¹ Usually “Lunda empire” in the literature. But the unity of the Lunda ensemble did not consist in more than the acceptance of the Mwaant Yaav as a ruler superior to others . Moreover, some of these polities were founded by Rund settlers. Hence by analogy “commonwealth” seems more appropriate than “empire.”

² Pogge, P.Im Reiche des Muala Jamwo (Berlin, 1880) 224–26Google Scholar; Schütt, O., Reisen im südwestlichen Beckcn des Congo (Berlin, 1881) 79Google Scholar (data from 1878/79), briefly mentions “Bangala Quinguri” as a kinsman of “Matiamvo” who helped the Portuguese against the queen of “Ginga oder N'Gola” and ousted the king of the Pende.” The rest of his account makes evident that he derived it from a Portuguese source, probably Neves, discussed below.

³ de Carvalho, H.A. Dias, Ethnographia e Historia tradicional dos povos da Lunda (Lisbon, 1890), 53–112, esp. 58–82.CrossRefGoogle Scholar

⁴ Miller, J.C. “The Imbangala and the Chronology of Early Central African History,” JAH, 13 (1972), 570–74.CrossRefGoogle Scholar His Kings and Kinsmen (Oxford 1976), 114–223Google Scholar, does not date this. Hence my own dating in Curtin, P.D.et al, African History from Earliest Times to Independence (2d. ed.: London 1995), 226.Google Scholar

⁵ Vellut, J-L., “Notes sur le Lunda et la frontière luso-africaine (1700-1900),” Etudes d'histoire africaine, 3 (1972), 65–69Google Scholar; Miller, Kings, 112n1.

⁶ de Heusch, Lue, Le roi ivre ou l'origine de l'étal (Paris 1972).Google Scholar

⁷ Hoover, Jeffrey, “The Seduction of Rweej” (PhD, Yale University, 1978), 157–175Google Scholar for the historiography of the tradition; 211-43 for the rejection of the Kinguri exodus.

⁸ Ibid., 244-87. The dating proposed for the earliest colony at Mukulweji is ca. 1670-1700.

⁹ Ibid., 238, and ca. 1600 for the rule of Rweej.

¹⁰ Thornton, John K.,”The Chronology and Causes of Lunda Expansion to the West, c.1700-1852,” Zambia Journal of History 1 (1981), 7.Google Scholar Thornton knew about, but did not have access to, Hoover's work.

¹¹ Ibid., 11n10. Cavazzi visited Cassange from 1660 and may also have received further information from Antonio da Serravezza, who lived at Cassange's court from 1655 to 1659. Cf. Saccardo, G., Congo e Angola (3 vols.: Lisbon, 1982), 1: 504–05, 523, 524.Google Scholar Hence the record of the traditions dates most likely from 1655-60.

¹² This meaning occurs in the Wambo variant of Ovimbundu. Cf. Miller, , “Imbangala,” 565 n63Google Scholar, and other dictionaries. In modern Kimbundu dictionaries Kinguri means “forefather,” “begetter,” “procreator,” “root,” etc., and nguri or nguri ana, “trunk of descent,” i.e., the person from whom a family descends. The earliest reference I could find, however, occurs in da Matta, J.D. Cordeiro, Ensaio de Diccionario Kimbundu-Porluguez (Lisbon 1893), 125Google Scholar: nguriana: “mã;e de filhos.” In Kongo (1650) ngudi/nguri meant “mother (in general)” and with -ankama, “master, lord.” In modern Kongo the term means “mother, age, authority” and “clan or family.”

¹³ de Montecúccolo, J.A. Cavazzi, Descrição histórica dos três reinos: Congo, Matamba e Angola trans. de Legguzano, G.M. (2 vols.: Lisbon 1965), 1:190 (Book II, #31).Google Scholar

¹⁴ Miller, , “Imbangala,” 568.Google Scholar

¹⁵ In essence all the chiefs of Songo, including apparently the king of Bihe himself.

¹⁶ Graça, J., “Expedição ao Muatayanvua,” Boletim da Sociedade de Geógraphia de Lisboa, 9 (1890), 391.Google Scholar

¹⁷ Neves, A.R., Memoria da Expedição a Cassante … em 1850 (Lisbon, 1854), 96–97.Google Scholar

¹⁸ de Lacerda, Paulo Martins Pinheiro, “Noticias do Paiz de Quisama,” Annnes do Concelho Ultramarino (Parle não official) (1846), 123–25.Google Scholar

¹⁹ Magyar, Làszlo, Reisen in Süd-Afrika in den Jahren 1849 bis 1857 (Pest 1859), 266–69.Google Scholar A comparable tradition was later reported by Capello, H. and Ivens, R., From Benguella to the Territory of the Yacca, trans. Elwes, A. (London, 1882), 157–58Google Scholar, but the leader here was Muzumbo-Tembo.

²⁰ Santos, M. E. Madeira, editor of Porto, A.F. Ferreira da Silva, Viagens e apontamentos de um Portuguese em Africa (2 vols.: Coimbra, 1986), 1:282.Google Scholar

²¹ Pogge, , Im Reiche, 224Google Scholar; Carvalho, , Ethnografia, 538, 577, 578, 585, 590, 595, 622–23.Google Scholar Carvalho does not explicitly acknowledge Lourenço Bezerra as a source, contrary to the impression conveyed in Thornton, , “Chronology,” 2–3Google Scholar, note 14. But Antönio Bezerra, his kinsman, was the official interpreter of the expedition (Carvalho, , Viagens, 104Google Scholar). The other main source which Carvalho acknowledges as “later than Bezerra” was the latter's kinsman, Rocha. Moreover, it is evident from all his reports that Ambaquistas in general were his main advisors and guides.

²² Vellut, , “Lunda,” 133–34.Google Scholar

²³ Pogge, , Im Reiche, 224.Google Scholar

²⁴ Carvalho, , Ethnografia, 1:77Google Scholar, mentions Njinga and “the already established fort of Massangano.” Pogge, , Im Reiche, 225Google Scholar,”Ginja.”

²⁵ “Explorações dos Portuguezes no interior d'Africa meridional,” Annaes Maritimos e Colonines, parte não official, 3a/6 (1843), 238–39.Google Scholar

²⁶ “Mueller's Bericht über seine Reise zu Muata-Kumbana” in von Wissmann, Hermann, Im Innern Afrikas (Leipzig, 1891), 101CrossRefGoogle Scholar (Kahungula's report), 108 (Pende want to meet their “cousin”).

²⁷ Vansina, JanThe Children of Woot (Madison, 1978), 37.Google Scholar

²⁸ See below; for instance, the genealogy explaining the position of Mai.

²⁹ Carvalho, , Ethnographia, 59–102.Google Scholar

³⁰ Capello, /Ivens, , Benguella, 1:157–58; 190–92.Google Scholar

³¹ Cardoso, Fonseca, Em terras do Moxico (Porto, 1919), 14–21Google Scholar; text reproduced in Bastin, M.-L.Tshibinda Ilunga, héros civilisateur (2 vols.: Brussels, 1966), 2: xxvii–xxix.Google Scholar This text from 1903-04 was recorded at a minor Cokwe court.

³² No accounts from any Minungo chief are available, nor are records after 1920 for Shinji or Luchazi. Hoover, “Seduction,” 19n19, mentions an archival reference to an otherwise unknown 1916 report that apparently was a draft of a draft (1928) of a 1938 article on Rund history.

³³ Hoover, , Seduction: 215, 219, 220. 311–15Google Scholar; Papstein, Robert, “The Upper Zambezi: A History of the Luvale People, 1000-1900” (PhD., University of California, 1978), 129–30Google Scholar (“Nama-Kungu” clan); idem., ed., The History and cultural Life of the Mbunda Speaking Peoples (Lusaka, 1994), 1 (Kola, chieftainess Naama); Neves, , Memoria, 96Google Scholar (Nhâma); de Magalhães, A.A., “Origem dos Basongo,” Mensàrio administrativo, 15 (1948), 34Google Scholar (Candembe).

³⁴ Papstein, , “History,” 30–50Google Scholar (Lovale in northern Rhodesia); Hoover, , “Seduction,” 303–05Google Scholar (Cokwe, 1960).

³⁵ Ibid., 214-26.

³⁶ I do not discuss the impact of the story on the historiography of Angola here. Suffice it to say that the demise of the Kinguri exodus requires major revisions to Miller's Kings and Kinsmen.

³⁷ Schütt, , Reisen, 136.Google Scholar Mai was the youngest son of “Mutombo Mucalla” [Mukulu] and a brother of Muene Canchica (Kanyok), Muata Cassongo (the main Luba ruler in Katanga) and “Quibinda Ilunga.” See also ibid., 145, 150. All other known references to Mai's supposed origin agree. In 1855 Livingstone noted that Mai was an independent chief. Schapera, Isaac, ed., Lvingstone's African Journal (2 vols.: London, 1963), 2:245.Google Scholar For the fullest account see Carvalho, , Ethnographia, 98–99.Google Scholar

³⁸ Leitão, M. Correia, “Viagem,” in Dias, G. Sousa “Uma viagem a Cassange nos meados do século XVIII,” Boletim da Sociedade de Geografia de Lisboa, 56 (1938), 25.Google Scholar Leitão's report (ibid., 9-30), which dates from 1756, is not well known and its contents are frequently misunderstood, despite Vellut's, J.-L. “Relations internationales du Moyen-Kwango et de l'Angola dans la deuxième moitié du XVIIIe s.,” Etudes d'Histoire Africaine, I (1970), 75–135Google Scholar, which is devoted to an analysis of this report.

³⁹ Dias, , “Viagem,” 10Google Scholar (“aluluas”), 25 (“molua”), and 28 28 (“maluê”). Molua soon became the standard designation.

⁴⁰ Almost certainly raffia cloth. If so, this is the earliest known reference to raffia cloth from beyond the Kasai river. Note that slaves are not mentioned but probably were already a major object of trade.

⁴¹ Donge could refer to several Ndonji titles and territories in Angola, e.g., to the areas later labeled Cokwe or Minungo, or even to the Jaga realm in northernmost Matamba mentioned by Cavazzi, , Congo, Matamba e Angola, 1:21, 177Google Scholar; 2:196: (books 1:16; 11:4,6, VII:31).

⁴² de Cadornega, A. de Oliveira, História geral das guerras angolanas (3 vols.: Lisbon, 1942), 3:219–20.Google Scholar The original dates to 1681. Embaca obviously refers to Ambaka, the fort in Angola, and Calunga, “sea,” as well as the term cited for “ship” are kimbundu, hence the tales refer to Angola, not to Mozambique.

⁴³ Several authors have studied the internal chronology of the Yaka kingdom in depth. In the latest effort Van Roy, H., Les Byaambvu du Moyen-Kwango (Berlin, 1988), 73, 178–79Google Scholar, concluded that Muteeba Ndziimbu was the first king who began to create a stable state and dates this process to 1710-1740/50. Considering Leitão's evidence, the closure date is only a little too early. As a result, the proposed later chronology of the Yaka kingdom and surroundings seems to be only twenty years or less too early. In contrast, not much attention has been paid to internal chronology elsewhere and the proposed datings need to be shortened, sometimes by a full century, to the mid- (or in the Mbunda case perhaps even to the late) eighteenth century.

⁴⁴ Dias, , “Viagem,” 19–21.Google Scholar The fact that the title of the Shinji lord Malundo was claimed to be imported by Rund conquerors in Carvalho's time should not be taken to mean that this was already so in 1756. The fact that Leitao distinguished Malundo and the other realms just east of the Kwango from the Rund, to the point that he gives separate ethnographic descriptions for each, suggests that the title and the dynast of 1756 was not of Rund origin.

⁴⁵ Birmingham, David, Trade and Conflict in Angola (Oxford, 1966), 150 (1762), 152–53 (1767).Google ScholarThornton, “Chronology,” 8, 13n53 (1762)Google Scholar is incorrect in talking about a flood of refugees “from the ‘Moluas’.”

⁴⁶ For these sources see Vellut “Relations” and “Lunda”. Also Vansina, Jan, Kingdoms of the Savanna (Madison, 1966), 217Google Scholar (map with limits of Rund homeland before the mid-nineteenth century).

⁴⁷ Neves, , Memória, 98.Google Scholar The Shinji names occur in Carvalho, , Ethnographia, 92–93.Google Scholar Practically all the others are found on map 2 of Schütt, Reisen.

⁴⁸ This point cannot be developed here. Still, briefly, there was a “Lovar” type between the upper Kasai and uppermost Zambezi, a “Kwango” type along the latter, a Pende type along the middle Kwilu, and a Munjumbo a Kalunga type between the uppermost Cuanza and the Cunene.