ArticlePDF Available

Evidence for Pre-Clovis occupation at the Gault Site (41BL323), central Texas

Authors:
M.B. Collins
M.B. Collins
M.B. Collins
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Bruce Bradley at University of Exeter
Bruce Bradley
Download full-text PDF
Read full-text
Download citation
Content uploaded by Bruce Bradley
Author content
All content in this area was uploaded by Bruce Bradley on Dec 02, 2022
Content may be subject to copyright.
EVIDENCE FOR OLDER-THAN-CLOVIS AT THE GAULT SITE, TEXAS
Michael B. Collins with contributions by Sergio Ayala, Bruce Bradley, Steve Forman, Charles Frederick, Jennifer Gandy, Anna Gilmer, Robert Lassen, Brendan Nash, Jill Patton, Linda Perry, Jack Rink, Marilyn Shoberg, Charles A. Speer, Clark Wernecke, Tom Williams, and Nancy Velchoff
Figure 1. Regional setting of the Gault
site, Texas (41BL323)
Figure 2. Gault site map showing location of Area 15
and the positions of 2002 and 2007 test units
Figure 3. Photograph of a portion of the west wall of excavated
block in Area 15 with the approximate vertical distributions of
time-diagnostic artifacts as observed during excavation. The
excavation is ca. 3.6 meters deep, necessitating the stepped
excavation shown here
CLOVIS FIRSTERS—READ THIS:
An underlying question in this project is, does this site meet the criteria for acceptance as evidence
of “preClovis?” At Gault, there is no doubt that the more than 16,000 Older-than-Clovis artifacts are
of human manufacture; though sparse and poorly preserved, all fauna from the OTC component are
extinct Pleistocene species (horse and mammoth); there are more than 2.5 meters of alluvium with
stratified Clovis and younger components overlying this component; in places there is a break in
artifact frequencies immediately below Clovis before increasing again in the OTC deposit;
preliminary OSL dating of OTC includes two results (11-03, 11-04) that appear to be reversals, that
in fact, date fill in an disturbance (root?) and cut-and-fill feature and two results (11-05b, 11-06) that
we consider incongruously old (further efforts at dating are underway); luminescent dates above
OTC are in correct stratigraphic order, and are consistent with age estimates based on time-
diagnostic artifacts; numerous professional archaeologists witnessed and/or participated in this
excavation at all stages and a few have also been part of the laboratory analyses; multiple lines of
evidence indicate that downward movement of artifacts has been minimal; and, other than the OTC
blade technology, which is similar to that of Clovis, there are no technological or typological
continuities between the lower component and Clovis, nor any typological counterparts to the OTC
Figure 4. Bivariate plot of optically stimulated
luminescence dates by depth and cultural time
periods.
Figure 5. Synoptic geologic profile of west
wall of Area 15 excavation from elevations
92.05m to 93.55 m with the approximate
boundaries of the Clovis and Older-than-
Clovis components (from Gilmer 2013).
Figure 6. Comanche Peak limestone bedrock as
exposed in the floor of the Area 15 excavation
block. Fluting is interpreted as the result of
protracted stream flow of ancestral Buttermilk
Creek across the bedrock surface under three
different orientations of the stream, earliest
(green), intermediate (blue), and latest (red).
Photo by Gregg Cestaro.
Figure 7. Area 15 cultural chronology based
on time-diagnostic artifact distributions and
plotted against the available luminescent
dates.
Figure 8. 3-D plot of Clovis (green) and Older-
than-Clovis (red) artifacts, bedrock (gray), and
large limestone rocks (blue) in the lower 2
meters of excavated Area 15 (view is looking
south). Clovis artifacts are a point, blades,
diagnostic debitage, bifaces, core, and unifaces.
Older-than-Clovis component artifacts lacking
established age assignments consist of
potentially time-diagnostic blade cores, blades,
tools on blades, bifaces, flake cores, and tools
on flakes. The anomalous pile of large,
irregular limestone rocks measures 108 cm E-
W, 79cm N-S and 51cm high, consists of clasts
up to 35 cm across and 15 cm thick and lacking
in stream rounding, partly supported by soft
sediment, and neither clearly natural nor
human-made. A few non-diagnostic flakes
were present throughout. Stratigraphically, it
extends downward from the base of the Clovis
component through most of the underlying
deposits.
Figure 9. Vertical distributions of flakes by weight and
count below 93.00m elevation in the Clovis and Older-
than-Clovis components (modified from Gilmer 2013).
Figure 10. Backplot of diagnostic Andice,
Bell notching flakes, debitage, and point
fragments. At ca.6,000-5900 B.P. this is the
earliest Middle Archaic interval of Central
Texas.
Figure 11. Comparison of
representative Clovis and Older-
than-Clovis artifacts indicative of
blade technology.
Figure 12. Comparison of
representative Clovis and Older-
than-Clovis artifacts indicative
of biface technology.
Figure 13. Comparison of
representative Clovis and Older-
than-Clovis artifacts indicative of
flake technology.
Excavations, 2007-2014, in Area 15 of the Gault Site (Figures 1, 2) specifically targeted evidence
for an Older-Than-Clovis (OTC) occupation seen in two test units in 2002 and 2007 (Figure 2).
Excavation of a large block penetrated 3.5 meters from the surface to bedrock (Figures 3-6) through
culture-bearing alluvial fill. Physical and cultural stratigraphic evidence as well as luminescent dating
are consistent in showing a coherent sequence of Older-than-Clovis, Clovis, Late PaleoIndian, Early
Archaic, and Middle/Late Archaic occupations over an apparent span of more than 14,000 calendar
years (Figures 4, 7-9). This effort addressed seven specific questions: (1) What is the context and
integrity of the Clovis component? This component, roughly 35 cm in thickness, rests in largely
undisturbed alluvial deposits of high integrity (Figure 5). Twenty nine diagnostic stone artifacts and
30,000 waste flakes were recovered. (2) What is the depositional context and integrity of the underlying
component? This component, roughly 80 cm in thickness, is in minimally disturbed water-lain deposits
that yielded 45 formal artifacts and 16,000 waste flakes (Figures 5, 8). There is good separation
between Clovis and underlying material across 10 of the 12 excavated squares, but clearly some mixing
in one limited area; there is also an unexplained, anomalous stone pile near the eastern margin of the
excavation, extending downward from the base of the Clovis component (Figure 8). (3) What is the
nature of an apparent depositional discontinuity between the Clovis-bearing deposit and the one below
it? This discontinuity (Figures 5, 8, 9) is characterized by a decrease in artifacts and an increase in soil
carbonates in the 10 to 15 cm below Clovis, which appears to indicate a reduction in human activity and
possibly an environmental change. (4) What is the age and duration of the earlier component?
Preliminary luminescent dates on soils suggest an age range from greater than 14,000 to near 13,200
years for the early component (Figures 4,7,8). (5) What does the earlier assemblage indicate about
adaptive behavior? Details are lacking, but broadly, the OTC adaptations seem to be that of generalized
hunter-gatherers, a primary dependence on local tool-stone sources, and a generalized stone tool
production technology (Figures 11-13). The extreme contrast in projectile point forms between Clovis
and OTC (Figure 12) indicates contrasting weapons systems. Poor preservation of plant and animal
remains severely hampers interpretations of diet and perishable material culture, but meager evidence
suggests Clovis fauna (bone) and flora (starch grains) resources to be similar to OTC. Use wear on stone
tools are also similar. (6) Is the earlier component a possible technological progenitor of Clovis?
Evidently not. Although blade production technology seems to be somewhat similar in OTC and Clovis,
the production of bifaces and flake tools are entirely different, particularly the nature of the projectile
points. (7) What do differences or similarities between the Clovis and earlier components reveal about
possible alternative histories of occupation and cultural succession at this locality and beyond? Locally
there seems to be a break in the cultural succession between OTC and Clovis at Gault in addition to the
technological contrasts. In broader contexts of North America and the Western Hemisphere, this is not
a surprising finding given the complex mosaic of archaeological manifestations older than and
contemporary with Clovis.
Acknowledgements: Partial support from NSF Grant 0920549, a grant from Summerlee Foundation and from many
donors as well as volunteer efforts on the part of scores of individuals—all of which is greatly appreciated.
Clovis OTC Clovis OTC Clovis OTC
projectile points in any later cultures of this region.
... Pre-Clovis implies a linear relationship between the well-defined, wide-spread, fluted-point, Clovis and the sites that pre-date them, suggesting observable technological and/or cultural continuity (Haynes 2015). "Earlier than Clovis" or "Older than Clovis" has been adopted as the preferred terminology by other investigators to avoid implying a direct relationship between Clovis and the preceding groups (Collins and Bradley 2008;Collins et al. 2014). ...
... In this model, the Clovis toolkit is a unique adaptation brought onto the continent by a group with little relationship to the peoples already there. At Gault and Paisley Caves, two generally well-accepted Earlier than Clovis sites, the lithic assemblage is distinctly non-Clovis (Collins and Bradley 2008;Jenkins et al. 2012;Williams et al. 2018). Projectile points recovered from the earlier than Clovis horizons at both locations are technologically distinct stemmed projectile points lacking the hallmark fluted lanceolate formats of later Paleoindians. ...
... Friedkin sites of central Texas in the early 2010s as one of the best documented earlier than Clovis sites was a major catalyst for reassessing other potential early sites across the state (Collins and Bradley 2008;Michael R. Waters et al. 2011;Waters et al. 2018;Williams et al. 2018). ...
New Excavations of Late Pleistocene Deposits at Bonfire Shelter: A Geoarchaeological Approach to Determining the Origins of Bone Bed 1
Thesis
Full-text available
  • Dec 2020
Bonfire Shelter is a stratified rockshelter site in Val Verde County, Texas with multiple archaeological components spanning the Paleoindian through Late Prehistoric Periods. The shelter is primarily known as the site of two well-documented bison kills: one in the Archaic (Bone Bed 3) and one in the Late Pleistocene (Bone Bed 2). Excavators in the 1960s and 1980s argued that a third bone bed, designated Bone Bed 1, comprised entirely of extinct Pleistocene megafauna, is also the result of human activity. If unambiguous evidence of human activity is identified, Bone Bed 1 may predate the appearance of Clovis or related Early Paleoindian traditions in the region. This thesis presents the results of new excavations and geoarchaeological analyses conducted to evaluate the formation processes associated with Bone Bed 1 and their implications for potential archaeological deposits. In the summer of 2017, Texas State University’s Ancient Southwest Texas Project initiated new excavations of Bone Bed 1. Intact portions of the Bone Bed 1 substrata with in situ faunal remains were identified at the base of test units dating to the 1960s and 1980s. A series of 11 test units and one column sample excavated in this area reidentified and confirmed the Bone Bed 1 stratigraphy and faunal assemblage reported by Bement (1986). Sediment from each column sample strata, including three strata related to Bone Bed 2, was evaluated with a suite of geoarchaeological analyses to better understand the formation processes contributing to the Late Pleistocene deposits at Bonfire Shelter. Targeted microartifact sampling was conducted to identify ephemeral traces of human activity potentially overlooked by previous investigators. A functional model exploring plausible scenarios that could account for the presence of the Pleistocene faunal assemblage at Bonfire Shelter was developed based on ethnoarchaeological accounts, modern proxy studies, and known archaeological sites of similar antiquity. Geological, faunal, and potentially cultural evidence was synthesized using this model to identify the “best fit” scenario for each Bone Bed 1 stratum. While no conclusive evidence of human activity was identified, this thesis provides valuable insight into the dynamic conditions at Bonfire Shelter in the Late Pleistocene and refines the chronology of Bone Bed 1 by over 1,000 years, providing critical context for newly identified Early Paleoindian activity elsewhere in Mile Canyon.
View
... A single radiocarbon date of 12,460 ± 490 RCYBP (13,402-16,100 calBP) obtained by Bement (1986), indicates that Bone Bed I predates the commonly-accepted age range for the Clovis period. In light of new evidence for pre-Clovis occupations from the Gault and the Debra L. Friedkin sites in Central Texas (Collins and Bradley 2008;Waters et al. 2011) and the recent discovery of intact Paleoindian deposits at Eagle Cave less than 1 km to the south (Koenig et al. 2017a:89), the Late Pleistocene deposits at Bonfire shelter warrant careful reevaluation. Bone Bed 2: competing interpretations Dibble and Lorrain (1968), along with Bement (1986), Prewitt (2007) and others (e.g. ...
New investigations at Bonfire shelter, Texas examine controversial bison jumps and bone beds
Article
Full-text available
  • Sep 2020
Bonfire Shelter (41VV218) is a nationally significant site in the Lower Pecos region of the West Texas borderlands that contains a record of episodic use by hunter-gatherers spanning at least twelve millennia. At least two major bison hunting episodes are evident at Bonfire Shelter, one associated with Paleoindian Plainview and Folsom projectile points (Bone Bed 2), and another associated with Late Archaic Castroville and Montell points (Bone Bed 3). The approximately 12,000-year-old layers comprising Bone Bed 2 may represent the oldest and southernmost bison jump in North America, but this interpretation is the subject of recent debate. In addition, older deposits containing Rancholabrean fauna but lacking stone tools (Bone Bed 1) date to approximately 14,000 years ago and are proposed by previous researchers to be at least partially the result of human activity. This article reviews the issues surrounding Bone Bed 2 and Bone Bed 1 and presents new radiocarbon dates, artifacts, features, along with some initial observations and ongoing plans for renewed field investigations at Bonfire Shelter carried out by the Ancient Southwest Texas Project at Texas State University.
View
... The northern part of North America was repeatedly glaciated with ice covering the areas of plausible (or sometime the only possible and passable) spatial movement such as the NW coast of British Columbia and the western Alberta prairies (Fig. 1). Despite a certain progress in the earliest American studies indicating human presence in distant regions on the Great American Plains between 15 300-13 000 yr BP, i.e., prior to the Clovis complex (McAvoy & McAvoy, 1997;Adovasio et al., 1990Adovasio et al., , 1998Webb, 2005;Holen, 2006;Joyce 2006;Collins & Bradley, 2008;Gilbert et al., 2008;Lowery et al., 2010), no consensus has been reached about the timing of the initial human migrations entering the Americas, the level of technology and the associated typological variety of stone tools the early inhabitants of the new continent brought with them. The North American archaeologists still tend to look for assemblages with bifacial or micro-blade flaking patterns in geologically recent deposits, assuming a possible continuity with the Siberian Upper Palaeolithic projectile point and palaeo-Arctic technologies, respectively, or follow possible evolutionary links with the Clovis complex in immediately timely preceding archaeological inventories (Waters et al., 2011). ...
View
... Excavations at Gault, under the direction of Michael Collins since 1998, have only uncovered an estimated 2e3% of these Clovis deposits (Collins, 2010;personal communication). The Gault Site has produced compelling evidence for a pre-Clovis or Older Than Clovis (OTC) occupation (Collins and Bradley, 2008;Williams et al., 2018) and is several hundred meters from the Debra L. Friedkin site, also with pre-Clovis deposits (Jennings, 2012;Waters et al., 2011b;Jennings and Waters, 2014). ...
A blended model of mobility behavior: Clovis period hunter-gatherers at the Gault Site
Article
  • Jul 2019
  • QUATERNARY SCI REV
This paper reviews previous work conducted on Laser Ablation – Inductively Coupled Plasma – Mass Spectrometry (LA-ICP-MS) sourcing 33 Clovis period projectile points from the Gault Site (41BL323) in Central Texas. This paper includes new geochemical data from four new primary geologic sources to increase an Edwards Plateau region chert database to 302 primary geologic samples from 204 samples with which to test against the geochemistry of the 33 Clovis artifacts. These artifacts are then assessed with regards to their form, breakage patterns, and distance to source to suggest a blended model of mobility for Clovis period hunter-gatherers in Central Texas. This data suggests that the pattern of movement across the landscape of the Edwards Plateau for these hunter-gatherers may have been a blended pattern alternating between collector and forager based on changing distribution of resources seasonally. This patterning may have been influenced for the Gault Clovis hunter-gatherers reliant on availability of chert resources as well as ecotonal density and diversity of plants and animals; especially, as gauged by Clovis site types and distribution spread throughout much of the Southern Great Plains.
View
... for decades researchers believed that Clovis artifacts were the material remains of the initial inhabitants of the continent. However, recent investigations at the Gault (Collins and Bradley 2008) and friedkin (Waters et al. 2011) sites in Bell County, and other sites across North America, have yielded cultural materials below Clovis levels with accompanying radiocarbon ages that predate Clovis. Currently, little is known about these pre-Clovis peoples, but it is becoming clearthat they were widespread across the continent centuries before Clovis hunters and gatherers roamed North America. ...
National Register Testing at 41HM46, Hamilton County, Texas: CR 294 Bridge Replacement at the Leon River
Article
  • Jan 2011
Prewitt and Associates, Inc., conducted test excavations at site 41HM46 in Hamilton County, Texas, to determine its eligibility for listing in the National Register of Historic Places. The work was performed in September 2003 in conjunction with a proposed bridge replacement on County Road 294 over the Leon River. The excavations consisted of three backhoe trenches, eight shovel tests, and seven hand-dug test units totaling 7 m3. Excavations yielded a small assemblage of chipped stone artifacts (tools, cores, and unmodified debitage), two features, and other cultural materials that appear to be associated with occupations ranging from the Late Archaic to possibly the Late Prehistoric. Although most of the artifacts and other cultural materials were recovered from an undisturbed cumulic soil, there was no clear vertical separation of deposits or discrete components. The low artifact frequency, lack of well-defined features, and scarcity of potential temporal indicators such as diagnostic tools or charcoal for radiocarbon dating make the site unlikely to yield important information. Based on these factors, it is recommended that 41HM46 be judged ineligible for listing in the National Register of Historic Places.
View
... The northern part of North America was repeatedly glaciated with ice covering the areas of plausible (or sometime the only possible and passable) spatial movement such as the NW coast of British Columbia and the western Alberta prairies (Fig. 1). Despite a certain progress in the earliest American studies indicating human presence in distant regions on the Great American Plains between 15 300-13 000 yr BP, i.e., prior to the Clovis complex (McAvoy & McAvoy, 1997;Adovasio et al., 1990Adovasio et al., , 1998Webb, 2005;Holen, 2006;Joyce 2006;Collins & Bradley, 2008;Gilbert et al., 2008;Lowery et al., 2010), no consensus has been reached about the timing of the initial human migrations entering the Americas, the level of technology and the associated typological variety of stone tools the early inhabitants of the new continent brought with them. The North American archaeologists still tend to look for assemblages with bifacial or micro-blade flaking patterns in geologically recent deposits, assuming a possible continuity with the Siberian Upper Palaeolithic projectile point and palaeo-Arctic technologies, respectively, or follow possible evolutionary links with the Clovis complex in immediately timely preceding archaeological inventories (Waters et al., 2011). ...
View
... In general, the Buttermilk Creek Complex tools and cores are small and lightweight, and represent a toolkit designed for high residential mobility. Artifacts in a similar, dated geological context are reported at the Gault site, Texas, just 250 m upstream (Collins and Bradley 2008). Morrow and others (2012) recently questioned the accuracy of the OSL ages and the geological context of the artifacts at the site, but provided no new data from the site to support this claim. ...
The First Americans: A Review of the Evidence for the Late-Pleistocene Peopling of the Americas
Chapter
Full-text available
  • Jan 2014
Archaeological evidence from North America shows that Clovis complex sites date between 13,000 and 12,600 cal yr BP. The evidence for the Clovis complex pre-dating 13,000 cal yr BP is equivocal. Artifact assemblages are found at a number of sites in South America that date to 13,000 cal yr BP that have no affinity to Clovis. These data show that both continents were contemporaneously occupied at 13,000 cal yr BP. The absence of Clovis artifacts in Beringia and the geographic concentration of the Clovis complex in North America south of the continental ice sheets indicate that Clovis likely originated somewhere in the continental United States from an antecedent group. Several archaeological sites in North and South America provide credible evidence for the occupation of the Americas before Clovis, and these sites date back to ca. 15,000 cal yr BP, some 2000 calendar years before Clovis. Artifacts from these sites show that these early people made and utilized bifaces, blades, bladelets, and osseous tools. A migration of people into the Americas, a few thousand years prior to Clovis, is supported by the modern genetic evidence. The presence of people in the Americas prior to Clovis also correlates with the initial decline in megafaunal populations during the fifteenth and fourteenth millennia. We propose the term “Exploration Period” to encompass the sites predating the emergence of the recognizable hallmarks of the Clovis complex. © 2014 by Kelly E. Graf, Caroline V. Ketron, and Michael R. Waters. All rights reserved.
View
Discovering World Prehistory: Interpreting the Past through Archaeology
Book
  • Jan 2022
Discovering World Prehistory introduces the general field of archaeology and highlights for students the difference between obtaining data (basic archaeology) and interpreting those data into a prehistory, a coherent model of the past.
View
Envisioning a Western Clovis Ritual Complex
Article
  • Jun 2021
It is becoming increasingly apparent that the initial migrants into the Americas arrived via the Pacific coastal route prior to the inception of the Clovis complex. It further appears that these initial immigrants had non-Clovis technology and a generalized (or broad spectrum) economy with little use of the megafauna in their midst. Once developed, Clovis, marked by a different lithic technology, quickly spread over most of the Americas as far south as Panama, and is associated with the exploitation of proboscideans. This begs the question of why these animals were not exploited by the apparently earlier people. It is speculated herein that the seemingly abrupt appearance of Clovis, its focus on proboscideans, and its sudden demise reflects the appearance of a technologic and organizational system focused on the ritual exploitation of mammoths operating within preexisting Paleoindian societies, herein called the Western Clovis Ritual Complex.
View
North America before Clovis: Variance in temporal/spatial cultural patterns, 27,000-13,000 cal yr BP
Article
Full-text available
  • Jan 2014
Significant new archaeological discoveries made over recent decades present a wide range of contrasting cultural patterns across North America during various portions of time between 27,000 and 13,000 cal yr BP. Each of these patterns is now represented by multiple sites within distinctive environmental settings. The extent and variance of this rich archaeological fabric indicates a much more complex process of peopling North America (and the Western Hemisphere), with multiple cultural origins and a long period of human presence prior to the advent of the distinctive Clovis manifestation, or coexisting with but differing from Clovis. The focus is on North America with limited mention of selected datasets in South America. © 2014 by Kelly E. Graf, Caroline V. Ketron, and Michael R. Waters. All rights reserved.
View
ResearchGate has not been able to resolve any references for this publication.