation, Brown University.

Bell, Nawal Moccos 1972, ‘The Age of Mansa Musa of Mali: Problems in Succession and Chronology’, International Journal of African Historical Studies 5, 2 (221-34).

Bird, Charles S. 1972, ‘Heroic songs of the Mande hunters’, in R. M. Dorson (ed.), African Folklore, New York: Doubleday.

Bird, Charles S. 1974, The Songs of Seydou Camara, trans. Charles Bird with Mamadou Koita and Bourama Soumaouro. Bloomington, Indiana: African Studies Center, Indiana University.

Bird, Charles S. 1977, Review of Gordon Innes, Sunjata: Three Mandinka Versions. Research in African Literatures 8, 1 (353-69).

Bird, Charles S. and Kendall, Martha 1980, ‘The Mande Hero. Text and context’, in Explorations in African Systems of Thought, I. Karp and C. Bird (eds.) Bloomington: Indiana University Press.

Boston, J. S. 1964, ‘The Hunter in Igala Legends of Origin’, Africa 43 (116-26).

Camara Laye, 1978, Le Maître de la Parole. Paris: Plon.

Camara Laye 1980, The Guardian of the Word, Kouma Lafôlô Kouma, trans. J. Kirkup. Glasgow: Collins/Fontana.

Camara, Seydou 1979, ‘À propos de quelques centres de traditions historiques du Manden’, Études Maliennes, 2 (25-39).

Camara, Sory 1976, Gens de la parole, essai sur la condition et le rôle des griots dans la société malinké. Paris and The Hague: Mouton.

Cashion, Gerald 1982, Hunters of the Mande. A behavioural code and worldview derived from a study of their folklore. Ph.D dissertation, Indiana University.

Cissé, Y. T. 1964, ‘Notes sur les sociétés de chasseurs Malinké’, Journal de la société des Africanistes 34 (175-226).

Cissé, Y. T. and Kamissoko, Wa 1975, L'Empire du Mali. [First SCOA seminar, Bamako.] Paris: SCOA.

Cissé, Y. T. and Kamissoko, Wa 1977a, L'Empire du Mali. [Second SCOA seminar, Bamako.] Paris: SCOA.

Cissé, Y. T. and Kamissoko, Wa 1977b, Mansa Kankoun Moussa le Pèlerin. Provisional text and translation. [Third SCOA seminar, Niamey.] Paris: SCOA.

Cissé, Y. T. and Kamissoko, Wa 1988, La Grande Geste du Mali. Des origines à la fondation de l'empire. Paris: Karthala-ARSAN.

Conrad, D. C. 1981, The Role of Oral Artists in the History of Mali. Ph.D thesis, S.O.A.S., London University.

Conrad, D. C. 1985, ‘Islam in the Oral Traditions of Mali: Bilali and Surakata’, Journal of African History 27, 1 (33-49).

Courlander, Harold 1978, ‘Three Soninke Tales’, African Arts 12, 1.

Delafosse, M. 1972, Haut-Sénégal-Niger. (Three volumes, first published 1912). Paris, Maisonneuve et Larose.

Diabaté, Massa Makan, 1970a, Kala Jata. Bamako: Editions Populaires.

Diabaté, Massa Makan, 1970b, Janjon et autres chants populaires du Mali. Paris: Présence Africaine.

Diabaté, Massa Makan, 1975 (ed. and trans.) L'aigle et L'épervier ou la geste de Sunjata. Paris: Pierre Jean Oswald.

Dumestre, Gérard 1979, (ed. and trans.) La Geste de Segou. Classiques Africaines 19. Paris: Armand Colin.

Dupuis-Yakouba, A. 1911, Les Gow ou Chasseurs du Niger, Légendes songai de la région de Tombouctou. Paris: Ernest Leroux.

Feierman, Steven 1974, The Shambaa Kingdom. A History. Madison: University of Wisconsin Press.

Fontenrose, Joseph 1959, Python. A Study of Delphic Myth and its Origins. Berkeley: University of California Press.

Frobenius, Leo 1913, The Voice of Africa. Two volumes. London: Hutchinson and Co.

Graves, Robert. 1955, The Greek Myths. Two volumes. Harmondsworth: Penguin Books

Humblot, Paul 1951, ‘Episodes de la légende de Soundiata’, Notes Africaines 52.

Innes, Gordon 1974, (ed. and trans.) Sunjata: Three Mandinka versions. London: S.O.A.S.

Jackson, M. 1977, The Kuranko: Dimensions of Social Reality in a West African Society. London: Christopher Hurst.

Jackson, M. 1979, ‘Prevented successions: A commentary upon a Kuranko Narrative’, in Fantasy and Symbol. Studies in Anthropological Interpretation, ed. R. H. Hook, London: Academic Press.

Jackson, M. 1982, Allegories of the Wilderness: Ethics and Ambiguity in Kuranko Narratives. Bloomington: Indiana University Press.

Jackson, M. 1986, Barawa and the Way Birds Fly in the Sky. Washington: Smithsonian Institution Press.

Johnson, J. W. 1978, The Epic of Sun-Jata: an attempt to define the model for African epic poetry. Ph.D. dissertation, Indiana University.

Johnson, J. W. 1979, (ed. and trans.) The Epic of Sun-Jata according to Magan Sisoko. Two volumes. Folklore publications group, Indiana University: Bloomington.

Johnson, J. W. 1986, The Epic of Son-Jara. Bloomington: Indiana University Press.

Konaré Ba, A. 1983, Sunjata, le fondateur de l'Empire du Mali. Dakar-Abidjan-Lomé: Nouvelles Éditions Africaines.

Levtzion, Nehemia 1963, ‘The Thirteenth- and fourteenth-century kings of Mali’, Journal of African History 4, 3.

Levtzion, Nehemia 1973, Ancient Ghana and Mali. London: Methuen.

Levtzion, Nehemia and Hopkins, J. F. P. 1981, (eds. and trans.) Corpus of early Arabic sources for West African history. Cambridge: Cambridge University Press.

Leynaud, E. and Cissé, Y. T. 1978, Paysans Malinké du Haut Niger (Tradition et développement rural en Afrique Soudanaise). Bamako: Édition Imprimerie Populaire du Mali.

Meyer, Gérard 1987, (ed. and trans.) Contes du pays malinké. Paris: Editions Karthala.

Monteil, Charles 1953, ‘La légende de Ouagadou et l'Origine des Soninké’, in Mélanges Ethnologiques, Mémoires de l'Institut Français d'Afrique Noire 23, Dakar: I.F.A.N. (359-408).

Monteil Charles 1968, Les Empires du Mali. Étude d'Histoire et de Sociologie Soudanaises. First published 1929. Paris: G.-P. Maisonneuve et Larose.

Monteil, Charles 1977, Les Bambara du Ségou et du Kaarta (Étude Historique, Ethnographique et Littéraire d'une peuplade du Soudan Français). First published 1924. Paris, G.-P. Maisonneuve et Larose.

Moser, Rex E. 1974, Foregrounding in the “Sunjata”, the Mande epic. Ph.D dissertation, Indiana University.

Neumann, Erich. 1970, The Origins and History of Consciousness. Princeton: Bollingen/Princeton University Press.

Niane, D. T. 1960, Soundjata ou L'Épopée Mandingue. Paris, Présence Africaine.

Niane, D. T. 1965, Sundiata, an Epic of Old Mali. London: Longman.

Niane, D. T. 1974, ‘Histoire et tradition historique du Manding’, Présence Africaine 89 (59-74).

Niane, D. T. 1975, Recherches sur l'Empire du Mali au Moyen Age. Paris: Présence Africaine.

Niane, D. T. 1984, ‘Mali and the second Mandingo expansion’, General History of Africa IV, Africa from the Twelfth to the Sixteenth Century, (ed.) D. T Niane. Paris, London and Berkeley: Unesco/Heinemann/University of California Press (117-71).

Pageard, Robert 1958, ‘Notes sur les rapports de “senankouya” au Soudan francais particulièrment dans les cercles de Ségou et Macina’, Bulletin de L'Institut Francais D'Afrique Noire, série B, XX, 1-2 (124-41).

Pageard, Robert 1961, ‘Soundiata Keita et la tradition orale’, Présence Africaine 36 (51-70).

Raglan, Lord 1936, The Hero. A Study in Tradition, Myth and Drama. London: Methuen.

Sahlins, Marshall 1987, Islands of History. London and New York: Tavistock Publications.

Sidibé, Mamby 1959, ‘Soundiata Keita, Héros historique et légendaire, empereur du Manding’, Notes Africaines 82 (41-51).

Sidibé, Mamby and Djian, J. no date, Veillée avec le vieux Mamby Sidibé. Bamako: Ministère de la Jeunesse, des Sports, des Arts et de la Culture (Direction Nationale des Arts et de la Culture, Division Arts et Lettres).

Smith, Abdullahi 1987, ‘The Legend of the Seifuwa: a Study in the Origins of a Tradition of Origin’, in Kwanashie, G. A., Mangvwat, M. Y., Ahmed, M. A., Bala Usman Y. (eds.), A Little New Light: Selected Historical Writings of Abdullahi Smith, vol. I, Zaria: The Abdullahi Smith Centre for Historical Research (22-56).

Thoyer-Rozat, Annik 1978, (ed. and trans.) Chants de Chasseurs du Mali par Mamadu Jara. Paris: Privately published.

Tiemoko Kone 1970, Soundiata. Ed. and trans. Lassana Doucoure and Madame Marta. Bamako.

Vries, Jan de 1963, Heroic Song and Heroic Legend, trans. B. J. Trimmer. London: Oxford University Press.

Wrigley, Christopher C. 1973, ‘The Story of Rukidi’, Africa 43 (219-35).

Zeltner, Franz de 1913, Contes du Sénégal et du Niger. Paris: Ernest Leroux.