IS. PROJECT REPORT RECENT ARCHAEOLOGICAL RESEARCH IN THE VOLTA BASIN VOLTA BASIN RESEARCH The construction of the Akosombo Dam has necessitated a crash- I brought the matter in March 1960 to the attention of programme of rescue-work in the very large area to be flooded by the Volta Lake. Mr. H.D. Col I ings, then Director of the Ghana Museum; but he seems to have failed to bring it to the Board, with the result that nothing was done until early 1963, when Professor Irwin, on behalf of the Historical Society of Ghana, launched the Volta Basin Research Project. After many months of wearisom estimating, discussion with the National Council for Higher Education, correspondence and organisation within the University, the Ghana Government made a generous grant in February 1964. Archaeology was one of the disciplines approved for special support, because what we should fail to find now we should never find. Unidersity-departments had indeed already started work the previous July, in confidence that support would be forthcoming; and Professor Irwln himself had already left the country before he saw the project which he had initiated going ahead. The Volta Basin Research Project comprises two classes of work: research arising from the creation of the lake and likely to continue indefinitely especially in biology; and salvage-research which must be done before the lake is f u l l. In the latter class archaeology has held by far the largest place, and there are now two research-fellows in the field, especially appointed for this work, M r. R.N. York and Mr. D. Mathewson. Their before they are flooded. task is to excavate all important sites The first stage, however, was to carry out as complete a survey as possible of the area to be flooded. A good many sites had been previously known; but I have been able to add greatly to this list. This survey has not been completed. Lack of time and inaccessibility made it impossible to reach many areas on the lower Volta and in the Afram Plains, even though the Ghana Army conducted me for,a month by boat to regions which I could not have reached of my own accord. Work was delayed also by the floods of 1963. The Afram Plains have been particularly difficult of acess. Riverside villages could be reached only by boat, and navigation in the dry season was extremely dangerous. Ther were no paths, so except when the grass was burnt, one could neither walk inland nor see what might be there. It. PROJECT REPORT 19, would have required ten years to do a proper survey, wok ing for short periods each year, of the right bank of the Volta from the Pra to the Afram. There is however reason for thinking that the inaccessible sour them part of the Volta Basin was anciently sparsely inhabited, and that we have not there lost many sites. Other parts of the valley are more accessible, provided one is prepared to use one's two feet for locomotion and not to rely entirely on four wheels. A large number of sites has been listed, from small hamlets perhaps not more than a century old to large towns of many ages and to find-spots of palaeolithic tools. Doubtless others will still be found. I cannot at the moment give totals; but I can give numbers in sample-areas, which may illustrate the scope of the project and the wealth of remains that we are recording and saving. In the floodable area on the left bank of the Volta from Kete Krachi to the mouth of the Daka, a distance of about forty miles, we have found thirty-nine sites. In the Oti valley between Bladjai and Wasa- wasa, on the right bank, over about thirty miles, we have recorded seventeen sites. Near the head of the Black Volta Lake, from Buipe Bridge to Kadelso, on both banks for about eight miles we have found mounds above the flood-line; and this number will certainly be increased,as there are sites close to Morno, which have not yet been recorded. • Up to the present detailed work has been confined to two classes of site, remains of the old stone-age and towns and villages of the proto historical period, a few going back into the middle ages. On the whole, medieval sites are difficult to locate. Mounds, if there were any, have been levelled, and the habitation layer is usually buried under a foot or more of soil. Lone baobabs, so reliable an indicator of sites from the sixteenth century onwards, hardly last in West Africa from medieval times, though in the Zambesi valley trees are known which are over a thousand years old. One or two neolithic sites are on the list for excavation; but being far upstream they have had to wait their turn. Village-sites of the XVI-XIX centuries are common. There was almost ribbon-development along the east bank below Kete Krachi. Their economy must have depended partly on fishing; at the same time, the Volta was unsafe from slave-raiders, and nearly every site lies behind the bank concealed from the river by a strip of gallery-forest, and close to a creek where canoes could be hidden. The pottery is local and fairly uniform; occasional painted sherds from Gonja and European imports from the coast indicate limited contact with the outside world. PROJECT REPORT 20.- All these sites seem to be earlier than the Kwahu push across the Volta, which probably dates from the XIX century. Above Otisu Mr. York set out to discover Kwahu terracottas, some of which were already known; he found an older village, but not a single terracotta in i t. Near Ohia Mankyene is a large village, from its pottery Kwahu or Ashanti; but it has no older pottery of the riverside-peoples, who were presumably proto- Krachis. The O ti also was dangerous, though it is more difficult to navigate, and its meanders would enable defenders on land to slip across from one ambush to another, while the attackers toiled round loop after loop of the river. A huge fortress was constructed at Kitari; Kajeso, farther upstream, is built on a precipice with a steep hollow way to the river, which could easily be defended. The archaeologist has neither the time nor the competence to collect oral traditions about his sites. He also rather despises material finds. But if traditions are to be collected about these towns and villages it must be done urgently. From below Kete Krachi the people have been herded into large resettlement-villages. It is difficult to imagine less promising ground for the preservation of tradition, which always has a strongly local flavour. The other work which archaeologists have hitherto been doing in the Volta Basin is more closely allied to geology. The study of stone tools in their geological settings enables us to guess the antiquity of man, and examination of old muds and gravels provides a reconstruction of the ecological setting. Samples have been taken for pollen-analysis and for radiocarbon-dating. It must be remembered that Ghana has a series of palaeolithic sites far and away superior to any other territory of West Afri ca; and half of these lie east of the Volta, many in the area to be flooded. Ghana's wealth of material as compared with West Africa from Senegal to Dahomey; a good many further sites, which have not yet been published; but even with this addition, Ghana would be far ahead of all her neighbours together. Of the Cheliean Ghana has 45 - 55 sites, the rest of West Africa hardly three; of the Sangoan Ghana has about 280, the rest 75, of which two thirds are in Togo and Dahomey. I give figures for a few palaeolithic cultures, to demonstrate I do not include Nigeria, because recent work there has revealed The real value of this opportunity of intensive excavation in the Volta valley has been the discovery of a very large series of remains which will form a standard for work in other parts of Ghana. The Volta- valley sites may not always have been the largest or politically the most PROJECT REPORT 21. important; there were other tribes in other parts of the country. But in future reference will have to be made to the Volta valley in any work that is carried out. It must be remembered that other artificial lakes are being planned in Ghana. Let us hope that their areas will be explored as intensively as that of Lake Volta. O. Davies. See below for a list of Oral tradition collected by Research Assistants of rhe Institute- Oral Tradition of villages in the Volta Flood Area recorded by (1) K. Ameyaw, and (11) E.Y. Aduamah Research Assts. in the Institute cc. II No. AM II it II II II II II II II It II II II II II II II II II II II II II II it II II •i n it it II it ti it M it it II it n II II II ii .. n II ii II it II it II it .. II n H II it II 91 II 1 . 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Tradition of Nketepa n 11 11 II 11 .1 " II •• . gg II II II It II II II II II II Worobong Mankrong; Asaresu, Yereka Asuboni Ac hi (Ekye) Tease Pitiko Shewohoden, Nsunua, etc. Kotoso Aframso, Obra, etc. Odotom (Begoro area) Chebi Samankwaye, etc. Odotom (Kwahu) Afram Plains, (OkwOtou section) A jade Akroso Apai-Ahenkro Apaso Ahenkro Yamboso Otiso Dukoman Ahenkro/Supom Motodua Asabi PROJECT REPORT 22 COLLECTION OF ORAL TRADITION FROM THE VOLTA BASIN, MAY 1963 - APRIL The field work was carried out in four stages; viz: STAGE 1. The eight Ewe traditional areas on the east side of the Volta affected by the flooding of the Volta namely Kpa hne, Tonkor, Botoku, Tsoho, Wusuta, Awate, Averne, Tsome and villages thereunder each of them. (See the attached Map One). STAGE 2 Ewe settlements in the Afram Plains between the """"""" Afram rivor in the south and the Nyuinyui stream in the north (see the attached Map Two). STAGE 3 The migrant Ewe fishermen or the Tongus as they are called who occupy both banks of the river from Ajena area northwards to a short distance beyond Kete-Krachi. STAGE 4 The Akan-speaking areas of Buem-Akan, Dodi, and Anum. (For the first mentioned area see Map 3). The collections are catalogued as follows:- N o .l No..2 N o. 3 No.. 4 N o. 5 No.6 N o. 7 N o. 8 No.9 Kpalime Todome, Kpalime Kaira. Tonkor Botoku Tsoho Wusuta Awate and Jifaji Aveme Tsome Ewes in the Afram Plains comprising settlements of Akangla, Atikplale, Awatokplui, Alehere Chemfe, Nframa, Burupai, Fanu, Nyinapong, Anidze, Dzanikata, Adzeyife, Sodzife, Gadibafe, Oprida, Amuvinu, Nyigbenu and Dadiase. N o. 10 Dodi-Anum area consisting of Labolabo and Dod? PROJECT REPORT No. 11 No. 12 23, Buem-Akan Area consisting of Nkonya, Bowiri, Kwamekrom, Tapa, Sakada, Krakwamekofe, Mangoase, Takroso, Breniase, Odei, Akanie, Zongo Dawuda, and Okomfokrom. The migrant Tongu fishermen consisting of Bejamsotornu, Oborborkro, Nkatikwatornu, Nsunua, Apretsi, Supon, Aboi, Santase, Odiewu, Dodife, Katsra, Akrokai, Taglase, 3uira Kawnoso, Akroso Kpome, Fahiakobo, Akukordzie, Akawakpe, Kodofe, Kpebe, Sakachire, Atixenu, Mangoase, Didim, Ampote, Dayinu Doguikofe, Agodeka, Apoesihu, Dzemeni, Dawukofe, Adoma Apasu, Fiewukofe, Sefenukofe, Miawone, Asikoko, Mem, Batanya and Abehediem. RESEARCH IN NORTHERN GHANAIAN LANGUAGES Language research is proceeding along two main lines: (a) Language analysis and description; (b) Comparative studies. a. Language analysis: The Northern languages at present being principally studied In these, analysis is being undertaken largely from are: Dagbani, Kasem, Dagaari, Kusal, Sisala, Vagala, Konkomba, Basal, and Bimoba. recorded texts, with a view to describing their structure, which in most cases has not been studied hitherto; even in Dagbani there is a notable lack of reliable description. An article on 'The phonology of the nominal in Dagbani1 has appeared as pre-print from the Institute's new Bulletin, and phonological studies of Kasem and Sisala have appeared as the first two volumes of the 'Collected Language Notes', published by the Institute.