Problems in the interpretation of radiocarbon dates: the Pastoral Neolithic of East Africa

For the Pastoral Neolithic of East Africa radiocarbon dates suggest two apparent anomalies in the archaeological record: pastoralism in the Central Rift at perhaps 7000 bp, and the very long duration of particular pottery traditions. This paper examines the dating evidence closely, in particular the assertion that apatite gives more reliable dates than collagen. Both the assertion and the anomalies appear to be false. It is shown that there is no firm evidence for pastoralism in East Africa before about 4000 bp. Similarly, the Narosura tradition is of shorter duration and more recent than previously supposed. The meagre dating evidence for the Namoratunga sites and its implications are also assessed. Examination of the dates for the Pastoral Iron Age shows that the Turkwel tradition is dated only by a single date and that the appearance of twisted cord rouletted pottery in the present millennium post-dates the end of stone tool manufacture by several centuries. The paper concludes with a plea for more critical assessment of radiocarbon dates and for greater rigour in their presentation.