1. de Heusch, Luc, Le roi ivre, ou l'origine de l'Etat (Paris, 1972)Google Scholar; idem, “What Shall We Do With the Drunken King?” Africa 45(1975), pp. 363-72.

2. I use the term ‘Lunda’ for the broad Lunda world spanning the savanna, including the Ruund, the Ndembu or “Southern Lunda,” the Yaka, the Lua of the Kwango, the Eastern Lunda of the Luapula, and occasionally other more peripheral groups.

I use the term ‘Ruund’ for the specific people at the center of the Lunda state among whom I did my own field work, those sometimes called “Northern Lunda” or “Lunda of Mwant Yav.” They differ in language and culture from the other associated Lunda groups with whom they shared their political system. Today's Ruund largely include what were formerly distinct groups, at least politically: Malas, Matab, etc. The Ruund are centered in Zaire in Shaba's Kapanga and Sandoa zones, but they extend across northeast Angola into Bandundu province, Zaire.

3. de Heusch, , Le voi ivre, p. 15.Google Scholar All translations are my own.

4. Names are written as pronounced by the Ruund people themselves except in citations. Customary uRuund spelling is used except that unvoiced final /u/ is written w to avoid confusion for the reader. The letters t, d, s, and z have palatalized (“whistled”) pronunciations before /u/; ng always represents /n/ as in English “sing”. Other consonants are similar to their English values, and the five vowels correspond to the International African Alphabet. Vowel length is not represented here; the ethnic name contains a true double vowel, Ru(w)und. For further information, see Vincke, Jacques L., “Aspects de la phonologie et de la morphologie de la langue lunda (Ruund)” (unpublished doctoral thesis, Université officielle du Congo, Elisabethville, 1966)Google Scholar; Hoover, J. Jeffrey, “An uRuund-English Dictionary” (unpublished manuscript, 1975–1977Google Scholar; available from author).

5. nswan mulapw and kanampumb in Ruund.

6. This is my own summary of the generally accepted elements. Two of the most popular and accessible versions are those of Carvalho and Duysters. de Carvalho, Henrique Dias, Ethnographia e historia tradicional dos povos da Lunda (Lisbon, 1890), pp. 60–112CrossRefGoogle Scholar; also abbreviated and translated in Turner, Victor W., “A Lunda Love Story and its Consequences,” Rhodes-Livingstone Journal, 19(1955), pp. 1–26Google Scholar; Duysters, Léon, “Histoire des Aluunda,” Problèmes d'Afrique Centrale, 12(1958), pp. 75–98.Google Scholar By far the most detailed version is Di-Diaka, Kamweny, “Le symbolisme des rites d'investiture chez les Aruund” (mimeographed mémoire, Institut Supérieur Pédagogique, Lubumbashi, 1974).Google Scholar

7. de Heusch, , Le roi ivre, pp. 15–36.Google Scholar

8. Ibid, pp. 37-111.

9. Ibid, pp. 112-77.

10. Ibid, p. 276.

11. Ibid, p. 190.

12. Ibid, p. 193.

13. Ibid, pp. 188-98.

14. Red, black, and white are the only color adjectives among the Ruund as well as among the Ndembu from whom Turner drew the data re-analyzed by de Heusch. This is a widespread feature among Bantu languages. Cf. Turner, Victor W., “Color Classification in Ndembu Ritual: A Problem in Primitive Classification” in his The Forest of Symbols (Ithaca, 1967), pp. 49–92.Google Scholar

15. de Heusch, , Le roi ivre, pp. 206–16.Google Scholar

16. Ibid, pp. 221-29.

17. Carvalho, , Ethnographia, p. 223.Google Scholar

18. Bastin, Marie-Louise, “Tshibinda Ilunga: héros civilisateur. A propos de statuettes tshokwe représentant un chef-chasseur” (Brussels: mimeographed, 1966).Google Scholar

19. Biebuyck, Daniel, “Fondements de l'organisation politique des Lunda du Mwaantayaav en territoire de Kapanga,” Zaïre, 11(1957), pp. 787–817.Google Scholar

20. Frans Roelandts, MS, reprinted in Bastin, “Tshibinda Ilunga,” appendix, xvi. My informants said rather that nkond was an nkumbw, “praise-name” for the crocodile. My copy of the Bastin essay was destroyed and thus unavailable for comparison, but another Roelandts manuscript (for which I thank William Pruitt Jr.) does not support all of de Heusch's references.

20. Crine-Mavar, Fernand Bruno, “Aspects politico-sociaux du système des terres des Luunda Septentrionaux,” in Biebuyck, Daniel, ed., African Agrarian Systems (London, 1966), p. 167.Google Scholar The reference is incorrectly given by de Heusch.

22. Baumann, Herman, Lunda, Bei Bauern und Jägern in Inner-Angola (Berlin, 1935), p. 159.Google Scholar

23. Bastin, , Art décoratif tshokwé (Lisbon, 1961), 1:p. 102.Google Scholar

24. Crine-Mavar, , “Un aspect du symbolisme luunda: l'association funéraire des Acudyaang,” Miscellanea Ethnographica, [Annales du Musée royal de l'Afrique centrale, sciences humaines, 46], (Tervuren, 1963), p. 101.Google Scholar

25. de Heusch, , Le roi ivre, p. 226.Google Scholar

26. Ibid, p. 227.

27. Ibid, pp. 230-99.

28. Ibid, p. 258.

29. Ibid, pp. 274, 296.

30. Ibid, p. 298.

31. Ibid, p. 276; Vansina, Jan, Kingdoms of the Savanna (Madison, 1966).Google Scholar

32. For one methodological critique of Lévi-Strauss and his methods, along with those of others proclaiming a method for myth analysis, see Nathhorst, Bertel, Formal or Structural Studies of Traditional Tales, [Stockholm Studies in Comparative Religion, 9], (Stockholm, 1970).Google Scholar

33. de Heusch, , Le roi ivre, p. 235.Google Scholar

34. Ibid, p. 13.

35. Duysters, , “Histoire des Aluunda,” pp. 81–86Google Scholar; Pogge, Paul, Im Reiche des Muata Yamvo (Berlin, 1880), pp. 224–26Google Scholar; Carvalho, , Ethnographia, pp. 58–76Google Scholar; van den Byvang, M., “Notice historique sur les Balunda,” Congo, 1(1937), pp. 429–35Google Scholar; Struyf, Yvo, “Kahemba: Envahisseurs Badjok et conquérants Balunda,” Zaïre, 2(1948), pp. 373–75Google Scholar; Labrecque, Edouard, “Histoire des Mwata-Kazembe,” Lovania, 16(1949), pp. 9–33Google Scholar; 17(1950), pp. 21-48.

36. de Heusch, , Le roi ivre, pp. 218, 297.Google Scholar

37. Torday, Emil, On the Trail of the Bushongo (London, 1925), pp. 240–41.Google Scholar

38. de Heusch, , Le roi ivre, p. 297.Google Scholar

39. Elsewhere he does imply a Kuba to Lunda transmission. Ibid, p. 218.

40. Vansina suggests that a possible source of the episode may be the biblical story of Noah put into circulation in the Kwango from a mission at Okango from before 1610 to 1640. However, we lack sufficient documentation from the Kwango area itself to know the true spread of the episode. Personal communication, 25 September 1976.

41. de Heusch, , Le roi ivre, pp. 191, 218, 227.Google Scholar

42. See Chapter 2, “The Linguist and the Historian,” and Appendix 1 in Hoover, J. Jeffrey, “The Seduction of Ruwej: Reconstructing Ruund History” (Ph.D. thesis in progress, Yale University).Google Scholar I have not worked with the Yaka language, but it has been classed in Bantu Zone H (with kiKongo and kiMbundu) while the others mentioned are all in the revised Zone K, but quite diverse. See van den Eynde, K., Eléments ale grammaire Yaka (Kinshasa, 1968), pp. 5–6.Google Scholar

43. The evidence against Ruund circumcision rites, although never collected together previously, seems quite convincing. Among the Ruund themselves in the twentieth century circumcision took place as soon as boys were old enough to care for the wounds themselves without burdening their mothers. Informants pointed out this as a contrast with neighboring peoples when such ethnic minorities performed mukanda dances in the Musumb public square during folklore festivals sponsored by the muant yav. Old men said that even when they were boys mukand did not amount to much for the Ruund. Hospital-trained nurses took over the role of practitioner in Kapanga zone, expecially when numerous rural clinics were in operation. In the early colonial period, Montenez, while describing the intermingled ethnic groups of the Sandoa-Dilolo area, distinguished the Ruund by their circumcising at younger ages, as individuals rather than in groups, and without feasting; his article on the practice, while purporting to cover both Cokwe and Ruund, uses Cokwe terms exclusively. The lack of circumcision/initiation camps has also been noted for the Ruund-speaking in Kaninchin of Kasai, just outside the political sway of the mwant yavs. Montenez, P., “Notes sur l'identité coutumière des indigènes d'origine lunda,” Bulletin des jurisdictions indigènes et du droit coutumier congolais, 4(1936), p. 275Google Scholar; idem, “Notes sur la circoncision chez las indigènes du territoire de Sandoa,” ibid, 6(1938), p. 242; Dizez, Musas Samal Mwin-Katim, “Histoire des Kanintshin: Quelques perspectives sur l'histoire ancienne des Etats Lunda” (mimeographed mémoire, Université Nationale du Zaïre, Lubumbashi, 1974), pp. 30–31.Google Scholar

According to a personal communication from William Pruitt, Jr., the Salampasu immediately north of the Ruund adopted the mukanda complex only in the west, where more heavily influenced by the Lwalwa; the practice was abandoned early in the colonial period with no pressure from the government or missions. Pruitt believes it was introduced in the late nineteenth century when the Cokwe were the dominant economic and military agents in the area.

Other evidence comes from Ruund emigrant groups in the far west and far east. Struyf writes that the Kahemba Lunda did not circumcise (with the mukanda initiation rite?) before the Cokwe invasions. Cunnison notes that the Eastern Lunda practice lapsed at the beginning of the colonial period, attributing this to missionary opposition. At a minimum it must mean the Lunda nobility did not consider circumcision camps an ethnic hallmark or an important cultural institution; perhaps the few missionaries had little to do with the Eastern Lunda dropping the custom. Struyf, , “Kahemba,” p. 386Google Scholar; Cunnison, Ian, The Luapula Peoples of Northern Rhodesia (Manchester, 1959), pp. 166–67.Google Scholar

Finally, linguistic clues indicate that circumcision, at least as a conspicuous cultural institution, was not always practiced in the Ruund homeland. The Ruund term kandandj, “newly circumcized youth,” cannot be cognate with the similar Cokwe term, itself a borrowing. The tonal patterns for the widespread term mukanda in savanna languages suggest several west-to-east chains of transmission. The widespread term kayanda dealing with aspects of either mukanda or mungonge initiation may be akin to the Ruund term aMayand for various small groups to their northwest practicing these cults. The Salampasu term mubindji “circumcised person” is related to widespread terms from the Lulua-Kasai area for “savage” or “slave.” See Hoover, “An uRuund-English Dictionary,” entries for each term; idem, “The Seduction of Ruwej,” Chapter 5 and Appendices 3 and 4.

The essential point is that the mukanda circumcision/initiation rite in no way reflects a “Lunda” world view but rather one spreading from central Angola. The physical procedure may have been customary for a considerable period among the Ruund if not the Salampasu, but it had minor political or cultural importance.

44. First, for the Malas, mungony is the name of a dance and not of the initiation ceremony. Crine-Mavar, “Un aspect du symbolisme luunda.” The term is possibly cognate with mungonge or mungongi in other languages, however.

Second, while the Malas are today considered Ruund and speak a Ruund dialect, they were apparently classified differently before the colonial period. Carvalho groups them with the atuNgomb (Cokwe or oviMbundu traders associated with cattle), Minungu, Lwena, and Kosa as “aioko” which he translates as “those who left.” This etymology for the ethnic term Cokwe is probably spurious, but it was not a brotherly act for his informants to lump the Malas with the others at a time when the Cokwe were overrunning the Ruund heartland. Carvalho, , Ethnographia, pp. 90, 101.Google Scholar Duysters lists the Malas as distant neighbors of the proto-Ruund of Iyal a Mwakw. Duysters, , “Histoire des Aluunda,” p. 81.Google Scholar The Ruund court has pursued a “Greater Lunda” policy since the colonial period in which such minor distinctions disappear and even Cokwe and Salampasu are embraced as “true Ruund.” See Bustin, Edouard, Lunda Under Belgian Rule (Cambridge, 1975), pp. 167–68, 174, 193–95CrossRefGoogle Scholar; Hoover field notes, especially conversation with Rev. Nawej Iyav (André), Nansaka Makondi, Chiyaz Khoj during dances in Musumb public square, 28 Nov. 1976.

Carvalho mentions “mufua mungongue” as a charm to protect against madness. Mufw is “dead person” in Ruund, but Carvalho does not mention any association with funerals, nor is he clear whether it is mungonge which is believed to cause madness. Carvalho frequently includes non-Ruund data without distinguishing it; in any case, his transcription indicates the term was borrowed rather than a Ruund cognate of the Cokwe, Pende, etc. words. Carvalho, , Ethnographia, p. 242.Google Scholar

My own inquiries into mungony and the atudyang, based on Crine-Mavar's essay, always elicited responses that “that is a custom of the Malas Un northern Sandoa zone]; we don't know much about that here in Kapanga.”

45. de Heusch, , Le roi ivre, pp. 221–29.Google Scholar

46. See each entry in Hoover, “An uRuund-English Dictionary.” Iyaz a ngomb was described first as a rather rare green water snake of about 1.5 meters, and secondly as a mythic serpent reputed to dig rivers and eat people. The actual snake is associated with the mwant yav because of the black bead-like markings on its head, reminiscent of a chief's chibangul “crown,” and through its use in certain charms for the mwant yav. It is an nkumbw “praise-name,” of the mwant yav. It appears in the nkumbw of the mwant chinan published by Lucas as Chiyaz a Ngomb; he translates it first as the mythical serpent', then as mwant yav.

Lucas, Stephen, “Baluba et Aruund” (thèse du 3e cycle, Ecole Pratique des Hautes Etudes, Paris, 1968), 2:p. 88.Google Scholar

Chinowej is the name of the harmless blind or worm snake (genus Typhlops) which appears to have two heads and is considered good, unlike other snakes. It is also used as a term of great respect for a well-mannered person and, most commonly, as a personal form of address for God. Crine-Mavar, , “L'historié traditionnelle du Shaba,” Cultures au Zaïre et en Afrique, 1(1973), pp. 90–92.Google Scholar

47. Hoover, field notes.

48. Carvalho communicates the capital's definite order, but his map has problems. The present Musumb faces west rather than east. It corresponds with Carvalho's map only if everything keeps absolute position north and south of the axis except the “legs” which keep theirs relative to the king. Carvalho places the rukonkish, the king's “mother,” on his right or “father's” side, opposite her consort and subjects. Perhaps he drew the map from memory and from notes placing groups partly by relative and partly by absolute positions. Carvalho, Ethnographia, map facing 226. I was unable to verify any animal imagery for the city plan. However, the word mes can variously mean ‘eyes,’ ‘face,’ ‘front,’ and ‘vanguard.’ Carvalho may have sketched his map before he was familiar with it, relying on the descriptions of informants and interpreters. By intent or mistranslation, the zoomorphic image may have occurred to someone during what would have been a difficult explanation.

49. Hoover, field notes.

50. de Heusch, , Le roi ivre, p. 201.Google Scholar

51. Carvalho, , Ethnographia, pp. 66, 67.Google Scholar

52. Duysters, , “Histoire des Aluunda,” pp. 82–83Google Scholar; Ngand Yetu (Cleveland, Transvaal, 1963), pp. 11–12Google Scholar; Mwazaz, J. Kasong Mastak, “Risang'u rakin ra disambishin da ngand yet'u ya ruund ni Want Yav wet'u” (unpublished MS, Kapanga, 1948, 1968), pp. 8, 14–15Google Scholar; Gubbels, Jeroom, “Nsang ja Aruund” (mimeographed, Mission Catholique, Kapanga, 1963), pp. 6–7Google Scholar; “Empire Lunda” (mimeographed, Groupement des Associations Mutuelles de l'Empire Lunda, Musumb, 1960), pp. 4–5Google Scholar; testimony, mwin ohipet Munung (Daniel), Mwajinga, Nov. 1973; Kabamb Kateng, “Akaruund” (unpublished MS, Musumb, n.d.).

53. de Huesch, , Le roi ivre, p. 276.Google Scholar

54. Ibid, p. 199.

55. Ibid.

56. Duysters, , “Histoire des Aluunda,” p. 81.Google Scholar

57. de Heusch, , Le roi ivre, p. 179.Google Scholar

58. Duysters, , “Histoire des Aluunda,” p. 81.Google Scholar

59. nswan mulapw Ditend, uyang rituals, Musumb, 8 Aug. 1974, 20 July 1975.

60. de Heusch, , Le roi ivre, p. 179.Google Scholar

61. Duysters, , “Histoire des Aluunda,” p. 83Google Scholar; Van den Byvang, , “Notice historique,” p. 431Google Scholar; Ngand Yetu, pp. 12-13; Gubbels, , “Nsang ja Aruund,” p. 5Google Scholar; “Empire Lunda,” p. 5; Kabamb, “Akaruund.”

62. Duysters, , “Histoire des Aluunda,” p. 61Google Scholar; Ngand Yetu, p. 13; Gubbels, , “Nsang ja Aruund,” p. 7Google Scholar; Kabamb, “Akaruund,” Hoover, field notes, nkunibw ‘praise-name’ of the mwad mwish.

63. de Heusch, , Le roi ivre, p. 187Google Scholar; Carvalho, , Ethnographia, p. 73.Google Scholar The other such Ruund testimony is given in Lucas, , “Baluba et Aruund,” 1:p. 205.Google Scholar In a 1966 interview the nswan murund increased her own importance vis-a-vis other senior titles at the court by denying her sterility, denying that Chibind Yirung ever held the rukan “bracelet of office,” and ignoring the role of the first rukonkish. Lucas realized that the account was unusual. I have no information to support this testimony, nor direct refutation from the present nswan murund. However, the nswan murund with whom Lucas spoke is well remembered for having misled Crine-Mavar on one occasion and was revoked by mwant yav Mushid Muteb before his 1968 investiture for her “troublesomeness” and “senility.”

In Struyf's account on pp. 370-71, the sterility is displaced onto an aunt who is contrasted to the mother of Ruwej; the names given are Lukonkesha and Kamonga. In Ruund tradition Kamong a Ruwaz was Chibind Yirung's second wife under Ruwej and the first to be given the title of rukonkish. Since the genealogy Struyf gives here contradicts one given on p. 358, as well as all other testimonies, it can be considered corrupt. Struyf, “Kahemba.”

Roland reports a Sanga tradition that Ruwej had children by earlier consorts. Only Duysters states flatly that she was sterile; Mastak says she was childless with Chibind Yirung. Other accounts merely imply sterility. Custom at the court throws an interesting light on the Roland statement. The rukonkish must remain sterile; as the “mother of the mwant yav”, it is said that any son would become mwant yav automatically. The same restriction does not apply to the nswan murund, the heir of Ruwej; as “mother of the Lunda” her children do not constitute the same threat to normal succession. Roland, Hadelin, “Résumé de l'histoire ancienne du Katanga,” Problèmes sociaux congolais, 61(1963), p. 23Google Scholar; Duysters, , “Histoire des Aluunda,” p. 84Google Scholar; Mastak, , “Risang'u rakin,” p. 6Google Scholar; Hoover, field notes.

Thus the traditions are more complex than de Heusch realizes. The “normative” version which attributes no issue to Ruwej may not have intended to imply definite sterility.

64. Lucas, , “Baluba et Aruund,” 1:pp. 179–85.Google Scholar

65. The nswan murund (the successor to Ruwej) considers the kabungs to be her mwanamakwams, “children of my mother.” The term is frequently used with the general sense of “kinsman,” but here it is used more specifically as “sibling” since the nswan murund is also considered a kabung and since the others are the perpetual “uncles” of the mwant yav.

66. Mastak, , “Rusang'u rakin,” pp. 15–16Google Scholar; Ngand Yetu, p. 25.

67. de Heusch, , Le roi ivre, p. 206.Google Scholar De Heusch does not give the source of his translated reference with the form “hunts-manship and chieftainship.” The Ruund form ends with uyaj ni ukaleng, “glory and mukaleng-chieftainship” and is much closer to versions known from other Lunda areas. These variants are discussed in a note in Hoover, “The Seduction of Ruwej,” Ch. 5.

68. Turner, , “Color Classification in Ndembu Ritual,” pp. 59–92.Google Scholar

69. Reefe, Thomas Q., “Traditions of Genesis and the Luba Diaspora,” History in Africa, 4(1977), pp. 183–206.CrossRefGoogle Scholar

70. Lucas, “Baluba et Aruund.”

71. de Heusch, , Le roi ivre, pp. 205–06.Google Scholar