Read an Excerpt
Clovis Lithic Technology
Investigation of a Stratified Workshop at the Gault Site, Texas
By Michael R. Waters, Charlotte D. Pevny, David L. Carlson Texas A&M University Press
Copyright © 2011 Michael R. Waters, Charlotte D. Pevny, and David L. Carlson
All rights reserved.
ISBN: 978-1-60344-467-5
CHAPTER 1
The Gault Site and the Investigation of Excavation Area 8
The Gault site (41BL323), located in central Texas, is one of the most important Clovis sites in North America. This single site contains more Clovis artifacts than any other 13,000-year-old site west of the Mississippi River. These artifacts were found in primary contexts within undisturbed late Quaternary deposits.
Because Gault was a quarry, workshop, and camp, it provides an unprecedented look at Clovis lithic technology. Bifaces, in all stages of reduction, chronicle the Clovis trajectory of biface manufacture. Hundreds of cores, core-tablet flakes, and blades document the trajectory of blade production. Just as important are the tens of thousands of pieces of debitage left behind from these tool-making activities. These and other tools and faunal remains provide a window into the Clovis past.
This volume presents a comprehensive summary of the studies conducted at Excavation Area 8 of the Gault site. This area, known informally as the Lindsey pit, was excavated by faculty and students from Texas A&M University in 2000 and 2001. Excavation Area 8 was a lithic workshop where dense concentrations of Clovis artifacts were recovered from two geological units.
Location and Physiographic Setting
The Gault site covers roughly 16 ha in Bell County, Texas, at the headwaters of Buttermilk Creek (figure 1) in a small valley incised into limestone bedrock in the Edwards Plateau Physiographic Province. Buttermilk Creek is a perennial first-order stream and flows from west to east for 13 km to its confluence with Salado Creek, which eventually empties into the Brazos River. The creek is recharged with water from gravity springs that tap the Edwards Aquifer at its head and along its course (figure 2). These springs have rarely gone dry during historic times.
The valley at Gault is asymmetrical, with steep bedrock slopes on the south side and a gently sloping surface to the north. The bedrock in the Buttermilk Creek drainage basin is composed entirely of the Lower Cretaceous Comanche Peak, Edwards, and Georgetown formations. The Comanche Peak and Georgetown formations are primarily chalky limestones and marls with some shale. The Edwards Formation, which overlies the Comanche Peak Formation and underlies the Georgetown Formation, consists of fine-grained limestone and dolomite with nodules and thin lenses of chert (Adkins and Arick 1930; Barnes 1981; Sellards et al. 1932). These rocks have undergone physical and chemical weathering, and the products of these weathering processes are the source of the clastic sediments making up the colluvial and alluvial deposits created during the late Quaternary. These formations have also been deformed and fractured during post-depositional tectonic activity.
The valley has thick alluvial and colluvial deposits supporting a diverse array of trees and other vegetation. The valley slopes are composed of bedrock covered in some places with a thin veneer of colluvium. Chert abundantly crops out in seams on the slopes. The surrounding uplands are characterized by thin soils and rocky limestone surfaces that support a vegetation of trees and grass.
At a broader scale, the Gault site lies at the eastern edge of the Edwards Plateau in a unique ecotonal setting, near the Balcones Escarpment. East and south of the Escarpment lie the Blackland Prairies and the Gulf Coastal Plain. These areas provide a diversity of floral and faunal resources, only a short distance from the site, that are different from those of the Edwards Plateau.
History of Research
The first professional excavations were undertaken in 1929–1930 by J. E. Pearce of the University of Texas (Collins 2002, 2007). He named the site and excavated trenches through a large burned rock midden, defining the prehistoric sequence for central Texas. Pearce excavated only the Late Prehistoric and Archaic levels of the site and never discovered the rich Paleoindian layers.
Before, during, and after the Pearce excavations, the site was dug extensively by relic hunters in search of projectile points. Most of this indiscriminate digging was concentrated in the artifact-rich Late Prehistoric and Archaic layers. Some collectors, however, dug a little deeper and found Late Paleoindian, Folsom, and Clovis materials.
The first collector to bring the Clovis component at Gault to the attention of professional archaeologists was David Olmstead in 1990 (Collins 2002). He showed Michael Collins and Tom Hester, of the Texas Archaeological Research Laboratory in Austin, four engraved stones associated with a Clovis projectile point. In 1991, Collins and Hester conducted a small test excavation in the area of the Olmstead discovery and confirmed the presence of Clovis artifacts at the site. Uncontrolled digging continued at Gault until 1998, when Howard and Rickey Lindsey purchased the property that encompassed most of the site and ended the digging.
In 1998, Howard Lindsey excavated a pit on the south side of the creek to look for artifacts. In the process, he discovered a mammoth mandible and numerous Clovis artifacts. Lindsey contacted Collins and Hester, who spent a few days recovering the mammoth remains and artifacts from the excavated sediments. Recognizing the importance of the site and wishing to pursue scientific investigations, Collins negotiated and received permission to excavate on the property for a three-year period from June 1999 to June 2002.
Hester and Collins invited Harry Shafer and Michael Waters of Texas A&M University to investigate the site with them as coprincipal investigators. Together they received some of the initial funding to investigate the site. After examining the site, Shafer and Waters decided to conduct a block excavation adjacent to the pit dug by Lindsey in 1998 (Excavation Area 8; figure 3). Here the stratigraphy was well defined, Clovis lithics were known to be abundant, and the potential for preservation of faunal remains was great. Shafer and Waters conducted field investigations in the spring semester of 2000. In the summer of 2001, David Carlson (Texas A&M University) returned with a field school and continued excavations along the south wall of Excavation Area 8. During this period, Collins excavated other portions of the site. Waters worked as the site geoarchaeologist, recording stratigraphic sections across the site over the duration of the investigation. Tom Stafford (Stafford Research Laboratory) obtained radiocarbon dates from Excavation Area 8, and Steve Forman (University of Illinois) obtained luminescence ages from the sediments at Excavation Area 8 and elsewhere at the site.
In May 2002, archaeological investigations ceased at the Gault site. The excavation areas and trenches were filled and the site returned to pasture land. In 2007, Collins purchased the Lindsey property, resumed archaeological investigations, and donated the property to the Archaeological Conservancy. He continues to investigate the site. Waters continues to record the geology of new exposures and excavations.
The Texas A&M University Excavations and Analyses
The Texas A&M University excavations were confined to Excavation Area 8. In 2000, 26 one-square-meter units were excavated, 22 within a contiguous block (figures 4 and 5). Within the block, the upper sediments of most excavation units were quickly removed to Level 8 (elevation 95.97 m above datum), with only the location of diagnostic artifacts noted. Standard archaeological methods were then used to excavate the remaining late Quaternary deposits overlying the limestone bedrock. Excavation units were dug in 5 cm levels using bamboo tools to avoid damaging the artifacts. The stratigraphy within each excavation unit was monitored and recorded. Most artifacts and other items of interest larger than 2.5 cm were piece-plotted to 1 cm accuracy in the three-dimensional site grid. All excavated sediments were screened through 0.25-inch mesh; a 14 percent sample from each level was screened through 0.125-inch mesh. Excavation unit N1018/E983 served as the witness column in the central part of the block. A 50 cm2 bulk sample was collected from the northwest quadrant of this excavation unit following the same methodology used in the rest of the block, but artifacts were not piece-plotted. This sample was microscreened in the laboratory. After the sample was removed, the remaining 75 percent of the unit was excavated in the same manner as the other excavation units. The Texas A&M University excavation crew consisted primarily of graduate students and some undergraduates (figure 6). Charlotte Pevny supervised the laboratory. Numerous undergraduate students helped process and analyze the collected materials.
During the 2000 season, nearly 67,000 lithic artifacts, including bifaces, points, blades, cores, endscrapers, and debitage, were recovered (table 1). In addition, approximately 5700 faunal specimens were collected. Analysis of this material occurred over the next nine years and resulted in five M.A. theses and two Ph.D. dissertations. William Dickens (2005) studied the blade and biface technology. James Wiederhold (2004) conducted a detailed analysis of the endscrapers, with special emphasis on usewear. Ashley Smallwood (2006) and Scott Minchak (2007) examined the usewear on the bifaces and blades, respectively, and conducted numerous experimental studies to interpret the usewear observed on Clovis artifacts. Charlotte Pevny (2009) studied the debitage and edge-modified flake tools. Dawn Alexander (2008) conducted a site-formation study, which included artifact orientation and refit analyses. Heidi Luchsinger (2002) conducted a micromorphological study of the Excavation Area 8 sediments. Eric Bartelink and Jason Wiersema, both doctoral students at the time, analyzed the faunal remains from the site as a special study. Brandy Gibson (1997) conducted a geomorphic reconnaissance of the entire Buttermilk Creek floodplain several years before excavations began.
Additional work in Excavation Area 8 undertaken by David Carlson in 2001 resulted in the excavation of 15 one-square-meter units along the south wall of the 2000 excavation area and to the east along the south wall of the Lindsey pit (figure 1). In this area, Clovis-age deposits were largely absent because they pinched out or were eroded. However, the remaining deposits contained a Late Paleoindian and Archaic record. This material was processed and cataloged, but it is not reported here.
Objectives and Organization
This volume has two primary objectives. The first is to present a comprehensive study of the material recovered from the Clovis components at Excavation Area 8. This includes a description of the Clovis artifacts and faunal remains, a reconstruction of Clovis blade and biface manufacture, and the identification of spatially organized activities at the Clovis workshops. The second objective is to use this information along with data from other sites to examine the broader issues of Clovis technology, subsistence, and mobility. Even though Archaic, Late Paleoindian, and Folsom artifacts were collected at this locality, only the Clovis assemblage is discussed in this volume. Further, this study focuses exclusively on the Clovis-age artifacts and faunal remains recovered during the 2000 Texas A&M University excavations. Only Clovis materials from secure geological contexts are presented here; artifacts that may be Clovis but were recovered from equivocal stratigraphic contexts, younger deposits, or the surface of the excavation block are not included. Chapter 2 presents the geoarchaeological studies of Excavation Area 8. Chapters 3 through 6 cover all aspects of the organization of Clovis lithic technology, including detailed discussions of blade, biface, and other tool technologies. Chapter 7 presents the results of microscopic usewear studies of the lithic assemblage. The faunal remains from Excavation Area 8 are discussed in chapter 8. Chapter 9 provides an analysis of the spatial distribution of artifacts and faunal material. A synthesis of the data from Excavation Area 8 is presented in chapter 10 and includes discussions of how this portion of the Gault site was used by the Clovis inhabitants and broad issues of Clovis technology, subsistence, and mobility. Much of the information presented in these chapters is a synthesis of theses or dissertations completed at Texas A&M University. However, because additional studies since their completion have yielded new data, there are some discrepancies between what is reported here and in the theses and dissertations. In all cases, this book represents the most current and revised information about Excavation Area 8.
CHAPTER 2
Stratigraphy, Chronology, and Site Formation
Buttermilk Creek divides the Gault site into two geologically distinct areas (figure 1). The south side of the creek is dominated by fans that emanate from two north-south trending tributary channels. Fan deposition is characterized by coarse-grained clastic sediments. As the fans aggraded, they also were inundated periodically by floodwaters from Buttermilk Creek. These overbank flood events left deposits of fine-grained clays. This interplay of fan and floodplain processes created a complex interfingering facies relationship of alluvial fan gravels with finegrained floodplain clays and silts. The area north of the creek is characterized by a modern floodplain and two alluvial terraces. Deposition in this region is dominated by channel and overbank processes from the creek with some colluvial input from the adjacent slopes to the north. Channel and channel margin sediments underlie the lower terrace, and vertical accretion floodplain sediments dominated the upper terrace. Colluvial sediments, consisting of sands and angular gravels, are derived from the adjacent slopes and interfinger with the floodplain deposits.
This chapter describes the late Quaternary geology of Excavation Area 8, which lies on the south side of the creek. The age of deposits at Excavation Area 8 were determined by the presence of known-age diagnostic artifacts, primarily projectile points, and luminescence ages. Ultimately, we did not use the radiocarbon method to date the deposits at Excavation Area 8 because materials suitable for radiocarbon dating were absent or gave erroneous dates.
Late Quaternary Geology and Geochronology
The surficial landscape of the Gault site south of Buttermilk Creek is dominated by two alluvial fan complexes separated by a bedrock ridge. To the west of the ridge is a large alluvial fan complex that slopes toward Buttermilk Creek and is dissected by two ephemeral stream channels. To the east of the interfluve is a small tributary fan complex. The sediments deposited in these areas derive from both Buttermilk Creek and the adjacent slopes and tributary streams. Poorly sorted gravel deposits characterize the fan deposits and attest to high-energy, episodic, and rapid deposition. Well-sorted clay and silt sediments derive from Buttermilk Creek during over-bank flood episodes.
Excavation Area 8 is situated at the nose of the bedrock ridge on the edge of both fan complexes. At this location, floodplain deposition dominates over fan deposition, unlike the adjacent fans. Here a thick late Quaternary sediment sequence is preserved (figures 7-11). This location is unique to the Gault site; it has a well-defined stratigraphy that buried a long archaeological sequence and was not eroded by later geological processes.
At Excavation Area 8, the late Quaternary sediments lie unconformably on eroded limestone bedrock. This bedrock is highly weathered and slopes downward to the north (figure 9). The first late Quaternary sediments deposited on the bedrock were those of Unit 1. Unit 1 consists of framework-supported irregular rounded limestone cobbles with a few chert gravels that form a 4.5 m long and 0.5–0.1 m thick body that rests on the gently sloping limestone (figure 10). These gravels were deposited by high-energy flow within a small tributary channel. No artifacts or faunal remains were found in this unit.
Unit 2 unconformably overlies Unit 1. It is a gravel deposit of subangular to sub-rounded limestone and chert cobbles. This deposit rises and thickens to the north, creating a shallow depression to the south (figure 7). The gravels represent high-energy deposition within a channel of a tributary fan stream. No artifacts were found in Unit 2, but Clovis-age artifacts were found on the surface of the gravel. The remains of horse, mammoth, and turtle were found within the gravels.
The slight depression created to the south of the Unit 2 gravel body occasionally filled with water, creating a shallow pond. Within this pond, clays accumulated (Unit 3a). This clay, which ranges from 10 to 25 cm thick, is dark brown in color and has undergone pedogenic alteration. Wedge-shaped peds have developed in the clay, with prominent slickensides along the ped faces. Calcium carbonate nodules within the clay range from 0.5 to 3.0 cm in diameter. Unit 3a pinches out at both the north and south ends of the excavation block (figure 12a). Artifacts within Unit 3a are coated with calcium carbonate on both the top and the bottom surfaces. A dense concentration of Clovis artifacts, including three diagnostic projectile points (figures 13 and 14; table 2), was recovered from Unit 3a.
(Continues...)
Excerpted from Clovis Lithic Technology by Michael R. Waters, Charlotte D. Pevny, David L. Carlson. Copyright © 2011 Michael R. Waters, Charlotte D. Pevny, and David L. Carlson. Excerpted by permission of Texas A&M University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
