SOURCE: Sienkewicz, Thomas J. “The Greeks Are Indeed Like the Others: Myth and Society in the West African Sunjata.” In Myth and the Polis, edited by Dora C. Pozzi and John M. Wickersham, pp. 182-202. Ithaca, N.Y.: Cornell University Press, 1991.

[In the following essay, Sienkewicz discusses similarities between the Sunjata and Greek myths, particularly in addressing the “tension between diversity and unity.”]

The mark Greek mythology has left on Western thought and culture is undeniable and indelible. The use of Greek myths as standard themes and points of reference for Western artists, writers, and poets, and the survival and flourishing of Greek mythology in the modern world, have been interpreted by some as a sign of the universality of Greek mythology, of its unique adaptability to different and changing social and cultural needs.

This view of Greek mythology as both unique and universal among the world's mythologies has been eloquently voiced by George Steiner, who proposes that the Greek language and the Greek myths are inseparably linked and that the syntax of the Greek language and the myths expressed therein reflect basic, universal human experiences and forms of expression.¹ In Steiner's view this basic bond between Greek mythology and the Greek language creates a collection of myths which incorporates all the diverse modes of human language and experience. In short, the Greeks have said it all and little original material has been left by the Greeks for later generations to develop. Hence the special place of Greek mythology and the reason why the Greeks are different from the others.

Marcel Detienne has challenged this Hellenocentric view.² Focusing on “misunderstandings” between traditional Hellenists and structuralists, Detienne rejects the Hellenists' quest for the so-called original version of a myth in favor of his own structuralist preference for reading mythic multiforms within their ethnographic contexts. It is the variants that show the essence of myth as “recurrence and repetition in variation.”³ Such variation, caused by inevitable reworking of myth to adapt to changing times and different locales, emphasizes a bond between myth and society, long recognized, especially by Malinowski in his well-known theory of charter myths.⁴

As the previous chapters show, muthos emerges and develops in the context of a society, a polis. Each polis could and did adapt traditional tales to changing social and political needs. Greek myths reflect the society in which they flourished and reveal a tension between the poleis and pan-Hellenism, between local diversity and pan-cultural unity. Pan-Hellenism necessitated the rejection, revision, and reinterpretation of many local myths.

In this tension between local variants and pan-cultural canonicity the Greeks are not unique. Similar tendencies can be noted in medieval Europe and in the Islamic world. In all three contexts there is a diversity of local multiforms overspread and controlled by a dogmatic religious and cultural center. Just as Delphi and Athens often functioned as processing centers for a pan-Hellenic version of a local myth, so too did medieval Rome serve as the religious and cultural unifier of a disparate and diverse Christian Europe. Indeed, Rome has long maintained the authority to canonize or reject particular beliefs and doctrines. A combination of an imprimatur and an index of forbidden books in the Roman church aimed at determining the official versions of hagiographic and other texts. In a similar way, Mecca has functioned for the Islamic world, not only as a religious center, but also as a source of many pan-cultural myths. Yet in all three areas particular variants have continued to thrive despite such unifying tendencies. It is through the flexibility of mythic variants that broad cultural groups can maintain their sense of unity without completely sacrificing their sense of local pride.

Interest in the relationship between myth and society has led modern scholars to the living mythologies of the so-called underdeveloped countries. Such study is often primarily anthropological in nature and, until very recently, it started from an assumption of Western cultural superiority which labeled such cultures “primitive.” Yet all human cultures are now recognized to be much closer in terms of intellectual development, and “primitive” has come to be seen essentially as an invalid assessment of the myopic West.⁵

In this chapter I first examine the relationship between myth and society in one Third World culture, that of the Manding of West Africa, and then compare this study of Manding poetry in its social context with the Greek experience. A tension between poetry and society, between muthos and polis, is revealed in several significant features of Manding oral poetry, in which the dynamics of multiformity versus canonicity operate within the context of the limited literacy given in that society. At the end of this chapter I compare briefly the role of oral poetry in Manding and Greek cultures and consider the relationship between literacy and canonicity in both cultures.⁶

Manding society possesses both a vigorous pan-cultural tradition and a constant pull toward diversity. Crossing several modern political boundaries, including those of Mali, the Gambia, and Ivory Coast, and speaking a number of closely related languages, such as Malinke, Bambara, and Soninke, the Manding are an ancient people with a long and well-established sense of history and a living oral literature.

The Manding oral tradition includes praise songs, creation myths, circumcision songs, hunting songs, and heroic legends.⁷ Sufficient material has been collected, especially of Manding heroic legends, to demonstrate the existence of diverse multiforms within this oral literature.⁸ These multiforms tend to be localized in origin; that is, each Manding area has its own preferred version of the tradition. The persistence of such multiforms is due to an oral tradition that allows for local diversity and for adaptation over temporal and spatial distance. The Manding peoples believe that their oral stories retell the experiences of their common past, yet the diversity of their multiforms shows the ability of these stories to adapt to changes of time and locality.

Many of the features of oral poetry noted in modern Yugoslavian and ancient Greek epics by Albert Lord can be found also in the Manding tradition.⁹ Conservatism and fluidity create a dynamic tension within this oral culture, which simultaneously retains its traditional stories and adapts them to changing performative situations. Oral literature represents both past and present transformations of society. It is a record of the past as perceived in the present. Essentially performative by nature, such oral literature is recreated each time it is presented to an audience. No two oral performances of the same myth are ever identical word for word. Manding singers are aware that multiforms of the same story exist; for example, one singer says:

You see one griot,(10)
And he gives you an account of it one way,
And you will find that that is the way he heard it;
You see another griot,
And he gives you an account of it in another way,
And you will find that what he has heard has determined his version.

Innes 1974.145, ll. 8-13.

Built into the griot's statement is the assumption that there is a common core to this legend, upon which multiforms are built.

Charles S. Bird has noted several distinctive features of the Manding language. One characteristic is a marked cohesiveness within the Manding language group—a “high degree of mutual intelligibility among all the dialects.”¹¹ Manding speakers are able to communicate easily with each other, even while using different dialects. A parallel can be noted here with ancient Greece, where dialects were also mutually understandable. Such linguistic cohesiveness provides a ready means to establish unity within diversity and is thus critical for the creation of a common culture.

Bird accounts for the cohesiveness of the Manding group by a linguistic “wave theory” characterized by both a core of grammatical rules which all the dialects share and a diverse group of overlapping rules shared only by particular areas.¹² This wave theory reflects the Manding's simultaneous diversity and unity on the level of language.

Linguistic simplicity, avoidance of redundance, is another important feature of Manding. Bird argues that such simplicity made Manding an important trade language and fostered extensive communication both within and beyond the Manding-speaking region.¹³ This linguistic feature provides further structural support for a unified Manding society.

Communication in the Manding area has been encouraged, too, by a relatively stable political situation lasting for several centuries. Bird points out that the Manding peoples have always been associated with great empires that facilitated trade in the area and provided protection from some of the violent vicissitudes of war.¹⁴ The Manding empire also fostered a common political bond, creating a great source of pride, identity, and unity for its diverse population.

All of these characteristics of the Manding, which Bird notes in the context of extralinguistic factors affecting linguistic change, also illustrate a significant tension between diversity and unity on linguistic, political, and cultural levels. The Manding are a group of many peoples speaking separate dialects, but with a common language and history. While the Manding peoples see themselves as sharing a mutual heritage and genealogy, their common culture, retaining distinctly local elements, maintains, from tribe to tribe and place to place, differing versions of this heritage.

The primary vehicle for the transmission of this common culture is Manding oral literature, the centerpiece of which is an epic about the hero Sunjata (1230-1255), founder of the Manding empire and its greatest chief.¹⁵ Remembered in West Africa through oral tradition for seven centuries, Sunjata's life receives some independent evidence in the chronicle of Ibn-Khaldūn, a fourteenth-century Arabic writer.¹⁶ Although the historical basis of the hero myth, that is, Sunjata's actual existence, is not questioned by modern historians, the legend itself has been handed down orally with noteworthy variation according to locality. The entire oral tradition, filled with contradictory details, cannot be accepted as fact; it is simply too fluid, too marked by local variants. Sunjata the hero has become more important than Sunjata the man. He has become the tradition's unifying element and the genealogical center to whose life and history all the various ruling families have traced their ancestry. Sunjata's legend has become the cultural linchpin that unites diverse traditions, genealogies, and histories.

Multiforms of the Sunjata contain several thematic parts: the miraculous conception, birth, and childhood of the hero, his exile, and his defeat of the Susu king Sumanguru to become ruler of the Manding. These sections may be expanded or contracted, but usually they appear in every version. To this narrative core, several subtales are sometimes added, including, at the beginning, the story of Sunjata's mother and the source of her special powers, and at the end, a narration of various battles and events following the defeat of Sumanguru. Some discretion is thus allowed in the presentation of the tale.

The length of the Sunjata varies considerably from version to version. Bird notes that one version may be told in a single evening whereas another may take up to thirty hours to recite.¹⁷ Such variation in length is a general characteristic of oral epic poetry notable too in Greece and in Yugoslavia. Shorter and longer versions, both equally valid depending upon context, can exist for the same episode. Lord says of a Yugoslavian singer that “the length of his songs and the degree to which he will ornament them will depend on the demands of the audience.”¹⁸ Lord also notes that Parry's interest in Peter Vidić's song resulted in a second version almost twice as long as the first.¹⁹ An example of a Greek tale told in both longer and shorter versions can be found by comparing Odyssey ix-xii with Odyssey xxiii 300-343.

The epic nature of the Sunjata has been the object of much discussion as a result of Ruth Finnegan's controversial “Note on Epic.”²⁰ Although Finnegan argues in this brief treatment of the subject that no “relatively long narrative poems” have been composed in Africa, there is no question that both the Sunjata and other African poems have many affinities with such primary, oral epics as the Iliad and Odyssey or those in the Yugoslavian tradition.²¹ Finnegan's negative statement concerning African epic is based upon texts that are not reliably transmitted. A good example of such poor transmission is Niane's edition of the Sunjata (1965), which offers a reworked prose form very different from the performed poetic version. On the other hand, multiforms of the Sunjata such as those recorded by Innes (1974) and by Johnson (1979, 1986) exhibit, in fact, the poetic language that Finnegan thinks African epics lack.

Like the Homeric epics, the Manding epic uses a special poetic language that separates it from everyday speech. Bird has noted the restricted nature of Manding epic diction, which is both archaic and dialectal.²² The language of the epic is thus simultaneously conservative and local in nature. Like the Manding language itself, the diction of the Sunjata is both unifying, as the language of a national epic, and diverse, in its preference for dialectal forms.

Several aspects of the Sunjata particularly illustrate the tension between diversity and unity in Manding society: the effect of the audience upon the variant, the relationship between the past and the present within the epic, and the conflict between Islam and animism. In each of these areas can be noted a balance between variants, an acceptance of multiforms recognizing the coexistence of diversity and unity within the same epic. The Sunjata thus becomes a primary vehicle for expressing both local individuality and national conformity in a single context.

The relationship between audience and variants in the Manding epic is especially revealing. Perhaps in no other oral epic tradition is there such significant evidence for the way a particular audience can affect the performance of the singer and the multiform that is created. Thanks to the care with which field researchers such as Gordon Innes and J. W. Johnson have recorded the circumstances of particular performances of the epic, it is now possible to illustrate how the composition of the audience can determine the multiform. I examine such variation of song in several different versions.

The first is Bamba Suso's version, recorded by Innes (1974) as it was performed at a school in Brikama in the Gambia.²³ This multiform has a notable emphasis on genealogies, aetiologies of surnames and the foundation legends of towns near Brikama, whereas all of this information is notably lacking in other variants. These features may be due to the fact that the griot was asked to tell the pupils “something of the history of their people” and “to give as full an account as possible.”²⁴ Brikama itself receives special notice, although it is not mentioned in other variants. Most notably, in Bamba's multiform, Brikama is settled by Sankareng Madibe Konte, Sunjata's maternal grandfather and one of his generals.²⁵ The present chief of Brikama claims descent from Sankareng.²⁶ In this context it is significant that the special arrow that causes Sumanguru's defeat is shot by Sankareng in Bamba's song and not elsewhere.²⁷

The second version I refer to is that of Dembo Kanute, also recorded by Innes. This is the only published variant not gained from contrived circumstances, that is, not sung for and before a fieldworker with recording equipment. No outsider was present at the performance of Dembo's song, sung at the griot's own suggestion. The host himself, Seni Darbo, taped the performance.²⁸ In Dembo's version, it is a Darbo who shoots the arrow that defeats Sumanguru.²⁹

A third example of the relationship between the site of the performance and the song can be seen in the version of Magan Sisòkò recorded by Johnson (1979). Sisòkò, singing at Kita in Mali, interrupts the narrative in order to sing the praises of this town:

This is Kita.
Mount Kita and Budòfò.
O city of Magan, the Tall!
O city of holy men!
Tayakun and Tayaba.
This is Kita!
Kita's ancient name was Mount Geni.
Let us turn to Kita now.

Johnson 1979.127-128, ll. 2130-2137³⁰

A foundation legend of Kita follows these lines.

Clearly the real pressures of local interests and individual listeners have determined the forms of Bamba's, Dembo's, and Sisòkò's songs. The singers must balance the tendency toward a unified tale of Sunjata with the demands of local pride and relevance.

Integration of more recent historical events into the traditional epic is another feature of the Sunjata. Allusion to nineteenth-century jihads or Moslem religious wars, and to leaders such as the insurrectionist Musa Molo is sometimes woven into the narrative of Bamba Suso's song.³¹ Fode Kaba, archenemy of both Molo and the French colonials, is mentioned in the songs of both Bamba Suso and Banna Kanute.³² These references provide historical and thematic links between the distant epic past of Sunjata and the twentieth-century audience. Just as in Homer, anachronisms easily slip into the narrative: for example, the use of guns in Sunjata's battles, which took place before the introduction of gunpowder.³³

A striking illustration in the Sunjata of the tension between diversity and unity is the integration of Islam into the originally animistic epic. Here one can see the traditional religion of the Manding coming to terms with the increasing presence of Islam within its society. Islamic influence seems to have appeared in the area as early as the century following that of Sunjata, when one of his successors, Mansa Musa (1373/4-1387/8), made a pilgrimage to Mecca, the rumor of which reached even European ears. Despite the Islamic faith of Musa, Sunjata himself was probably not a Moslem and Manding rulers and peoples may have retained their animistic faith for several centuries. Only in the early nineteenth century did Islam begin to make larger inroads in the area, mostly due to population upheavals and traumatic jihads in the 1840s.³⁴ We can therefore assume that the largest integration of Islamic themes into the Sunjata occurred after this period.

Animistic themes are prominent in the Sunjata. Both Sunjata's mother and Sumanguru possess special magical powers conceived animistically. In several multiforms, a significant portion of the epic traces the animistic powers of Sunjata's mother back to the jinn of a buffalo woman.³⁵ The animistic powers of Sumanguru, who can be conquered only by a counterspell, are also a constant of the epic. Islamic influence emerges in other parts of the multiforms. In Magan Sisòkò's song, Sunjata's genealogy shows Islamic elements and Sunjata becomes a descendant of the Prophet.³⁶ In Banna Kanute's song, the first four Manding tribes to convert to Islam become major sources of support to Sumanguru in his conflict with Sunjata.³⁷

The tension between animism and Islam remains essentially unresolved both in the epic and in Manding society. Islam has not yet succeeded, if it ever will, in displacing all the animistic elements of the epic. The appearance of Islamic themes in the traditional Sunjata epic illustrates the way the historical process reflects social pressure within the epic. Because of changes in Manding society since the nineteenth century, Sunjata himself has developed Islamic ties, despite the historical improbability of such links.³⁸

The Manding singer shows himself aware of and affected by modern political pressures. Praise of current political leaders is not uncommon,³⁹ and fieldworkers have mentioned that modern politicians often patronize griots in the expectation that the singers will interweave references to them into their songs.⁴⁰

Traditionally the jeli was a member of a hereditary caste of singers who were linked, family by family, with specific tribal rulers. Thus a Kuyate has always been the jeli of a Keita ruler of Mali.⁴¹ The jeli was considered to be, not merely an entertainer, but also the preserver of tribal heritage through his song and the primary source of praise for his ruler. Furthermore, the traditional jeli served as the chief's spokesman or herald and advisor. Bala Faasigi Kuyate, Sunjata's griot, frequently has in the epic the role of the hero's spokesman; for example, it is Bala who calls Sunjata's troops together for him in Bamba's version of the Sunjata.⁴² In return for his services, the jeli looked to his chief as a patron who supported him financially. The dependence of the griot upon his chief is illustrated by a destitute Sunjata's efforts to support the griots in Bamba's version, where it is said that

They went and begged from him.
When they went and begged from him,
He did not have anything.
He went and got honey in the bush,
And brought it back for the griots.
Whatever he gave them, they did not scorn it.

Innes 1974.47, ll. 142-147

Religious ceremonies, such as the reroofing of the sacred hut in Kaaba (Kangaba) in southern Mali, have been important in the training of griots by providing opportunities for the sharing of songs among griots from different regions. Bird suggests that such occasions “exert a normalizing force on the society's linguistic behavior.”⁴³ These events may have also served as a normalizing force in the development of the Sunjata epic, as a means of national control over local multiforms. In this sense, Kaaba may, like Delphi, Rome, and Mecca, be an example of a center for pan-cultural assimilation of regional myths.

The advent of colonial rule and the deterioration of the power and wealth of tribal chiefs have meant a parallel decrease in the modern griot's duties and social status. Traditional occasions for performance of the griot's song have become more and more rare, and, without wealthy chiefs to support them, the singers have found it difficult to retain their position in society. The following lament, voiced by Banna Kanute, reveals the singer's frustration with the current situation:

In Sunjata's day a griot did not have to fetch water,
To say nothing of farming and collecting firewood.
Father World had changed, changed.

Innes 1974.159, ll. 320-322

The griots have lost not only their financial support but also several of the functions upon which their status was founded. No longer are they advisors and spokesmen to kings and chiefs. As noted above, the modern political situation has only marginal use for a jeli who is sometimes employed by aspiring politicians to sing their praises. To support themselves some griots have tried to adapt their song to different media, such as radio, in their own countries. Indeed the very survival of the singers and their oral tradition is currently in jeopardy. The pressures of the modern world may yet force the Sunjata into a static, uniform version, and the epic may take on pan-Manding features by losing its fluidity and variety. Changes in society and the influence of modern media may create what several centuries of oral tradition never did: a canonical, national form of the Sunjata.

The traditional role of the griot frequently emerges from the epic itself. The modern singer often reminds his audience of the once-important status of the jeli; for example, in Bamba's song, the only part of his father's property that Sunjata wants are the griots.⁴⁴ The prominence of the jeli is built into the very structure of the epic, in which the hero develops a close bond with his singer. In Bamba's version Sunjata even serves his unsuspecting singer flesh from his own thigh when both are starving on the savannah.⁴⁵ Frequently in the variants Sunjata is unable to accomplish his tasks without the assistance of the singer. Usually the singer's aid consists of verbal encouragement.⁴⁶ At the least, Sunjata needs his singer to voice his praises.⁴⁷ The jeli can also use his song to exhort an unwilling or incapacitated hero to action, or even to transform the hero's uninspiring or disreputable actions into praise. When Sunjata runs away in battle, the griots are able to save the hero's reputation by making the incident part of Sunjata's praise name.⁴⁸

The contrast between the griot's role in the epic and his place in the modern world thus illustrates the tension between a tolerant tradition that accepted diversity and a modern world that strives for homogeneity.

The hero's ties with other characters in the epic and with other members of his society reveal a great deal of variety and interchangeability of roles from multiform to multiform. In one version Sunjata performs a deed that is done by another character in another multiform. The most striking example of this interchangeability occurs in the balafon scene. In this episode, the first balafon, a type of xylophone, is discovered and played.⁴⁹ In Banna's version, it is Sunjata who obtains the balafon and plays it.⁵⁰ In Dembo's version, it is Sumanguru who performs the same deed.⁵¹ In several multiforms, the griot then comes upon the instrument, plays it without permission, and is hamstrung by the owner, whether Sunjata or Sumanguru.⁵² The parallels between these versions suggest a very close link between the hero and his enemy in the epic.⁵³

Another episode of the epic showing role interchangeability is the firing of a special arrow to defeat Sumanguru. In Banna's version it is Sunjata who makes the shot.⁵⁴ But in Bamba's song it is Sunjata's maternal grandfather, Sankareng.⁵⁵ Dembo sings that yet another man, the Darbo ancestor of his host, is the slayer of Sumanguru.⁵⁶

The Sunjata is an epic that praises the community as a whole, and its hero is someone with whom everyone in his society is closely associated. Not only is Sunjata, as the victor over Sumanguru, the savior of his society, but, as the representative of his society, Sunjata also becomes, through his victory, the means to the heroic apotheosis of his whole society. His epic becomes the Manding national epic and the source of unity out of diversity.

Through its hero the Sunjata epic becomes a communal experience that unites all members of society, not only in the epic past but also in the performative present.⁵⁷ The epic is more than the tale of its characters; it is at the same time about its audience. The bond between the Sunjata and the audience is created by various literary features of the Manding epic. Sometimes the audience is integrated into the epic in direct address by the griot. Such references to the live audience are not uncommon in the Manding epic, in the form of an address. During Bamba's performance the musical accompanist asks the griot questions and receives responses.⁵⁸ In several versions the singer directly addresses his host several times. Banna addresses his host and wishes him long life.⁵⁹ Dembo and Sisòkò end their songs by singing directly to their hosts.⁶⁰ These references can be at times exhortations to do something in life worthy of memory, such as the following plea by Dembo to his host:

Seni Daabo, do something, sir;
Life consists of doing something,
Not of doing everything,
For there is no end to that, and failure wins no support.

Innes 1974.269, ll. 67-70

These references can also be critical, such as the following rebuke by Banna:

But the griots have a complaint, Demba:
Don't you know
That an ordinary narrator and an expert singer are not the same?

Innes 1974.201, ll. 1264-1266

Direct address in the Sunjata is often linked with the use of praise songs. Sung in a style different from the narrative sections of the epic,⁶¹ these passages of praise are usually directed toward characters in the epic. For instance, a standard praise song of Sunjata, based upon the names of the hero's father and mother, is

Sukulung Kutuma,
And Sukulung Yammaru,
Naareng Makhang Konnate,
Cats on the shoulder,
Simbong and Jata are at Naarena.

Innes 1974.149, ll. 116-120

Members of the audience can also be included in such praise songs. Banna sings the following praise of his host, Bakari Sidibe:

Praise be to God, Master of the worlds.
Kibili Demba, thank you.
Salimata Dembo,
Husband of a Jebate woman, Demba, thank you.
If you take it from here,
You must join it to Karata.
You come from
Jallo and Jagite,
Great Sidibe and Sankare.
Dembo, who used to live with the people of the Island,
Dembo, son of a Sanyang woman, thank you.

Innes 1974.201, ll. 1253-1263

This passage combines praise of Sidibe with references to his ancestors and his place of origin and thus makes the performance of personal interest to the singer's audience. The Sunjata serves as a mechanism of praise simultaneously for past and present members of Manding society.

Past and present are also united in the epic by association of various towns, regions, and family names with episodes in the hero's life. Thus, in Bamba Suso's song, the foundations of many towns in the Gambia are traced back to deeds of Sunjata and his generals.⁶² Particularly common are the aetiologies of surnames, which are sometimes even traced to actual phrases spoken by Sunjata. Bamba tells the following story to explain the origin of the family name Noomo:

When Sunjata had been told what had happened,
He sent Tira Makhang,
With orders to proceed against that king.
Tira Makhang captured the king,
Destroyed his residence,
Bound him,
And took him to Sunjata.
When Sunjata saw him,
He said to him, “Noomo” ‘Defeated’.
Even today the surname Noomo is in existence.

Innes 1974.83, ll. 932-941

For the listener with the surname Noomo, this passage makes the epic real and personal; thus the modern Manding listener becomes a part of the Sunjata.

References to several Manding social customs within the epic create further points of identity for the audience. The ceremony surrounding the circumcision of young boys is an important tradition among the Manding and becomes, in several versions of the Sunjata, the occasion when the crippled hero learns to walk.⁶³ In Bamba's song the circumcision ceremony is linked not only with the walking episode but with another test based upon magic trousers.⁶⁴ In Banna's version the references to the circumcision rite itself are lengthier and more detailed and include several attempts by the antagonist to dispose of his enemy.⁶⁵ Plot and social custom are combined in these passages to create for the Manding audience a further link between the hero and themselves.

Manding family and clan structure are integrated into the epic. Polygamy sometimes dominates the story through a bitter rivalry of Sunjata's mother and another wife of his father.⁶⁶ Maternal ties are particularly important in Manding culture and are emphasized in the epic by Sunjata's ties with his mother.⁶⁷ In Niane 1965 and Johnson 1979 the hero's special powers are derived explicitly from his mother's animistic associations with the jinn of a buffalo woman. The exiled hero's return home and victory over Sumanguru are also frequently linked with the impending death of his mother. The most elaborate use of this theme appears in versions where the hero demands and receives two supernatural signs joining his mother's death with Sunjata's victory over Sumanguru.⁶⁸

Ties between brother and sister are likewise strong in Manding society and are reflected in the close relationship between Sunjata and his sister Sugulun, another necessary adjunct to Sunjata's heroic power. One of the standard and most memorable episodes of the epic is the seduction of Sumanguru by Sugulun, who thus obtains for her brother the knowledge necessary to defeat the Susu king.⁶⁹

Relationships between modern clans and tribes are explained and reinforced in multiforms of the Sunjata; for example, a strongly Islamic version gives great prominence to the four Manding families that first converted to Islam.⁷⁰ Traditional ties between certain ruling families and certain griot families such as the relationship between the Kuyates and the Keitas,⁷¹ appear in other variants.

All of these features of the Sunjata show how the West African epic operates as a social unifier joining diverse peoples in the present by creating communal unity with the past. The living epic is a dynamic link between past and present, between hero and audience, between local diversity and “national” unity.

The Manding culture shares with the Greek culture a dynamic blend of local and national trends. Tension between diversity and unity is especially strong in the Manding region, which encompasses an area larger than France and includes over four million speakers. The Manding language serves as a principal medium of communication in a significant part of West Africa, and the large number of speakers and their broad geographic spread are considered unusual in an African context. So also the Greek language became widespread in the Mediterranean world beginning at least as early as the age of colonization in the eight century b.c., and it became the lingua franca there in the Hellenistic Age. Manding is thus a special case in Africa, just as Greece was special in the ancient Mediterranean world.

This social and cultural dynamic within the oral Manding tradition parallels the relationship between Greek muthos and polis, especially in the early stages of pan-Hellenism before local multiforms were displaced by more generic, pan-Hellenic versions. A good example of this process in Greece is the canonization of the tale of the conflict between Achilles and Agamemnon over such variants as the conflict between Achilles and Odysseus at Troy.⁷² The multiformity of these quarrel scenes in the Greek tradition calls to mind a similar multiformity in the Sunjata epic, where particular deeds are attributed to different characters in different variants. The Greek tradition thus reflects a process of selection and canonization which the Manding tradition lacks.

The Manding epic differs from the Greek epic tradition in its extensive use of direct address. Homer's audience hardly ever intrudes into the Iliad or the Odyssey, whereas the membership of a particular Manding audience can even determine the specific multiform used by the singer. The exclusion of the audience from the Greek epic may be the result of canonization, which is inevitably a process of universalizing the audience. It is possible that earlier multiforms of the Greek tradition prior to canonization contained references to a specific audience, as do Odysseus' tales to the Phaiakians and Eumaios.⁷³ Pan-Hellenism, however, brought the need to remove localized features of the epic, and the audience lost its diversity as the epic itself was made more universal.

In one significant respect the Manding epic tradition is different from its Greek counterpart: it lacks a fixed text. Despite the inroads of literacy and mass media in the twentieth century, the tradition has remained oral and there is as yet no discernible tendency toward canonization of versions. Local variants are very strong.

Particularly noteworthy in the West African context is the coexistence of oral literature and limited literacy. There is a long tradition of literacy in the Manding area, restricted to particular strata of society in an Islamic context. Universal literacy was never the goal of Manding society, where only the marabout, or Islamic teacher, needed to know how to read in order to study the Koran. The function of literacy in Manding society has thus been primarily religious, not secular.⁷⁴ In a similar way, literacy was generally restricted to the religious classes of medieval Europe, where not even the ruling nobility were necessarily literate. In the Manding area, oral tradition, not writing, has been the vehicle for preserving and remembering past events. Thus, in this part of West Africa, castes of unlettered poets—those generally known to us as griots—transmitted oral history to a largely illiterate society that received Islamic teaching from a literate religious class (the marabouts).

The existence of an oral literature in Manding culture with its limited literacy raises important questions about the relationship between the introduction of writing in Greece and the establishment of the Homeric texts. It is often thought that the appearance of the Greek alphabet caused immediate transcription of oral literature such as the Homeric epics and that there was a clear-cut transition in Greece from an illiterate or oral society to a literate one. Goody assumes an unusually rapid increase in literacy in Greece between the eighth and fifth centuries b.c. and an unusually high literacy rate in fifth-century Greece. Yet there is no firm evidence for this. Certainly literacy was common in fifth-century Athens, especially among citizens and metic merchants, but what percentage of the population did this represent?⁷⁵ Athenian society had a very restricted citizenship that excluded all women and slaves. Preinscribed ostraka ‘shards’ discovered in the Athenian agora ‘place of assembly’ suggest that literacy was not all that common among citizens, even in the fifth century,⁷⁶ and it is not insignificant that Greek literature until the end of the fifth century was primarily performative in nature, meant to be presented orally even if not so composed. Such performative literature includes epic, lyric, tragedy, and comedy. In this context, it is also noteworthy that one of Athens' greatest philosophers, Socrates, wrote no philosophical treatises, but taught orally, and is known only through the writings of his pupils Plato and Xenophon and the burlesque comedy of Aristophanes. Even some of the treatises of Aristotle, a century later, survive only via a semi-oral medium in the form of lecture or student notes, not in Aristotle's own published text. In the case of the Greek mythic corpus, it is not unlikely that Hesiod and Homer existed in fixed, oral versions decades before these works were written down. Rhapsodes, reciters of fixed Homeric texts, were probably active in Greece long before the first known Homeric recension by the Peisistratids.⁷⁷ Clearly literacy did not displace oral composition and presentation in Greece overnight.⁷⁸ The transition from an orally oriented society to a literate one is a slow, subtle, and not necessarily inevitable process, as can be demonstrated in the case of the Manding.

Ruth Finnegan especially questions whether narrative form is a prominent feature in such African epics as the Sunjata, which is, to a large extent, interspersed with praise songs. Yet the frequently panegyric nature of the Sunjata does not detract from its narrative focus on the life of Sunjata; rather, as Johnson has noted, the combination of narrative with praise song emphasizes the multigeneric nature of African epic.⁷⁹ Nagy has pointed out the relationship of epos ‘poetic utterance’ and ainos ‘praise’ in the Homeric tradition.⁸⁰ Mixing of genres is thus a structural expression of diversity and unity within both the Manding and Greek epics.

Johnson also emphasizes the multifunctional features of the Sunjata, which offers a model for Manding social relations, especially among clan families.⁸¹ The epic reinforces, too, certain social and religious institutions and practices, such as circumcision, and serves as a medium of national unity among the Manding. Yet even this characteristic is not restricted to Africa, for the Greek epic served similar multifunctional purposes, especially in its pan-Hellenic stage as the Greek national poem.⁸²

The Sunjata centers around a single character in much the same way that events in the Iliad turn around Achilles, who sulks in the background for much of the first half of the epic while other characters, such as Menelaos, Diomedes, and Ajax, occupy the limelight. However, unlike the heroes described in Sir Maurice Bowra's heroic world, which is isolated from the real world, Sunjata's actions do not set him apart from his contemporary society and from the modern Manding world.⁸³ He is not a hero who is different and solitary. He is a Manding hero because he is an active member of his society and because other members of his society help him. West African epic emphasizes, rather than the isolated individual as a hero, the heroic individual as a member of a community of equals. Sunjata cannot defeat Sumanguru without the direct assistance of his sister, his mother, his griot, and the community itself. Sunjata's success is their success, too.⁸⁴ Unlike the Greeks, for whom there was an unbridgeable gap between the present and the Heroic Age, the Manding have a sense of communality with their past. Sunjata is as much a hero of the present as he is of the past.

Sunjata's ability to unify Manding people across temporal boundaries has been noted by Massa M. Diabaté, a prominent Malian scholar and poet, who has said that the Sunjata epic “attributes the entire cultural experience of a society to one character who has made a mark on his time … and deriving all past and present values of that society from this character, thus rendering the epic a source of identity serving to distinguish that group from others.”⁸⁵ Analysis of the Sunjata epic in terms of Manding society thus illustrates important features of an oral tradition in its social context. The tension between diversity and unity which exists in the Manding tradition provides significant perspective for the Hellenist and demonstrates that the bond between muthos and polis is not unique to the Greek experience. Every oral society creates its own peculiar relationship between its myths and its sociocultural contexts. The Greeks are indeed like the others.

Notes

1. Steiner 1984.135-138. For example, Steiner argues that the myth of Prometheus is connected with the future tense and that of Narcissus with the first-person singular verb.

2. Detienne 1979. The title of this chapter alludes to Detienne's sarcastic title: “The Greeks Are Not like the Others.”

3. Detienne 1979.6.

4. See Malinowski 1931 or excerpts in Lessa and Vogt 1979.63-72.

5. The similarities between “civilized” and “primitive” are discussed in Goody 1977.

6. Not the least significant of the parallels between the West African Manding culture and the ancient Greek culture is that the Manding poetic tradition is the primary vehicle of myths of foundational value to the Manding society.

7. On circumcision songs, see Innes 1972; on hunting songs, see Bird 1972; on creation myths, see Dieterlen 1957.

8. On multiforms of Manding heroic legends, see Zemp 1966.

9. See Lord 1960.

10. While a singer of songs like the Sunjata (the equivalent of a Greek aoidos ‘bard’) is known in some parts of the Manding world as a jeli, “griot” is more common in English. The etymology of “griot” is mysterious. Possibly derived from the Portuguese word criar ‘to educate, to instruct’, “griot” comes into English from the French griot, widely used in the 1960s to refer to West African singers. There is an extensive bibliography on the etymology of griot. See Labouret 1951, who argues a Portuguese provenance.

11. Bird 1970.148.

12. Bird pp. 150-152.

13. Bird p. 154.

14. Bird p. 153.

15. English spellings of Manding words, including proper names, vary greatly. Thus Sunjata also appears as Sundiata, Sun-Jata, and Son-Jara.

16. Ibn-Khaldūn 1868.

17. Bird 1970.156-157.

18. Lord 1960.25.

19. Lord p. 113.

20. Finnegan 1970.108-110.

21. On the epic form in Africa, see Biebuyck 1968, Biebuyck and Mateene 1978, Okpewho 1979, Johnson 1980 and 1986, and Seydou 1983.

22. Bird 1970.157.

23. See Innes 1974.37-38 for a description of the circumstances of Bamba's performance.

24. Innes p. 370.

25. See Innes p. 85, l. 958 for Sankareng's settlement of Brikama.

26. Innes pp. 128-129, note to l. 957.

27. Innes p. 79, ll. 836-840.

28. See Innes pp. 264-265 for a fuller description of the circumstances of this performance.

29. Innes p. 309, ll. 956-972.

30. See also Johnson 1979.162, l. 2889.

31. Innes 1974.97-99, ll. 1223, 1240, 1242, 1248, 1297.

32. For Bamba's references to Fode Kaba, see Innes 1974.97-99, ll. 1240, 1242, 1249, 1280. For Banna's, see Innes 1974.185, ll. 895, 896, 901.

33. See Innes p. 231, ll. 1918-1921.

34. For more detailed discussion of these nineteenth-century events, see Quinn 1972 and also Schaffer and Cooper 1980.

35. Jinn, an Arabic word borrowed by the Manding, refers to powerful animistic spirits. Jinn is also the source of the English word “genie.” Versions of the Sunjata with the buffalo woman episode are recorded by Johnson 1979 and 1986 and Niane 1965.

36. See Johnson 1979.34-37, ll. 1-72. The fluidity of Sunjata's genealogy within the oral tradition can be compared with the changing genealogies of the Tiv people, noted by Goody and Watt 1968.31-34.

37. These families are the Siises, the Jaanes, the Kommas, and the Tures. See Innes 1974.157-181. Innes discusses the prominence of Islam in this version on p. 241, note to l. 83.

38. Conrad 1985 offers a detailed discussion of the integration of Islam into Manding oral tradition.

39. Dembo refers to Fili Daabo, a Malian politician and Dembo's patron. See Innes 1974.267, l. 4 and 313n.

40. See Innes 1974.10-11.

41. The relationship between the Kuyates and the Keitas is illustrated in Bamba Suso's song (Innes 1974) and in Niane's 1965 version of the epic. Bamba especially emphasizes these ties with the following statement: “There is a special relationship / Between the members of the Keita family and the members of the Kuyate family. / Even today, if a member of the Kuyate family deceives a member of the Keita family, / Things will go badly for him” (Innes 1974.61, ll. 470-473).

42. Innes 1974.61-71.

43. See Bird 1970.156.

44. Innes 1974.45, ll. 125-127.

45. Innes pp. 59-61, ll. 437-475.

46. In Banna's song, Bala Faasigi sings Sunjata's praises as a song of encouragement as the hero prepares for his final encounter with Sumanguru. See Innes 227, ll. 1844-1863.

47. E.g., see Innes p. 155, ll. 244-246.

48. Innes p. 71, ll. 684-691.

49. On the importation of the xylophone to West Africa, see Jones 1971.

50. Innes 1974.209-213, ll. 1465-1526.

51. Innes pp. 273-275, ll. 133-203.

52. E.g., Innes pp. 213-215, ll. 1548-1564, and Niane 1965.38-40.

53. A similar relationship exists between Hermes and his antagonist, Apollo, in the Homeric Hymn to Hermes. The newborn Hermes, who had invented the lyre, stole the sacred cows of Apollo but returned them at the command of Zeus. Apollo was charmed by the instrument, which Hermes played and sang to, and thus the two gods exchanged privileges. Apollo, receiving the lyre from Hermes, bestowed upon him the staff that would mark him as cowherd and poet.

54. Innes 1974.231, ll. 1937-1939.

55. Innes p. 79, ll. 835-842.

56. Innes pp. 309-311, ll. 956-980.

57. Even so did ancient Greek performative poetry create kharis ‘mirth’, a bond between poet and audience and a communal act of the society to which they belonged. This can be illustrated with examples in epic (Odyssey ix 3-11) and, especially, in epinician poetry, such as Pindar Olympian 7.93-94, 14.13-16. For further illustrations of this phenomenon in oral cultures, see Edwards and Sienkewicz 1990.

58. Innes 1974.67, ll. 594-596; 69, ll. 614-625; 71, ll. 661-664, and 77, ll. 794-797.

59. Innes p. 171, ll. 574-579.

60. Innes p. 311, ll. 1016-1020, and Johnson 1979.194, ll. 3628-3631.

61. See Innes 1974.17-20 for a discussion of the different modes of song in the Sunjata.

62. For example, Innes p. 85, l. 958, the settlement of Brikama, where Bamba's performance takes place, is traced to Sankareng, Sunjata's general and grandfather.

63. On the importance of this ceremony, see Schaffer and Cooper 1980, and Innes 1972.

64. Innes 1974.43-45, ll. 85-121.

65. Innes pp. 181-207, ll. 845-1405. The circumcision episode covers about one-third of this version.

66. See Niane 1965 and Johnson 1979 and 1986 (version of Fa-Digi Sisòkò).

67. See Schaffer and Cooper 1980.87-90.

68. See Innes 1974.287, ll. 447-477, Niane 1965.45-47, Johnson 1979.165-172, ll. 2980-3147, and Johnson 1986.164-165, ll. 2410-2450.

69. See Innes 1974.73-79, ll. 693-842; 215-219, ll. 1565-1689; 303-307, ll. 829-913; Johnson 1979.182-183, ll. 3381-3415; also Johnson 1986.171-173, ll. 2668-2741.

70. See Innes 1974.157-181, ll. 265-844.

71. See Niane 1965.1 and Innes 1974.61, ll. 471-475.

72. Nagy 1979.42-58 has shown that the Homeric texts contain evidence for a multiform that focused on the Achilles-Odysseus conflict. Creation of a pan-Hellenic version necessitated the suppression of such a variant in favor of the Achilles-Agamemnon conflict upon which the Iliad is based.

73. See, e.g., Odyssey ix 2-18 and xiv 462.

74. On the association of literacy and Islam in West African countries, see Goody 1987.125-138.

75. See Harvey 1966.

76. See Harvey pp. 590-593 and Hands 1959.76-79.

77. For a detailed ancient illustration of a fourth-century rhapsode in action, see Plato's Ion. On performance of oral poetry in ancient Greece, see Nagy 1990b, chaps. 1 and 2.

78. See Havelock 1986.

79. See Johnson 1980.321.

80. Nagy 1979.222-242.

81. Johnson 1980.319-320.

82. On multifunctionalism in Greek myths, see Kirk 1974.

83. Bowra 1964.

84. See Bird 1976.

85. From Diabaté's unpublished doctoral thesis at the University of Paris. Translated by Seydou 1983.49-50.

Bibliography

Biebuyck, D. 1968. The Mwindo Epic. London.

Biebuyck, D., and K. Mateene. 1978. Hero and Chief: Epic Literature from the Banyanga, Zaïre Republic. Berkeley.

Bird, C. S. 1970. “Development of Mandekan (Manding): A Study of the Role of Extra-linguistic Factors in Linguistic Change.” In Languages and History of Africa, ed. D. Dalby, 146-159. New York.

———. 1972. “Heroic Songs of the Mande Hunters.” In African Folklore, ed. R. Dorson, 275-293. New York.

———. 1976. “Poetry in the Mande: Its Form and Meaning.” Poetics 5:89-100.

Bowra, M. 1964. “The Meaning of the Heroic Age.” In The Language and Background of Homer, ed. G. S. Kirk, 3-28. Cambridge.

Conrad, D. C. 1985. “Islam in the Oral Traditions of Mali: Bilali and Surakata.” Journal of African History 26:33-49.

Detienne, M. 1971. “Orphée au miel.” Quaderni Urbinati di Cultura Classica 12:7-23. In Gordon 1981. Cambridge.

———. 1979. Dionysos Slain. Baltimore. Trans. M. Muellner and L. Muellner of Dionysos mis à mort, Paris, 1977.

Dieterlen, G. 1957. “The Mande Creation Myth.” Africa 27: 124–38.

Edwards, V., and T. J. Sienkewicz. 1990. Oral Cultures Past and Present: Rappin' and Homer. Oxford.

Finnegan, R. 1970. Oral Literature in Africa. Oxford.

Goody, J. R. 1977. Domestication of the Savage Mind. Cambridge.

———. 1987. The Interface between the Written and the Oral. Cambridge.

Goody, J. R., and I. Watt. 1968. “The Consequences of Literacy.” In Literacy in Traditional Societies, 27-68. Cambridge.

Hands, A. R. 1959. “Ostraka and the Law of Ostracism: Some Possibilities and Assumptions.” Journal of Hellenic Studies 79:69-79.

Harvey, F. D. 1966. “Literacy in the Athenian Democracy.” Révue des Études Grecques 79:585-635.

Havelock, E. 1986. The Muse Learns to Write: Reflections on Orality and Literacy from Antiquity to the Present. New Haven, Conn.

Ibn Khaldūn, Walī al-Dīn ‘Abd al-Rahmān b. Muhammad. 1868. Kitāb al-‘Ibār wa-diwan al-mubtada qa‘l-Khabar. Bulak. = Histoire des Berbères et des dynasties musulmanes de l'Afrique septentrionale. Trans. Commission internationale pour la traduction des chefs d'oeuvre. Beirut, 1956–1959.

Innes, G. 1972. “Mandinka Circumcision Songs.” African Language Studies 13:88-112.

———. 1974. Sunjata: Three Mandinka Versions. London.

Johnson, J. W. 1979. The Epic of Sun-Jata according to Magan Sisòkò. Bloomington, Ind.

———. 1980. “Yes, Virginia, There Is an Epic in Africa.” Research in African Literatures 11:308-26.

———. 1986. The Epic of Son-Jara: A West African Tradition. Bloomington, Ind.

Jones, A. M. 1971. Africa and Indonesia: The Evidence of the Xylophone. 2d ed. Leiden.

Kirk, G. S. 1974. The Nature of Greek Myths. New York.

Labouret, H. 1951. “À propos du mot ‘griot’.” Notes Africaines 50:56-57.

Lessa, W. A., and E. Z. Vogt. 1979. Reader in Comparative Religion: An Anthropological Approach. 4th ed. New York.

Lord, A. 1960. The Singer of Tales. New York. Reprinted 1976. Cambridge, Mass.

Malinowski, B. 1931. “Culture.” In Encyclopedia of the Social Sciences, ed. E. Seligman and A. Johnson, vol. 4. New York.

Nagy, G. 1979. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry. Baltimore.

———. 1990b. Pindar's Homer: The Lyric Possession of an Epic Past. Baltimore.

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

Okpewho, I. 1979. Epic in Africa. New York.

Quinn, C. A. 1972. Mandingo Kingdoms of the Senegambia. Evanston, Ill.

Schaffer, M., and C. Cooper. 1980. Mandinko: The Ethnography of a West African Holy Land. New York.

Seydou, C. 1983. “The African Epic: A Means for Defining the Genre.” Folklore Forum 16.47-68.

Steiner, G. 1972. Language and Silence: Essays on Language, Literature and the Inhuman. New York.

———. 1984. Antigones. New York.

Zemp, H. 1966. “La légende des griots malinké.” Cahiers d' Études Africaines 6:611-42.