Skip to main content Accessibility help
Hostname: page-component-59b7f5684b-2bkkj Total loading time: 0.511 Render date: 2022-09-26T15:54:58.714Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "useRatesEcommerce": false, "displayNetworkTab": true, "displayNetworkMapGraph": false, "useSa": true } hasContentIssue true


Published online by Cambridge University Press:  02 November 2010

CEMAf, France and Syracuse University
CEMAf, France and Syracuse University


The Ghanaian forest was well settled by agricultural communities prior to the opening of the Atlantic trade in the late fifteenth century. The most prominent of these settlements were earthworks sites, construction of which began in the first millennium ce and continued until their abrupt abandonment prior to the mid-fifteenth century. In this article, previous archaeological data are evaluated in light of current research to provide a plausible alternative hypothesis for the history of the Akan, placing that history in a much broader and deeper context.

Research Article
Copyright © Cambridge University Press 2010

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.)


1 Akan is a linguistic classification referring to several closely related languages extending from south-eastern Côte d'Ivoire to the Volta river in eastern Ghana. Although it is a linguistic classification, a high degree of cultural homogeneity and subsistence practices resulting from shared historical processes characterizes groups within the language family. For further discussion, see for example M. E. K. Dakubu, The Languages of Ghana (London, 1988); Kiyaga-Mulindwa, D., ‘The “Akan” problem’, Current Anthropology, 21–4 (1980), 503–6CrossRefGoogle Scholar; E. Schildkrout (ed.), The Golden Stool: Studies of the Asante Center and Periphery (New York, 1987).

2 For review of the early textual sources see P. E. H. Hair, The Founding of Castelo de São Jorge da Mina: An Analysis of the Sources (Madison, 1994); and A. Teixeira da Mota and P. E. H. Hair, East of Mina: Afro-European Relations on the Gold Coast (Madison, 1988). See also J. B. Ballong-Wen-Mewuda, La vie d'un comptoir portugais en Afrique Occidentale (Lisbon, 1993); J. Vogt, Portuguese Rule on the Gold Coast 1469–1682 (Athens, OH, 1979).

3 See Hair, Founding, pp. 55–6, n. 37. See also C. R. DeCorse, ‘Varied pasts: history, oral tradition, and archaeology on the Mina coast’, in J. F. Brooks, C. R. N. DeCorse, and J. Walton (eds.), Small Worlds: Method, Meaning, and Narrative in Microhistory (Santa Fe, 2008), 77–96.

4 G. L. Chouin, Eguafo: un royaume africain ‘au cœur françois’ (1637–1688) (Paris, 1998); K. Y. Daaku, Trade and Politics on the Gold Coast, 1600–1720 (London, 1970); Y. Deffontaine, Guerre et société au royaume de Fetu, Ghana: 1471–1720 (Ibadan, 1993); Feinberg, H. M., ‘Africans and Europeans in West Africa: Elminans and Dutchmen on the Gold Coast during the eighteenth century’, Transactions of the American Philosophical Society, 79: 7 (1989);CrossRefGoogle Scholar J. Fynn, Asante and its Neighbors 1700–1807 (London, 1971); R. Kea, Settlements, Trade, and Polities in the Seventeenth-century Gold Coast (Baltimore, 1982); T. Rivière, Economie et politiques des peuples Ga d'Accra (Ghana) au XVIIe siècle (unpublished PhD thesis, Centre de Recherche Africaine, Université de Paris I, 1994); E. Terray, Une histoire du royaume abron du Gyaman (Paris, 1995); P. Valsecchi, I Signori di Appolonia: poteri e formazione dello stato in Africa occidentale fra XVI e XVIII secolo (Rome, 2002).

5 The archaeological data referred to includes ongoing work by the Central Region Project. Currently, the project is focusing on the survey of archaeological sites between the Pra River basin in the west and the Kakum River in the east. See C. R. DeCorse, An Archaeology of Elmina: Africans and Europeans on the Gold Coast, 1400–1900 (Washington, DC, 2001), 4–6; G. L. Chouin, ‘Forests of power and memory: an archaeology of sacred groves in the Eguafo polity, southern Ghana, 500–1900 A.D.’ (unpublished PhD thesis, Syracuse University, 2009); S. Spiers, ‘The Eguafo kingdom: investigating complexity in southern Ghana’ (unpublished PhD thesis, Syracuse University, 2007); E. Carr, ‘“They were looking for white jobs”: the archaeology of postcolonial capitalist expansion in coastal Ghana’ (unpublished PhD thesis, Syracuse University, 2001); G. L. Chouin, ‘Archaeological perspectives on sacred groves in Ghana’, in M. J. Sheridan and C. Nyamweru (eds.), African Sacred Groves: Ecological Dynamics and Social Change (Oxford, 2008), 178–94; Cook, G. and Spiers, S., ‘Central Region Project: ongoing research on early contact, trade and politics in coastal Ghana, ad 500–2000’, Nyame Akuma, 61:1 (2004), 1728Google Scholar; C. R. DeCorse, G. Cook, R. Horlings, A. Pietruszka, and Spiers, S., ‘Transformation in the era of the Atlantic world: the Central Region Project, coastal Ghana 2007–2008’, Nyame Akuma, 72 (2009), 8593Google Scholar.

6 See I. Wilks, Forests of Gold: Essays on the Akan and the Kingdom of Asante (Athens, OH, 1993).

7 For example see DeCorse, C. R., ‘Coastal Ghana in the first and second millennia ad: change in settlement patterns, subsistence and technology’, Journal des Africanistes, 75 (2005), 4352Google Scholar; P. Shinnie, ‘Early Asante: is Wilks right?’, in J. Hunwick and N. Lawler (eds.), The Cloth of Many Colored Silks: Papers on History and Society Ghanaian and Islamic in Honor of Ivor Wilks (Evanston, IL, 1996).

8 For a review and synthesis, see Chouin, ‘Forests’, ch. 9.

9 Features of this kind, as well as hilltop sites, are an ancient settlement form in West Africa and they have wide distribution, including both Atlantic and pre-Atlantic period sites. See C. R. DeCorse, ‘Fortified towns of the Koinadugu plateau: northern Sierra Leone in the Atlantic world’, in C. Monroe and A. Ogundiran (eds.), Landscapes of Power: Regional Perspectives on West African Polities in the Atlantic Era (forthcoming); G. E. Brooks, Landlords and Strangers: Ecology, Society and Trade in Western Africa, 1000–1630 (Boulder, 1993) 238–9, 246, 250, 255–6; Chouin, ‘Forests’; Chouin, G. L., ‘Sacred groves as historical and archaeological markers in southern Ghana’, Ghana Studies, 5 (2002), 177–96Google Scholar; Chouin, ‘Archaeological perspectives’, 178–94.

10 N. R. Junner, ‘Ancient trenches in the Birim district’, in Report on the Geological Survey Department for the Financial Year 1933–1934 (Accra, 1934), 14. Junner was then director of the Gold Coast Geological Survey. For a review of the initial reports of earthworks and a map of their distribution, see D. Kiyaga-Mulindwa, ‘The earthworks of the Birim valley, southern Ghana’ (unpublished PhD thesis, Johns Hopkins University, 1979), 19–23; see also F. J. Kense, ‘Archaeology in Anglophone West Africa’, in P. Robertshaw (ed.), A History of African Archaeology (Portsmouth, NH, 1990), 135–54.

11 H. J. Braunholtz, ‘Report by Dr. H. J. Braunholtz on pottery etc., excavated from an ancient entrenched site near Abodum, Birim district’, Report on the Geological Survey Department for the Financial Year 1935–1936 (Accra, 1936), 34–6; idem, Archaeology in the Gold Coast’, Antiquity, 10 (1936), 469–74CrossRefGoogle Scholar.

12 Shaw, T., ‘Prehistory and archaeology in the Gold Coast’, in Conférence Internationale des Africanistes de l'Ouest, II (Dakar, 1951), 467–99Google Scholar.

13 O. Davies, ‘Excavations at Kokobin and the entrenchments in the Oda area’, in Archaeology in Ghana (London, 1961), 25. See also idem, West Africa before the Europeans (London, 1967), 287–90.

14 Davies, ‘Excavations’, 25.

15 See ibid. 15.

16 Ibid. Areas were also selected and cleared in the southern and eastern ramparts, as Davies looked for evidence of a palisade. These areas, although marked ‘excavation’ on the site map, seem not to have been excavated. Other stratigraphic information was obtained from two existing surveyors' pits and a cut made for the railway, the latter providing an interesting section of the eastern rampart. See ibid. plates 3 and 4.

17 Kiyaga-Mulindwa, , ‘Earthworks’; see also idem, ‘Social and demographic changes in the Birim valley, southern Ghana, c. 1450 to c. 1800’, Journal of African History, 23 (1982), 6382CrossRefGoogle Scholar. Kiyaga-Mulindwa, a Ugandan, may have been inspired by the research design and approach to the use of oral traditions developed in the framework of previous archaeological research on the earthworks of western Uganda, which had taken place before Idi Amin seized power in 1971. See for instance Robertshaw, P., ‘Archaeological survey, ceramic analysis, and state formation in western Uganda’, African Archaeological Review, 12 (1994), 105–12CrossRefGoogle Scholar; Robertshaw, P. and Taylor, D., ‘Climate change and the rise of political complexity in western Uganda’, Journal of African History, 41 (2000), 45CrossRefGoogle Scholar.

18 Kiyaga-Mulindwa, ‘Earthworks’, 179–89.

19 ibid. 31, 186, 228 n. 1. The collection of oral traditions was included in the project's research design. While some traditions were recorded, Kiyaga-Mulindwa was clearly disappointed by ‘the quality and utility of tradition’ in the Birim valley (ibid. 33). He pointed out a variety of problems, including the predominance of traditions belonging to dominant groups and families, as well as the ‘compression’ of local traditions in a chronology dominated by Asante history. He mainly seems to have overestimated the potential of oral traditions to directly complement archaeological data. Elsewhere, we have noted that the interface between oral historical data and archaeological data can be very narrow: G. L. Chouin and C. R. DeCorse, ‘Trouble with siblings: archaeological and historical interpretations of the West African past’, in T. Falola and C. Jennings (eds.), Sources and Methods in African History: Spoken, Written, Unearthed (Rochester, 2003), 7–15; DeCorse, ‘Varied pasts’.

20 The sites were mainly selected on the basis of their accessibility from Akim Manso. The other sites served to provide comparative data and were used to elaborate the earthwork/Atwea theory of population replacement developed by Kiyaga-Mulindwa: Kiyaga-Mulindwa, ‘Earthworks’, 78.

21 ibid. 95–6.

22 Two new series of dates were obtained by James Boachie-Ansah in 2009 from a test excavation outside and inside the enclosure of a newly documented earthwork at Ngyeduam. Dates obtained outside the enclosure suggest a possible reoccupation of the earthwork in the second half of the fifteenth century, while dates obtained from trenches excavated within the enclosure belong to both the first and second millennia ce. See J. Boachie-Ansah, ‘Who were the builders of the earthworks in the forests of southern Ghana?’, Studies in the African Past, 8 (forthcoming). Future archaeological investigations of the ditch are planned to provide comparative data with that of Akrokrowa (personal communication from James Boachie-Ansah, 2009).

23 Vivian, B. C., ‘Recent excavations of Adansemanso’, Nyame Akuma, 46 (1996), 3742Google Scholar; Shinnie, P., ‘Early Asante and European contacts’, Journal des Africanistes, 75 (2005), 2542Google Scholar.

24 In 1936, Braunholtz pointed to the absence of European imports – tobacco pipes in particular – in the assemblage excavated by Junner (see above) and suggested ‘an antiquity of at least three or four centuries’. Braunholtz further remarked that ‘the weathered condition of many of the sherds, and the silting up of trenches also suggest some age’, before concluding that all evidence pointed ‘to the sites having been deserted by the 16th century, if not earlier, and the accumulation of material suggests a lengthy period of occupation previous to this’. In his study of Kokobin, Davies did not go much further than Braunholtz, even judging ‘hazardous’ the latter's suggestion of ‘an antiquity of at least three or four centuries’ (Davies, ‘Excavations’, 25). He only noted that the earthworks were ‘likely to be roughly contemporary’ and that they belonged ‘to the same cultural stage’ (ibid. 26). Finally, he seems to have accepted Wilks's suggestion that ‘the entrenchments are fortified settlements built by the Akwamu in the second half of the seventeenth century to resist pressure from Akim to the west’. This comment was added to the paper in the form of a nota bene and Davies did not provide any specific comment on Wilks's suggestion. He later revised his position, however, and considered the earthworks as dating to ‘pre-European times’ and as ‘proto-Akan’ (Davies, West Africa, 289–90).

25 Kiyaga-Mulindwa, ‘Social and demographic changes’, 73–7. Bellis criticized Kiyaga-Mulindwa's interpretation, pointing out that the transition in pottery tradition, as noted above, was not a phenomenon limited to the Birim valley but is observable throughout the forest area of Ghana. See J. O. Bellis, ‘A late archaeological horizon in Ghana: proto-Akan or pre-Akan’, in E. Schildkrout (ed.), The Golden Stool: Studies of the Asante Center and Periphery (New York, 1987), 47; DeCorse, Archaeology, 118, Wilks, I., ‘The forest and the Twis’, Transactions of the Historical Society of Ghana, new series, 8 (2004), 50–1Google Scholar.

26 Kiyaga-Mulindwa, ‘Earthworks’, 91, 97, 100–2.

27 The information given by Kiyaga-Mulindwa has been recalibrated using the online version of the program CALIB Rev.5.0.1, written at the Quaternary Isotope Lab of the University of Washington and maintained by Paula and Ron Reimer. The program is based on calibration datasets ratified at the 18th International Radiocarbon Conference. See Reimer, P. J. et al. , ‘IntCal04 terrestrial radiocarbon age calibration, 26 – 0 ka BP’, Radiocarbon, 46 (2004), 1029–58CrossRefGoogle Scholar. We thank Jean Polet for calling our attention to the program.

28 Kiyaga-Mulindwa, ‘Earthworks’, 97. Unfortunately, Kiyaga-Mulindwa did not provide us with much stratigraphic detail regarding the samples on which these radiocarbon dates are based. SI-2717 and SI-2718 for example, were collected ‘from the middle of the ditch’.

29 James Bellis expressed reservations about the sample's context, noting that ‘this date was derived from a carbon sample stratigraphically located on an old surface below the earliest construction of the earthwork. It is apparently associated in no specific way with earlier cultural material (e.g., a clearly defined hearth or rubbish pit containing cultural association)’ (Bellis, ‘Late archaeological horizon’, 45; personal communication 2009).

30 Kiyaga-Mulindwa, ‘Earthworks’, 97.

31 ibid. 93.

32 Sutton, J., ‘Akim Kotoku’, Archaeology in Ghana, 2 (1981), 6Google Scholar.

33 Kiyaga-Mulindwa, ‘Earthworks’, 188.

34 See ibid. 96–7. This stratigraphic section is redrawn in Fig. 1 of this article.

35 Kiyaga-Mulindwa, ‘Earthworks’, 188.

36 Kiyaga-Mulindwa provides no detailed information about the nature of the material culture associated with SI-2716: ibid. 92, 196.

37 ibid. 188.

38 The abandonment of the earthworks is discussed below. For a more detailed discussion on the reoccupation of earthworks in the historic period, see Chouin, ‘Forests’, ch. 9.

39 Chouin, ‘Forests’. Abrem Berase is the capital of the Abrem Traditional Area, located in the Komenda-Edina-Eguafo-Abrem district of the Central Region of Ghana.

40 This sample was recovered from the north-west part of Unit L, embedded in reddish-gray clay characteristic of the culturally sterile subsoil. It is worth noting that cultural material was found in the southern part of the unit to a depth of 260 cm below the surface. Beta-184399 was given in the Report of Radiocarbon Dating Analyses submitted 12 November 2003 by Beta Analytic Inc. Conventional Radiocarbon Age was 6240±40 BP, 13C/12C ratio used was −26·25‰.

41 A probable abandonment in the fourteenth century is based on our interpretation of Beta-184397 at a 1σ calibrated age that offers a more precise dating interval (c. 1304–65 ce). Chouin, ‘Forests’, ch. 9, notes the need for a more detailed study of the stratigraphy of the Akrokrowa earthwork.

42 Wilks, ‘Forest’, 48–52.

43 We note similar discussions concerning the earthworks of western Uganda. See, for instance, Robertshaw, P., ‘The age and function of ancient earthworks of western Uganda’, Uganda Journal, 47 (2001)CrossRefGoogle Scholar.

44 N. R. Junner, ‘Archaeological notes: ancient fortified settlements’, Annual Report of the Geological Survey, 1931–1932 (Accra, 1932), 16.

45 Davies, ‘Excavations’, 15.

46 Davies, West Africa, 287–8.

47 ibid. 289.

48 Wilks, I., ‘The rise of the Akwamu empire, 1650–1710’, Transactions of the Historical Society of Ghana, 3 (1957Google Scholar), quoted in Kiyaga-Mulindwa, ‘Earthworks’, 27.

49 Kiyaga-Mulindwa, ‘Earthworks’, 196.

50 Siddle, D. J., ‘The evolution of rural settlement forms in Sierra Leone circa 1400 to 1968’, Sierra Leone Geographical Journal, 13 (1969), 34Google Scholar; the author quotes several nineteenth-century European sources. On Sierra Leone, see also DeCorse, C. R., ‘An archaeological survey of protohistoric defensive sites in Sierra Leone’, Nyame Akuma, 19 (1980), 1417Google Scholar; C. R. DeCorse, ‘Fortified towns’.

51 See J. Fairhead and M. Leach, Misreading the African Landscape: Society and Ecology in a Forest–Savanna Mosaic (Cambridge, 1996).

52 P. Ozanne, quoted in Kiyaga-Mulindwa, ‘Earthworks’, 27.

53 See for instance Ray A. Kea, ‘Trade, state formation, and warfare on the Gold Coast, 1600–1826’ (unpublished PhD thesis, University of London, 1974), 91.

54 P. Ozanne, ‘Ghana’, in P. Shinnie (ed.), The African Iron Age (London, 1971), 48–9.

55 Sutton, ‘Akim’, 7–8.

56 Wilks, Forests, 95.

57 Shaw, ‘Prehistory’, 496.

58 Davies, ‘Excavations’, 16, Fig. 2.

59 Kiyaga-Mulindwa, ‘Earthworks’, 95–6.

60 J. G. R. Christaller, A Dictionary of the Asante and Fante Language Called Tshi (Chwee, Twi) (Basel, 1881), 266.

61 Pescheux, G., ‘Centre, limite, frontière dans le royaume Asante précolonial’, Journal des Africanistes, 74 (2004), 191Google Scholar, our translation.

62 Polet, J. and Saison, B., ‘Enceintes fortifiées de la Séguié (Côte d'Ivoire)’, Recherche, Pédagogie et Culture, 55 (1981), 55Google Scholar.

63 Connah, G., ‘Archaeological research in Benin City, 1961–1964’, Journal of the Historical Society of Nigeria, 2 (1963), 473Google Scholar; Norman, N. and Kelly, K., ‘Landscape politics: the serpent ditch and the rainbow in West Africa’, American Anthropologist, 106 (2004), 102CrossRefGoogle Scholar.

64 The Plague was probably caused by the gram-negative coccobacillus, Yersinia pestis. Y. pestis (bubonic plague) is transmitted to humans through flea bites. It affects the lymphatic system, engendering a strong immune reaction that causes extensive inflammation, resulting in swelling and internal bleeding. The Plague was referred to as the ‘Black Death’ because accumulated blood under the skin and in the extremities of the body dries and turns black. Victims die from septicemia or organ failure caused by the drop in blood pressure and the lack of circulating oxygen. The disease can also establish itself in the lungs, becoming highly contagious in the form of pneumonic plague. In pre-antibiotic times, pneumonic plague killed in a maximum of three days. For more information, see for instance (last visited 31 July 2010).

65 See, for example, S. Cohn, The Black Death Transformed: Disease and Culture in Early Renaissance Europe (London, 2002); F. Cartwright, Disease and History (New York, 1991).

66 For example, see J. Iliffe, Africans: The History of a Continent (Cambridge, 1995), 47–8, 67.

67 ibid. 47.

68 M. Posnansky, ‘Prelude to Akan civilization’, in E. Schildkrout (ed.), The Golden Stool: Studies of the Asante Center and Periphery (New York, 1987), 17. Cartwright notes that the Black Death spread ‘southward into Africa, eastward into China, and northward to Russia and the Scandinavian countries’, describing it as a ‘world-wide pandemic’ (Cartwright, Disease, 32).

69 Chouin initially provided a brief review of possible supportive evidence (‘Forests’, ch. 9), but there is a need for a complete re-evaluation of the archaeological dataset in Africa. In West Africa, Jenne Jeno and the associated sites in the inland Niger delta were abandoned prior to the beginning of the fifteenth century (see R. J. McIntosh, The Peoples of the Middle Niger (Oxford, 1988), 203). Settlement occupations in Sierra Leone may show a similar break in occupation (DeCorse, ‘Fortified towns’). With regard to other parts of Africa, it is perhaps worthy of note that Zimbabwe complex sites were also abandoned before the mid-fifteenth century (I. Pikirayi, The Zimbabwe Culture: Origins and Decline of Southern Zambezian States (New York, 2001), 150–1.

70 Entrenchment sites have also been found in south-eastern Côte d'Ivoire. A cluster of more than a dozen earthworks, similar to those in the Birim valley, were discovered at La Séguié in the late 1960s during forest clearance. In 1969 and 1971, Jean Polet conducted excavations at Enceinte no 2. In the absence of radiocarbon dates and data on the age of the entrenchments, he has recently suggested that these sites may need reassessment (J. Polet, personal communication, 2009).

71 While there is no question that the European capitalist economy and its associated technologies had dramatic consequences, these were nonetheless articulated within the framework of local responses and cultural traditions. For example, see DeCorse, Archaeology, 175–92. See also A. B. Stahl, Making History in Banda: Anthropological Visions of Africa's Past (New York, 2001), 12–15.

Cited by

Save article to Kindle

To save this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Save article to Dropbox

To save this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about saving content to Dropbox.

Available formats

Save article to Google Drive

To save this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you used this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about saving content to Google Drive.

Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *