free trial issue                                             subscribe                                                  back issues

Athena Review, Vol.3, no.2: Peopling of the Americas 

The Gault Site, Texas, and Clovis Research

Michael B. Collins

University of Texas at Austin

Humankind evolved somewhere in eastern Africa, conservatively, some two million years ago. With tools and an enlarged brain, humans began to expand into new territories and new environments, eventually filling most of the African continent and much of Eurasia (Schick and Toth 1993). By 40,000 years ago, modern forms of humans occupied most of the habitable parts of Europe, Asia, and Australia (Lourandos 1997). This remarkable history of human “adaptive radiation,” as the process is known, lacked only a final chapter - the peopling of the Western Hemisphere. When, whence, by whom, along what route(s), under what conditions, and with what cultural baggage were North and South America first colonized? This is one of the most engaging questions facing archaeologists, anthropologists, linguists, human biologists, geologists, paleontologists, paleogeographers, and paleoclimatologists, and their efforts of late have created a lively and exciting research atmosphere. At the crux of their inquiry is an intriguing prehistoric culture in North America known as “Clovis” (Stanford 1991). Long believed to have been the first culture in the Americas, Clovis is the benchmark against which all empirical and theoretical accounts regarding the arrival of people in the New World are measured. While many of us are engaged in the search for evidence of earlier, “preClovis,” archaeological manifestations, I am suggesting that careful scrutiny of the benchmark itself is overdue.

Recent research has revealed that Early Paleoindian cultural history in the south central US was more complex than previously thought and there are indications that the prevailing paradigm on the origin of New World cultures is in need of revision. No single site is responsible for these developments, but the Gault Site (fig.1) in Central Texas certainly weighs in with multiple contradictions to current theory.  

Gault is a large site, more than 800 m long and 200 m across (~16 ha) with abundant archaeological evidence for human occupations spanning the entire local prehistoric record of 11,000 RCYA. It occupies the constricted head of the valley of a small stream where reliable springs flow and abundant chert of extraordinary quality crops out. Today the locality is well watered and supports a diverse array of trees and other vegetation on deep soils in stark contrast to sparse xeric vegetation on thin, rocky soils on the immediately surrounding uplands. At a larger scale, this setting is in the Balcones Ecotone (a transition zone between two distinct habitats) where resources of limestone uplands mingle with contrasting landscapes occurring on adjacent coastal plains (fig.1). The Edward’s Plateau differs in its geology, soils, flora, and fauna from the Black Prairie region of the Gulf Coastal Plains. Gault today is a special place and evidently has been for a very long time.

[Fig.1: Physiographic map of Texas showing proximity of Gault to the rich environmental overlap zone (the Balcones Ecotone) between prairie and savanna habitats of the Coastal Plain and the Edward’s Plateau (Gault Archeological Project).]

In Central Texas, the archaeological record of the Paleoindian period consists primarily of chipped stone artifacts which are associated with preserved faunal remains at some localities. Paleoindian sites rarely have organized features, and burned rocks are extremely scarce. Beginning around 9,000 RCYA an abrupt shift occurred and sites post-dating this shift are characterized by enormous quantities of fire-cracked rocks and a variety of chipped and ground stone implements (Collins 1995). This marks the beginning of a long Archaic record that resulted from using heated stones for cooking in earth ovens. A variety of plant and animal remains are found preserved in some of these sites. Extensive early deposits containing Paleoindian artifacts and lacking burned rock remain almost undisturbed at Gault beneath mostly disturbed deposits with Archaic artifacts and copious amounts of fire-cracked rocks.

Major Archaeological Components at the Gault Site: The dates accompanying the following periods refer to the chronology known for Central Texas.

Late Prehistoric Period (1,200-400 RCYA): Regionally, a shift to bow and arrow technology followed by adoption of ceramics marks the beginning of the Late Prehistoric period, in Texas roughly 1,200 RCYA. This period is minimally represented at Gault by a few arrow points and by a couple of pottery sherds. No intact deposits containing isolable Late Prehistoric archaeological materials have been found. As Europeans moved into the region some 400 years ago, Native American lifeways underwent catastrophic changes and ultimately disappeared in less than two and a half centuries. Petroglyphs on the valley wall at the Gault Site may date from the tumultuous time when Europeans and Native Americans vied for control of the region. These meager remains do not present compelling evidence, but do suggest that hunting and gathering remained the primary subsistence strategy of Late Prehistoric and Historic period occupants of Gault, perhaps until the eighteenth or nineteenth century AD.

Archaic Period (9,000-1,200 RCYA): Our investigations have found that most of the Archaic-age deposits at the Gault Site have been completely disrupted. The soil is disturbed and we commonly find cans, bottles, and other modern trash throughout the midden. Early photographs and present day remnants indicate that these deposits were once at least 12 or 15 hectares in extent and from 0.1 to 2.5 meters in thickness. Artifacts diagnostic of the several style intervals of the Archaic - mostly fragmentary dart points - found in the disturbed deposits indicate use of the locality throughout the 8,000 or so years of the Archaic. Artifacts overlooked by looters include bones of various animals as well as tools indicating such activities as hunting, woodworking, plant-food processing, and production/maintenance of various kinds of perishable material culture. This was not a site devoted to a narrow set of specialized activities. In some excavated areas, we find a thin but intact layer of very early Archaic artifacts (dating to between 8,000 and 9,000 RCYA overlying earlier deposits of Paleoindian age. This intact layer is an excellent indication that no recent disturbance has befallen the deposits below. We know from investigations at other Archaic sites in the region that well-adapted hunter-gatherers created the middens at sites like Gault. Thus, even in their disturbed state, these Archaic deposits are prima facie evidence that the ecotonal setting of Gault was ideal for hunter-gatherers throughout its history of use (i.e. one can extrapolate back from the Archaic to the Paleoindian Period). Hunter-gatherers select such settings as part of a strategy of positioning themselves on the landscape where the greatest diversity of floral, faunal, and mineral resources occurs in the greatest proximity.

Paleoindian Period (>12,000-9,000 RCYA): Below the Archaic midden in most, but not all, areas of the site are deposits and soils containing modest numbers of stone artifacts, including distinctive styles of projectile points, diagnostic of several intervals in the Late Paleoindian period. Age estimates for these materials are extrapolated from elsewhere and span the interval between about 10,000 and 9,000 RCYA. Although these deposits are relatively undisturbed, their yield of information has not been great thus far. There are numerous items from Gault that appear to fit within the Late Paleoindian, but are not previously named or recognized forms. At present, we know little more than the existence of Late Paleoindian remains at Gault, but future investigations can be expected to increase our knowledge significantly.

[Fig.2: Folsom age (10,900-10,200 RCYA) artifacts from Gault include projectile point types Midland (a,c-e) and Folsom (h), Midland point manufacturing failures (b), Folsom point manufacturing failures (f-g), and ultra thin bifacial knives (i) (photo: Gault Archeological Project).]

Two style intervals are well established for the Early Paleoindian subperiod of the region, later Folsom and earlier Clovis, and evidence of both is found at Gault. Folsom is technologically distinctive and represents a specialized, nomadic bison hunting lifeway of ca. 10,900 to 10,200 RCYA (around 12,900-11,000 calendar years ago). Folsom sites are found in areas that were in, or immediately adjacent to, grasslands at the time and faunal remains, if preserved, always include bison bones. The Folsom tool kit of stone is light in weight but afforded lethal weapons, efficient butchering/meat processing tools, and effective scrapers for hideworking. Its purposes and its portability bespeak a nomadic hunting way of life. Gault is not an ideal locality for bison-hunters, but it is close to the prairies of the Gulf Coastal Plain and afforded such resources as high quality tool stone so important to specialized hunters. Not surprisingly, Folsom remains at Gault are relatively sparse and are dominated by debris from the manufacture of stone tools (fig.2) and exhausted or broken points discarded at the end of their usefulness.

In contrast, Clovis is abundantly represented at the site. Several hundred thousand pieces of stone, bone, ivory, and teeth from Gault can be attributed to the Clovis interval (11,200-10,900 RCYA; 12,900-12,550 cal BP). Most are debris from stone tool manufacture, but a diverse array of tools (figs.3,4) occurs as well, along with bones of several kinds of animals. Clovis is pivotal in New World prehistory and this component at Gault is the focus of our investigations.

[Fig.3: Clovis points at Gault manifest a wide array of conditions and forms. Some were never finished (j), others were used and broken (a-d,h,k), and still others were extensively resharpened (e-g,l); lanceolate (a-c,g-h), recurvate (d-f,i-j), and triangular (k-l) outlines were in use (photo: Gault Archeological Project).]

In a few areas of the site, excavations have revealed small numbers of artifacts in strata beneath well-defined layers of Clovis artifacts (fig.5). It is not clear at this time whether the underlying materials are early and sparse Clovis manifestations or if they represent a human presence at the site prior to Clovis. Seeking to answer this question is of paramount interest in our investigations.

The Gault Site and the “Clovis-First” Paradigm: Dates for Clovis occupations across North America fall primarily in a three to four hundred year interval, near the end of the last glaciation (Haynes 1992, 1993). Clovis is found all across North and Central America from the Pacific to the Atlantic and from the southern fringe of Canada to Costa Rica. At numerous sites (notably areas of Blackwater Draw, Colby, Dent, Domebo, Lehner, Lange/Ferguson, Miami, Murray Springs, Naco, and Sheaman), Clovis assemblages are dominated by projectile points found in association with skeletal remains of mammoths. Finds such as these gave rise to a theory that people who used Clovis points were specialized mammoth hunters who continuously moved across the landscape in search of herds. According to the theory, their weaponry and skill were so effective that they were at least partly responsible for the extinction of mammoths at the end of the Pleistocene. Excavators of these and other Clovis sites have found virtually nothing archaeological underlying any Clovis deposit.

[Fig.4: Clovis chipped stone tools at Gault include choppers (a), adzes (b), bifacial knives (c), end scrapers on blades (d), gravers on blades (e), and serrated blades (f) (photo: Gault Archeological Project).]

Because there seemed to be no older cultural remains in the Western Hemisphere, Clovis came to be considered the archaeological signature of the founding population in the Americas. And, because the early dates for Clovis coincided with two remarkable glacial phenomena, a plausible explanation for the origin of this founding population emerged and soon became entrenched. So much of the Earth’s hydrosphere was locked up in glaciers and ice sheets that oceans worldwide were lowered by 100 meters or more. A broad isthmus emerged from the floor of the Bering Sea and connected what are now Siberia and Alaska with a land bridge known as Beringia. Around 12,000 to 11,000 RCYA (13,600-12,550 calendar years ago), Beringia was almost ice-free. It was also thought there was a narrow “ice-free corridor” along the Mackenzie River valley between the Cordilleran glaciers of western Canada and the vast Laurentian ice sheet that covered most of the rest of Canada. (This theory, however, has been challenged by recent finds in both geology and cave fauna; see Dixon, this issue). Some 11,500 years ago, Asians are theorized to have walked out of Siberia, across Beringia, down the purported ice-free corridor, and out onto the north central Great Plains to become the Clovis mammoth hunters and the progenitors of the American Indians. In a few generations, these colonists quickly spread out to fill much of the Western Hemisphere all the way to the southern tip of South America.

Gault is one of several Clovis sites along the Balcones Ecotone in Central Texas (Meltzer and Bever 1995). Each of these sites is near good springs at outcrops of abundant, high-quality chert, and is strategically situated in relation to diverse floral and faunal resources. The sophisticated articulation between humans and their landscape seen in this local pattern is repeated at hundreds of local analogs across the continent. This pattern closely matches those of the Early and Middle Archaic pattern that followed (cf. Johnson 1991, 1995; McKinney 1981), and strongly suggests a generalized hunting and gathering adaptation.

[Fig.5: Chipped stone artifacts from layers below well defined Clovis deposits at Gault; it is not yet known whether these are earlier Clovis or preClovis cultural remains (photo: Gault Archeological Project).]

A second critical flaw in the Clovis-first paradigm is emerging from careful study of Clovis subsistence. Many early discoveries of Clovis sites (1930s through 1950s) occurred because someone noticed large bones and investigated. As often as not, these turned out to be kill sites. Other less eye-catching Clovis sites have been found and investigated, especially in recent decades. These have shown that the Clovis lifeway was much more generalized with evidence for exploitation of diverse animals, large and small. Much less diet breadth is seen for the later, more specialized bison hunters (such as Folsom).

Frogs, birds, and small mammals were on the Clovis menu at Gault, along with horse, bison, and mammoth. These hints of diet breadth at Gault are all the more significant given the very adverse conditions for bone preservation at the site and the nascent stage of our faunal analyses. Besides the exploitation of diverse fauna, Clovis inhabitants at the Gault Site engaged in a variety of activities using their stone tools, including digging,  working wood (fig.4b), and cutting grass as indicated by microscopic wear patterns.

In the succession of Clovis components at Gault, there is an interesting change in the remains of large fauna. Lower deposits contain bones of mammoth, horse, and bison whereas higher ones have only bones of bison, suggesting that the Clovis interval at Gault spanned the extinction of horse and mammoth in the local region. If this is true and if the theoretical proposition is correct that Clovis weaponry is specialized for the taking of mammoth, then there should be a change in technology at the point in time when mammoths disappear from the regional fauna. No such change has been perceived in our studies.

A third flaw seems to be the emphasis current theory places on the mobility of Clovis hunters. Specialized hunters of big game must move as needed in response to herd movement or depletion in order to reliably obtain  their prey. Because of the mammoth’s large size and slow rate of reproduction, any hunters who were largely  dependent upon them would likely deplete local herds in a short period of time unless they had developed a highly  effective way of preserving meat over long periods of time. Otherwise, as the theory acknowledges, they would have to move often.

Evidence for high mobility permeates the existing Clovis archaeological record. Sites are often small and artifact numbers are low in most Clovis components. Clovis peoples were moving tool stone great distances and virtually  every Clovis site yields multiple examples of exotic stone. However, it is not clear how this movement occurred  (Meltzer 1989). Competing hypotheses call for highly mobile hunters frequently making tools at one place and  disposing of them in an exhausted state at another place, exchange between groups, or long-term retention of  favored exotic stone among less mobile groups. Interest in this problem is acute because of the remarkable  distances stone traveled. For example, a Clovis point made of obsidian originating in the state of Querétaro in  Central Mexico was found at Kincaid Shelter in west Central Texas, a distance of 1,000 km (Hester et al. 1985).  Distances of 200 to 500 km are commonly indicated (Meltzer 1989).

At Gault is found one of the largest of the known Clovis occupation areas, covering at least 3 ha (and probably more) within the larger site. Debris is abundant at places in the Clovis component, sometimes reaching 200 pieces  per 5 cm level in a 1m by 1m excavation area. With less than one percent of the area excavated, we have recovered  some 300,000 Clovis pieces. Clovis people were at Gault repeatedly over a significant period of time as indicated  by the average 30 to 40 cm thickness of Clovis-age deposits. Direct dating has not been possible for the Clovis  component at Gault, but the deposits and their contents indicate that considerable time is represented. Much of  the natural deposition that encased the Clovis materials is of clay that resulted from low-energy, over bank  flooding of the nearby stream. Typically in Central Texas this mode of deposition builds up deposits at an average  of less than a millimeter per year, which provides a very crude indication that 400 or more years could be represented by the host deposits at Gault.

As mentioned, remains of mammoth and horse disappear in the upper levels at Gault, tentatively indicating that these two animal forms became locally extinct during the Clovis tenancy. At the very top of the Clovis section  are also found artifacts diagnostic of the Folsom archaeological interval that succeeds Clovis in the Plains and Southwest. These facts suggest that the entire temporal span of Clovis could be represented at Gault. This is a  finding inconsistent with a model of near constant mobility.

A fourth flaw is the existence of sites in North and South America that indicate a human presence prior to Clovis. These include Monte Verde in southern Chile with a radiocarbon age of ca. 12,500 years ago,  Meadowcroft in Pennsylvania with stone artifacts dating at least as old as 14,000 RCYA, Cactus Hill in Virginia  with ca. 15,000 year old non Clovis artifacts underlying a Clovis component, and burned and cut bones at Cueva Quebrada in Texas with wood charcoal dated between 14,000 and 12,800 RCYA. There are some possible shared  traits of lithic technology between Clovis and the assemblages at two of these early sites.    

Small, lanceolate, bifacially-flaked, unfluted projectile points and small prismatic blades are characteristic of the preClovis components at Meadowcroft (PA) and Cactus Hill (VA). The points are not Clovis points and the small blades are not Clovis blades, but at a general level of comparison, it can be said that bifacial, lanceolate  projectile points along with prismatic blades are at the center of the Clovis lithic technology as well as that of at  least two preClovis sites. The significance of this general comparison is yet to be ascertained.  

[Fig.6: Clovis prismatic blade cores (a-d) and blades (e-k) are abundant at Gault (photo: Gault Archeological Project).]

There is also an intriguing history of mammoth-bone flaking in Clovis as well as preClovis sites in North America. Flaking bone requires a technique similar to that used in flaking stone where the right amount of force  has to be applied at the correct angle to the correct place. Very hard objects (stones, presumably) were used to deliver considerable force to a small, precisely chosen spot on each of the flaked bones, creating a fracture similar to that of such stones as chert. Bone, also like stone, could be flaked by high-energy forces in nature. A critical  factor in all of the sites discussed below is an absence of any evidence for natural forces capable of flaking mammoth bone, leaving humans as the most likely agent.

Excavations at Owl Cave (Idaho) and Lange/Ferguson (South Dakota) recovered Clovis chipped stone artifacts associated with percussion flaked mammoth long bones; a mammoth sacrum bone was also flaked at  Lange/Ferguson. A disarticulated mammoth skeleton at the Lindsay site (Montana) was buried in a deposit of  loess (wind blown silt). The gentle settling of wind blown dust on this skeleton left delicate bones such as ribs intact. Bones were stacked in an unnatural way and several were fractured while green; nine pieces of hard sandstone were also present, but no artifacts diagnostic of Clovis were found. The Lindsay mammoth died ca. 11,200 RCYA, well within the Clovis time interval.

Two sites in Nebraska (Jensen and La Sena) and one in Kansas (Lovewell) are similar in that disarticulated mammoth skeletons were found in deep loess or fine alluvial deposits. At each site, some bones were in unusual  positions, delicate bones such as scapulas and ribs were intact, and long bones were fractured from powerful  blows with a hard object. No stones of any kind were found. Radiocarbon ages for these three sites are ca. 14,000,  18,000, and 18,000 RCYA, respectively. Duewall-Newberry (Texas) is an undated site with similar evidence in a  low-energy flood deposit of the Brazos River and Cooperton (Oklahoma), dated to ca. 17,000 RCYA, is also  similar except that it rests in a higher-energy gravel (though evidently not one capable of flaking dense mammoth bones).

A long North American tradition of flaking mammoth bone seems to be represented by this evidence. And, clearly that practice continued into Clovis times.  

Another recent development in Clovis investigations is discovery of Clovis components that predate earlier estimates for the Clovis time interval of between 11,200 and 10,900 RCYA.

As originally conceived, the Clovis-first theory presumed that the first inhabitants of the Americas were similar to modern northern Asians since they came out of Asia and were ancestral to later Native Americans. This is not  supported by the physical characteristics of the earliest human skeletons in North and South America who bear  features of southern Asians (and, occasionally, even Europeans;see Brace, this issue).

Finally, a premise upon which the Clovis- first model rests is that there had to be a land bridge from the Old to the  New World in order for the first humans to make the journey (“A land bridge is regarded as being the first necessity” [Coles and Higgs 1969:419]). This presumption excludes water travel from consideration when investigators evaluate alternative models for the peopling of the Americas.

In sum, at Gault and at many other Clovis sites and a handful of preClovis sites is found evidence that is inconsistent with the prevailing, Clovis-first paradigm for the peopling of the Americas. So what replaces this  theory?

Clovis Origins: One of the perplexing questions regarding Clovis is, “what are its origins?” Nowhere along the theoretical route out of northeastern Asia, across Beringia, and down the ice-free corridor is there known to be a  clear antecedent culture to Clovis. Nor has one been found within the Clovis realm. Either a yet to be discovered (or recognized) culture ancestral to Clovis existed somewhere in Eurasia, or Clovis developed out of roots in the  New World. Or, perhaps most likely of all, Clovis was a vigorous hybrid of cultures and technologies.

[Fig.7: Fragments of engraved stones recovered from Clovis deposits at Gault in 2000 and 2001 (photo: Gault Archeological Project).]

There are many traits in Clovis that are shared with Upper Paleolithic cultures of Western Europe (Collins n.d.), most notably Solutrean (21,000 to 16,000 RCYA) and Magdalenian (16,000 to 11,000 RCYA) (Boldurian and Cotter 1999:95-97; Bordes 1968:213-219; Haynes 1982, 1987; Wilmsen and Roberts 1978:99-100, 180). Except for the youngest phases of the Magdalenian these cultures significantly predate Clovis and they are geographically far removed. Furthermore, the shared traits do not form a coherent entity but appear rather widely spread in Paleolithic time and space. Nonetheless, some of the similarities are great enough to stimulate speculation over possible historical connections. (1) Clovis and Solutrean bifaces are flaked in remarkably similar fashion and there are lanceolate, bifacial points in both [fluted in Clovis, not fluted in Solutrean]; (2) Clovis blades and blade tools are similar to those of both the Solutrean and the Magdalenian; (3) Clovis bone and ivory rods, some pointed and some not, resemble bone and ivory pieces in both the Magdalenian and the Solutrean; (4) a single example of a perforated baton or shaft wrench in Clovis has counterparts in both Solutrean and Magdalenian; (5) another single artifact in Clovis with Magdalenian equivalents is a barbed harpoon tip; (6) portable art objects in the form of small engraved stones is a trait shared between Clovis and the Magdalenian (figs.7 ,8); (7) and, finally, in the Magdalenian and probably in Clovis, ochre was used as an ingredient in the adhesives that held bone or stone objects in their wooden hafts. These shared traits do not constitute a case for direct historical connection between Clovis and the western European Upper Paleolithic, but they do suggest that at least some elements of Clovis technology have remote roots in the western European Paleolithic tradition as do some of the cultures of the Russian Steppe.

There are small blades and unfluted, lanceolate bifacial points at Meadowcroft, PA and Cactus Hill, VA, two North American sites predating Clovis. Dates for these materials at Meadowcroft are near 14,000 RCYA; at  Cactus Hill, dates are closer to 15,000 RCYA (16,300 and 16,950 cal BP, respectively). A succession of relatively  minor technological transformations from the Meadowcroft and Cactus Hill points and blades could result in the  larger blades and fluted lanceolate points of Clovis. The findings at these and other preClovis sites are not yet  robust enough to be universally accepted, so it is too early to say that they represent a New World origin for  Clovis, but this is a hypothesis that will undergo much scrutiny as time goes on.

The Search for preClovis: Some prehistorians argue that, given the amount of archaeological activity in the Americas (especially North America), any preClovis archaeological manifestation would have been found by  now if it existed (Jelinek 1992). A number of factors severely limit the confidence we can place in this position, as  examples from the South Central part of the United States clearly illustrate.

[Fig.8: Engraved limestone pebbles (a-d) found at Gault in 1990 by D. Olmstead (photo: Gault Archeological Project).]

Probably the most significant factor limiting what we know and are likely to know of preClovis is the nature of that record. Simply put, we do not know what we are looking for. In all likelihood, any peoples in the New World prior to Clovis were generalized hunters and gatherers who left meager evidence of their existence. The search for preClovis sites, then, is a search for an archaeological signal of  low-visibility in secure, datable geologic contexts. Given this objective, how do the practices and procedures of archaeologists measure up?

Many archaeologists working on and near the Southern Great Plains, the region I know best, like a majority of their colleagues in North America, lack training in the earth sciences and are only superficially conversant with  Quaternary stratigraphic and pedogenic (soil formation) evidence for the age, integrity, and depositional context  of deposits and landforms relevant to archaeological inquiry.  

Much archaeological investigation here as elsewhere is done as a result of mandated “cultural resource management” where areas to be affected by public-sector land modifications are surveyed for sites, sites are  evaluated for significance, and steps are taken to mitigate impacts to the more significant sites. Relatively junior  and inexperienced crews conduct a majority of the surveys. Members of these crews often have little knowledge of  Quaternary geology. Criteria for site significance are bureaucratic contrivances that tend to favor large, highly visible sites.

Finally, besides being of low visibility, preClovis sites must have been few in number and not all of them have survived the unrelenting forces of nature. Of all of the sites that existed 12,000 years ago, a significant number  have been lost to erosion by wind and water.

Yes, we may have done a lot of archaeology in North America, but no, we have not eliminated the chance that a preClovis record exists. Low site density, low site visibility, no known distinctive artifacts, and a common lack of  appropriate qualifications on the part of a majority of archaeologists combine to produce a very low chance that  preClovis sites are being discovered. Even when one is discovered, the odds are against its significance being recognized or any further work being done.  

Remedies: Our quest is for the best possible archaeological account of the earliest peoples in the Americas. If Clovis is, in fact, the earliest cultural evidence, we need to have confidence in that finding. If there were people  here before Clovis times, unequivocal evidence of that fact is needed. Regardless the outcome of our quest, our search procedures must be improved. This should begin with improved awareness of the issues and data  requirements in a broader sector of practicing archaeologists. We need to vacate the assumptions and expectations derived from the Clovis-first model. Cultural resource management criteria for site significance need to be  expanded to include ephemeral sites in geologic contexts indicating Late Glacial ages. There are increasing numbers of professional archaeologists in the Americas with backgrounds in the earth sciences, but many more are needed. As more qualified investigators emerge, there should be more targeted searches conducted where the emphasis is careful scrutiny of intact deposits of Late Glacial age. With these changes in place for a number of years, we should be able to say with meaningful confidence what the early archaeological record of the Americas is really like.

Conclusions: Evidence at the Gault site significantly increases a growing body of data that are discordant with theClovis-first model. The size, density, and thickness of Clovis deposits are in conflict with the notion of high mobility. Specialized hunters of mammoths would have little reason to frequent a locality like Gault, whereas it is ideally suited to the needs of generalized hunter-gatherers. The array of activities indicated by the diverse tools and their use wear contradicts specialization.  

In my view the archaeological manifestation that we call “Clovis” is more complex than our usual accounts would suggest. Clovis occurs in diverse habitats and some of the more significant sites, such as Gault,  Thunderbird, and Aubrey, have the distinctive signature of a generalized adaptation rather than a specialized mammoth hunting lifeway. Clovis is too common, too widespread, and too well adapted in too many environments  to satisfy the theoretical expectation of the first colonizers in the New World. There are also numerous  archaeological indicators of an earlier human presence. Comprehensive and objective review of the empirical and logical bases for the Clovis-first theory is in order, and I predict that it will be replaced with a significantly different account of the peopling of the Americas.

Clovis looks to me like a technocomplex - a constellation of technologies shared by multiple ethnically distinct peoples over a wide area. Such a configuration could have come about if one or more technological innovations were introduced into a North America sparsely populated by groups of successful hunters and gatherers, each intimately familiar with their own environment. What to archaeologists 12 millennia later looks like a widespread expression of a single culture may, in fact, have been a relatively superficial set of shared material traits employed  by groups who spoke different languages and lived by different cultural codes. The speed with which this thing we call Clovis spread across the continent is easier to envision if it were spreading among many peoples who already knew the lore of their many homelands.

These ideas may be proven wrong. I only ask that this be thoroughly demonstrated, not assumed or asserted.

[Note: This is an abridged version of the article, whose full text and illustrations appear in the printed issue of Vol.3, no.2 of Athena Review.]


Adovasio, J.M. 1993. “The ones that will not go away.” In O. Soffer and N. Praslov (eds.) From Kostenki to Clovis. pp.199-218. Plenum, NY.

Amick, D.S. 1999. Folsom Lithic Technology. Ann Arbor, International Monographs in Prehistory.

Anderson, A.D. (ed.). 1975. “The Cooperton Mammoth: An Early Man Bone Quarry.” Great Plains Journal 14(2):130-173.

Banks, L.D. 1990. From Mountain Peaks to Alligator Stomachs: A Review of Lithic Sources in the Trans-Mississippi South, the Southern Plains, and Adjacent Southwest. Oklahoma Anthropological Society, Memoir 4.

Bement, L.C. 1986. Excavation of the Late Pleistocene Deposits of Bonfire Shelter, Val Verde County, Texas. Texas Archeological Survey, Archeology Series 1. Univ. Texas at Austin.

Bird, J.B. 1938. “Antiquity and Migrations of the Early Inhabitants of Patagonia.” Geographical Review 28:250-275.

Bird, J.B. 1969. “A Comparison of South Chilean and Ecuadorean ‘Fishtail’ Projectile Points.” The Kroeber Anthropological Society Papers 40:52-71.

Black, S.L., W.L. Ellis, D.G. Creel, and G.T. Goode. 1999. Hot Rock Cooking on the Greater Edwards Plateau: Four Burned Rock Midden Sites in West Central Texas. Studies in Archeology 22, Texas Archeological Research Laboratory, University of Texas at Austin and Archeology Studies Program, Report 2, Environmental Affairs Department, Texas Dept. of Transportation.

Boldurian, A.T. and J.L. Cotter. 1999. Clovis Revisited. Philadelphia, University of Pennsylvania Museum.

Bordes, F. 1968. The Old Stone Age. New York, McGraw-Hill.

Broster, J.B. and M.R. Norton. 1993. “The Carson-Conn-Short Site (40BN190): An Extensive Clovis Habitation in Benton County, Tennessee.” Current Research in the Pleistocene 10:3-4.

Brown, K.M. 1976. “Fused Volcanic Glass from the Manning Formation.” Bulletin of the Texas Archeological Society 47:189-207.

Coles, J.M. and E.S. Higgs. 1969. The Archaeology of Early Man. New York, Frederick A. Praeger.

Collins, M.B. 1976. “Terminal Pleistocene Cultural Adaptations in Southern Texas.” Prétirage, Colloque XVIII, Habitats Humains Antérieurs a L’Holocène en Amérique. IXe Congrès, Union Internationale des Sciences Préhistoriques et Protohistoriques. Nice, France.

Collins, M.B. 1990. “The Archeological Sequence at Kincaid Rockshelter, Uvalde County, Texas.” Transactions of the 25th Regional Archeological Symposium for Southeastern New Mexico and Western Texas. pp.25-33. Midland, Texas, Midland Archeological Society.

Collins, M.B. 1994. “Cueva Quebrada Shelter, Texas.” Current Research in the Pleistocene 11:26-28.

Collins, M.B. 1995. “Forty Years of Archeology in Central Texas.” Bulletin of the Texas Archeological Society 66:361-400.

Collins, M.B. 1998a. “Interpreting the Clovis Artifacts from the Gault Site.” TARL Research Notes 6(1):4-12.

Collins, M.B. (ed.). 1998b. Wilson-Leonard, An 11,000-year Archeological Record of Hunter-Gatherers in Central Texas. Studies in Archeology 31, Texas Archeological Research Laboratory, University of Texas at Austin and Archeology Studies Program, Report 10, Environmental Affairs Division, Texas Department of Transportation.

Collins, M.B. 1999. Clovis Blade Technology. Austin, University of Texas Press.

Collins, M.B. n.d. “Comparing Clovis and the Western European Upper Paleolithic; What are the Rules of Evidence?” Beyond Clovis. Corvallis, Center for the Study of the First Americans.

Collins, M.B. and K.M. Brown. 2000. “The Gault Gisement.” Current Archeology in Texas 2(1): 8-11.

Collins, M.B., T.R. Hester, and P.J. Headrick. 1992. “Engraved Cobbles from the Gault Site, Central Texas.” Current Research in the Pleistocene 9:3-4.

Collins, M.B., T.R. Hester, D. Olmstead and P.J. Headrick. 1991. “Engraved Cobbles from Early Archeological Contexts in Central Texas.” Current Research in the Pleistocene 8:13-15.

Ferring, C.R. 2001. The Archaeology and Paleoecology of the Aubrey Clovis Site (41DN479), Denton County, Texas. Denton, Center for Environmental Archaeology, Department of Geography, University of North Texas.

Figgins, J.D. 1933. “A further contribution to the antiquity of man in America.” Proceedings of the Colorado Museum of Natural History 12(2):4-10.

Frison, G.C. 1982. “The Sheaman Site: A Clovis Component.” In G.C. Frison and D. Stanford (eds.) The Agate Basin Site. pp.143-157. New York, Academic Press.

Frison, G.C. and B. Bradley. 1980. Folsom Tools and Technology of the Hanson Site, Wyoming. Albuquerque, University of New Mexico Press.

Frison, G.C. and B. Bradley. 1999. The Fenn Cache: Clovis Weapons and Tools. Santa Fe, One Horse Land & Cattle Co.

Frison, G.C. and D. Stanford. 1982. The Agate Basin Site. New York, Academic Press.

Frison, G.C. and Todd, L.C. 1986. The Colby Mammoth Site: Taphonomy and Archaeology of a Clovis Kill in Northern Wyoming. Albuquerque, Univeristy of New Mexico Press.

Gardner, W.M. 1974. The Flint Run Complex: Pattern and process during the Paleoindian to Early Archaic. Occasional Publication No. 1:5-47. Washington D.C., Archaeology Laboratory, Catholic University of America.

Hall, D.A. 1995. “Bones of Nebraska Mammoths Imply Early Human Presence.” Mammoth Trumpet 10(1):1, 4-7.

Hall, D.A. 1997. “Flaked Bone Suggests Americans Used Mammoths as Tool Materials.” Mammoth Trumpet 12(3):7-8.

Hannus, L.A. 1989. “Flaked Mammoth Bone from the Lange/Ferguson Site, White River Badlands Area, South Dakota.” In R. Bonnichsen and M. Sorg (eds.) Bone Modification. Orono, Center for the Study of the First Americans, Institute for Quaternary Studies, Univ. of Maine.

Haury, E.W. 1953. “Artifacts with Mammoth Remains, Naco, Arizona I: Discovery of the Naco Mammoth and the Associated Projectile Points.” American Antiquity 19:1-14.

Haury, E.W., E.B. Sayles, and W.W. Wasley. 1959. “The Lehner Mammoth Site, Southeastern Arizona.” American Antiquity 19:1-14.

Haynes, C.V., Jr. 1982. “Were Clovis Progenitors in Beringia?” In D.M. Hopkins, J.V. Matthews, Jr., C.W. Schweger, and S.B. Young (eds.) Paleoecology of Beringia. pp.383-398. New York, Academic Press.

Haynes, C.V., Jr. 1987. “Clovis Origins Update.” The Kiva 52(2):83-93.

Haynes, C.V., Jr. 1992. “Contributions of Radiocarbon Dating to the Geochronology of the Peopling of the New World.” In Taylor, Long, and Kra (eds.) Radiocarbon after Four Decades. pp.355-374. New York, Springer-Verlag.

Haynes, C.V., Jr. 1993. “Clovis-Folsom Geochronology and Climatic Change.” In O. Soffer and N. D. Praslov (eds.) From Kostenki to Clovis. pp.219-236. New York, Plenum Press.

Hemmings, T.E. 1970. Early Man in the San Pedro Valley, Arizona. Unpublished Ph.D. dissertation. Tucson, Univ. Arizona.

Hester, J.J. 1972. Blackwater Locality No. 1: A Stratified Early Man Site in Eastern New Mexico. Fort Burgwin Research Center Publication 8. Rancho de Taos, New Mexico.

Hester, T.R., G.L. Evans, F. Asaro, F. Stross, T.N. Campbell and H. Michel. 1985. “Trace Element Analysis of an Obsidian Paleo-Indian Projectile Point from Kincaid Rockshelter, Texas.” Bulletin of the Texas Archeological Society 56:143-153.

Holen, S.R. 1996. “The Lovewell Mammoth: A Late Wisconsinan Site in North Central Kansas.” Current Research in the Pleistocene 13:69-70.

Holliday, V.T. 1997. Paleoindian Geoarchaeology of the Southern High Plains. Austin, Univ. Texas Press.

Holliday, V.T., C.V. Haynes Jr., J.L. Hofman, and D.J. Meltzer 1994. “Geoarchaeology and Geochronology of the Miami (Clovis) Site, Southern High Plains of Texas.” Quaternary Research 41:200-213.

Jelinek, A.J. 1966. “An Artifact of Possible Wisconsin Age.” American Antiquity 31:434-435.

Jelinek, A.J. 1992. “Perspectives from the Old World on the Habitation of the New.” American Antiquity 57:345-347.

Jennings, J.D. (ed.). 1978. Ancient Native Americans. San Francisco, Freeman & Co.

Johnson, E. 1987. Lubbock Lake: Late Quaternary Studies on the Southern High Plains. College Station, Texas A&M Univ. Press.

Johnson, E. 1989. “Human-Modified Bones from Early Southern Plains Sites.” In R. Bonnichsen and M. Sorg (eds.) Bone Modification. Orono, Maine, Center for the Study of the first Americans, Inst. of Quaternary Studies, Univ. Maine.

Johnson, L. 1991. Early Archaic Life at the Sleeper Archaeological Site, 41BC65, of the Texas Hill Country, Blanco County, Texas. Texas State Department of Highways and Public Transportation, Pulications in Archaeology, Report 39, Austin.

Johnson, L. 1995. Past Cultures and Climates at Jonas Terrace, 41ME29, Medina County, Texas. Office of the State Archeologist Report 40, Texas Department of Transportation and Texas Historical Commission, Austin.

Leonhardy, F.C. 1966. Domebo: A Paleo-Indian Mammoth Kill in the Prairie-Plains. Contributions of the Museum of the Great Plains 1. Lawton, Oklahoma.

Lourandos, H. 1997. Continent of Hunter-Gatherers. Cambridge, Cambridge University Press.

Lundelius, E.L., Jr. 1984. “A Late Pleistocene Mammalian Fauna from Cueva Quebrada, Val Verde County, Texas.” Carnegie Museum of Natural History, Special Pub. 8:456-481.

McAvoy, J.M. and L.D. McAvoy. 1997. Archaeological Investigations of Site 44SX202, Cactus Hill, Sussex County, Virgina. Research Report Series 8, Virginia Department of Historic Resources. Richmond.

McKinney, W.W. 1981. “Early Holocene Adapations in Central and Southwestern Texas: The problem of the Paleoindian-Archaic Transition.” Bulletin of the Texas Archeological Society 52: 91-120.

Meltzer, D.M. 1989. “Was Stone Exchanged among Eastern North American Paleo-Indians?” In C. Ellis and J. Lothrop (eds.) Eastern Paleo-Indian Lithic Resource Use. pp.11-39. Boulder, Westview Press.

Meltzer, D.M. and M.R. Bever 1995. “Paleoindians of Texas: An Update on the Texas Clovis Fluted Point Survey.” Bulletin of the Texas Archeological Society 66:47-81.

Miller, S.J. 1989. “Characteristics of Mammoth Bone Reduction at Owl Cave, The Wadsen Site, Idaho.” In R. Bonnichsen and M.H. Sorg (eds.) Bone Modification. Orono, Center for the Study of the First Americans, Institute for Quaternary Studies, University of Maine.

Redder, A.J. 1985. “Horn Shelter Number 2: The South End, A Preliminary Report.” Central Texas Archaeologist 10: 37-65.

Sanders, T.N. 1990. Adams: The Manufacturing of Flaked Stone Tools at a Paleoindian Site in Western Kentucky. Buffalo, Persimmon Press.

Schick, K.D. and N. Toth. 1993. Making Silent Stone Speak. New York, Simon and Schuster.

Sellards, E.H. 1938. “Artifacts Associated with Fossil Elephant.” Bull. of the Geological Society of America 49:999-1010.

Sellards, E.H. 1952. Early Man in America. Austin, University of Texas Press.

Stanford, D. 1982. “A Critical Review of Archaeological Evidence Relating to the Antiquity of Human Occupation of the New World.” In Ubelaker and Viola (eds.) Plains Indian Studies: A Collection of Essays in Honor of John C. Ewers and Waldo R. Wedel. pp.202-218. Smithsonian Contributions to Anth. 30. Washington, D.C., Smithsonian Press.

Stanford, D. 1983. “Pre-Clovis Occupation South of the Ice Sheets.” In R. Shutler, Jr. (ed.) Early Man in the New World. pp.65-72. Beverly Hills, Sage Pub.

Stanford, D. 1991. “Clovis Origins and Adaptations: An Introductory Perspective.” In R. Bonnichsen and K.L. Turnmire (eds.) Clovis: Origins and Adaptations. pp.1-13. Corvallis, Center for the Study of the First Americans, Department of Anthropology, Oregon State University.

Steele, D.G. and D.L. Carlson. 1989. “Excavation and Taphonomy of Mammoth Remains from the Duewall-Newberry Site, Brazos County, Texas.” In R. Bonnichsen and M. Sorg (eds.) Bone Modification. pp.413-430. Orono, Center for the Study of the First Americans, Inst. for Quaternary Studies, Univ. Maine.

Steele, D.G. and J.F. Powell. 1994. “Paleobiological Evidence of the Peopling of the Americas.” In R. Bonnichsen and D.G. Steele (eds.) Method and Theory for Investigating the Peopling of the Americas. pp.141-164. Corvallis, Center for the Study of the First Americans, Department of Anthropology, Oregon State University.

Watt, F.H. 1978. “Radiocarbon Chronology of Sites in the Central Brazos Valley.” Bulletin of the Texas Archeological Society 49:111-138.

Wilmsen, E.N. and F.H.H. Roberts. 1978. Lindenmeier, 1934-1974, Concluding Report on Investigations. Washington, D.C., Smithsonian Institution Press.

Wormington, H.M. 1957. Ancient Man in North America. 4th edition. Denver, Denver Museum of Natural History.

Athena Review Image Archive™   |  Paleoanthropology in the News   |  Guide to Archaeology on the Internet   |   Free issue   |   Back issues

Main index of Athena Review   |   Subject Index   |   Travel Pages   |   Galleries and Museums   |  Ad rates   |  Current issue index |

Copyright  ©  1996-2003  Athena Publications, Inc.  (All Rights Reserved).