The Gault Site

For most students of archeology, "Clovis" conjures up a vision of a distinctive spear point like the one shown on the left and of small groups of "big game" hunters killing Ice-Age elephants as they migrated across North America. For over five decades the Clovis-first hypothesis—the idea that Clovis hunters were the first people to explore the New World—has been a fundamental part of the story of the peopling of the Americas. Clovis peoples with their remarkably sophisticated hunting technology were seen as the first pioneers, highly mobile hunters who walked to North America via the Bering Land Bridge. Once below the Ice Sheet, small groups of Clovis hunters and their families expanded rapidly across the continent killing mammoths so effectively that the species was pushed over the brink of extinction. Or so the standard story goes.

In the past decade, the Clovis-first hypothesis has come under attack from many directions. There have been multiple claims of earlier, pre-Clovis sites in North and South America, none totally accepted, but several very credible. And then there are the early skeletal remains found in North America, such as the controversial Kennewick Man, whose features are said to more closely resemble certain Caucasian populations than later Native Americans. Adding to this, some stone tool experts have made a case for a close resemblance between Clovis technology and that of Late Paleolithic Europe. In the last few years these claims, pro and con, have been reviewed in Time, Newsweek, New Yorker, National Geographic and on the Discovery Channel, among others.

Meanwhile at the Gault site deep in the heart of central Texas, Clovis culture is being reconsidered week by week, midway through a planned five-year dig. The emerging view hardly resembles the Clovis story known to generations of archeology students. Instead of a new group of people exploring an unknown land, we seem to see a people thoroughly familiar with their surroundings. Instead of highly mobile elephant hunters, we see what looks like a full-blown generalized hunting and gathering culture living in the same kind of places and doing many of the same kinds of things that characterized Archaic-era life all across the continent a few thousand years later. This is more than a new spin, this is a whole new way of thinking about what is still, to many, America's earliest recognizable culture.

The evidence presented here is so new that most of it has yet to be reported, at least not in a proper scientific sense. Research at Gault site continues as does laboratory research and reporting. Several of the nation's leading Paleoindian experts and specialists in various subfields and dozens of their undergraduate and graduate students are working on different aspects of the site. In other words, what follows is merely a glimpse of what is to come, a peek behind the scenes and into the head of Dr. Michael B. Collins, the lead researcher who heads up the project. This is what he now (2001) thinks Clovis life may have been like based on the emerging evidence. Like any good scientist, he reserves the right to change his mind. Collins fully expects the next few years will bring even more surprises at Gault and elsewhere that will help paint a more complete and accurate picture of the Clovis past.

Born and raised in Texas, Collins knew in high school that he wanted to become an archeologist (see "A High School Student Discovers Bone"). The Gault Project brings together the three principal themes of his career: lithics, early Paleoindians, and geoarcheology. His interest in the early peoples of the New World started when he first found Paleoindian artifacts around the playa lakes near Midland, Texas, while still a teenager. He studied archeology and geology at the University of Texas and then went on to get a Ph.D. from the University of Arizona. As a stone tool expert he is well known for his work with Tom Dillehay on the extremely early Monte Verde site in Chile as well as a recent book, Clovis Blade Technology (1999, UT Press). His interest in geoarcheology, the marriage of archeology and geology, developed as a natural outcome of his interest in Paleondian cultures. Collins has worked on archeological projects in many states, countries, and continents, none (with the possible exception of Monte Verde) more exciting or important than the Gault site. Being able to investigate a world-class archeological site less than an hour's drive from either his Austin home or his Williamson County farm is a dream come true for "Dr. Clovis."

Ecotones and Endless Flint

To understand why Clovis peoples repeatedly came to Gault and apparently stayed for quite some time, you first have to know where it is and what it had to offer. The Gault site is located in central Texas about 40 miles north of Austin, halfway between Georgetown and Killeen (Ft. Hood). It sits near the head of a small creek in a small wooded valley just at the point where a number of springs come together to form a clear, cool vigorous stream that has never gone dry in historic times. This valley is one of many that cut through the eastern flank of the vast limestone Edwards Plateau that stretches far to the south and west. The Gault site is in the northern part of the Texas Hill Country within the Lampasas Cut Plain, where, as its name implies, the limestone plateau is somewhat flatter and "cut" by many incised stream valleys.

A three-hour walk down the creek valley, the limestone country ends abruptly at the Balcones Escarpment and the Black Prairie begins. This geological fault zone is one of the most impressive ecotones in North America, places where different environments come in contact. Moving east along the creek, almost everything changes in just a few miles—geology, hydrology, soils, plants, and animals.

The Black Prairie, so named for its rich black "gumbo" clay soil, was a grassland for at least the last 15,000 years and still would be today if the grazing, fire control, agriculture, and concrete had not all but spelled its doom. Spanish explorers in the early seventeenth century rode their horses through grass so thick and deep in many places that only the mounted riders could see where they were going. They found plenty of buffalo and antelope to hunt on the Black Prairie but when they turned west and entered the rugged up and down world of the Texas Hill Country, they had to rely on deer and turkey. They found dense, towering bands of hardwood forest along the streams and an oak savannah in the uplands with mixed grasses and trees, many of them stunted and confined to mottes by periodic range fires.

Moving from the grand scale to the local, the Gault site itself sits on a smaller scale ecotone that is still obvious today even to the casual visitor. The road to the site leads through the typical rocky limestone rolling hills with very little soil and lots of cedar (juniper), live oak, mesquite and prickly pear. As you approach the site, the road drops off into the valley—only about 45 feet lower, but what a difference. The deep, well-watered soils provide habitat for huge hardwood trees—burr oaks, walnuts, pecans, ash, elm, bois d' arc, and a dozen more species including willow and cottonwood. In a word, it is lush. While we don't have an accurate idea of what the local vegetation was like in Clovis times, the contrast between valley bottom and the surrounding uplands would have been just as stark.

So the Gault site was located on ecotones, large and small, in a small, protected wooded valley with a spring-fed stream. But it had one other important thing going for it—a nearly inexhaustible supply of extremely high quality flint (chert). The flint occurs as stream-worn cobbles along the creek and it weathers out of the bedrock along the valley slopes and in the uplands surrounding the site. For over 13,000 years people have been picking up flint here and yet modern flintknappers have hauled away pickup loads of it. And still you cannot walk 10 feet without seeing pieces of flint. Some of the larger nodules are the size of a fat watermelon.

They Came To Stay

Most known Clovis sites fall into one of four categories. By far the most numerous are places where isolated finds of Clovis points are made. The next most common are kill sites, places like Lehner and Murray Springs in southeastern Arizona or Domebo in south-central Oklahoma where elephant bones and Clovis artifacts were found together. And then there are Clovis caches, isolated places were Clovis points, bifaces, blades, or blade cores are found in tight piles thought to represent hidden stashes. Finally, there is the rarest category, camps—places where Clovis peoples stayed put long enough for considerable debris to build up. Some camps occur in rockshelters and others in open settings, but most known Clovis camps appear to be the result of fairly brief stays. In contrast, the Gault site is clearly a major base camp, a place where people returned repeatedly and probably stayed for lengthy periods of time.

How do we know? Well for one thing it is a very large site. Imagine a football field. Now add a second one beside it. Now add two more pairs of fields end to end. And this is only the core area of the site measuring about 80 by 300 yards where Clovis materials are known to be concentrated. The entire Gault site covers an area about 90-100 yards wide by about 650 yards long. Not all of this area has been tested yet, but enough to get a fairly good idea of what lies beneath the surface. The evidence is not spread uniformly; some areas have much greater artifact densities than others. And it is also clear that some of the deposits have been washed away by floods—concentrations of Clovis artifacts have been found within gravel deposits along the creek. It is not just large, but incredibly rich. The Clovis deposits average about 40 centimeters (16 inches) thick but are sometimes twice that or more, and in places the deposits contain unbelievably large numbers of Clovis artifacts. Collins guesses that the Gault site may have already yielded as much as 60% of all excavated Clovis artifacts known today.

Base camps, as the name implies, are places where people stayed for a while and ventured out from, repeatedly. One characteristic they have is "assemblage diversity"—lots of different kinds of artifacts. The diversity already recognized at Gault is astonishing. There are many Clovis points—finished points, worn out points, half-made points, resharpened points, and lots of fragments. And Clovis bifaces—big heavy bifaces, small thin bifaces, bifaces broken in manufacture, and several kinds of specialized bifaces. There are also hundreds of Clovis blade cores and blades—large ones, small ones, crested outer blades, thin inner blades, broken blades, and used blades. And then there are the blade tools—end scrapers made on blades, serrated blades, blades with sharp graver-like beaks, and blades with incredible use-wear traces. The finding of several adzes or wood-working tools was quite unusual—this tool form was not known previously from other Clovis sites, although it occurs more commonly at later sites. Another interesting artifact from Gault is a bone or ivory rod, found in the ancient gravel deposits of the creek amid definite Clovis artifacts. It is very similar to specimens found in other Clovis sites.

Among the unprecedented finds at Gault are the incised stones—smallish, smooth limestone rocks and chert (flint) flakes that have various patterns and designs formed by shallow lines almost certainly made with sharp flint flakes. More than 100 of these "mobile art" objects are known from Gault, from early Paleoindian contexts as well as later Archaic-age deposits. The Clovis-age specimens may represent the earliest examples of representational art in North America.

The artifact diversity obviously means that many different tasks were carried out at the site and probably elsewhere by work parties who returned to Gault. The Gault researchers are not ready to enumerate these tasks in much detail, but some patterns are already clear. First and perhaps foremost, the Gault site was a major tool-making locality—a lithic workshop where a great many stone tools were made, most out of flint. All stages of tool-making were carried out from the first stages of "primary reduction" (breaking up large cobbles into usable pieces) to the final stages of putting the finishing touches on a completed artifact. And beyond—there are many resharpened and broken tools at Gault that show people were "retooling"—taking the time to replace broken and worn parts (such as the tips of spears) with sharp new ones. While evidence is less direct, a great deal of wood working, binding, and so on must have taken place there as well. Tiny gravers—very small, delicate stone tools with sharp beaks—hint at some sort of specialized work—scarifying (scratching the skin to draw blood or allow tattoo pigment to absorb) or incising bone, wood, or stone?

Although bone preservation is generally poor in the Clovis deposits, there are enough bones of mammoth, horse, and bison to suggest that these animals were killed not too far away and at least partially butchered at the site. Many tools speak to hunting and butchering—Clovis points, bifaces and sharp blades with meat polish, and heavy choppers probably used to dismember large animals. Endscrapers made on blades suggest that hide working was another typical activity.

One of the most promising avenues for documenting the daily activities of the Clovis peoples who stayed at Gault is through use-wear analysis of stone tools. Using this approach, Gault staff researcher Marilyn Shoberg has already identified three very different tasks that Clovis blades were used for: butchering, grass-cutting, and woodworking. She looks at individual stone tools under 200x magnification using a binocular microscope with polarized light and special eye pieces designed just for this kind of work. At 200x, the edge of a Clovis blade looks like an alien world complete with craters and mountains. These micro-topographic features are the actual surface of the stone tool.

When stone tools are used repeatedly, once-sharp edges become dull and rounded and polish forms. Different kinds of contact materials create different kinds of polish—high polish, dull polish, domed polish and more. And when hard particles, such as sand grains adhering to a hunk of mammoth, come into contact with the polished edges, they leave striations—scratches and gouges. The orientation of these blemishes tells Shoberg which direction the tool was being pulled (or pushed).

The patterns seen on archeological specimens must be compared to experimental tools used for known purposes on known materials. These serve as control or reference samples.

The picture above is the surface of a serrated Clovis blade. The shiny polish is the kind that is associated with cutting meat—butchering. Notice the criss-crossed striations. The triangular pattern left by the intersection of the striations is very characteristic of a tool used repeatedly to cut the meat of a large animal. The different striation directions imply that the blade was held in different ways or at least at different angles.

Now compare this with the first picture on the right showing the magnified edge of a Clovis blade with obvious polish. Notice that most of the surface is completely smooth and the edge heavily rounded. The polish is continuous, smooth, and has numerous small pits and medium-sized comet-shaped pits. This kind of polish is consistent with use in processing plant material high in silicates, and is often referred to as "sickle gloss." The edge (at the bottom of the picture) is heavily rounded, which means that this tool was used for quite some time.

Finally, compare the first two photos with the second (lower) picture on the right showing the magnified edge of an experimental blade used to cut Little Bluestem grass, the kind that formerly covered much of the Black Prairie. After 2000 strokes, characteristic silica polish is just forming in a continuous band along the edge of the blade, and pockmarked areas of polish extend back from the edge. The areas of polish are not as large and continuous on the experimental blade when compared to that on the archeological tool, suggesting that the experimenters have many more thousand strokes to put in. But the characteristic form of the polish on the replica tool, with small pits and linear features parallel to the edge of the blade, is very similar to that found on the archeological tool. This comparison supports the hypothesis that the prehistoric blade was used to harvest and/or process grass or reeds.

Shoberg has just begun to look at the Clovis tools from Gault. She and other researchers will spend hundreds of days staring under the microscope, taking notes and photographs, and comparing archeological tools with more experimental ones. Tedious work to be sure, but the result will be a much more complete view of what Clovis peoples were doing with their finely made stone tools.

How Long Did They Stay?

Among the things that are missing from the Gault site are organic remains, especially charred plant remains. The preservation conditions are such that so far not a single charcoal fragment has been found from the lower Paleoindian layers (although many matrix samples have been saved for further analysis). Charred plant remains could reveal many important clues and they could be used for radiocarbon dating. At present there are no radiocarbon assays of Clovis age from the Gault site. Several lines of evidence suggest that Clovis peoples visited the site over a long period of time and may have stayed here for prolonged periods.

Update: Since the exhibit was created in 2001, the Gault project has obtained a series of infrared stimulated (IRSL) dates from soil samples, a process that can determine when minerals in the soil were last exposed to the sun. The resulting dates nicely match the relative dates indicated by the distinctive artifact styles and give dates of around 13,000 years ago for the Clovis occupations at Gault.

The stratigraphy of the Gault site is very complex. Because it lies within a narrow stream valley with deep deposits, permanent springs, and multiple channels coming together, the character of the deposits can change dramatically over the space of a few meters (6-10 feet). It was a "dynamic" environment, meaning that things could and did change quickly. Major floods, for instance, changed the stream course repeatedly, ripping up old deposits and creating new ones. As the continuing excavations connect now-isolated excavation units and as various geological and soils experts study the many samples that have already been taken, entire books will probably be written about the stratigraphy of the Gault site. But based on what we know today, there are at least three distinct Clovis "components" at Gault.

That is, in various places in the site there are at least three distinct layers containing Clovis artifacts. Their nature implies that these formed over extended periods of time, decades at least and probably centuries. In the lowest and hence earliest Clovis deposits, there are the bones of mammoth, horse, and bison, all species that became extinct at the end of the last Ice Age. In the later, upper two deposits the only large bones that have been found are those of extinct bison. Given this, it is possible that mammoth and horse became extinct (at least locally) during the Clovis era at Gault. It is obvious that a considerable period of time elapsed during the Clovis occupations at Gault, several hundred years at least and probably more. Based on the IRSL dates, the Gault site could have been occupied as early as 12,000 B.C. and as late as 10,900 B.C.

Textbooks will tell you that Clovis points were specifically designed for mammoth hunting. Yet, in the upper (and latest) Clovis deposits at Gault it appears that no change in weaponry was made after the extinction of the mammoths. The Clovis points are still the classic form and they are found with bison remains. Collins thinks this is another indication that Clovis technology was a generalized rather than specialized one. The prevailing concepts are ripe for reconsideration.

This still leaves the question of the nature of the occupations: were these intermittent, as seems most likely, or continuous? Researchers aren't sure yet and may never be certain, but there are indications that the occupation(s) may have been lengthy. The most obvious is the sheer quantity of materials and size of the site—this must have taken either lots of visits or a fair number of people over lengthy spans of time.

Another clue is the amount of exotic lithic materials. One of the most characteristic aspects of the stone tools at most Clovis sites is that the discarded, worn-out-stone tools are more often than not made of exotic non-local materials. Many studies have shown that Clovis peoples routinely carried or traded flint and other stones hundreds of miles from their sources. But at Gault, exotic materials are very uncommon, even among the worn-out tools such as Clovis points that were broken and dulled and resharpened until they were too small to be useful. Although no counts are yet available, the overwhelming majority of all stone tools and tool-making materials at Gault are made of the local flint. This strongly suggests that Gault hunters commonly started and ended their trips at Gault and that they did this throughout the history of the site.

Putting It All Together: Clovis Reconsidered

Based on the evidence just reviewed and on arguments and data presented by many other researchers in recent years, a new view of Clovis culture is taking hold. Clovis peoples may not have been the pioneers who first settled North America. Leaving aside the direct evidence for preClovis sites, there is this: Clovis artifacts are known from all 48 of the lower states plus southernmost Canada, Mexico, Costa Rica, Guatemala, and northern South America. These continent-wide localities occupy a tremendous range of environments, from coastlines to mountains and almost everything between. The inescapable conclusion from these facts alone is that most Clovis peoples were NOT highly mobile, specialized mammoth hunters—they were generalized hunter-gatherers who must have relied on animals of all sizes and a great many plants. The emerging data from the Gault site lend very strong support to this interpretation.

In addition to what has already been mentioned, here are several more examples of the kinds of evidence from Gault that reinforce this view. Among the bones found in the Clovis deposits are turtle bones, burned frog bones, burned bird bones, and small mammals yet to be identified. In Clovis faunal assemblages across North America, the most commonly identified animals are not elephants—they are turtles. And the Clovis diet was not based on animals alone. This, of course, is obvious anyway because humans can't live for long on just meat. But at Gault, use-wear studies are already finding evidence of a wide range of contact materials including the stunning example of the Clovis blade with the highly developed use-wear signature of grass cutting. While grass-cutting may not have been for food (the tall grasses of the Black Prairie are ideally suited for thatching and bedding), it is another indication of the diversity of behaviors that are being documented at the Gault site.

It should not seem surprising at all that Clovis peoples were more than one-dimensional. They were, as we are learning at Gault and elsewhere, much more interesting people who adapted to a wide range of environments and climates and behaved like generalized hunters and gatherers of the later Archaic cultures of North America. Contrast this with Folsom culture, the quintessential specialized big game hunters of the Plains. All known Folsom sites occur within or near the Great Plains and prairies of the midcontinent. And all of the Folsom sites with animal bones have extinct bison bones. No Folsom caches are known, suggesting perhaps that Folsom peoples were so mobile and so focused on "encounter" hunting strategies, that they did not plan ahead by caching materials for the future. Folsom peoples did not make much use of the prismatic blade technology, perhaps because blade cores are heavy things that don't travel well. The ultra-thin Folsom bifaces are light and the thinning flakes struck from them well suited for making Folsom points and other artifacts. Clovis is anything but Folsom-like.

For now, this is where we leave the story of Gault and of Clovis reconsidered. Clovis culture will never be seen again in the same light, as it has for so long. The Clovis-first hypothesis—with all its implications of specialized big-game hunters—is all but shattered. While it hasn't been proved beyond doubt that preClovis peoples existed in North America, Clovis peoples weren't anything like those described by the long-held ideal. If they were the first pioneers, as many archeologists still believe, they were exceptionally adaptable and extremely fast learners (not to mention prolific breeders) for whom killing mammoths was just one of many successful strategies. The results of work at the Gault site in central Texas will help lead researchers, students, and the public to a much more sophisticated and accurate understanding of Clovis culture.