Coming to America

By MICHAEL D. LEMONICK

KNOW-IT-ALLS ARE fond of pointing out that the only true Americans are the descendants of the diverse tribes found by Columbus and Cortes when they first arrived in the New World. That's wrong, of course. Even North and South American Indians had immigrants for ancestors: northeastern Asians who crossed from Siberia to Alaska in prehistoric times across the bridge of land that then spanned the Bering Strait.

But when did these adventurous souls reach Alaska? What kind of people were they? How fast did they spread down through the Americas? For decades, archaeologists felt sure they knew the answers: the first Americans were skilled hunter-gatherers and toolmakers who arrived about 11,500 years ago and moved rapidly southward, reaching deep into South America within about five centuries as well as helping drive to extinction such prehistoric mammals as mastodons and woolly rhinos.

Now, a competing theory about the original Americans, once touted by only a small band of renegade archaeologists, has become too compelling to ignore. Its thesis is that the first migration took place not 11,500 years ago but 20,000 or 30,000 or even 50,000 years ago. Although the evidence is still sketchy, archaeological digs in Chile, Brazil, Venezuela, the U.S. and Canada have yielded tantalizing clues that this radical notion might be correct. "This is a hot area of research," says Dennis Stanford of the Smithsonian Institution. "Man's origin in the New World is one of the major unanswered questions of archaeology."

The old assumption that humans arrived in the Americas about 11,500 years ago is known as the Clovis hypothesis. The name comes from the 1933 discovery of a fluted stone spearpoint dated to that era in a pile of mammoth bones near Clovis, New Mexico. Over the years, similar spearpoints were unearthed all over North America, all apparently about the same age. Because the weapons, known as "Clovis points," were so widespread, and because essentially no artifacts at all were found in older sediments, archaeologists and anthropologists concluded that the Clovis people were the first and that they came over from Asia in the last years before melting glaciers and rising sea levels submerged the Bering land bridge.

The theory has flaws, though. One is the idea that prehistoric people could have populated an entire continent in a mere 500 years, the span between the time of the presumed land migration and the time by which Clovis spearpoints had been deposited throughout North America. Even more problematic are signs of very early culture in South America. "Humans don't sprint through their environment," says Mercyhurst College archaeologist James Adovasio. "But that's what the Clovis guys would have us believe. There's no analogue for that in archaeological history."

More doubts arise from studies of linguistic and genetic diversity. Modern North and South American Indian languages presumably evolved from a single ancestral tongue, but they differ so greatly that it is hard to imagine how this could have happened in just a few score centuries. Similarly, the mitochondrial DNA in the cells of Native Americans differs so much from tribe to tribe that a single, relatively recent ancestral group seems unlikely.

Still, if the Clovis people were not first, where is the evidence of their predecessors? According to a growing group of archaeologists, the signs of the original Americans can be found in several places:

Monte Verde, Chile

Perhaps the most convincing candidate for a pre-Clovis site is Monte Verde, on the Chinchihuapi Creek in southern Chile. A team led by University of Kentucky archaeologist Tom Dillehay discovered indisputable traces there of a human settlement that was inhabited between 12,800 and 12,300 years ago. Usually all scientists can find from that far back are stones and bones. In this case, thanks to a peat layer that formed during the late Pleistocene era, organic matter was mummified and preserved as well.

In a decade of digging, Dillehay's team found an unparalleled array of artifacts, including not only stone tools and animal bones but also chunks of mastodon meat, wild potatoes, and seaweed and other plants that must have been imported from the Pacific coast, some 40 miles away. The archaeologists discovered fire pits surrounded by burned wood chips, wooden lances with hardened tips, wooden basins containing seeds, grindstones -- and a human footprint. The foundation of a wishbone-shaped structure held the remains of more than 20 types of medicinal plants, some of which bore marks that may be the imprints of human molars. Most intriguing of all, the scientists unearthed wooden foundations and crude timbers that Dillehay believes supported an oval, tentlike dwelling similar to late Pleistocene shelters found on the Siberian plains. Says he: "We know these people exploited a wide variety of resources stretching from Monte Verde all the way to the coast. They used wood, ate plants, fashioned stone tools and from time to time captured game animals, such as mastodons and paleollamas."

All the artifacts from Monte Verde have now been subjected to dozens of radiocarbon analyses -- a standard archaeological dating technique in which the amount of radioactive carbon in an organic specimen is used to calculate its age. Dillehay says he is "very confident" that he has found remnants of a culture that existed some 125 centuries ago.

Dillehay has also uncovered traces of what may be an even older campsite nearby on a buried promontory. The evidence: 26 fractured stones, some of which were clearly worked by human hands, as well as three clay-lined pits containing charcoal that may be nearly 33,000 years old. Although radiocarbon dating supports this idea, Dillehay is reluctant to draw any conclusions. "The older level is a hell of a problem," he says, "and it simply will not go away. The more I look at the evidence, the more it looks like it represents human culture, but intellectually I still can't accept that humans were in the New World earlier than 15,000 to 20,000 years ago."

Meadowcroft Rock Shelter, Pennsylvania

Although the age of the earliest objects from Meadowcroft remains controversial, this rock shelter 30 miles southwest of Pittsburgh has long been considered one of North America's most promising pre-Clovis sites. Among the findings: charcoal, pieces of bone and antler (some scored with knife marks) and charred fragments of basketry that are estimated to be between 12,000 and 15,000 years old. There is also an assortment of non-Clovis blades and points. Says Mercyhurst's Adovasio, who has studied Meadowcroft for nearly 20 years: "It may well be the oldest archaeological site in North America."

Bluefish Caves, Yukon Territory

Discovered in 1975 by researchers with the Archaeological Survey of Canada, these caves in the remote northern Yukon have yielded flaked stone tools that are 10,000 to 13,000 years old, what appear to be butchered mammoth bones 15,500 to 20,000 years old and bone tools from perhaps 23,500 years ago. To date, however, the researchers have been unable to find any hearths or other cultural features.

Taima-Taima, Venezuela

An ancient water hole called Taima-Taima in northern Venezuela became the deathbed of a young mastodon -- killed, apparently, by some of the first Americans. The site, excavated by Alan Bryan and Ruth Gruhn of the University of Alberta in Edmonton, appeared to contain 13,000-year-old mastodon bones, one of them embedded with a pointed stone projectile. Mixed in were stone tools and rounded pebbles that could have been made only by humans. Some archaeologists, however, believe the artifacts found at Taima-Taima became interspersed with the mastodon bones as the water level in the hole rose and are therefore much younger than they seem. Even so, there is strong interest in the site because its reputed age is close to that of Monte Verde.

Pedra Furada, Brazil

Of all the plausible places for early human settlement of the Americas, Pedra Furada, located in a region of dramatic sandstone cliffs in the arid outback of northeastern Brazil, is probably the most exciting -- and most disputed. When archaeologist Niede Guidon of the School for the Advanced Study of Social Sciences in Paris first excavated the site in 1978, she found cave paintings, ash-filled hearths and what she believes are stone tools that are at least 30,000 and perhaps more than 50,000 years old. Says Guidon: "I was the first person to be surprised. I believed the standard theories." Each successive radiocarbon test, though, bore out her initial findings. She became a convert -- and an untiring champion -- of the pre-Clovis theory.

But convincing fellow scientists has been a battle. Guidon's conclusions have been greeted with skepticism by many archaeologists. One problem, explains Randall White of New York University, is that the "tools" from the deepest levels at Pedra Furada are mixed with naturally fractured river gravel. This suggests that the geological layer was not laid down in an orderly way. The stone flakes could easily have been churned together with much older river rock before settling. Moreover, they might not be human-made at all; the artifacts themselves could have formed by natural erosion.

Critics have similar doubts about charcoal Guidon believes came from ancient fireplaces. "Radiocarbon dating is tried and true," explains archaeologist David Meltzer of Southern Methodist University. "The problem is linking the dating of objects to human occupation. How do you know it was a piece of charcoal touched by human hands and not just a piece of burned tree?" Brian Fagan of the University of California at Santa Barbara is a bit more blunt: "I think Pedra Furada is absolute horse manure."

That kind of derision doesn't faze the feisty Guidon. On the charcoal deposits, she argues, "If they had been left by forest fires, carbon deposits would have been found scattered across a wide area." They are not. In many cases, the charcoal is ringed by stones, says Guidon, which is strong supporting evidence that these were man-made hearths, not natural formations. Besides, the area was a humid, tropical rain forest 30,000 years ago, and natural fires would have had a hard time getting started.

The artifacts in deep layers don't trouble her either. They couldn't have been washed in from elsewhere and mixed, she says, because the rock shelter where they were found is more than 60 ft. above the surrounding terrain. Nor < could the objects have tumbled down from higher up on the cliff, says Guidon, since the cave is protected by a massive rock overhang that would have kept out both falling rock and flowing water.

Those who remain skeptical of pre-Clovis findings are most troubled by the ambiguous nature of many of the artifacts. To make a convincing case for a pre-Clovis culture, says Cornell archaeologist Thomas Lynch, "recognizable artifacts from that period must be dispersed over a broad area, reflecting the movement of primitive peoples from place to place. A Clovis point is just as recognizable as the tail fin on a 1952 Cadillac."

Like its more conventional counterpart, the pre-Clovis theory has some logistical problems. If humans got to South America by 13,000 years ago, they would have had to cross the Bering land bridge many thousands of years earlier. That would have been no problem, but heading south from there would have been tough: ice sheets -- or the inhospitable terrain they left behind -- cut off virtually all access to the bulk of North America from Alaska between 20,000 and 14,000 years ago. Guidon's rather controversial answer: maybe the immigrants came over to South America in boats directly from Asia.

Despite all the doubts and unanswered questions, the case for pre-Clovis Americans is gaining ground. Even if the discoveries at Pedra Furada fail to satisfy the critics, sites such as Monte Verde and Meadowcroft are powerful testimony that early migrations did take place. However the first immigrants got to the New World, and whatever the reason why they left behind so little physical evidence, it has become difficult to deny their existence -- and increasingly likely that earliest American history will have to be rewritten.

With reporting by Andrea Dorfman/New York and Ian McCluskey/Sao Raimundo Nonato