Published online by Cambridge University Press: 22 January 2009
The Ashanti bureaucracy as we know it from the nineteenth century owed its structure and character to reforms in government initiated by Osei Kwadwo (1762–77) and pushed forward by his successors Osei Kwame (1777–ca. 1801) and Osei Bonsu (ca. 1801–1824). Many previously hereditary offices became appointive: biographical studies of leading officials in the early nineteenth century show that they were appointed, and promoted, by the king on merit, without regard to their origins. Freemen and slaves, Ashanti and non-Ashanti, were recruited into the bureaucracy. Administrative skills tended to be transmitted through patri-kin, and in the course of the nineteenth century powerful groups grew up which, generation after generation, provided the king with trained officials. An administrative class emerged, distinguished from other elements in society by its dependence upon the king for its very existence. Elaborate checks were instituted to prevent the bureaucracy transforming itself from a controlled to a ruling one.
Bureaucratization in Ashanti produced little quantitative change in the pattern of office growth. Its effects were upon the quality of government. A distinction began to take shape between the king acting in his private capacity through his palace functionaries, and in his public capacity through his bureaucratic officials. Government became increasingly efficient, and was extended over fields of activity previously untouched. The growth of the bureaucratic process was reflected in the increasingly absolutist nature of the Ashanti state.
1 See, for example, the description of Dompoase in 1817, in Bowdich, T. E., Mission from Cape Coast Castle to A shantee (1819), 28.Google Scholar
2 Rattray, R S., Ashanti (1923); Religion and Art in Ashanti (1927); Ashanti Law and Constitution (1929).Google Scholar
3 Bowdich (1819), passim; also Bowdich, An Essay on the Superstitions, Customs, and Arts common to the Ancient Egyptians, Abyssinians, and Ashantees (1821).Google Scholar
4 Wilks, I., ‘Ashanti government in the nineteenth century”, in West African Kingdoms, ed. Kaberry, P. (in the Press).Google Scholar
5 See Wirtschaft und Gesellschaft, III, 6 (1922), 650–78.Google Scholar
6 The Institute of African Studies, University of Ghana, has two series of relevant records: IAS/AS—Ashanti stool histories recorded by Mr J. Agyeman-Duah of Kumasi, and IAS/AS/CR—transcripts of proceedings of the Kumasi Divisional Council. Among actor-reports I would also include the records of the Dutch, Danish, and English establishments on the Gold Coast, preserved in the metropolitan archives of the three countries.Google Scholar
7 Bowdich (1821), p. 54, also p. 21.Google Scholar
8 Robertson, G. A., Notes on Africa (1859), 199.Google Scholar
9 Bowdich (1819), 246, also 236, 252 n.Google Scholar
10 Bowdich (1819), 83. The first appointments of resident commissioners in the Accra area were in fact made in 1776, the last year of Osei Kwadwo's reign, when Boakye, Ankra and Nkansan assumed responsibility for the Dutch, English and Danish forts respectively (see letter from Factor van der Peuye to Director General, dd. Accra, 9 November 1776, in the Elmina Journal, Archives of the Dutch Settlements on the Guinea Coast, General State Archives, The Hague (Carson 608)).Google Scholar
11 The career of Opoku Frefre is reconstructed mainly from IAS/AS/CR 101: Constitutional dispute relating to the Gyasewa Stool succession, Kumasi Divisional Council, hearing commenced 21 December 1939; and from numerous references in Bowdich (1819), J. Dupuis, Journal of a Residence in Ashantee (1824). Opoku Frefre appears to have retired from active control of financial affairs in or about 1818, probably after his public display of wealth late in 1817, reported by Hutchison, in Bowdich (1819), 395, and remembered in tradition, see AS/CR/101. He died c. 1826.Google Scholar
12 This summary of the career of Agyei synthesizes the slightly conflicting accounts of Bowdich (1819), 248–9, and R. Lee, Stories of Strange Lands (1835), ch. v.Google Scholar
13 I have outlined the career of another important official, Muhammad al-Ghambac, in ‘The position of Muslims in metropolitan Ashanti in the early 19th century’, in Islam in Tropical Africa, ed. I. Lewis (in the Press). Muhammad al Ghambac was a son of the Imam of Gambaga in Mamprussi. He arrived in Kumasi in or shortly before 1807, and was taken into the king's service. Ten years later he was one of the king's closest advisers, with special responsibility for the affairs of the northern provinces. Some of his correspondence, in Arabic, is preserved in Cod. Arab CCCII, Oriental Section, Royal Library, Copenhagen.Google Scholar
14 Commission dd. 31 July 1875, in L'Explorateur, III (1876), 238. I am indebted to Mrs Marion Johnson for drawing my attention to this document.Google Scholar
15 Ellis, A. B., The Land of Fetish (1883), 188–91;Google ScholarAffairs of the Gold Coast and Threatened Ashanti Invasion (1881), 51–4: letter from Owusu Ansa to Gov. Ussher dd. Cape Coast, 17 May 1880.Google Scholar
16 Affairs of the Gold Coast (1881), 135–6.Google Scholar
17 Hutton, W., Voyage to Africa (1824), 128;Google ScholarDupuis (1824), xxviii and 184.Google Scholar
18 See, for example, Cruickshank, B., Eighteen Years on the Gold Coast of Africa, 1 (1853), 340–1.Google Scholar
19 Diary, of Asante, David, 1884, in Mitteilungen der geographische Gesellschaft zu Jena, iv (1886), 32.Google Scholar
20 See Gros, J., Voyages, aventures et captivité de J. Bonnat chez les Achantis (1884), 178. See also, for example, Affairs of the Gold Coast (1882), Lonsdale's Mission to Kumasi, Salaga, Yendi, etc., 58, 68, and 79.Google Scholar
21 ‘Les Achantis d'après les relations de M. Bonnat’, in L'Explorateur, 1, ii (1875).Google Scholar
22 Little investigation has been made of the adwumfo—goldsmiths, blacksmiths, umbrella-makers, etc.—but it seems clear that in the transmission of such crafts claims of patrifiliation were most relevant; see, for example, Rattray (1923), 301.Google Scholar
23 For which in general see Fortes, M., ‘Kinship and marriage among the Ashanti’, in African Systems of Kinship and Marriage, ed. Radcliffe-Brown, A. R. and Forde, D. (1950), 252–84;Google ScholarFortes, M., ‘The “submerged descent line” in Ashanti’, in Studies in Marriage and Kinship, ed. Schapera, I. (1963), 58–67.Google Scholar
24 I have often heard elderly men from such groups remark, ‘My wife's brothers cannot take my sons away from me while I am alive. If they try, I shall sue them.’ The retention of sons is also facilitated by the levirate.Google Scholar
25 Rattray (1929), 92, probably overestimates the importance of slaves in the formation of these groups.Google Scholar
26 These observations are based mainly upon interviews with holders of senior esomdwa in Kumasi, 2–6 August 1965. My field-notes are deposited in the Institute of African Studies, University of Ghana, and will be referred to as Kumasifleld-notes (1965).Google Scholar
27 For some effects of such a change in role, in another Context, see Wilks, I., ‘Akwamu and Otublohum: an eighteenth-century Akan marriage arrangement’, in Africa, XXIX, 4 (1959), especially 398.Google Scholar
28 Bowdich (1819), 296–7.Google Scholar
29 There were two occupants of the Pinanko stool before Asumadu, but these are not strictly regarded as occupying the post of Gyaasewahene, since it was only with Asumadu that the stool was vested by the king with administrative functions.Google Scholar
30 This genealogy is based upon IAS AS/15 IAS AS/CR/101; Kumasi field-notes (1965); information collected by Mr A. C. Denteh, 1928–1965; and also upon miscellaneous references in archival and published sources too numerous to be detailed here. I have reconciled, at some few points, slightly conflicting evidence. Kwame Tuah, the eleventh occupant of office, was appointed by the British administration, then the superior authority, but acting, as all members of the Gvaasewa I have spoken to agreed, quite constitutionally.Google Scholar
31 This genealogy is based upon IAS AS/41; Kumasi field-notes (1965); and also upon miscellaneous documentary sources.Google Scholar
32 IAS AS/157.Google Scholar
33 Rattray (1929), 92, 118.Google Scholar
34 The adehyedwa are, in general, stools which originated independently of the king, i.e. either pre-Ashanti stools which were incorporated into the new kingdom in the late seventeeenth century, or non-Ashanti stools which subsequently transferred allegiance to Ashanti. The great majority of these stools are matrilineal; one of the few exceptions is the Asafo stool of Kumasi, IAS AS/38. With the adehyedwa, and distinct from the mmammadwa, are the poduodwa, stools created by the Ashanti kings but vested in a lineage, e.g. the Bantama stool of Kumasi, IAS AS/3g-40. The adehyedwa, poduodwa and esomdwa (including mmamrnadwa) are all established stools, in the sense that their occupants swear allegiance to the king; they are thus distinguished from the abusuadwa or family stools, which are non-established and the concern only of the lineage.Google Scholar
35 The term was suggested to me by Mr P. C. Gibbons.Google Scholar
36 For state trading, see the excellent description in Hayford, Casely, Gold Coast Native Institutions (1903), 95–6. For the Batahene see IAS AS/I; Kumasi field-notes (1965).Google Scholar
37 P.R.O., T. 70/41: letter from T. E. Bowdich dd. Kumasi, 9 July 1817; see also Bowdich (1819), 83.Google Scholar
38 State Archives, The Hague: Dutch settlements on the Guinea Coast, 39, Journal of the Visit to Kumasi of W. Huydecoper, entries for and 26 April 1817.Google Scholar
39 Based upon IAS AS/73; Kumasi field-notes (1965).Google Scholar
40 State Archives, The Hague: Dutch settlements on the Guinea Coast, Elmina Journal, letter from Van der Peuye dd. Accra, 9 November 1776, and subsequent letters.Google Scholar
41 Affairs of the Gold Coast (1882): Lonsdale's mission, 63–4.Google Scholar
42 I hope to publish in a forthcoming paper the evidence for Ashanti's diplomatic relations, in the nineteenth century, with the caliphate of Sokoto, the imamate of Futa Jallon, and with the kingdoms of Dahomey and Mossi.Google Scholar
43 Bowdich (1821), 28; (1819), 298.Google Scholar
44 Bowdich (1819), 289.Google Scholar
45 Ricketts, H. J., Narrative of the Ashantee War (1831), 125.Google Scholar
46 Affairs of the Gold Coast and Threatened Ashanti Invasion (1881), 153–6.Google Scholar
47 See Claridge, W. W., A History of the Gold Coast and Ashanti, II (1915), ch. xix.Google Scholar
48 Arabic MS: Al-risālah bi 'l-malak yasamma Būakī, Balme Library, University of Ghana (Arabic version of the report of Boakye Tenten's mission to Prasu and Elmina).Google Scholar
49 Affairs of the Gold Coast (1882), 66.Google Scholar
50 Dupuis (1824), 235–6.Google Scholar See also Priestley, M. and Wilks, I., ‘The Ashanti kings in the eighteenth century: a revised chronology’, J. Afr. Hist. I, 1 (1960), 93.Google Scholar
51 Respectively IAS AS/a; 9; 147; 42; 21 and 45.Google Scholar
52 For the Malinke see Bowdic (1821), z, together with Dupuis (1824), 124 n. and xxxvii. For the Hausa see Affairs of the Gold Coast (1881), 51 et seq.Google Scholar
53 Letter from Hutchison dd. Kumasi, 3 February 1818, in T. E. Bowdich, A Reply to the Quarterly Review (1820).Google Scholar
54 Kumasi Divisional Council records: minutes of 12 January 1961. This stool was unusual among esomdwa in that it always appears to have passed to brothers and nephews, but not to sons.Google Scholar
55 Bowdich (1819), 294.Google Scholar See also Dupuis (1824), 505.Google Scholar
56 For the Tandoh case see Bowdich (1819), 823; Huydecoper's Journal, entries for 8, 16, 22 Sept., 18, 23 Oct., 7, 9, 54, 15, 89 Nov., Dec. 1816 and 7 Jan. 1817.Google Scholar
57 Hutchison's Diary, 7 and 8 November 1817, in Bowdich (1819), 393. For Adusei, see IAS AS/127.Google Scholar
58 Bowdich (1819), 255.Google Scholar
59 Ibid. 293.
60 Ibid. 295.
61 Ibid. 97. IAS AS/75.
62 Hutchison's Diary, 14 November 1817, in Bowdich (1819), 395. This event is also remembered in tradition; see IAS AS/CR/101.Google Scholar
63 Bowdich (1819), 256. IAS AS/81.Google Scholar
64 Ibid. 255.
65 For the Kommenda case, see Bowdich (1819), 87–8, 101, 113, 121. For the Cape Coast case see PRO. T. 70/41, letter from the king of Ashanti to Governor, Cape Coast Castle, dd. Kumasi, 1 October 1817.Google Scholar
65 Casely, Hayford (1903), 95–6.Google Scholar
66 Bowdich (1819), 293.Google Scholar
68 Ibid. 309.
69 Ibid. 294.
70 Huydecoper's Journal, entry for 7 May 1817.Google Scholar
71 Bowdich (1819), 123.Google Scholar See also Meredith, H., An Account of the Gold Coast of Africa (1812), 168–9.Google Scholar
72 Bowdich (1819), 123.Google Scholar
73 See IAS AS/1–162.Google Scholar
74 Wilks, , ‘The growth of the Akwapim state: a study in the control of evidence”, in The Historian in Tropical Africa, ed. Vansina, J., Mauny, R. and Thomas, L. V. (1964), 396 et seq. In this paper I suggested that the normal distribution of reign lengths is linked with other regular distributions, e.g. age at coming to office, age at death. It is now clear to me that the crucial factor is the succession system, i.e. the ratio of lateral (brother, etc.) to vertical (son, nephew, etc.) succession. In Kumasi, therefore, the strong tendency to constancy in length of tenure of office reflects the incidence of fraternal succession, which is similar for all types of stools.Google Scholar
75 The data were mainly collected in 1964. The assumption is made that, where T = 1, the occupant is in mid-tenure; hence T = 1 corresponds to 1964 less a half of C, strictly 1955–56.Google Scholar
76 Weber, III, 6 (1922).Google Scholar
77 Bowdich (1819), 296.Google Scholar
78 See, for example, Huydecoper's Journal, 2 December 1816 et seq.; Freeman, T. B., Journal of Two Visits to the Kingdom of Ashanti (1843), 18.Google Scholar
79 See, for example, Bowdich (1859), 232, 296; Lee (1835), 174.Google Scholar
80 There was one major limitation upon spatial continuity. The amanto, the original Ashanti towns founded around Kumasi in the late seventeenth century, preserved into the nineteenth century a largely autonomous character, and the king's bureaucracy possessed no jurisdictions within the territories of the chiefs of Bekwai, Juaben, Kokofu, Nsuta and Mampong. Attempts in the nineteenth century to remove this anomaly brought Kumasi into conflict with these towns, and contributed greatly to the collapse of Ashanti in the face of the British challenge.Google Scholar
81 Bowdich (1819), 253.Google Scholar
82 L'Explorateur, 1 (1875), ii.Google Scholar
83 Cruickshank, 1 (1853), 341–5.Google Scholar
84 State Archives, The Hague: Dutch Settlements on the Guinea Coast, 384, despatch from Daendels dd. 6 December, 1816.Google Scholar
85 Dupuis (1824), lxi.Google Scholar
