Monographs presenting detailed regional overviews and including a wide variety of site types have become a rare commodity in archaeological literature. The Middle Stone Age (MSA) of Nigeria in its West African context is the exception to this trend. Primarily based on the author's own research conducted in Nigeria in the 1970s, the volume also offers an overview of Middle Stone Age and Sangoan sites in West Africa in a “country-by-country fashion” (p. 1). The volume comprises 317 images, including figures, photographs recording the condition of now inaccessible sites, illustrations of profiles and stratigraphy as well as drawings of lithic artefacts and 27 tables, providing a detailed historical record for the sites discussed. Allsworth-Jones's major concern is to ensure that the debate on the Middle Stone Age, which is crucial to the deep history of modern humans, takes into account the archaeological record from West Africa. His aim, therefore, is to make these data available to the scientific community working elsewhere in Africa, where often the West African data are so little known “that it is sometimes assumed that there is no such sequence” (p. xiii) in that region. The author holds himself partly responsible for that situation, and the volume is intended to compensate for his dearth of published results or for having published results in relatively inaccessible journals. This I do not consider a fundamental flaw, however; local journals are often the first point of reference for national scholars.
The organisation of the book is geographic, rather than chronological, technological or thematic; this is because, as the author correctly notes, colonisation has impacted the diverging research traditions—including typology and terminology—as well as the geographic and political boundaries. Table 1 shows the correlation between past and present terminology, providing an overview of geological stages expressed as pluvials or interpluvials, alongside their equivalent stages as they appear in older publications and reports. The relatively poor absolute dating evidence in West Africa hampers correlation of any other data from the archaeological sites.
Allsworth-Jones defines Middle Stone Age sites by the presence of Levallois technology. The Sangoan has long been, and indeed remains today, a floating concept without a chronological anchor. As Allsworth-Jones notes, “in West Africa, assemblages called Sangoan have been defined by different authors in different ways” (p. 232). For this reason, Allsworth-Jones calls for a new assessment of the lithics, by a single team to provide a standardised typology and agreed terminology for this material; he demonstrates how this might work by applying the method to the Palaeolithic sites that he discovered in Nigeria. A more descriptive approach is employed in Chapter 2 to catalogue the artefacts that he collected in the mining area north of the Jos Plateau, from the stone debris dumps and those discovered in situ in sections and in excavated layers.
In Chapter 4, his regional overview, the author points out that on the rare occasions that assemblages with a Levallois component are found in stratigraphic contexts in West Africa, they are stratigraphically above Sangoan or Acheulean and below Aterian or Late Stone Age. For Allsworth-Jones, this is a clear indication of the reality of the Middle Stone Age. West Africa is “not lacking in characteristic assemblages”, but clarifying their meaning “depends upon finding sites in acceptable stratigraphic context” (p. 224), and more particularly in contexts that allow for absolute dating. The large distances between the sites hamper their stratigraphic correlation, making assessment of the variability, idiosyncrasy or similarity of environmental settings or geological periods very challenging. Fortunately, over the last two decades, more OSL dates have become available, although they are still limited to relatively few sites such as Birimi in Ghana (±41 and ±24 ka), Ounjougou in Mali (spanning from 70–25 ka), and the Falémé Valley in Senegal (extending to ±63 ka). These OSL dates situate Middle Stone Age assemblages in a time frame that began at least 70 ka ago, in Marine Isotope Stage (MIS) 3. The radiocarbon date for Bilma in Niger is associated with a Levallois-dominated industry and fits into MIS 3. To these examples cited in the book, we may add the open-air site of Njuinye in southern Cameroon, where a date of 34.7 ka comes from the lower layers containing quartz stone industries with a possible Middle Stone Age affiliation, according to the excavators Mercader and Martí (Reference Mercader, Martí and Mercader2003). In contrast to the MIS 3 chronological timespan are the Mid Holocene radiocarbon dates from sites such as Zenabi, Banke and Nok, all on the Jos Plateau in Nigeria. Allsworth-Jones suggests that these aberrant dates may best be explained by post-depositional disturbance. Yet these sites are not far from the Shum Laka rockshelter in Cameroon where blades and triangular points, reminiscent of Levallois, and large bifacially flaked tools are found in levels dated to around 7000 BP (e.g. fig. 5 in Lavachery Reference Lavachery2001). I agree with Casey (Reference Casey and Mercader2003) that this is far from conclusive evidence of a Holocene date for Levallois-influenced materials elsewhere in West Africa, but also that the presence of Levallois technology may not, in itself, be an indicator of the Middle Stone Age. This intriguing dating evidence should be an incentive for more intensive research on the Middle Stone Age in West Africa.
Allsworth-Jones's detailed archival records and geographic overviews provide an excellent account of the peculiarities of industries with a Levallois component, and the volume is a highly recommended point of departure for identifying areas suitable for in-depth research on this crucial period in the deep history of modern humans in this part of Africa.
