¹ Edward Searle, Komenda, to CCC, 14 Feb. 1695, Rawlinson C.746, #1402. Unless otherwise indicated, all letters cited here originated from Komenda. The numbering of the letters in the Rawlinson collection conforms to a guide prepared in 1972, copies of which are located in the Bodleian Library and several other institutions.

² I concentrate here on the seventeenth and eighteenth centuries for three reasons. First, it was the time in which John Kabes flourished. Second, the earlier period was not characterized by frequent socio-economic contact, was monopolized by the Portuguese, and is deficient in documentation. Finally, to include the nineteenth century, with the shift to legitimate commerce and its ramifications (particularly the markedly increased capacity and propensity of the Europeans to interfere, whether with gunboats or Bibles) would invalidate much of the analogy.

³ There is debate about the role and composition of the grumetes and about the various meanings of the term in the literature of the period. See especially Nardin, J., ‘Recherche sur les “gourmets” d'Afrique occidentale’, Revue française d'histoire d'outre-mer, liii (1966), 215–44CrossRefGoogle Scholar; Rodney, Walter, A history of the Upper Guinea coast, 1545–1800 (Oxford, 1966), 77. 202Google Scholar; Boulègue, Jean, Les Luso-Africains de Sénégambie (Dakar, 1972), 11–14Google Scholar; Curtin, Philip D., Economic change in pre-colonial Africa: Senegambia in the era of the slave trade (2 vols.: Madison, 1975), I, 95–7Google Scholar. For the lançados the most detailed treatment is da Silva, M., ‘Subsídios para o estudo dos lançados na Guiné’, Boletim Cultural da Guiné Portuguesa, xxv (1970), 25–40, 217–32, 397–420, 513–60.Google Scholar

⁴ Nardin, , ‘Recherche,’ 234–5.Google Scholar

⁵ Rodney, , Upper Guinea, 215Google Scholar speaks of the grumetes as ‘completely detribalized’ — perhaps too severe a description.

⁶ For the Caulkers see Scotland, D. W., ‘Notes on the Banana Islands, A.D. 1462–1846’, Sierra Leone Studies, n.s. xi (1958), 149–60Google Scholar; Fyfe, Christopher, A history of Sierra Leone, (London, 1962), 10, 81, 248–50Google Scholar; Caulker, Patrick, ‘The autochthonous people, British colonial policies, and the Creoles in Sierra Leone’, Ph.D. Thesis, Temple University, 1976.Google Scholar

⁷ Depending on how one defines ‘class’ it may be possible to argue that the grumetes did not constitute one, but I can envisage no definition of the term which would encompass the coastal middlemen traders of the Gold Coast. Hence I would disagree with Daaku, K. Y.'s description of them as ‘a new class’ in chapter V of his Trade and politics on the Gold Coast, 1600–1720 (Oxford, 1970)Google Scholar. For the Akani (ethnic group, social class, trading diaspora?) see Boahen, A. A., ‘Arcany or Accany or Arcania and the accounts of the sixteenth and seventeenth century Europeans' records’, Transactions of the Historical Society of Ghana, xiv (1974), 105–12Google Scholar; Kea, R. A., ‘Trade, state formation, and warfare on the Gold Coast, 1600–1826’, Ph.D. Thesis, University of London, 1974, 88–155.Google Scholar

⁸ See Horton, Robin, ‘From fishing village to city-state: a social history of New Calabar’, in Douglas, Mary and Kaberry, Phyllis, eds., Man in Africa, (Garden City, 1971), 38–58Google Scholar. Several scholars, most notably E. J. Alagoa, have posited pre-contact dates for the establishment of several of the most important Niger delta states and argue that they only became more complex in response to the trade. Discussion of this point is deferred to the conclusion of this paper.

⁹ See instance, Lever, J. T., ‘Mulatto influence on the Gold Coast in the early nineteenth century’, African Historical Studies, iii (1970), 253–61CrossRefGoogle Scholar. None of the important Cape Coast leaders, e.g., Edward Barter, Thomas Awishee, or Cudjoe Caboceer, have yet received extended attention.

¹⁰ For purposes of this paper ‘Komenda’ refers to the town of that name on the coast (which was variously named by the Fante, Portuguese, Dutch, and English) while ‘Eguafo’ refers to the state then usually called ‘Grand Commany’, to which the town then belonged.

¹¹ Bosman, Willem, A new and accurate description of the coast of Guinea, (London, 1705). 30–1Google Scholar. For the personal relations between Kabes and Bosman see below, pp. 16–7. The only other detailed source for Kabes so far utilized by historians is the diary of Baillie, William, agent at Komenda from 1714 to 1716, T70/1464, PRO.Google Scholar

¹² E.g., Claridge, W. W., A history of the Gold Coast and Ashanti (2 vols.: London, 1915), I, 146–7.Google Scholar

¹³ Daaku, , Trade and politics, 115–27.Google Scholar

¹⁴ I refer here to the correspondence in Rawlinson C.745–7, Bodleian Library, Oxford, and the James Phipps papers in the Public Record Office, Chancery Masters Exhibits, C113/261–95.

¹⁵ Daaku, , Trade and politics, 115Google Scholar; Harper, C. H., Report of Commission of Enquiry … into the relationship of the stools of Nkusukum and Commendak (Gold Coast. Sessional Paper #5 of 1917/18), 25Google Scholar; editorial notes to 1967 reprint of Bosman, , Description, 514.Google Scholar

¹⁶ In 1702 the Director-General at Elmina spoke of Kabes as ‘nephew to the Great Mercador Aban’ of Elmina: William de la Palma to X, 26 June 1702, WIC 198. The English agent at Komenda noted that Kabes had a ‘cozen’ in Dutch service: William Cross to CCC, 18 Mar. 1686, Rawlinson C.745, #830. See also pp. 6, n. 23.

¹⁷ Cross to CCC, 18 Mar. 1686, Rawlinson C.745, #830; similarly, Cross to CCC, 14 Sept. 1686, Rawlinson C.746, #1013.

¹⁸ The first mention of this John Cabessa seems to be in 1663 when he was called ‘the Captain of the Hill’ in Kormantin and was firmly engaged in the English interest there, so much so that in a skirmish with the townspeople he was wounded in the head. Thomas Davies et al., Kormantin, to East India Company, London, 4 Mar. 1665, Original Correspondence, E/3/27, fos. 194^v-195, India Office Records, London.

¹⁹ Great Britain. Calendar of State Papers, Colonial Series: America and the West Indies, 1661–1668, (London, 1880), 294Google Scholar. Perhaps not entirely coincidentally some later Komenda traditions associated Commeh, who they thought had founded the town, with Kormantin. See Henige, D., “The problem of feedback in oral tradition: four examples from the Fame coastland’, J. Afr. Hist., xiv (1973), 229–30.Google Scholar

²⁰ The name Kabes and its variants was presumably of Portuguese origin (‘cabeça’ meaning ‘corporal’ and the like) but in this instance at least it seems to have become somewhat of a family name. There is no evidence that Kabes was either Christian or educated. The few letters in the Rawlinson correspondence from him seem to have been written for him by someone at Komenda fort.

²¹ Compare, for example, the terms of the agreement of 1686 with the actions of the Eguafo rulers in 1714, below, pp. 16–17.

²² Komenda Diary sub 29 May 1716, T70/1464, PRO.

²³ Tando had fled to Elmina from Komenda in 1714 when he was described as a ‘nephew’ of Kabes. The Dutch alleged that Tando had been born at Komenda: Elmina, Journal sub 27 Dec. 1714, WIC 124Google Scholar. I owe this reference to Albert van Dantzig.

²⁴ Phipps, et al. to RAC, 2 July 1722, C113/274, 240–241^vGoogle Scholar. The Dutch also feared disorders and their Agent at Fort Vredenburg ‘urgently’ requested powder: Elmina, Journal, 23 June 1722, Furley Collection, Balme Library, University of Ghana, 1715–30, 1718–23, pp. 223Google Scholar. The details of this succession are discussed below, pp. 13–4.

²⁵ For French activities see Wiltgen, R. M., Gold Coast mission history, 1471–1880 (Techny, Ill., 1956), 42–54, 59–60, 64–6Google Scholar, and sources cited there; Villault de Bellefond, Nicolas, Relation des costes d'Afrique appélés Guinée (Paris, 1669), 191–3Google Scholar; Roussier, Paul, ed., L'établissement d'lssiny, 1687–1702 (Paris, 1935), x–xviiiGoogle Scholar. For a recapitulation of the Dutch and English activities see Daaku, , Trade and Politics, 78–80Google Scholar, and van Dantzig, A. and Priddy, B., A short history of the forts and castles of Ghana (Accra, 1971), 10.Google Scholar

²⁶ See, for instance, Lionel Staveley, Kormantin, to East India Company, 1 May 1658, Original Correspondence, E/3/25, #2648, India Office Records, London.

²⁷ Bosman, , Description, 29–30Google Scholar; undated memorandum of Dalby Thomas [?], C113/291, 24–24^v, PRO. See also van Dantzig, A., ‘English Bosman and Dutch Bosman: a comparison of texts’, History in Africa, ii (1975), 214, n. 26.Google Scholar

²⁸ James Nightingale to CCC, 16 Nov. 1681, Rawlinson C.745, #125; same to same, 19 Jan. 1682, Rawlinson C.745, #223.

²⁹ Daaku, , Trade and politics, 118–23.Google Scholar

³⁰ Undated and unsigned treatise (probably Dalby Thomas, , c. 1708), C113/274, 272Google Scholar. Some of Kabes's activities as a supplier as discussed in Daaku, , Trade and politics, 120–2.Google Scholar

³¹ Patrick, Edward to CCC, 19 Nov. 1717, 3 Dec. 1717, C113/275, 126, 127Google Scholar. Sometimes, though, corn was shipped from Cape Coast to Komenda: e.g., Thomas, to RAC, 25 Apr. 1707Google Scholar, in Davenant, , Writings, 5: 218.Google Scholar

³² Komenda Diary sub 21 June 1715, 28 June 1715, 23 July 1715, T/70/1464, PRO.

³³ William Ronan to CCC, 30 Nov. 1691, Rawlinson C.747, #2788. Ronan was one of the Chief Agents at Cape Coast, sent down to supervise the intended re-establishment.

³⁴ Ronan to CCC, 30 Nov., 2 Dec. 1691, Rawlinson C.747, #2789, 2790. The attempt was abandoned immediately after the last letter.

³⁵ Thomas Willson to CCC, 1 Dec. 1694, Rawlinson C.747, #3046; Edward Searle to CCC, 8 Feb. 1695, Rawlinson C.746, #1395; same to same, 15 Feb. 1695, Rawlinson C.746, #1406. For Kabes's instrumental help in building the factory in 1686 see Cross to CCC, 8 Dec. 1686, Rawlinson C.745, #1107.

³⁶ For such payments see below, pp. 10.

³⁷ Du Casse's account in Roussier, , L'établissement, 16–26Google Scholar; Daaku, , Trade and politics 80–1.Google Scholar

³⁸ Cross to CCC, 26 Mar. 1686, Rawlinson C.745, #835.

³⁹ That is, to Apr. of 1688, after which there are no more letters until Nov. of 1691.

⁴⁰ See below, pp. 15–16. Perhaps Kabes realized that the English felt a greater need to remain at Komenda than did the Dutch, already established at both Elmina and Shama. Still, there is ample evidence that the Dutch courted Kabes assiduously.

⁴¹ Dalby Thomas opined that Komenda was ‘like to be the best [point of trade] on the coast’ and the Mareen of Eguafo, the official designated to deal with the Europeans on commercial matters, had gone to reside there. Thomas, to RAC, 1 Aug. 1706, T70/5,. 17^v-18Google Scholar. See also Daaku, , Trade and politics, 116–8.Google Scholar

⁴² Komenda Diary sub 29 May 1715, 11 Aug. 1715, T70/1464, PRO.

⁴³ Bennett, Thomas, CCC, to RAC, 3 Jan. 1717, T70/6, 45Google Scholar. Bennett complained that Kabes was ‘never punished for clandestine trading while Commenda was held on the Company's account but has been since it was on the Chief's [Bleau's] account’: Ibid. 47. Details of Bleau's and Baillie's arrangement are not available but presumably they differed little from the terms under which Bennett held Accra at about the same time. For these see C113/292, 16–16^v. We do have the instructions given to Edward Patrick, Baillie's successor at Komenda, dd. 15 June 1716. C115/261, PRO.

⁴⁴ John Rootsey to CCC, 6 Apr. 1697, Rawlinson C.746, #1947; Komenda Diary sub 23 Feb. 1715, 29 Mar. 1715, 10 May 1715, T70/1464, PRO.

⁴⁵ E.g., see Cross to CCC, 14 July 1686, Rawlinson C.745, #930; Samuel Chambers to CCC, 5 July 1687, Rawlinson C.747, #2302; Robert Elwes to CCC, 25 Aug. 1687, Rawlinson C.747, #2355; Searle to CCC, 13 Feb. 1695, Rawlinson C.746, #1395; Gerrard Gore to CCC, 7 July 1698, Rawlinson C.747, #3790.

⁴⁶ For some indication of the variety of goods stocked by the Royal African Company see Davies, K. G., The Royal African Company, (London, 1957), 170–9, 350–7Google Scholar, and Johnson, M., ‘The Ounce in eighteenth-century west African trade’, J. Afr. Hist., vii (1966), 197–214.CrossRefGoogle Scholar Neither of these discusses the varying demand from place to place on the coast.

⁴⁷ If anything, the local correspondence suggests that Dutch-made goods, especially Flemish sayes and brass pans (which were lighter in weight than those manufactured in England) were in greater demand than comparable English goods.

⁴⁸ Clearly price was an important factor, but quality and utility also mattered. We know that price levels were negotiable (and negotiated); so too, we must assume, were these non-price factors but since any such accommodations would not appear in the account books, there is no useful way to measure them.

⁴⁹ Barbot, Jean, ‘Journal du voyage de Guinée, Cayenne et Illes Antilles de l'Ameréque, A^os 1678 et 1679’, British Museum, Add. 28788.Google Scholar

⁵⁰ E.g., Bosman, , Description, 27Google Scholar; Barbot, , A description of the coasts of North and South-Guinea, (London, 1732), 154Google Scholar; de Bellefond, Villault, Relation, 191–2Google Scholar; Du Casse in Roussier, , L'établissement, 10, 33Google Scholar. Dapper, Ogilby and other compilations naturally reflect this as well.

⁵¹ For instance, the ground rent for Komenda fort was paid to the Head Caboceer of Komenda rather than to the agent of the Eguafo ruler after the 1730s. T70/411, 23^v; T70/417, 10, PRO.

⁵² What part he did play will be discussed in the next section.

⁵³ Cross to CCC, 8 Dec. 1686, Rawlinson C.745, #1107.

⁵⁴ This judgment is based on numerous letters in the Rawlinson series from 1685 to 1688 and 1694 to 1696.

⁵⁵ T70/1463 sub 26 Feb., 28 Feb., 14 Mar., 16 Mar., 20 Mar. 1704. The English seemed to feel (and, in the event, correctly) that the Eguafo ruler would be more amenable to their threats than Kabes. Pace Daaku, , Trade and politics, 122Google Scholar, there is no evidence that Kabes played any part in persuading the Eguafo ruler to disavow his sale of the hill to the Dutch.

⁵⁶ Komenda Diary sub 24 Nov. 1714, T70/1464, PRO.

⁵⁷ Searle to CCC, 23 Feb. 1695, Rawlinson C.746, #1419; same to same, 3 Mar. 1695, Rawlinson C.746, #1431.

⁵⁸ Howsley Freeman to CCC, 21 May 1695, Rawlinson C.746, #1511- More on Kabes and the Adorns is in Gerrard Gore to CCC, 8 Feb. 1700, in Davenant, , Writings, 5: 205–6Google Scholar. In these cases the distinction between panyarring as a kind of personal, civil action and warfare as a public, military action is not always clear. That is, Kabes may merely have been panyarring in some of the instances cited.

⁵⁹ Thomas Willson to CCC, 30 Nov. 1694, Rawlinson C. 747, #3045; Searle to CCC, 8 Feb. 1695, Rawlinson C.746, #1395.

⁶⁰ Gore to CCC, 21 June 1698, Rawlinson C.747, #3778.

⁶¹ Komenda Diary sub 4 Nov. 1714, 10 May 1715, 16 July 1715, 15 Aug. 1815 T70/1464; Williams, William to CCC, 17 Jan. 1718, C113/275, 131Google Scholar; Phipps, et al. to RAC, 2 July 1722, C113/274, 240^v.Google Scholar

⁶² Resolution of Elmina Council dd. 8 Apr. 1717, WIC 124.

⁶³ Phipps, et al. to RAC, 2 July 1722, C113/274, 240–241^v See above, pp. 6.Google Scholar

⁶⁴ Ibid.; Valckenier, , Elmina, , to Tinker, , CCC, 5 Feb. 1724, NBKG 91, FC 1715–30, 1724–6, pp. 110Google Scholar. Remember that Kabes had a son named Aq[g]ua who might, in the course of succeeding his father, have come to be known as ‘Ahenaqua’. Elsewhere he was called the ‘descendant’ of Kabes: Elmina, Journal sub 13 Feb. 1723, FC1715–50, 1718–23, pp. 263.Google Scholar

⁶⁵ Husband and Chalmers, , CCC, to Elmina, 24 Nov. 1743, FC1731–57, 1740–6, pp. 188.Google Scholar

⁶⁶ T70/1063 sub 28 Oct. 1793; T70/1069 sub 4 Jan. 1796; various entries in T70/1121, T70/1122, T70/1123; Watts, Martin to CCC, 9 May 1779, UAC International, London.Google Scholar

⁶⁷ A point noted by Daaku, , Trade and politics, 115.Google Scholar

⁶⁸ Robert Elwes to CCC, 24 Oct. 1687, Rawlinson C.747, #2337; same to same, 12 Aug. 1687, Rawlinson C.747, #2342.

⁶⁹ Cross to CCC, 24 Oct. 1686, Rawlinson C.745, #1047.

⁷⁰ Freeman to CCC, 20 Aug. 1698, Rawlinson C.747, #3828. Gore remained with the RAC and eventually rose to be the senior of the three Chief Agents at Cape Coast, dying there in 1717: Phipps, and Bleau, to RAC, 6 Feb. 1717, T70/6, 48, PRO.Google Scholar

⁷¹ Gore, , Phipps, , and Bleau, , CCC, to RAC, 23 Mar. 1715, T7O/6, sGoogle Scholar; Francia, Anthony, CCC, to RAC, 11 Mar. 1715, T70/6, 9Google Scholar. Francia thought this a pretext to replace him with Baillie, ‘a friend of Mr Bleau'’ who held Komenda on his own account. In this he was probably at least partly right, but the Agents' argument, whether or not specious, suggests the attitude of the RAC toward placating Kabes. For a full account of Francia's alleged activities see unsigned [but probably Thomas Bennett] Report dd. 21 Jan. 1716, 0113/294, 20–7, PRO. Another account noted that Kabes had reported Francia's illegal activities to the authorities at Cape Coast: ‘Papers relating to a Portuguese ship run on shore at Accra, [December], 1715’, C113/291, 3. PRO.

⁷² As far as I know, there is no information as to whether the agents sent to Komenda were first evaluated for a capacity to exist in harmony with Kabes; apparently, they were not. The Agents at Cape Coast Castle could, if occasion demanded, serve as a convenient counterpoise to their representative at Komenda who served as a ‘lightning rod’. This occurred several times in addition to the three instances of replacement.

⁷³ Cross to CCC, 30 July 1686, Rawlinson C.745, #957; same to same, 20 Nov. 1686, Rawlinson C.745, #1082; Willson to CCC, 1 Dec. 1694, Rawlinson C.747, #3046; Searle to CCC, 8 Feb. 1695, Rawlinson C.746, #1395.

⁷⁴ To judge, that is, from the comments in Elwes to CCC, 15 Jan. 1688, Rawlinson C.747, #2558.

⁷⁵ William Ronan to CCC, 30 Nov. 1691, Rawlinson C.747, #3788; same to same, 30 Nov. 1691, Rawlinson C.747, #3789.

⁷⁶ Ronan to CCC, 2 Dec. 1691, Rawlinson C.747, #2790. According to Ronan: ‘We must make a country voyage of it’.

⁷⁷ Searle to CCC, 8 Feb. 1695, Rawlinson C.746, #1395. According to Bosnian, , Description, 31Google Scholar, Kabes ‘invited’ the English, but this is not corroborated by the Komenda correspondence. Bosman's chronology is weak here, for he seems to place the ‘invitation’ after an event which is to be dated later in 1695.

⁷⁸ Pearson, Josiah, Anamabu, to RAC, 23 Mar. 1704, T70/5, 9, PROGoogle Scholar. Daaku, , Trade and politics, 118–9Google Scholar, discusses this episode. The Phipps papers, especially C113/280, offer much comment on the perversities of Thomas's behaviour.

⁷⁹ Cross to CCC, 14 Sept. 1686, Rawlinson C.745, #1013. Du Casse ascribed the panyarring to Kabes's friendly treatment of the French in 1685. Roussier, , L'établissement, 16–7.Google Scholar

⁸⁰ Nikolaas Sweerts, Elmina, to CCC, 21 Sept. 1686, Rawlinson C.745, #1017.

⁸¹ Cross to CCC, 25 Sept. 1686, Rawlinson C.745, #1025; same to same, 24 Oct. 1686, Rawlinson C.745, #1047.

⁸² By his own account Bosman was in charge of Fort Vredenburg sometime in 1695 when it was attacked by the West India Company's ‘enemies’: Description, 27–8Google Scholar. This must have been the attack on the Dutch fort (‘they plied them pretty thick until 1 in the morning’) described by Searle in his letter of 2 Mar. 1695, Rawlinson, C.746, #1431. Bosman also described a skirmish with Kabes: Description, 30–1Google Scholar. See also De la Palma to X, 26 June 1702, WIC 98, who wrote that Kabes refused to come to Elmina ‘as a result of his fear of William Bosman who, on the orders of Director-General [Jan] Staphorst [who held office from 1694 to 1696] had the design of shooting him with a pistol’.

⁸³ Freeman to CCC, 26 June 1695, Rawlinson C.746, #1544.

⁸⁴ Cross to CCC, 26 Mar. 1686, Rawlinson C.745, #835. Cf. the account of Du Casse in Roussier, , L'établissement, 16.Google Scholar

⁸⁵ Du Casse in ibid., 23. The context does not suggest that the goods had been entrusted to Kabes during the French reconnoitering expedition in 1686.

⁸⁶ Alagoa, E. J., A history of the Niger delta, (Ibadan, 1972), 138–53Google Scholar; Jones, G. I., The trading states of the Oil Rivers, (London, 1963), 133–4Google Scholar; Latham, A. J. H., Old Calabar, 1600–1891, (Oxford, 1973), 11–3, 36–7, 45–6.Google Scholar

⁸⁷ Alagoa, , ‘Long distance trade and states in the Niger delta’, J. Afr. Hist., xi (1970), 317–29Google Scholar; idem, History, 134, 138–9, 144, 149–50, 152–3, 158–9Google Scholar; Isichei, Elizabeth, The Ibo people and the Europeans, (London, 1973), 32.Google Scholar

⁸⁸ Ogedengbe, K. O., ‘Aboh chronology’, paper presented at the Annual Meeting of the African Studies Association, Philadelphia, 1972, where a date of c. 1680 is suggestedGoogle Scholar; In Onitsha the first historical (i.e., non-mythicized) ruler seems to be Chima in the fourth generation before the oba ruling in 1832: Henderson, Richard N., The king in every man (New Haven, 1972), 447–8.Google Scholar

⁸⁹ For much of Kabes's career and for most of the period of English activity on the Gold Coast before the nineteenth century we must depend on tertiary reports—abstracts made in London of syntheses from the outforts prepared by the Agents at Cape Coast Castle. To appreciate the problem we might compare the three-word abstract (‘John Cabess dead’) in T70/7, 34, with the more than 400-word description of his death and obsequies in C113/274, 240–241^v. This renders it somewhat hazardous to venture statements such as ‘all these [untoward] incidents [between Kabes and the English] occurred between 1714 and 1716 when the tactless Baillie was posted to Komenda’: Daaku, , Trade and politics, 127Google Scholar. We now know differently.