On a hillside by the Savannah River (South Carolina, USA) an archaeologist and a graduate student  had reason to think they were in the presence of a breathtaking discovery. Or at the least, they were on to something more than 20,000 years old that would throw American archaeology into further turmoil over its most contentious issue: When did people first reach America, and who were they? The sandy soil of a trench walls was flecked with pieces of chert, the source of flint coveted by ancient toolmakers. Some of the stone flakes appeared to be unfinished discards. Others had the sharp-edged look of more fully realized blades, chisels and scrapers. Long ago, it seemed, Stone Age hunter-gatherers had frequently stopped here, and, perhaps, these toolmakers were among the first Americans.
     The archaeologist, Albert Goodyear of the University of South Carolina, excised a chunk of chert. Its sides, he said, had all the marks of flint-knappers' work. They had presumably smashed one cobble against another, leaving fracture lines through the rock, and then recovered thin slices for making sharp tools. "This is not a natural occurrence," Goodyear said, "Too many blows have been struck." If he is right, American prehistory is being extended deeper in time at this remote dig site near Barnwell.
     Judging by the depth of sediments, the site may have been a toolmaking center at least 7,000 years earlier than the arrival of the Clovis people. Once thought to be the earliest Americans, Clovis hunters left their finely worked fluted projectile points across the United States over five centuries, beginning 13,000 years ago.
     Robson Bonnichsen, who is director of the Center for the Study of the First Americans in College Station, Texas, and has examined some of the possible artifacts, said, "If the preliminary findings hold, this is a tremendous discovery." But he cautioned that "a lot of hard research needs to be done to really test this thing thoroughly." A hurdle, scientists said, may be to establish that the stone pieces are indeed human-made tools. Many a presumed pre-Clovis site has failed to gain scholarly acceptance over the question of whether stone pieces that look like tools were the work of early humans or of nature.
     Bonnichsen said much of the 16,000-year-old chert material previously excavated "looks really good" and might well be tools. But it is too soon, he added, to render a nature-versus-culture verdict on the stone pieces from the greater depths and earlier ages at the site excavated by Goodyear. More experimental work is required to understand how the chert could have been modified into tools.
     Goodyear said he planned a wider and more intensive search next year. Sarah Sherwood, an anthropologist at the University of Tennessee, is to visit the site to investigate for signs of bone and plant remains, possible evidence for cooking fires, and to determine whether the remains are indeed from a fireplace and are not an accumulation of ash deposited by river floods.
     "If this is 25,000 years old, and I think it is," said Goodyear, "Then scientists will come here from all over the world to see for themselves, and they will argue about it for another 10 years."

Sources: International Herald Tribune, The New York Times (1 July 2004)