A very ancient Y chromosome mutation has been found that occurs in Siberia as well as Europe. This mutation is boldly linked with the bearers of the Aurignacian culture who entered Europe 40,000 years ago. The culture appears in Siberia at about the same time, as if these early people had spread both east and west.
Dr. Underhill and his colleagues associate another mutation, which is common in India, Pakistan and Central Asia as well as Europe, with the people of the Kurgan culture who, according to one theory, expanded from southern Ukraine and spread the Indo-European languages.
Dr. Underhill's report tries to make the grand synthesis between archaeological and genetic data, but it will probably be some time before the specialists in each area agree on how the two types of data should be associated.
"It is very exciting that the geneticists now have internal dating procedures, but really I think the dates are very loose indeed," Dr. Renfrew said in an interview.
Geneticists believe that the world outside Africa was populated by the migration of a very small number of people who left east Africa about 50,000 years ago. These modern humans, with their more advanced and inventive culture, are thought to have displaced the archaic hominids like the Neanderthals, which had emigrated from Africa many thousands of years earlier.
These Paleolithic populations created sophisticated stone tools and left evidence of their advanced culture in the cave paintings of southern France, dating to at least 30,000 years ago. Although anatomically modern humans first appear in Africa about 150,000 years ago, their archaeological remains show little sign of modern human behavior.
Dr. Richard Klein, an archaeologist at Stanford University, has suggested that some genetic change, perhaps as profound as the invention of language, occurred in Africa around 50,000 years ago, and that it was these behaviorally modern humans who both spread within Africa and populated the rest of the globe.
This thesis was challenged at the Cold Spring Harbor conference by two archaeologists, Dr. Sally McBrearty of the University of Connecticut and Dr. Alison Brooks of George Washington University. They argued that each of the components said to characterize the Paleolithic revolution in human behavior, like stone blades, long distance trade and art, can be found in Africa at earlier dates.
"So all the behaviors of the Upper Paleolithic have an African pedigree," Dr. McBrearty said. The behaviors were gradually assembled as a package and exported, "which is why it appears suddenly in Europe 40,000 years ago," she said.
Dr. Klein said in an interview that he doubted some of the early dates proposed by Dr. McBrearty and Dr. Brooks, and that even if the dates were correct, modern behaviors conferred such an advantage that they should appear in a broad pattern, not just at the handful of places cited by his critics. To understand what happened in the past, it is necessary to look for patterns and ignore the "noise," he said.
The synthesis of archaeology with population genetics may provide a basis into which a third discipline can join, that of historical linguistics. Most linguists insist that languages change so rapidly that their roots cannot reliably be traced further back than 5,000 years. Only a few, like Dr. Joseph Greenberg of Stanford, believe that some elements of language remain constant, enough to reconstruct all the world's languages into just 14 superfamilies of a much great antiquity.
The signature of these ancient superfamilies can be seen in the geographic distribution of languages, Dr. Renfrew said. In some areas of the world, like the Caucasus, New Guinea and South America, there are many language families packed into a small area, which he called a mosaic zone. In other areas, a single language family covers a broad area or spread zone. The Indo-European languages, which stretch from Europe to India, are one such example. Another is Afro-Asiatic, the superfamily that includes the languages of Ethiopia and Somalia and Semitic languages like Arabic and Hebrew.