Dr. Andrea Fitting mapping at Meadowcroft, ca. 1975.
In our earthbound arena, the hydrogen peroxide flotation process not only efficiently broke down the clay or silt bonds in sticky sediments, but it sharply reduced the damage to any charred items, meaning, among other things, that we could send much cleaner samples to the radiocarbon labs for dating.
Implementing all these procedures was time-consuming in the extreme and very expensive. Indeed, by 1975, the “summer” field season at Mead-owcroft went from two months (mid-June to mid-August) to four months with six-day workweeks and twelve-hour workdays. People would still be working at the site when the weather turned cold and the leaves began to change in late September. Fortunately, we enjoyed heavy financial support virtually from the outset. This included funding from the University of Pittsburgh and the Miller family via the Meadowcroft Foundation, plus grants from the National Geographic Foundation, the National Science Foundation, the Buhl Foundation, the Leon Falk Family Trust, and several private sources. These not only included a former student of mine from Youngstown State University and a longtime friend, John Boyle, but also his father, Edward (who passed away as this chapter was being written).
The actual research all this funding made possible was tedious, painstaking, highly technical, and very slow moving. In fact, it was as far from the Indiana Jones stereotype of instantaneous and spectacular discoveries as imaginable. Through the twenty-seven years of the entire Meadowcroft project there were very few dramatic discoveries, but instead the agonizingly slow accumulation of minute bits of information, the full significance of which is only now emerging. One of the most important aspects of all we accomplished at Meadowcroft is the often overlooked fact that it provided a nearly unique sequence of human habitation over a period of some 16,000 years.
In retrospect, it is easy to see that this kind of archaeology is not for everyone, and I can fully understand why some archaeology students (and their postmodernist professors) would prefer to debate whether the world exists from the comfort of a cozy bar rather than spend a four-month field season excavating microstrata with razor blades or staring through a microscope at sand grains or flotation samples. It takes a special sort of mentality to engage in the kind of unremitting, tedious labor we insisted on, and it is not difficult to see why the Meadowcroft/Cross Creek project was filled with strange personalities. After twelve hours of troweling thin layers of dirt and dust, you go crazy at night. Running, weight lifting, drinking, fornicating, staring off into space and babbling incoherently— you do almost anything for relief. It takes a truly bizarre person to live that way for months on end, and we had tents full of them.
Whatever the toll all this took on young and eager psyches, the precision we wanted to teach at Meadowcroft—the extent to which one could use the most modern and sophisticated techniques to obtain, refine, and analyze data—paid off in ways that we certainly could not have imagined when we began. For once the pre-Clovis dates came out and some of the profession went into attack mode, we were well armed with the most precise close-focus data ever extracted from a North American archaeological site of this or probably any other time period.
Photo micrograph of sediment sample from lowermost Stratum IIa at Meadowcroft Rockshelter. Sample shows bone fragment (B) and eggshell fragment (E) and also indicates that the sediments were not affected by groundwater percolation.
While the ultimate “payoff” of our high-resolution data-recovery methods in terms of the pre-Clovis/Clovis wars was years away, there wasalso an immediate return. By the end of the second year of the project, we knew that the Cross Creek drainage had not assumed its present form until late Pleistocene times. Specifically, we determined that the actual genesis of Meadowcroft Rockshelter had begun between 40,000 and 23,000 B.P. But, assuming that anyone was around to visit it in these early years, Meadow-croft was probably not a very attractive place to hang out in much before 23,000 to 21,000 years ago. Up until then, it was subject to periodic flooding by a Cross Creek that had not yet cut down to its current bed level. After 21,000 years B.P., the flooding ceased, leaving the site high and dry, and by 20,000 years ago people could have lived there quite comfortably. But this, of course, was also the time when the glacier had reached its most southerly growth, a phenomenon called the glacial maximum. What of the great glacier towering just to the north of Meadowcroft Rockshelter?
Even before all of our geological and other analyses, we knew that forty-seven miles north of Meadowcroft lies Moraine State Park, which marks the southern terminus of the last glacial advance at the height of the last glacial period. We also knew that by 14,000 years ago, the glacier had receded to where the city of Erie now lies, 125 miles to the north. And the presence of the glacier in our general neighborhood would become an enormous and a nearly endless bone of contention once we published those pre-Clovis dates from Meadowcroft. I would drastically underestimate, in fact, how geologically naive many of my colleagues were (and still are in many cases).
At the time, I did not think much about my nearly fanatical drive to make Meadowcroft the best excavation ever, but I realize now with utter clarity that it was me showing Jennings that I could be even better than he at his own game—fieldwork. My dissertation had involved no fieldwork (just the study of collections of materials), but I had stayed the course under his ferocious rudeness, as so many others had not. Maybe I wanted to hear him say something that sounded like approval. He never did. Once, years after the first Meadowcroft data was published, he allowed as how he thought my excavation was “evidently” done with great care, but he never visited the site.
Recently, a friend of mine was talking to an amateur archaeologist in Pennsylvania who said that among those folks and some of my students, I have now become known as The Dark Lord. “Very demanding,” the amateursaid. That came as a big surprise and brought to mind Gandalf's admonition to the hobbit who likened one of his companions to the evil Lord of the Rings. “There is,” Gandalf said with finality, “only one Dark Lord.” I confess I am very demanding, but, compared to Jennings, at least some people think I'm a hell of a lot nicer.
ECO - SQUABBLE AT MEADOWCROFT
When my colleagues were confronted by the state of the natural world in the immediate vicinity of the Meadowcroft Rockshelter beginning 16,000 years ago, the Clovis Firsters simply couldn't believe it. Far from being the barren tundra they expected, what with the proximity of the glacier, the area around Meadowcroft could not even have been characterized as a bo-real or northern forest, though some boreal tree species like spruce were present. In fact, what with the oak and hickory trees and white-tailed deer, it was not all that different from today. Since this evidence from Meadow-croft did not accord with the maps the paleoecologists had drawn up, many of them and their archaeological friends said the radiocarbon dates from the earliest strata at Meadowcroft had to be wrong
What all this told me was that these critics were basically ignorant of glaciers, glacial geology, and even the nature of the paleoecological record, as well as the methods by which paleoecologists gather and interpret data. I also realized they apparently had an unusually hard time simply reading the geological and other reports emanating from Meadowcroft.
If you stand next to an alpine glacier of the sort that persist to this day on mountaintops in the Pacific Northwest, at noon on a cloudless day you can often go shirtless. To insist that a white-tailed deer (or, at least, its black-tailed western counterpart) couldn't live in the vicinity is just silly. To this day, some critics haven't got it through their heads that at the time in question the glacier was not looming up on the nearby horizon, it was more than a hundred miles away. But back in the late seventies, when the pre-Clovis dates from Meadowcroft were announced, some people were pursuing the craft of paleoecology, or, more accurately, examining its conclusions, with a set of built-in blinders. It was simply taken as a given that open land next to a glacier had to be uniformly frozen—underlaid by permafrostand overlaid by bleak treeless
tundra. But all the evidence at that time in the Northeast came from relatively high elevations, which is to say places where it is always colder. Lower down, warmer temperatures are the norm and growing seasons are longer. Today Meadowcroft, which is 853 feet above sea level, enjoys fifty-nine more frost-free days than Pittsburgh, which is only thirty miles north and east but, much more important, 370 feet higher.
Unglaciated sites with late Pleistocene botanical material near Meadowcroft Rockshelter. Note: The map shows the ice front during the glacial maximum, ca. 20,000 B.P.
Evidence of the ecosystems in glacial times was, and still is, derived largely from fossil pollen grains and other plant remains extracted from former bogs and lakes, and in the seventies and eighties very few such sites had been analyzed for data from as far back as 20,000 years ago. (And those were all high-altitude sites.) In 1980, only some eight sites in the entire unglaciated American Northeast had yielded plant material of Pleistoceneage (see map on page 175), hardly a sharp-focus picture for an area of more than 290,000 square miles. By the early 1990s, the picture remained soft. Data from fossil pollen for 15,000 to 18,000 years ago had been obtained from only about twenty-five sites in the entire eastern half of the North American continent, from Texas to Hudson Bay and points east.
To this day, much of the paleoclimatic data from eastern North America consists of pollen grains, and even for palynologists, fossil pollen can be tricky. For one thing, some pollen is more recognizable than others— especially pollen from the coniferous trees such as hemlock, spruce, and pine, which include a great many northern species. Pine pollen is especially resistant to decay. As a result, pine will often be overrepresented in a given sample—both in fact and in interpretation. At the same time, deciduous trees that occur in warmer climes, such as maple and chestnut, have pollen that is particularly subject to decay from microbes and natural oxidation. Pollen is best preserved when there exists a consistent pattern of either wet or dry conditions, of cold, or of acidity. These are all conditions that are scarce in the Northeast. Yet another problem with making interpretations of climate based on such remains is that many kinds of trees do not indicate, or circumscribe, a particular set of climatic conditions. Today's white pine can indicate cool or warm or even boreal conditions, depending on the species, and identification at the species level is often difficult with fossil pollen grains. Beyond all that, the pollen of many tree species is easily caught up by the wind and transported considerable distances. This is an excellent strategy for spreading your kind if you are a plant, but it makes it hard for a palynologist to say that the tree that gave rise to the pollen he has found used to grow nearby.
Pollen analysis, in other words, tends to be better for giving a general and fairly soft-focus climatic picture of a large region rather than for providing details about the local microclimate in a specific locality. Yet even with all of those caveats about the pollen record, and despite a good deal of other evidence to the contrary, many palynologists today still insist on the unbroken, solid, uniform stripe of tundra along the edge of the Laurentide glacier, based on the high-elevation sites they are most familiar with.
The paleoecologist may, however, have other evidence with which to work. In the assessment of ancient ecosystems—preservation permitting—you may recover bigger pieces of botanical evidence in addition to, or even instead of, pollen. At Meadowcroft, pollen is poorly preserved but macro-botanical remains are not. In fact, the site yielded more than 1.4 million items, including everything from seeds to moderately large portions of tree limbs with and without bark. Since most of the bigger stuff is not designed for long-distance travel, it is a pretty fair indicator both of what was growing near the site as well as what humans and other animals were bringing into it. Significantly, while most of the plant remains postdated 10,000 B.P., enough charred remains were present in the lowest levels to show that deciduous trees were present at the time humans first visited the site some 16,000 years ago.
Back in those early days of the Meadowcroft excavations (and to some extent still today), critics sought any conceivable way to shoot down our results. So one early line of attack was that the radiocarbon dates for all this material had to be wrong because such temperate creatures as deer and deciduous trees simply couldn't exist that near the glacier. It was all tundra and/or boreal forest near the glacier in those days, they said. Q.E.D., quod erat demonstrandum. Thus is demonstrated Adovasio's folly. The critics could ignore the elevation factor because the local experts in palynology also ignored elevation. This confusion, by the way, is a hang-up mostly among eastern paleoecologists. In the West, people such as David Madsen at the Utah Geological Survey understood at least two decades ago that elevation plays a key role in localized climate and weather, a bit of Uniformitarianism that remains a principle that it is unwise to overlook. As a matter of fact, the critical role of elevation was thoroughly understood by the 1890s, when C. Hart Merriam, a Yale-trained biologist on a government science expedition to the Southwest, elucidated the concept called Merriam's Life Zones. This showed that, on the mountains and mountain chains of the West, there are essentially stripes of vegetation types as you go higher and higher, ranging from hot desert cacti to boreal trees such as Engelmann spruce to windswept tundra. Moisture tends to increase, and temperature to decrease, the higher you go. Indeed, you can travel the ecological equivalent of two thousand miles from, say, Arizona to Hudson Bay, simply by climbing the mountains ranged east of Tucson where I went to college.
Even so, in the late seventies and early eighties, the scoffers were bayinglike coyotes over what they thought was the dead meat of Meadow-croft.
Enter Kathleen Cushman, a young Pennsylvania woman who had majored in art history at Ohio Wesleyan University, then gotten a master's degree in Old World prehistory from the University of Chicago. In 1976, at Texas A & M, she began to work on a Ph.D. degree in biology, in particular in paleoecology, and even more in particular in the state of the environment just south of the Laurentide Ice Sheet at Meadowcroft, which she knew was “one of the few well-documented archeological sites in the northeastern United States from which several types of fossil botanical material have been recovered.” Five years later, in 1981, she had proved my point about the “reality” of that alleged stripe of tundra tucked up against the ice from the New Jersey shore to the Rockies.
Her comparative analysis of the eight other sites in the Northeast where plant remains had then been analyzed suggested that elevation above sea level might well be as important a factor as proximity to the glacier. High on the Allegheny Plateau, conditions tended then to be more boreal (as they do today). Yet, in all eight sites, there was clear evidence of the presence of some deciduous trees such as hickory, birch, beech, and oak among the pines and spruce. Even at the glacial maximum, in many instances the forest composition was not unlike its modern composition. Elsewhere, she noted, a site from south-central Illinois showed that some 11,000 years ago, oak, elm, and ash were growing amid the spruce a mere fifteen miles from the limit of the glacier. She concluded, in the cool language of science, that the ice's influence “on vegetation may not have been as great as has been previously believed.”
Clearly, then, the countryside along the edge of the glacier must have been a mosaic, a patchwork of different ecosystems. Depending on elevation, local weather conditions, even which way a gorge faced, it varied from tundra to at least partly deciduous woodland. It varied enough to support many kinds of life from mastodons to deer, from boreal owls to songbirds and of course to enterprising humans, who, while probing the inviting tributary to the Ohio River called Cross Creek, would have looked up at one point and noted an inviting rockshelter nestled into the hillside, facing south, sun-dappled, a good place to tarry a while.
But such a picture continued to be anathema to many of my colleagues. As late as 1989 and even on into the 1990s, this confusion of expectations with facts continued. In that year, Dena Dincauze, already a persistent critic of Meadowcroft and card-ca
rrying Clovis Firster from the University of Massachusetts, complained in a review of the then-available scientific literature devoted to Meadowcroft about the “absence of the extinct Pleistocene fauna” in the controversial pre-Clovis level called Stratum IIa and the “presence of hardwood species macrofossils” there. Once again, at least, in Dena's mind, in spite of evidence accumulating to the contrary, there should not be the likes of oak trees and deer so close to the looming edge of the glacier; thus the dates had to be wrong. This is about the same as saying that because Frenchmen usually speak French and I have reported and even tape-recorded a Frenchman saying something in English, he cannot be French. In other words, her complaint is not reasoning in any sense normally ascribed to the process of science. In science, facts are facts, and expectations are continuously adjusted in the light of facts.
Why many eastern scholars have not given up their belief in the map is beyond me. But you can still see it in textbooks despite the fact that Mead-owcroft and other sites have now shown conclusively that it never existed. I recall asking John Guilday once, after he had examined tens of thousands of animal bones from the site, if he was disturbed by the fact that temperate species had turned up at Meadowcroft just after the glacial maximum, and he said he wasn't. Clearly, at Meadowcroft, he said, the transition from Pleistocene to Holocene was something of a nonevent—a near-perfect echo of the conclusion reached by Kathy Cushman with the plant remains.
The First Americans Page 20