Genes, Peoples, and Languages

  • Luigi Luca Cavalli-Sforza
Farrar, Straus & Giroux: 2000. 216 pp. $24
Credit: DAVID NEWTON

In the early twentieth century, a uniquely American version of anthropology emerged under the direction of prominent scholars such as Franz Boas, Alfred Kroeber and Edward Sapir. What set this school apart from its European counterparts was its efforts to construct a scientific approach to deep-time human history, and the catholic inclusion of evidence from ethnography, archaeology, linguistics and human biology (then called physical anthropology). Today, tensions within anthropology threaten to explode this model of a ‘holistic’ approach to the study of human cultures. Luigi Cavalli-Sforza, a geneticist whose research has depended on close collaboration with archaeologists and historical linguists, gives us ample reason to celebrate anew this holistic vision of anthropology.

For several decades, Cavalli-Sforza and his colleagues have pursued a variety of research projects involving not only the mapping of human genetic variation, but also the correlation of geographic patterns of variation with patterns of language distribution, and with archaeological evidence for ancient population movements. The underlying theme integrating these disparate studies traces the biological and cultural evolution of modern Homo sapiens throughout the past 100,000 years.

Over the course of Cavalli-Sforza's career, the methods for studying genetic variation have themselves been radically transformed, from the relatively crude mapping of ABO blood groups, to the sequencing of DNA, and even the recovery and sequencing of ancient DNA itself. Genes, Peoples, and Languages tells the story of this research odyssey in a style that is highly accessible to intelligent readers from any background, without presuming prior grounding in genetics, statistics or, for that matter, linguistics and archaeology.

In his effort to make his research methods transparent to those without a background in human genetics, however, Cavalli-Sforza ends up devoting much of the first half of his book to a kind of primer in genetics, evolutionary trees and multivariate statistics. This may deter some readers, for the fascinating problems of deep-time human history do not begin to emerge until the fourth chapter. But for those who persist — or simply skip ahead — the intellectual rewards are ample.

In the chapter “Technological revolutions and gene geography”, Cavalli-Sforza tackles one of the great problems of historical anthropology: the expansion of modern humans out of Africa, and subsequent diasporas that correlated with major periods of demographic growth. Here he draws on his work with archaeologist Albert Ammerman in demonstrating a major population-expansion-driven (as opposed to strictly cultural) diffusion from the Near East into Europe, associated with the spread of agricultural peoples. But while the genetic landscape of modern Europe still bears the signature of this major diaspora, other ‘streams of gene diffusion’ are also evident in statistical patterns of gene distribution, including an expansion from the region north of the Black Sea. This appears to have been associated with the expansion of pastoral nomads who originally domesticated the horse.

Although he dwells at greatest length on Europe, Cavalli-Sforza also discusses evidence for several other major diasporas, including the expansion of Bantu language-speaking groups in sub-Saharan Africa, and the remarkable dispersal of the Austronesian-language speakers. This latter dispersal would ultimately result in the human colonization of the most remote Pacific islands, including Easter Island and Hawaii.

There are, of course, always pitfalls in multidisciplinary research waiting to entrap unwary scientists stepping beyond the familiar terrain of their own discipline. Writing on language, Cavalli-Sforza is liable to raise some hackles, for not all of his claims fall within what is currently accepted by mainstream historical linguists or prehistorians. This derives, in part, from Cavalli-Sforza's desire to have a ‘big picture’ classification, or family tree, of world languages, which can then be correlated with his genetic tree of human populations. Unfortunately, the classic methods of historical linguists have serious limitations once one attempts to move back in time to a level deeper than that of major language families. Cavalli-Sforza thus relies heavily on the work of linguists Joseph Greenberg and Merritt Ruhlen, especially the latter's highly controversial ‘super-family’ classification.

For example, in the key diagram comparing genetic and linguistic trees, Cavalli-Sforza identifies a ‘Melanesian’ population, and groups it genetically most closely with ‘Polynesians’ and ‘Micronesians’, and more distantly with a ‘New Guinean’ population. Linguistically, the Melanesians and New Guineans are depicted as linked together in an ‘Indo-Pacific’ family, while the Polynesians and Micronesians are identified as part of the large ‘Austronesian’ language family.

However, the very concept of a discrete ‘Melanesian’ population was rejected many years ago, and the notion of an ‘Indo-Pacific’-language family was similarly abandoned. Rather, the people who occupy the islands of Melanesia (excluding New Guinea) display tremendous genetic variation. This is the result of a specific and complex history of human migrations, including an early phase of hunter–gatherer expansion around 40,000 years ago, and a later phase of population intrusion by Austronesian speakers about 3,500 years ago — reflected archaeologically in the Lapita cultural complex. This latter intrusion resulted in virtually all island Melanesian populations speaking Oceanic languages (a branch of Austronesian).

Other regional specialists will doubtless find similar problems with Cavalli-Sforza's efforts to correlate genes and languages at such a broad level. But this does not mean that the effort to construct such worldwide correlations is futile, for there is clearly much historical truth in Cavalli-Sforza's models. As a first approximation, the history of human diasporas painted in this book shows a vast improvement in our knowledge, which has been made possible by rapid advances in method. And not just in biological science, for the historical linguists themselves are now tackling the thorny issues of deep-time language reconstruction more intensively than ever before. My point is simply that Cavalli-Sforza's models will need more work, and much refinement, before they will satisfy all the critics.

In his final chapter Cavalli-Sforza ranges widely over some of the most profound questions of human evolution, including the role of culture as a means of biological evolution, the modes by which culture is transmitted, and the emergence of complex hierarchical social structures. The ideas raised here reinforce how essential it is for social scientists and biologists not to be isolated — as they so often are in the intellectual structures of modern academia — and to realize how much they need each others' insights.

Edward Sapir eloquently laid out a multidisciplinary methodology for historical reconstruction in his now little-read 1916 monograph Time Perspective in Aboriginal American Culture: A Study in Method (Canada Department of Mines). At the time, the most sophisticated discipline contributing to the tracing of deep-time human history was linguistics. Sapir himself lamented that he had probably “undervalued the data of physical anthropology”. Little could he have anticipated the advances that would come with the revolution in molecular biology and genetics.

Cavalli-Sforza and his colleagues have now given historical anthropology sophisticated tools to look at human variation, and to read that variation as a ‘text’ of human history. Cavalli-Sforza writes that “to some, history is not a science”, but he has helped to show that history is susceptible both to the methods of science and to the power of a multidisciplinary approach.