Archaeology and Human Genetics: Lessons for Both
Brown, Keri A., Pluciennik, Mark, Antiquity
The introduction of new scientific techniques into archaeology has often been surrounded by polemics which go beyond disagreements over the validity of methodology (e.g. Renfrew 1973; Thomas 1990). Typically, humanities-based archaeologists may be perceived as Luddite in their anti-science rhetoric, while scientists may be accused of ignorance of cultural and interpretative process. Many of these responses relate to disciplinary boundary maintenance -- the protection of particular spheres of knowledge, status and authority. The recent relationship between archaeology and genetics is a case in point. We examine these issues in relation to current studies of human DNA and later prehistoric and historic archaeologies, rather than the other major focus of much early DNA work, the origins and dispersal of modern humans.
History and background
Genetics first impinged on most archaeologists in the early 1970s, with the work of geneticist Cavalli-Sforza and archaeologist Ammerman on the transition to farming in Europe (Ammerman & Cavalli-Sforza 1971; 1973; 1984). They suggested that the population spread of Neolithic farmers from southwest Asia should still be evident in modern gene frequencies and distributions. Subsequent research by Cavalli-Sforza and others (e.g. Cavalli-Sforza et al. 1994; Sokal et al. 1991) expanded to cover many more areas, samples and genetic loci. Such work has also attempted to elucidate the histories of supposedly distinct populations such as `Celts' or `Indo-Europeans' in conjunction with allegedly linked characteristics including language, ethnicity, subsistence practices and material culture (e.g. Barbujani et al. 1994; Bodmer 1993; Cavalli-Sforza 1996; Moral et al. 1994; Renfrew 1992; Sokal et al. 1993).
Much of this work has been extensively criticized (e.g. Clark 1998; Fix 1996; McEachern 2000; Richards et al. 1996; Sims-Williams 1998a). The causes and directionality of theclines (gene frequency gradients) have been disputed. It now seems clear that Principal Component Analysis maps are palimpsests of gene flows (Renfrew 1998). In general, such modelling offers poor time resolution for events and processes understood in social and cultural terms. Because of the lack of stability of provenance of modern samples, a necessarily fuzzy picture is given. Most of the interpretations were concerned with processes of movement and mixture dating back many millennia. However, they tended to treat populations as stable in terms of linked genetic and cultural features, and the possible relationships of biological data to ethnic labels, artefacts and language have been poorly understood (Lasker & Crews 1996; Moore 1995; Pluciennik 1996a; Sims-Williams 1998b; Zvelebil 1995; 1998). Geneticists often used some sort of `tree' model as though they were dealing with discrete populations and fixed characteristics. Diagrams of difference between genetic samples were often misunderstood as representing `real' phylogenies (cf. Sykes 1999: 132). Elucidating the scale and nature of locally variable biological interactions may require investigation of ancient DNA in conjunction (where available) with historical, textual and archaeological (including bone isotope) evidence.
Many interpretations based on genetic data have thus been at odds with understandings of historical processes by many archaeologists. At present, using contemporary data we are able to model past genetic histories within certain parameters, but these are inevitably constrained by inherent imprecision in detailing the scale, nature and context of the processes of gene flow, and dating the origins of genotypes. Yet this interface between the biological and cultural in specific communities and conditions at particular times is precisely what may be of prime interest. However, such work has also pointed to areas of genetic anomaly or complexity, and can also offer another method of approaching time-depths in distinct genetic entities (lineages) or populations. …