"Tracking the First Americans" by Brian Fagan in "Archaeology" (November/December 1990, pp. 14-20) "Ancient Find, But How Ancient?" ...The headline in "The New York Times" for April 17 stood out from a mass of political news. Correspondent James Brooke had traveled deep into north- eastern Brazil in search of the first Americans. He had found them at the Boqueirao de Pedra Furada, deep in the arid Sao Raimundo Nonato region. Here French archaeologist Niede Guidon and Italian colleague Fabio Parenti were digging in a remote area famous for its fine rock paintings and prolonged human settle- ment. Pedra Furada is a deep rock-shelter, with fill more than 55 feet in depth. Like other archaeological sites in the Sao Raimundo Nonato, the shelter contains abundant, and undoubted, evidence of human occupation as far back as about 10,000 years ago. But Guidon and Parenti claim that this remote shelter was occupied much earlier in prehistory, as early as 47,000 years ago. That is a staggeringly early date for human settlement in the Americas, more than 30,000 years earlier than the 15,000- year-old date that represents conventional archaeological wisdom. If the new date is correct, then human beings were living in the New World at a time when Neanderthals flourished in Europe. Boqueirao de Pedra Furada first made headlines in the British scientific journal "Nature" in 1986. Guidon and her colleagues then claimed they had found 32,000-year-old hearths and stone artifacts in the depths of the Furada. Not only that, they wrote of painted rock fragments going back to at least 17,000 years ago, some of the earliest cave art in the world. The 32,000-year date caused considerable surprise. It was very early, but not impossible, if the evidence for it was adequately documented. Now Guidon and Parenti have raised the stakes dramatically. They claim to have 21 radiocarbon and thermoluminescense dates associated with human occupation that go back from 14,300 years ago to 47,000 years ago. The two excavators also report that they have discovered charcoal and burned rocks arranged in hearths going back more than 30,000 years in an "unbroken se- quence," to quote the "Times". The 47,000-year date, according to the two, is associated with an ash-filled hearth ringed with a circle of stones. What are we to make of such claims? Are they based on meticulously collected archaeological data and careful analyses of cava strata, designed to identify and eliminate all possible sources of nonhuman formation of "artifacts" and "hearths"? Or have the excavators been carried away by their discoveries, by the quest for that holiest of archaeological Holy Grails--the very first Native Americans? I have a horrible feeling that for Pedra Furada the latter is the case. Even after eight years researching and writing about the settlement of the New World, I continue to be amazed at the passions that the debates over the first Americans raise in the archaeological breast. It is as if there is a competition to find the earliest, to break new ground, a competition where specious claims are more important than solid, indisputable archaeological evidence. The recent "Times" story does not help. It carries not only the 47,000-year claim, but also quotes two well-known scholars. One describes the site as "hot stuff," while the other argues that Guidon's early date has been "pretty well deep-sixed" by scholars working in the field. Strong stuff, but anyone who follows the subject soon becomes inured to claims and counterclaims. Ever since I wrote "The Great Journey", a book about the peopling of the Americas, a torrent of correspon- dence has crossed my desk. Many letter writers want me to take sides, to state that I either support the notion of 30,000- or 40,000-year-old settlements in the Americas, or that I join the "diehards" who believe no one ventured into this part of the world until after the Ice Age, some 12,000 years ago. Which side am I on, they want to know. The story in the "Times" is clearly written with side-taking in mind. What absolute nonsense this is, and how counterproductive for archaeology. The debate over who first settled the Americas and when is important, but it should be entered into with great care. To do otherwise is to demean archaeology as a serious science. I then recalled a report in the April 1990 issue of "Ameri- can Antiquity" detailing how archaeology at early sites should be conducted. In it, James Adovasio, now director of the Archaeo- logical Institute at Mercyhurst College, and two colleagues summarize a decade of chronological work at the once controver- sial Meadowcroft Rock-shelter, 30 miles southwest of Pittsburgh. Throughout the 1970s, Adovasio and a multidisciplinary research team searched the deep deposits of Meadowcroft with a fine- toothed archaeological comb. Adovasio excavated the intricate layers of the rock-shelter, passing through no less than 11 strata. With comprehensive dating, he was able to document a remarkable continuity of human occupation from about 700 to 12,000 years ago--and considerably earlier. Initial radiocarbon dates from the earliest levels suggested that humans might have occupied the site as early as 19,500 years ago, a little less than 30,000 years after the alleged settlement in northeastern Brazil. While everyone accepted the chronology for the later levels at Meadowcroft, many experts were worried by the possibility of contamination in the lowermost levels, either as a result of coal particles in the deposits or through human disturbance. The "American Antiquity" essay is a carefully reasoned analysis of the controversy and a line-by-line response to critics' concerns. Adovasio describes how all possible sources of contamination were removed from the radiocarbon samples and how a new accelerator method was used to check the accuracy of dates determined by other methods. The Adovasio essay is a model of just how a controversial dating question should be handled. The tone is sober, the data comprehensive, the statistical arguments telling. We can be confident that the question of the age of this most important North American site is as settled as it ever will be. As Adovasio remarks, "it is important to note that the earliest Meadowcroft dates that have extensive artifactual associations do not argue for any radical extension of the 15,000-year baseline. "Perhaps," he says, "humans were at Meadowcroft some 2,000 or 3,000 years earlier than the first, well-documented appearance of Clovis people about 11,500 years ago in much of the Americas." There is a lesson in all this. Adovasio's chronology was questioned by other scientists not on the grounds that it was wrong, but because they were concerned that there might be undetected contamination in the samples. Over a period of years. Adovasio investigated this possibility with great thoroughness, confirmed the basic validity of his dates, and produced as refined a chronology as he could. How, he says, it is time to "address the issue of other potential Meadowcrofts." He is correct, but there is definitely a right and a wrong way to do it. Meadowcroft is a long-term project that has cost tens of thousands of dollars and engaged the best efforts of a talented research team for a decade. A great deal of effort has gone not only into documenting and publishing the finds from the shelter, but also into understanding the complicated sequence of geologi- cal and climatic events that contributed to the formation of the occupation deposits at Meadowcroft. In short, we know as well as we ever will how the early deposits were formed and how old they are. Have Guidon and her colleagues worked with the same scien- tific meticulousness? Guidon and Parenti's claims fly not only in the face of established chronologies, but in that of scientific archaeology as well. For a start, many archaeologists are deeply suspicious of the Brazilian site. They argue that the hearths were the result of natural forest fires, that runoff in the vicinity of the rock-shelter caused mixing in the deposits, that the quartzite "artifacts" from the site result from rock falls from the cliffs above. How can one make such judgments without seeing the place first hand, the excavators argue? As for the arti- facts, Parenti says he is certain of at least 200 or 300 of the artifacts. According to the "Times", he implies he is more qualified to make that judgment than his critics are because he is more accustomed than American archaeologists to examining the kinds of primitive artifacts found with people who lived more than 40,000 years ago. This is hardly a reasoned and sober response to legitimate scientific objections. Few American scholars have visited Sao Raimundo Nonato, which is far from the beaten track. This means that Guidon and her team of excavators have been working in much greater isola- tion that James Adovasio did at Meadowcroft. Dozens of scholars visited his excavation, to the great intellectual benefit of all concerned. Apart from inevitable language barriers, those who labor at Pedra Furada have few first-hand opportunities to discuss with other experts their stratigraphic and geological observations, or their claims for hearths and early stone tools. In their eagerness to defend their cherished claims, the excavators of Boqueirao de Pedra Furada have missed the real point that their critics are trying to make. These experts are not necessarily saying that Guidon is wrong, nor are they taking sides. What, they ask, is the basis for the claim for a 47,000- year-old settlement in the Americas other than simple observation and instinct? For instance, how did the deposits in which the hearth and its stones were discovered form? How did the ash concentration accumulate? As a result of natural fires, or through deliberate human actions? Have flakes from the early levels been fitted together to reconstruct flaking practices and stone technology? What possible natural causes for flaking quartzite could account for the "tools," if any? To date, it is fair to say that we have only ever-earlier datings and flat claims to work with. Judging from the available literature and conversations with many of my colleagues, these claims are based on the notion of "I observed this; trust my observations." The careful work at Meadowcroft has set a standard; personal convic- tion is no longer sufficient. What is needed at Boqueirao de Pedra Furada is another Meadowcroft project, where every possibility is exhaustively and laboriously worked through. Putting it bluntly, a claim for 47,000-year-old human settlement in the Americas has to be documented beyond a reasonable doubt by methods that use modern science to eliminate the possibilities that the "hearths" and "stone tools" were created by natural phenomena. Any other approach is intellectually and archaeologically bankrupt.