The First Americans

by James Adovasio

  The people who made these tools were no novices. Instead they imported high-quality materials: Kanawa chert from West Virginia, Flint Ridge material from Ohio, Pennsylvania jasper, and Onondaga chert from New York. They used sophisticated knapping techniques to make elegant tools. In other words, these were no simple-minded “rock-bashers” but instead highly skilled craftspeople. Most of the blades we found were small, almost microblades. In other excavations elsewhere, archaeologists had found similar items, a widespread technology in which, for example, a spear or dart might be made by fitting several of these small blades into slots around the end of the spear, making a multiedged weapon of considerable effectiveness. So here had been people familiar not only with that technology but with the flaking of larger points as well. The Miller lance-olate was, we learned, vaguely similar to examples found in Oregon and Texas and not totally unlike examples found in the Great Plains. But theMiller point antedated all those other examples, and this led us to wonder if it might be ancestral to all or, at least, some of them.

  The oldest finished lithic tool from Meadowcroft. Called the Mungai knife, after a local farmer, it was recovered from a ca. 16,000 B.P. living surface.

  The other tools we found were similar to one degree or another to those found at other sites, though earlier. The rhomboid “knives” were like those found elsewhere in both eastern and western Pennsylvania; the graver and retouched flakes resembled those found in Nova Scotia as well as New Mexico's Blackwater Draw, where fluted Clovis points were first found in the 1930s.

  In all, from all eleven levels we would eventually collect 2 million artifacts and plant and animal remains, and spend years sorting through it and making sense of it all. But midway through the second season, on July 13, we got the first radiocarbon dates. We had requested eleven dates to be run from five different strata, and that, in itself, is a lot of dates to be run. Doing so is expensive, and it was not uncommon for a dig to be “completed” with only a handful of radiocarbon dates, or even as few as one, which as often as not called such data into question. I wanted to avoid that, among other things.

  The deepest date derived from a stratum that contained no cultural associations (that is, nothing man-made like a tool or hearth) was anywherefrom 19,000 to 28,000 years old. The most recent stratum we dated went back to only 375 B.C. plus or minus 75 years, and others ranged as far back as 2870 B.C. plus or minus 85 years. And then a huge leap—the next two dates were shockers: 12,900 B.C. and 13,170 B.C.

  The radiocarbon chronology from Meadowcroft Rockshelter.

  And that meant three or four thousand years before Clovis.

  Excavations at Meadowcroft, facing southeast. Dr. David Clark, site supervisor, is in foreground. Students in the background are working on Pleistocene living surfaces.

  Maybe, we all thought, because the gap was so great from about 3000 B.C. to 12,000 B.C., the early dates had to be wrong. On the other hand, the radiocarbon lab was utterly professional and almost always totally reliable. And so also, I knew, were the diggers at Meadowcroft.

  One of these was David T. Clark, our first crew chief—a former U.S. marine who had gone on to graduate from Ohio State University. The prototypical hard case, he was the most meticulous fieldworker I have ever seen or probably will ever see. An enormous amount of closed-site, rock-shelter, and cave excavation involves slowly unearthing material micro-layer by microlayer with a trowel. It can seem endless, finicky, and tedious, but it is upon the care with which this task is done that literally everything else of archaeological importance rests. Clark was an utter genius with a trowel, which he wielded like a laser scalpel, and the students he taught became so proficient that they could hear it or feel it when the trowel hit another microlayer—a subtle shift, an almost imperceptible transition, marked only by a textural change from silty sand to sandy silt or vice versa.

  Clark was taciturn, severe, rigid, and uncompromising. If a studentmade the same mistake twice, he or she would not work another day for him. He had no patience with screwups and expected the people working in his particular area—the so-called Deep Hole, which contains the deepest and oldest levels—to start work a full hour earlier than everyone else and stay on an hour later. Illness or hangover was no excuse. Given his almost monomaniacal concern with precision, I had no doubts that the samples we had sent to the radiocarbon lab were absolutely kosher in terms of context and association.

  General view of excavations at Meadowcroft, ca. 1976, facing southeast.

  After straw-bossing the major excavations at Meadowcroft, Dave eventually moved on, doing his dissertation work at Pyramid Lake out in the western Great Basin in 1977. I got a call from him a little while later.

  “It's gonna be tough here,” he said. I asked him why.

  “The crew chief is sorta different,” he said. I asked him what he meant.

  “Well, she—uh—uh—she doesn't wear a shirt.”

  I told him that he'd just have to deal with it. Dave completed his dissertation, got his Ph.D., and wound up teaching elementary school in Washington, D.C., and working part-time in the archaeology department of Catholic University there. In some ways, I suppose, the high point of hiscareer was his years as crew chief at Meadowcroft. We replaced him there with a former student of mine from Youngstown State University, Mike Beckes. Beckes was another man of superb technical skills, a former Hell's Angel type and more than a bit of a hard case himself, but one who was somewhat more approachable than Clark.

  Another regular was Joel Gunn, a quiet, patient, and insightful man who had a thorough knowledge of cave stratigraphy from working in Europe on several digs, as well as a thorough appreciation of how people used caves and rockshelters again and again over long periods of time. He had, and indeed still has, a high regard for the profound effect climate and climatic change can have on the nature of human cultures. He kept us attuned to these questions as the excavations continued and various kinds of material—including huge quantities of climatically sensitive plant and animal remains—turned up.

  We called Joel Dr. Jekyll or Mr. Hyde depending on whether he had spent any time with a container of John Barleycorn. A thoroughly charming and usually mild-mannered guy, he was much married (and much divorced). Once into the “loon water” (as we always called it), he was capable of a Fredric March type of face change and no end of outrageous activity.

  Interestingly, when Joel retook possession of his body, he usually had no recollection whatsoever of any of Hyde's excessive behaviors. In fact, over time, I became convinced that the two Joels actually did not know of each other. Though today he would probably be locked up for some of his deeds, this was the seventies and we were not members of a department of classics or art history. Most of the archaeology students of the time were people who would be more comfortable on a front-end loader or in a roadside saloon than at the Princeton Club in a Hepplewhite chair.

  The Dark Lord had been among the first field archaeologists to insist that other specialists be on hand from the outset of an excavation, such as geologists, sedimentologists, and floral and faunal experts—people who could assess the many kinds of evidence on the spot as well as later in the lab. In this, the Meadowcroft Rockshelter dig was especially blessed. We had the services of geologist Jack Donahue from the beginning and on through virtually all the years of the dig. More on him later. Another specialist who was with us throughout most of that period was John Guilday,a vertebrate paleontologist who identified the hundreds of thousands of animal remains we came across. He was a polio victim and spent his professional life in an iron lung. His expertise was so great that the Carnegie Museum of Natural History was happy to build him a lab in his house, and there, with grad students or, more often, his wife, Alice, holding up each bone and turning it for him as he directed, he identified every creature with a backbone that we found at the site. He was, to put it simply, the very best in his field.

  It is important to have this kind of continuity of personnel in a dig that extends over a lo
ng time. Otherwise, you can get people with different criteria and different agendas, and the turnover can be a plague to the final assembling of all the data.

  From the outset, well before we discovered that we were going to be shoved into the Clovis/pre-Clovis wars, I had wanted this excavation to be a model of its genre, an unsurpassed use of the finest and most precise archaeological excavating techniques, including any new technique that came along. In the secret recesses of my brain, I actually wanted it to be the best piece of field archaeology ever undertaken.

  To that end and from its inception, the Meadowcroft/Cross Creek project was a multidisciplinary undertaking. The central goal or theme of the operation was the systematic acquisition, analysis, and integration of any and all data bearing on the archaeology, history, paleoecology, geology, geomorphology, pedology, hydrology, climatology, and floral and faunal succession of the entire Cross Creek drainage. Moreover, I wanted the data gathered, analyzed, and interpreted with as great a degree of precision and with the most sophisticated methodologies of which any of the project staff were cognizant. Also, crucially, we were fortunate that the project could be carried out virtually without temporal or fiscal constraints. Heavily funded from the beginning to the present, it was designed to epitomize the so-called state of the art.

  ROCKS, DIRT, AND DUST

  Again, from the outset of the Meadowcroft/Cross Creek project, and even before the first trowel full of sediment was ever removed from the rockshelter,we wanted to learn as much as possible about this small stage upon which any prehistoric actors and actresses might have played out some of their own particular destinies. That meant, of course, understanding the geology of the rockshelter itself and the drainage of which it was a prominent part, not to mention whatever effects on the shelter's climate and condition the waxing and waning of the glaciers had produced.

  Geology, of course, is not simply about the grand events of the past— the great movements of the earth's plates, the forging of mountains, the eruptions of volcanoes, the carving by water's inexorable hand of the Grand Canyon. It is also about much smaller and, in geological terms, shorter events. Even in 1973, it was possible to identify microgeological events and processes that had taken place not over eons but rather in actual calendar years and even far briefer periods. It was, and remains, my contention that to understand the activities of humans in the past you must first define and understand the landscape with whose trajectory they have intersected. In short, without knowing what has happened to the stage over time, how can you study the prehistoric performers?

  Even as a freshman back in the sixties at the University of Arizona, I had never understood why such a huge chasm (pedagogically) existed between geology and archaeology. I took huge doses of geology as an undergraduate and later as a graduate student. I learned everything I could about sedimentology, erosion, microgeology. My first lecture before a class was actually as an unpaid assistant in geology, not anthropology. My archaeology mentors, notably including Emil Haury, often stressed the importance of geology, and of course Vance Haynes was already teaching geoarchaeology at Arizona at that time. The importance of the geological stage was further hammered into my head at Utah by both Jennings and Aikens, as well as fellow students such as David Madsen. Small wonder, then, that for me the elucidations of the geology and archaeology of prehistoric sites were inseparable activities.

  This perspective was not particularly unique then and, indeed, stretched back to the beginnings of both fields in Europe in the nineteenth century. At that time, those engaged in prehistoric archaeology often were trained first as geologists. Today, regrettably, geology has been sharply deemphasized in the training of most undergraduate and graduate archaeology students, with profoundly negative consequences. Indeed, while thelast several generations of American archaeology students may be much more theoretically elegant than I or my generation, many are methodologically bankrupt. In any case, it is often easier simply to summon a compliant geologist to a site and then employ his or her services on an as-needed basis. Regrettably, most geologists (willing or not) are not normally trained to address issues important to the archaeologist or to think in much more temporally circumscribed terms than eons or millennia. In such situations, the geologist and archaeologist can be likened to ships passing in the night without significant communication or meaningful interchange.

  In any event, before fieldwork began at Meadowcroft, I knew what we wanted to ascertain geologically. First, we needed to understand the longhistory of the Cross Creek drainage itself and how long it had been in its present configuration. Next, we needed to know in as sharp a detail as could be managed the precise evolution of the rockshelter over time. How and by what means had sediment been introduced or “emplaced” in the rockshelter? Had the larger rocks, which littered the hillside below the site, once been part of the overhang or roof? If so, how big had the site once been? Material from small grains to boulders will, over time, fall from the ceiling of a rockshelter, a process called spalling. We wanted to develop a record of spalling incidents and how they had affected the human use of this shelter.

  Meadowcroft senior geologist Dr. Jack Donahue (left) and excavation supervisor Dr. David Clark sampling sandstone in the Meadowcroft Rockshelter cliff face.

  To help answer these questions, I recruited a more-than-willing Pitt geologist, Jack Donahue, who along with his students remained active at Meadowcroft for decades. Jack was at the time a coal geologist, a field that by then was not likely to develop much more deeply or widely than it already had. Jack was ready for a career switch, and he threw himself into what he would in many ways help to develop—the field of geoarchaeology. Indeed, he would become the first editor of that subfield's main scientific journal, Geoarchaeology, and he remains active in geoarchaeological research to this day.

  General view of the latest late Pleistocene living surface at Meadowcroft prior to the removal of roof blocks, facing southwest.

  To unravel the complex history of Cross Creek, Jack and his students initiated a geological assay of the Cross Creek drainage and the rockshelter that was (and still is) unparalleled in scope, scale, and resolution, at least in the realm of cave and rockshelter excavations. To all of the previously published geological material about the area, they added the findings of field reconnaissance on foot and by air. They studied backhoe trenches and deep-core borings in the surround (activities that were part of a conveniently timed soil conservation project).

  Late Pleistocene Clovis-age living surface with rock fall removed, facing south.

  As for the rockshelter itself, we first mapped it in detail, including all the rocks lying on the surface. We extracted a series of rock samples all the way up the seventy-two-foot-thick sandstone rock unit that made up the walls and roof of the rockshelter, and from this we could see, via microscopic study, even minute changes in the composition of the sandstone.These changes would let us “fingerprint” rocks we found as we dug down and match them with the places on the roof from which they had spalled.

  Again, before the excavation process began, we inventoried all the plants growing around the rockshelter and up the slopes above, and we stripped an area up to seventy feet from the rockshelter of all vegetation, the better to understand the nearby topography.

  Once excavation began, with a north–south trench running from outside the dripline to the back wall of the shelter, an exhaustive series of collecting and sampling protocols was invoked. We made three-dimensional maps of the site every foot or so as we descended, showing not only any sign of a human presence, but also rock spalls down to the size of a silver dollar and anything else that indicated the amount of floor space that was available and used. We cut twelve sampling columns straight down through the site to the shale bedrock below, to keep track of such things as the ratio of sand to silt particles over time. We looked in these samples for pollen, for elements such as nitrogen, phosphorus, and others that result from the decay of plant materials, and for the remains of small ani
mals such as rodents and snails that are sensitive to small climatic changes. We studied individual sand grains at the atomic level with scanning electron microscopes to see if they had been moved around by water at any time.

  Perhaps the most novel (and at the time unprecedented) geological effort began in 1974, when we placed an aluminum tray on the sloping roof of a protective wood structure we had by then erected over the rockshelter floor to protect it from rain, curious hikers, or vandals, as well as permit work to proceed in foul weather. Into the tray we swept sand and rock fragments from the roof—a job we did daily for the next four years of fieldwork. With this material matched daily to the weather humidity, temperature, and so forth, we had a useful gauge of weather conditions during earlier times at the rockshelter. In this same vein, we used holding tanks to trap all sediment washing into the rockshelter's ends during rainstorms, another way of studying the thick sediment pile that accumulated in the rockshelter over time.

  We soon became the first archaeological excavation to have a hookup at the site to a mainframe computer (at Pitt) and certainly one of the first ever to have such complete computerized records of every detail of the excavation.

  As happens at most professional excavations, we sieved virtually everything through a series of increasingly tiny meshes via water flotation. If the sample came from a fireplace or had any charred material, we processed it using hydrogen peroxide, the very thing used to create blondes in beauty parlors. The British archaeologist who invented the technique told us about it and, in keeping with the plan to use any useful technique, we ordered up two fifty-five-gallon drums of the stuff. The student we assigned to draw some peroxide from the barrel put on a protective rubber apron and gloves and nevertheless soon found that the peroxide was eating through them and his shoes. We thus learned the difference between the highly diluted solution used as hair dye and antiseptic and 100 percent laboratory-grade hydrogen peroxide, which is a component in some rocket fuels and was actually tried out as a propellant for British submarines.