Hostname: page-component-cd9895bd7-mkpzs Total loading time: 0 Render date: 2024-12-26T04:58:23.509Z Has data issue: false hasContentIssue false
  • English
  • Français

Reflections on Early Interlacustrine Chronology: An Essay in Source Criticism1

Published online by Cambridge University Press:  22 January 2009

David P. Henige
David P. Henige
Affiliation:
University of Birmingham
Get access Rights & Permissions [Opens in a new window]

Extract

The scholar interested in the early history of the interlacustrine area has an unusually large corpus of traditional evidence available. These sources are among the most detailed in tropical Africa and several full-scale works have been based on them. This paper seeks, through textual analysis, to demonstrate that some of the most important of these sources have been influenced by the content of earlier writings and by each other, and that their corroborative value is very small.

Three problems are of particular interest here—the alleged contemporaneity of Nakibinge of Buganda, Olimi Rwitamahanga of Bunyoro, and Ntare Nyabugaro of Nkore; the Biharwe eclipse and its ascribed dates; and the value for chronology of the accounts of Nyoro invasions southward. Emphasis on these aspects has meant that while the questions of the Bacwezi, the Nkore capitals, and the Nyoro tombs have been taken into account, specific attention has not been paid to them here.

The present paper seeks only to suggest possible alternatives to the presently accepted reconstruction of early interlacustrine history, and argues that the nature of our evidence, once divested of its synthetic accretions, precludes the development of comprehensive hypotheses. It is important at this stage to attempt a full reassessment of these traditional sources through comparative textual analysis and through the extensive use of archival documentation which may illuminate more clearly the milieu in which the traditional historiography of the interlacustrine region developed.

Type
Research Article
Information
The Journal of African History , Volume 15 , Issue 1 , January 1974 , pp. 27 - 46
Copyright
Copyright © Cambridge University Press 1974

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

2 Head, M. E., ‘Inter-tribal history through tribal stories’, Uganda Journal [henceforth UJ], X (1946), 108.Google Scholar

3 Oliver, R., ‘Reflections on the sources of evidence for the pre-colonial history of east Africa’ in Vansina, J., Mauny, R. and Thomas, L-V. (eds.), The Historian in Tropical Africa (London, 1964), 316.Google Scholar

4 Kaggwa, Apolo, Ekitabo kya Basekabaka be Buganda (London, 1901 with further editions in 1912 and 1927);Google ScholarK. W. ‘Tito Gabafusa Winyi], ‘The Kings of BunyoroKitara’, UJ, III (1935), 155–60; IV (1936), 75–83; V (1937), 53–67;Google ScholarNyakatura, J. W., Abakama ba Bunyoro Kitara (St Justin, Quebec, 1947);Google ScholarKatate, A. G. and Kamagungunu, L.Abagabe b'Ankole (Kampala, 1955).Google Scholar

5 For a convenient summary see Oliver, R., ‘Discernible developments in the interior, c. 1500–1840’, in Oliver, R. and Mathew, G. (eds.), History of East Africa (Oxford, 1963), 180–9.Google Scholar

6 The Galla hypothesis was first advanced by Speke in 1863 and was generally accepted for the next 75 years. Speke, J. H., Journal of the discovery of the source of the Nile (London, 1863), 246–8.Google Scholar The Bito were identified as Luo by Crazzolara, J. P., ‘The Lwoo people’, UJ, v (1973/1938), 1–21.Google Scholar

7 Oliver, ‘Discernible developments’, 172.Google Scholar

8 E.g., Cohen, D. W., ‘A survey of interlacustrine chronology’, J. Afr. Hist. XI (1970), 177–94;CrossRefGoogle ScholarPosnansky, M., ‘Towards an historical geography of Uganda’, East African Geographical Review, I (1963), 10–12;Google Scholaridem, ‘Kingship, archaeology and historical myth’, UJ, XXX (1966), 4–5; Karugire, S. R., A history of the kingdom of Nkore in western Uganda to 1896 (Oxford, 1971), 16–32et passim;Google ScholarDunbar, A. R., A history of Bunyoro-Kitara (Nairobi, 1965), 34–38;Google ScholarUzoigwe, G. N., ‘Succession and civil war in Bunyoro-Kitara’, International Journal of African Historical Studies, VI (1973), 49–55.CrossRefGoogle Scholar

9 See below, pp. 8–9.Google Scholar

10 Most clearly demonstrated in Cohen, ‘Survey’, chart opp. p. 190.Google Scholar

11 See the sources cited in footnotes 8 and io. For the literature on the Biharwe eclipse, see section II below. In referring to Nyoro rulers mentioned by K. W., I use a prefix indicating their order in his list in order to suggest their likely chronological placement as he interpreted it.Google Scholar

12 Gorju, J., Entre le Victoria, I'Edouard et I'Albert (Rennes, 1920), 155–61.Google Scholar

13 Kaggwa, Apolo (ed. and tr. by Kiwanuka, M. S. M.), The Kings of Buganda (Nairobi, 1971), 28.Google Scholar

14 Gorju, Entre le Victoria, 62–4.Google ScholarCf. Fisher, R., Twilight tales of the black Baganda (London, 1911), 137–9.Google Scholar

15 Ibid.

16 Kaggwa, Basekabaka, 316. I have used the 1912 edition of this work when referring to material in it concerning the other states in the area which is not included in Kiwanuka's edition. A later edition published in 1927 is identical in content but the pagination varies slightly.Google Scholar

17 The role of the Bacwezi in determining the early chronology of the interlacustrine area is, like everything about them, subject to much debate. See particularly Wrigley, C. C., ‘Some thoughts on the Bacwezi’, Uganda Journal, XXII (1958), 11–17. In the conclusion of this paper I briefly address myself to this problem.Google Scholar

18 Gorju, Entre le Victoria, 157, 159. Although most of Gorju's account of the Bacwezi and the Bito was based on Ruth Fisher's work, she did not identify the opponent of Cwamali as Ntare Kiitabanyoro. This was probably an inference drawn by Gorju from the Nkore ruler's sobriquet.Google Scholar

19 Ibid. 1958–1959, 160–1. Elsewhere Gorju dated these events to ‘the beginning of the eighteenth century’, Ibid. 152. Apparently the first specific claim that Buganda was founded by the twin brother of the first Nyoro ruler appeared in L-M-A Linant de Bellefonds, ‘Itinéraire et voyage de service fait entre le poste militaire de Fatiko et Ia capitale de M'tesa, roi d'Uganda, fév-juin 1875’, Bulletin de la Société Khédiviale de Géographie de l'Egypte, I (1876/1877), 12 de Bellefonds was told this by two rivals for the office of mukama (king) of Bunyoro.Google Scholar

20 On dating of the Biharwe eclipse see, e.g. Gray, R., ‘Annular eclipse maps’, J. Afr. Hist. IX (1968), 148–9Google Scholar and Cohen, D. W., ‘The Cwezi cult’, J. Afr. Hist. XXVII (1968), 652.Google Scholar

21 Kiwanuka, M. S. M., ‘The empire of Bunyoro-Kitara: myth or reality?’, Canadian Journal of African Studies, II (1968), 32–2, 39–40, 43;Google Scholaridem, The history of Buganda from the foundation of the kingdom to 1900 (London, 1971), 274–5. Here Kiwanuka comes close to rejecting the Olimi reference in K. W. but refrains from actually doing so.

22 E.g. Sykes, J., ‘The eclipse at Biharwe’, UJ, XXIII (1959), 44–50;Google ScholarHaddon, E. B., ‘Kibuka’, UJ, XXI (1957), 111–19;Google ScholarGray, J. M., ‘The solar eclipse in Ankole’, UJ, XXIII (1963), 217–22.Google Scholar

23 K. W., ‘The Kings of Bunyoro-Kitara’, 79.Google Scholar

24 D. P. Henige, ‘K. W.'s Nyoro kinglist: oral tradition or the result of applied research?’, a paper presented at the Fifteenth Annual Meeting of the African Studies Association, Philadelphia, Pa., 8–11 Nov. 2972. An expanded version of this paper will be forthcoming shortly, while a briefer look at K. W.’s work is in my The Chronology of Oral Tradition: Quest for a Chimera (Oxford, 1974), 105–14.Google Scholar

25 I have used Nakibinge in titling this section although there is no direct traditional evidence that he was necessarily a contemporary of Nyabugaro. This is a convention of the historiography and in fact matters little since it is Olimi I who ties the package together. In this discussion of Ganda chronology, Kiwanuka (Buganda, 275) cites a ‘tradition’ that recorded an eclipse in the reign of Mulondo. Actually this seems to have been no more than the testimony of a single informant who may have inferred an eclipse to Mulondo’s time on the basis of familiarity with K. W.'s account, which definitely placed it after the death in battle of Nakibinge.Google Scholar

26 Katate and Kamagungunu, Abagabe. I have used a typescript translation of this work by John A. Rowe in the Center for Research Libraries, Chicago, Illinois.Google Scholar

27 And, perhaps significantly, this appearance seems to have been selective and perhaps confined to literate informants like Katate. For instance it does not appear in the traditions of Nkore collected by Fr. Nicolet, apparently, from internal evidence, in the mid-1940s. See his ‘Régions qui se détachèrent du Kitara et devinrent des royaumes indépendants’, Annali Lateranensi, XXXIV–XXXVI (1970/1972), 234.Google Scholar

28 Karugire, Nkore, 249.Google Scholar

29 Ibid. citing Katate and Kamagungunu as the Nkore traditions in question.

30 Oliver, , ‘The traditional histories of Buganda, Bunyoro, and Nkore’, Journal of the Royal Anthropological Institute, LXXV (1955), 114–17, provided a preliminary analysis of Katate and Kamagungunu's work, which had just appeared but he made no effort at extensive exegesis.Google Scholar

31 See, among others, Sykes, ‘Eclipse at Biharwe’, 44–50;Google ScholarGray, J. M., ‘The eclipse in Ankole’, 217–21;Google ScholarHaddon, ‘Kibuka’, 114–17;Google ScholarVansina, J., L'évolution du royaume rwanda dès origines à 1900 (Brussels, 1962), 54–5.Google Scholar

32 E.g. Gray, , ‘Kibuka’, UJ, xx (1956), 69–70;Google Scholaridem, ‘Eclipse in Ankole’, 218–19. Only recently, with the refinement in the measurement of eclipse paths, the bases of choice has included questions of magnitude and likely visibility. Gray, R., ‘Eclipse maps’, J. Afr. Hist. vi (1965), 251–62CrossRefGoogle Scholar and idem, ‘Annular eclipse maps’, J. Afr. Hist. x (1968), 147–57. But it will be argued later in this paper that such considerations are often relatively tin important.

33 This conclusion assumes that the eclipse sighted by Nyabugaro was different from that sighted by Juuko and the same as that sighted by Olimi according to K. W.Google Scholar

34 Gray, J. M., ‘The early history of Buganda’, UJ, II (1934/1935), 268;Google ScholarGray, R., ‘Eclipse maps’, 257.Google Scholar

35 Kaggwa, Basekabaka, 316.Google Scholar

36 I have used the 1912 edition of Basekabaka because it was the first to contain capsule histories of other interlacustrine states. I have not seen the 1901 edition but presume that an account of Juuko's eclipse appeared in it. However, Stanley's account of Ganda history collected in 1875 did not mention art eclipse in connexion with Juuko: Stanley, H. M., Through the Dark Continent (2 vols., London, 1878), II, 350. Nor did a manuscript written in igoo by Auguste Achte, a White Father, ‘Histoire des rois baganda’ in White Fathers' Archives, Rome.Google Scholar

37 Kaggwa, Basekabaka, 316.Google Scholar

38 K. W., ‘Kings’, 79.Google Scholar

39 Katate and Kamagungunu, Abagabe, 53–4.Google Scholar

40 Gray, ‘Eclipse in Ankole’, 220.Google Scholar

41 In his articles K. W. did not name the place where Olimi I allegedly saw this eclipse. The locale of Biharwe (‘people say the eclipse was seen at Biharwe’) was supplied to Gray, evidently in response to importunities, much later, probably in 1955 or 1956. Gray, ‘Kibuka’, 69.Google Scholar

42 For Mbaguta see Williams, F. Lukyn, ‘Nuwa Mbaguta, Nganzi of Ankole’, UJ, x (1946), 324–35, and Karugire, Nkore, 3.Google Scholar

43 Briefly, this argument involves a comparative analysis of the several published and unpublished Nkore royal genealogies which suggests that several ekyebumbe (usurpers) were transformed into legitimate rulers by Katate and Kamagungunu—a process which, inter alia, lengthened the genealogy by several generations. In addition there are indications that the exploits of Ntare Kiitabanyoro were attributed partly to Nyabugaro who may not have had an independent existence at all.Google Scholar

44 The arguments supporting this interpretation are rather complex and cannot be included here.Google Scholar

45 Oliver, R., ‘Ancient capital sites of Ankole’, UJ, XXIII 56.Google Scholar

46 Oliver, ‘Ancient capital sites’, 56, 59–60. Nicolet, ‘Régions qui se détachérent du Kitara’, 237–8, 262, assigned several more Sites to Ntare Kiitabanyoro than did Katate and Kamagungunu.Google Scholar

47 The problem is complicated by the conflicting opinions regarding the number of capitals of early Nkore rulers. Roscoe, John, The Banyanhole (London, 1923), 36, stated that capital sites were changed with great frequency in the early period. Oliver (‘Ancient capital sites’, 62) argued on the basis of Katate and Kamagungunu that ‘the frequent changing of sites was a late nineteenth century innovation’ and that ‘in earlier centuries a single site was occupied for a whole reign’. The information in Nicolet differs from both of these interpretations for he shows the earliest rulers with several capitals, the middle group (with the exception of Ntare Kiitabanyoro) generally with only one site, and all the rulers from Kahaya I (c. 1800) with several sites. Nicolet, ‘Regions qui se détachèrent du Kitara’, 261.Google Scholar

48 For the eclipse of 1680 see Kiwanuka, Buganda, 276. In accordance with the analytical approach of this paper, it should be added that it is not inconceivable that the Nkore account of the eclipse in Kaggwa's 1912 edition of Basekabaka may have been designed to emulate the Juuko reference in the 1901 edition.Google Scholar

49 Gray, R., ‘Annular eclipse maps’, 151. Even on the basis of genealogical reckoning, Juuko could be dated as late as a notional 1732, although Kiwanuka, History of Buganda, 285, prefers to date him rather earlier than 1680.Google Scholar

50 Kaggwa, Basekabaha, 316.Google Scholar

51 K. W., ‘Kings’, 79.Google Scholar

52 This same point was made to me by Dr H. F. Morris of the University of London.Google Scholar

53 Vansina, L'évolution, 86, et passim.Google Scholar

54 Ibid. 54. Gray, ‘Annular eclipse maps’, 151–2, preferred to date this eclipse to 1796 on the basis of magnitude and likely visibility. Vansina's acceptance of the Nyoro ruler as ‘Cwa’ was based on the work of Alexis Kagame who in his later writings called this invader ‘Cwa son of Nyabongo’. See, e.g. his Abrégé de l'ethno-histoire du Rwanda (Butare, 1972), 71–2. He attributes this designation to ‘les mémorialistes Rwandais et les aèdes du genre dynastique’ but at the moment I favour the opinion that it is feedback from Torelli's list in Gorju, Entre le Victoria, 64. The exact synchronization: ‘sur lea deux hates le nombre des monarques eat égal: 19 depart et d'autre’ seems to be the principal basis for Kagame's argument. Surely it is too much to expect Rwanda traditions to remember the name of the father of a foreign invader. See also p. 42 below.Google Scholar

55 Vansina, L'évolution, 54, 56.Google Scholar

56 Vansina, ‘The use of oral tradition in African cultural history’ in Gabel, C. and Bennett, N. R. (eds.), Reconstructing African Cultural History (Boston, 1967), 62.Google Scholar

57 Van Noten, F. L., Les tombes du roi Cyirima Rujugira et tie la reine Nyirayuhi Kanjogera: description archéologique (Tervuren, 1972).Google Scholar

58 Personal communications from Richard Sigwalt and David W. Newbury.Google Scholar

59 Gray, ‘Eclipse maps‘ and idem, ‘Annular eclipse maps’.Google Scholar

60 Herodotus, , The Periian Wars, I: 74.Google Scholar

61 E.g. Oliver, ‘Ancient capital sites’, 56–60;Google ScholarCohen, ‘Survey’, 187–8.Google Scholar

62 Wilson's account in Johnston, H. H., The Uganda Protectorate (2 vols., London, 1900), II, 596–8;Google Scholar Fisher's account in ibid. 598–9; Roscoe, , The Bakitara or Banyoro (Cambridge, 1923), 88–9.Google Scholar

63 Czekanowski, Jan, Wissenschaftliche ergebnisse der Deutschen-Zentral-Afrika Expedition 1907–8, VI/I, Ethnographie-Anthropologie, I, Forschungen (Leipzig, 1917), 56–7.Google Scholar

64 M. Posnansky in introduction to reprint edition of Fisher, , Twilight tales (London, 1969), xxvi.Google Scholar

65 Ibid. 133–4 and chart opp. xxviii.

66 Ibid. 138.

67 Meldon, J., ‘Notes on the Bahima of Ankole’, Journal of the African Society, VI (1906/1907), 136–53, 234–49.Google Scholar

68 Ibid. 242.

69 Kaggwa, Basekabaka, 316.Google Scholar

70 Ibid. 318–9.

71 K. W., ‘Kings’, 79. According to K. W., Nyabugaro fled to Kagera island, an action attributed to Ntare Kiitabanyoro in Kaggwa. In any case the place of refuge of the Nkore ruler seems on the face of it to be an unlikely detail to have been preserved by Nyoro traditions.Google Scholar

72 Ibid. 81–2.

73 Cf. ibid. Kaggwa, Basekabaka, 315, and Fisher, Twilight tales, 133–4.

74 K. W., ‘Kings’, 82. K. W. mentioned Mugambwa as the leader of the Nkore forces, another datum mentioned in Kaggwa.Google Scholar

75 Fisher, Twilight tales, 137–9, distinguished between an Olimi and an Isansa, whom she called Olimi's son.Google Scholar

76 K. W., ‘Kings’, 56.

77 Kaggwa, Basekabaka, 318. Karugire, Nkore, 157, argues that Karaiga's tenure was too short to have allowed him to engage in ‘a foreign war’ and rejects K. W.'s account on this basis. It would be better, it seems, to reject it on the basis of ex silentio, since the tenure or secureness of Karaiga's rule would have had no influence on any decision of the Banyoro to invade Nkore.Google Scholar

78 K. W., ‘Kings’, 57.Google Scholar

79 Kaggwa, Basekabaka, 319–20.Google Scholar

80 One of the most suspicious aspects of K. W.'s account is the astonishing amount of circumstantial (not to mention corroborative) detail regarding Nkore arid Buganda. For instance Kaggwa mentioned only one mukama of Bunyoro by name whereas K. W. found it necessary to name no fewer than i i Buganda rulers, often as patent obiter dicta. Early Nkore traditions called all Nyoro rulers Chwarnale while K. W. is able to name at least 4 abagabe. It is more than a little paradoxical that all non-Nyoro names in K. W. are known from other sources while so few Nyoro rulers are.Google Scholar

81 For the irredentism of Bunyoro during this period see Beattie, John, The Nyoro State (Oxford, 1971), 82–7. This is confirmed by a personal communication, 17 Oct. 1972, from Fr Antoine Caumartin, a White Father who spent over thirty years in the area.Google Scholar

82 Karugire, Nkore, 156–8, argues from different premises that Nyoro traditions are confused regarding their relations with Nkore during this period.Google Scholar

83 K. W., ‘Kings’, 81.Google Scholar

84 In fact, it is hard to believe that raids of this kind were not a fact of life in this area with the numerous herds of milk cattle as the prizes sought. Over the course of time it would become more and more difficult for tradition to distinguish among any of these as being more important than others although ‘invasions’ which resulted in the occupation of the royal capital, for instance, would tend to be remembered the longest from the perspective of the defeated but not necessarily from that of the victor who might be inclined to regard this accomplishment as a casual by-product of a campaign with different ends.Google Scholar

85 See Ishumi, A. G. M., ‘The kingdom of Kiziba’, Journal of World History, XIII (1971), 721–4.Google Scholar

86 Rehse, H., Kiziba, Land und Leute (Stuttgart, 1910), 245–6;Google ScholarCésard, E., ‘Le Muhaya’, Anthropos, XXXII (1937), 42–5.Google Scholar

87 Pagès, A., Un royaume hamite au centre de l'Afrique (Brussels, 1933).Google Scholar Earlier accounts of Rwanda such as Czekanowski, Forschungen, 287–92, provided no details of early rulers.Google Scholar

88 See above, note 54.Google Scholar

89 Pagès, Un royaume hamite, 558.Google Scholar

90 Ibid. 558 n.

91 It is not clear how this co-optative process might have occurred. We know that K. W. in his researches travelled as far as Rwanda and this may have affected later Rwanda traditions. Or it may simply be a matter of Kagame's being influenced by the material in Fr Gorju's work.Google Scholar

92 See Ford, J. and de Z. Hall, R., ‘The history of Karagwe (Bukoba District)’, Tanganyika Notes and Records, no. 24 (1947), 3–25.Google Scholar

93 John Roscoe, in entitling his book on the Banyoro The Bakitara, did so because he felt they should more properly be termed so.Google Scholar

94 Speke, J. H., Journal of the discovery of the source of the Nile (London, 1863), 261–2.Google Scholar

95 K. W., ‘Kings’, 156, 76.Google Scholar See also his introduction to Dunbar, A. R., A history of Bunyoro-Kitara (Nairobi, 1965), V.Google Scholar

96 Ibid. x–xi.

97 Linant de Bellefonds, ‘Itinéraire’, provides the only dissenting voice in this otherwise harmonious chorus by stating that the name Bunyoro came into vogue when Buganda seceded. His source, however, was Mutesa himself (Ibid. 63), who may have exaggerated the importance of Buganda in this early period. Certainly Toro lay precisely in the region described by Speke and Dunbar as quintessentially Kitara. See also Beattie, Nyoro state, 56.

98 Ingham, K., ‘The amagasani of the Abakama of Bunyoro’, UJ, XVII (1953), 139–45.Google Scholar

99 Hofmayr, W., Die Schilluk. Geschichte, Religion und Leben eines Niloten-Stämmes (Mödling, 1925), 63. Significantly, Hofmayr stated that the reigning Shilluk ruler ‘settled the controversy’ over the earlier ruler's existence by erecting the tomb.Google Scholar

100 E.g., Posnansky, M., ‘Kingship, archaeology, and historical myth’, UJ, XVII (1966), 5–6.Google Scholar

101 See, for example, Oliver, R., ‘A question about the Bachwezi’, UJ, XVII (1953), 135–7;Google Scholar Wrigley, ‘Some thoughts’; Mungonya, Z. C. K., ‘The Bacwezi in Ankole’, UJ, XXII (1958), 18–21;Google ScholarBeattie, J., ‘Group aspects of the Nyoro spirit mediumship cult’, Rhodes-Livingstone Journal, xxx (1961), 11–34;Google ScholarCory, H., ‘The Buswezi’, American Anthropologist 57 (1955), 923–32;CrossRefGoogle ScholarPosnansky, , ‘Bigo bya Mugenyi’, UJ, XXXIII (1969), 143–4,Google ScholarWrigley, , ‘The story of Rukidi’, Africa, XLIII (1973), 228–32.Google Scholar

102 Schnitzer, E., ‘Journal einer Reise von Mrúli nach der Hauptstadt Unyoro's mit Bemerkungen über Land und Leute’, Petermann's Mitteilungen, XXIV (1879), 182.Google ScholarCf. Casati, G., Ten years in Equatoria and the return with Emin (2 vols., London, 1891), II, 273.Google Scholar The account in Grant, James A., A walk across Africa (London, 1864), 292–3, conforms in some of its details to the modern interpretation.Google Scholar