Footprints on the Landscape

Material Information

Footprints on the Landscape The Historical Ecology of Hunter-Gatherers in the Archaic Southeast
Waggoner, James
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (165 p.)

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Committee Chair:
Sassaman, Kenneth E.
Committee Members:
Oyuela-Caycedo, Augusto
Brandt, Steven A.
Southworth, Jane
Graduation Date:


Subjects / Keywords:
Archaeology ( jstor )
Area surveys ( jstor )
Coastal plains ( jstor )
Ecology ( jstor )
Forests ( jstor )
Highlands ( jstor )
Hunter gatherers ( jstor )
Landscapes ( jstor )
Stone tools ( jstor )
Wetlands ( jstor )
Anthropology -- Dissertations, Academic -- UF
City of Gainesville ( local )
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Anthropology thesis, Ph.D.


Hunter-gatherer populations worldwide burn vegetation and use other means to modify their environments for a variety of reasons. Archaeologists acknowledge that hunter-gatherers of the ancient past likely impacted their environments much like modern foragers, but they do not agree on the scale or consequences of the impacts. Many habitats worldwide depend on the regular application of fire to be sustained. Lightning strikes are one source of ignition for fire-dependent communities, but humans may have been equally active in their spread and perpetuation. One such fire-dependent community is the longleaf pine-wiregrass forests of southwest Georgia, U.S. Greatly reduced by Euroamerican land-use practices, longleaf pine-wiregrass forests can be sustained only when subjected to fire every one to three years. The economic value of these forests to humans is significant, if indirect. When in close proximity to wetlands, longleaf pine-wiregrass forests provide ample edge and browse for white-tailed deer and other fauna important to ancient foragers. Through the use of fire, humans had the potential to manage the primary production of forests and thus influence the degree to which settlement of upland localities could be sustained. Archaeological survey in the Dougherty Plain of southwest Georgia was conducted to collect regional-scale data on the land-use patterns of hunter-gatherers who resided in the area during the Late Archaic period (ca. 5000-3000 BP). Lithic artifacts from 446 components were classified into four site types indicative of functional variations among locations in a settlement-subsistence system. Bridging arguments linking patterns of land-use to the predictability of resources, and hence use of fire, were deduced from theories of technological organization. Results of this analysis suggest that resource predictability enhanced by burning supported settlement consisting of habitation throughout the interriverine zone at locations of upland wetlands, coupled with short-term camps and hunting stations in upland tracts adjacent to wetlands. These are the same upland tracts that hold the greatest potential for fire management in the Dougherty Plain. Future research should target upland wetlands in these locales to seek sediment cores with direct evidence for burning. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (Ph.D.)--University of Florida, 2009.
Adviser: Sassaman, Kenneth E.
Electronic Access:
Statement of Responsibility:
by James Waggoner.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Waggoner, James. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
Resource Identifier:
489256527 ( OCLC )
LD1780 2009 ( lcc )


This item has the following downloads:

Full Text




2009 James Cowan Waggoner, Jr. 2


To my parents, James Cowan and Marjorie Sc hear Waggoner, whose unfailing love and support has given me the strength it takes to succeed 3


ACKNOWLEDGMENTS While I take full responsibility for the thoughts and ideas presented in this dissertation, I must acknowledge several individu als whose contributions made it a more solid and viable piece of work. This dissertation c ould not have been completed wi thout the patient guidance and invaluable advice of my committee members: Ken Sassaman (chair), Steve Brandt, Augusto Oyuela-Caycedo, and Jane Southworth. This research was successful thanks to the cooperation of the Un iversity of Georgia Laboratory of Archaeology, especia lly Mark Williams who afforded valuable advice and use of facilities. David Middleton generously provided housing (for almost 2 years!) at DeSoto Springs Plantation, as well introductions at the Jones Center, Tall Timbers, and to other sites and persons throughout southwest Georgia. I am thankful for Doug Logan, Manager of the DeSoto Springs Plantation, who provided logistic al support, friendship, & unusual headgear. Lindsay Boring, Director of the Joseph Jones Ecol ogical Research Center, gave pe rmission to survey the center property. Woody Hicks and Kay Kirkman, also of the Jones Center, helped me understand the ecosystems of southwestern Georgia and the karst topography and hydrol ogy of the underlying Dougherty plain. Jean Brock, IT manager of th e Jones Center, provided valuable GIS data. Charlie Tarver, owner of Keel Creek, allowed me to work on his plantation which borders the Jones center. Thanks are owed to the Antonio Waring La b of West Georgia and the Archaeology Lab of South Georgia College for collections acce ss. Frankie Snow, Chris Trowell, Keith Stephenson, and Dan Elliott provided key advice regarding fieldwork throughout the course of his research. Dave Crass, State Archaeologist with the Georgia Department of Natural Resources, and Steve Ruckel, wildlife biologist with the Wildlife Resources Division of the 4


Georgia Department of Natural Resources gave me support and permission to Elmodel Wildlife Management Area. Special thanks to Marvin Singletary for access not only to his collections but also to his significant knowledge about southwestern Geor gia's past and present cultural landscapes. Collectors who provided access to collections and valuable insights, including Doug Tarver, Roger Birkhead, Bobby Bass, and Charles Colton, who shared a plethora of knowledge of the Elmodel WMA. A large part of what made th is research possible was the fi eld work assistance I received from so many incredible individu als. I was assisted in the field by Carol E. Colaninno, Deborah Mullins, and the tireless crew of the Chickasa whatchee archaeological survey, including Morgan Ritchie, Tisha Entz, Gary Christopherson, the la te Jim Chamblee, Eric Marks, Elaine Juzwick, and Sammy Smith. Dr. James C. Waggoner also jo ined me in the field on several occasions. Landowners and land managers who gave pe rmission to work on private land: Roland Wetherbee, Al Sudderth, Bill Jordan, Louis Ha tcher, Brian McClure, George Brannon, Bennie Hewett, J.S. Hawke, Hal Holton, Rodney Locke, Butch & Holley Miles, a nd W. Kent Sanders. Several individuals have given me so much over the past several years, that it would take pages to properly show my true thanks. John Chamblee, helped me get started in the very beginning, joined me in field, helped me gain access to many properties, gave me database support and offered countless hours of general cons ultation, but also has been a great friend and source of encouragement. Victor Thompson, has been there from the beginning of my academic life, through the indoctrination into the field of archaeology, a nd has become one of my best friends. 5


Neill Wallis and Michelle LeFebvre are treme ndous; they would do anything for me and I would do anything for them. Drew Kitchen is a good friend who has been there for me and helped me blow off steam on our weekly rides on the Hawthorne trail. Tom Pluckhahn, is a damn good archaeologist and friend. Deborah Mullins is an excellent field archeologist who not only joined me digging in the field, but was th ere for me to discuss archaeology, academia and life. Scott Palumbo is a good archeologist who started the doctoral program with me and who helped spark my interest in coll ege football. Chris Morehart, w ho I met at FSU, is one of the best theoreticians that I have ever known. I must also list Sarah Cevone and Jackie O' (Hall) among the friends that have been there for me as sources of emotiona l support through the long and enduring process of finishing a dissertation. I thank Jonathan Merten for sharing my love of Indian food. And above all, I thank my family. My pa rents, James and Marjorie Waggoner, have given me love and support during this long proce ss. I also thank my brothers Robert and Gordon, sister Lyric, and other members of my friends and family, who never understood why it took so many years to write that paper but who are proud of me nonetheless. 6


TABLE OF CONTENTS Page ACKNOWLEDGEMENTS.............................................................................................................4 LIST OF FIGURES.........................................................................................................................9 LIST OF TABLES................................................................................................................. ........10 ABSTRACT...................................................................................................................................11 CHAPTER 1 PREHISTORIC HUNTER-GATHERER HIST ORICAL ECOLOGY IN THE ARCHAIC SOUTHEAST....................................................................................................................13 Background: The Archaic Peri od in the Eastern Woodlands..................................................15 Changing Perspectives on Hunter-Gatherers...........................................................................19 The Theoretical Approach of the Dissertation.........................................................................24 Fire in the Forest: The Influence of Anthropogenic Burning..................................................26 Research Objectives........................................................................................................... ......29 Revaluating the Interior Coastal Plain...............................................................................30 Rethinking Hunter-Gatherers.............................................................................................31 Organization of the Study........................................................................................................32 2 HUNTER-GATHERERS, THE ENVIRONMENT, AND HISTORICAL ECOLOGY.........34 Background, Definition and Application of Historical Ecology.............................................37 An Environment of Hunter-Gatherers.....................................................................................41 The Development of a Pristine Nature.................................................................................44 Hunter-Gatherers and the Do mestication of the Landscape....................................................49 The Archaeology of Historical Ecology and Anthropogenic Fire...........................................58 3 THE CHICKASAWHATCHEE ARCHAEOLOGICAL SURVEY: ENVIRONMENTAL SETTING, PREVIOUS ARCHAEOLOGI CAL WORK, FIEL D METHODS, AND SITES.................................................................................................................................65 Previous Archaeological Research Related to the Late Archaic.............................................70 Field Methods................................................................................................................. .........80 Individual Survey Areas....................................................................................................... ...81 Chickasawhatchee WMA...................................................................................................82 The Jones Center at Ichuaway...........................................................................................83 Sudderth Farms............................................................................................................... ...86 Magnolia Plantation High..................................................................................................86 7


Red Bluff Plantation......................................................................................................... .86 Bermuda Plantation........................................................................................................... .90 Keel Creek Preserve.......................................................................................................... .90 Elmodel WMA.................................................................................................................. .93 Locke Farms.................................................................................................................. .....93 Ken Sanders.................................................................................................................. .....93 Preliminary Results of Survey................................................................................................. 97 4 THE MATERIALITY OF HUNTER-GAT HERER LANDSCAPES: STONE TOOL ANALYSIS, DISTRIBUTION, AND LANDSCAPE COVER........................................98 Material Residues and Interpretatio ns of Hunter-Gatherer Land Use.....................................99 Previous Applications in the Southeast..................................................................................102 Beyond Foragers and Collectors: Incorporating Historical Ecology into Hunter-Gatherer Land Use..........................................................................................................................108 Application in the Analysis of Recovered Materials From the Chickasawhatchee Archaeological Survey....................................................................................................110 Parameters for the Development of Cate gories of Hunter-Gatherer Activity Areas.............114 Summary and Conclusions....................................................................................................119 5 STONE TOOLS AND FIRE ON THE LANDSCAPE.........................................................121 Using Fire to Increase Resource Pr edictability in Southwest Georgia..................................123 Variation in Stone Tool Assemblages....................................................................................127 Discussion and Conclusions..................................................................................................136 6 HUNTER-GATHERER LIFEWAYS IN THE INTERIOR COASTAL PLAIN: CONCLUSIONS AND FUTURE DIRECTIONS...........................................................139 LIST OF REFERENCES.............................................................................................................147 BIOGRAPHICAL SKETCH.......................................................................................................165 8


LIST OF FIGURES Figure page 1-1 Study area in the upper Gulf Coastal Plain of Georgia......................................................13 3-1 Location of Survey Tracts within the Dougherty Plain.....................................................65 3-2 Chickasawhatchee Survey Area.........................................................................................81 3-3 Ichuaway Survey Area....................................................................................................... 84 3-4 Sudderth Farm Survey Area...............................................................................................86 3-5 Magnolia Plantation Survey Area......................................................................................87 3-6 Red Bluff Plantation Survey Area.....................................................................................88 3-7 Bermuda Plantation Survey Area.......................................................................................90 3-8 Keel Creek Preserve Survey Area......................................................................................91 3-9 Elmodel WMA Survey Area..............................................................................................93 3-10 Locke Farms Survey Area.................................................................................................94 3-11 Ken Saunders Survey Area................................................................................................95 5-1 Type I sites with hafted bifaces bifaces, flakes, cobbles, and cores...............................130 5-2 Type II sites with hafted bifaces, bifaces, and flakes......................................................131 5-3 Type III sites with just flakes...........................................................................................1 32 5-4 Type IV sites with just haft ed bifaces, notably isolated finds.........................................133 9


LIST OF TABLES Table page 5-1 Assemblage Typology and Inferred Site Function..........................................................127 5-2 Absolute and Relative Frequencies of S ites by Assemblage Type and Survey Tract.....129 10


Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FOOTPRINTS ON THE LANDSCAPE: THE HISTORICAL ECOLOGY OF HUNTERGATHERERS IN THE ARCHAIC SOUTHEAST By James Cowan Waggoner, Jr. August 2009 Chair: Kenneth E. Sassaman Major: Anthropology Hunter-gatherer populations worldwide burn ve getation and use other means to modify their environments for a variety of reasons. Ar chaeologists acknowledge th at hunter-gatherers of the ancient past likely impacted their environm ents much like modern foragers, but they do not agree on the scale or consequences of the imp acts. Many habitats worldwide depend on the regular application of fire to be sustained. Lightning strikes are one source of ignition for firedependent communities, but humans may have been equally active in their spread and perpetuation. One such fire-dependent community is the longleaf pine-wiregrass forests of southwest Georgia, U.S. Greatly reduced by Euroamerican land-use practices, longleaf pine-wiregrass forests can be sustained only when subjected to fire every one to three years. The economic value of these forests to humans is significant, if indirect. When in close proximity to wetlands, longleaf pine-wiregrass forests pr ovide ample edge and browse for white-tailed deer and other fauna important to ancient foragers. Through the use of fire, humans had the potential to manage the primary production of forests and thus influe nce the degree to which settlement of upland localities could be sustained. 11


Archaeological survey in the Dougherty Plai n of southwest Georgia was conducted to collect regional-scale data on the land-use patt erns of hunter-gatherers who resided in the area during the Late Archaic period (ca. 5000-3000 BP). Lithic artifacts from 446 components were classified into four site types indicative of functional variations among locations in a settlementsubsistence system. Bridging arguments linking pa tterns of land-use to th e predictability of resources, and hence use of fire, were deduced from theories of technological organization. Results of this analysis suggest that resource predictability enhanced by burning supported settlement consisting of habita tion throughout the interriverine zone at locations of upland wetlands, coupled with short-term camps and hun ting stations in upland tracts adjacent to wetlands. These are the same upland tracts that hold the greatest potential for fire management in the Dougherty Plain. Future research should target upland wetlands in these locales to seek sediment cores with direct evidence for burning. 12


13 CHAPTER 1 PREHISTORIC HUNTER-GATHERER HISTORICAL ECOLOGY IN THE ARCHAIC SOUTHEAST This research addresses anthropogenic im pacts to regional-scale environments by hunter-gatherers. More specifi cally, it examines the extent to which Native Americans used fire to influence the dist ribution of plant and animal re sources. The study area is the Gulf Coastal Plain of the lower Southeast (F igure 1-1), a region th at once had extensive longleaf pine-wiregrass forest c over and a significan t historical associ ation with fire. This area is noteworthy because it has often been portrayed as lacking adequate food resources to support large numbers of prehis toric people, especially during the middle Holocene when pine species supplanted the oa k-hickory-southern pi ne forest (Fish and Fish 1977; Delcourt and Delcourt 1985; La rsen 1980; Watts 1971; Watts et al. 1996; Whitehead 1965). However, upland ponds in th e Coastal Plain were likely focal points of human settlement due to the plant and anim al species they supported, and could have enabled substantial human settlement through the Holocene (Brooks et al. 1996; Waggoner 2003). Research pertaining to Late Archaic landuse in the Middle A tlantic region, the Mid-South, and north Florida has shown that wetland food re sources were a consistent factor in hunter-gathe rer settlement decisions (Custe r and Bachman 1986; Gardner 1978; Milanich 1994; Price and Brown 1985). Caroli na bays, for example, are common in the South Carolina Coastal Plain and investigatio ns have demonstrated their importance to hunter-gatherers throughout th e Archaic Period (Brooks et al. 1996). These and other upland wetlands combine with upland pine ar eas and local streams systems to create a landscape that is unique to the Coastal Plain.


Figure 1-1. Study area in the upper Gu lf Coastal Plain of Georgia. In this dissertation, I examine the relati onship between fire susceptibility and overall land-use patterns ba sed on site data from the Chickasawhatchee and IchauwayNochaway Archaeological Surveys in southwest Georgia. Site distri butions suggest that Late Archaic hunter-gatherers in the interior Coastal Plain purposely placed longer-term habitation sites in locations that were not su sceptible to fire and helped maintain an intentional mosaic of edge habitats or ecoto nes comprised of riveri ne localities and piney uplands interspersed with small wetlands using broad-scale burning. From this perspective, hunter-gatherers were actively engaging the environment in ways that call into question the long-held assumption that th ey did not make widespread environmental 14


impacts. Indeed, in some areas like the in terior Coastal Plain, hunter-gatherer groups may have made a greater impression on the land scape than later, more sedentary people. The lack of attention given to this leve l of hunter-gatherer landscape manipulation by archaeologists relates directly to the history of hunter-gathe rer studies in North American archaeology. With regard to archaeology in the S outheast, it stems from development of the evolutionary cultural categories used to organize prehistoric social groups based on their material remains, notab ly their subsistence technology. Background: The Archaic Period in the Eastern Woodlands The Archaic Period (10,000 to 3,000 BP) in the eastern woodlands of North America represents a series of dramatic socio-cultural change s among hunter-gatherer populations. Following the close of the Plei stocene geologic epoch, groups of huntergatherers spread out across the continent. W ith time, regional specializations emerged as people dealt with unique cultura l and environmental systems. Archaeologists working in recent years have successfully identified a number of changes that occurred during the late Middle and Late Archaic (6000 to 3000 BP) such as population increase, decreases in territorial ranges, the expa nsion of long-distance exchange networks, increased sedentism, the development of pottery tec hnology, the first constr uction of monumental architecture, and the beginnings of cultural complexity (Anderson and Joseph 1988; Blanton and Sassaman 1989; Goodyear et al. 1979; Jefferies 1996; Marquardt 1985; Russo 1994; Sassaman 1993, 2004; Sassaman a nd Anderson 1995; Saunders et al. 1994; Saunders et al. 1997). Increased sedentism and social complexity are particularly noteworthy in this discussion because it is believed to coincide with the exploitation of abundant food 15


resources occurring in restrict ed localities on the landscape. An outstanding example of this is the large shell middens in the Green and Tennessee river valleys in Kentucky and Tennessee, respectively. Known initially to archaeologists as the Shell Mound Archaic," such shell-bearing sites also occur in Louisiana, Georgia, Florida, and northern Alabama (Claassen 1996:235; Russo 1996). An unintentional shortcoming of this research has been an emphasis on hunter-g atherer landscape ma nipulation through the construction of shell mounds only in areas where the environment is naturally rich in resources that leave conspicuous traces (i.e. shell middens), while less obvious environmental niches have been given considerably less attention. The above listed expansion of characteristics related to Archaic Period huntergatherers represent not only the hard work and creativity of interested investigators but also a wider intellectual accep tance that these sorts of cultural achievements happened much earlier than was previously believe d. Southeastern archaeology had long been mired in an evolutionary trajectory of cu lture change that did not acknowledge or consider complexities and cultural successes of early peoples (Russo 1994), an attitude that can easily be discerned with a review of historical characteri zations of temporal periods in the Southeast. Generalizations of the Archaic Period for example, have underscored its portrayal as a time of no madic hunter-gatherers who lived in small groups, did not possess pottery, did not practice plant or animal husbandry, and relied primarily on stone tools (Willey and Phillips 195 8:107). It was believed that not until the later Woodland Period (3000 to 1000 BP) did pe ople begin to experiment with plant domestication, which brought w ith it the need for longer-te rm storage and preparation techniques that required potte ry. Following this line of thinking, people also did not 16


possess the political organizational capabilities requi red by the construction of monumental architecture until much late r during the Mississippian (1000 to 1500 BP) (Smith 1986; Steponaitis 1986). While these s tages of advancement have long since been more thoroughly clarified, broad-scale alterations to the environment by huntergatherers has continued to receive little attention from the southeastern archaeological community. With regard to the Archaic in eastern North America, hunter-gatherers are more often than not cast agains t the static backdrop of the environment. It was held to influence their movements and land-use decisi ons, which were predicated on factors that were beyond their control. Their movement s are considered largely seasonal and in response to the availabi lity of food resources. Recognition of landscape manipulation during the later Mississippian Period has become commonplace in recent years (Del court and Delcourt 2004; Hammett 1992; Wagner 2003). This has also coincided with a much wider acceptance of native peoples having the ability to influence the broader lands capes in which they live (McGlade 1995). In many instances these practices had been going on long before the arrival of Europeans. North American examples include the fore sts of New England, the Cumberland Plateau region in eastern Kentucky, th e lower Little Tennessee Ri ver valley in Tennessee, the Green Bay area of northern Wisconsin, a nd southwestern Oregon (Cridlebaugh 1984; Delcourt et al. 1998; Dorney and Dorney 1989; LaLande and Pullen 1999; Patterson and Sassaman 1988). Landscape manipulation is by no means restricted to North America, as Heckenberger et al. (2003) ha ve recently demonstrated in the Amazon rainforest. Taken together, these studies represent a dramatic departure from earlier descriptions which 17


depict the region as a pristine natural envi ronment as opposed to one that was heavily influenced by human activity. Land-use activities relate d to Late Archaic (5000 to 3000 B.P.) hunter-gatherers across the interior Coastal Plain will be th e primary focus in this dissertation. The significance of this pursuit is based on the present state of knowle dge regarding people who were living in the area dur ing this time, as well the dramatic increase in the number of Late Archaic sites in the region. This expansion in site distribution is thought provoking because it comes on the heels of what is commonly referred to as the Middle Archaic problem by archaeologists working in the interior Coastal Plain. Various aspects of the Middle Archaic cultural chronolo gy in the interior Coastal Plain are not well established, which has led to a number of working models to account for it. Models range from the outright abandonment of the region due to assumed resource impoverishment coinciding with the widespread influx of pine species, to the presence of a yet-to-be recognized cultural chronology ba sed on stemmed hafted bifaces as opposed to tapered and rounded ones that are recognized farther north in th e Piedmont (Blanton and Sassaman 1989; Elliott and Sassaman 1995; Faught and Carter 1998; Kowaleski 1995; Sassaman 1991). While addressing problems related to th e Middle Archaic cultural chronology is beyond the scope of this dissertation, counter ing several of the preconceived notions related to the environmental pa ssivity of hunter-gatherers and life in the interior Coastal Plain are its focal points. Much more em phasis will be placed on the agency of huntergatherers to impact the environments in whic h they live whether from an intentional or unintentional standpoint. Greater focus is also be placed on understanding the 18


outstanding biodiversity of the interior Coastal Plain. Both of these problems may be addressed through emphasizing th e historical ecology of the region, paying attention to human activities in addition to the ecologica l significance of fire and the potential relationship between the two. Emphasis will be given to the overall distribution of archaeological sites recorded during the fieldw ork portion of this project and their broadscale relationship to landscape patches within the project area. Taking this approach will allow for the reevaluation of land-use during the Late Archaic, the ecology of huntergatherers, and the overall environmental pe rception of the interior Coastal Plain. Changing Perspectives on Hunter-Gatherers Tracing the study of prehistoric hunter-gathe rers back through the development of southeastern archaeology finds them firmly ensconced in the Archaic Period. While we now better understand that groups living before and after this expansive time period also likely pursued an existence predicated on hunting and gath ering, no other cultural time period in North American archaeology is more synonymous with this way of life. The heavy influence of the environment has been li nked with the Archaic from its beginnings. In its early inception in the archaeological literature, the "Archaic" stage was largely described from a technological standpoint. W illiam Ritchie (1932) fi rst coined the term after he encountered pre-ceramic artifact midde ns in the state of New York. It was later expanded to encompass areas in the lower S outheast through the work of W.S. Webb and David DeJarnette in Kentucky and Alabam a (Webb and DeJarnette 1942) and further efforts suggested the term may have been a pplicable to archaeological materials as far west as California (Willey and Phillips 1958:104). The Archaic stage was ascribed more of an environmental emphasis in the Southeas t by Joseph Caldwell (1958). His theory of 19


primary forest efficiency, was predicated on the idea that the Archaic was a time of settling in by which groups mapped on or became accustomed to their surroundings and began to take advantage of a plentiful and diverse natural resource base (Caldwell 1958). Contemporary research pertaining to prehistoric hunter-gatherers in the Southeast is pervasive and covers a nu mber of unique temporal a nd geographic areas (e.g., Gibson and Carr 2004; Randall 2007; Sassaman 2005; Saunders and Hays 2004; Thompson 2006). Much of our greater understanding of hunter-gatherer lif estyles in North America, however, has grown out of early attempts to create a near continental-wide stage based initially on similarities between assemblage s of stone tools with little regard for individual developments stemming from re gional variation. A portion of the problem stems from the development of cultural chronol ogies based on the recovery of temporally sensitive diagnostic artifacts which have been extrapolated to cover areas where they may not be applicable (e.g., Broyles 1966; Coe 1964). With time, greater recognition of local variations was achieved as differences among material remains came to the forefront. An early discussion of prehistoric hunter-gatherer lifeways and their material correlates in the eastern woodlands was pres ented by Willey and Phillips (1958) in their overview of the Archaic Stage for the regi on. Their discussion tied in to the greater issue of the Archaic Stage as a whole, it s relatedness to the previous Lithic Stage, and the overall utility of assigni ng it early and late divisions. The Lithic Stage as defined by Willey and Phillips coincided with what present researchers designate as the Paleoindian Period. With regard to the Archai c, the authors were quick to highlight the many parts of its linked material culture su ch as flaked tools and ground stone. They 20


even went as far to question the utility of hunting and gathering as a descriptor for Archaic groups and suggested that the hunting of large mammals may be more accurately associated with Lithic Stage groups while the Archaic represented the earliest groups dependent on gathering (Willey and Phillips 1958:107). The "Archaic" as described by Willey and Phillips (1958) is interesting because it was the first use of the term in a cultural developmental context. An outstanding point of their discussion was the emphasis placed on the development of tem porally sensitive categories for lifestyles based on the interpretation of ma terial remains. Unlike late r discussions, little notice was paid to the overarching environments and th e potential effects they may have had on the people who lived in them. In contrast to Willey and Phillips ( 1958) general characterization of Archaic groups, Jesse Jennings (1968) and Joseph Caldwe ll (1958) paid much greater attention to the environment in their discussions of the period. Both archaeologists countered the strong generalizations regarding hunters and gatherers as presented by Willey and Phillips and instead cast them as being mu ch more dynamic and in tune with their surroundings. In this light they were percei ved as being much more sophisticated with regard to their regional specializations. Caldwell's Trend and Tradition in the Eastern Woodlands was a refreshing example of the th en-new approach to the study of hunters and gatherers. He opposed the stance of Willey and Phillips as being too overly simplified in scope and instead posited that through time people mapped on and adapted to the various environments in whic h they lived (Caldwell 1958). From this perspective regional differences could be explained through environmental variability. 21


In many ways Caldwell's reevaluation of Ar chaic peoples and their relationship to the environment set the stage for later devel opments that took place during the processual phase of Americanist Archaeo logy. Researchers associat ed with this paradigm downplayed independent invention, migration, and diffusion as the primary drivers of culture change as championed by culture hist orians. Under their burgeoning influence, changes in culture were increasingly viewed as occurring in conjuncti on with shifts in the environment. A great deal of research pertaining to hunter-gat herers was undertaken during this time as archaeologists such as Lewis Binford and others sought out common experiences with contemporary living groups in order to understa nd how they organized themselves in relation to encountering food resources. This approach certainly moved well beyond the scope of Willey and Phillips but it also broached much broader aspects of how people viewed and interacted with the encompassing landscape (Binford 1979, 1980; Kelly 1983). One of the most influe ntial approaches to categorizing huntergatherer forms of social organization duri ng this time was Binford's (1980) foragercollector dichotomy. In this model Binf ord predicted that based on the overall distribution of natural resources that people would organize themselves along a continuum from foraging to collecting. Following the model, foragers would be residentially mobile and would move their entire residence to exploit a new foraging zone. At the opposite end collectors were expe cted to primarily in habit longer-term base camps and would utilize a series of small, short-term extractive camps to acquire resources which would then be removed to the centralized locality (Binford 1980). Importantly, people were not bound to either mode of resource acquisition as both methods could be used interchangeably depending on the overall distribution and 22


frequency of plant and animal foods. While it has since been critici zed for its lack of emphasis given to land-use decisions made not only in relation to f ood resources but also within a broader social sphere of interact ion stemming from relationships with other groups of people living on the landscape (W eissner 1982), Binford's model was highly influential and has continued to be a drivi ng force underlying attempts to understand the archaeological signatures of hunter-gatherers. All of the above mentioned approaches to the archaeological record of huntergatherers have greatly advanced our current und erstanding of hunter-gatherers in general. In many instances each researcher built on th e foundation of the other to create a more precise, detailed picture of the cultural activities or behavioral adaptations of past groups. There is, however, a similar theme among them based on their charac terization of huntergatherers and their relationshi p with the environment. All too often, hunter-gatherers are investigated against the back drop of environmental variability. This approach does not implicate human ingenuity or the ability of peopl e to act in their own self interests in the face of environmental strife. While few people today look to Caldwell's model of "primary forest efficiency" for theoretical guidance, he was partially accurate in his portrayal of people bein g in touch with their surroundings to the degree that they could efficiently choose from the environmental and material resources surrounding them. Working from a better informed perspective about ecological system s and how they can be impacted by people, as well as more sophisticated data collec tion and evaluation to approach these issues at a regional level, c oupled with the new agency-centered historical ecology approach utilized in this study, presents the opportunity to develop more thorough models of land-use that encompass additional variables beyond just climate 23


shifts and general forest cover. We now ha ve the capacity to better evaluate human-landuse interactions and can approach landscapes as places that are actively shaped by the people who live in them. Following the early infl uence of Caldwell, it is time to consider that people not only had an intimate unders tanding of their surroundings but that they also had the knowledge base to actively in fluence their broad-scale distribution and predictability of resources that were important to them. Historical ecology offers th e advantage of considering the implications of human land-use decisions as having the potential to dramatically shape ecosy stems. Its use is outlined in the following section. This persp ective serves to treat the landscape itself as an artifact marked by the behavior of peopl e and allows for the broader evaluation of the classification of hunter-gatherers, partic ularly in relation to the age-old culturalevolutionary typology used to categorize socie ties through time (Sahlins and Service 1960; Service 1962). While Archaic Period peop les did not likely domesticate individual plant or animal species in the widely accepted sense, they surely possessed the capacity to domesticate broader landscapes by influenc ing species distributions. Similar to the early arguments concerning hunt er-gatherers in the South east and their ability to construct large earthen and shell structures and otherwise partic ipate in socially complex activities usually reserved for larger agricultural societies, clear divisions such as hunter-gatherer and agri culturalists may be too broad and rigid in their current implementation to accurately characterized people in the past. The Theoretical Approach of the Dissertation Historical ecology provides the theoretical platform for this research. It is amenable here because of its emphasis on the recursive relationship between culture and 24


nature which is important when considering aspects of humanized landscapes. The final outcome of this research is potentially significant fo r three reasons. Firs t, it counters the long-held perception of environmental pa ssivity among hunter-gat herers. Second, it helps better evaluate people in an environment previously in terpreted as having little to offer them. And third, the extent to which people impacted forest cover and how this may have in turn affected settlement stra tegies can be quantified. This dissertation compliments a growing body of research that ut ilizes a historical-eco logical approach as a viable alternative to explore the dy namic relationship between people and the environment (Balee 1998; Crumley 1994; Marqua rdt and Crumley 1987). As noted, this allows for a more encompassing approach when considering this relationship, which differs from the deterministic elements of cultural ecology, and the lack of environmental consideration of many st rictly socially derived theories. Historical ecology considers that people have had a long-standing relationship with the environment and in many instances have altered it (Crumley and Marquardt 1990; Cr umley 1994). This dir ectly relates to the ongoing debate regarding the extent of environmental impact by Native Americans (Cronon 1996; Denevan 1992; Vale 2002). The landscape is not just the physical e nvironment; it is also historically and culturally determined (Balee 1998:16). From this perspectiv e, the relationship of people and the environment is much more interac tive and territori ality goes beyond mere access to natural resources when people have a vested interest in a region ba sed on the degree to which they have made, modified, or inherited it. This interaction manifests itself in what Marquardt and Crumley (1987:1) refer to as th e landscape, or the sp atial and historical relations of people and their environment. In this way, the environment is more than a 25


back drop against which the lives of people pl ay out, but rather is actively engaged and manipulated. The entire landscape can be view ed as an artifact, or product of human activity based on the distri bution and manipulation of food resources. Hammett (1992:133) echoes this statement in her Functi onal landscape, which she defines as a landscape that meets the needs of the local inhab itants. The appropriate ness or fit of the overall structure or design of a landscape is part of a regions ecological history. Following Marquardt and Crumley (1987), a landscape is a region recognized for its scale and distinctiveness compared to neighboring regi ons. The interior Coastal Plain of the lower southeastern Unite d States is amenable to a historical ecological research strategy. It meets the regional criteria es tablished by Marquardt and Crumley (1987:3) based on its distinctive spatial configuration compared to th e Piedmont and Coastal Zone, with which it shares a northern and sout hern conterminous boundary, respectively. Fireadapted ecosystems comprise a significant porti on of the forest cover in the Coastal Plain. Burning is a historical com ponent in many areas based on th e occurrence of fire-adapted plant and animal species. Fire in the Forest: The Influence of Anthropogenic Burning Research over the last twenty year s provides increasing evidence that Native Americans actively modified broad areas of North America (Lewis 1982; Patterson and Sassaman 1988; Pyne 1982, Stewart et al. 2002). Historical accounts indicate that fire was used extensively in many of these pur suits (Bonnicksen 2000; Pyne 1982; Stewart et al. 2002) and that various habitats were burned to control the overall distribution, diversity, and abundance of plant and animal resources (Lewis 1982: 45). Intentionally set fires helped create and maintain a series of microhabitats, or edge habitats that were 26


useful for hunting and gathering (Krech 1999; Lewis 1982:51). Evid ence of intentional burning by hunter-gatherers has been reported from Au stralia (Head 1996), North America (Lewis 1977, 1982); South America (M eggers 1966), and Asia (Lewis 1972). At contact, Europeans found areas on th e North American continent which had likely long been maintained by anthropogenic fire, evidenced by an open, park-like forest cover across broad areas of the continent (Bromley 1935; Cronon 1983; Denevan 1992; Krech 1999). While historical evidence suggest s that fire was regul arly implemented by Native Americans, Bonnicksen (2000:147) sugges ts that similar strategies extend back to the Paleoindian Period. He notes that fires were set to ame liorate dense underbrush, and encourage the growth of grasses, forbs, shr ubs, and nut-bearing trees that were attractive to game animals. Anthropogenic fire as oppos ed to lightning fire is largely held to account for many fires in prehistory, based mo stly on the idea that the latter is too infrequent in many areas to regularly pr ecipitate fire (Bonni cksen 2000:146; Pyne 1982:12). Fires set by humans would likely be necessary to regularly maintain many areas. Fire has a pronounced effect on ecosystem s in that it burns undergrowth and returns nutrients to the soil in the form of ash. Following the initial burn, insects descend on an area to take advantage of scarred trees. Burned areas in turn attract birds and other insect predators. Herbivores are also attracted to burned areas after vegetation begins to regenerate. Overall, there is an increase in plant and animal species following a burn (Whlelan 1995:238). The initial increase is short-lived, with the effects of fire lasting five to six months before the system be gins to stabilize with a subsequent decrease in species diversity. Additionally, some plan t species require fire to produce seeds. 27


Lewis (1982) suggests that fires set in th e early spring would encourage vegetation regrowth by as much as two to three weeks ear lier than if left to sprout on its own. The ability to influence vegetati on patterns would likely impact hunter-gatherer land-use. This is significant when considering the floral composition of the Coas tal Plain and social territoriality during th e Archaic Period. The interior Coastal Plain is a notab le geographic region for this proposed research endeavor because large portions of it are (or were) covered by fire-adapted ecosystems. Regular, low-intensity fires in longleaf pine-wiregrass ecosystem are necessary to ensure forest health and longevi ty. This is significan t considering recent emphasis on the infrequency of lightning induced fire (Pyne 1982). The relationship of Native Americans and fire has been studied in other areas of the North American woodlands such as Maine, Minnesota, and Alaska (Anderson 1979; Patterson and Sassaman 1989; Shackleton 1979). Davis ( 1933) has suggested that by the time of European contact, prehistoric burning had extensively modified the forests of the Northeast. Bromley (1935) also states that at contact, eastern forests had an open, parklike appearance, a point reiterated by Krech (1999). Current models of Coastal Plain land-use during the mid-Holocene do not consider the full extent of the cultural landscap e. In this case, fire may have been a powerful tool employed by Coastal Plain inhabi tants to influence the ecological structure of the region. Recent research in the southeastern Coastal Plain of Georgia suggests that fire increased in relation to human settlement during the Archaic Period (Lamoreaux 1999; Seilestad 1994). With the means to influe nce food resources, it is likely that the Coastal Plain was much more culturally dynamic than has prev iously been considered. 28


This adds an additional social component to Archaic Period hunter-gatherers in the lower Southeast considering the id ea of distinctive social groups inhabiting different physiographic areas, i.e. Piedmont vs. Coastal Plain. The landscape is considered to have changed through time by becoming more soci ally subdivided (Anderson 1996; Sassaman et al. 1988). In addition to experien cing environmental change wrought by climate shifts, the Coastal Plain environment was also likely regularly ma nipulated by people to encourage the predictability of food resources. Research Objectives Data used in this dissertation were collected during four (2003-2006) summer seasons of field work in southwest Georgi a. The project area encompasses diverse biomes, with riverine settings a nd pine forests interspersed w ith wetlands of varying size and structure. Survey was undertaken with the hope of recordi ng a large number of surface exposed sites that could be collect ed. Historic and contemporary land-use management strategies aided in this endeavor because years of ine fficient agriculture and timber practices left many sites deflated and lying on the ground surface. While this made distinguishing intersite activity areas difficult it did allow for sites to be located and delineated. Evaluation of these sites is based on the relationship of stone tools such as hafted bifaces, unhafted bifaces, cores, c obbles, and debitage recovered from them. Combinations of these artifacts or lack thereof are used to develop site types for each locality. This allows for the enhancement of a local-level model of land-use for the interior Coastal Plain and mark s the first research objective he re. An undertaking of this sort has never been attempted in this area a nd its completion here will provide a better understanding of how people were living in the region during th e Late Archaic. Similar 29


undertakings have been successfully completed in areas further north in the Coastal Plain such as the Middle Savannah River Valley (Sa ssaman et al. 1990) but none this far to the south. Many Late Archaic land-use mode ls focus on areas of high food resource productivity such as riverine or coastal settings Similar models for areas of the interior remain undeveloped. A consequence of this si tuation is the forcing of land-use models into the region that were developed from recognized patterns in other areas. For example, nuts have long been held as an im portant food source for people inhabiting the eastern woodlands (Caldwell 1958, Gardner 1997:161), and land-use models for Late Archaic hunter-gatherers in th e Piedmont and interior Plat eau are generally based on the availability of these and othe r seasonal resources. Nutproducing species of hardwoods are present in the Coastal Plain but not in th e same frequency as farther north and west. It is not however, without its own types of other resources, particularly wetlands, conducive to sustained human settlement. There is a need for a more thorough understanding of local-level land-us e in the interior Coastal Plain. Revaluating the Interior Coastal Plain The interior Coastal Plain is unique and is generally characterized as being a coarse-grained landscape (Sassaman 1991). This characterization is held to have impacted prehistoric land-use decisions. All too often, the region is cast as having little to offer people in the way of food resources due to the pine species being the predominant forest cover (Fish 1976; Lars en 1980). Coupled with the pyrogenic dependency of longleaf pine that dominated the region, a distinctive picture of human existence in the Coastal Plain becomes apparent As a hypothesis in n eed of testing, life in the interior Coastal Plain was distinctiv e and required a differe nt type of land-use 30


compared to people living in coastal areas or in the Piedmont. The combination of ponds, upland pine areas, and local streams and rivers creates a patchwork or ecotone landscape that is unique to the Coastal Plain. Instead of being portrayed as a region that lacked adequate food resour ces to support large numbers of people in prehistory, especially during the middle Holocene when pine species supplan ted the oak-hickorysouthern pine forest (Delcourt and De lcourt 1985; Watts 1971; Watts et al. 1996; Whitehead 1965), the Coastal Pl ain can instead to be viewed as the partial product of human activity. In contrast to previous descriptions of it as a homogenous region, the Coastal Plain actually encompasses many di verse ecosystems like the longleaf pinewiregrass forest which has been characteri zed as having some of the highest levels biodiversity among global ecosy stems (Early 2004; Forman 1979; Goebel et al. 1997). Rethinking Hunter-Gatherers Landscape management by hunter-gatherers has not been emphasized in previous models of Late Archaic land-use in the lower Southeast. As noted earlier, this stems largely from an evolutionary model that si tuates humans to a passive and dependent relationship on natural environm ents until they harnessed technologies and organizational structures to produce food. Their lack of attrib utes associated with larger societies such as group size, stratified classe s, and agriculture places them at the lower end of the development scale. Consequent ly, their ability to modify th e environment is given scant attention, or precluded altoge ther. Besides accounting for a consideration of increased food resources, land management also epito mizes an investment in the landscape, which has implications for territoriality, group permanence, and sociocultural complexity. By considering more fully the potential hunter-gatherers have to create environments of 31


their design, and thus affect the course and pace of change, we can appreciate that the cultural evolutionary boundary separating food-producers from foragers is untenable. Organization of the Study Following this introductory chapter, th e second chapter will outline historical ecology as the guiding theoretical perspective of this dissertation. This will proceed with an overview of its development based on its re lationship with the br oader field of ecology and how it has developed under the guise of anthropologists and archaeologists. While this is presented initially from a historical perspective it ultimately provides a more indepth view of its applicati on in this dissertation beyond what is stressed in the introduction above. Preliminary emphasis is given to the development and meaning of historical ecology followed by examples of loca lized applications which highlight the use of fire by native peoples to improve the f ood resource predictability in the various landscapes in which they live. Chapter 3 provides an introduction to th e research area within the expansive territory of the interior Coastal Plain. It covers the parameters of the individual survey areas and where they occur within the greater project area. Each of the survey areas is relatively unique based on forest cover, elevation and hydrology so they will be presented individually and their significance related to the broader project. Field methods used to record archaeological sites in the project area are also highlighted. This is followed by an introduction and overview of the longleaf pi ne-wiregrass ecosystem that once covered a large swath of the interior Co astal Plain. A major component of this undertaking is to highlight this forest covers biodiversity compared to previous characterizations by archaeologists. Included in this is a detailed discussion of the role th at fire plays within 32


this ecosystem, specifically how regular, low-intensity fires are needed for its continued upkeep and overall good health. A discussion of the frequency of lightni ng strikes is also included here. Its inclusion is important because natural lightning induced fires have long been held to be responsible for the sp read of pine species throughout the Coastal Plain. Wetland resources and their place within in the project area will also be described Chapter 4 will cover the analysis of st one tools recovered during the survey. Discussion will focus on the variables for individual tools and debitage as outlined above. Using this as a starting point I will evaluate the relationship between tools and debitage at individual localities in order to create site s types in the project area. This will be followed in Chapter 5 by a broader analysis of sites in relation to fire potential. Conclusions from the project and future directions additional research should follow are outlined in the sixth and final chap ter of this dissertati on. This section will provide the culmination of all the work presented here. As was stated above, a primary goal of this work is to present a reconstructi on of Late Archaic life in the interior Coastal Plain within the framework of historical ecology. By pursuing this line I intend to cast prehistoric hunter-gathere rs who lived in the region as actively engaging the environments in which they lived instead of being environmentally passive. I also hope to move beyond the conclusions drawn in this di ssertation and use them as a stepping stone for further research in the interior Coastal Plain, such as palynological work to gain a better grasp on prehistoric fore st cover, and to quantify ch arcoal counts from organic sediments to better evaluate the extent to which local hunter-gatherers were actively managing areas of the region with fire. 33


CHAPTER 2 HUNTER-GATHERERS, THE ENVIRONMENT, AND HISTORICAL ECOLOGY Native peoples are often assumed to pursu e a harmonic existence within pristine natural environments. Whether this assumpti on is explicitsuch as the classic image of Iron Eyes Cody, also known as The Crying I ndian, who appeared in the Keep America Beautiful anti-litte r campaign of the early 1970sor imp lied more recently in debates regarding the broad-scale fate of Amazonian rainforest s (Heckenberger et al. 2003), native identities are frequently conscripted and recast as being environmental stewards (Kottak 1999:26). This is particularly true of Native Americans whose imagery is often associated with a distant, idyllic time in which they lived in peace and harmony with nature and took only what they needed from the land. Following this line, it was not until the arrival of Europeans with their industries of greed and destruction that the previously unmolested environments of the Americas began to decline (Krech 1999:16). An outstanding historical component is also associated w ith this issue as Europeans and later European Americans ofte n struggled to incorporate native peoples into their broader world view, a dilemma that is easily seen in early depictions of them. More often than not, these illustrations empha size heavily the naturalness of the worlds in which native peoples lived. Observations of these constructe d identities are also distinguished in the names used to define them, such as the Noble Savage, or more recently as the Ecological Indian (Krech 1999:17). Even the label savage is noteworthy here, as pointed out by Krech ( 1999), which was implemented by European scholars as a means to represent Native Americans as pursuing an animal-like existence or otherwise living in a state of nature. Combined, these presentations of both nature and indigenous peoples have worked to skew the general perceptions of how they are 34


viewed and understood. The implications for th is are significant gi ven that stereotypes associated with both Native Americans and the broader landscape ar e often misguided, the former for having little impact on natu re and the latter for being pristine and untouched. These stereotypes ha ve also been extended to th e portrayal of prehistoric hunter-gatherers, especially in relation to the lack of attent ion given to them regarding their ability to stimulate environmental change A portion of this problem stems from the ways in which contemporary hunter-gatherers were initially charac terized and studied, first by explorers and settlers a nd later by anthropologists. C oupled with this is the early portrayal and management practic es related to historic peri od forests which were often considered to have been relatively unaff ected by the long-term cultural practices of people. Debate regarding the extent of native in fluence on broader-scale landscapes has become commonplace in recent years, especi ally considering growing conservation efforts to restore some areas back to a perc eived pre-colonial stat e of nature. Greater consideration of long-term human influence has also become routine as environmental historians have placed increased emphasis on long-term anthropogenic impacts as well as some ecologists questioning the balance and pr edictability of climax ecosystems (Cronon 1995:25). These renewed efforts to place na tive peoples within a wider sphere of environmental influence have not been wit hout their share of critics however, as many researchers still question the ut ility or the accuracy of qua ntifying the affects people may have had on the landscapes in which they lived (Forman and Russell 1983; Russell 1983; 86; Vale 2002). While research lately has focused greater attention on the exploits of native groups, they all too of ten concentrate heavily on agriculturalists and changes they 35


made to enhance conditions that were c onducive for growing domesticated crops. Unfortunately, within this line of thinking, hunter-gathere rs and the environment are often assumed to be synonymous. A reevaluation of regions like the interi or Coastal Plain offers the opportunity to not only develop new models of prehistoric la nd-use that incorporate the utilization of broad-scale areas but also to rethink the utility of the general dichotomies such as hunter-gatherers and agriculturalists that have been used to categorize people in the past. In many instances, these categories may be too narrow in scope, particularly regarding hunter-gathe rers and the relationships they developed with their surrounding landscapes. The remainder of this chapter will focu s on the study of hunter-gatherers and how it has ultimately impacted how anthropologi sts perceive them, alternative ways of thinking about them, and how romanticized perceptions of the environment have furthered skewed ideas related to native peoples and environmental passivity. The history and shortcomings of each issue will be outlined to lay the groundwork for an indepth discussion of the relevance of this approa ch at the conclusion of the chapter. As a starting point, however, I pres ent an overview of the re lationship between ecology and anthropology and the overall re search potential of historical ecology in relation to both, especially regarding the increas ed consideration of anthropogeni c fire, as being useful for the reinterpretation of hunter-gatherer land-use in the prehis toric past. Discussion of these points will set the stage for a presentation of historical ecology and how it offers a good opportunity to reevaluate how archaeol ogists think about hunter-gatherers, the environment, and interactions between them. This will be followed by an overview of 36


various methods in which researchers have used to evaluate human-environmental impacts and examples of how people in the pa st have implemented anthropogenic fire to enact broad-scale environmental manipulation. Background, Definition and App lication of Historical Ecology In its purest sense, ecology is the st udy of organisms in relation to their environment. Taking this perspective, many anthropologists have co-opted and modified aspects of ecological theory fo r their own research agendas. The two fields have thus enjoyed a long collabora tive relationship beginning in th e 1950s with the development and implementation of cultural ecology by Julian Steward (Netting 1986:1). Steward first coined the term cultural ecology in an attempt to apply eco logical concepts to account for the association between particular societies and their na tural environments. He contended that small-scale societies were heavily influenced by the natural environment. Even though his use of ecology was criticized for being environmentally deterministic, Stewards application of cultu ral ecology was relevant for its time as he was ultimately taking culture as his unit of analysis and attempting to account for culture change. Of course, th is reliance on the environment as an explanation for culture change would also become a central component, as discussed in Chapter 1, of the processual archaeology m ovement during the 1960s. Due in part to simultaneous occurrences in both the fields of ecology and anthropology, the focus of eco logical anthropology shifted during the 1960s, especially after the publication of Rappaports (1968) Pigs for the Ancestors Based on his work with the Tsembaga Maring in Papua New Gu inea, Rappaport espoused a different form of ecological anthropology. As an analytical alternative to cultures of Stewardian 37


cultural ecology, Rappaport stressed the struct ure and function of ecological populations and ecosystems (Biersack 1999:8; Kottak 1999 :24). Rappaports focus was primarily systemic and he used the Maring kaiko as an example to emphasize the capacity of negative feedback within a system. For example, Rappaport interpreted the kaiko from a functional perspective which he believed wa s predicated on the need to maintain a balance between, pigs, warfare, and people (Biersack 1999:6). He thus defined the kaiko as a pig sacrifice to ancestral guardians in times of war. Warfare was avoided while debts to supernatural and ordi nary allies were outstanding (Biersack 1990:6). The sole opportunity to fulfill this debt was through the slaughter of pigs. To prepare for this event pigs were accrued and fattened. With time, as the herds expanded in size, increased pressure was placed on Maring women who were charged with caring for them. When the demands on their labor exceeded its breaki ng point they would encourage the massive slaughter of pigs and the kaiko would then be underway. Rappaport (1968:6) contended that this practice contributed a number of posit ive elements to the groups well-being. It first restored the human to pig ratio to a bear able level, it helped maintain local biotic communities, and it limited the frequency of fighting as well as distributed a substantial supply of pork across a large regional popula tion (Biersack 1999:6). While Rappaports application of ecology provided new insi ghts into how human culture could be analyzed and understood within a broader eco logical sphere, it was nonetheless criticized for its functionalism and systems-based approach (Biersack 1999:7). Rappaports use of ecology is relevant re garding the application of historical ecology to the research problem presented in th is dissertation for a number of reasons. As a starting point, histori cal ecology represents a depa rture from the way in which 38


Rappaport used ecology in his attempt to ac count for human activ ity within a wider ecological context. His systems-based appr oach relied on the concept of the niche regarding an organisms place in the environm ent. Following a summary of this line of reasoning by Ingold (1992), researchers associ ated with Rappaports brand of ecological anthropology tended to view the environmen t as being comprised of a series of many alcoves into which organisms fitted themselves. If an alcove was abandoned or otherwise unoccupied then another organism could take up residence within it (Ingold 1992:41). While Rappaport did ma ke a bold step in recognizing differences between a groups cognized environment or how people pe rceived their world, and their operational environment, which is how it actually opera tes, he nonetheless separated nature and culture into separate parts as opposed to rec ognizing them as interc onnected entities. Historical ecology accentuates the diverg ence from this line of thinking as it acknowledges that the environment is hi storically and culturally produced through human-nature interactions (Balee 1998:16; Biersack 1999:9). By focusing on the production of space and its associated features and accompanying activities, it becomes apparent that humans leave a pronounced footprint on the landscape (Crumley and Marquardt 1990; Crumley 1994; Kirch and Hunt 1997; Lansing 1991). Crumley (1994) takes this a step further in her discussion of aspects of the physical landscape which retain the practices, ideas, and thoughts of the people who have lived there. Pursuing a historical ecological perspective thus forces a more encompassing approach when considering the relationship be tween people and the environm ent, notably a deeper-time perspective than cultural ecol ogy. Some of the earliest roots of this thinking may be traced back to the writings of Marx and E ngels where they argued that a nature which 39


preceded human history no longer exists anywhere (Biersack 1999:9). Simply put, people have long had a hand in shaping environments. This is an especially important point for considerat ion when thinking about the potential of impacts made to the envir onment by hunter-gatherers. Beyond just reevaluating the impacts of hunter-gathere rs, Balee and Erickson (2006) emphasize the need to focus attention on the historical landscape as a starting point. For them, the historical landscape is a multidimensional ent ity that is ultimately the product of longterm human activity (Balee and Erickson 2006:1) The physical landscape can thus be read like a text because it is the reflection of the cultures that in teracted with it and inferences about them may be made based on the material remains of traces of patterned behavior that are inscribed on the landscap e (Balee and Erickson 2006:2). While other sciences may distinguish betw een landscapes with or without human influence, they consider the human species as a keystone species that alters landscapes primarily through disturbance mechanisms, such as fire (Balee and Erickson 2006:2). These practices, they contend, have in creased the richness and availability of food resources in different times and places by augmenting biodiversity. By acknowledging this perspective, the human species can be understo od as a principal mechanism of change in the natural world, a mechanism qualitatively as important as natura l selection (Balee and Erickson 2006:5). Environments, therefore, ar e in a sense adapted to the sociocultural and political systems that have coexiste d with them for potentially long spans of historical time. Evidence for these long-term connections can be observed in the present landscape based on the previous selection of certain economic species (Balee and Erickson 2006:7). As will be demonstrated, historical ecology offers a new perspective 40


on the environment and how people interact with it. This is significant considering the ways in which both have previously been ch aracterized by archaeologi sts, historians, and natural scientists. An Environment of Hunter-Gatherers From its earliest inception in the anthropologi cal literature, the term huntergatherer has described peoples who did not practice agriculture but instead relied on hunting and gathering as a means to acquire their daily sustenance. Steward (1949) for example, presents a brief description of th is lifestyle in his discussion of the "PreAgricultural Era" of human-cultural developm ent. He goes on to depict them as people whose technology is primarily geared towa rds "satisfying biological needs for food, clothing, and shelter" (Steward 1949:10). People enacting landscape-level change do not play a factor until his later discussion of gr oups involved in incipient agriculture. Much of the early research related to the history and discussion of hunter-g atherers is based on work in Africa, which presents a number of historical problems related to colonization and a general lack of written records within the region. While the importance of categorical te rms like hunter-gatherers cannot be denied because they provide a common language that multiple resear chers can rely on to communicate, they fall short in some inst ances when people are pigeonholed as opposed to being understood within their own contexts. A prime example of this stems from the anthropological discussion regarding its use based on research relate d to groups centered in Africa. Much of the ear ly research related to hunte r-gatherer ethnographic research was undertaken in conjunction with the San groups in the Kalahari Desert region of the southern portion of the contin ent. Early anthr opological interest in the region began 41


during the early 1950s as anthropologists were drawn to study people living in the Kalahari due to their perceived remoteness and isolation from other people. These characteristics would also prove influentia l as cultural evolutionism and human ecology emerged at the forefront of anthropologi cal concerns (Wilmsen 1983:10). Huntergatherers, in this regard, were believed by many to encompass the essence of the human condition more than any other groups who could be studied. This was the precise goal of Richard Lee and others who ventured to the area to study people whom they believed to be relics of a far off distan t past that had not changed fo r generations (Kottack 1999:31; Stiles 1992:14; Wilmsen 1983:10-11). The study of them in this capacity was believed to have had the potential to lend in sight into the lives of earlier Paleolithic hu nter-gatherers (Kurtz 1994:68). From this perspective they were held to represent a form of affluent society or basic human ecology that epitomi zed the original condition of humankind and a basic human adaptation (Kurtz 1994:69; Leacock and Lee 1982:5; Sahlins 1972; Stiles 1992:14). An immediate downside to th is approach was the intense focus placed on the basic niche of hunter-g atherers place in nature w ith food-gathering being the primary regulator in how they lived, a position that came at the expense of understanding their greater relationship to the surrounding landscape (Wilmsen 1983:12). This is in spite of the fact they are usually examin ed from an ecologica l standpoint (Netting 1986:8). The perception of Kalahari groups as being cut off from the outside world and thus representing people leading an isolated existence was challenged in what has come to be known as the Great Kalahari De bate when Edwin Wilmsen (1983) openly questioned Lee and others on the grounds that the temporally frozen hunter-gatherers 42


they sought to study had actually not lived liv es in isolation but had in fact been influenced by relationships with neighbori ng groups who practiced agriculture (Kurtz 1994:68; Wilmsen 1989). As a result, increase d emphasis has been placed on the greater world system in which Kalahari groups are a part of (Myers 1988:263). For example, it is now generally understood that groups livin g in the region represent a great deal of diversity regarding their subsistence pract ices, social organi zation, and localized identities. These characteristics pertain not only to Kalahari gr oups but also huntergatherers in general. Often their identities are fluid and may be used interchangeably as circumstances present themselves. For example, agriculturalists may assume practices normally associated with hunter-gatherers in ti mes of stress. There are also examples of hunter-gatherers who were once agriculturalists. A lternatively, people believed to live exclusively as hunter-gatherers may also establish complex trading relationships with their neighbors as a means to acquire resources they desire (Grinker 1994). While debate concerning the utility and accuracy of anthropological terms such as hunter-gatherers is ongoing, an important point to consider in the discussion here is the influence that early ethnographies, such as those undertaken in the Kalahari, had on archaeology, particularly during the development of the New Archaeology during the 1960s and 1970s. Culture as mans primary means of adaptation was held up as a central tenant of this arising paradigm. In this capacity, the environment was often cast as a backdrop against which human culture de veloped. Changes related to culture were often believed to be the result of environmen tal flux. An immediate shortcoming of this approach relates directly to the study of hunter-gatherers and their ability to enact change within the environments they live. This is clear when considering that archaeologists 43


aligned within the new arch aeology actively sought out et hnographic expe riences with contemporary groups of hunter-gatherers whom they believed held ties to a distant past even though they often lived in marginalized settings and had social relationships with non-hunter-gatherers as was discussed above (Binford 1977; Gould 1978). A primary point to remember here is th at the environment was again he ld up as a backdrop against which humans were reactionary actors who were only capable of long-term cultural change in response to widespread environmenta l shifts. In contrast to this prevailing picture, hunter-gatherers should be viewed from a more holistic perspe ctive. While they certainly have the capacity to actively enact broad-scale environmental change they may not necessarily intend to though their actions have the potential unintended consequence of doing so. Pursuing this line requires alte rnative ways of thi nking first about the environment and also how hunter-gatherers percei ve and interact with it, points which are covered in the following sections. The Development of a Pristine Nature Examining perceptions of the various environments of North America from a historical perspective highli ghts the intense romanticism th at coincides with not only their early depictions but modern day assessments of them as well. Much of this idealism began during the 19th century in the writings of He nry David Thoreau, John Muir, and other writers who created and inadvertently perpetuated the myth of a pristine, underpopulated landscape that had expe rienced little influence at the hands Native Americans prior to the arrival of Europeans (Denev an 1992:369). As early American settlers explored and established perman ent residences in different re gions of the continent they often over-emphasized their own achievements at the expense of those made by native 44


populations which further exacerbated the general belief in a previously untouched landscape. Considering that nature is a cultural construct, interpretations and ideas related to it are highly susceptib le to shifts within the broader cultures to which it is connected. For example, in a modern context, wilderness generally refers to areas that have not been fully impacted by civilization which make them islands in an otherwise sea of pollution and environmental degradation (Cronon 1995:69). Cronon (1995) underscores this point in his discussion of modern, urban-dwelling people who eagerly seek out what they believe to be wild, spiritually fulfilling experiences with nature. This view starkly contrasts with earlier inte rpretations of wilderness which tended to emphasize its perceived negative qualities such as emptine ss, savagery, desolation, and barrenness with its only real value being that it might be reclaimed by peopl e and used in a positive manner (Cronon 1995:70-71). To be lost in the wilderness during the early to mid 19th century was a precarious situa tion indeed. By the late 19th century, however, many of the negative connotations related to supposed areas of wilder ness had given way to more modern preservationist ideals as espoused in the writings of Thoreau and Muir who likened interacting with nature to an religious experience (Cronon 1995:76). Coupled with this picture was a general fascination with the American frontier whose mythology also changed through time as more areas of North America became settled. It was during this time that the frontier became intertwined with the broader American identity based on the pervasive belie f that democracy and the national identity had largely been borne out of the actions of European Am ericans who moved into the wild country where they were able to shed the trappings of civiliz ation and reconnect 45


with their primal energies. Similar to the quasi-religious leanings of Thoreau and Muir, wilderness and the frontier were held up as plac es of renewal and thus became an integral component of the American character (Cronon 1995:76). This highli ghts the constructed aspect of nature because it was during this time that the push to set aside natural areas for preservation based on the belief that the frontier and thus a part of the American identity was disappearing (B owden 1992:22; Pyne 1982:17). Interestingly, it was people primarily asso ciated with urban settings who were the driving force behind many early preservation efforts. Cronon (1995:79) suggests that because a metropolitan existence contrasted so drastically with ideas of a natural wilderness, the romanticized, biased view of wilderness was enhanced. Urbanites developed a wilderness in their own imag e and thus furthered the myth of a rough country as virgin, an illusion that they were seeing their na tion in its pris tine, original state. Nature in this cap acity, had to be natural, pris tine, remote from humanity, and untouched by a common past (Cronon 1995:79-83). Bowden (1992) also takes a historical slant in his discussion of ideas related to the development of a mythical, pristine landscape. He divides what he calls The Invention of American Tradition into four phases that roughly coincide with the EuropeanAmerican settlement of various regions of No rth America. The four phases are further subdivided into two major headings, those rela ted to wild people and wild lands before settlement and those of the American people and the landscapes they produced (Bowden 1992:4). All four phases developed much later than the achievements they proposed to depict. Inaccurate images were usually associated with the first phase which later became ensconced and accepted within th e wider historical consciousness. These 46


images would soon after be picked up and perpetuated by writers, such as Thoreau and Muir, and eventually established as an invented tradition that was in time culturally internalized and presented as unquestionable fact (Bowden 1992:4). By the fourth and final phase the tradition pervaded all aspects of the broader culture and became a part of the general knowledge base where it appeared in textbooks related to geography and history and was otherwise taught to the ge neral public as comm on knowledge (Bowden 1992:4). There is also a familiar theme that runs through each of the phases that developed during the mid to late 19th century long after the actual events had taken place and following the final demise of the pre-American landscape. Each phase, however, worked to highlight the achievements of American settlers. They presented a much exaggerated view with regard to the hostil ity of the environments they encountered during settlement. The landscape was held to be much more hosti le to incoming settlers than it had been in reality. Examples include, the Desart wildernesse and Primeval forest of New England, along with the infertile virgin prai ries and the Great American Desert of the Midwest (Bowden 1992:5-15). Combined, th e taming of these geographic areas was believed to be comparable to the accomplis hments that had previously occurred in Europe. An immediate downside to the accepta nce of these beliefs was the belittling of the real achievements that had been made by Native Americans after hundreds if not thousands of years of actively interacting with the Nort h American landscape. For example, early depictions of the eastern woodl ands point out mature forests with an open understory that reflected fr equent ground fires (Denevan 1992:371). Foot and horse travel in this setting was re latively easy in spite of late r stories related to the rugged 47


nature of pre-colonized North American lands capes. Davis (1933) has suggested that by the time of European contact, prehistoric burning had extensively modified the forests of the Northeast. Bromley (1935) also states that at contact, eastern forests had an open, park-like appearance, a point reiterated by Kr ech (1999). In spite of these observations, the environmental accomplishments of Native Americans were all but ignored, first by Europeans and later by American settlers. The above discussion points out the obvious need to account for the landscape management practices of native peoples, and this has become more commonplace in recent years (Bowden 1992; Cronon 1983; Dene van 1992; Krech 1999; Pyne 1982; Vale 2002; Wagner 2003). Even rain forests, wh ich typically stand as the symbol for pristine environments have been reevaluated to incorporate indige nous practices of land management (Cronon 1995:82; Heckenberger et al. 2003). However, these endeavors have tended to focus more on people who pract iced agriculture. H unter-gatherers, on the other hand, have received little attention and ar e generally believed to leave or have left a small overall footprint on the landscape. Th is is in spite of a growing body of evidence that they can in fact have a dramatic imp act on local environments. In Australia, for example, it has been proposed that people si gnificantly altered fire-r egimes to the point that they impacted forest cover at a con tinental scale at which it caused broad-scale climate change that drove some Pleisto cene mammals to extinction (Miller et al. 1999:208). This is an inte resting point considering that it happened during the Pleistocene and also that is was at the hands of Austra lian hunter-gatherers. A good North American example is Chapman et al. (1982) suggestion that human-induced fire influenced the ecological makeup of areas in the Little Tennessee River Valley. 48


With regard to Native Americans and ot her indigenous groups we must not only strip away preconceived notions regarding the environment a nd their relationships with it but this should also be taken this a step fu rther to incorporate hunte r-gatherers for further consideration. There is a growing body of evidence th at suggests people may significantly alter landscapes through the most unintentional of actions. The first step towards achieving this goal is to give greater consideration to the ways in which huntergatherers perceive and intera ct with the environment. As previously mentioned the environment should be thought of as more than just a backdrop against which human action takes place. It ought to be approached from the perspective that its composition is the byproduct of long-term human interaction. Hunter-Gatherers and the Domestication of the Landscape Conceptualizing hunter-gatherers requires that they not only be placed within a context directly related to their surrounding landscapes but that they may also impact them, either actively through out right intentional ma nipulation but also as the unintended consequences of their normal activities. This stands in contrast to the general perception of them as having only a mild impact on th e environment. Consideration of huntergatherer activities should extend beyond just finding food, their relationship with other people such as agriculturalists, or the percei ved impacts of other cultural spheres such as capitalism when developing models of land-us e related to them. However, it is often implicitly assumed that hunter-gatherers view the environment in a fashion similar to the scholars who study them. This creates the in accurate perception of a dichotomy between nature and culture, which is not entirely corre ct. It becomes an even more contentious point when we consider that n ature is a cultural construct and thus subject to criticism 49


when considering a humanized perspective. As opposed to a western dichotomy of nature and culture, hu nter-gatherers differ in their view s of the environment which is at odds with many of the land-use models devel oped to account for the materiality of the activities left behind by th eir prehistoric counterp arts (Head et al. 2002). Two outstanding issues relate to reevaluating the perception of hunter-gatherers. Rather than just asse ssing the behavior of hunter-gatherers based on activities related to acquiring food resources on a random encounter basis, as is proposed by many land-use models following Binford (1980), greater cons ideration should be given to ways in which they mold and shape the landscapes in which they live. For example, we must move beyond the idea of plant species dom estication as the only means enacting environmental change. It is also important to consider the domestication of the broader landscape as a means for hunter-gatherers to manipulate the distribut ion and diversity of plant and animal species in or der to enhance the predictabili ty of encountering them. At present, the evolutionary developmental dic hotomy as perceived by most anthropologists may be too abrupt. These rigi d categories have inadvertently resulted in a narrow view of hunter-gatherers and in many instances thei r role as landscape managers has been obscured and they are relegated to the role of endless wanderers always in search of food, seeking out their environmental niche, or ot herwise leading a hand to mouth existence. There is a need for more of an intermediate level where greater cons ideration is given to the possibility that people already had str ong ideas related to landscape manipulation prior to plant and animal domestication wh ich requires genetic ch ange brought about by people who were selecting for certain genetic ch aracteristics. This pre-genetic selection 50


"stage" is difficult to account for based on the present system which contrasts marginalized hunter-gatherers a nd incipient agricultural groups. As outlined above, there has also been much criticism related to the use of contemporary hunter-gatherers to develop mode ls of prehistoric la nd-use. This stems primarily from the acknowledgement that many of the contemporary groups studied, primarily in Africa, were not as isolated as once believed and that they had in fact had extensive interactions with other people who were not hunter-gatherers. Coupled with the fact that they often lived in marginalized settings made them inappropriate for the development of land-use models that potential ly operated at continental-level scales. Placing an increased emphasis on landscape alte ration not only gives greater agency to hunter-gatherers but it also opens more geographic regions for consideration by archaeologists. In this case, the interior Coastal Plain potentially may have been much more amenable to settlement with regard to food resources than has been previously considered. This is based not only on the pot ential of hunter-gathere rs to enact landscape change but it also takes the study of them to a level beyond just settlement or subsistence niches in which they are commonly associated (Head et al. 2002:175). It also stems from the possibility that their percep tions of the environment may be much different than has formerly been suggested. Giving greater attention to the ways in which hunter-gatherers perceive the landscapes around them not only highlights th eir management strategies but it also illustrates their similarities with some of the practices of agricultural groups. Head et al. (2002), for example point out that hunter-g atherers often display qualities usually reserved for agriculturalists such as be ing active and knowledgeab le land managers, 51


transforming environments in both deliberate an d unconscious ways at a variety of scales, and having intimate attachments to partic ular places (Head et al. 2002:174). Head et al. (2002) argue that a conseque nce of the acceptance of the similarities between hunter-gatherers and agriculturalists has caused a breakdown of the general dichotomy between hunter-gathere r and agricultural space. Fu rther, they go on to discuss how hunter-gatherers sub-divide the landscape in relation to their inte ractions with it. The authors base their arguments on long-term ethnographic encounters with Australian aborigines. They also point out the tende ncy of archaeologists to pigeon-hole huntergatherer plant use in subsistence or ecologica l terms since they are generally presumed to bound and enclose spaces with complex social rather than material means (Head et al. 2002:175). To counter this na rrow viewpoint they forward the terms country and garden as a way to which Australian hunter-gat herers observe their world. In their view, country is a place that gives and receives life, it is not just imagined or represented, but rather is lived in and lived with (Head et al. 2002:176). Garden, in contrast, is considered to represent more intensified modifications, boundedness or patchiness, at various locations in space (Head et al. 2002:176). Country and garden are further subdivided into two forms of domains wh ich the authors posit as both knowledge and practical consciousness that ar e social in their overall c onception (Chase 1989:46; Head et al. 2002:175). Domus corr esponds to hunter-gatherer phy sical domains that operate around a specific point in time and space whic h may be distinguished from domiculture which represents parcels of knowledge, stra tegies and actions, in cluding those relating to the management of plants and more specifically knowledge and activity bundles which relate temporally to a specifi c habitat (Head et al. 2002:176). 52


Yam digging was a primary activity of these people who relied on them not only as a source of food but also as a means for marking the landscape. The distribution of yams was based partially on geology as rocky soils were perceived as growing better quality yams but they were also anthropogenic because people actively sought out these particular soils and subsequently planted their yams patches in them (Head et al. 2002:181). In many instances patches were also given names. Material evidence for this activity occurs in not only the large piles of dirt and rocks left behind by digging but also by their patchy distribution. Patches are eas ily discernible by the large holes created by digging for yams. Yam removal not only includes digging but also moving large boulders and rocks by rolling them out of the way and by the piling up of smaller rocks which creates rings around individual holes (Head et al. 2002:180). Quarrying of toolquality stone was also an activity associated with yamming due to the large amounts of stone that was excavated while people were digging for yams. The authors noted encountering several areas of stone flaki ng and quarrying, with many large cores and flakes that were associated with yam patches (Head et al. 2002:181). Taken together, yam and quarry patches a nd their associated human activities also represent aspects of both gar den and country as discu ssed by Head et al (2002). They occur as discrete and di stinctive physical lo calities that are pa tchily distributed within a wider area (country) as a result of geology and they also correspond with nodes of intensive human labor (garden) that are gi ven names and further in corporated into the understanding of the broader landscape (Head et al. 2002:182). These are important points to consider when rethi nking hunter-gatherers and their interactions with nature. They have the ability to control the broa der distribution of food resources that are 53


important to them. In this case yamm ing was not only a heavily influenced anthropogenic activity but it also integrated with quarrying to acquire stone for the production of stone tools. Both of these point s will be returned to later in the broader discussion of connection of people and fire in the interior Coastal Plain of southeastern North America. In their discussion of hunter-gathere r pyrotechnic practices in the Pacific Northwest, Lewis and Ferguson (1999) present the concepts of fire yards and fire corridors. They are based largely on a cros s-cultural comparison of similar activities by other hunter-gatherer groups from around the world. Both yards and corridors are associated with fire-dependent ecosystems which are easily iden tified by their fire mosaic qualities (Lewis and Ferguson 1999:164). The authors also point out that these types of ecosystems are highly conducive to hunting and gathering because of the significant levels of ecological diversity they contain. Fire mosaics may be caused by both natural and human-induced fires, though th ey may be differentiated based on their size, frequency, and severity. Natural fires, for example, may be characterized by their larger size, low frequency, and higher intensit y because they are usually responsible for complete forest or stand replacement (Lew is and Ferguson 1999:165) Those fires set by people, particularly hunter-gatherers on the other hand, may be distinguished from natural fires since they are smaller, more frequent, and of a low level of intensity. Lower intensity fires stem from the increased fr equency which does not allow excess fuel to build up which would cause potentially c onflagratory burning conditions. These characteristics are a direct result of hunte r-gatherers altering prev ailing fire regimes by 54


burning during alternate seasons, with increased frequencies, and under safer, more managed conditions (Lewis and Ferguson 1999:165). Within these contexts fire yards an d fire corridors constitute areas of controlled burning by hunter-gat herers. Fire yards are openings or clearings within larger forested zones that are maintained by burning while fire corridors are linear areas along streams, sloughs, ridges, and trails that are similarly maintained through regular burning (Lewis and Ferguson 1999:167). Both fi re yards and corridors provide a greater abundance of plant and animal resources and a much greater level of hunting predictability. Maintaining open grasslands especially increases the abundance of animal populations (Lewis and Fergus on 1999:171). Similar to garden and country forwarded by Head et al. 2002, fire yards and corridors represent yet another means that huntergatherers subdivide and interact with their native environments. Based on her observations in California, Anderson (2006) suggests that Native Americans had been altering local environments for up to 12,000 years prior to the arrival of Europeans. By practicing a number of different management techniques, such as burning, coupled with carefully planned harvesti ng intervals, they were able to control the distribution and population density of a number of plant and animal species that were important to them (Anderson 2006:1). With regard to the discussion here, regular burning was successfully employed by California hunter-gatherers to improve habitat for game animals, eliminated troublesome brush, and reduced the potential for catastrophic fires. Similar to the earlier discussion re garding indigenous views of nature, Anderson (2006) notes that native California groups di d not distinguish betw een managed land and wild land. She goes on to discuss the fact th at the word wilderness is actually absent 55


from tribal vocabularies (Anderson 2006:3). In fact, the idea of w ilderness stems from a European psyche accustomed to a Neolithic way of life in which people distinguish between areas of human manufact ure and those deemed to be of natures responsibility. In many instances, as pointed out by Anders on (2006:3) contemporary indigenous people refer to areas of wilderness in a negative context because they have not been cared by people for a long time. Finally, Ingold (2000) discusses the si gnificance of personal experiences that hunter-gatherers have as they interact with the landscapes around them. Citing the earlier work of Turnbull (1965), he emphasizes the paternalistic relations hip that the Mbuti Pygmies of the Ituri Forest ha ve with the natural world around them (Ingold 2000:43). They refer to it as a mother or father because it provides them with food, warmth, shelter, clothing, and all the other basic necessities of life (Ingold 2000:43; Turnbull 1965:19). This parent/child relationship is important to consider because it not only reveals the intimacy which hunter-gatherers may express toward the environment but it is also reflects close relationship they have in understanding it. For example, Ingold (2000:47) points out that experience gain ed through human-organism-environment interactions, provides the raw material of sensation, that along with food that huntergatherers take home with them. He notes that similar to the way people become familiar with other people, they also beco me familiar with the environment by spending time with it. In this capacity, hunter-gatherers often consider time spent on forays in the forest to be useful, whethe r or not it returns anything to consume (Ingold 2000:47). Time spent in the forest fosters human-environment relations, people get to know it as they would a relative because of the time they spend with it. Following the previous 56


discussion regarding the nature-culture dichotomy, Ingold (2000) discusses the connection between the culture of hunter-gatherers and nature in which he concludes that the two are an integrated enti ty. This view is particul arly important when thinking about the level of knowledge that people ma y have with regard to their immediate environs. All combined, these examples of hunte r-gatherer-landscape re lations exemplify the need to examine the environment as mu ch more than just a backdrop that both dictates human behavior and also serves as a setting against which it plays out. The cases above illustrate that hunter-gatherers can have not only intensely intimate relations with the environments in which they live but can also have a powerful hand in shaping them and thus controlling the broadscale distribution of plant and animal resources. They are much more than the wandering nomads that they have been presented as for years in the archaeological literature. Following Head et al. (2006:176), if we are dismantling the binary divisions between humans and the environment we must take plant ecology and biogeography seriously as a potential arti fact (Head et al. 2002:176). Taking the landscape as a unit of analysis may allow for the familiarity that prehistoric huntergatherers had with the environment based on the material distri bution of artifacts stemming from their various activities in relation to forest cover. In this case, taking a closer look at the distributi on and dynamics of the interior Coastal Plain may reveal a significant anthropogenic component that has no t previously been c onsidered, especially in relation to anthropogenic fire. Aboriginal people are widely reported as describing their burning activities as cleaning up the c ountry but this same t ype of action has not been considered for the interior Coastal Plain in spite of the fact that it is often referred to 57


as a fire forest (Head 1996; Head et al 2002; Pyne 1982). In many areas research suggests that people likely made sweeping imp acts through the use of fire. It has been proposed that at contact, Eu ropeans found a North American continent whose ecosystems had long been influenced by the actions of people (Cronon 1983). That such a prolonged span of time transpired between the initial contact and later excursions by explorers and settlers led to an inaccurate perception of the pre-contact landscape. Denevan (1992) notes that many areas rebounded ecologically during the interim, which helped Europeans create the image of the land as sparsely populated and little impacted by Native Americans prior to their arrival. In actuality, it was very much a widowed land in the sense proposed by Jennings (1975). A similar situation occurred in Australia as aborigines were openly killed off or otherwis e forced off the land (Head et al. 2002:175). Anthropogenic impacts to North American landscapes likely had an extensively protracted history beyond just the park-like appearance as experienced by the initial European explorers (Anderson 2006). Just how far these land mana gement practices go back in time in the Southeast is not presen tly known. A goal of th is dissertation is to extend evidence for management back at least to the Late Archaic period. In this respect, it is a goal of this research to determine the extent to which Late Archaic hunter-gatherers affected the longleaf pine-wiregrass ecosystem through the use of fire. The Archaeology of Historical Ecology and Anthropogenic Fire There are a number of methods that may be utilized to identify and quantify the extent of anthropogenic fire. It is, howeve r, difficult to distinguish between human-set and natural, lightning-induced fires. This difficulty can be overcome using multiple lines of evidence. For example, fire histories of certain geographic localities have been 58


reconstructed using tree cores, analysis of palynological data, and fire scars (Wagner 2003:136). The analysis of polle n can provide a great deal of insight into past forest cover to reveal changes through time. Changes such as a broad-scale shift from fire vulnerable to fire tolerant species may be indicative of human influence. Pollen and charcoal samples taken from wetland or lake bo ttom sediment cores can also be useful in determining the size and genera l locality of fires in the past (Lamoreaux 1999; Seilestad 1995). Increased levels of charcoal fragment s are an indicator of anthropogenic fires (Clark 1988). A close inspection of charco al fragments may also reveal evidence pertaining to whether or not a pa rticular fire was small and localized or a larger regional one. This is based on the overall size of charco al particles and their relation to the size of the body of water from which the sediment core from which they were derived was taken. Smaller bodies of water capture larger, local-level particles while larger ones usually contain smaller fragments that may have been carried from fires that were much farther away (Clark 1988; Patterson and Backman 1988). Intervals between fires may be determined from the examination of fire scars from tree cores taken from old trees (Huffman 2006). The biogeographic distribution of certain plant species may also be indicative of a fire-maintained anthropogenic landscape. Th is is especially noteworthy when thinking about various ecological aspects of the longl eaf pine-wiregrass ecosystem which will be discussed more in-depth in the next chapter. Looking at its overal l distribution and what it requires for continued maintenance and longevity also reveal insights into its long-term interaction with humans. In pa rticular, it requires regular low-intensity fires to ensure its overall health and well-being (Christensen 1987). The anthropogenic contribution to 59


perpetuating longleaf pine forests in the inte rior Coastal Plain has long been a point of contentious debate. Quantifying the potential ex tent of this practice will ultimately need to be pursued using the methods discussed a bove. For this dissertation however, the material remains of various activity areas, wh ich will be further outlined in Chapter 4, will be compared to potential for fire and sustained habitation. As discussed above, hunter-gatherers often play a significant role in shaping landscapes by influencing the distribution of economically usef ul plant and animal species. This should be reflected in the overall distribution of particular site types in the project area. In this case, it is likely that Late Archaic hunter-gatherers were usi ng fire to maintain an open-mosaic forest cover of longleaf pine which would also have been interspersed w ith upland ponds. This mosaic quality would have created a series of ecotones or edge habi tats. This effect would further contribute to in creased levels of biodiversity. People intentionally set fires for different reasons and i ndigenous burning practices are well represented in the ecological and archaeological literature. While some may have started from meager beginnings such as driving game, they may have potentially taken on a life of their own as their long-term consequences became r ealized. This in turn may have further influenced the positioning of habitable pla ces within the broader social landscape. Examples of hunter-gatherer induced fire s are common in the archaeological and ethnographic literature. They stem from work in a number of different and unique geographic locations such as North America, Europe, South America, Australia, and Africa (Kalisz et al. 1986; Lewis 19 77, 1980, 1982; Meggers 1966; Patterson and Sassaman 1989; Pyne 1982; Stewart et al. 2002 ; Yellen 1977). In general, fire is a powerful tool in the hands of hunter-gat herers and it has been wielded in both 60


contemporary and prehistoric settings to exact the landscape-level manipulation of economically important plant and animal species. Fire was used in various habitats for a number of reasons such as to control the overall distribution, dive rsity, and abundance of plant and animal resources as well as create and maintain a series of micro, or edge habitats that were useful for hunting and gathering (Krech 1999; Lewis 1982:45-51). Deer in particular, thrive in managed hab itats because of improved browse. This has been illustrated in both Eur opean and North American contexts (Klein 1970; Lay 1956; Stellars 1976). Based on findings in Europe, Sassaman ( 1988) has suggested that Mesolithic hunter-gatherers aggressively altered upland ar eas in north-central Eu rope with fire to create better growing conditions for hazel ( Corylus spp.) and their nut yields, which were highly desirable for humans. Regular controll ed burning in the uplands is believed to have created amenable growing conditions in which hazel trees thrived. Sassaman (1988:4) notes that Mesolithic hunter-gatherers used fire to carve out and maintain patches in mature forest. While Sassamans findings are based on work in the temperate forests of north-central Europe they are noneth eless useful here as they illustrate how hunter-gatherers can impact local environmen ts through the implemen tation of fire. The use of fire in this Europe an context is very similar to Lewis and Fergusons (1999) discussion of hunter-gatherer fire yards in the Pacific Northwest. Carved out patches in otherwise hardwood dominated forests created higher levels of biodiversity by fashioning a series of mosaic quality grassy patc hes within the prevailing forest cover. This is also similar to observations fro m northeastern North America. Charcoal records from sediment cores taken from bot h inland and coastal locations in southern 61


New England suggest that indigenous burning was customary in the region during times of high population density, and possibly bega n as early as the Late Archaic period (Patterson and Sassaman 1988). From pond sediment cores taken at Cliff Palace Pond in the mountains of Kentucky, Delcourt and Delcourt (1998) argue that Late Ar chaic hunter-gatherers relied on fire to clear forest openings on upper slope s and ridge tops to promote the growth of early successional flora. Palynological eviden ce from the area suggests that local huntergatherer groups were encouragi ng the growth of plants that are generally associated with the Eastern Agricultural Complex (Wagner 2003:138). Lewis (1982:50) notes that hunter-gatherers living in nor thern Alberta burned the forests for a number of reasons that included reducing volatile summer wild fire conditions and improving browse for large game animals such as moose that would move into recently burned areas to consume freshly sprouting green shoots. Berry producing plants were also stimulated by fire and it al so worked to thwart the raging, bloodthirsty populations of black flies and mo squitoes that could easily ma ke life miserable for people living nearby. Fires were also set along stream channels to create grassy borders between the waters edge and the surrounding forest. Th is enhanced streamside habitat attracted rabbits, moles, and other small mammals which in turn appealing to larger predators like weasels, coyotes, wolves, and foxes (Lewis 1982:50). The shorelin e edges of lakes and wetlands were also fired to improve habitat for migratory waterfowl. Perhaps no continent is more associated with fire than Australia (Lewis 1982; Pyne 1997). Examples of hunter-gatherer use of fire to manipulate and control Fire Stick have been illustrated across the continent. Known as fire stick farming, local hunter62


gatherers made good use of fire in virtually every conceivable environment. They successfully shifted the fire-regime at virtually a continental scale fr om one of infrequent conflagratory fires to one based on much smalle r frequent fires. This dramatic shift was possible due to native familiarity with local burn cycles. By changing the seasonality and frequency of fires Australian aborigines were able to successfully a lter the distribution of plant and animal species that were important to them (Lewis 1982:63). Another good example of burning by hunte r-gatherers stems from Yellens (1977) observations among the !Kung San of the Kalahari He notes that the !Kung set fires in late summer to encourage the growth of new grass (Yellen 1977:34). Yellen also points out that if misused burning can easily lead to a detrimental outcome given the areas susceptibility to drought. Of course, Yellens perception could easily stem from his own lack of understanding of the utility of burning within the broader ecological landscape in which the Kung live. For example, he draws his conclusion regarding the proposed sensitivity of the vegetation but goes on to comme nt that data related to annual rainfall as well as climatic succession remains an intriguing and largely unanswered one (Yellen 1977:34). Much of Yellens conclusion stems fr om the context of the time in which he was writing. He makes note of the grassy ar eas that expand outward from the Aha Hills at the expense of larger trees and shrubs but fails to make a connection between that expansion and indigenous burning practices (Yellen 1977:34). Perhaps if hed spent more time in the area he might have made a connection between the two. Of course, focusing again on the context of the time, Ye llen was observing the foraging practices of the Kung to make better informed models of hunter-gatherer land use. As was discussed 63


earlier, incorporating their ability to inflic t landscape-level vegetation change was not a top priority of researchers. Given the preceding overview of hunter-gathe rer practices of broadcast burning, it is plain to see that fire is an important tool that can be wi elded by people who are otherwise not generally considered to have a significant impact on the regions in which they live. This is certainly the case with regard to the interior Coastal Plain of southeastern North America. This line of thinking, combined with the ecological dynamics of the longleaf pine-wiregrass ecosy stem which will be discussed in detail in the next chapter, creates a picture of life in the region that has not been previously considered. In all lik elihood, Late Archaic hunter-gatherers in the region used fire in much the same way that it was used in othe r parts of the world. Open grasslands are highly productive areas and keeping them open would have been in th e best interest of the people who lived close to a nd interacted with them. Likewise, the many wetlands and stream banks conducive to habitation would have been enhanced through controlled use of fire. 64


CHAPTER 3 THE CHICKASAWHATCHEE ARCHAEOLOGICAL SURVEY: ENVIRONMENTAL SETTING, PREVIOUS ARCHAEOLOGICA L WORK, FIELD METHODS, AND SITES Data used to address the research proble ms presented in this dissertation were recovered during the Chickasawhatchee Archaeol ogical Survey in southwest Georgia. In this chapter, I present the details of the surv ey project, such as the spatial parameters and orientation of the project area, relevant environmental variab les, the findings of previous archaeological research in the region, and me thods used to discover and record sites in the project area. Survey work commenced during the summer field season of 2003 and continued through the 2004, 2005, and 2006 seasons. The early portion of the survey concentrated on site discovery in the Chickasawhatchee Cr eek watershed and the areas surrounding the Chickasawhatchee Swamp (Chamblee 2006), wh ile later efforts were focused on the Ichuaway-Nochaway Creek drainage. The project area is unique becau se of its size and location deep within the Gulf Coastal Plai n. It is comprised of several individual disconnected survey areas that encompa ss a diverse range of topographical and ecological features. The entire project area lies within the Doughert y Plain (Figure 3-1) which is a sub-province of the interior Coastal Plain. The Dougherty Plain is characterized as a broad, relatively flat physiographic district with a level to ge ntly rolling topography and a northwest heading geophysical orientation (Murray 1961). Topographic relief in the Dougherty Plain is greatest near the Fall Line Hills, which stands as its northern boundary, at roughly 250 feet above sea level and 77 feet above sea level in its southern expanses in Seminole County. It terminates 65


Figure 3-1. Location of Survey Tr acts within the Dougherty Plain. 66


where the Fall Line Hills and the Tifton Up land meet. Karst topography, formed on the underlying Ocala and Suwanee Limestones, under lies the plain which has contributed to the formation of numerous sinkholes in the district. Gum ponds and Cypress swamps are common hydrologic features of this landscape. Cypress swamps are easily recognized by their rounded outline with the tallest trees occurring in the center and sma ller tress moving out toward the edges. Gum Ponds are marked by the lack of cypress tr ees which are supplanted by swamp black gum and other lowland tree species (Wharton 1978:79). Both types of ponds are categorized as depressed surface hydric features, charac terized by their shallow depths that may or may not contain water, a result of overall prec ipitation and/or their de gree of closeness to the underlying limestone bedrock (Wharton 1978) Periodic natural fires that occur during periods of relative dryness and low-water levels serve to regenerate ponds and increase their biotic potential (Wharton 1978:74) The approximate origins of these small bodies of water are not entirely known though some have been attributed to sea-level rise during the Pleistocene and subs equent higher water levels in the Coastal Plain by the mid-Holocene (Gaiser et al. 2001; Hoyt and Hails 1967; Murray 1961; Wharton 1978:81). Cypress swamps and Gum ponds are home to a diverse array of plant and animal species. Numerous insects, amphibian s, reptiles, birds, and mammals depend on the ponds as a source of water and breeding ar eas. The ecotonal zone or edge habitat surrounding the ponds also attracts a high percenta ge of wild life and is an integral hub for the surrounding pine flat lands of the Co astal Plain (Wharton 1978:82). Based on the underlying limestone bedrock, which is part of the greater Floridian Aquifer, many of the ponds are attributed to sinkhole activity. Others have been compared, at least 67


ecologically, to Carolina Bays which have received a great deal of attention from archaeologists working in South Carolina (Brooks et al. 1996; Brooks et al. 2001; Wharton 1978:80). Late Archaic si tes associated with similar karstic features have also been studied in Florida (Bullen et al. 1974; Bullen and Dolan 1959; Neill 1964; Wharton et al. 1982). Shifts toward increased use of wetland resources during the Late Archaic period are well documented across the eastern Woodlands (Brooks et. al 1996; Custer and Bachman 1986; Elliott and Sassaman 1995; Gardner 1978; Milanich 1994; Price and Brown 1985). Wetland environments offer an a rray of plant and animal resources that were beneficial to local hunter-gatherers, and they harbor, under certain conditions enormous productivity, at last seasonally. Wetland ecosystems like the upland ponds discussed above occur across the expanse of interior Coastal Plain. Although they may not compare with the wetlands of bottomland sy stem in terms of extent and productivity, upland ponds and bays indeed attracted human settlement throughout the Holocene, and, at times, appears to have sustained settle ment of prolonged and/or repeated duration. The economic and ecological significance of upland wetlands in the project area is contrasted with the now-de funct conceptual bias of arch aeologists that the interior Coastal Plain was relatively unpr oductive and thus in capable to sustaining significant human presence (Larson 1980; Fish and Fish 1976). Even though recent archaeological endeavors have countered previous claims (Brooks and Canouts 1984; Brooks et al. 1996), parts of the region are of ten referred to as the pine barrens, which downplays, if not denigrates, its actual ecol ogical richness. Few archaeolo gical interpretations of the region get beyond dealing with what most perceive of as a homogenous sea of pine trees. 68


Most often overlooked are the wetland system s discussed above. Just giving greater consideration to upland ponds does much to a dvance a picture of heterogeneity in the interior Coastal Plain. The seemingly e ndless vegetative monotony which so often negatively characterizes the region primarily stems from unsustainable land management practices in the mid to late 19th and early 20th centuries as settlers and industrialists began pushing deeper into the interior Coastal Plai n as the country expanded westward from the initial coastal settlements. Successive wa ves of people and indus try concluded in the outright destruction of the longleaf pine-wireg rass ecosystem that once blanketed a large portion of the region. This was followed by its replacement with nonlocal, fastergrowing pine species that could be more fr equently harvested for timber. Traveling through the region today it is ha rd to see beyond the sea of pi ne trees that are actually continuous rows of slash and loblolly pine s grown on industrial-style plantations. The trees are planted in near pe rfect rows with little if any underbrush which gives the landscape an overly sterile appearance. In contrast to modern pine plantatio ns, the historic l ongleaf pine-wiregrass ecosystem of the project area embodied much greater levels of eco logical diversity. Longleaf pine grows in a number of diverse environments based on elevation and soil moisture, and it is thus subdivided by ecol ogists along vegetational lines. Interspersed with the upland ponds and wetlands, the longleaf-wiregrass communities of interriverine landforms represented some of the most diverse biomes of the Coastal Plain in precolonial times. The perpetuation of divers ity in longleaf pine communities, of course, hinges on a high-frequency fire rotation. 69


That the project area was forested in l ongleaf pine is readily accepted based on historical records. Unfortunately, the demise of this once grea t forest system in the early late 19th and early 20th centuries left an obvious void in the natural landscape, and it created a formidable obstacle in fully unders tanding the ecosystem at a landscape-level scale and how people may have interact ed with it in the distant past. Previous Archaeological Research Related to the Late Archaic Many of the earlier archaeological findings in southwest Georgia were associated with the construction of large hydroelectric f acilities, river channel improvement projects, or otherwise focused directly on the Chatta hoochee and Flint River flood plains or their associated tributaries (Boyd 1958; Bulle n 1950, 1958; Caldwell 1978; Fish and Fish 1976: Fish and Mitchell 1976; McCluskey 1976). Ot her projects have focused on specific sites considered to be culturally signi ficant (Fairbanks 1940; Pluckhahn 2003; Sears 1956; Steinen and Crawford 1990). More re cent projects, like this one, have concentrated on examining private collections and/or surveying a nd excavating in upland localities (Elliott 2004; Elliott and Dean 2006; Waggoner 2003). Research designs associated with the earlier re servoir projects were not conducive to finding Late Archaic sites in the region wh ich were usually small in size and composed primarily of debitage and stone tools. They instead tended to concentrate on larger later period sites with ceramics a nd or mounds (Fish and Fish 1976:3). For example Kelly (1950) located and excavated only a single Archaic site during the entire survey coinciding with the construction of the Jim Wo odruff dam at the confluence of the Flint and Chattahoochee rivers. The site was disc overed at Lane Springs. A torrential storm that dropped fifteen inches of rain over a tw enty-four hour period resu lted in the wash out 70


of a sand-ridge on Spring Creek (Kelly 1950:29) The sand-ridge was in close proximity to a runoff channel of Lane Springs and the churning current deposited hundreds of worked flints in its former position (Kelly 1950:29). A. R Kelly was notified of the situation as he and his University of Georgi a field school crew were working just down stream at the site of the Woodruff dam. Ke lly (1950) noted that th e flints had extended down into the ridge to a depth of roughly five feet. He categorized the tools as handchoppers or fist axes and likened them to tools of the European Neolithic. Kelly also described a large assortment of oval flint blad es, a number of grindi ng stones and a single mortar that had been collected by local school children, all of which he attributed to prepottery or Archaic activit y (Kelly 1950:29). Partially as a consequence of insufficient survey strategies implemented by earlier projects, southwest Georgia, as a region, has been cited as having a relatively low number of Late Archaic sites (Ande rson 1996:170). It may also potentially result from an unrecognized divergent patte rn of land-use in the interior Coastal Plain in contrast to other areas in the Southeast that have received greater atten tion from archaeologists. For example, major Late Archaic shell midden sites have been excavated in the Savannah River Valley that reflect a heavy reliance on riverine food resources such as shellfish (Claflin 1931; Sassaman et al. 2006). In contrast, no she ll middens with Late Archaic features or components have been reported in either the lower Flint or Chattahoochee River valleys, though Kelly did excavate a la rge shell midden at White Springs which is close to the point where Spring Creek enters the Flint River. He described it as a prehistoric clambake or oval depression dug into the terra ce with a sizable portion of Weeden Island pottery (Kelly 1950:30). 71


Aside from Kellys (1950) work at Lane Springs, only a few Late Archaic sites were studied during the early reservoir proj ects in the Flint and Chattahoochee River valleys. An important discovery was Ripley Bullens (1958) excavat ion of J-5, a deeply buried multi-component site located along the banks of the Chattahoochee River. The site was located on the second natural levee of the Chattahoochee River, just to the north of the confluence of the Flint and Chattahoochee rivers. All vegetation in the vicinity of the site had been cleared since it would be inundated within the reservoir. Clearing initially revealed the largest Fort Walton (M ississippi) period village along the west side of the river (Bullen 1958:333). Artifacts at the site were recovered near the ground surface so the archaeological investigations were initially focused on the Fort Walton component but the emphasis shifted when rese rvoir engineers worki ng at the site dug a ditch from the levee to the Chattahoochee whic h cut a deep profile across the J-5 site. Large stemmed hafted bifaces were recovered from back dirt removed from the ditch. Bullen profiled the ditch and identified fourteen separate archaeological zones. Zone 9 is most relevant to the discussion here becau se Bullen (1958) placed it chronologically in the Late Orange period based on the presence of Orange Plain pottery. Seven soapstone vessel sherds were also recorded in the z one (Bullen 1958:337). Bullen (1958) noted that the cultural remains recovered from zone 9 resembled those of people living on a river levee. Mussel shells were common throughout the zone. A wide variety of stone tools were also recovered from this zone. They included medium and large hafted bifaces, soapstone vessel fragments, and Orange Plai n fiber-tempered pottery (Bullen 1958:338). Bullen did not attempt to assign type names to any of the hafted bifaces he recovered. Based on the photographs in his report, however most may be categorized as Savannah 72


River and Abbey hafted bifaces which are two of the most common types of Late Archaic hafted bifaces in the Coastal Plain. Several reworked hafted end scrapers were also excavated. Basal morphology in many instan ces can be attributed to either Savannah River or Elora hafted bifaces prior to be ing broken and reworked into hafted end scrapers. Botanical and faunal remains reported from the site include nuts, deer, turtle, opossum, lynx, muskrat, beavers, sna ils, and mussels (Bullen 1958:339). In addition to the J-5 site, Bullen al so recorded and excavated J-18 which provided valuable information regarding the La te Archaic in the interior Coastal Plain. The J-18 or Tan Vat site was situated on a low rise on the eastern edge of Tan Vat pond which Bullen interpreted as a relic channel of the Chattahoochee River (Bullen 1958:323). Bullen tested the site by excavating a trench in four-inch levels that was 10 feet wide and 50 feet long (Bullen 1958:323). Like that at the J-5 site, the Late Archaic component at J-18 was represented by Orange Plain fiber-tempered pottery and soapstone vessel sherds, though both were recovered in low numbers (Bullen 1958:324). A variety of stone tools were also repor ted at J-18. They were described as scrapers, drills, and utilized flakes. Hafted bifaces and other stone tools were also reported. Again, Bullen did not attempt to place them into typol ogical categories but based on the photographs from his report, however, a majority of them can be categorized as Savannah River and Elora hafted bifaces (Bullen 1958:357). A la rge pitted sandstone slab along with 26 broken quartz cobbles, one chert and four quartz hammer stones were also recovered (Bullen 1958:224). The first broad-scale archaeological survey s away from the major rivers in the region were implemented by St einen (1976), Fish and Fish (1976), and Fish and Mitchell 73


(1976). The three projects vari ed in scope but each contri buted valuable information regarding details of Archaic land-use in the Dougherty Plain. St einens (1976) project was a county-wide reconnaissance survey of Early County, Georgia which was primarily geared toward identifying and recording form ative period sites in the vicinity of the Kolomoki site, which has been interpre ted as a major Middle Woodland ceremonial center in the region. Even though a central goal of Steinens survey was to recover data associated with the middle Woodland, he also recorded a number of wh at he referred to as preceramic or Archaic sites. He made two observations concerning Late Archaic land use in the area, the firs t being that almost any, high, dry area between creeks will have some evidence of an Archaic occupatio n (Steinen 1976:71), and second, he noted that Late Archaic hunter-gatherers had adap ted their land use to open areas within the Dougherty Plain (Steinen 1976:71). Combining data on bot h Archaic and formative period land-use, Steinen (1976: 73) developed a land-use model that emphasized the broad-scale use of the uplands during the Archaic followed by a shift in the direction of alluvial areas during the formative period which he attributed to an increased reliance on horticulture. Fish and Fishs (1976) work in the Dry Creek watershed and Fish and Mitchells (1976) study in the Big Slough watershed repres ent the earliest large systematic surveys in southwest Georgia. Dry Creek is a trib utary of the Flint River. Archaeological surveys of both watersheds were precipitated by proposed channel improvement projects. The Dry Creek project area covered portions of Seminole, Miller, and Early counties, which are southwest of the Chickasawhatchee Archaeological Survey. The full extent of the watershed lies entirely w ithin the Dougherty Plain and is thus important to the 74


Chickasawhatchee Survey project area given its close proximity. Fish and Fish (1976:2) noted the occurrence of upland ponds with cypress and tupelo gum in the upper reaches of the watershed. They also reported that Late Archaic sites were commonly associated with these upland wetlands. Fish and Fishs (1976) survey was significant for the time due to its focus on interriverine areas which had previously received little attention from archaeologists. Pursuing this strategy, they identified 119 si tes, of which 71 could not be organized chronologically because of a lack of diagnostic artifacts. Of the remaining sites, thirty-three produced Archaic-style ha fted bifaces while the final twelve were ceramic age based on the recovery of ceramics or coeval projectile points (Fish and Fish 1976:9). A second important development within th e Dry Creek project was the authors advancement of site types based on the recovery of temporally sensitive hafted bifaces, chipped stone tools, and debita ge. This achievement was es pecially relevant in that virtually no attempts had been previously made to decipher Late Archaic land-use in Dougherty Plain portion of the in terior Coastal Plain. Debitage was divided into flakes of bifacial retouch, flakes of normal percussion, and formless debris (Fish and Fish 1976:7). Tool categories consisted of chopper, plane, thin biface, thick biface, projectile point, drill, and graver. A simple index of diversity was calculated to quant ify divergent artifact densities between sites (Fish and Fish 1976:7). This strategy was presumed to shed light on the differing activities carried out at each site. For example, a high level of diversity would represent a greater numbe r of activities whole a lowe r level would indicate fewer activities had been carried out at a particular locality (F ish and Fish 1976:9). Based on this technique, the authors forwarded th ree primary site types for the Dry Creek 75


watershed. Class I represented the material remains of special activity areas, Class II sites were characterized as short-term camp sites, while Class III sites were held to be permanent to semi-permanent habitation sites or base camps. All combined, the three site types were believed to correspond to a patter n of seasonal resource procurement in which resources were attained and processed at the same locations (F ish and Fish 1976:13) Similar to the Dry Creek survey, Fish and Mitchells examination of the Big Slough watershed also generated a considerab le amount of new data related to Late Archaic land-use. The Big Slough is a tributary of the Flint River and runs through Grady and Mitchell counties, which are south of th e Chickasawhatchee Archaeological Survey. Fish and Mitchell divided the Big Slough project area along the Dougherty Plain and Tifton uplands which is another major physiogr aphic province of the interior Coastal Plain. The Tifton uplands were directly to the east of the Big Slough project area and had a reconstructed vegetation cove r of pine interspersed with in a wiregrass savannah (Fish and Fish 1976:21). An upland solution escarpment marks the boundary of the Tifton Upland in the Big Slough watershed. The bounda ry is easily distinguished from both the Dougherty Plain and the Tifton Up land due to its greater topogr aphic relief, which is also bisected by streams, and presence of less num erous but deeper solution sinks compared to the Dougherty Plain (Fish and Mitchell 1976:3). The findings in the Big Slough watershed provided valuable insight into prehistoric life in the inte rior Coastal Plain. Fish and Mitchell (1976) not only demonstrated an intensive use of areas known for being forested in pine trees but also that sites tended to commonly be located in upland areas. Prior to the survey, Late Archaic upland use in the area was rela tively unknown, though Fish and Mitchell 76


(1976:11) discovered that the Big Slough actually experienced its most intensive use during the Late Archaic. They recorded 89 sites all of which dated to the Late Archaic based on the presence of 132 temporally se nsitive hafted bifaces (Fish and Mitchell 1976:11). The most commonly encountered diag nostic hafted bifaces were the Savannah River, Clay, and Wade types, all of whic h suggest a time range from 5000 to 4000 BP (Fish and Mitchell 1976:14). Th ey accounted for 89 of the 132 hafted bifaces recovered during the survey (Fish and Mitchell 1976:14). Flakes and debitage were categorized using the organizational strategy from the Dry Creek Survey. The same tool types were also used from the Dry Creek Survey with th e added criteria that artifacts had to display signs of intentional retouch, pecking, or, grinding to be ca tegorized as tools (Fish and Mitchell 1976:13). ). Site types included loci of simple specialized activities, temporary campsites or short-term base camps, and long-term base camps (Fish and Mitchell 1976:17). Specialized activity s ites had no formal tools or only a single tool, while temporary campsites and short-term base camps had evidence for a broader range of activities such as hafted bifaces, choppers, thick bifaces, and a vari ety of scraper types (Fish and Mitchell 1976:19). As implied by the name, long-term base camps would have a broader range of tool types to infer an even greater number of activities. Unlike the Dry Creek Survey, Fish and Mitchell (1976) not onl y developed site types but also presented de tails pertaining to their occurrence across the broader landscape. As a starting point, Fish and Mi tchell (1976:21) argued that the high number of recovered hafted bifaces suggested that Late Archaic activit ies in the Big Slough watershed were primarily focused on hunti ng activities (Fish and Mitchell 1976:21). With regard to the other physiography in the pr oject area, a greater frequency of 77


specialized activity sites were recorded in the upland areas of the solution escarpment was higher than in any of the other phys iographic regions cove red in the Big Slough Survey (Fish and Mitchell 1976:21). Fewer s ites of all types were encountered in the Dougherty Plain proper but a number of special ized activity sites were located along the boundary between the Dougherty Plain and the Solution Escarpment (Fish and Mitchell 1976:21). The zone between the Dougherty Plain and the Solution Escarpment also had the highest number (n = 10) of short-term camps than any of the other physiographic provinces in the project area. The closed ca nopy of the pine fore st and its suppression of food resources were held up by the investigators as the primary reason behind the lower incidence of sites in the Dougherty Pl ain. Fish and Mitche ll (1976:21) also noted the fewest number of specialized activity si tes per square kilome ter occurred along the perimeter of the Big Slough. All factors co mbined, including the disparity between the numbers of specialized activity sites, shortterm and temporary base camps, and longterm camps, led the authors to suggest that La te Archaic sites recorded in the vicinity of the Big Slough were part of a broader but unknown territorial pattern (Fish and Mitchell 1976:22). This review of findings from previous archaeological research in the vicinity of the Chickasawhatchee project area provides a starting point by to examine the areas culture history and also evalua te patterns of local Late Ar chaic land-use. Artifacts related to the Late Archaic culture history are re presented in southwest Georgia by Savannah River, Clay, Wade, Abbey, and Elora hafted bifaces. All combined, these temporally sensitive diagnostic artifacts represent a time range of 4150 to 3000 BP (Whatley 2002). While somewhat brief, this span of culture history also serves as a comparative body for 78


the hafted bifaces recovered during the Chicka sawhatchee Archaeological Survey. It also exemplifies the dominance of Late Archaic o ccupation in the region. It also highlights land use in upland areas far removed from riverine settings that have not been given full consideration in previous studies. Much of th e work associated with the earlier reservoir projects was not conducive to locating Late Archaic sites and, given this limitation, archaeologists had only a rudime ntary conceptual framework to adequately place sites within a broader context. Fish and Mitchells (1976) attempt to place the Late Archaic sites recorded in the Big Slough survey into a broader, yet unide ntified territorial round was a fortuitous approach to dealing with Archaic Period sites. Fish and Mitchells (1976) observed pattern of upland use is particul arly relevant to the discussion here, especially considering that the predominant site type recorded was related to specialized activities. That specialized activity sites were routinely re corded in the uplands supports the pattern of these areas being used for hunting activities. Evidence illustrating the importance of upland ponds in Late Archaic land use is also presented. While the focus of Steinens (1975) survey was primarily concerned w ith locating Middle Woodland sites, his discovery of Archaic sites in areas of highe r elevation between creek s further confirms a pattern of upland utilization dur ing the Late Archaic in the interior Coastal Plain. His acknowledgement that Late Archaic hunter-gathe rers had adapted their land use to open areas within the Dougherty Plain will also be relevant to the discus sion of longleaf pinewiregrass fire management. All previous archaeological work in the vicinity of the Chickasawhatchee Archaeological survey combines to present a picture of wide-ranging Late Archaic land79


use. The deeply buried Archaic sites reported in the riverine floodplai ns of the Flint and Chattahoochee rivers (Bullen 1958; Kelly 1950), reflect the patterns expected for this time period as espoused by river-centric set tlement models (Stoltman 1974). Floodplains are typically assumed to be resources ri ch and actively sought by hunter-gatherers seeking food. Many Late Archaic land-use mode ls are based on data from sites outside the interior Coastal Plain. The pattern of upland activity as presented by Fish and Fish (1976), Fish and Mitchell (1976) and Steinen (1975) runs count er to models that tend to emphasize floodplains, though it may reflect a different pattern of land use than has been previously considered, especia lly considering the importance of fire to prehistoric forest cover in the interior Coastal Plain. Othe r than areas in direct proximity to the Chickasawhatchee Swamp, no riverine flood plain areas were explored in the Chickasawhatchee Archaeological Survey. In fact, most of the survey focused on upland locales. Methods associated with research in these areas are the subject of the next section. Field Methods Field methods consisted of opportunistic a nd systematic survey. Their application varied by individual survey area based on vege tation cover and surface visibility. Some areas had exposed ground surfaces because they were plowed for agriculture while others were primarily forested in secondary growth longleaf pine for cons ervation or loblolly and or slash pine that were destined for the pulp mill. Plowed areas with prepared crop rows also made systematic survey feasible by providing regularly occurring rows that were walked at evenly spaced 10 meter intervals. Opportunistic methods of survey were 80


implemented in forested areas where dirt ro ads or randomly available areas with surface visibility were examined for sites. A Garmin GPXV hand-held GPS unit was used for all aspects of the survey work. In addition to taking standard coordinate points, the GPS also had a track function that took points on regular intervals to record di stance and space covered. For the project, tracks were taken and kept as records of areas surveyed. Tracks were downloaded at the end of each working day and imported into a GIS program to create a map of areas covered. This method was very useful in both types of survey because it provided a visual record of where surveyors had been ove r the course of each day. Sites were also recorded using the GPS. Individual points we re taken at smaller sites while larger sites were outlined using the track function. Dirt roads were sometimes examined in forested areas that otherwise lacked adequate surf ace visibility. Following the conclusion of fieldwork, all lithic artifacts were washed, analyzed, and categorized based on the criteria outlined in the next chapter. Individual surv ey areas are discussed in the subsections that follow. Individual Survey Areas The separate survey areas within the Chickasawhatchee Archaeological Survey are unique topographically, in thei r general elevation, and in their relati onship to standing water. Each survey area is decribed sepa rately in the following subsections. Final conclusions regarding land use in the project area will be divided into uplands and lowlands based on individual survey areas. 81


Chickasawhatchee WMA The Chickasawhatchee Wildlife Management area (Figure 3-2) is located in Calhoun, Dougherty and Baker counties in the central portion of the D ougherty Plain. It covers 19,700 acres and encompasses a large portion of the Chickasawhatchee swamp. Elevation in this survey area ranges from 76.2 me ters above sea level at its highest points and 45.72 meters at its lowest. Figure 3-2. Chickasawhatchee Survey Area. 82


Vegetation cover consists of bottomland ha rdwoods and pine uplands. Because of a lack of large areas of open ground, survey of the Chickasawhatchee WMA consisted of walking dirt roads and inspecting food plots for hunting. While not the most optimal conditions, both the roads and f ood plots crossed a variety of landform types at multiple elevations thus providing a worthy representative sample. A landform within the Chickasawhatchee WMA known as the Mountain was surveyed by a field crew from West Geor gia College in the late 1970s when it was discovered that the site was be ing looted. The landform had b een recently clear cut so a series of dog leash surveys were implemented across it. The results of the survey were never reported and the artifacts have since been curated at the Antonio J. Waring, Jr. Laboratory of Archaeology at the State Un iversity of West Georgia in Carrolton, Georgia. I borrowed the artifacts to analyze and incorporate the results into the larger database of the Chickasawhatchee Archaeological Survey. The Jones Center at Ichuaway Ichuaway is a privately owned 29,000 acre outdoor laboratory dedicated to the ecological study of the longleaf pi ne forest (Figure 3-3). At present, much of the forest cover consists of grassland ha bitats, with the remainder cons isting of agricultural fields, wetlands, and riparian hardwood hammocks. Overall elevation on the property ranges from highs of 64.31 m to lows of 26.52 m. Agricultural fields comprise only a small portion of Ichuaway. Additionally, survey of the property took place in the summer when planted crops were in the fields, wh ich made for poor surface visibility and only limited access. The lack of plowed arable land at Ichuaway was offset by roughly 50 miles of sandy maintenance roads. Ichuawa ys maintenance roads are divided into 83


primary, secondary, and tertiary roads. As implied by the names, primary roads receive much more vehicle traffic than secondary a nd tertiary roads and require more routine maintenance. All of the roads on Ichuaway are located on the na tural ground surface which is generally composed of sandy soils. No gravel or riprap is laid down to create a road bed. There is limited erosion in hillier areas but for the most part sites can be easily discovered and delineated by recording artifacts on the road surface or in thinly vegetated areas along side it. Secondary and tertiary roads were especi ally useful in this regard because they are only cut with a bush hog pe riodically to keep them from becoming overgrown. Like the Chickasawhatchee WMA, Ichuaways roads cross over both upland and lowland areas of the propert y and thus provide a good repres entative sample off all of the landforms in the survey area. In addi tion to the artifacts rec overed during the field survey of Ichuaway, a private collection bel onging to Roger Birkhead was analyzed and incorporated into the project database. Bir khead was an intern at the Jones Center for five years in the late 1990s. He collected artifacts on the property during his time there. Fortunately, he had the wherewithal to thoroughly document the locations where he recovered the artifacts. He donated a sizable portion of his collection, consisting mostly of pottery and debitage, to the Jones Center upon his departure to attend graduate school at Auburn University. He took all the diagnos tic hafted bifaces, the remainder of his collection, from the property with him when he moved to Alabama. I visited Roger during the spring of 2007 to document his colle ction. It consisted primarily of Late Archaic hafted bifaces, which prove d useful to this study. 84


Figure 3-3. Ichuaway Survey Area. 85


Sudderth Farms Sudderth Farms (Figure 3-4) is a small, privately owned farm located in the northeastern corner of Cal houn County along the central west ern edge of the Dougherty Plain. There is a large wetland along its east ern and southern edges in addition several smaller upland ponds directly in the survey area. Elevations range from 100.58 m in higher areas to 78.94 m at its lowest point. As a farm, it is actively plowed and there is little surface vegetation to hinde r visibility. The area was surveyed in the late spring just prior to planting so surface visibility was amen able to high intensive survey. Fields were walked at evenly paced 10-m intervals to locate and record archaeological sites. Magnolia Plantation High Magnolia Plantation (Figure 3-5) is an industrial style farm located in east-central Calhoun County near the border with Doughert y County in the northern most portion of the central Dougherty Plain. It has the highest elevation in the Chickasawhatchee Archaeological Survey. The areas highest elevation ranks at 158.49 m while its lowest point is 109.73 m. There is also a large wetla nd that lies along the survey areas eastern edge. Smaller ponds are interspersed within the survey area. Open fields at Magnolia were amenable to full coverage systematic su rvey. Transects were paced across the fields at 10-m intervals. Red Bluff Plantation Red Bluff Plantation (Figur e 3-6) is a hunting planta tion in northern Baker County in the central portion of the Doughert y Plain. It shares a conterminous boundary on the southeastern edge of the Chickasa whatchee Wildlife Management area and encompasses a large number of wetlands due to its close proximity with the 86


Figure 3-4. Sudderth Farm Survey Area. 87


Figure 3-5. Magnolia Plantation Survey Area. 88


Figure 3-6. Red Bluff Pl antation Survey Area. 89


Chickasawhatchee Swamp. Elevation levels range from highs of 76.20 m to lows of 45.72 m. Because this area is primarily used for modern hunting survey consisted of the examination of dirt road surfaces and sma ll feed plots for deer and other wildlife. Bermuda Plantation Bermuda Plantation is a privately owne d agricultural plantation located in southeastern Calhoun County (Figure 3-7). It is just to the southwest of the Keel Creek Preserve. Elevations on the property r un from a high of 66.14 m to 54.86 m in lower areas. The survey area consisted of two row s north to south trending linear lines of round fields. An area interspersed with small wetlands lies between the two lines. Keel Creek Preserve The Keel Creek preserve is a privately ow ned that sits on the southwestern corner of the Chickasawhatchee WMA (Figure 3-8). It is primarily forested in pine trees and interspersed with small upla nd wetlands. Hardwood species do occur on the property but they are primarily confined to areas of lowe r elevation. Elevation ranges from a high of 67.06 m to a low 53.34 m. The property is managed for timber harvests and also conservation with the long term goal of refore sting the area in longleaf pine. As a result, much of the property is forested and there were few open areas to examine besides dirt roads, firebreaks, and food plots. 90


Figure 3-7. Bermuda Plantation Survey Area. 91


Figure 3-8. Keel Creek Preserve Survey Area. 92


Elmodel WMA The Elmodel Wildlife Management Ar es is a 7000 acre state owned wildlife preserve in central Baker County, an area th at is the central portion of the Dougherty Plain (Figure 3-9). It had prev iously been an active farm prior to being acquired by the state. Several large fields were inspected fo r sites. Long term ma nagement plans for the Elmodel call for encouraging the planting and re growth of longleaf pine forest in upland areas. Much of the property is routinely cut an d plowed to help implement this plan. As a result, survey conditions were optimal for survey. The Elmodel also encompasses the confluence of the Chickasawhatchee and Ichua way Nochoway creeks. There are also a number of upland wetlands on the property. Its highest point of elevation measures located 57.91 m while lower el evations fall to 42.67 m. Locke Farms Locke farms is a small privately owned fa rm in southeastern Terrell County, an area that lies just outside the western boundary of the D ougherty Plain (Figure 3-10). Elevations on the property run from a hi gh of 100.58 m to a low of 78.94 m. The area fortuitously rests between a seri es of upland wetlands. Two la rge wetlands are just off its eastern edge while several smaller ones lie close to the west. Ken Sanders Ken Sanders is a small privately owned farm just outside the northwestern boundary of the Elmodel WMA in central Baker County (Figure 3-11). It is primarily an upland area with few upland wetlands directly within the area surveyed, though there is a single pond located about 300 meters to the east There are however, a number of ponds 93


Figure 3-9. Elmodel WMA Survey Area. 94


Figure 3-10. Locke Farms Survey Area. 95


Figure 3-11. Ken Saunders Survey Area. 96


in the northern portion of the Elmodel which would have likely aff ected land use in the area enclosed by Ken Sanders. Preliminary Results of Survey A total of 738 archaeological components were recorded during the Chickasawhatchee Archaeological Survey. Of those, 446 have components believed to date to the Archaic Period, and, of these, 143 ar e definitively Late Ar chaic in age, based on presence of diagnostic hafted bifaces. Mo st of the sites with Archaic components may be characterized as lithic scat ters, consisting of stone tool s and the debitage left behind from their production and maintenance. Following fieldwork, all artifacts were ba gged and boxed up for transport back to the Laboratory of Southeastern Archaeology at th e University of Florid a. There at the lab they were washed, sorted, and analyzed. They were later entered into an Access database along with locational coordinate s to facilitate a thorough exam ination of where particular artifact types are generally occurring across the landscape. The attributes measured during the initial analysis, their significance, and the steps taken to examine Late Archaic land use in the interior Coastal will be discussed in the next chapter. 97


CHAPTER 4 THE MATERIALITY OF HUNTER-GATHERER LANDSCAPES: STONE TOOL ANALYSIS, DISTRI BUTION, AND LANDSCAPE COVER Archaeological sites recorded during the fieldwork portion of the Chickasawhatchee Archaeological Survey span the entire temporal spectrum from the Paleoindian through Historic periods. Late Archaic sites, however, were the most commonly encountered and thus provide an opp ortunity to quantify artifacts from them and evaluate the extent of land-use practices of people who lived in the area during this time. Most of the Late Archaic sites may be characterized as lit hic scatter, consisting of stone tools and the debitage left from th eir production and maintenance. Sites often drastically differed in size from five to ten meters in diameter to la rger areas upwards of 200 m across. Recovered stone tools from the recorded sites represen ted several forms of hafted and unhafted bifaces, as well as unifaces, hammerstones, and grinding stones, which stand as the material remains of dis tinctive tool-using activities. Uses of these tools likely included a number of activities such as butchering animals, scraping hides, processing plant food resources, or producing and repairing stone implements. A close inspection of them will certainly reveal deta ils of their individual use lives but when contextualized within a broa der collection they have the greater potential to reveal intricate components of th e land-use systems in which they were a part. Inferring Late Archaic land-use patterns from these sites will involve developing meaningful site categories that accurately en compass the range of activities based on the stone tools and debris they contain. As a final product, I hope to develop a model of land-use that not only interprets the recorded s ites in relation to each other but also relates 98


them to the broader landscapelevel ecology of the interior Co astal Plain, and, ultimately, the anthropogenic actions that rendered the landscape suited to sustained settlement. Material Residues and Interpretations of Hunter-Gatherer Land Use The analysis of stone tools relative to their distribution within a larger landscape has been approached from a number of temporal, spatial, and geographic perspectives. This subject relates to both the study of materials from pr ehistoric sites and also the ethnoarchaeological examination of activity areas created by contemporary living groups of hunter-gatherers. As the field of archaeol ogy began to take on a more scientific bent in the early 1960s, archaeologists became more interested in artifacts as a means to interpret past behavior rather than just se rving as temporal markers, but were also believed to be reflective of th e behaviors that were carried out at particular localities (Binford and Binford 1966; Bordes 1961; Schiffer 1976). With time, as more ethnoarchaeological research came to the forefr ont, this emphasis on site activities was greatly expanded. No longer were individual sites simply seen as places to be studied to gain insight into past human behavior but ra ther as parts of a mu ch greater systems of land use (Binford 1977, 1978, 1979; Yellen 1977). Binford (1980) generalized patterned variation observed in land-use (or settlement-subsistence) stra tegies among hunter-gatherers in his forager-collector model. Binford argued that variations in the frequency and structure of settlement moves and associated foraging practices are best understood as adaptations to effective temperature of a given region of the world, pa rticularly as it determined the time-apce variations of biotic resources of economic value to humans. For example, he posited that foraging was a more conducive procurement strategy for people living in highto 99

PAGE 100

moderate temperate regimes with little variab ility in the distribution of resources across space and through time. Sites of residence were established at locations of food, and then relocated to a new location as the source of food had dropped below the point of diminishing returns. Because foragers were believed to pursue resources on an "encounter" basis, they were not expected to leave behind functionally specific sites related to their activit ies (Binford 1980:5-10). The material remains from this type of land-use were predicted to resu lt in a somewhat continuous sp atial distribution of similar occupations (Binford 1980:7). This contrasted with collector land-us e which was expected to leave behind clustered distributions of both base camps and stations from a strategy entailing logistical procurement of resources from a central base. Collector strategies are expected to exist whenever there are marked variations in the time-space distribution of resources. Whereas environments of low effective temp erature tend to support heterogenic resource structure, Binford noted that any conditions that prevent a group for relocating residential bases will lead to a collect or strategy of land use. From a technological perspectiv e, tool kits were expect ed to vary with land-use strategy and thus serve as an archaeological proxy for mobility, settlement range, and resource procurement (Binford 1980; Shott 1986). It was posited, for instance, that forager toolkits would be less diverse, less formalized, less specialized, and more expedient, than those of collectors. Sassaman et al. (1990) illustrate this point in their discussion of sites in the Mi ddle Savannah River valley of th e Southeast. They posited that collector technology should be formal, por table, efficient, anticipatory, and oriented toward specific tasks which contrasts with th e more situational and generalized strategies 100

PAGE 101

of foragers (Sassaman et al.1990:221-222). Site s left by collectors should reflect this diversity with residential bases, locations field camps, stations, and caches being spatially arranged based on their relationshi p to diverse resource patches (Binford 1980:12). With regard to the material residues creat ed by the activities of both foragers and collectors, Binford (1980:12) was quick to point out that the two land-use strategy were not mutually exclusive, but, as a model, repr esented a spectrum of approaches to land use that ranged from simple to complex. Rare ly have archaeologists acknowledged that a land-use strategy incorporated both forager and collector aspects in the seasonal round. The chief exception is the model posited by Anderson and Hanson (1988) for Early Archaic land use in the Savannah River valle y, in which they argued that both strategies were used in response to seasonal changes a nd physiographic contrast s within basin-wide mobility ranges. The underlying inferential value of Binfor ds model is that in tersite variability can be expected when a large number of di fferent activities took place across different locations with tools that were differentially designed, manufacture d, used, and discarded, in response to their intended roles in a to tal settlement-subsiste nce system (Binford 1979:271). This should be further reflected in the distinctive places on the landscape where tools are recovered by archaeologists. As a cautionary note, distinguishing among site types, such as collector bases and short-term procurement camps, may be problematical. Because mobile groups will periodically shift their ranges in ways that restructure the time-spa ce relationship among resources, sites may be reoccupied by th e same people but serv e different purposes 101

PAGE 102

(Binford 1982:10). For example, a longer-term o ccupational area initiall y in the center of a range may later fall on its outer periphery wh en a group adapts its settlement pattern to incorporate unexploited areas ou tside its original range. Rather than the number of diverse activities taking place as would be expe cted of materials recovered from a site previously used as a base camp, a new series of shorter-term tasks would replace them. Consequentially, the overall dens ity of archaeological materials at the site would increase even though the activities re lated to their deposition w ould be quite different. Differentiating between these land-use strate gies may be more easily accomplished in a stratified situation but they are still useful in surface contexts if certain diagnostic artifacts are present. Shorter-term activiti es such as routine tool maintenance leave behind a narrower range of artif acts such as flakes but areas that are inhabited for much longer periods of time will reveal material resi dues of a greater dive rsity of activities. Despite these potential drawbacks, much of the early work conducted by Binford and others to infer behaviors from the material remains of past hunte r-gatherers has been highly influential among members of the southeastern archaeological community. Previous Applications in the Southeast The common occurrence of lithic scatters recorded during the Chickasawhatchee Archaeological Survey is similar to earlier find ings farther to the north in the Piedmont of Georgia and South Carolina during the mid 1980s when archaeologists relied heavily on the forager/collector dichotomy in analyses of artifact assemblages. An outstanding problem during this time was related to unders tanding the nearly ubiquitous occurrence of quartz lithic scatters (Canouts and Goodyear 1985:180). They were accurately characterized as being relatively small in size and lacking sign ificant levels of 102

PAGE 103

stratigraphy due to natural deposition processe s in the uplands coupled with erosion stemming from many years of inefficient farming methods during the historic period. Many were often ignored outright as having little to offer in the way of research potential (Canouts and Goodyear 1985:180). Given their extensive geographic distribution they were commonly defined as limited activity sites that represented the remains of resource exploitation or other distinctive functions su ch as hunting, nut-gathering, or lithic material acquisition areas (Canouts a nd Goodyear 1985:185; Goodyear et al.1979). Artifacts recovered from these sites were usually found on the surface or in shallow plow zone deposits and assemblages reportedly consis ted of small thinning flakes to whole and broken bifaces, which correspond with the varyin g stages of stone tool production such as resharpening, breakage, and discard (C anouts and Goodyear 1985:186). One of the primary issues stemming from this research problem was the apparent absence of any recognizable patterning in th e location of sites (Goodyear et al. 1979:255-261). They were seemingly recorded anywhere and ever ywhere regardless of spatial or locational attributes such as slope, el evation, or topography. House and Ballenger (1976) took this observation a step further in thei r analysis of site distributi on in relation to streams where they concluded that there was little correlati on between the internal composition of sites and their relationship to local ar ea streams of varying sizes. Problems associated with adequate ly defining the spatial and locational occurrence of the sites discu ssed above also directly relates to the difficulty of understanding the patterns of Middle Archai c land use in the both the Piedmont and Coastal Plain physiographic provinces. Arch aeologists have long r ecognized a dramatic increase in the number of sites in the Piedmont that date to this time period, a pattern that 103

PAGE 104

also coincides with a near absence of similar sites to the south in the interior Coastal Plain (Blanton and Sassaman 1989; Ca nouts and Goodyear 1985; Sassaman 1991; Stoltman 1974). Explanations for these divergent dist ributions of sites include changes in the environment, resistance to social oblig ation, and a general de crease in population or "abandonment" of the Coastal Plain (A nderson 1991; Blanton and Sassaman 1989; Faught and Carter 1998; Sassaman 1991). Most notable for th e discussion here is the characterization of Middle Archaic groups in the Piedmont as pursuing a foraging strategy of resource procurement based on r ecovered artifact assemblages that were generally identified as small, low-density scatters. Tools associated with these sites are technologically redundant with little differences in density or diversity between sites (Blanton and Sassaman 1989:53; Sassaman 199 1). Sassaman (1991:31) notes that Piedmont Middle Archaic tools fall within f our general classes; hafted bifaces, other bifaces, unifaces, and utilized flakes with de bitage occurring as thinning flakes, other flakes, and chunks. Though much of the discu ssion in this paragraph is focused on the Middle Archaic it is nonetheless relevant for the broader issue of understanding Late Archaic land use in the interior Coastal Plain given that archaeologists interested in this research problem have relied on stone tool assemblages to develop sociocultural categories for prehistoric hunter-gatherers. Assemblage redundancy has also been forw arded as the primary evidence that the people who produced them were foragers who moved as an entire residential unit from camp to camp rather than using a system of base and logistical camps. Arguments for this type of strategy have been predicated largely on the homogenous forest cover in the Piedmont during the mid-Holocene (Bla nton and Sassaman 1989; Sassaman 1991). 104

PAGE 105

Resources were fairly evenly spaced and predictable so they could be easily exploited as opposed to the patchy, heterogeneous resour ce base in the Coastal Plain (Sassaman 1991:36). In their discussion of Late Archaic la nd use in the middle Savannah River Valley, Sassaman et al. (1991) recognized three distinct types of lithic assemblages. They were noted as biface production areas, larger sites with a greater diversity of tool forms located on terraces relatively close to the Savannah River, and lower density remains located along upland tributaries (Sassaman et al. 1991:311). Based on the materials recovered from these site types, Sassaman et al. ( 1991:314) presented a hypothesis for Late Archaic land use which focused on a series of lowla nd aggregation camps that were occupied by several groups on a seasonal basis prior to dispersing into the uplands. Much of the evidence for multi-household aggregation s ites was based on the occurrence of high levels of seasonally available food resources during the spring such as migrating sturgeon which are large in size and could be taken in great numbers. As opposed to relying on the forager/collector categories as pro posed by Binford (1980), Sassaman and his colleagues instead developed aggregation and dispersed settlement categories to account for hunter-gatherer land use duri ng the Late Archaic. Regardless, their arguments are still predicated heavily on recovered stone tools, the debitage left over from their production, and other exotic lithic mate rials such as soapstone. While the sorts of sites and assembla ges discussed above are similar to many recorded during the survey work in southwest Georgia, a nota ble difference may lie in the ecological differences between the Piedmont a nd interior Coastal Plain. Most important is the discrepancy in forest cover between the two regions. The Piedmont consists 105

PAGE 106

primarily of mixed hardwood and southern pine forests, which contrasts with areas to the south which prehistorically would have been predominantly forested in longleaf pine and wiregrass, particularly in upla nd areas (Earley 2004). As prev iously discussed, this forest regime is often referred to as a "fire fo rest" given its predilection for regular, lowintensity fires that ensure forest health, maintenance, and longevity (Earley 2004; Pyne 1982). Rather than just viewing sites in the Coastal Plain from strictly the perspective of their proximity to resource areas such as str eams or rivers it may also prove useful to examine their spatial occurrence in their relationship to landscape patches. In this instance, patches include upland areas that would have been forested in longleaf pine and wiregrass, along with depressional wetl ands which together would have created a patchwork mosaic of ecotones with higher levels of biodiversity. While sites will certainly occur in areas where there are plen tiful aquatic resources they should also be located in upland locales where people coul d potentially take a dvantage of an area's unique resource base. This type of site distribution may be similar to the findings of John Keller (1982) based on archaeological work in the Coastal Plain uplands of Louisi ana and east Texas in which he made a connection between the occu rrence of lithic scatters and upland areas located in pine forests. His work was unique in this regard because he was an early proponent of relating lithic sca tters with particular types of landscape cover. Many of the conclusions drawn by Keller (1982) were base d on the recovery of small, homogenous, low-density lithic scatters recovered in interriverine areas of the interior Coastal Plain. As opposed to larger base camp sites in rive rine settings, these upland sites were quite small and lacked much internal variability (Keller 1982:40). Keller did not attempt to 106

PAGE 107

differentiate between scatters but he did notice some minor po tential preferences such as a decrease in evidence of for heat-treating of materials recovered in the uplands and a slight tendency for site locations to occur on finger ridges that extended into lowland areas. It spite of these observations he was still quick to point out that the overall distribution was for the most part unpred ictable (Keller 1982:42) Even though sites contained abundant lithic artifacts, most artif acts were debitage, while hafted tool forms and utilized flakes comprised only minor pa rts of assemblages. The methodology to deal with these sites was principa lly concerned with establis hing a functional relationship between the attributes of recovered mate rials and specific behavioral activities. Conclusions from Keller's study emphasized that flake production was the primary goal of tool production, which was geared towards butchering animals such as deer, turkey, and quail, all commonly found in upland localitie s. In this capacity the sites likely represented the remains of single butchering events (Kel ler 1982:40). Evidence in support of this hypothesis was found in the ed ge angles of many flakes, which Keller considered to be conducive to butchering ac tivities. It bears mentioning that Keller attempted to incorporate ecologi cal elements of the longleaf pine forests in which they were located into his model of land use, particularly the potential relationship between people and the perpetuation of the fire forest. Much of this chapter's discussion thus far has focused on the occurrence of lithic scatters and the problems they have presen ted for archaeologists attempting to accurately interpret them. As has been shown, research ers have tended to emphasize various aspects of hunter-gatherer lifestyles such as soci ocultural organization based on the composition of landscape cover and degree of mobility. What is often overlooked in these pursuits is 107

PAGE 108

the understanding that landscape cover can be hi ghly localized and that people can play a significant role in the broadscale distribution of plant a nd animal species. A primary goal in pursuing this line is the development of a model of land use that will not just highlight the internal variability and locationa l attributes of the r ecorded sites but also relate them to the greater ecological com ponents of the longleaf pine forests of the interior Coastal Plain. Beyond Foragers and Collectors: Incorporating Historical Ecology into Hunter-Gatherer Land Use Many of the models of land use discussed above are based heavily on the early ideas of Binford and others, particularly regarding the dichotomy of foragers and collectors. While this work has been inst rumental in the development of a greater understanding of prehistoric land use related to hunter-gatherers, they fall short in their overall characterization of peopl e and their relationshi p to the environment. A number of factors are involved in this problem such as th e origins of the Archa ic Period in general and its initial emphasis on people mapping on to the environment and adapting to it (Caldwell 1958). Culture was further viewed as an adaptive technique for people dealing with environmental strife during the era of the Processual or New Archaeology of the 1960s and 1970s, when it was held up as the human species primary means of adaptation. Rather than simply relying on independent invention, diffusion, or population movements to account for culture change during the cu ltural historical paradigm of American archaeology, the envir onment was swapped as the primary driver for culture change. In many instances the e nvironment is inadverten tly cast as a static backdrop against which human culture plays ou t. Changes in cultural activities have 108

PAGE 109

traditionally been interpreted as coinciding with changes in the environment. Taking more of a historical ecological perspectiv e here I hope to reca st prehistoric huntergatherers in the interior Co astal Plain as actively engagi ng their surroundings and in many instances having a significant influence in shaping them. Early criticism of Binfords (1980) la nd-use model focused on its lack of emphasis regarding relations among different social groups occupying the same landscape. (Weissner 1982), for example, c ountered that while f ood resources certainly played a central role in seasonal moveme nts, they were also influenced by the relationships between people. Other shortcom ings, as are emphasized stem from his lack of attention given to the abil ity of hunter-gatherers to shap e the landscapes in which they live. This position was not restricted to Binford. In his discussion of environmental features within the Kalahari Desert re gion as experienced by the Dobe' Kung, Yellen (1977) identifies the occurrence of small, sh allow, seasonal pans that fill with water during periods of high rain. He also menti ons the !Kung practice of starting late summer fires to encourage th e growth of new grass around th em (Yellen 1977:34). Even though Yellen was of the opinion that material remain s provide an exact refl ection of the system of which they were a part, he did not cons ider the relationship of site location to landscape cover except vaguely in his discu ssion of seasonal camps. The lack of attention paid to hunter-gatherer landscape modi fication in early land-use models is likely a consequence of the harsh, marginalized en vironments of modern groups. The frozen landscape of the arctic, for example, wher e many ethnographic observations have been made, presents little opportunity bey ond local impacts for people living there. 109

PAGE 110

Application in the Analysis of Recovered Materials From the Chickasawhatchee Archaeological Survey As a starting point for the research presented here, I accept that sites recorded in the Chickasawhatchee and Ichuaway-Nochaw ay creek drainages are only a partial refection of the activities that were carried out at these vari ous locations prehistorically. Because stone tools and debitage were the only materials recovered they merely represent a small portion of overall site function as ther e were also likely organic media that did not preserve. Secondly, because most of the mate rials presented here were recovered from surface contexts their utility is somewhat limited. Site density, for example, is difficult to assess and quantify because there is little in the way of horizontal or temporal control beyond the presence or absence of diagnosti c artifacts. As was discussed above, settlement pattern strategies can be shifted to counter decreases in resource availability. Based on this reality, a site that once served as a longer term hab itation area can later become the locus of shorter occupations and a se ries of different activit ies. This type of change can dramatically affect the overall assemblage composition and density of a site. Keeping such analytical biases in mind, I am interested in examining the spatial relationship between certain di agnostic materials and natura l features such as rivers, ponds, streams, and upland patches. What pe ople doing at sites associated with natural features and interassemblage differences am ong such sites should be reflected in the types and proportions of lithic artifacts. Understanding site activities on a site by site basis will be determined by the types of artifacts recovered from them, such as hafted bifaces, unhafted bifaces, flakes and other de bitage, cores, hammerstones, and grinding stones. 110

PAGE 111

Hafted bifaces are the most diagnostic arti fact type recovered during this project. Made from chipped stone, they are comprise d of two elements, th e blade which is the working portion of the tool, and the haft element used to affix the tool to a handle or shaft (Andrefsky 1998:21). The latter is considered to be temporally sensitive since they have been observed to change through time based on their recovery from in-situ stratified contexts (e.g., Coe 1964). Late Archaic hafted bifaces are noted for their usually broad stems and blades which are hallmarks of the period (Bense 1994:85). Hafted bifaces may also be placed within a broader realm w ith bifaces known as formal and expedient technologies. As implied by the names, formal tools are those that require a great amount of effort in their production and expedi ent tools are those that do not (Andrefsky 1998:30). Hafted stone tools recorded during this project were analyzed based on their overall size, shape, degree of completeness, breakage patterns, heat treatment, and the temporal sensitivity of their intact hafting elements. While not being as temporally sensitiv e as hafted bifaces, unhafted bifaces are unique in their own right since they may be used as tools but they also can be relied on as a source of flakes. Several prominent lithic specialists have commented on the long-term production of stone tools (Andrefsky 1998; Bi nford 1979; Kelly 1988). Kelly (1988:718) in particular notes that bifaces are flake or co re blanks that have been reduced from either face using percussion to produce a bifacial cutt ing edge. Taking this as a starting point, hafted and unhafted bifaces fit into a trajectory which consists of a series of five stages of production (Andrefsky 1998:31). Earlier stage to ols are useful as choppers or otherwise heavy maintenance tools given their robust si ze. They can be resharpened as they become dull with the subsequent flakes also be ing used as cutting tool s. As they become 111

PAGE 112

progressively thinner from continued sharpeni ng, their basal areas may be worked into hafting elements for placement in a handle (Andrefsky 1998:30). Following Fish and Fish (1976), it may also prove fruitful to examine the spatial occurrence of thick and thin unhafted bifaces, especially if they were potentially being used as cores. Thicker ones may be more apt to be recovered fr om areas where people were spending longer periods of times versus thinner ones, which would have a narrower cutting edge,making them more effici ent for cutting or butchering. Flakes may be divided along two lines, they can be byproducts of the production of stone tools and they can also be used as simple cutting tools for short-term tasks. Recovered flakes from the Ichuaway-Nochaway and Chickasawhatchee creek surveys were analyzed for their size, whether or not they were heat treated, and whether they were whole, or were the proximal or distal portions of the orig inal flake. This analytical scheme is a variation on the theme origina lly presented by Sullivan and Rozen (1985). While their original study has been criticized heavily due to problems of replicating some of their analytical categories through experi mental archaeology, it is still held in high esteem given its emphasis on typological categ ories for debitage as opposed to taking an attribute based approach (Andrefsky 1998:111; Shott 1994). As implied by the name, whole flakes are those that ha ve their proximal, medial, and distal sections intact. A proximal flake refers to a flake fragment in which the striking platform and bulb of percussion is still present (Andrefsky 1998:88). In contrast, a distal flake is missing its bulb and striking platform. Whole and proximal flakes were noted for whether they had a lipped or unlipped bulb. Lipped bulbs are more likely to be indicative of being removed from bifaces or bifacial cores where as unlipped flakes are more likely to have been 112

PAGE 113

removed from amorphous cores (Odell 2004:121; Shott 1994:78; Sullivan and Rozen 1985:758). Cores are non-bifacial tools that were used as a source of flakes to be subsequently used as or further made into tools (Andrefsky 1998:80). There are a number of types of formal core types such us unidirectional, multidirectional, bipolar, and amorphous which have been recognized by archaeologists working in divergent contexts around the globe. Cores recovere d from the field were measured for their dimensional characteristics such as maximum length, width, and thickness. They are an interesting find considering the density of natural chert source s in the Doughtery Plain. In addition to being a source of raw material th ey may also have been used as cutting or chopping tools so their locational attributes are as noteworthy as their physical characteristics (Andrefsky 1998:80). Hammerstones are used to detach flakes from cores and bifaces in hard-hammer percussion (Andrefsky 1998:11). Several hamm erstones were recovered over the course of this project. All of the recorded hamme rstones are river cobbles. Evidence that they were used as hammerstones comes from the distinctive wear patterns on their elongated edges. Continued use has a tendency to remove the outer river worn cortex to reveal the inner quartzite interior. With regard to analysis, all hammerstones were examined for their overall size dimensions and presence or ab sence of wear. Some of the cobbles were small and were unlikely to have been usef ul as hammerstones. Therefore the overall analysis of their distribution will focus on thei r presence or absence at sites coupled with whether they have surface wear. While not used as hammerstones, smaller stones were 113

PAGE 114

still transported to these interriverine ar eas, potentially during later periods for the purpose of burnishing pottery. Besides being used as hammerstones, rive r cobbles may also be used for grinding stones. A number of these stones were also recorded over the course of the project. These stones were likely used to grind up plant foods and potentially hard seeds so they could be more easily consumed. This process usually requires grinding one stone against another to achieve the desired effect of wearing down resistan t hulls and cellular structures. Tools of this sort are often r ecovered from archaeologi cal contexts in pairs and referred to by a number of descriptors su ch as manos and metates, millstones, and nutting stones (Odell 2004:79). As opposed to hammerstones, which tend to have significant amounts of wear on their edges, grinding stone s have pronounced areas of surface wear on one or both faces. Similar to the way the recorded hammerstones were analyzed above, grinding stones were measured for their overall size dimensions and their presence or absence will be utilized to quantify the broader-scale analysis of site distributions. Combined, the en tire list of artif act types will be examined based on the discussed attributes in order to determine as pects of past land use. The boundaries of which are the topic of the following section. Parameters for the Development of Categor ies of Hunter-Gatherer Activity Areas The development of categories of hunter-gatherer activity areas will be based on a number of variables related to the production, maintenance, and use of stone tools. Stone tools used as a descriptive term here enco mpasses all hafted bifaces, bifaces, flakes, hammerstones, and grinding stones. As was previously mentioned they mark the only existing material remains left on any of the sites recorded during the project. Many 114

PAGE 115

activities may be inferred from them base d on their relationship to other materials recovered from any particular site. For example, large numbers of flakes found in association with other forms of debitage along with exhausted and failed bifaces may be indicative of a retooling area. As a starting point, I will begin with a number of queries related to recovered flakes. Using this as a foundation I will then create categories to represent the range of activiti es undertaken by prehistoric hu nter-gatherers who inhabited portions of the interior Coastal Plain. The goa l of this undertaking is too divide artifacts into classes based on their locat ional recovery and their asso ciation with other artifacts and overlaying that with lands cape-level vegetation coverage. Flakes and debitage stand as outstanding markers for the activity of producing or maintaining stone tools. For the purpose of study here I will examine the size of flakes relative to location following Kelly (1988:721). In his discussion of flake type distributions, Kelly (1988) outlined his argum ent for the initial pr oduction of bifaces in residential areas which were later used as cores at logistical sites. Archaeological evidence for this distribution should occur in residential sites having large numbers utilized biface-reduction flakes, whereas logistical sites would have much fewer examples of this artifact type. Virtually no utilized flakes were recorded during the survey which may result from the common o ccurrence of natural chert outcrops in the project area. Even though few utilized flak es were noted, Kelly's (1988) argument may still be valid if overall flake size is cons idered. The distribution of flakes by size may lend insight into aspects of overall land-use since larger fl akes may be indicative of earlier stages of biface production where as smal ler flakes may be more reflective of tool maintenance and therefore take place in ar eas where people were spending less amounts 115

PAGE 116

of time or at least were focusing on specific activities. Following this line of reasoning, flake size and location may be reflective of the overall activities taking place at an individual locality such as larger flakes being recovered at quarri es and/or habitation sites while smaller ones will occur at special-act ivity areas. Larger flakes would be more likely to occur during the ini tial phases of tool production wh ile smaller flakes are likely to result from resharpening and maintenance. The spatial occurrence of flakes by size will be accentuated by an examination of the distribution of whole and proximal lippe d flakes, which sites have the highest frequencies of them, and where they are locat ed within the broader landscape. Like the use of other categories, this one may be indi cative of areas where people were strictly performing maintenance level activities regard ing stone tools as oppos ed to outright tool production especially if the lipped flakes te nd to be small in size. As previously mentioned, lipped flakes are more likely to re sult from bifacial thinning as opposed to unlipped flakes which are greater evidence for core reduction (Odell 2004:121; Shott 1994: 78; Sullivan and Rozen 1985:758). Taking this a step further, a lower frequency of lipped flakes may be indicative of speciali zed activity areas where people would have primarily been maintaining tools. This dist ribution may also be sp atially significant if it illustrates that many of these specialized activ ity areas fall in upland locales especially considering the role of fire in them. A third level of analysis will include th e overall distribution of heat-treated and non-heat-treated flakes. Heat treating is an inte nsive activity requiri ng the controlled use of fire to thermally alter th e crystalline propertie s of poor quality Coastal Plain chert. This type of undertaking would be more likely to occur at a longer-term habitation site as 116

PAGE 117

opposed to an area of more limited activity as suggested by Keller (1982). Heat-treated flakes should also be expected to be encountered in locations of specialized activity as a result of the further bifacial reduction of tools made from already heat-treated Coastal Plain chert, but not in the sa me quantities as areas where initial heat treating and tool making took place. The distribution of flakes by size and li pping will also be expanded to include their association with bifaces. Flakes in th is portion of the analysis should reflect a greater frequency of lipping since they were being removed from bifacial cores or were the result of the thinning of bifaces. Similar to the discussion above, this distribution should be reflective of areas where people were both producing and maintaining stone tools which were also likely to occur in areas of increas ed habitation. Kelly (1988) echoes this position in his overv iew of the archaeological co nsequences of stone tool production. He argues that there should be a concentration of bifaci al flaking debris in residential sites which is generally dominated by small bifacial retouch debris along with a positive correlation between biface fragment s and bifacial debris (Kelly 1988:723). This should also be followed by a low inciden ce of biface reduction flakes used as tools at these locations. Flakes with lipping and broken bifaces could also be potentially located in non-residential areas though they will be in fewer in number compared to places where people were spending prolonged periods of time. While flakes with lipping are indicative of the production and care of bifaces, flakes without lipping stand as evidence that they originated from cores or from a preliminary stage of tool production (Sullivan and Rozen 1985). Close inspection of their distribution is notable here because there are two potential localitie s where they may be 117

PAGE 118

located. Based on this range of potential use they may be found in association with high frequencies of flakes and debitage left ove r from people making stone tools. In this capacity they may also be associated with a broad range of other materials. They may also be recovered from more ephemeral si tes though the frequency of other materials associated with them should be much lower in comparison to places where people were spending longer periods of time. The distribution of cores will also be gauged against the size distribution of flakes. Ephemeral sites s hould have a high frequency of small flakes whereas longer-term sites will be more representative of flakes from the entire range of the reduction process (Johnson 1996:119). Sites with cores are also of interest here given their special re lationship to both the availability of raw materials a nd for the production of flakes for use as tools and also to be made into tools. Despite their utility as a source of flakes, cores may also be used as cutting or chopping tools. Hafted bifaces recovered during the projec t were analyzed for their shape, size, breakage patterns, and temporal sensitivity. To broaden this analysis beyond the realm of just physical attributes and use-wear, I also examine their sp atial distribution across the landscape, particularly with regard to sites that were redundantly inhabited through time. The Late Archaic spans a range of roughly three thousand years. Even though stemmed, broad-bladed hafted bifaces are the hallmark of this time period there is also some degree of diversity among types through time. Looki ng at the distribution of hafted bifaces through time based on chronological position may reveal changes in land use patterns that occurred over the course of the Late Archaic. This follows from Sassaman's (1991) analyses of Middle Archaic materials from s ites in South Carolina in which he noted the 118

PAGE 119

locational differences between single and mu lti-component sites. Reoccupation at some sites in the project area may highlight preferences for certain areas or landforms through time. This could also be an important variab le to explore under the rubric of fire ecology especially if people were aware that certain areas were more suscep tible to burning than others. Longer-term sites would therefore be more likely to occur in areas that were less susceptible to burning. Even though their occurrence was rare comp ared to those of other locations, sites with just hafted bifaces will also be examined to see where they fall within the greater landscape. In all cases, biface-only sites consis ted of a lone hafted biface recovered in an area that was otherwise devoid of flakes, debitage, or any other form of stone tool. What this unique occurrence represents at this time is not known but it may be considered a site type by itself without the aid of further analysis given th e lack of other outstanding materials. In all likelihood they represent isolated activities such as butchering or alternatively they could also denote an epis ode where the tool was dropped, misplaced, or discarded. Summary and Conclusions Variation in the composition and dens ity of lithic assemblages across archeological sites has long been considered indicative of land-use patterns among people who use mobility as a strategy to exploit resources. In gene ral, assemblages consisting of simple, informal tools that are redundant acro ss locations in a region are indicative of a forager strategy, which is expected to o ccur under conditions of limited time-space variation in the availability of resources. In contrast, assemblages consisting of diverse, formalized tools that vary from location to location are indicative of a collector strategy, 119

PAGE 120

which is expected to occur under marked ti me-space variation in the availability of resources. We can also assert that, in general, the durati on and/or redundancy of use of particular sites on the landscap e will vary positively with the density and diversity of lithic artifacts. The distribution of lithic raw materials greatly affects these generalized relationships, particularly when raw material s are highly localized a nd short in supply. The present study area, however, is rich in si liceous stone suited to chipped stone tool manufacture, and river cobbl es throughout the region provide sources of hammerstones and grinding stones. The differential availability of lithic raw material across the study area is thus not expected to be a major cause of interassemblage variation in assemblage composition or density. Efforts to model hunter-gatherer land-use to date have not seriously considered the notion that anthropogenic changes to th e landscape have the potential mitigate the direct relationships between toolkit design a nd the time-space distribut ion of resources. The application of management practices, lik e burning, would have lessened the demand on stone tools alone to facilita te the acquisition of food resour ces, not in the direct sense of tool-using activities, but rather in the way stone tools are responsive to systems-wide variables such as the predictability of resour ce-seeking activities. In the next chapter, the structural relationship betw een stone tools and anthropoge nic fire will be examined. 120

PAGE 121

CHAPTER 5 STONE TOOLS AND FIRE ON THE LANDSCAPE The significance of the technological or ganization approach to lithic artifact analysis is that it relates properties of stone tools to dimensions of environmental and behavioral variation far beyond the use of the tools themselves. Chief among the relevant environmental variables is the predictability of food resources, notably where and when they can be located and effectively exploi ted. In environments that are highly predictable, tools can be desi gned for specific tasks at specif ic places on the landscape at specific times of the year. When resources are not predictable, tools need to be generalized or multi-purpose, because tool us ers do not know in advance where and when resources will be encountered. As reviewed in Chapter 2, fire has an integral role in determining the composition, structure, and seasonal availabil ity of many resources important to humans, and fire can be implemented by humans to aff ect these conditions. It follows that the application of fire, like the design and depl oyment of stone tool technologies, can be considered a tool that humans use to meet th eir economic needs. But because fire has the potential to alter the time-spac e distribution of food resource s, its deliberate use, as a management tool, has the potential to affect the design and deployment of foodacquisition and processing technologies by m itigating the direct relationship between stone tools and environment, as proposed in models of technological organization. Some general hypotheses can be deduced from the foregoing observations. If fire was utilized to increase the predictability of the time-space distribution of food resources requiring stone tools to procure and process, then stone tool s would be more functionally 121

PAGE 122

specific and less generalized than under conditions of lower pred ictability. A corollary to this hypothesis is that stone t ool assemblages would show gr eater intersite variation under conditions of enhanced predictability b ecause tool-using tasks would dovetail into specific locations on the lands cape and recur with sufficient redundancy (say, year after year) to enhance functional specif icity. In general, the regular use of fire to enhance the predictability of food resources on the lands cape would encourage the development of collector strategies of re source procurement and the formalized, curated toolkits conducive to logistical mobility. After nearly three decades of research into the technological organization of Archaic hunter-gatherers in the Southeast, it is widely accepted that collector strategies became more common in the Late Archaic period as populations became more densely packed into various localities and maintained more perman ent, or at least redundant residential bases from which dive rse logistical forays were launched to locate and procure necessary food resources (e.g., Amick and Carr 1996; Anderson 1996). Although some archaeologists have given considerable thought to the notion that these conditions promoted the domestication of weedy and starchy seed-bearing species of economic value to humans (Gremillion 1996), no one ha s formally hypothesized that the routine application of fire to enha nce resource predictability was a major impetus to the emergence of collector strategies and in creased land-use redundancy. Coupling the widespread evidence for controlled use of fire by hunter-gatherers worldwide, the sustainability of fire-adapted biomes such as the longleaf pine -wiregrass forest of southwest Georgia renders such a hypothesis worthy of investigation. 122

PAGE 123

Using Fire to Increase Resource Pred ictability in Southwest Georgia As discussed earlier, the longleaf pine-wiregrass fore sts of southwest Georgia depend on frequent fire (every one to three years) to maintain its richness and health. It is, in fact, one of the most fire-dependent ecosystems in North America (Christensen 1987). Although longleaf pine itself may not ha ve factored into human economies in any significant fashion, it is the open understory of the fire forest that is critical. When subjected to regular fire, longleaf pine stands develop into ta ll, widely spaced trees whose crowns rarely touch, enabling sunlight to reach the ground to promote to the growth of wiregrass, legumes, forbs, and herbaceous pl ants. Legumes are espe cially noteworthy for their role in replacing nitrogen loss through frequent burns (Haine s et al. 1999), and because they are preferred foods for white-tailed deer (Czuhai and Cushwa 1968). The economic significance of the longleaf pine-wiregrass ecosystem for Archaic hunter-gatherers is not known directly, but in sofar as deer, turkey, and herbaceous plants factored into their diets, this ecosystem ha d potentially great signifi cance. Lightning fire alone may not have been sufficient to mainta in this ecosystem over long periods of time (i.e., centuries), and some foresters have s uggested that natural fires must have been supplemented by native burning (e.g., Robbins a nd Myers 1992). The extent, periodicity, and intensity of native fires are unknown to us, but sufficient knowledge is known about the fire ecology of longleaf pine-wiregrass ecosys tems to suggest that fire would not have produced uniformly positive effects if applied indiscriminately. That is, human-set fires, assuming they were set regularly, needed to be controlled in order to achieve productive, sustainable conditions. As is true in ecosystems throughout th e world, fire control in the longleaf-wiregrass ecosystems of southwest Ge orgia is a function of fuel, moisture, wind, 123

PAGE 124

and topography. Stream systems provide bot h the topographic and moisture gradients that enable burns to be controlled, but their sideslopes and bottomlands support not longleaf pine, but deciduous ve getation, which, in its own righ t, is highly significant for humans, but far less abundant (in ancient times) than longleaf pine forests in the greater southwest Georgia region. Upland ponds throughout the study area en hance the potential for controlled burning provided by stream systems. In fact upland wetlands are arguably more likely to have figured into fire rotations given th eir abundance and physiographic position relative to the longleaf pine-wiregrass biome. Both increased edge (pond margins) and overall greater soil moisture contribute to fire control by bounding spaces to burn and limiting the loss of organic matter in soil. As disc ussed earlier, archaeol ogists are beginning to appreciate that upland wetlands throughout the southeastern Coastal Plain long attracted human settlement. As in the use of controll ed burns by Indians of Alberta (Lewis 1980), Archaic denizens of the Dougherty Plain may have set pond edges on fire to both enable access to open water and to improve the leve l of primary production for creatures up the food chain. The relationship between longleaf pine-wiregrass forests and upland wetlands is indirect with respect to possible native burni ng. These two features of the greater region occur in relatively close proxim ity to one another (<1 km) in some locales, but they are often more distant. A host of environmen tal factors intervene to create tremendous diversity in ground cover, even lo cally, and to this we can add the controlled use of fire. Setting aside cause and motive, if Late Archaic populations of the Dougherty Plain targeted upland wetlands for settlements of s ubstantial duration (in regional pattern of 124

PAGE 125

seasonal movement, over decades), fire may have become a necessary tool to sustain sufficient economic returns wit hout relocating to a new localit y. And if the use of fire was intensified to sustain settlement in a given locality, longleaf pine-wiregrass communities would potentially persist, even spread, in locations proximate to upland wetlands of targeted settlement, all else being equal. Again, this is not a direct spatial relationship between upland wetla nds and longleaf pine, but rath er a pattern of regional, or landscape-level land-use that is best expressed by the contrasts of physiography, hydrology, and vegetation cover be tween the northern and sout hern portions of the study area. Situated squarely in the Dougherty Pl ain, the ca. 40-km long expanse of the project area expresses some differences with potential consequences for human settlement and controlled use of fire. By virtue of the survey design, the study area can be divided it into a north tract consisting largely of the Chickasawhatachee survey (see Chapter 3), and a south tract c onsisting largely of the Ichuaway survey. In general, the north tract has more and more reliable surf ace water than the south tract. At the south end of the north tract, stream courses leav e a near-surface limestone substrate and enter an unconsolidated sandy plain, which encourag es the development of narrower, incised channels and floodplains. Occasional submerge nce and reemergence of channels occurs in the south tract at locations of karstic erosion. Upland wetlands decrease in size and frequency from north to south in the sout h tract, leaving only stream channels as a reliable source of perennial water and wetland habitat. Both tracts are capable of sustaining longleaf pine-wiregrass ecosystems in a variety of settings, and both very well may have been dominated by these forests in 125

PAGE 126

precolonial times. Irrespective of original condition, the north trac t is less capable of supporting large, unbroken tracts of longleaf pine-wiregrass communities because it is punctuated so frequently by wetlands, upland and bottomland. Of course, this very condition of patchiness or heterogeneity offe rs countless benefits to organisms, like humans, with varied diets, as well as optimal habitat for controlled bu rning at the scale of the locality. The overall greater moisture re gime of the north tract makes it much less risky for burning than its southern counterpart. Ultimately, from an archaeological point of view, it is the predictability of the environment that is most relevant as this is what enables the development of routine practices that leave distinctive archaeologica l residues. For this project, the residues available for study are the stone tool and tool debris scat tered across the surfaces of hundreds of sites. It remain s now to bridge the hypothese s for fire use to observable dimensions of variation in lithic assemblages. The sorts of test implications pertinent to assemblages of stone tools are enabled by comparisons between the north and south survey tracts. Given the contrasting conditions of the north and south tracts for settlement and potential use of fire, we can propose a number of test implications for variation in stone tool assemblages at sites between the tracts. These imp lications follow from the gene ral hypotheses stated above, and with the assumption (that awaits testi ng) that the north tr act could be more successfully managed with fire to yield gr eater predictability than its southern counterpart. 1. Lithic assemblages will be relatively diverse and dense more often at sites in the north tract by virtue of longer, more repeated, and/or more intense habitation. 126

PAGE 127

2. Lithic assemblages at sites adjacent to upland ponds in the north tract will be as diverse and dense as sites adjacent to stream channels and other bottomland wetlands in the north tract. 3. Lithic tools from sites of the north tract will exhibit a greater level of functional specificity than tools from sites in the south tract. 4. Sites in the north tract will exhibit a higher level of interassemblage variation than sites in the south tract. The balance of this chapter examines the data available to address each of these test implications. Variation in Stone Tool Assemblages Drawing on the discussion of lithic artifact types in Chap ter 4, a simple typology of assemblages provides the necessary units for co mparison outlined in the test implications above. The sorting criteria for these types and their inferred function are listed in Table 5-1. This typology is not meant to imply that these types are mutually exclusive, for variation in duration or reuse of sites can alone account fo r differences between Types II and III, for example. However, given that these samples were collected from surfaces subject to all manner of post-depositional tr ansformations, including previous collection, the analysis here must remain coarse-grain ed until better controls are enabled. Despite the limitations of the data, the hope remains that patterning would be sufficiently robust to emerge from large-sample comparisons be tween survey tracts, as well as locational contexts within tracts. 127

PAGE 128

Table 5-1. Assemblage Typology and Inferred Site Function. Type artifacts function I cobble tools, bifaces, cores, and flakes prolonged/repeated habitation II bifaces and flakes short-term camp III only flakes transient use, butchering IV isolated hafted biface field loss, hunting The absolute and relative frequencies of assemblages by type and survey tract are given in Table 5-2. Components included in this analysis include any site with a demonstrable Late Archaic component (i.e., diagnostic hafted biface of Late Archaic age) and all other lithic sites lack ing pottery, which are presumed to be Archaic in age. Two uncontrollable biases are intr oduced by including sites lacking diagnostic Late Archaic artifacts. First, some of the sites lacking pottery are undoubt edly of the period of pottery making (i.e., post-3500 B.P.), merely locations that did not result in the breakage and deposition of pottery vessels. Second, some of the lithic site s undoubtedly predate the Late Archaic period. Debitage of early Ho locene age is often rendered conspicuously different from later debris because of an advanced level of weathering, but mid-Holocene (Middle Archaic) debitage is not easily distinguished for debr is of the subsequent period. Obviating these biases is the sheer domina nce of Late Archaic hafted bifaces in collections from the project area, signifying a marked intensity of land-use during this period, a pattern not uncommon for other locales of the upper Coastal Plain (e.g., Sassaman et al. 1990). 128

PAGE 129

As the totals in Table 5-2 show, Type III assemblages are the most common across the board. Type II sites are second in frequency, followed by Types VI and I. It makes intuitive sense that the densest (Type I) and least dense (T ype IV) assemblage types are relatively rare, the former because they map on to locations most desirable for human occupation, which are relatively rare, and the latter because they are like the proverbial needle in the haystack. The most frequent type, Type III, is useful not only for testing inferences about land-use pattern s, but also for checking observed patterning against sample bias. That is, the distribution of this highest frequency site type provides a basis for eliminating survey bias as an expl anation for the lack of lower-frequency site types in areas populated by Type III sites. None of the assemblage types as so limited in frequency as to undermine any attempt to locate nonrandom pa tterning in their distributions. The total sample of 446 sites is evenly divided between the north and south tracts (Table 5-2). By assemblage type, the frequencies of sites between survey tracts reveal marked differences. Figures 5-1 through 5-4 illustrate these differe nces graphically and provide an inferential basis for detecting di fferences in land-use patterns. The most dramatic difference is the clear tendency fo r Type I sites to occur in the north tract (Figure 5-1). The lack of Type I sites in south tract (Table 5-2) is not quite accurate, for it excludes a few such sites just beyond the major polygon of the tract. Also, there is a decided collector bias in the south tract that may have removed from some sites hafted bifaces and/or cobble tools (John Chamblee, personal communication, 2009), reducing these assemblage to Types II and III. One can add to these biases a relative lack of cobbles to the south. Still, considering these and other bias es, the tendency for Type I 129

PAGE 130

Table 5-2. Absolute and Relative Freque ncies of Sites by Assemblage Type and Survey Tract. Type North Tract South Tract Total I n 32 0 32 row% 100.0 0.0 100.0 col% 14.2 0.0 7.2 II n 88 37 125 row% 70.4 29.6 100.0 col% 38.9 16.8 28.0 III n 87 151 238 row% 36.6 63.4 100.0 col% 38.5 68.6 53.4 IV n 19 32 51 row% 37.3 62.7 100.0 col% 8.4 14.5 11.4 Total n 226 220 446 row% 50.7 49.3 100.0 col% 100.0 100.0 100.0 sites to be located in the north tract is highly robust. Consid eration of locational patterning among Type I sites within th is tract is taken up further below. 130

PAGE 131

Figure 5-1. Type I sites with hafted bif aces, bifaces, flakes, cobbles, and cores. 131

PAGE 132

Figure 5-2. Type II sites with hafted bifaces, bifaces, and flakes. 132

PAGE 133

Figure 5-3. Type III sites with just flakes. 133

PAGE 134

Figure 5-4. Type IV sites with just ha fted bifaces, notably isolated finds. 134

PAGE 135

Type II sites likewise show a nonrandom tende ncy to be located in the north tract (Figure 5-2). Well over two-thirds of the 125 si tes of this type are located in the north tract. This trend is reversed with Type III (Figure 5-3) and Type IV (Figure 5-4) sites, which are almost twice as frequent in the south tract as they ar e in the north tract. In short, the frequency distributions of site types by survey tract provide ample evidence to infer different patt erns of land-use to the north and south of the study area. There are absolutely no data to determin e if these divergent patterns are due to differential land-use, cultural divisions, or even chronological variations masked by the coarseness of lithic biface typology. One can la ment the lack of better data to address these issues, but instead we will review the four test implications listed above and draw whatever inferences these data allow. The first test implication derived from th e general hypothesis about use of fire is that lithic assemblages will be relatively divers e and dense more often at sites in the north tract by virtue of longe r, more repeated, and/or more in tense habitation. This appears to be borne out in the data available, alt hough the sampling biases mentioned earlier may lessen robustness of the patter n. Given its abundant surface water, the north tract clearly held great potential for not only riverine hab itation, but also interrive rine settlement in proximity to wetlands and ponds. The second implication states that lithic assemblages at sites adjacent to upland ponds in the north tract will be as diverse and dense as sites adjacent to stream channels and other bottomland wetlands in the north tract. This is generally true, but not all upland ponds with sites yielded diverse and dense assemblages, and some of the bottomland locations in the north tract produced assemblages of marked density and diversity. The 135

PAGE 136

overall pattern is one that is being observed elsewhere in the upper Coastal Plain (e.g., Brokks et al. 1996, 2001), namely, that upla nd wetlands were capable of sustaining settlement as well as riverine sites. In the south tract of the projec t area, upland wetlands are not as numerous and reliable as in the north tract, and they have a greater tendency to dry out during prolonge d periods of drought. The third implication is that lithic tools from sites of the north tract will exhibit a greater level of functional specificity than tool s from sites in the south tract. For this we have only anecdotal evidence, as well as th e observation that cobble tools are diverse and relatively common at sites in the north tract. Better evidence for functional specificity may be subtle, as in the shape and angl e of edges on standard hafted bifaces. Finally, the fourth implication is that sites in the north tract will exhibit a higher level of interassemblage variation than sites in the south tract. Th is is patently obvious given the greater evenness across site types within the north tract. No single site type accounts for more than 39 percent of all sites in the north tract. In contrast, Type III sites comprise almost 69 percent of all sites in the south tract. North tract sites, as a population, evince a much greater degree of inte rassemblage variation than those of the south tract. Discussion and Conclusion The relatively marked difference between the surface assemblages of lithic artifacts from sites in the nor th and south survey tracts is most likely due to divergent land-use patterns. There are many similarities in the environments of both tracts, but the differences outweigh the similarities, at least as they affected the dist ribution of classes of lithic artifacts recovered in this study. Th e chief difference would appear to be the 136

PAGE 137

greater amount of surface water in the north tract. Both areas have upland ponds, but those of the north tract are larger, more numerous, and more consistently watered compared to those of the south tract. Th ey were clearly capable of attracting and sustaining human settlement, and they would have positioned groups in close proximity (ca. 1-5 km) to upland stands of longleaf pi ne. The upland landforms that historically supported these stands contain an abundance of so-called lithic scatters (Type II and III sites, in particular), the sorts of residues we could expect from hunting forays away from sites of habitation. The south tract abounds in these site types, plus it contains an abundance of isolated hafted biface finds, i ndicative perhaps of losses during hunts (e.g., embedded in deer that escaped after being hit, or unretrieved projectiles from shots that missed their targets), but also possibly due to collector bias. Biases notwithstanding, the two survey tracts contain measurably different archaeological records, and the distinction is stro ng enough to be tentatively regarded as land-use variations. But to what extent ar e the differences the result of functional differentiation within one set tlement-subsistence system, as opposed to distinct patterns of cultural difference, coeval or not? Too fe w data are available to answer this question with confidence, but some tent ative answers can be proffered. Assuming that the differences observed in the study are indicativ e of functional or seasonal variation within a single, unified se ttlement-subsistence system, the north tract was the nexus of long-term and repeated sett lement, whereas the south tract encompassed a range of short-term, perhap s seasonally-specific activities. Given the physiographic and hydrologic differences between the tw o tracts, these inferences are hardly provocative. Potential exists for fire ma nagement in both areas to improve their 137

PAGE 138

economic utility and predictability, but they w ould have likely been treated differently for natural and cultural r easons. The natural reasons cente r on the distribu tion of surface water, ambient soil moisture, and surface topography. The cultural reasons center on the sensibilities about the proximity of locations for burning to habitable places. Without question, this level of functional differentia tion is well within the range of settlement scale among contemporary hunter-gatherers (K elly 1995), and even a minor fraction of the presumed annual mobility range of Earl y Archaic foragers of the Savannah River valley (Anderson and Hanson 1988). Alternatively, Late Archaic archaeologi cal records elsewhere in the lower Southeast show a marked le vel of group differentiation a nd niche specialization (e.g., Anderson 1996; Sassaman et al. 1988). If th is sort of process accounted for the assemblage variation between the north and so uth tracts, then we would be observing the residues of local groups with di stinctive land-use st rategies: a southern group oriented to stream channels and utilizi ng upland landforms for only s hort-term, specialized purposes (i.e., hunting deer), and a northern group that targeted upland wetlands, as well as stream channels, for habitation, and mapped onto a mo re heterogeneous landscape with greater site functional specificity. Fire may have been equally central to either strategy/group, but its implications for broader patterns and processes vary with context, a subject addressed in the closing chap ter of this dissertation. 138

PAGE 139

CHAPTER 6 HUNTER-GATHERER LIFEWAYS IN THE INTERIOR COASTAL PLAIN: CONCLUSIONS AND FUTURE DIRECTIONS Incorporating historical eco logy into preexisti ng models of Late Archaic huntergatherer land use provides the opportunity to develop new ideas related to life in the interior Coastal Plain. Previous land-use pa tterns pertaining to pe ople living in the region during the Late Archaic have been characteri zed by a shift from a system of residential mobility as utilized by groups during the previous Early and Middle Archaic to one marked by restricted or entrenched logi stical mobility (Amick and Carr 1996:15; Sassaman et al. 1988:90). Settlement models deve loped to address site distributions in the interior Coastal Plain have tended to place a great deal of emphasis on riverine resources, such as floodplains and more specifically edge swamps and sloughs that were exposed to seasonal flooding (Stoltman 1974; Turnbaugh 1975). The utility and accuracy of the Middle Coastal Plain Flood Plain Swamp Model, developed in the 1970s to address the patterning of Late Archaic archaeological si tes has been upheld by subsequent surveys and excavations at several locales in the in the Coastal Plain (Elliott and Sassaman 1995:142). A shortcoming of these models is the l ack of emphasis they place on the potential that hunter-gatherers had significantly altered the distribution of food resources that were important to them. There is also a lack of attention paid to upland locales and their susceptibility to fire, which were likely areas of exploitation by hunter-gatherers using fire, following Lewis and Ferguson (1999) a nd Sassaman (1988). Similar upland areas in the Southeast are now receiving greater at tention for their wetlands (e.g., Brookes et al. 1996), though hunter-gatherer burnin g is not considered. It is now generally accepted 139

PAGE 140

that river-based settlement was augmented by movement away from rivers and into interriverine areas on a seasona l basis marked by periods of group aggregation in the lowlands (Anderson and Joseph 1988; Anderson 1996:170; Goodyear et al.1979; Sassaman et al. 1990:311). Giving greater consideration to activities such as an thropogenic burning augments preexisting models of land use. For example, regular burning in upland areas, especially those interspersed with ponds and wetla nds, would have created and preserved a patchwork of ecotones (Denevan 1992; Lewis and Ferguson 1999). Empirical evidence to fully test this hypothesis in the project ar ea is lacking due to a variety of factors. However, it has been possible to draw some conclusions and develop additional research questions based on artifact and site patte rning that can be additionally explored, augmented, and tested in the course of future research endeavors. By manipulating the distribution of economic resources people would have enhanced the predictability of the resources in some areas. This would also be reflected in the types of artifacts found in particular localities, especially in relation to forest cover. Following Balee and Erickson (2006), the la ndscape must be considered to be a reflection of the cultures that interacted with it. From this perspective the distribution and sustainability of the l ongleaf pine-wiregrass ecosystem may potentially be thought of as being anthropogenic in nature. Thinking of the interior Coastal Plain as a managed landscape (Denevan 1992) may help account for th e distribution of artifacts recovered at Late Archaic sites in the region. For example, rather than just thinking of upland areas as a relative no mans lands due to their distance to fl owing water, they may be 140

PAGE 141

reconsidered following a combination of Le wis and Fergusons (1999) fire corridors and Head et al. (2002) concept of garden. It is likely that large porti ons of the uplands were fi red wholesale in a similar fashion to Lewis and Fergusons (1999) fire corridor. This c ould have been easily accomplished by taking advantage of natural top ographic features such as ridges. The already dry uplands with their sandy soils would have been prime for burning. However, upland ponds and wetlands would likely have needed special attention following Head et al. (2001) garden concept. This would pr obably have stemmed from differing types of vegetation growing in direct proximity to the wetlands. For example, hardwood tree species would likely have grown direc tly around pond and wetland areas given their greater access to groundwater. These hardwoods would have stood in stark contrast to the surrounding longleaf pine forest though they would have contribut ed to higher levels of biodiversity in the imme diate areas around the ponds and wetlands. While a few large hardwood trees would certainly be acceptable, thickets of small trees would not. The combination of increased surface soil mois ture and hardwood leaf litter that would otherwise be non-conducive to burning would ensure that pond and wetland edges received special attention to make sure flam es adequately spread into those areas and burned the vegetation evenly. Following Head et al. (2002), wetlands therefore likely acted as individual patches and represent domiculture based on their distribution within the broader sphere of human/environmental in teraction or domus. They occur as discrete and distinctive physical localities th at are patchily distribu ted within a wider area (country) as a result of geology and they also correspond with nodes of intensive 141

PAGE 142

human labor (garden) that are further in corporated into the understanding of the broader landscape (H ead et al. 2002:182). From a material perspective, these sorts of landscape-level activities should be reflected in the types of artifac ts recovered from these areas. With regard to the interior Coastal Plain, upland areas were likely main tained as hunting areas given their high levels of biodiversity which would be e nhanced by burning. Based on this assumption both whole and broken finished stone tools used for hunting and/or butchering along with debitage associated with their maintenan ce would be expected in upland locales. Following this line, flakes should also be small since they would be produced by the maintenance rather than the production of stone tools. Following Ingold (1992), there may also have been unintended consequences in that some areas like the uplands may have been rendered useless for long-term se ttlement due to regular burning. Longer-term habitation sites may have been placed in locatio ns that were not as susceptible to fire. They will also display a wider variety of ar tifacts which stand as material evidence for the greater number of activitie s that were undertaken there. In the long-term, the utility of anthropogeni c fire in the interi or Coastal Plain will have to be gauged against the reality of lightning-induced fires. It has long been held that the longleaf pine-wiregrass ecosystem was maintained by natural lightning fires. However, were those types of fires frequent e nough to ensure the stab ility of the system? Was the natural cycle consistent enough to bur n large swaths of fore st? Climate can be highly unpredictable. In the case of longleaf pi ne forests, they need to be subjected to low-intensity fires every one to three years to be sustained. If they do not burn regularly 142

PAGE 143

various hardwood species, especially oaks, ga in a foothold. With time, they will outcompete the slower-growing pine for nutrien ts and especially access to sunlight. People compensated for this unpredictability by taking matters into their own hands. That the forest existed at all in the extent that it did at c ontact likely stands as evidence for its long term inte raction with people. Humans leave a pronounced footprint on the landscape (Crumley and Marquard t 1990; Crumley 1994; Kirch and Hunt 1997; Lansing 1991). The primary issue here is that hunter-gatherers inte ract and domesticate the landscape at an immense scale. Their connection with it is more significant than simply wandering through it in search of food. Taking a long-term perspective on the ro le of fire in culture change, two alternative scenarios warrant consideration. First, fire may have been one among many responses to economic shortfalls that we re precipitated by decreased mobility, higher population density, and increased settlement permanence or redundancy. This is the ageold problem of economic intens ification that structures many arguments about culture change in anthropology. In the case of th e Dougherty Plain, the ar gument is that Late Archaic groups made increasingly permanent a nd redundant use of prime real estate, in this case, terraces and floodplains of major st ream channels, and especially locations of stream confluence. Expansion into upland areas, where surface water was available, is one among many solutions to demographic crow ding and diminished economic returns in bottomland locales. At locations of upla nd settlement groups would have eventually reached the point of diminishing returns and either relocated to other locales in the uplands, or, when relocation was not an opti on, make adjustments to enhance or expand 143

PAGE 144

the resource base of the fixed area surrounding encampments. Fire, in this scenario, is one of several strategies that could ha ve been employed to meet this demand. Alternatively, the use of fire in upland locales by river-based groups experiencing economic and demographic stress could have pre-adapted the upland biome for subsequent relocation of some or all of the riverine groups to the uplands. Either scenario implies that group dispersa l into the uplands was an alternative to the usual river-based settlement of the Late Archaic period. The question remains: were groups pulled into upland habitat or ra ther pushed there by economic stresses stemming from overexploitation of bottomland locations? We can add to this another questions deduce from a political economic pers pective. That is, were movements into the uplands precipitated by intergroup conflicts or some sort of stress of a political nature? In this regard, the uplands provided refuge, and were structured in a way that disallowed the formation of large social aggregates. Unlike the large floodplain and terrace locales along major str eams, landforms associated with upland wetlands are smaller and more dispersed, encouraging gr eater local group autonomy, and placing a cap on demographic expansion, at least locally. The use of fire would have enabled this sort of realignment to occur with a high level of success. Throughout the greater So utheast Coastal Plain, upland dispersal was a common strategy of settlement beginning in the late Late Archaic pe riod (after ca. 3500 BP), and continuing into the Early Woodland period. Insofar as fire was among the tools people used to enable this widespread change we should consider more seriously its role in the emergence of the so-called Easter n Horticultural Complex (Gremillion 1996). Indeed, the appearance of native domesticates after ca. 3000 BP is now tied to use of 144

PAGE 145

upland habitat in the lower Midwest and Midsout h. Fire likely played a role in this development, either directly in the sense of augmenting habitat to improve the conditions of domestication, or more generally in c ontributing to a cultural disposition for manipulating, or domesticating nature. In Neolithic Europe, the ideological shift from viewing nature as independent of human action to one that recognizes and natura lizes human actions to alter environments is a process Hodder (1990) refe rs to the domestication of Europe. We do not have sufficient evidence from the Southeast U.S. to determine if a similar shift occurred in the transition from hunting and gathering to farmin g. Full-scale agriculture in the Southeast would not appear until about 1000 BP., 2000 years af ter the Late Archaic period. Still, it is worth considering how the deliberate use of fire to alter vegetation and landscape may have predisposed native pe oples in the Southeast to the adoption of farming many centuries later. Finally, the problem of equifinality loom s large. In studies worldwide, the association between evidence for fire and human settlement is robust. However, arguments for causation are neve r conclusive. That is, we simply do not have convincing evidence in virtually any claim of anthr opogenic burning that huma ns did not simply respond to natural-set fires. One could easily imagine that fires set naturally were readily observable by ancient people, even at great di stances. Knowing that the consequences of fires one, two, or more years hence were bene ficial, humans could have simply scheduled their settlement moves in accordance with natu ral fire. This assumes, of course, that humans had unfettered access to territory, which is generally not the case. To attribute all 145

PAGE 146

human use of burned areas to the rhythms of nature alone requires that we fully document those rhythms and the consequences of utility for humans. The need for better, more thorough data on lightning-set fires has already been noted. Other needs for future research cente r on direct evidence for fire and direct associations between fire and human settlement Site-specific data are lacking. Sediment cores from upland wetlands in the Dougherty Plain are needed to document both the incidence of fire from charcoal and the conse quences of fire from pollen. Efforts to date have not been fruitful, largely because so many of the upland wetla nds were subject to seasonal drying, or have been destroyed by modern land alterations. The greatest potential for finding preser ved sediment records is in the wetlands of the Chickasawhatchee area (i.e., the north tract). Compared to other areas of the Dougherty Plain, this tract has more and better watere d ponds, and thus greater potential to locate deposits that have not dried out or been completely destroyed in the modern era. Ultimately, one could assemble a convincing argument for anthropogenic burning one site at a time. However, this strategy alone is not likely to illustrate the scope and significance of native uses of fire if indeed this was a regional, landscapescale strategy that truly structured the world in which they lived. This deeper historical approach to the problem benefits from a wider array of data sets and scales of analysis that are both temporally and spatially expansive. This disse rtation is an initial st ep in that direction and hopefully will inspire others to carry this line of research forward to demonstrate that even hunter-gatherers, with their simple tec hnology and mobile lifest yle, left large and lasting footprints on the la ndscape of the Southeast. 146

PAGE 147

LIST OF REFERENCES Amick, Daniel, and Philip Carr 1996 Changing Strategies of Technological Organization. In Archaeology of the Mid-Holocene Southeast edited by K.E. Sassaman and D.G. Anderson, pp. 41-56. University Press of Florida, Gainesville. Anderson, David. G. 1991 Examining Prehistoric Settlement Distribu tion in Eastern North America. Archaeology of Eastern North America 19:1. 1996 Approaches to Modeling Regional Settlement in the Archaic Period. In Archaeology of the Mid-Holocene Southeast edited by K. E. Sassaman and D. G. Anderson, pp. 157-176. University Press of Florida, Gainesville. Anderson, David. G., and Glen. T. Hanson 1988 Early Archaic Settlement in the S outheastern United States: A Case Study from the Savannah River. American Antiquity 53:262-286. Anderson, David. G., and J. W. Joseph 1988 Prehistory and History along the Uppe r Savannah River: Technical Synthesis of Cultural Resource Investigations Interagency Archaeologi cal Services Division, National Park Service. Anderson, M. Kat 2006 Tending the Wild: Native American K nowledge and the Management of Californias Natural Resources University of Californi a Press, Berkeley and Los Angeles. Anderson, R. S. 1979 A Holocene Record of Vegetation and Fire at Upper South Branch Pond in Northern Maine. M.S. Thesis, University of Maine. Andrefsky, William 1998 Lithics: Macroscopic Approaches to Analysis Cambridge University Press, Cambridge. Balee, William (editor) 1998 Advances in Historical Ecology Columbia University Press, New York. Balee, William, and Clark Erickson (editors) 2006 Time and Complexity in Historical Ecology: Studies in the Neotropical Lowlands Columbia University Press, New York. 147

PAGE 148

Bense, Judith A. 1994 Archaeology of the Southeastern United States: Paleoindian to World War II Academic Press, Orlando. Benson, Robert. W. 1995 Quartz Passports of the Morrow Mount ain Middle Archaic in the Savannah River Drainage. Early Georgia 23(1):1-19. Biersack, Aletta 1999 From the New Ecology to the New Ecologies. American Anthropologist 101:5-18. Binford, Lewis R. 1977 Forty-seven Trips: A Case Study in the Character of Archaeological Assemblage Formation. In Tools as Cultural Marke rs: Change, Evolution, and Complexity edited by R.V.S. Wright, pp. 24-36. Australian Institute of Aboriginal Studies, Canberra. 1978 Nunamiut Ethnoarchaeology Academic Press, New York. 1979 Organization and Formation Processe s: Looking at Curated Technologies. Journal of Anthropological Research 35:255-73. 1980 Willow Smoke and Dogs' Tails: Hunt er-Gatherer Settlement Systems and Archaeological Site Formation. American Antiquity 45:4-20. Binford, Lewis R., and S.R. Binford 1966 A Preliminary Analysis of Functional Variability in the Mousterian of Levallois Facies. In Recent Studies of Paleoanthropology edited by J.D. Clark and F.C. Howell, pp. 238-95. American An thropologist 68:Part 2. Blanton, Dennis. B., and Kenneth E. Sassaman 1989 Pattern and Process in the Middl e Archaic of South Carolina. In Studies in South Carolina Archaeology in Honor of Robert L. Stephenson edited by A. G. Goodyear and G. T. Hanson, pp. 53-72. Anth ropological Studies 9. South Carolina Institute of Archaeology and Anthropology, University of South Carolina, Columbia. Bonnicksen, T. M. 2000 America's Ancient Forests: From the Ice Age to the Age of Discovery John Wiley & Sons, New York. Bordes, Francois 1961 Mousterian Cultures in France. Science 134:803-810. 148

PAGE 149

Bowden, M. J. 1992 The Invention of American Tradition. Journal of Historical Geography 18:326. Boyd, Mark 1958 Historic Sites in and Adjacent to the Jim Woodruff Reservoir, FloridaGeorgia. Bureau of American Ethnology Bulletin 169, pp. 195-314. Smithsonian Institution, Washington D.C. Bromley, S. W. 1935 The Original Forest Type of Southern New England. Ecological Monographs 5(61-89). Brooks, Mark J., and Valetta Canouts 1984 Modeling Subsistence Change in the Late Prehistoric Period in the Interior Lower Coastal Plain of South Carolina Anthropological Studies 6, Occasional Papers of the South Carolina Institu te of Archaeology and Anthropology, University of South Carolina, Columbia. Brooks, Mark J., Barbara E. Taylor and J. A. Grant 1996 Carolina Bay Geoarchaeology and Holocene Landscape Evolution on the Upper Coastal Plain of South Carolina. Geoarchaeology 11(6):481-504. Brooks, Mark J., Barbara E. Taylor, Peter A. Stone, and Leonard R. Gardner 2001 Pleistocene Encroachment of the Wateree River Sand Sheet into Big Bay on the Middle Coastal Plai n of South Carolina. Southeastern Geology 30:241-257. Broyles, Bettye J. 1966 Second Preliminary Report: The St. Albans Site, Kanawha County, West Virginia West Virginia Geological and Economic Survey, Charleston. Bullen, Ripley P. 1950 An Archaeological Survey of the Ch attahoochee River Valley in Florida. Journal of the Washington Academy of Sciences 40:101-125. 1958 The Bolen Bluff Site on Paynes Prairie, Florida. Social Sciences Contributions 4. Florida State Museum, Gainesville. 1975 A Guide to the Identification of Florida Projectile Points Kendall Books, Gainesville. Bullen, Ripley P., and Edward M. Dolan 1959 The Johnson Lake Site, Marion County, Florida. The Florida Anthropologist 12:77-94. 149

PAGE 150

Bullen, Ripley P., Otto Jahn, and Mark J. Brooks 1974 Some Tests at the Zellwood Site (O r-17) Beside Lake Apopka, Florida. Florida Anthropologist 27:62-76. Caldwell, J. R. 1958 Trend and Tradition in the Prehistory of the Eastern United States. Memoir 88. The Illinois State Museum & Amer ican Anthropologica l Association, Springfield. 1978 Report of the Excavations at Fa irchilds Landing and Hares Landing, Seminole County, Georgia. Submitted to Southeast Archaeological Center, Tallahassee. Canouts, Velleta, and Albert C. Gooyear, III 1985 Lithic Scatters in the South Carolina Piedmont. In Structure and Process in Southeastern Archaeology, edited by R.S. Dickens, Jr., and H.T. Ward, pp. 180194. University of Alabama Press, Tuscaloosa. Chamblee, John F. 2006 Landscape Patches, Macroregional Exchanges, and Precolumbian Political Economy in Southwest Georgia Ph.D. dissertation, Department of Anthropology, University of Arizona, Tucson. Chapman, Jefferson. 1976 The Archaic Period in the Lower Litt le Tennessee Valley: The Radiocarbon Dates. Tennessee Anthropologist 1(1):1-12. Chapman, J., P. A. Delcourt, P. A. Cr idlebaugh, A. B. Shea and H. R. Delcourt 1982 Man-Land Interaction: 10,000 Years of American Indian Impact on Native Ecosystems in the Lower Little Tenn essee River Valley, Eastern Tennessee. Southeastern Archaeology 1(2):115-121. Chase, A. 1989 Domestication and Domicult ure in Northern Australia: A Social Perspective. In Foraging and Farming: The Evolution of Plant Exploitation edited by D.R. Harris and G.C. Hillman, pp. 42. Unwin Hyman, London. Christensen, N. L. 1987 The Biogeochemical Consequences of Fire and their Effects on the Vegetation of the Coastal Plain of the Southeastern United States. In The Role of Fire in Ecological Systems edited by L. Trabaud, pp. 1. SPB Academic Publishing, The Hague. Claflin, William H.J. 1931 The Stallings Island Mound, Columbia County, Georgia Peabody Museum of American Archaeology and Ethnology Papers 14(1). 150

PAGE 151

Claassen, Cheryl 1996 A Consideration of the Social Organi zation of the Shell Mound Archaic. In Archaeology of the Mid-Holocene Southeast edited by K.E. Sassaman and D.G. Anderson, pp. 235-258. University Press of Florida, Gainesville. Clark, James S. 1988 Stratigraphic Charcoal Analysis on Pe trographic Thin Sections: Application to Fire History in Northwestern Minnesota. Quarternary Research 30:81-91. Coe, Joffre. L. 1964 Formative Cultures of the Carolina Piedmont The American Philosophical Society, Philadelphia. Cridlebaugh, Patricia 1984 American Indian and Euro-American Impac t upon Holocene Vegetation in the Lower Little Tennessee River Valley, East Tennessee Unpublished Ph.D. dissertation, University of Tennessee, Knoxville. Cronon, William 1983 Changes in the Land Norton, New York. Cronon, William (editor) 1996 Uncommon Ground: Rethinking the Human Place in Nature Norton, New York. Crumley, C. L. (editor) 1994 Historical Ecology: Cultural Knowledge and Changing Landscapes School of American Research Press, Santa Fe. Crumley, C. L., and W. H. Marquardt 1990 Landscape: A Unifying Concep t in Regional Analysis. In Interpreting Space: GIS and Archaeology edited by K. M. S. Allen, S. W. Green and E. B. W. Zubrow. Taylor & Francis, London. Custer, Jay F., and David C. Bachman 1986 Prehistoric Use of Bay/Basin Feat ures on the Delmarva Peninsula. Southeastern Archaeology 5:1-10. Czuhai E., and C. T. Cushwa 1968 A Resume of Prescribed Burnings on the Piedmont National Wildlife Refuge. U.S. Forest Service Note SE-86. Davis, M. B. 1983 Holocene Vegetational History of the Eastern United States. In Late Quaternary Environments of the United States edited by J. H.E. Wright, pp. 166181. University of Minnesota Press, Minneapolis. 151

PAGE 152

Delcourt, Paul. A. 1980 Goshen Springs: Late Quaternary Vegetation Record For Southern Alabama. Ecology 6(2):371-386. Delcourt, P. A., and H. R. Delcourt 1981 Vegetation Maps for Eastern North America. In Geobotany II edited by R. C. Romans, pp. 123-165. Plenum Press, New York. 2004 Prehistoric Native Americans and Ecol ogical Change: Human Ecosystems in Eastern North America since the Pleistocene Cambridge University Press, Cambridge. Delcourt, P. A., H. R. Delcourt, C. R. Ison, W. E. Sharp, and K. J. Gremillion 1998 Prehistoric Human Use of Fire, the Eastern Agricultural Complex, and Appalachian Oak-Chestnut Forests: Pale oecology of Cliff Palace Pond, Kentucky. American Antiquity 63:263-78. Denevan, William. M. 1992 The Pristine Myth: The Landscape of the Americas in 1492. Annals of the Association of American Geographers 82(3):369-385. Dorney, Cheryl H., and John R. Dorney 1989 An Unusual Oak Savanna in Northeas tern Wisconsin: the Effect of IndianCaused Fire. American Midland Naturalist 122:103-113. Earley, Lawrence S. 2004 Looking for Longleaf: The Fall and Rise of an American Forest. University of North Carolina Press, Chapel Hill. Elliott, Dan T. 2004 Southwest Georgia Archaeological Survey LAMAR Institute. LAMAR Institute Publication 60. Elliott, Daniel T., and Tracy M. Dean 2006 Navigable Waterways Project, Phas e I: Upper Chattahoochee and Toccoa Rivers. Lamar Institute Report No. 97. Elliott, Daniel. T., and Kenneth. E. Sassaman 1995 Archaic Period Archaeology of The Georgia Coastal Plain and Coastal Zone Georgia Archaeological Res earch Design Papers 11. Re port 35. University of Georgia Laboratory of Archaeology, Athens. Fairbanks, Charles H. 1940 Mound C, Macon Group. Newsletter of the South eastern Archaeological Conference 2:4. 152

PAGE 153

Faught, Michael K., and Brinnen Carter 1998 Early Human Occupation and Environmental Change in Northwestern Florida. In As the World Warmed: Human Adaptations Across the PleistoceneHolocene Boundary, edited by B. Eriksen and L.G. Straus. Quaternary International 49/50:167-176. Fish, Paul R., and Suzanne K. Fish 1976 Patterns of Prehistoric Site Distribution in Effingham and Screven Counties, Georgia University of Georgia Laboratory of Archaeology Series Report Number 11. Department of Anthropology, Un iversity of Georgia, Athens. 1977 Prehistoric Settlement in the Dry Creek Watershed. University of Georgia Laboratory of Archaeology Series Report Number 14. Department of Anthropology, University of Georgia, Athens. Forman, R. T. T. 1979 Pine Barrens: Ecosystem and Landscape Academic Press, New York. Forman, R. T. T., and E. W. B. Russell 1983 Evaluation of Histor ical Data in Ecology. Bulletin of the Ecological Society of America 64:5-7. Gaiser, E. E., B. E. Taylor, and M. J. Brooks 2001 Establishment of Wetlands on the So utheastern Atlantic Coastal Plain: Paleolimnological Evidence of a Mid-Holo cene Hydrologic Threshold from a South Carolina Pond. Journal of Paleolimnology 26:373-391. Gardner, Paul 1994 Carbonized Plant Remains from Dust Cave. Journal of Alabama Archaeology 40:189-207. 1997 Ecological structure and be havioral implications of mast exploitation. In People, Plants, and Landscapes: Studies in Paleoethnobotany edited by K.J. Gremillion, pp. 161. University of Alabama Press, Tuscaloosa. Gardner, William. M. 1978 Comparison of Ridge and Valley, Blue Ridge, Piedmont, and Coastal Plain Archaic Period Distribution: An Ideali zed Transect. Unpublished manuscript in possession of the author. Gibson, Jon L., and Philip Carr (editors) 2004 Signs of Power: The Rise of Cu ltural Complexity in the Southeast University of Alabama Press, Tuscaloosa. 153

PAGE 154

Goebel, P. C., B. J. Palik, and L. K. Kirkman 1997 Landscape Ecosystem Types of Ichauway Technical Report 97. Joseph W. Jones Ecological Research Center, Newton, Georgia, USA. Goodyear, Albert. C., III, J. H. House, N. W. Ackerly 1979 Laurens-Anderson: An Archaeological Sur vey of the Inter-Riverine Piedmont Anthropological Studies 4. South Ca rolina Institute of Archaeology and Anthropology, University of South Carolina, Columbia. Gould, Richard 1978 The Anthropology of Human Residues. American Anthropologist 80:815-35. Gresham, Thomas. M., R. Jerald Ledbetter, Robbie F. Ethridge and T. J. Price 1989 Archaeological Investigations of th e Mill Creek Site, Americus, Georgia. Southeastern Archaeological Services, Athens, Georgia. Griffin, J. W. 1974 Investigations in Russell Cave. Ru ssell Cave National Monument, Alabama National Park Service, Atlanta. Grinker, Roy Richard 1994 Houses in the Rainforest: Ethnic ity and Inequality among Farmers and Foragers in Central Africa University of Californi a Press, Berkeley and Los Angeles. Hainds, Mark J., Robert J. Mitchell, Brian J. Palik, Linds ay R. Boring, and Dean H. Gjerstad 1999 Distribution of Native Legumes (L eguminoseae) in Frequently Burned Longleaf Pine (Pinaceae)Wiregrass (Poaceae) Ecosystems. American Journal of Botany 1999:1606-1614. Hammett, J. E. 1992 Ethnohistory of Aboriginal Landscape s in the Southeastern United States. Southern Indian Studies 41:1-50. Head, Lesley 1996 Rethinking the Prehistory of Hunter-G atherers, Fire and Vegetation Change in Northern Australia. The Holocene 6:481-7. Head, Lesley, Jennifer Atch ison, and Richard Fullagar 2002 Country and Garden: Ethnobotany, Archaeobotany and Aboriginal Landscapes Near the Keep River, Northwestern Australia. Journal of Social Archaeology 2:173-190. 154

PAGE 155

Heckenberger, Michael J., Afukaka Kuikuro, Urissap Tabata Kuikuro, J. Christian Russell, Morgan Schmidt, Carlos Fausto and Bruno Franchetto 2003 Amazonia 1492: Pristine Forest or Cultural Parkland. Science 301:1710-1713. Hodder, Ian 1990 The Domestication of Europe: Structure and Contingency in Neolithic Societies. Blackwell, Oxford Hoyt, John H. and Hails, J.R. 1967 Pleistocene Shoreline Sediments in Coastal Georgia: Deposition and Modification. Science 155:1541-1543. House, John H., and David L. Ballenger 1976 An Archaeological Survey of Inte rstate 77 Route in the South Carolina Piedmont. Research Manuscript Series 104. South Carolina Institute of Archaeology and Anthropology, University of South Carolina, Columbia. Huffman, Jean Marie 2006 Historical Fire Regimes in Southeas tern Pine Savannas. Unpublished Ph.D. dissertation, Louisiana Stat e University, Baton Rouge. Ingold, Tim 1992 Culture and the Perception of the Environment. In Bush Base: Forest Farm. Culture, Environment and Development edited by E. Croll and D. Parkin. Routledge, London. 2000 The Perception of the Environment: Essays on Livelihood, Dwelling, and Skill Routledge, London. Jefferies, Richard W. 1996 The Emergence of Long-Distance Exchange Networks in the Southeastern United States. In Archaeology of the Mi d-Holocene Southeast edited by Kenneth Sassaman and David Anderson, pp. 222-34. University Press of Florida, Gainesville. Jennings, Jesse 1968 Prehistory of North America McGraw-Hill, New York. Johnson, Jay K. 1996 Lithic Analysis and Questions of Cultural Complexity: The Maya. In Stone Tools: Theoretical Insights into Human Prehistory edited by G. Odell, pp. 159180. Plenum, New York. Johnson, Jay. K., and Samuel O. Brookes 1989 Benton Points, Turkey Tails, and Cache Blades: Middle Archaic Exchange in the Midsouth. Southeastern Archaeology 8(2):1989. 155

PAGE 156

Kalisz, Paul J., Alan W. Dorian, and Earl L. Stone 1986 Prehistoric Land-Use and the Distri bution of Longleaf Pine on the Ocala National Forest, Florida: An In terdisciplinary Synthesis. The Florida Anthropologist 39:183-193. Kelly, Robert L. 1983 Hunter-Gatherer Mobility Strategies. Journal of Anthr opological Research 39:277-306. 1988 The Three Sides of a Biface. American Antiquity 53:717-34. Kelly, R. L., and L. C. Todd 1988 Coming into the Country: Early Paleoindian Hunting and Mobility. American Antiquity 53:231-44. Kirch, Patrick V., and T. L. Hunt (editors) 1997 Historical Ecology in the Pacific Islands Yale University Press, New Haven. Kirkland, D. S. 1994 Ten Millennia of Human Landscape Us e on the Interior Coastal Plain of Georgia Unpublished M.A. Thesis, University of Georgia. Klein, D.R. 1970 Food Selection by North American Deer and Their Response to OverUtilization of Preferred Plant Species. Symp. of British Ecol. Soc. 10:25-44. Kowaleski, Steven. A. 1995 Large-Scale Ecology in Aborigin al Eastern North America. In Native American Interactions: Multiscalar Analys es and Interpretations in the Eastern Woodlands, edited by M. S. Nassaney and K. E. Sassaman, pp. 147-173. University of Tennessee Press, Knoxville. Kottak, Conrad 1999 The New Ecological Anthropology. American Anthropologist 101:23-35. Krech, Shepard, III 1999 The Ecological Indian W.W. Norton & Co mpany, New York. Kurtz, Donald V. 1994 Winnowing the Great Kalahari Deba te; Its Impact on Hunter-Gatherer Studies: A Review of Current Literature. Political and Legal Anthropology Review 17:67-80. 156

PAGE 157

LaLande, Jeff, and Reg Pullen 1999 Burning for a Fine and Beautiful Op en Country: Native Uses of Fire in Southwestern Oregon. In Indians, Fire and the Land in the Pacific Northwest edited by R. Boyd, pp. 255-276. Oregon State University Press, Corvallis, Oregon. Lamoreaux, H. J. K. 1999 Human-Environmental Relationships in the Coastal Plain of Georgia Based on High-Resolution Paleoenvironmental Records from Three Peat Deposits Ph.D. Dissertation, University of Georgia, Athens. Lansing, J. Stephen 1991 Priests and Programmers: Technolog ies of Power in the Engineered Landscape of Bali. Princeton University Press, Princeton, NJ. Larson, Lewis 1980 Aboriginal Subsistence Technology on the Southeastern Coastal Plain during the Late Prehistoric Period University Press of Florida, Gainesville. Lay, Daniel W. 1956 Effects of Prescribed Burning on Fo rage and Mast Production in Southern Pine Forests. Journal of Forestry 54:582-584. Leacock, Eleanor, and Richard Lee 1982 Politics and History in Band Societies Cambridge University Press, Cambridge. Lewis, Henry T. 1972 The Role of Fire in the Domestication of Plants and Animals in Southwest Asia: A Hypothesis. Man 7:195-222. 1977 Maskuta: The Ecology of Indian Fires in Northern Alberta. Current Anthropology 7:15-52. 1982 Fire Technology and Resource Manageme nt in Aboriginal North America and Australia. In Resource Managers: North American and Australian HunterGatherers, edited by N.M. Williams and E. S. Hunn, pp. 45-67. vol. AAAS Selected Symposia 67. Westview Press, Boulder, Colorado. Lewis, Henry T., and Theresa A. Ferguson 1999 Yards, Corridors, and Mosaics: How to Burn a Boreal Forest. In Indians, Fire and the Land in the Pacific Northwest edited by R. Boyd, pp. 164-184. Oregon State University Press, Corvallis, Oregon. Marquardt, William H. 1985 Complexity and Scale in the Study of Fisher-Gatherer-Hunters: An Example from the Eastern United States. In Prehistoric Hunter-Gatherers: The Emergence 157

PAGE 158

of Cultural Complexity edited by T.D. Price and J.A. Brown, pp. 58-98. Academic Press, Orlando. Marquardt, W. H. and C. L. Crumley 1987 Theoretical Issues in the An alysis of Spatial Patterning. In Regional Dynamics: Burgundian Landscapes in Historical Perspective edited by C. L. Crumley and W. H. Marquardt, pp. xx-xx. Academic Press, San Diego. McGlade, James 1995 Archaeology and the Ecodynamics of Human-modified Landscapes. Antiquity 69:113-133. Meggers, Betty 1966 Ecuador. Thames and Hudson, London. Milanich, Jerald T. 1994 Archaeology of Precolumbian Florida University Press of Florida. Gainesville. Miller, Gifford H., John W. Magee, Beverl y J. Johnson, Marilyn L. Fogel, Nigel A. Spooner, Malcolm T. McCulloch, Linda K. Ayliffe 1999 Pleistocene Extinction of Genyornis newtoni: Human Impact on Australian Megafauna. Science 283:205-208. Murray, Grover E. 1961 Geology of the Atlantic and Gulf Coasta l Province of North America. Harper, New York. Myers, Fred 1988 Critical Trends in the Study of Hunter-Gatherers. Review of Anthropology 17:261-282. Neill, Wilfred T. 1964 Trilisa Pond, An Early Site in Marion County, Florida. Florida Anthropologist 17:187-200. Netting, R. Mc. 1986 Cultural Ecology (2nd ed.) Waveland Press, Prospect Heights, Ilinois. Odell, George 2004 Lithic Analysis Plenum, New York. Patterson, W. A., III, and K. E. Sassaman 1988 Indian Fires in the Preh istory of New England. In Holocene Human Ecology in Northeastern North America edited by G. P. Nicholas, pp. 107-135. Plenum, New York. 158

PAGE 159

Pluckhahn, Thomas 2003 Kolomoki: Settlement, Ceremony, and Status in the Deep South, A.D. 350 to 750. University of Alabama Press, Tuscaloosa. Price, T. Douglas, and James A. Brown 1985 Aspects of Hunter-Gatherer Complexity. In Prehistoric Hunter-Gatherers: The Rise of Cultural Complexity edited by T.D. Price and J.A. Brown, pp. 3-20. Academic Press, New York.. Pyne, S. J. 1982 Fire in America: A Cultural History of Wildland and Rural Fire University of Washington Press, Seattle. 1995 World Fire: The Culture of Fire on Earth University of Washington Press, Seattle. Randall, Asa R. 2007 St. Johns Archaeological Field Sc hool 2005: Hontoon Island State Park. Technical Report 7, Laboratory of Sout heastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Rappaport, Roy 1968 Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People Yale University Press, New Haven. Ritchie, W. A. 1932 The Lamoka Lake Site. In Researches and Transactions of the New York State Archaeological Association pp. 79-134, New York. Robbins, L. E., and R. L. Myers 1992 Seasonal Effects of Prescribed Burning in Florida: A Review Tall Timbers Research, Inc., Tallahassee, FL, Miscellaneous Publication Number 8. Russell, E.W.B. 1983 Indian-set Fires in the Forests of the Northeastern United States. Ecology 64:78-88. Russo, Michael 2004 Measuring Shell Rings for Social Inequality. In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Phillip J. Carr, pp. 26-70. University of Alabama Press, Tuscaloosa. Sahlins, Marshall D. 1972 Stone Age Economics Aldine, Chicago. 159

PAGE 160

Sahlins, Marhall D., and Elman R. Service 1960 Evolution and Culture University of Michig an Press, Ann Arbor. Sassaman, Kenneth E. 1988 Mesolithic Fires in North-Central Eu rope. A Statement of Field submitted in partial fulfillment of Ph .D. program requirements, Department of Anthropology, University of Massachusetts, Amherst. 1991 Adaptive Flexibility in the Morrow Mountain Phase of the Middle Archaic Period. South Carolina Antiquities 23(1 & 2):31-41. 1993 Early Pottery in the Southeast: Tr adition and Innovation in Cooking Technology University of Alabama Press, Tuscaloosa. 2004 Complex Hunter-Gatherers in Evol ution and History: A North American Perspective. Journal of Archaeological Research 12:227-280. 2005 Poverty Point and Structure, Event, Process. Journal of Archaeological Method and Theory 12:335-364. Sassaman, K. E., and D. G. Anderson 1995 Middle and Late Archaic Archaeologic al Records of South Carolina. South Carolina Institute of Archaeology and Anthropology Savannah River Archaeological Research Papers 6. South Carolina Institute of Archaeology and Anthropology, University of South Caro lina, University of South Carolina, Columbia. Sassaman, K. E., M. J. Brooks, G. T. Hanson and D. G. Anderson 1990 Native American Prehistory of the Mi ddle Savannah River Valley: A Synthesis of Archaeological Investigations of the Savannah River Site, Aiken and Barnwell Counties, South Carolina Savannah River Archaeological Research Papers 1. South Carolina Institute of Archaeol ogy and Anthropology, University of South Carolina, University of South Carolina, Columbia. Sassaman, Kenneth. E., Glen. T. Hanson and Tommy Charles 1988 Raw Material Procurement and the Reduction of Hunter-Gatherer Range in the Savannah River Valley. Southeastern Archaeology 7(2):79-94. Sassaman, Kenneth E., Meggan E. Blessing, and Asa R. Randall 2006 Stallings Island Revisited: New Ob servations on Occupational History, Community Patterning, and Subsistence Technology. American Antiquity 71:539566. Saunders, Joe W., E.T. Allen, and R.T. Saucier 1994 Four Archaic? Mound Comple xes in Northeast Louisiana. Southeastern Archaeology 13:134-152. 160

PAGE 161

Saunders, Joe W., R. D. Mandel, R. T. Saucier, E. T. Allen, C. T. Hallmark, J. K. Johnson, E. H. Jackson, C. M. Allen, G. L. Stringer, D. S. Frink, J. K. Feathers, S. Williams, K. J. Gremillion, M. F. Vidrine and R. Jones. 1997 A 5400-5000 B.P. Mound Complex in Louisiana. Science 277:1796-1799. Saunders, Rebecca, and Christopher Hays (editors) 2004 Early Pottery: Technology, Style, an d Interaction in the Lower Southeast University of Alabama Press, Tuscaloosa. Schiffer, Michael B. 1976 Behavioral Archaeology Academic Press, New York. Sears, William H. 1956 Excavations at Kolomoki: Final Report University of Georgia Press, Athens. Seielstad, Carl. A. 1994 Holocene Environmental History at Chatterton Springs on the Southern Coastal Plain of Georgia M.A. Thesis, University of Georgia, Athens. Service, Elman 1962 Primitive Social Organization Random House, New York. Shackleton, J. 1979 Paleoenvironmental Histories from Wh itefish and Imuruk Lakes, Seward Peninsula, Alaska M.Sc. Thesis, Ohio State University, Columbus. Shott, Michael 1986 Technological Organization and Se ttlement Mobility: An Ethnographic Examination. Journal of Anthropological Research 42:15-51. 1994 Size and Form in the Analysis of Flake Debris: Review and Recent Approaches. Journal of Archaeological Method and Theory 1:69-110. Smith, Bruce D. 1986 The Archaeology of the Southeastern United States: From Dalton to de Soto. In Advances in World Archaeology (vol. 5) edited by F. Wendorf and A. Close, pp. 1-92. Academic Press, Orlando. Snow, Frankie 1977 A Survey of the Ocmulgee Big Bend Region. Early Georgia 5(1&2):37-60. Steinen, Karl 1976 An Archaeological Survey of Early County, Georgia. Miscellaneous Project Report Series Number 2. Flor ida State Museum, Gainesville. 161

PAGE 162

Steinen, Karl T., and Thomas J. Crawford 1990 The Sonny Lee Site: Shifting Sands a nd Archaeological Site Interpretation on the Gulf Coastal Plain. Early Georgia 18:65-81. Steponaitis, Vincas 1986 Prehistoric Archaeology of the Southeastern United States, 1970-1985. Annual Review of Anthropology 15:363-404. Steward, Julian 1949 Cultural Causality and Law: A Trial Fo rmulation of the Development of Early Civilizations. American Anthropologist, New Series 51:1-27. Stewart, O. E., et al. 2002 Forgotten Fires: Native Americans and the Transient Wilderness University of Oklahoma Press, Norman. Stiles, Daniel 1992 The Hunter-Gatherer Revisionist Debate. Anthropology Today 8:13-17. Stoltman, James B. 1974 Groton Plantation: An Archaeological Study of a South Carolina Locality. Peabody Museum Monographs, No. 1, Cambridge. Sullivan, Alan P., and Kenneth C. Rozen 1985 Debitage Analysis and Ar chaeological Interpretation. American Antiquity 50:755-779. Thompson, Victor 2006 Questioning Complexity: The Prehisto ric Hunter-Gatherers of Sapelo Island, Georgia. Unpublished Ph.D. dissertation, Depa rtment of Anthropology, University of Kentucky, Lexington. Turnbaugh, William 1975 Toward an Explanation of the Broadpoint Dispersal in Eastern North American Prehistory. Journal of Anthropological Research 31:51-68. Turnbull, Colin 1965 Wayward Servants: The Two Worl ds of the African Pygmies Natural History Press, Garden City, NY. Vale, Thomas R. 2002 Fire, Native Peoples, an d the Natural Landscape Island Press, Washington. 162

PAGE 163

Waggoner, James. C. Jr. 2003 Gum Ponds and Cypress Swamps: Late Archaic Use of Upland Interriverine Resources in the Dougherty Plain of Southw est Georgia. Paper presented at the Southeastern Archaeological Conference, Charlotte, North Carolina. Wagner, Gail 2003 Eastern Woodlands Anthropogenic Ecology. In People and Plants in Ancient Eastern North America, edited by P.E. Minnis, pp. 126-171. Smithsonian, Washington. Ward, Traewick H.. and Stephen. Davis. Jr. 1999 Time Before History: The Archaeology of North Carolina. The University of North Carolina Press, Chapel Hill and London. Watts, W. A. 1971 Postglacial and Interg lacial Vegetation History of Southern Georgia and Central Florida. Ecology 52(676-690). 1971 Late-Quaternary Vegetation History at White Pond on the Inner Coastal Plain of South Carolina. Quaternary Research 13:187-199. Watts, W. A., E. C. Grimm and T. C. Hussey 1996 Mid-Holocene Forest History of Flor ida and the Coastal Plain of Georgia and South Carolina. In Archaeology of the Mi d-Holocene Southeast edited by K. E. Sassaman and D. G. Anderson, pp. 28-38. Univer sity Press of Florida, Gainseville. Webb, W. S. and D. L. DeJarnette 1942 An Archaeological Survey of Pickwick Ba sin in the Adjacent Portion of the states of Alabama, Mississippi, and Tennessee Smithsonian Institution, Bureau of American Ethnology, Washington, D.C. Weissner, Polly 1982 Beyond Willow Smoke and Dogs Tails : A Comment on Binfords Analysis of Hunter-Gatherer Settlement Systems. American Antiquity 47:171-178. Whatley, John 2002 An Overview of Georgia Projectile Points and Selected Cutting Tools. Early Georgia 30:7-133. Wharton, Charles H. 1978 The Natural Environments of Georgia Georgia Department of Natural Resources, Atlanta. 163

PAGE 164

Wharton, Charles H., Wiley M. Kitchens, Edward C. Pendleton, and Timothy W. Sipe 1982 The Ecology of Bottomland Hardwood Swamps of the Southeast: A Community Profile U. S. Department of the Interior, Fish and Wildlife Service, Biological Services Program, Washington, D.C. Whelan, R. J. 1995 The Ecology of Fire Cambridge University Press, Cambridge. Whitehead, R. 1965 Palynology and Pleistocene Phytogeography of Unglaciated Eastern North America. In The Quaternary of the United States edited by H. E. Wright and D. G. Frey. Princeton University Press, Princeton. Willey, Gordon. R. and Phillip Phillips 1958 Method and Theory in American Archaeology University of Chicago Press, Chicago. Wilsen, Edwin 1983 Land Filled with Flies University of Chicago Press, Chicago. Yellen, John E. 1977 Long Term Hunter-Gatherer Adap tation to Desert Environments: A Biogeographical Perspective. World Archaelogy 8:262-274. 164

PAGE 165

165 BIOGRAPHICAL SKETCH Although James Cowan Waggoner, Jr. was born on Staten Island, New York, (on July 9, 1971), he spent most of his life and archaeological career in Ge orgia. He received a B.A. in History from Middle Georgia College (now Georgia Colleg e and State University) in 1997, where he received credit for taking an thropology and archaeology classes at the University of Georgia. Jamie participated in his first archaeologica l field school in 1996, helping Mark Williams map and test the Lamar Site. After a couple of years doing Cultural Resource Management archaeology for Southern Research, Inc., Jamie enrolled in graduate studies at Florida State Un iversity in 1999. He received his M.A. in Anthropology from FSU in 2002 and entered th e University of Floridas Ph.D. program in the same year. In addition to his first fi eld school and his dissertation fieldwork along the Chickasawhatchee and Ichawaynochaway Cr eeks of southwestern Georgia, Jamie has participated in field projects in central Georgia, northeastern Ge orgia, Florida, and Mexico.