Professor of African and Islamic Archaeology

‘The Archaeology of Ritual, Medicine, Shrines, and Sacrifice among the Talensi of Northern Ghana’. 2004-2011.

Click images to read the captions.

The Tong (Tengzug) Hills are a small chain of hills in northern Ghana (c.5km x 3km) occupied by the Talensi ethno-linguistic group. Survey recorded 79 sites in the Tong Hills ranging from extant shrines to grinding hollows and tree refuges, to archaeological site complexes in rock shelters. Twenty-four excavation units were also completed. These have allowed the reconstruction of settlement chronology and through the material present, primarily ceramics, provided an insight into regional contacts (see Ghana Talensi 6).

No trace of the Middle Stone Age (MSA) was found in the Tong Hills. The Late Stone Age occupation of the Hills is probably represented in the deposits of the rock shelter that is now the Tongnaab Yaane shrine, as indicated by a single OSL date obtained from an excavation adjacent to the shrine (2500+/-235 (726-256 BC). It is very unlikely to have been a shrine at this time. In contrast, the OSL date of 440+/-55 (AD 1514-1624) obtained from another excavation in the shrine surroundings is related to the use of Tongnaab Yaane as a shrine.

The Early Iron Age is represented by two OSL dates from the base of the deposits in a rock shelter, Hyena Cave (e.g. 1465+/-180 [AD 364-724]), and two dates from the Nyoo shrine (see Ghana Talensi 1). This shrine, survey indicated, was far from being a ‘natural’ sacred grove but instead was an enshrined archaeological site comprising an extensive area of stone arrangements infilled with smashed and deliberately deposited ceramics, an active sacrificial area, and a zone covered with 142 standing stones, both erect and recumbent (see Publications). Excavations below the standing stones indicated that they had served a ‘ritual’ purpose, with repeat deposition of materials such as iron bracelets and complete ‘capped’ pots interred below each standing stone (see Ghana Talensi 1 and Ghana Talensi 9). These share resemblances but not exact parallels with contemporary Talensi ritual deposition practices in relation to shrines (see Ghana Talensi 1). Similarly, precise parallels for the excavated stone arrangements and pottery spreads do not exist. The stone arrangements bear comparison with contemporary stone settings used as seating places by Talensi elders during the pre-farming season Golib or Gologo festival (see Publications and Ghana Talensi 4), but pottery is not deposited. It is possible that the pot spreads were associated with feasting, or represent, perhaps, the ritual return of pots to the medium from which the clay they were made with was sourced, the Earth.

The Later Iron Age (post AD 1000) is represented by dates from the Touwang and Zandoya sites (e.g. 680+/-80 [AD 1248-1408]), and the Nyoo shrine, and based upon the archaeological chronology it can be suggested that the expansion of settlement in the period c. AD 1100-1500 perhaps represents an increased need for security, or an overall increase in population. Hence the Hills and their immediately surrounding area were made a focus of occupation. Prior to this the Hills were seemingly less intensively occupied. Nyoo, however, might have been a ritual centre, visited by surrounding populations for ritual gatherings and activities, for a considerable period prior to the more intensive occupation of the Hills.

More recent occupation, excluding the material from Tongnaab Yaane, is largely unrepresented in the excavated sites except at Hyena Cave. The absence of tobacco pipes, except at these sites, provides a chronological marker as tobacco smoking in West Africa, for which these pipes were presumably used, does not predate AD 1492 and European discovery of the Americas. A link with the British ‘pacification’ of the Hills in 1911 was also recovered in the form of a .303 rifle bullet that had been fired into a stone arrangement.

Other excavations were completed adjacent to the entrance of the Tongnaab Yaane shrine. One unit close to the area where ‘witches’ possessions and clothing are discarded prior or subsequent to their being cured of witchcraft provided an OSL date of AD 1514-1624 (see Ghana Talensi 7). This date is conceivably linked with the operation of the shrine and perhaps begins to historically contextualise its operation for franchising purposes, that are today achieved via the agency of boarchii and boarbii (see Ghana Talensi 7).

Traded items such as glass beads and cowry shells were very rare in the excavations. However, ceramics offer a tangible source of evidence for assessing potential contacts. Parallels between aspects of the ceramic assemblage from the Tong Hills and those from elsewhere in Northern Ghana include, to the far west, parallels in rim form and decoration with Birifoh-Sila Yiri, to the west, similarities in vessel form and decoration with ceramics from Yikpabongo in Koma Land, and likewise similarities in rim form and decoration to the southwest to Daboya on the White Volta River (see Ghana Talensi 9). Considering parallels in other directions is hampered by an absence of research and/or publication. Ritual links and shrine connections also have implications for the movement of goods and people prior to European contact (as they do today) as pilgrims to the shrines in the Tong Hills brought with them gifts such as salt, cloth, and iron hoe blades.

Botanical and geological surveys of the Tongo Hills were also completed. The botanical surveys in shrines indicated that plants and trees useful to humans predominated even within so-called ‘natural’ sacred groves (see Ghana Talensi 3). Extensive inventories of medicinal plants were undertaken (see Ghana Talensi 8), and thirty one plant based medicines recorded along with twelve non-plant based medicinal substances. Samples of medicine clays from shrines were analysed by XRD and GC-MS. These indicated that the clays were natural and had not been altered by human action, suggesting that the perceived ‘power’ of the medicine was derived from its shrine association rather than pharmacological properties. Potsherds from a medicine equipment disposal context were also analysed using GC-MS and GC-C-IRMS, and compared with a modern analogue. This indicated that the modern example had been used to prepare different substances, possibly through a prior life to it being used as a medicine pot, in comparison to the archaeological potsherds (see Publications). Sacrifice practices and distribution patterns were also recorded (see Ghana Talensi 2), to assess their archaeological and medicinal implications, and comparative studies undertaken (see Togo – Lome Fetish Market). The geological survey attested to the movement of rock in the landscape, albeit at a local scale, but contributing to the importation of, for example, standing stones in the Nyoo shrine (see Ghana Talensi 5).

The research was made possible through the kindness and co-operation of the Talensi themselves, and especially through the efforts of the Chief and Golibdaana, Honourable John Bawa Zuure. Permission was negotiated with the Talensi communities, and via sacrifice with the shrines themselves. The research was completed in co-operation with the Ghana Museums and Monuments Board, Dr Ben Kankpeyeng of the University of Ghana, Legon, and Dr Rachel MacLean, also of the University of Manchester. The fieldwork and post-excavation analysis was funded by the British Academy (2004-2008), the University of Manchester (2006), the British Institute in Eastern Africa (2005), and from 2008 by the Wellcome Trust.