Something Fishy in the Neolithic? A Re-Evaluation of Stable Isotope Analysis of Mesolithic and Neolithic Coastal Populations

Article excerpt

Introduction

The case for a rapid abandonment of marine resources along the coastlines of Northwest Europe at the Mesolithic-Neolithic transition (c. 4000 cal BC) has recently gained momentum with a series of palaeodietary stable-isotope investigations on human bone (Richards & Hedges 1999a, 1999b; Schulting & Richards 2002a, 2002b; Richards 2003; Richards, Price & Koch 2003; Richards, Schulting & Hedges 2003). When Tauber (1981) first announced the results of isotope analysis of human skeletons in Denmark over 20 years ago, his results had little impact on wider archaeological interpretation. The sample size of human skeletons was small, chronologically dispersed, and confined to Denmark. It was possible to argue that the apparently dramatic change in his data, from Mesolithic individuals who are mostly marine protein to Early Neolithic individuals who ate none, would be moderated by larger and more geographically varied samples, not least because most of his Neolithic samples were inland and most of his Mesolithic ones coastal. The more recent work, however, seems to confirm the original pattern with a larger sample of material, not only in Denmark, but more widely in Britain, as well as France (Shulting & Richards 2001) and Portugal (Lubell et al. 1994).

Moreover, the newer generation of results uses stable isotopes of nitrogen as well as carbon. The carbon in marine organisms is obtained from seawater, which is enriched in [sup.13]C relative to [sup.12]C, whereas the heavier isotope of nitrogen, [sup.15]N, is progressively enriched relative to [sup.14]N as organisms occupy higher trophic levels in food chains, whether they are marine or terrestrial. Cod, for example, are secondary carnivores, whereas cattle are herbivores, with corresponding differences in their nitrogen isotope composition. The fact that two independent parameters apparently point in the same direction seems to confirm the reliability of the results.

The shift from Mesolithic diets dominated by marine protein to Neolithic diets in which marine protein is absent appears to be abrupt, widespread and sudden. Some commentators have used the isotope results to suggest that Neolithic peoples did nor simply ignore marine resources in favour of the new agricultural economy, but actively rejected them as part of a new web of food taboos, religious beliefs and myths about the sea (Richards 2003; Thomas 2003). Others have urged caution (Bailey & Milner 2002; Liden et al. 2003; cf. Parkington 1991 and Sealy & van der Merwe 1992), drawing attention to potential biases and untested assumptions in the isotope technique and contradictions with other sources of palaeodietary information.

Our aim in this paper is to take a critical look at the isotope results in relation to other sources of evidence and to reconcile the apparent differences between them, with particular emphasis on coastal sites in Britain and southern Scandinavia. Our starting point is the clear contradiction between archaeological sources of information retrieved from early Neolithic sites, which show substantial evidence for the continuation of marine subsistence activities, and the current interpretations of isotope data, which claim proof of the abandonment of marine subsistence in the Neolithic. We consider three issues:

1 Archaeological evidence for palaeodiet

2 Sample bias in the human skeletons used for isotope analysis

3 The interpretation of stable isotope data

The archaeological evidence

Denmark has one of the richest late Mesolithic and Neolithic archaeological records in coastal western Europe, thanks to isostatic uplift of the ancient shoreline in the north of the country. The Mesolithic Ertebolle culture, 5400-3900 cal BC, is famous for its large number of coastal sites and shell mounds with their abundant evidence of fishing, shell-gathering and sea-mammal hunting (Andersen 1993, 2000). …