The First Americans

by James Adovasio

  I followed Dillehay and made my own sarcastic comments about the Old Faith and the doubters still carping about matters at Meadowcroft that had long been settled. I suggested that it was time to ignore altogether “the embittered acolytes of a religion that dies harder than Bruce Willis or Rasputin.”

  It was not all attack and defend, however. People from numerous fields, including linguistics and molecular genetics, discussed relatively new findings in this field that tended to support the pre-Clovis point of view. Other reports chronicled what appeared to be other pre-Clovis sites. Many of them were given over to the propounding of new pre-Clovis speculations and hypotheses about the timing and routes by which people might have come. And at one point Michael Collins, an old hand at the study of late Pleistocene America, exclaimed, “My God, this is an exciting time. Let's enjoy the fun and get on with the positive aspects of researching this.”

  His comment reminded me of another I had read earlier by DouglasPrice of the University of Wisconsin at Madison. It was a chapter he wrote called “The View from Europe” in a symposium volume on the peopling of the New World that Dillehay and Meltzer had put together in 1991. Price pointed out that there were considerable similarities to be seen in the peopling of North America (with people heading south into previously uninhabited country) and the peopling of northern Europe, especially Scandinavia, with people moving north into country available once the ice began to recede. The evidence was fairly clear now, he wrote, that these pioneers into Europe's recently deiced lands had not been “mobile, carefree, happy bowmen” as previously thought. Instead, such things as cemeteries of this period found in Denmark suggested that they had, on the contrary, been largely sedentary, both inland and on the coasts, and that marine and plant foods had been far more important in their diets than red meat.

  As Price argued, the mammoths and woolly rhinos of the area were already largely extinct by the time northern Europe began to warm up; they were not in any case found in association with human remains or artifacts. The pioneers did wipe out the last populations of aurochs, elk, and reindeer, but only on remote islands of the northern seas. Indeed, the auroch persisted until it was finally rendered extinct in Poland in the nineteenth century. So there was no wholesale slaughter of meat on the hoof in northern Europe to match the supposed blood-drenched blitzkrieg of Clovis Man.

  One of the great differences between European and American archaeology of this period is that in northern Europe preservation is better. Thanks to the lay of the land, there are vastly more sites, specifically a large number of lakes, wetlands, and peat bogs (wet sites like Monte Verde) where lithics and perishable artifacts remain more intact. As Price pointed out, in North America there are only a few late Pleistocene sites, and these are mostly temporary camps such as Meadowcroft Rockshelter. We have come up with practically no base camps or settlements, though they almost surely existed, as they did in South America. Price advised that we in the Americas spend a lot more time looking in old wetlands and along old lakeshores.

  With exceptional courtesy, Price also mentioned another difference between the two continents' styles of archaeology. In discussing the Hamburgianculture, one of the characteristic tool kits that had moved north in Europe as the glacier receded, he pointed out that there was great uncertainty as to whether it had originated in France or as far east as Ukraine. But, he said, the matter of origins generates a great deal more “heat” in America than in Europe—why, he wasn't sure. Perhaps, he went on, there was more to learn in America about our Ice Age past from cooperation than from acrimonious exchange.

  CHAPTER TEN

  THREE-LEGGED STOOLS AND SKULL WARS

  For many of the True Faith, the Clovis Bar was confirmed in the early 1980s by three other lines of evidence arising from linguistics and human biology. Archaeologists will steadfastly and to a person say that if there's no evidence of “it” in the ground, we cannot really believe that “it”— whatever “it” may be—happened. But they are perfectly happy to have people from other disciplines come along with theories, evidence, or even just opinions that confirm their own views. To have three legs on the stool of one's argument is reassuring. Four is even better.

  Just at the time in the seventies and eighties when Meadowcroft and Monte Verde had begun challenging the whole fabric of archaeological “truths” about the peopling of the New World, there came along three lines of inquiry: (1) linguistic studies, (2) the study of tooth morphology, and (3) a forensic technique of studying genes that had led to the recently popularized notion that we humans are all the children of a single, original female whom Western science named, with breathtaking originality, Eve. From these various lines of inquiry there arose a nearly euphoric sense of the justness and truth of Clovis First.

  LANGUAGE, TEETH, AND GENES

  Despite the fact that archaeology and linguistics fall under the academic umbrella of anthropology, at least in the United States, most American archaeologists are not trained in linguistics or are even much aware of how linguists work. Yet for years scholars of the indigenous languages of North and South America had complained that 11,200 years simply wasn't anywhere near enough time for the nine hundred or so separate languages that were present in the New World at the time of Columbus's arrival to arise from one invasion of fifty or a hundred would-be big-game hunters who presumably spoke one tongue. In North America alone there were some three hundred spoken languages deriving from either six or eight different root language stocks called phyla. For example, one such phylum is called Macro-Siouan, and it includes not only the languages of such Plains tribes as Crow, Mandan, Omaha, Winnebago, and Sioux but also the languages of the Mohawks and other Iroquoian tribes of New York State and Canada, as well as the Cherokees, who were from the Carolinas, the Catawbas, the Caddoan tribes, and the Yuchis. By linguistic reckoning— which calculates how long it takes a fundamental root word such as that for “mother” or “sun” to change from the original—it seemed clear that even 15,000 years was insufficient time to account for such remarkable diversity.

  One of the North American language phyla includes the ten languages of the Eskimos (Inuits, preferably) and Aleuts in the far north; another includes the thirty-eight languages called Athapaskan—spoken mostly by northern Canadian tribes but including the Apaches, the Navajos, and some California groups. It has long been taken as certain that the Athapaskan-speaking peoples—also called Na-Denes—were relatively late arrivals on the continent, followed yet later by the Inuits and Aleuts. If true, this left an impressive 250 languages spoken by the Indian tribes in the present United States alone. Though nothing compared to the polyglot world of New Guinea, where some 700 languages used to exist, those 250 languages represent a lot of linguistic evolution.

  Could some of the linguistic diversity of the Americas, therefore, havedeveloped in Asia? And would that suggest that several different groups of people arrived here at different times? Or was it essentially one group of people that wandered out on the misty steppes of Beringia and wound up populating a hemisphere, with the exception of the late-arriving Na-Denes, Inuits, and Aleuts?

  The complaints and suggestions of the linguists fell on archaeological ears that were mostly deaf. Whatever linguistic methods of computation were, they surely did not have the weight of in-the-ground, tangible artifacts that could be excavated, fondled, and dated by radiocarbon analyses. Archaeologists could ignore the theories of linguists as though they were a cloud of gnats flying around their heads. That is, until they found Joseph Greenberg, a linguist who entered the fray in the early 1980s with a different and convenient theory.

  Greenberg announced that all the other native languages of the Americas except Na-Dene, Aleut, and Inuit had arisen from a single root language he called Amerind. Part of his reasoning was that all of the Amerind languages—about nine hundred in all—share approximately 280 words, as well as a few grammatical forms. Most notably, and unique among the languages of the world, their first-person pronouns begin with
n and their second-person pronouns with m. This was strong evidence that a single group of people sharing a single language had been the first arrivals—read Clovis— particularly when it was added to evidence arising from other nonarchaeo-logical fields, such as the anatomy of teeth.

  All human beings share the same basic dental patterns, but different groups share certain minor, or secondary, dental characteristics. All of us, for example, have thirty-two adult teeth including our first molars, but some groups of people have different numbers of roots on the first molar. In all, human teeth have more than twenty different secondary characteristics. Christy Turner, a biological anthropologist at the University of Arizona, undertook a study of some 200,000 teeth from 9,000 prehistoric American Indians, along with thousands of teeth from Siberia, southern Asia, Africa, and Europe. Based on all the secondary characteristics, it is possible not only to distinguish Asian teeth from African and European teeth but also to separate the Asian teeth into two separate groups. One group, called “Sundadont,” is characteristic of teeth from Southeast Asia and begins to show up in the fossil record about 30,000 years ago. Fromthis group the second, known as “Sinodonts,” evidently evolved, and they characterize most northern Asians and all the native populations of the Western Hemisphere.

  One of the key traits of Sinodont teeth, which arose in Asia about 20,000 years ago, is extra ridges on the insides of the upper incisors, giving them a distinctive shovel shape. Sinodonts also possess three roots on the lower first molars, and there is, of course, variation among them. Na-Dene teeth are different from Inuit-Aleut teeth, and both groups' teeth differ from those of the rest of the Indian people of the hemisphere. Seemingly a perfect match for Greenbergian linguistics, paleodentistry added its own approximate time scale as well. And both these promising lines of evidence pointed to a single arrival. The Clovis First archaeologists, the great majority of North American archaeologists at the time, were delighted—even more so when the molecules that make up the genes began to tell the same story.

  In the 1980s, the idea of working out the entire genetic code of Homo sapiens was a dream that seemed a long way from fruition. What with out-breeding, inbreeding, and natural mutations, human genetics can be pretty messy. But there is another type of DNA found in small organelles within cells called “mitochondria,” which are responsible for energy metabolism within the cells. Inherited exclusively from the mother, this DNA mutates naturally five or ten times faster than the DNA found in the cell's nucleus, which is the result of the seemingly haphazard mixing of DNA from both parents. Scientists believed they could clock the changes in mitochondrial DNA, known as mtDNA: on average, every one million years between 2 and 4 percent of mtDNA will have mutated of its own accord.

  Soon—to simplify an extremely complex matter a bit—it was found that groups of Indians geographically widely separated nonetheless had nearly identical variations in their mtDNA. This meant that the group ancestral to them all had to have been very small. Q.E.D. So here was evidence for Paul Martin's band of rampaging Clovis hunters. As the biochemists peered more closely, though, what had been thought to be a clear picture became much murkier. Since the variations in mtDNA common to the native people in this hemisphere share only a few characteristics with those found in Asians, the variation clock must have begun to tickonce people arrived here, not before. Also, it began to appear that the clock had begun ticking between 21,000 and 42,000 years ago.

  LANGUAGE STOCKS AND HAPLOGROUPS

  It soon turned out that the more or less single story told by genes, teeth, and language had big problems. The neat biological and linguistic tale that matched the tale of Clovis First began to unravel. According to the evolutionary rates postulated for teeth, the ancestral people of all three distinct groups of native people began to differentiate only some 13,000 years ago. The rates of tooth evolution also suggested that the Inuits preceded the Na-Denes to this hemisphere, which flew in the face of all other evidence holding that the Inuits came after the Na-Denes. Furthermore, the 13,000-year-old onset of differentiation suggested by the teeth differs from the genetic evidence, which showed the differentiation as beginning more than 20,000 years ago and the Na-Denes preceding the Inuits. In the meantime, forensic scientists at the Federal Bureau of Investigation, who use mtDNA to “fingerprint” victims and criminals, began finding that reading mitochondrial evidence was more complicated than it at first seemed. The molecular biologists would soon agree, and they had to refine their own techniques.

  Meanwhile, Greenberg's nearly pan-Indian root language, Amerind, was coming in for voracious counterattack. Chiefly, it was seen as an over-simplification, disguising important differences and overemphasizing similarities. Students of historical linguistics were—not to put too fine a point on it—outraged (though not speechless with rage; academics are never speechless). A not atypical example was a paper prepared for a 1994 conference on methods of investigating the peopling of the Americas by Ives Goddard of the Smithsonian and Lyle Campbell of Louisiana State University. They referred to Greenberg's treatment of language as “superficial” and his argument as “specious” to an extent even greater “than other types of data used by prehistorians.” Greenberg's so-called multilateral comparison of many languages looking for similar-sounding words with similar meanings had nothing whatsoever, they said, to do with the historical waythat languages might have come about. For example, the criteria Greenberg established for Amerind also include Finnish. The actual history of languages is far too complicated for that sort of analysis: languages go extinct all the time, some groups of people are multilingual, changes occur at different frequencies, and words from several different languages can converge adventitiously. As for the first-person n and second-person m, the two historical linguists scoffed that not only is it not a unique pattern in the world (it is used in Swahili), but it is also not all that common in the Americas. As even Greenberg noted and then overlooked, the most common first-person sound in South American native languages is i.

  Having explained that postulating Amerind as a unitary language was “a vacuous hypothesis,” Goddard and Campbell had—at least to their own satisfaction—utterly destroyed the Greenberg analysis. They went on to say that so malleable are languages that historical linguistics cannot of its nature take languages very far back with any sense of confidence. That a multilateral comparison of European languages does show legitimate similarities is no surprise, since most are relatively recent anyway, having split off at most some six thousand years ago. Earlier than that, they said, tracking languages back to common roots is not presently possible. Certainly, tracking languages twice as far back—or farther—is out of the question. According to their analysis, a single root language for most American Indian languages is not ruled out, but neither are several root languages representing several different migrations of people in the deep past.

  So almost as soon as all this nonarchaeological evidence appeared to be standing stalwartly behind the comfortable archaeological dogma of Clovis First, the three-legged—indeed, counting Overkill, four-legged— stool was yanked away. It seemed that an equally good case could be made from linguistics, dentition, and genes that there was a long period, maybe 10,000 years or more, during which numerous groups already somewhat different from one another in various ways could have made several separate forays across the Bering land bridge or along its coasts. Indeed, some could have ventured out on the land bridge of Beringia and returned, others could have reached the new continent (not that it would have seemed to be a new continent) and gone back to Asia, and some could have perished without a trace. In any event, by the mid-nineties those archaeologists who needed a one-time arrival sometime shortly before Clovis cultureappeared in North America were back on their own, unbuttressed by dental studies, mtDNA, or most linguistic theory.

  At this point, traditional linguists were again confronted by another revolutionary in their midst, Johanna Nichols of the University of California at Berkeley. Among other claims
, she says she can track languages back beyond the 6,000-year cutoff point—in fact, way beyond. Somewhat in the tradition of the starship Enterprise, she has boldly gone into territory that most other linguists have not dared to explore. Although she is taken more seriously by the linguistic community than Greenberg, not everyone finds her work beyond doubt, and some harrumph that she is altogether wrong. What else is new in the world of social science?

  Nichols's method has little to do with the traditional linguist's comparison of words and phonemes to find similarities and therefore relationships. Comparative methods have so far left linguists without their holy grail, a single tree from which all languages have branched over time. Instead they are confronted by some two or three hundred separate trees (what Nichols calls “stocks”) that are hard to relate one to another. For example, the so-called Indo-European stock has 144 branch languages, all of which can be traced back about 6,000 years, beyond which is murk. Other stocks have fewer or more member languages; some, such as Korean, Basque, and maybe the language of the Zuni Indians, have only one. How these stocks are related remains enigmatic.

  Nichols's strategy is instead to determine the grammatical building blocks (the regularities) within and among language stocks and to map them. One might think of these as the girders and cement blocks, or the infrastructure, upon which a variety of differently adorned buildings can be erected. The building blocks tend to last, she says, while languages based upon them come and go. There are several dozen such building blocks, some of them fairly familiar, such as whether a verb typically goes at the end of a sentence (as in Latin), somewhere in the middle (as often is the case in English), or at the front. Some are quite arcane: for example, what Nichols calls “ergativity,” a feature of languages that use special prefixes or suffixes to modify a verb. For example, the tense of an ergative verb may be indicated by a prefix rather than a different form of the verb. Where we say “I shot a deer,” a speaker of such a language would say “I (past) shoot a deer.” Other languages, such as Chinese and Navajo, use vocal pitch ortone to change the meaning of words. (In Navajo, a tin-eared imitator, trying to say something as straightforward as “Hello,” may wind up, thanks to the wrong musical pitch, saying something scatological instead, such as “I fart often,” to the hilarity of his or her Navajo listeners.)