Elite women in the kingdom of kongo: Historical perspectives on women's political power

Journal of African History

Volumn 47, Issue 3, 2006, Pages 437-460

  Thornton, John K ^a  

^a BOSTON UNIVERSITY   (United States)

Author keywords

Angola; Democratic Republic of Congo; Politics; Precolonial; Women

Indexed keywords

CIVIL WAR; HISTORICAL GEOGRAPHY; POLITICAL PARTICIPATION; POWER RELATIONS; WOMENS STATUS;

AFRICA; CENTRAL AFRICA; DEMOCRATIC REPUBLIC CONGO; SUB-SAHARAN AFRICA;

EID: 33845343626     PISSN: 00218537     EISSN: 14695138     Source Type: Journal    
DOI: 10.1017/S0021853706001812     Document Type: Article

References (144)

1. Flora Edouwaye S. Kaplan (ed.), Queens, Queen Mothers, Priestesses and Power: Case Studies in African Gender (New York, 1997).
2. Powerful women have been a regular stock in trade in discussion of precolonial women, e.g. Catherine Coquery-Vidrovitch, African Women: A Modern History (Boulder, 1997 [original French 19941), 37-44
3. Iris Berger and E. Frances White, Women in Sub-Saharan Africa (Bloomington, 1999), 64-8, 77-80.
4. Sandra T. Barnes, 'Gender and the politics of support and protection in precolonial West Africa', in Kaplan (ed.), Queens, 1-18.
5. Ronald Cohen, 'Oedipus Rex and Regina: The queen mother in Africa', Africa, 5 (1977), 14-30,
6. see also Barnes's discussion of this in 'Gender and support', 2-4, 13-15.
7. For a study of Kongo women, with a focus on the non-elites (though some thoughtful material on elite women as well), see Susan Herlin Broadhead, ' Slave wives, free sisters: Bakongo women and slavery c. 1700-1850', in Claire C. Robertson and Martin Klein (eds.), Women and Slavery in Africa (Madison, 1983), 160-81.
8. John Thornton, 'Ideology and political power in Central Africa: The case of Queen Njinga (1624-1663)', Journal of African History, 32 (1991), 25-40.
9. Elaborating the details of this system requires more extensive treatment than is possible here, but even pre-1550 documentation generated by Kongo authorities allows us to see the dependency of office holding on the crown, the use of money salaries and the role of functional bureaucratic offices. The correspondence of Afonso I and Diogo I are printed in António Brásio (ed.), Monumenta Missionaria Africana (1st series, 15 vols.) (Lisbon, 1952-88) (henceforward abbreviated MMA), vols. I-II, IV and V.
10. For a description based on seventeenth-century materials, some written by visitors, others occurring incidentally in sources of Kongo authorship, see John Thornton, The Kingdom of Kongo: Civil War and Transition, 1641-1718 (Madison, 1983), 38-55
11. and idem,
12. 'The correspondence of the Kongo kings, 1614-35: Problems of internal written evidence on a Central African Kingdom', Paideuma, 33 (1987), 407-21.
13. The first comprehensive description of Kongo was produced by Duarte Lopes, Kongo's ambassador to Rome, around 1588, published and silently edited by Filippo Pigafetta, Relatione del Reame de Congo e la circonvincinta contrade (Rome, 1591). For examples of the crown and officers assigning offices and granting money salaries, see two late sixteenth-century provisions among the private papers and documents left in Rome by Antonio Manuel, Kongo's ambassador to the Vatican, upon his death in 1608: Archivio Segreto Vaticano, Arm 11, vol. 91, fols. 202, Provizão of Alvaro II to Antonio Manuel, 19 June 1599; 230, Provizão of Gonçalo da Silva Mendonça, Procurador of Congo, 19 Oct. 1593.
14. The term is first attested in the catechism of 1624, in a factional context, for catecheumens are enjoyed to obey 'our priests and those who govern the Republic' which is then rendered as 'Aculuntuêto anganga yoâna aludica edicanda' (seventeenth-century forms of this word still carried the class 5 prefix, which was lost in the eighteenth century, the initial 'e' is an elided definite article). See Mateus Cardoso (ed.), Doutrina christãa, mod. edn. by François Bontinck and D. Ndembe Nsasi, Le catechisme Kikongo de 1624: Réedtion critique (Brussels, 1978), 140-1.
15. For a more strictly familial context, see the dictionary of 1648, Biblioteca Nazionale Centrale Vittorio Emmanuele, Fundo Minori 1896, MS Varia 274, fol. 36v, 'Vocabularium Congoese, Hispanicum, Latinum...', 1648. In this Latin to Kikongo dictionary the term familia is rendered as 'dicanda', but in Latin a familia was a large social unit connected by kinship, patronage and slavery. See Charlton Lewis and Charles Short (eds.), Oxford Latin Dictionary (Oxford, 1962), 723.
16. Rodrigo Fontinha (ed.), Novo dicionário etimológico da lingua Portuguesa, rev. edn. by Joaquim Ferreira (Porto, n.d.), s.v. geração.
17. Undated letter attached to the royal inquest into the treason of Pedro Nkanga a Mvemba, 'Auto de Devassa', 12 Apr. 1550, MMA, 11, 261.
18. Discussion of the kanda is one of the bedrocks of modern Kongo ethnography, probably since its first formulations by missionary-anthropolgists like Joseph Van Wing or Leon Bittremieux. Clan histories are widely known today; it is the best starting point for discussions of oral tradition, and Jean Cuvelier's Kikongo language collection of clan histories, Nkutama a mvila za makanda (1st edn., Tumba, 1934, 4th edn. [revised by Joseph de Munck], Matadi, 1972) is regarded by many as authoritative.
19. For the role of the factions in the mid-seventeenth century, see Thornton, Kingdom of Kongo, 47-54.
20. Modern tradition since c. 1850 has focused on these named clans as the bedrock of social organization, and not surprisingly many historians have accepted this organization. For a full discussion of probable social organization in the earlier periods, and the origins of the modern clan system, see John Thornton, 'The origins and early history of the Kingdom of Kongo, c. 1350-1550', International Journal of African Historical Studies, 34 (2001), 7-12, and documentation cited there.
21. Jan Vansina, How Societies Are Born: Governance in West Central Africa before 1600 (Charlottesville VA, 2004), 88-99.
22. Recognizing this point differentiates my own work, which focuses on a state-centered locus, and restricts source material to that contemporary to the events, from that of Anne Hilton, The Kingdom of Kongo (Oxford, 1985), esp. 5-31,
23. and António Custodio Gonçalves, Kongo: Le lignage contre l'état (Évora, 1985), among others, which seek to wed modern anthropological observation (based on conditions in post-nineteenth-century society, generally at the village level) with older documentation.
24. For specifics on how one might use oral tradition collected both in earlier times and in the twentieth century, after social changes altered the nature of tradition, see Thornton, 'Origins'.
25. In this article, I use Portuguese and Italian kinship terms (translated into English) as I encounter them in the original sources. It should be noted that in Kikongo kinship terms are defined differently, generally with less specificity. E.g. the Kikongo term brother, mpangi, can also mean cousin (or sister), while father (se) can apply to uncles as well as fathers. Many writers are Kikongo speakers writing in Portuguese, for whom employing a term like cousin necessarily requires a full knowledge of the European kinship system. Some seem to have been well aware of this: For example, Afonso, in a letter of 1535 introducing his embassy, gives a very precise account of their kinship to him, 'Dom Afonso my nephew by brother ... dom Afonso my most distant nephew, and dom Amrrique my nephew by brother'. Afonso to Pope Paulo III, 21 Feb. 1535, MMA, 11, 39.
26. Coquery-Vidrovitch's observation that female power is most likely in matrilineal societies (African Women, 34) seems only marginally applicable in this case.
27. João de Barros, Decadas de Asia: Primeira Década (Lisbon, 1552), Book 3, 35.
28. Da Pina, Italian MS, 97rb-97va, printed in Carmen M. Radulet (ed.), O cronista Rui de Pina e a 'relação do reino do Congo' (Lisbon, 1992)
29. very similar passage in da Pina, 'Cronica del Rei D. Joam II, (c. 1515) cap. 63 (the only version to mention money in the gifts); both in Radulet (ed.), Cronista, 126-9, 151-3.
30. Her name is only given in the Portuguese version, De Pina, cap. 53, in ibid.
31. not in the corresponding section of the older Italian recension in ibid. 126.
32. These contemporary comments should discount the tradition that began circulating in the late seventeenth century which revealed that her son Afonso had buried her alive because she refused to part with a small idol after the institution of a strictly Christian law during his reign as king. For further discussion see John Thornton, The Kongolese Saint Anthony: Dona Beatriz Kimpa Vita and the Antonian Movement, 1684-1706 (Cambridge, 1998).
33. letter of 1531 cited below reveals that Leonor was the daughter of a king and mother of King Afonso, and she could not have been daughter of his father without being his sister, so she had to be daughter of one of her father's brothers. Seventeenth-century tradition, recorded by Mateus Cardoso (letter to Manuel Rodrigues, 1624, MMA, vii, 291) related that two 'cousins' (dous primos seus) of the son of Lukeni lua Nimi, the first king, held that office before Nzinga's own father and predecessor Nkuwu a Ntinu (alternatively Nkuwu a Lukeni, since Ntinu is a title). In Kikongo, there is no distinction between cousins and brothers (mpangi), but here it seems likely that the original genealogy did recognize this distinction, since his role as eldest son of his father would not make sense otherwise. Therefore Leonor Nzinga a Nlaza should have been the daughter of one of those cousins' fathers, who therefore would have borne Nlaza as his first name. Elsewhere, in his account, Morte de D. Alvaro III ElRei do Congo e eleição de D. Pedro Affonso', 1622, MMA, xv, 494. Cardoso gave the other cousin's name (here as a nephew of Lukeni) as Quinanga.
34. Note, however, that his name is given as 'Panso Aquitima' (Mpanzu a Kitima) in the oldest account of this event by de Barros, Decadas de Asia: Primeira Década, Book 3, cap. 10, 35, suggesting that his father was not João I Nzinga a Nkuwu, otherwise the second element of his name would be Nzinga. Kitima was probably not a personal name, but a nickname, relating to ferocity. In the seventeenth-century genealogical account of Mateus Cardoso, his name is given as 'Panzo-à-Muzinga' (Mpanzu a Nzinga) which would support the idea that he was a half brother: Cardoso to Rodrigues, 1624, MMA, vii, 292.
35. De Barros, Asia, Decadas de Primeira Década, Book 3, 35v.
36. Afonso mentioned writing an earlier letter that gave full details on his succession in his letter of 5 Oct. 1514, printed in MMA, 1, 295. While this letter, written probably in 1509, is no longer extant, it seems likely to have been de Barros's source. It may also have been a source for the earliest printed account of the event, Martin Fernandez de Enciso, Suma de Geographia q[ue] trata de todas las partidas & provincias del mundo... (Seville, 1519), 109-10 (penciled pagination in unpaginated text found in Biblioteca Nacional de Lisboa, Reservados 717v).
37. João III to Afonso (c. 1531), MMA, 1, 535.
38. Brásio dates the letter to 1529, but François Bontinck has argued cogently for the later date: See his 'L'ancien Congo et le Concile de Trente', Antennes, 2 (1962), 459-60.
39. If João was simply repeating what he had read in letters from Kongo, and this phrase repeats one made by Afonso, the Kikongo equivalent would probably have been mwana ye ngudi a ntinu, which might also mean 'niece and aunt of a king'.
40. Hilton has argued that this position was institutional, as a 'chief's co-wife', which she dubs nzimbu mpungu, on the strength of seventeenth-century documentation (Kingdom of Kongo, 38). I believe that the Kikongo title in her source would be better read as nsimba mpungi since the text reads 'Manicimbanpungui' (Biblioteca Nacional de Madrid, MS 3533, Antonio de Teruel, 'Descripcion Narrativa de la Mission de los PP Capuchinos ... en el Reyno de Congo', [1664], 124-5).
41. Thus arguing against the position of Susan Broadhead, borrowed from theoretical work of Karen Sachs, that wives represented subordinate figures as opposed to sisters, see 'Slave wives, free sisters', 160-2.
42. Afonso to Manuel 1, 5 Oct.1514, MMA, 1, 313.
43. João III to Afonso, c. 1531, MMA, 1, 533.
44. François Bontinck, 'De nouveau sur D. Afonso, roi de Congo', African Historical Studies, 3 (1970), 151-62.
45. Afonso had sought this dispensation for himself and a new wife, Caterina. This was probably not the Caterina mentioned as sister of Pedro, as Bontinck suggests (154 n. 9). For while it might be true, as he notes, that 'in a Bantu pen son or daughter, brother or sister, etc., have a more extended meaning than in the West', the confusion that could be created by pairs of words represented by the same Kikongo term (mpangi for sister/brother and mwana for son/daughter, nephew/niece) would not apply to sister/mother (mpangi/ngudi). Moreover, for Caterina to be Pedro's sister, she would also have had to be Afonso's daughter; impossible since he knew her only to be of the third degree of consanguinity to him.
46. The inquest has survived because it was submitted to Portugal to secure the extradition of one of the conspirators, Rodrigo de Santa Maria, who had served for many years as a Kongo royal factor in the Portuguese island colony of São Tomé.
47. Auto de Devassa, D. Diogo I, 10 Jan. 1552 (copying inquest of April 1550), MMA, 11, 261-2. It is not clear whether the Manynãbuã is the same sister who is there in Simgua and mother of Dom Bastiã (who elsewhere is identified as one of the principal plotters).
48. It is possible that in the confusion of translation, Pedro might have meant that Catarina was his cousin (daughter of his uncle or aunt and not of his father), since the same Kikongo word mpangi can mean either prima (female cousin) or irmã (sister) in Portuguese.
49. That such a descent name might be used on various occasions is confirmed by our first description of Kongo, in 1491, in which the king (João I, whose Kikongo personal name was Nzinga a Nkuwu) was called 'Monimolynymy' or 'Monymochanimj'. Both indicated descent from someone named Nimi: Possibly Nimi a Nzima, father of Lukeni lua Nimi. These two forms are given in the Italian and Portuguese recensions of de Pina's account: See Radulet (ed.), Cronista, 120 (Italian) and 148 (Portuguese).
50. The Italian form (Monymochanimj) might be glossed as mwene mukwa (the modern nkwa) Nimi, or lord owner of Nimi. The Portuguese version's middle syllable is less easily understood, but might be seen as the same if one allows that one or another scribe miscopied the abbreviated letter combination 'que' (q with a til over it, which was often rather carelessly left to drift upwards in the paleography of the day) in the original report as an '1'. João's personal name (Nzinga) is given in a near contemporary source, a letter of Gregorio de Quadra of 1520, published in Damião de Goes's chronicle, 'Cronica da Felicissimo Rei D. Manuel' of 1545, Pt. 4, cap. 9, extract published in MMA, xv, 59.
51. On this term see François Bontinck, 'Les quimbares: Note sémantique', Africa (Rome), 31 (1976), 41-55.
52. While daughter and sister of a king, she would not be the mother of one, as suggested by the tradition recorded by Mateus Cardoso in 1624 that Afonso had a son and three daughters who were either kings or the mothers of kings. The Christian names of two are given as Izabel and Ana, the third one, whose Christian name was not cited, was mother of Diogo and therefore not likely to be the Catarina, Mateus Cardoso to Jesuit Rector, 1624, MMA, vii, 292.
53. In twentieth-century Kikongo traditions, clans often took the names of their founder as their names, usually in the double form. However, there are a number of clans that take a title, like Mfutila (tax collector) or Capitau (Captain) as their origin: See Cuvelier (ed.), Nkutama, passim. This collection of names and summary traditions of some 500 clans is a good sample of clan naming practices.
54. Hence the historical proverb, 'Kinkanga, Kimpanzu, Kinlaza, makukwa matatu malambila Akongo' or 'Kinlaza, Kimpanzu, Kinkanga, the three stones on which Kongo cooked'. The proverb is reported in Cuvelier (ed.), Nkutama (1972 edn.), 73, from the clan motto (mvila) of the Kinkanga clan of Nkondo (near Songolo), Democratic Republic of Congo.
55. For further context see Thornton, 'Origins', 13.
56. On Álvaro's descent from Izabel lua Lukeni, see Cardoso to Rodrigues, MMA, vii, 292. Izabel and her relationship to her father could have been personally known to people still alive in 1624, as person of 75 years at that time would have been in his or her late teens at the time that Álvaro came to the throne, when her descent would have been much discussed.
57. Diogo I was a descendant of another of Afonso's daughters named Nzinga a Mvemba, and was therefore also Izabel Lukeni lua Mvemba's cousin, Pigafetta, Relatione, 58.
58. Notes of Cornelio Gomes for the 'emjenda a V Real Señoria', c. 1552, MMA, xv, 164.
59. Cornelio Gomes to Diogo Mirón, 26 Mar. 1554, MMA, xv, 181.
60. Pigafetta, Relatione, 58. This account gives his age at his accession in 1568 as 'around 25 years old'.
61. Pigafetta, Relatione, 57-8.
62. Suggested by François Bontinck in 'Un mausolée pour les Jagas', Cahiers d'études africaines, 20 (1980), 187-9.
63. Anonymous Jesuit letter, 15 Dec. 1587, MMA, iii, 350
64. Carta annua da Provincia de Portugal, 1588, ibid. iii, 378-9.
65. Richard Gray, 'A Kongo princess, the Kongo ambassadors, and the papacy', Journal of Religion in Africa, 29 (1999), 140-54.
66. Manuel Baptista Soares, 'Relação dos customes, ritos e abusos do Bispado do Congo', 7 Sept. 1619, MMA, vi, 379.
67. Writing two years later, Álvaro III described da Silva as executor of his father's will, Álvaro III to Paulo V, 25 Oct. 1617, MMA, vi, 289-90.
68. [Mateus Cardoso], 'Levatemento' 1622, MMA, xv, 530-7,
69. and 'Morte', ibid. 482-3.
70. 'Alcune Provisioni', summarizing an act of 20 Jan. 1616, MMA, vi, 252. The wording of the provision makes the unnamed 'mother of the same one [Alvaro]' go on the mission, who is then presented as uncle of Álvaro III. Her name is not known, but she was taken in war from the Marquis of Wembo, Mayala ma Nsamba by Álvaro II. It is certainly an exaggeration to describe her as a slave, as Hilton does (Kingdom of Kongo, 87), or to develop the idea that kings of this family consistently chose sons of 'slave wives' to succeed them.
71. Mateus Cardoso to Jesuit Rector, 1624, MMA, vii, 292.
72. [Mateus Cordoso], História do Reino de Congo (ed. António Brásio, Lisbon, 1969), cap. 15, fol. 16v.
73. Authorship is established in the French translation of François Bontinck, 'Histoire du royaume de Congo, 1624', Études d'histoire Africaine, 4 (1972), Introduction.
74. On Pedro's piety and the females at court, see Antonio Franco, Synopsis Annalium Societatis Jesu in Lusitania ab Anno 1540 usque ad Annus 1725 (Augsburg, 1726), 1624, no. 15, 241.
75. Cardoso, 'História', cap. 25, fol. 35v. The wording of the passage is confusing, however. He first mentions a variety of families who have taken Portuguese names (of which da Silva is most obvious in this period), and then notes that the royal house is called Afonso to indicate descent from the Christian king. In another paragraph he states that the two houses 'correspond to the House of Bragança and that of Aveiro, to which they marry their sons and daughters, except the prince who will succeed to the kingdom, because this they do with the House of Bata'. The latter statement may have been true in theory or tradition, but was manifestly not in practice.
76. da ida de o Padre Martens Cardoso da Companhia de Jesv fez ao Reino de Congo', 14 Sept. 1625, MMA, vii, 381.
77. Cardoso, 'Morte', MMA, xv, 483, 487, 493. The House of Soyo was probably not mentioned in the 'História', as it was not a royal one, save by its supplying of women to men who eventually did become kings. Christina's son, Pedro II, was born in 1573, and if she were twenty at the time he was born, she was then born in 1553. If her father was Miguel, and was also twenty when she was born, he would then have been in his late 50s when he became Count, thus rendering this relationship quite possible. If the House of Soyo were founded earlier than Miguel's reign (slightly before 1591), it would have been part of the royal family in accordance with Lopes's contention that the counts in the mid-1580s were always drawn from the royal house.
78. Johannes de Laet, Historic ofte Iaerlijck verhael van der verrichtingen der Geoctroyeerde West Indische Compagnie (Leiden, 1644, mod. edn. by S. P. L'Honoré Naber (4 vols.) (Hague, 1931-7), 66, notes that the Countess was sister of the king (then Garcia I, though since he had just been crowned it is possible that the Dutch believed her to be Garcia's sister when she was Pedro's, in which case she was daughter of Christina).
79. The name of the Count is given in 1627 as Paulo by Fernão de Sousa, letter to his sons, 1631 (summarizing correspondence of Ambrósio I, King of Congo, in 1627) in Beatrix Heintze, Fontes para a História de Angola no século XVII (2 vols.) (Wiesbaden, 1985-8), i, 275.
80. The count ruling in June 1626 was the paternal uncle of Garcia I, thus brother of Pedro II, Franco, Synopsis Annalium, 1626, no. 16, 247, who had to have been Paulo;
81. see also de Sousa to sons, in Heintze, Fontes, i, 247 (summarizing a letter from the oui vidor of Kongo).
82. Manuel Antonio was still ruling Soyo in 1624, Fernão de Sousa, 15 Aug. 1624, MMA, vii, 248.
83. Fernão de Sousa to King, 9 July 1626, MMA, vii, 461.
84. Additional details on the attack in Franco, Synopsis Annalium, 1626, nos. 15-16, 247
85. Fernão de Sousa to Governo, 15 Apr. 1626, MMA, vii, 432-3
86. Biblioteca Nacional de Lisboa, MS 241, Manuel Severim da Faria, 'Historia Portugueza', 229.
87. Franco, Synopsis Annalium, 1626, no. 16, p. 248. Franco, probably following the usage of the Jesuit living in São Salvador at the time, João da Pavia, uses the Latin term 'Dynasts' frequently to refer to powerful people at court. It appears to be his shorthand for these people rather than a separate named category of nobility or an institutionalized position. However, given da Pavia's close knowledge of Kongo's court, the possibility of some sort of institutional group can not be wholly ruled out.
88. Olfert Dapper, Naukeurige Beschrijvinge der Africaense Gewesten (2nd edn., Amsterdam, 1676), 211. Paulo was still ruling on 16 Feb. 1641 (Alvaro VI died on 22 Feb.), when he wrote to the Dutch West India Company, Algemeen Rijksarchief (Hague) Oude West Indische Compagnie, 56 (not consistently paginated, this letter carries an ink p. 36 and a penciled p. 153 on it).
89. In his letter to the Pope, da Silva argued that selection of the count is vested in Soyo and not the king, Daniel da Silva to Pope Innocenzio X, 25 Mar. 1648, MMA, x, 122-5, he also gives an account which is only clear if one follows the course of events in Franco, Synopsis Annalium.
90. Girolamo da Montesarchio, 'Viaggio al Gongo' (1668), mod. edn. by Calogero Piazza in La Prefettura apostolica del Congo all metã del XVII secolo. La relazione inedita di Girolamo da Montesarchio (Milan, 1976), with original foliation marked, 3v, in 1648 she was back in Soyo;
91. Leonor one of the members of the royal council in 1648 was sister of the king's wife and sister of the Count of Soyo, Biblioteca Nacional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 125.
92. Dapper, Naukeurige Beschrijvinge, 211-12
93. Da Silva to Pope, MMA, x, 124-5.
94. On the Soyo-Kongo dispute and its consequences, see Thornton, Kingdom of Kongo, 69-83.
95. Biblioteca Nacional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 118-20.
96. For the Duke's name, see da Montesarchio, 'Viaggio', 40v.
97. Biblioteca Nacional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 199, 151.
98. BN Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 152.
99. Ibid. 124.
100. Biblioteca Estense, Modena, MS Italicus 1380, Alpha N-9-7, Giacinto Brugiotti da Vetralla's untitled chronicle, c. 1658, ed. Giuseppe Monari da Modena in 'Viaggio al Congo...' (c. 1726), 459.
101. Biblioteca Nactional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 49.
102. Biblioteca Nactional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 49. 124-5.
103. Hilton argues (Kingdom of Kongo, 38) that Leonor Mwene Simb'a Mpungi was the 'king's female co-chief, invested like him with the powers of the other world', a suggestion for which there is no direct support in the text. She also contends that she must have been his paternal aunt, since she was head of the Kinlaza lineage ('Mani Mulaza' according to de Teruel, 'Descripcion Narrativa', 49 - on this see the preceding note). She alters the orthography of her second title to nzimbu mpungu and translates it as 'great nzimbu' (where nzimbu is the money used in the country). The text says 'simba' without a nasal (which should appear in this phrase by elision, and with an 's' not a 'z'), which makes her reading questionable. It may not, therefore, be the same title as the Marquis of Nzimba mpangi ('Nzimba npanghi' in his Italian orthography) from whose lineage Dona Beatriz Kimpa Vita would claim, in 1705, that the Virgin Mary descended.
104. See Bernardo da Gallo, 'Relazione dell'ltime Guerre civili del Regno di Congo...', 22 Mar. 1711, 305,
105. published in Teobaldo Filesi, 'Nazionalismo e religione nel Congo all'inizio del 1700: La setta degli Antonioni', Africa (Rome), 9 (1971), 268-303, 463-508, 645-68 (also published separately in Rome [1972], the edition marks the foliation of the original MS)
106. Hilton, Kingdom of Kongo, 92-3.
107. Saccardo argued that she was either a daughter of Pedro II and sister of Garcia I, or daughter of Álvaro II and sister of Álvaro III (Graziano Saccardo, Congo e Angola con la storia dell'antica missione dei Cappuccini [3 vols.] [Venice, 1982-3], 1, 453). Neither option is possible, since Garcia II was the same generation as both Garcia I and Alvaro III, and she would therefore be a cousin (prima) and not an aunt (tia). If she were the same princess who sought to obtain Carmelite nuns from the Spanish missionaries in 1585 or so, and was 15 at the time, she would have been 78 years old when de Teruel arrived in São Salvador.
108. Biblioteca Nacional de Madrid, MS 3533, de Teruel, 'Descripcion Narrativa', 49. One is tempted to see this as the Kinlaza lineage, which is not possible, since Álvaro and Garcia were the founders of that lineage that would later dominate Kongo's history, and she was from a different branch of the royal family by any reckoning.
109. Biblioteca Nacional de Madrid, MS 3533, Teruel, 'Descripcion Narrativa', 125.
110. For a careful study of the events using de Teruel's and da Vetralla's accounts, see Saccardo, Congo e Angola, 1, 453-5.
111. Da Montesarchio, 'Viaggio', 30-30v (the personal names of the Duke and Duchess are found on 6 and 6v).
112. Da Montesarchio, 'Viaggio', 20.
113. Giovanni Antonio Cavazzi da Montecuccolo, Istorica descrizione de' tre regni Congo, Matamba ed Angola (Bologna, 1687), Book 2, para. 76.
114. 'Relação da mais gloriosa e admirável Victoria que alançarão as armas de elRey D. Affonso VI' (1665), MMA, XII, 584-5.
115. Thornton, Kingdom of Kongo, 69-97.
116. Saccardo, Congo e Angola, 1, 455.
117. Antonio Gradisca da Zucchelli, Relatione del viaggio e missione di Congo nell'Etiopia inferiore occidentale (Venice, 1712), 225-6, 228
118. on the interpretation of her relationships, and the story of her faction, see Thornton, Kingdom of Kongo, 95-105.
119. For the dates of his reign, see Instituto Histórico e Geografica Brasileiro (Rio de Janeiro), lata 6, pasta 2, 'Catallogo dos Reis de Congo' (c. 1758), 5.
120. António de Oliveira de Cadornega, História geral das guerras angolanas (1680-81), ed. Joseé Matias Delgado and Manuel Alves de Cunha (3 vols.) (Lisbon, 1940-2, reprinted, 1972), 11, 362, 405. It is possible from Cadornega's description that there were two powerful Dona Anas, both holding the title Mwene Nlaza. One was the wife of Garcia, the other his sister. Cadornega's Ana Manimulaza was also the grandmother of Antonio I, which is not possible if he were Garcia II's son. But he refers to her withdrawing to the Mbidize River, and this is consistent with Ana Afonso de Leão's career. He also claimed she was a close relative to the Princes of Soyo (and hence a da Silva).
121. Girolamo Merolla da Sorrento, Breve e succinta relatione del viaggio nel regno di Congo nell'Africa Meridionale ... (Naples, 1692), 301.
122. Marcellino d'Atri, 'Giornate apostoliche fatte da me Fra Marcellino d'Atri ... 1690' [1702], in Carlo Toso (ed.), L'anarchia congolese nel sec. XVII. La relazione inedita di Marcellino d'Atri (Genoa, 1984) (original pagination marked)
123. D'Atri, 'Giornate', 344-5.
124. D'Atri, 'Giornate', 371.
125. Luca da Caltanisetta, 'Relatione della Missione fatta nel Regno di Congo per il Padre Fra Luca da Caltanisetta ... sino alla fine del 1701', in Raimondo Rainero (ed.), Il Congo agli inizi del setecento nella relazione del P. Luca da Caltanisetta (Florence, 1974) (original foliation marked), 103.
126. See Thornton, Kingdom of Kongo, 95-111
127. Thornton, Kongolese Saint Anthony, 24-6, 36-52, 75-7 and passim.
128. D'Atri, 'Giornate', 463-4
129. da Caltanisetta, 'Relatione', 90v-91.
130. See also Thornton, Kongolese Saint Anthony, 79.
131. Traditions noted in Mbanza Kongo, Aug. 2002.
132. The fuller story is in Thornton, Kongolese Saint Anthony.
133. Support for his descent comes from a local traditional chronicle of Matadi, collected by Jean Cuvelier in the late 1920s, published in part in Nkutama (1934 edn.), 72.
134. [Bernardo d'Asti], 'Missione in Practica', Introduction, para. 3, mod. edn. By J. Northumb, as La pratique missionnaire des PP. Capucins italiens das les royaumes de Congo, Angola et contrées adjacentes (Louvain, 1931).
135. Biblioteca Vaticana, Fondo Borgia Latina, MS 269, 45, Cherubino da Savona, 'Congo 1775: Breve Ragguaglio del Regno di Congo, e sue Missione scritto dal Padre Cherubino da Savona..'. at fol. 41, published in Carlo Toso, 'Relazioni inedite di P. Cherubono Cassinis da Savona sul 'Regno del Congo e sue Missioni"' in L'Italia Francescana, 45 (1975).
136. French translation was published in Louis Jadin, 'Aperçu de la situation du Congo et rite d'élection des rois en 1775, d'après le P. Cherubino da Savona, Missionaire au Congo de 1759 à 1774', Bulletin de l'Institute Historique Belge de Rome, 35 (1963). I have cited the MS foliation to allow comparison of texts.
137. Cuvelier (ed.), Nkutama (1934 edn.), 72. The chronology is approximate, the only datable person in the list is D. Garcia Nkanga a Mvemba who, it says, went to govern Kongo. He left for Kongo to become Garcia IV in 1743, and was replaced by Garcia Mbwandi, the sisters were his second and third successors. A variant found in the 1972 edition, 57, makes the first of the sisters his third successor. Tradition makes her member of a lineage called Kintumba, and specifically not a Kinlaza as her predecessors had been, so she is not likely to have been the Dona Isabela of the document, unless the Agua Rosadas's had a Kikongo clan name of Kintumba.
138. Da Savona, 'Breve Ragguaglio, 41V-42.
139. Academia das Cienças de Lisboa, MS Vermelho 296, Rafael de Castello de Vide, 'Viagem', 73.
140. Ibid. 141-2.
141. Archivio Provinciale de Cappuccini, Tuscany, Filippo Bernardi da Firenze, 'Relazioni d'alcuni Missionari Cappuccini Toscani singolarmente di P Lorenzo da Lucca ...', part 2, Relazione of 31 Dec. 1707, 317 (original pagination is given in the French translation of Jean Cuvelier, Relations sur le Congo du Père Laurent de Lucques (1700-1717 [Brussels, 1953]).
142. Da Savona, 'Breve Ragguaglio', 42.
143. Ibid.
144. Biblioteca Nacional de Lisboa, Códice 8742, 101V Regimento of Governor Francisco Innocenzio de Sousa Coutinho to Alvaro Rodrigues, Capitão Mor of Encoge, 28 Nov. 1766; 103, Circular letter to Queen of Huando, Brites Affonso da Silva, n.d. (c. late Nov. 1766).