Evidence for “pre-Clovis” human activity has been accumulating for decades as archaeologists have found a few unusually old sites in places as far apart as coastal Chile and central Pennsylvania. But there were always problems — a jumbling of deposits, uncertainties of dating — that made some archaeologists doubt the age of those discoveries.
The Texas finds, reported Thursday in the journal Science, are likely to persuade nearly everyone. The undisturbed condition of the site, a distinct layer of artifact-containing sediment below the Clovis deposits and dating that consistently puts that layer at 13,200 to 15,500 years old is what makes this discovery especially convincing.
“It pretty much closes it for me,” David G. Anderson, an anthropologist at the University of Tennessee in Knoxville, said of the debate about whether there were people in the Americas before the Clovis period, which began about 13,000 years ago and lasted less than 2,000 years.
“This is almost like a baseball bat to the side of the head of the archaeological community to say, ‘Wake up, there were pre-Clovis people here,’ ” said Michael R. Waters, the anthropologist at Texas A&M University who led the excavation.
Gary Haynes, an archaeologist at the University of Nevada at Reno and a skeptic of previous pre-Clovis claims, said: “This one comes closer than any of the others. I think it’s a half-step from finishing off the argument.”
The Clovis culture takes its name from the town in New Mexico where the characteristic artifacts were first excavated in the 1930s.
The newly unearthed objects come from a site northwest of Austin, which has been under excavation for several years along a waterway known as Buttermilk Creek. They consist of relatively crude scrapers, knife blades, broken and half-repaired spear points, and more than 15,000 flakes and chips testifying to human workmanship. They bear some similarity to Clovis tools, although not a clear one.
Whether the people who made them were related to the people who made the Clovis tools is uncertain. However, no bones or other DNA-containing materials were found, so the question can’t be answered.
“Cultural history and biological history do not have to go hand in hand. So there’s no way you can say they were related to each other,” said Eske Willerslev, director of the Center for GeoGenetics at the University of Copenhagen.
Anthropologists believe the first Americans migrated across a land bridge connecting eastern Siberia to Alaska. They then made their way south along the Pacific coast and through an “ice-free corridor” in glacier-covered North America. A site in southern Chile called Monte Verde, 35 miles inland from the Pacific, has evidence of human occupation at least 14,000 years ago.
Dried human excrement found in a cave in south-central Oregon has also been dated to 14,000 years before the present, so it comes from pre-Clovis people. Mitochondrial DNA indicates the producers of the feces were of Asian origin and could be — although are not necessarily — ancestors of American Indians.
Willerslev, who did that research, said there are three main possibilities for the relationship between the pre-Clovis and Clovis people.
They could have both been direct descendants from the same migrant group, with their tools evolving from the crude implements at the Texas site into the fine and highly consistent style known as Clovis.
Alternatively, the Clovis people could have come from Asia in a migration entirely separate from the earlier one. Once here, they could have made improvements on the tool-making of the pre-Clovis immigrants, or they might have brought an already more advanced technology.
The third and less likely possibility is that the pre-Clovis people were of a different ethnic origin, such as European. However, Willerslev said that “as things stand at the moment, I don’t think there’s much evidence that it’s non-Asians” who made the pre-Clovis tools.
Because there were no charcoal, seeds, skin or other materials derived from plants or animals at the Texas site, radioactive carbon dating couldn’t be used to determine the deposit’s age. Instead, the 13-member research team used “optically stimulated luminescence” dating, a technique developed in the 1980s that measures how long certain kinds of rock (typically quartz) have been out of the sunlight.
Waters, the lead archaeologist, said the site was apparently a campsite, not a tool production area. Layer after layer of undisturbed, sequential deposits show that prehistoric people visited the area on and off for thousands of years.
Ruth Gruhn, a retired professor of anthropology at the University of Alberta, visited the Buttermilk Creek site several years ago and agrees it dashes the 60-year-old “Clovis first” view of American settlement.
“It’s a very nice wooded valley near a stream. It probably doesn’t look very different from what it did when prehistoric people occupied it,” she said.