1. Strandes, Justus, The Portuguese Period in East Africa (2d ed.: Nairobi, 1962), 72.Google Scholar

2. Saad, Elias, “Kilwa Dynastic Historiography: A Critical Study,” HA, 6 (1979), 180–81.Google Scholar On such stylization of genealogies, see Bohannan, L., “A Genealogical Charter,” Africa, 23 (1952), 302–15.Google Scholar Saad's analysis of the Mahdali genealogy, in contrast to his discussion of the Shirazi, is on firmer ground. It must be noted, in passing, that much of his thesis leans on his translation of the Arabic ʿamr and ʿimara to denote the possession of a formal office of ʿamir. The text, however, does allow for the far more ambiguous translation of these terms followed by Free-man-Grenville. Saad's overall treatment of the text, while an interesting alternative, does not merit an especially great amount of confidence.

3. Wilkinson, J.C., “Oman and East Africa: New Light on Early Kilwan History from the Omani Sources,” IJAHS, 14 (1981), 272–306, esp. 286–300.Google Scholar

4. Spear, Thomas, “The Shirazi in Swahili Traditions, Culture, and History,” HA, 11 (1984)Google Scholar,

5. deV. Allen, J., “Swahili Culture Reconsidered: Some Historical Implications of the Material Culture of the Northern Kenya Coast in the Eighteenth and Nineteenth Centuries,” Azania, 9 (197 ), 105–38CrossRefGoogle Scholar; idem., “Swahili Culture and the Nature of East Coast Settlement,” IJAHS, 14 (1981), 306–35.

6. Horton, Mark, Shanga 1980: An Interim Report (Cambridge, 1980)Google Scholar, and personal communications; Phillipson, D.W., “Some Iron Age Sites on the Lower Tana River,” Azania, XIV (1979), 155–60CrossRefGoogle Scholar; Sinclair, P., “Chibuene - An Early Trading Site in Southern Mozambique,” in Allen, J. & Wilson, T.H., From Zinj to Zanzibar (Wiesbaden, 1982), 149–64Google Scholar; Hinnebusch, T., “The Shungwaya Hypothesis: A Linguistic Reappraisal” in Gallagher, J., ed., East African Cultural History (Syracuse, 1976)Google Scholar; Nurse, D., “History from Linguistics: The Case of the Tana River,” HA, 10 (1983), 207–38Google Scholar; idem., “A Hypothesis Concerning the Origin of Swahili” (mimeograph, 1982).

7. Martin, Bradford G., “Arab Migrations to East Africa in Medieval Times,” IJAHS, 7 (1974), 367–90Google Scholar; Freeman-Grenville, G.S.P., “The Coast, 1498–1840” in Oliver, R. & Mathew, G., eds., History of East Africa, I (Oxford, 1963)Google Scholar; and Chapter 3 of R.L. Pouwels, Horn and Crescent, Cultural Change and Traditional Islam on the East African Coast, 800–1900 (forthcoming).

8. Chapters 2 and 3 in Pouwels, Horn and Crescent. See also van den Bergh, L.W.C., Le Hadramout et les colonies arabes dans l'Archipel Indien (Batavia, 1886)Google Scholar, for corresponding effects on Malaysia and Indonesia. It is questionable whether or not the effects of these earlier phases would show up as strongly in linguistic evidence as it does, apparently, for the nineteenth century (Derek Nurse, personal communication) since an appreciable amount of older Swahili vocabulary, especially in the non-‘standard’ older dialects, has dropped out of use, according to A.S. Nabahani (interview, Malindi, July 1975).

9. For discussions of these critiques and examples of the ‘new’ oral historiography, see Pender-Cudlip, Patrick, “Oral Tradition and Anthropological Analysis: Some Contemporary Myths,” Azania, (1972), 3–24.Google ScholarMacGaffey, W., “African History, Anthropology and the Rationality of the Native,” HA, 5 (1978), 101–20Google Scholar; Miller, J., “Listening for the African Past,” in Miller, J., ed., The African Past Speaks (Folkestone, 1980), 1–59Google Scholar; Spear, T.T., “Oral Traditions: Whose History?” HA, 8 (1981), 165–81.Google Scholar

10. Miller, , “Listening,” 5–21Google Scholar; idem., “The Dynamics of Oral Tradition in Africa” in Bernardo Bernard!, ed., Fonti Orali (Milan, 1978).

11. Wilkinson, , “Oman,” 288.Google Scholar

12. Tolmacheva, Marina, “They Came from Damascus in Syria: A Note on Traditional Lamu Historiography,” IJAHS, 12 (1979), 259–69.Google Scholar

13. Wehr, Hans, A Dictionary of Modern Written Arabic (Ithaca, 1971), 410Google Scholar; Sacleux, Ch., Dictionnaire Swahili-Francais (Paris, 1939), 824.Google Scholar

14. The Kilindini, for example, claim to be from Shungwaya, as do the Bajun; see Guillain, Charles, Documents sur l'histoire, le géographie et le commerce de l'Afrique Orientale, (Paris, 1856), 3:240Google Scholar; Talbot-Smith, L., “Historical Record of Tanaland,” Kenya National Archives (KNA), DC/LAM/3/1, p. 44Google Scholar; Werner, Alice, “A Wwahili History of Pate,” Journal of the African Society, 14 (1915), 157n9Google Scholar; Nurse, Derek, “Bajun Historical Linguistics,” Kenya Past and Present, 12 (1980), 39Google Scholar; oral traditions on the Famau at Siyu collected by Howard Brown, personal communication. Probably the most revealing account, however, was reported by Elliott. According to a tradition he collected at Ngumi, the Bajun originally were “light coloured” non-Muslims who possessed no slaves and who lived on amicable terms with other mainland Africans. However, they were converted to Islam by “Arabs,” with whom hey “amalgamated,” but only “on the understanding that all their historical records and traces of a spearate nationality were destroyed.” Elliott, J.A.G., “A Visit to the Bajun Islands,” Journal of the African Society, 25 (1925/1926), 255.Google Scholar

15. Many Bajun in the Lamu region simply take names like Khazraji, Maawi (Umayyad), or Makhzumy. At Mombasa, some of the Taifa Ithnaashara have taken names like al-Kindi (Kilindini), al-Jaufi, Umeya, etc. For a few traditions in which various people claim noble Hijazi origins or claim to be Ansar, see “Hadlthi ya Wachangamwe” in Lambert, H.E., Chi-Jomvu and Ki-Ngare (Lampala, 1958), 89–90Google Scholar; “Usulu was Wajomvu,” in ibid., 70–71; and “Kurratil Ayun fx Nxsbatil Bajun” in Grottaneilli, V.L., Pescatori dell'Oceano Indiano (Rome, 1957), 365–67.Google Scholar

16. See B. Malinowski, “Myth in Primitive Psychology” in idem., Magic, Science and Other Essays (Glencoe, Ill., 1948); Bohannan, , “Genealogical Charter,” 302–15Google Scholar; Leach, Edmund, The Structural Study of Myth and. Totemism (London, 1967)Google Scholar; Beidelman, T.O., “Myth, Legend and Oral History,” Anthropos, 65 (1970), 74–97Google Scholar; de Heusch, Luc, Le roi ivre ou l'originie de l'etat, (Paris, 1972).Google Scholar

17. See the various essays in Miller, African Past Speaks, as well as idem., Kings and Kinsmen (London, 1976); Vansina, Jan, The Children of Woot (Madison, 1978)Google Scholar; and Feierman, Steven, Shambaa Kingdom (Madison, 1974)Google Scholar, among others.

18. Geertz, Clifford, The Interpretation of Cultures (New York, 1973).Google Scholar

19. Vansina, Jan, “Memory and Oral Tradition” in Miller, , African Past Speaks, 262–79.Google Scholar

20. Sahlins, Marshall D., “Culture and Environment: The Study of Cultural Ecology: in Sol Tax, ed., Horizons in Anthropology, (London, 1965), 215–30Google Scholar; idem., “The Segmentary Lineage: An Organization of Predatory Expansion,” American Anthropologist, 63 (1961), 322–45; Horton, Robin, “Stateless Societies in the History of West Africa” in Ajayi, J.F.A. and Crowder, M., eds., History of West Africa, I (New York, 1972), 78–119.Google Scholar

21. Sahlins, , “Segmentary Lineage,” 236–29Google Scholar; Horton, , “Stateless Societies,” 84–93Google Scholar; Newcomer, Peter J., “The Nuer Are Dinka: An Essay on Origins and Environmental Determinism,” Man, 7 (1972), 5–11.CrossRefGoogle Scholar

22. Southall, Aidan, “Nuer and Dinka Are People: Ecology, Ethnicity and Logical Possibility,” Man, 11 (1976), 482.CrossRefGoogle Scholar Similar observations were made by David Turton in an unpublished paper, “The Mursi and the Bodi: An Hypothesis Concerning War, Migration, and the Formation of Ethnic Groups in Southeastern Ethiopia.”

23. Southall, , “Nuer and Dinka,” 478.Google Scholar

24. Ibid., 468–78; Newcomer, “The Nuer.”

25. The summary which follows is based on wide range of sources, including works by H.N. Chittick, Christopher Ehret, Thomas Hinnebusch, Mark Horton, Derek Nurse, and Thomas Spear. For a detailed outline of this discussion, see Pouwels, Horn and Crescent, Chapters 1 and 2.

26. This term is used since it is recognized, of course, that a culture evolves, and therefore it is difficult to pin down precisely just when people now called Swahili ‘became’ recognizable as such. I would suggest that the peoples presently called Swahili exhibit a bundle of characteristics, including living on or near the East African littoral and islands; being primarily urban in origin and outlook; being Muslims; living in the characteristic rectangular house which is constructed out of coral, mangrove, and makuti; having multiple social allegiances; having Kiswahili as their first language; and being cultivators of the distinctive coastal crops, as well as fisherfolk and coastal traders. The so-called proto-Swahili, as such, would have been peoples who included at least some of these traits in their culture and who demonstrably were in the early stages of cultural and linguistic divergence from related African (principally Sabaki-speaking) groups, such as the Pokomo and the Mijikenda.

27. This name has not been settled. Some have called it Wenje ware, after the Tana river site where it was identified by Phillipson, D.W., “Some Iron Age Sites on the Lower Tana River,” Azania, 14 (1979), 155–60.CrossRefGoogle Scholar However, this is the same pottery originally found in Period la levels at Kilwa and was labelled “Kilwa Early Kitchen Ware” by its discoverer, H.N. Chittick. In deference to the site where it originally was identified and to Chittick's achievement(s), I suggest that ‘Kilwa ware’ should be an appropriate name for it.

28. Christopher Ehret, personal communication. Derek Nurse, “History from Linguistics” and personal communication, details four other languages which developed out of Sabaki: Malankote (Elwana), Comorian, Pokomo, and Mijikenda.

29. Mgomezulu, M., “Recent Archaeological Research and Radiocarbon Dates from East Africa,” JAH, 11 (1981), 447–48.Google ScholarSinclair, Paul, “Chibuene--An Early Trading Site in Southern Mozambique” in Allen, /Wilson, , From Zinj to Zanzibar, 149–64Google Scholar; H.N. Chittick, personal communication, 5 Sept. 1983.

30. See Freeman-Grenville, G.S.P., The East African Coast, Select Documents (hereafter SD), (Oxford, 1962), 20, 24Google Scholar; Lombard, M., The Golden Age of Islam, (New York, 1975), 179.Google Scholar

31. Locally-produced cloth was first associated with the rise to commercial prominence of Mogadishu and it remained an important ingredient of coastal trade well throughout the Portuguese period. Cloth still is woven by guildsmen in Mogadishu. Chittick, , Kilwa, 2:428–31Google Scholar, found, significantly, spindle whorls in Period lb levels at Kilwa. In its heyday, Pate, too, was famous for its “Pate cloth.”

32. See Datoo, B.A., Port Development in East Africa (Nairobi, 1975), 29–30Google Scholar; Chittick, , Kilwa, 1: 61–99Google Scholar; idem., “Kilwa: A Preliminary Report,” Azania, 1 (1966), 19; Horton, , Shanga 1980, 408Google Scholar and personal communication; Kirkman, James, “Historical Archaeology in Kenya, 1948–1956,” Antiquaries Journal, 37 (1957), 16–28.CrossRefGoogle Scholar

33. Horton, Robin, “On the Rationality of Conversion,” Africa, 41 (1971), 85–100, 101–08Google Scholar; Peel, J.D.Y., “Syncretism and Religious Change,” Comparative Studies in Society and History, 10 (1967/1968), 121–44.CrossRefGoogle Scholar

34. See “Memoir concerning the East Coast of Africa, 15 June, 1777 and 26 September, 1777” in Freeman-Grenville, G.S.P., The French at Kilwa Island, (Oxford, 1965), 92.Google Scholar In other parts of Morice's correspondence with J.F.C. de Cossigny, he observed the “predatory habits and ferocity” of Arab seamen which compelled both Africans and Swahili townspeople at Kilwa to distrust them. Ibid., 82.

35. Joao de Barros, quoted in Kirkman, James, Ungwana on the Tana (Hague, 1966), 10.Google Scholar The towns referred to specifically as walled were Ozi, Mombasa, Malindi, Mudio/ldio, and Kilifi. Other major towns to have been walled were Gedi, Lamu, Kilwa, and Pate. See various Portuguese accounts in SD, 66, 108, 134.

36. See Pouwels, Horn and Crescent, chapter 4. One Swahili historian actually does refer to the name of one town as meaning “refuge from one's enemies,” q.v. “Kitab az-Zanuj” in Cerulli, Enrico, Somalia, scritti vari editi ed inediti, (Rome, 1957), 1:329.Google Scholar

37. The literature on this is extensive, so only a few will be cited. See Guillain, , Documents, 2: 97Google Scholar; Stigand, C.H., The Land of Zinj (London, 1913), 124–25Google Scholar; Ingrams, W.H., Zanzibar, Its History and Its Peoples (London, 1931), 435Google Scholar; Caplan, Patricia, Choice and Constraint in a Swahili Community (London, 1978), 100–01Google Scholar; KNA, DC/LAM/3/2, p. 34.

38. An excellent description of a kuzunguo is detailed in Lienhardt, Peter, “A Controversy over Islamic Custom at Kilwa Kivinje, Tanzania” in Lewis, I.M., ed., Islam in Tropical Africa (London, 1966), esp. 377–79.Google Scholar See also Caplan, , Choice, 90Google Scholar; el-Zein, A.H., The Sacred Meadows (Evanston, 1974), 281–321.Google Scholar

39. Such, for instance, are the traditions associated with the arrival of the Abu Bakr b. Salim shurafa at Pate around the middle of the sixteenth century. In these stories, either it was the Orma or the Portuguese who were repelled. See Nasir, Sayyid Abdallah Ali, Al-Inkishafi, ed., Hichens, W. (Nairobi, 1972), 11Google Scholar; Zein, , Sacred Meadows, 39Google Scholar; Martin, B.G., “Arab Migrations to East Africa in Medieval Times,” IJAHS, 1 (1974), 381Google Scholar; Cerulli, , Somalia, 1:22–23Google Scholar; Stigand, , Zinj, 51Google Scholar; Talbot-Smith, , “Historical Record of Tanaland,” 64.Google Scholar

40. I originally made this point in an earlier article, “The Medieval Foundations of East African Islam,” IJAHS, 11 (1978), 394–96.Google Scholar

41. The following social analysis is outlined in greater detail in Pouwels, , “Islam and Islmaic Leadership in the Coastal Communities of Eastern Africa” (Ph.D., UCLA, 1979), 208–56.Google Scholar See also Caplan, , Choice, 19–58Google Scholar, for an excellent description.

42. Prins, A.H.J., Didemic Lama (Groningen, 1971)Google Scholar; Ranger, T.O., Dance and Society in Eastern Africa (Berkeley, 1975), 1920Google Scholar; Lienhardt, Peter, “The Mosque College of Lamu and Its Social Background,” Tanganyika Notes and Records, 53 (1959), 228–42Google Scholar; Pouwels, , “Islam and Islamic Leadership,” 234–44.Google Scholar

43. Prins, , Didemic Lamu, 23n15.Google Scholar

44. Fighting with food as a means of expressing status in wealth--or more accurately, the power to draw upon one's ‘social capital’ to provide wealth--is a familiar topic among students of Melanesian societies, but has not yet been explored extensively among Africanists. See, Young, Michael W., Fighting with Food: Leadership, Values and Social Control in a Massim Society, (Cambridge, 1971).Google Scholar Such activity, however, was observed in Lamu by British colonial administrators. See Skene, R., “Arab and Swahili Dances and Ceremonies,” Journal of the Royal Anthropological Institute, 47 (1917), 415–21, esp. 415–16Google Scholar, and Harries, Lyndon, Utenzi wa Mkunumbi: A Swahili Potlatch, (Nairobi, 1967).Google Scholar

45. Turner, Victor, The Ritual Process, Structure and Anti-Structure (Chicago, 1969), 83.Google Scholar

46. The sole exception is number 8. The recent (1980) date of its collection, however, probably has something to do with this. Many Swahili nowadays are beginning to think of themselves as Africans.

47. It must be remembered that present evidence suggests that the Swahili/Shirazi first appeared north of the Tana river.

48. This process was not always unidirectional. So-called Arabs sometimes reverted to being Shirazi in other circumstances. Lately, again, with independence, “Arabs” and “Shirazi” are calling themselves Africans in Kenya.

49. For information on Mogadishu see Guillain, Documents, passim; Chittick, H.N., “Medieval Mogadishu” in Allen, /Wilson, , From Zinj to Zanzibar, 48–54.Google Scholar On Siyu, interviews, Mombasa: Sayyid Athman b. Abdu'r-Rahman Saqqaf, 23 January and 3 February, 1975. On Lamu: numerous interviews obtained January-August 1975; Talbot-Smith, , “Historical Record,” 47–49Google Scholar; Clive, J., “A Short History of Lamu,” KNA DC/LAM/3/1, p. 5Google Scholar; Prins, Didemic Lamu; Lienhardt, “Mosque College,” passim. On Vumba Kuu: Hollls, A.C., “Notes on the History of Vumba, East Africa,” Journal of the Royal Anthropological Institute, 30 (1900), 282Google Scholar; McKay, William F., “A Precolonial History of the Southern Coast,” (Ph.D., Boston University, 1978), 51.Google Scholar On Zanzibar: Sacleaux, Charles, Grammaire des dialectes Swahilis, (Paris, 1909), xiv–xviGoogle Scholar; Appendix, no. 7. On Kilwa: Appendix no. 3; Morice in Freeman-Grenville, , French, 74, 79–81, 92.Google Scholar On the Comoros: Appendix, no. 6; Verin, Pierre, “L'introduction de l'Islam aux Comores selon les traditions orales,” in Allen, /Wilson, , From Zinj to Zanzibar, 193–200.Google Scholar

50. Taylor, W.E., African Aphorisms, (London, 1924), 80–83Google Scholar; Abdulaziz, M.H., Muyaka, Nineteenth Century Popular Poetry, (Nairobi, 1979), 19–21, 146–47Google Scholar; Idrisi in SD, 20.

51. Sassoon, Hamo, “Excavations at the Site of Early Mombasa,” Azania, 15 (1980), 1–42.CrossRefGoogle Scholar

52. Krapf, J.L., Dictionary of the Suaheli Language, (London, 1882), 270Google Scholar; al-Bauri, Mbwana b. Mbwanafundi, “News of Mombasa,” KNA CP/1/2, p. 14Google Scholar; Berg, F.J., “The Swahili Community of Mombasa, 1500–1900,” JAH, 9 (1968), 42–45.CrossRefGoogle Scholar

53. Sassoon, , “Excavations,” 39Google Scholar; Battuta, Ibn in SD, 31.Google Scholar The area associated with Shehe Mvita's town is corroborated by early sixteenth-century eyewitness accounts and maps: Freeman-Grenville, , SD, 52, 97Google Scholar; da Sa's map in Sassoon, , “Mbaraki Pillar and Related Ruins of Mombasa Island,” Kenya Past and Present, 14 (1982), 31.Google Scholar See also the discussion in Sassoon, , “Excavations,” 9–10.Google Scholar

54. See the “Mombasa Chronicle” in SD, 216–17; Berg, , “Swahili Community,” 46, 51Google Scholar; Gray, John M., The British at Mombasa, 1824–26 (New York, 1957), 7–9.Google Scholar

55. See Pouwels, Horn and Crescent, chapters 7 and 8.

56. Wilkinson, , “Oman,” 297–99.Google Scholar

57. The image of the cloth barrier or walkway to the mainland has several levels of significance: as separation of the island from a zone of unbelief (kufr)--like kuzunguo, already discussed; as a way of defining ‘legitimacy,’ land ownership, and a prevailing legal (e.i. Islamic) system, to be discussed in the conclusion; and, as is being argued here, as symbolizing the unique position of the Shirazi/Swahili in the trade that joined continental Africa to Indian Ocean commercial networks.

58. See the reports of de Barros, Hans Mayr, and Barbosa, in SD, 90–91, 103, 107, and 110.Google Scholar

59. Ibid., 127.

60. Freeman-Grenville, , French, 81–82.Google Scholar

61. Appendix, no. 3; Morice, in Freeman-Grenville, , French, 135–36.Google Scholar

62. Appendix, no. 3; SD, 221–23; Morice, in Freeman-Grenville, , French, 81, 135–36.Google Scholar

63. This appears tohave been quite ancient and widespread. The Mahdali sultan of the fourteenth-century Kilwa state was entitled the Abu'l-Mawahib--the ‘Father (Giver) of Gifts.’ Ibn Battuta, particularly, commented on this characteristic in SD, 32. Also, numerous interviews obtained in Mombasa and Lamu, 1975. See Horn and Crescent, Chapter 4. Prestation and redistribution--gift exchange--of course was the principal cement binding client to patron, senior to junior, ruler to subject all over Africa. See Meillassoux, Claude, “‘The Economy' in Agricultural Self-Sustaining Societies: A Preliminary Analysis” in Seddon, David, Relations of Production, (London, 1978), 127–57.Google Scholar

64. SD, 223.

65. Morice, in Freeman-Grenville, , French, 92Google Scholar; SD, 222–23.

66. As the author of the Kitab az-Zanuj says: “Persians for a certain while were powerful and authoritative in the cities and towns from Mogadishu to Kilwa.” Extensive fieldwork in the Lamu area traditions was carred out in 1974/75. Discussion of these traditions and their relationship to the material presented in this paper is to be found in Chapter 3 of Horn and Crescent.

67. Freeman-Grenville, , SD, 36–37.Google Scholar

68. Ibid., 89–90.

69. Ibid., 221–22.

70. Ibid., 297.

71. Saadi, Amur Omar, “Mafia--History and Traditions,” Tanganyika Notes and Records, 12 (1941), 25.Google Scholar

72. Dubins, Barbara, “A Political History of the Comoro Islands, 1795–1886” (Ph.D., Boston University, 1972), 19–20.Google Scholar

73. Kirkman, James, ed., “The Zanzibar Diary of John Studdy Leigh, Part I,” IJAHS, 13 (1980), 291–312.Google Scholar

74. Elizabeth Wangari Rugoiyo, oral tradition collected from Asmau Famau, Siyu, and published in Horton, Shanga, Appendix 5.

75. Stigand, , Land of Zinj, 29–30.Google Scholar

76. Cerulli, , Somalia, 1: 258–67Google Scholar; variation, ibid., 334.

77. Hollis, , “Vumba,” 282.Google Scholar

78. KNA DC/MSA/3/1, 197, or Werner, Alice, “Some Notes on the Wapokomo of the Tana River Valley,” Journal of the African Society, 12 (1912/1913), 366.Google Scholar

79. “Kurratil Ayun fi Nisbatil Bajun” in Grottanelli, , Pescatori, 366–67.Google Scholar

80. Freeman-Grenville, , SD, 83–84.Google Scholar

81. “Hadithi ya Wachangamwe” in Lambert, H.E., Chi-Jomvu and Ki-Ngare, (Kampala, 1958), 89–90.Google Scholar

82. “Usulu wa Wajomvu” in ibid., 70–71, 81.