1. By “historiography” I mean the methodology of writing history. The structural study of myth, in this discussion, applies more to the comparison of variant myths than to their interpretation. For a cogent critique of structuralist methodology in African history see Vansina, Jan “Is Elegance Proof? Structuralism and African History,”in HA, 10(1983); 307–48.Google Scholar

2. Peel, J.D.Y., “Making History: The Past in the Ijesha Present,” Man n.s., 19(1984), 112.CrossRefGoogle Scholar

3. E.g., Miller, J.C., ed. The African Past Speaks: Essays on Oral Tradition and History (Folkestone, 1980).Google Scholar

4. See Beier's, Ulli “Festival of Images,” Nigeria Magazine, no. 45(1953); 14–20Google Scholar; “The Historical and Psychological Significance of Yoruba Myths,” Odu, 1(1955);17–25Google Scholar; “The Oba's Festival, Ondo,” Nigeria Magazine, no. 50(1956); 238–59Google Scholar; A Year of Sacred Festivals in One Yoruba Town (Lagos, 1959)Google Scholar; “Before Oduduwa,” Odu, 3[n.d.]; 25–31.Google Scholar

5. See Harms, Robert, “The Wars of August: Diagonal Narrative in African History,” The American Historical Review, 88 (1983);809–34.Google Scholar

6. Lloyd, P.C., “Yoruba Myths--A Sociologist's Interpretation,” Odu, 2(1955); p.28.Google Scholar

7. Johnson, Samuel, The History of the Yorubas (London, 1921)Google Scholar, 3. For a more detailed discussion of the organization of arokin into aro and onisekere groups, of their relationship to dundun and bata drummers, and to the akunyungba and ayaba praise singers, see Agiri, B.A., “Early Oyo History,” HA, 2(1975);3.Google Scholar

8. This “puzzle” is most recently discussed in Robin Horton's comprehensive “Ancient Ife: A Reassessment,” in JHSN 9/4 (1979);69–149, esp. 119–28.Google Scholar

9. Lloyd, P., “Yoruba Myths,” 21Google Scholar; R.C.C. Law “Traditional History” in Biobaku, S.O., ed;, Sources of Yoruba History (Oxford, 1973)25–40.Google Scholar Cf. B.A. Agiri, “Early Oyo History,” 7, who glosses creation myths as “cosmological” and migration myths as “political,” a misleading distinction if both types have political implications.

10. But cf. Obayemi, Ade, “Ancient Ile-Ife: Another Cultural-Historical Reinterpretation,” JHSN, 9/4 (1979);151–85Google Scholar, for the strongest critique of the early Ile-Ife hegemony thesis. Obayemi argues that Ife became important as the manufacturing center of glass segi beads used to make the beaded crowns (ade) of sacred Yoruba kings.

11. Law, “Traditional History,” 210.Google Scholar

12. Johnson, History, 11Google Scholar; cf. “Adimun” in Abraham, R.C., Dictionary of Modern Yoruba (London, 1946), 14.Google Scholar

13. Folk-etymologies of kingship and chiefship titles are really statements about political status and are usually without sound linguistic foundation. See Peel, J.D.Y., “Kings, Titles and Quarters: A Conjectural History of Ilesha; II.” HA, 7(1980);225–35 et passimGoogle Scholar, for general features of Yoruba political titles and particular features of the Ilesha title system.

14. I collected an interesting variant of this tradition in Akure, merging the Ife and Oyo versions, in which the first Ooni was the son of a female slave who was impregnated by Oduduwa.

15. Beier, , “Oba's Festival“ 238.Google Scholar

16. Johnson, , History, 25.Google Scholar

17. Mabogunje, A.L. and Omer-Cooper, J.D., Own in Yoruba. History (Ibadan, 1971), 13.Google Scholar

18. Oni, J.O., A History of Ijeshaland (Ife, n.d.), 12–15.Google Scholar

19. Johnson, , History, 22.Google Scholar

20. For Ede cf. MacRow, D.W., “Natural Ruler--A Yoruba Conception of Monarchy,” Nigeria Magazine, no. 47(1955); 239Google Scholar, and Johnson, , History, 125Google Scholar; for Egba see Biobaku, S.O., The Egba and Their Neighbours (Oxford, 1957), 4Google Scholar and Johnson, , History, 8Google Scholar; for Ijebu see Lloyd, , “Installing the Awujale,” Ibadan, 12(1961), 7Google Scholar and Johnson, , History, 20.Google Scholar

21. I have discussed the semiotic principles of tonal transfer in the Yoruba drum language, and the performative functions of its utterances, in two working papers; “Tone-Bearing Units in Yoruba Speech Surrogates” (1982) and “Yoruba Talking Drums and Ritual Communication” (1981). The best ethnological study of Yoruba drums is Thieme, D., “A Descriptive Catalogue of Yoruba Musical Instruments” (Ph.D., Catholic University of America, 1969).Google Scholar

22. Peel's observation that in Ilesha “the sequence of the festivals of royal ancestors might seem likely to preserve the order of succession, perhaps even be oral tradition's mnemonic for it…” illustrates how sequence in ritual performance can recapitulate and preserve sequence in oral tradition: Peel, , “Kings, Titles and Quarters; I,” HA, 6(1979), 132.Google Scholar

23. Lloyd writes that “when an oba is installed he often re-enacts the final part of the journey of his ancestor to the town…Along the route are often shrines at which annual sacrifices must be made, each such action recalling the myth which describes the original purpose of the shrine. It is for these reasons that the route is only remembered within the kingdom;” Lloyd, , “Yoruba Myths,” 25.Google Scholar Agiri, “Early Oyo History,” relates that among the Alaafin's arokin, “it is the duty of the Ologbo to recite at one of the crucial installation ceremonies of the new Alaafin the legend concerning the migration of the ancestors of the Oyo kings,” and that “in precolonial times, any unsatisfactory rendering of the traditions on this occasion carried very dire consequences.”

24. Lloyd, , “Installing the Awujale,” 7.Google Scholar

25. Goody, Jack, Succession to High Office (Cambridge, 1966).Google Scholar

26. Johnson, , History, 20.Google Scholar

27. R.C.C. Law The Oyo Empire c. 1600–1836 (Oxford, 1977), 136.Google Scholar

28. Ibid., 135–36.

29. In “The Sociological Role of the Yoruba Cult Group,” American Anthropologist, 46(1944); 1–75Google Scholar, William Bascom classifed orisa cults into two types, “single-sib” and “multi-sib.” Whereas single-sib cults demarcate agnatic descent groups, multi-sib cults transcend unilineal principles of recruitment, and there he lets the matter stand. Had Bascom realized that corporate descent groups (“sibs” in his terminology) constitute political segments of quarters (adugbo), and that these constitute political segments of towns, he would have identified political segmentation as the unifying basis of “single-sib” and “multi-sib” orisa cults.

30. Based on Morton-Williams, Peter, “An Outline of the Cosmology and Cult Organization of the Oyo Yoruba,” Africa, 34(1964); 243–61.CrossRefGoogle Scholar

31. Cf. Ajayi, J.F. Ade, “The Aftermath of the Fall of Old Oyo,” in Ajayi, J. and Crowder, Michael, eds., History of West Africa, II, (London, 1974), 129–166Google Scholar; Lloyd, P.C., “The Traditional Political System of the Yoruba,” Southwestern Journal of Anthropology, 10(1954); 366–84CrossRefGoogle Scholar; Ogunba, O., “Ritual Drama of the Ijebu People: A Study of Indigenous Festivals” (Ph.D., University of Ibadan, 1967)Google Scholar; Peel, J.D.Y., Ijeshas and Nigerians: The Incorporation of a Yoruba Kingdom, 1890s to 1970s, (Cambridge, 1983)Google Scholar; Barber, Karin, “How Man Makes God in West Africa: Yoruba Attitudes Toward the Orisa,” Africa, 51(1981), 724–45.CrossRefGoogle Scholar

32. Peel, , “Kings, Titles and Quarters, I” 131–3.Google Scholar

33. Local accounts of political fission in Ayede, Isan, and Itaji kingdoms are documented in my “Rituals of Power: Yoruba Orisa Worship in Three Ekiti Kingdoms,” (Ph.D., Yale University, 1987).Google Scholar

34. According to Lloyd, “The subordinate towns usually retain their original myths, but any attempt to assert them, and thus threaten the integrity of the kingdom, was, in the past, met by raiding and possible destruction of the town.” Lloyd, , “Yoruba Myths,” 25.Google Scholar

35. Based on Wescott, J. and Morton-Williams, Peter “The Symbolism and Ritual Context of the Yoruba Laba Sango,” Journal of the Royal Anthropological Institute, 92(1962), 23–37.Google Scholar“The Festival of Iya Mapo,” Nigeria Magazine, 58(1958), 212–24.Google Scholar Cf. Beier, “Ori-Oke Festival, Iragbaji,” Nigeria Magazine, no. 56(1958); 65–83, for ar account of a similar subordinate town ritual, that of Iragbaji in relation to Ilesha.Google Scholar

36. Not all subordinate towns were conquered; some originated as farm settlements, while others sought the protection of more powerful allies.

37. Lloyd, , “Yoruba Myths,” p. 24.Google Scholar

38. In my “Rituals of Power,” the same formal relations obtain between lineages and their cults within quarters, and between quarters and their cults within towns, which in this essay I identify between capital and subordinate towns and their dominant orisa cults. Each political unit has its own historical traditions (itan), preserved by ritual, which often rival the oba's official history of the kingdom.

39. The classic studies of Old Oyo include Ajayi, J. Ade. and Smith, Robert, Yoruba Warfare in the Nineteenth Century (Cambridge, 1971)Google Scholar; Akinjogbin, I., “The Expansion of Oyo and the rise of Dahomey” in Ajayi, /Crowder, , eds., History of West Africa, II 304–43Google Scholar; Law, Oyo Empire; Lloyd, P.C., The Political Development of Yoruba Kingdoms in the Eighteenth and Nineteenth Centuries (London, 1971)Google Scholar; Morton-Williams, Peter, “The Yoruba Kingdom in Oyo” in Forde, D. and Kaberry, P.M., eds., West African Kingdoms in the Nineteenth Century (London, 1967), 36–69.Google Scholar

40. Lloyd, , Political Development 9Google Scholar; R.C.C. Law, “A West African Cavalry State: The Kingdom of Oyo” JAH,(1975), 1–15.Google Scholar

41. Law, Oyo Empire, 84–85.Google Scholar

42. Lloyd, , Political Development 10.Google Scholar

43. According to Mabogunje and Omer-Cooper, “The Alafin's position as representative of Sango was exploited to the full as a means of supporting his authority. The Sango cult was spread to every town under Oyo influence and organised in a hierarchy centered in the palace at Oyo. The Alafin's Ajele were often themselves Sango priests. This added to their authority at the courts of vassal rulers who were nevertheless divine kings in their own right. By passing into a state of spiritual possession, the Ajele whose personal status was the lowly one of a palace slave, would participate in divinity and be in a position to exercise authority equal to that of the vassal king.” Owu, 17.

44. Morton-Williams, , “Outline,” 225.Google Scholar

45. Fear of Sango did not always deter political revolt. The successful revolt of the Egba against Oyo, led by Lisabi sometime between 1775 and 1780, began with the massacre of the ilari throughout Egbaland. See Biobaku, , Egba, 8–9.Google Scholar

46. Law, Oyo Empire, 99, 112.Google Scholar

47. Ogunba, O., “Ceremonies” in Biobaku, , ed., Sources, 97.Google Scholar

48. Law, Oyo Empire, 241.Google Scholar

49. R. Horton, “Ancient Ife,”

50. Bascom, , “Sociological Role,” 37.Google Scholar

51. Abimbola, W., Sixteen Great Poems of Ifa (Paris, 1975) 2–4.Google Scholar

52. Dennett, R.E., Nigerian Studies (London, 1910) 88–90.Google Scholar

53. See McClelland, E.M., “The Significance of Number in Odu of Ifa,” Africa 36(1966), 421–31.Google Scholar

54. See Abimbola, W., “The Ifa Divination System,” Nigeria Magazine, no. 122/123(1977), 35–76Google Scholar; Bascom, William, Ifa Divination: Communication Between Gods and Men (Bloomington, 1969), 130.Google Scholar

55. Beier, . Year of Sacred Festivals, 56.Google Scholar

56. Bascom, , Ifa Divination, 141.Google Scholar

57. According to Abimbola Ifa, 15–16, iyere chanting preserves the “true” Ifa text against “spurious” revisions: “The chanting of iyere is a well-developed art among Ifa priests and it is usually done in choral form, led by someone who is a good chanter. To every complete sentence chanted correctly by the leader of the chant, the other Ifa priests chant han-in, meaning “Yes, that is right.” However, if the leader has chanted a sentence wrongly, the other priests inform him of this and tell him to correct his mistake. If he makes another mistake, he might be shouted down and he who is sure of himself immediately takes over from him. Where a priest makes serious mistakes while chanting and refuses to stop chanting in defiance of the wishes of the congregation, he might even be thrown out of the meeting in shame. By this rigid insistence on the correct recital of Ifa texts, Ifa priests have made it almost impossible for spurious passages to appear in the Ifa literary corpus.” Such a statement reflects more the ideological importance of textual authenticity than its actual achievement, a point developed in my conclusion.

58. See e.g. Beier, , Year 56–62.Google Scholar

59. Johnson, , History 48.Google Scholar

60. McClelland, “Significance,” 424n.

61. Abimbola, , Sixteen Great Poems of Ifa, 6.Google Scholar

62. See Beier, Year; Johnson, , History, 156Google Scholar, Law, Oyo Empire, 36, 88Google Scholaret passim.

63. Beier, , Year, 7, 12.Google Scholar

64. The second priest, the Olunwi is also referred to as Jagun or “warrior” in Beier, , “Obatala Festival,” 18–28.Google Scholar

65. Beier, , A Year of Sacred Festivals, 14.Google Scholar

66. In Notes sur le culte des Orisa et Vodun (Dakar, 1957), 459Google Scholar, Pierre Verger mentioned that Osalufon is the name given to Obatala in Ifon, but Beier may have been unaware of the myth's provenance. Orisa Onifon is worshipped in Ilesha (Peel, , “Kings, Titles and Quarters, II” 238Google Scholar Ifon traditions may have been incorporated and perpetuated.

67. From Beier, , Year, 14Google Scholar; cf. Verger, , Notes, 443Google Scholar and Adedeji, J.A., “The Place of Drama in Yoruba Religious Obserance,” Odu, 3(1966), 88–94.Google Scholar

68. Beier, , Year, 14.Google Scholar

69. The struggle between Oduduwa and Obatala more likely refers to a dispute over the kingship in Ife between two rival factions. See Adedeji, , “Place of Drama,” 90Google Scholar; Verger, , Orisha: les dieux Yorouba en Afrique et au Nouveau Monde (Paris, 1982) 252–54Google Scholar, in which “Obatala” may have solicited Igbo support.

70. Beier, , Year, 15.Google Scholar

71. Verger, , Notes, 459.Google Scholar

72. Associations between Esu and Sango on the laba panel are documented by Wescott, /Morton-Williams, , “Symbolism”, 23–37.Google Scholar

73. Notice how Sango remains blameless since the mistake is made by his servants.

74. See Beier, Year, 14Google Scholar; Verger, , Notes 440, 473–475Google Scholar; and Verger, , Orisha, 255–65.Google Scholar Law, Oyo Empire:, 88 documents that Ejigbo was once an Ife town that was later incorporated into Oyo.

75. See Verger, , Orisha, 252Google Scholar, and Johnson, , History, 27.Google Scholar In The Entrepeneur as Culture Hero, (New York, 1980), 123Google Scholar, B.I. Belasco approaches this interpretation when he argues that “Obatala's revolutionary fragmentation is to be understood in the context of Oyo's emerging imperial thrusts to the coast,” but he (109) accepts uncritically Beier's thesis that Obatala represents an “ancestral autochthonous culture.”

76. Not all ritualized mock-battles are, by implication, records of conquest. Peel, , “Kings, Titles and Quarters, II, 236–38Google Scholar, shows how ritualized conflict in Ilesha's Iwude Ogun between the king and his town chiefs represents an actual conflict which occurred between Owaluse and his townspeople, which became the paradigm for the “stereotypical reproduction” of similar conflicts, and of the contested relationship between king and people in general. Iwude Ogun thus represents the history of such conflict and the politics of its ritual mediation.

77. Law, Oyo Empire, 59.Google Scholar Obayemi, on the other hand, who accepts Beier's aboriginal theory, estimates that the Oduduwa “revolution” occured about eleventh/twelfth centuries; sees his “Ancient Ile-Ife,” 171.

78. Law, “Anthropological Models in Yoruba History,” Africa, 43(1973), 24.Google Scholar Law does not actually criticize anthropological models in general, but Lloyd's model of Yoruba political development and his teleological 'test' in Political Development.

79. Smith, Robert, “Event and Portent: The Fall of Old Oyo, A Problem in Historical Explanation,” Africa, 41(1971), 186.CrossRefGoogle Scholar

80. Here I follow Law's argument that “in an operational sense …the writings of the local historians have to be treated as 'primary' sources, since it is no longer possible to examine the sources on which they are based”: Law, “Early Yoruba Historiography,”HA, 3(1976), 76.Google Scholar

81. Law, “Traditional History,” 31.Google Scholar

82. Horton, , “Ancient Ife,” 118–28.Google Scholar

83. The quote, from ibid., 80, is based on Law's Review of Akinjogbin, I. A., Dahomey and Its Neighbours, JHSN 6/2(1968), 119–23.Google Scholar

84. Horton, , “Ancient Ife,” 92.Google Scholar

85. Ibid., 119. Note how “post-fifteenth century Ife” really concerns “mid-seventeenth century Ife” through to the fall of Oyo in ca. 1830.

86. Ibid., 121.

87. Ibid., 124. Horton argues that this interpretation accounts for two otherwise inexplicable features of Yoruba religion: the homogeneity of the odu Ifa throughout Yorubaland, and the coexistence in Yoruba kingdoms of both pan-Yoruba (Ife-centric)and purely local orisa rites and traditions.

Given more space for protracted polemic, I woijlfl show in greater detail how my “ritual systems” theory accounts for the first as an ideological claim (which is all that it amounts to anyway) and the second as built into the very idea of an Ife-centric ritual field.

88. Horton, , “Ancient Ife,” 128Google Scholar