Remapping Histories

Permanent Link: http://ufdc.ufl.edu/UFE0041824/00001

Material Information

Title: Remapping Histories Archaic Period Community Construction along the Middle St. Johns River, Florida
Physical Description: 1 online resource (428 p.)
Language: english
Creator: Randall, Asa
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010


Subjects / Keywords: anthropology, archaeology, archaic, florida, gis, hunter, lidar, shell, shellfishing
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: This study examines the significance of hunter-gatherer built landscapes through an analysis of Mount Taylor period (ca. 7300 4600 cal B.P.) shell mounds along the Middle St. Johns River in northeast Florida. Mount Taylor communities are best known for the incipient exploitation of shellfish and other aquatic resources in the region. In Florida and elsewhere, shell mounds are routinely interpreted as refuse accumulations, while their repeated occupation is taken to represent long-term continuities enabled by abundant wetland resources. Through a historiography of local research, I show that this model of shell mound growth reflects widely held anthropological assumptions regarding hunter-gatherer social simplicity and stasis, and obscures evidence for change through time. I develop an alternative framework based on practice theory to detail how shell mounds and associated places emerged as a historical process in which communities inscribed and politicized social memories through the deposition of shellfish and other materials within landscapes. In order to problematize the contexts of shell mound inhabitation, I examine the paleohydrology of the Middle St. Johns River. I also consider the evidence for continuities in subsistence, settlement, exchange relationships, and mortuary traditions throughout the Mount Taylor period. The social and ecological contexts provide a spatial and chronological framework for examining how Mount Taylor communities inhabited places. Because most shell mounds were destroyed by twentieth century land-use practices, I use historic observations and modern remote sensing data to develop a geospatial database detailing the location and organization of Mount Taylor places. The results of topographic and stratigraphic testing of non-mounded shell sites are also reported. This analysis details how Mount Taylor communities established and renewed settlements as witnessed in small-scale depositional practices. Finally, I reconstruct the histories of six Mount Taylor shell mounds based on stratigraphic testing. The results demonstrate that shellfishing was initiated at a time of considerable landscape instability by arguably diverse regional populations. After being established as places to dwell, some preexisting settlements were reconfigured as platform mounds upon which shellfish was deposited in ritualized sequences, while others were converted into foundations for mortuary mounds. Through time, other shell mounds experienced complex histories of abandonment, renewal, and transformation as well. I argue that along the St. Johns, Mount Taylor communities routinely referenced past places in order to construct new social histories that accommodated or denied social and ecological change.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: 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.
Statement of Responsibility: by Asa Randall.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sassaman, Kenneth E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041824:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041824/00001

Material Information

Title: Remapping Histories Archaic Period Community Construction along the Middle St. Johns River, Florida
Physical Description: 1 online resource (428 p.)
Language: english
Creator: Randall, Asa
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010


Subjects / Keywords: anthropology, archaeology, archaic, florida, gis, hunter, lidar, shell, shellfishing
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Abstract: This study examines the significance of hunter-gatherer built landscapes through an analysis of Mount Taylor period (ca. 7300 4600 cal B.P.) shell mounds along the Middle St. Johns River in northeast Florida. Mount Taylor communities are best known for the incipient exploitation of shellfish and other aquatic resources in the region. In Florida and elsewhere, shell mounds are routinely interpreted as refuse accumulations, while their repeated occupation is taken to represent long-term continuities enabled by abundant wetland resources. Through a historiography of local research, I show that this model of shell mound growth reflects widely held anthropological assumptions regarding hunter-gatherer social simplicity and stasis, and obscures evidence for change through time. I develop an alternative framework based on practice theory to detail how shell mounds and associated places emerged as a historical process in which communities inscribed and politicized social memories through the deposition of shellfish and other materials within landscapes. In order to problematize the contexts of shell mound inhabitation, I examine the paleohydrology of the Middle St. Johns River. I also consider the evidence for continuities in subsistence, settlement, exchange relationships, and mortuary traditions throughout the Mount Taylor period. The social and ecological contexts provide a spatial and chronological framework for examining how Mount Taylor communities inhabited places. Because most shell mounds were destroyed by twentieth century land-use practices, I use historic observations and modern remote sensing data to develop a geospatial database detailing the location and organization of Mount Taylor places. The results of topographic and stratigraphic testing of non-mounded shell sites are also reported. This analysis details how Mount Taylor communities established and renewed settlements as witnessed in small-scale depositional practices. Finally, I reconstruct the histories of six Mount Taylor shell mounds based on stratigraphic testing. The results demonstrate that shellfishing was initiated at a time of considerable landscape instability by arguably diverse regional populations. After being established as places to dwell, some preexisting settlements were reconfigured as platform mounds upon which shellfish was deposited in ritualized sequences, while others were converted into foundations for mortuary mounds. Through time, other shell mounds experienced complex histories of abandonment, renewal, and transformation as well. I argue that along the St. Johns, Mount Taylor communities routinely referenced past places in order to construct new social histories that accommodated or denied social and ecological change.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: 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.
Statement of Responsibility: by Asa Randall.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Sassaman, Kenneth E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041824:00001

This item has the following downloads:

Full Text




2 2010 Asa Robert Randall


3 To Jane


4 ACKNOWLEDGMENTS This study has been a long time in the making, and would have been impossible without the support of many i ndividuals and institutions. My committee deserves many thanks and praise. Ken Sassaman has been a true mentor, colleague, and friend. I have benef itted from his intellectual and inst itutional generosity more than I can express. Susan Gillespie has been an important influence throughout my graduate career. She has always challenged me to th ink critically about the relevance of my research, and for that I am very grateful Augusto Oyuela-Caycedo provided a needed alternative viewpoint on hunter-gatherers, which was shared over many cups of coffee. Joann Mossa introduced me to paleohydrology and freely offered books, references, and advice that positively s haped the direction of this research. I must also acknowledge my fellow graduate students and archaeology colleagues, to whom I am indebted for their stimulating discussions, help in the field, and camaraderie. In particular, I thank Meggan Blessing, K andi Hollenbach, Kendall Campbell, Clete Rooney, Zack Gilmore, Jason ODonoughue, Neill Wallis, and the late Jaime Waggoner. The bulk of the field work for this proj ect was conducted through the University of Florida St. Johns Archaeological Field School directed by Ken Sassaman. I am very thankful for the opportunity to divert fiel d school and laboratory efforts for my own research. The field work was directly enabl ed by the generosity of Florida State Parks and the Juniper Club, LLC. Much needed fi nancial support for radiocarbon assays published in this study was provided by the Florida Archaeological Council, the University of Florida Department of Anth ropologys Charles H. Fairbanks Award, and the Laboratory of Southeastern Archaeology. I also thank Jack Eckert of Harvard University who graciously helped me navi gate Jeffries Wymans journals, and Arden


5 Fontaine of the Volusia County Department of Public Works who shared LiDAR data with me. Throughout this process my family and friends have been emotionally encouraging and financially enabling. I am deeply indebted to my parents, who instilled in me a love of knowledge and dirt at an early age, and have never onc e questioned my choice of career. My aunt Jane, to whom this work is dedicated, opened up her home a decade ago, and I cannot thank her enough for everyt hing she has done for me. My brother Josh is one of my most important resour ces, and inspires me daily. He has always urged me to step outside my focused world, and has kept me balanced and aware of my surroundings. My cousins Ben and Stoney and friend Pete, have reminded me that life is filled with adventure and humor if you allow it. Finally, Maura has been a transformative addition to my life. Her support has been unw avering, especially when I was filled with doubt, and I look forward to giving back all the patience and love she has provided to me.


6 TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES ............................................................................................................ 9LIST OF FI GURES ........................................................................................................ 10ABSTRACT ................................................................................................................... 14 CHAPTERS 1 SHELL MOUNDS AND OTHER NONEVE NTS ...................................................... 16Inconceivable Agencies al ong the St. Johns River ................................................. 26A Persistent Other .................................................................................................. 40Monumental Di scord ............................................................................................... 47Chapter Over view ................................................................................................... 562 HUNTER-GATHERERS, LANDSCAPES, AND HISTORIES .................................. 65Mapping .................................................................................................................. 67Mapping Away History ............................................................................................ 73Inhabitation as History ............................................................................................ 77Practice and Co mmunity .................................................................................. 79Landscape and Pl ace ....................................................................................... 87Biographies of Plac e ........................................................................................ 95Towards Archaic Histories ...................................................................................... 993 THE ECOLOGIES OF PRECERAMIC ARCHAIC COMMUNI TIES ...................... 104Geomorphic Context of the St. Johns River .......................................................... 108Hydrogeology ................................................................................................. 109Economic Geology ......................................................................................... 112Relative Sea Level ......................................................................................... 113Regional Physi ography................................................................................... 114Holocene History of the Middle St. J ohns ............................................................. 117Impacts of Historic Land Use .......................................................................... 117Climate ........................................................................................................... 120Hydrology ....................................................................................................... 126Channel and Floodplain Organizati on ............................................................ 133A Dynamic Land ................................................................................................... 1414 A SHORT GENEAOLOGY OF MOUNT TAYLOR TRADITIONS .......................... 159Amidst the Rising Waters ...................................................................................... 160


7 Mount Taylor Chronology ...................................................................................... 164Settlement M obility ............................................................................................... 167Subsistenc e .......................................................................................................... 172Technolog y ........................................................................................................... 176Exchange .............................................................................................................. 179Mortuaries and Biog raphies .................................................................................. 184In the Wake of t he Mound Buil ders ....................................................................... 188A Long Running Record? ...................................................................................... 1905 DISCERNING MOUNT T AYLOR PLAC ES ........................................................... 195Modeling Shell Matrix Site s: Data and Me thods ................................................... 199Archaeological Data ....................................................................................... 199Geospatial Da ta .............................................................................................. 201Methods .......................................................................................................... 204Shell Matrix Site Variati on ..................................................................................... 205Ridges ............................................................................................................ 207Multi-Mound Complexes ................................................................................. 212Shell Fiel ds ..................................................................................................... 221The Structure of Regional Inhabitat ion ................................................................. 2246 MOUNT TAYLOR IN MICROCOSM ..................................................................... 247History of Invest igations ........................................................................................ 249Site Organiza tion .................................................................................................. 253Node 1 .................................................................................................................. 259Node 2 .................................................................................................................. 262Node 3 .................................................................................................................. 265Node 4 .................................................................................................................. 269Node 5 .................................................................................................................. 270Microscale Histories .............................................................................................. 2727 DIVERGENT HISTORI ES IN PLAC E ................................................................... 307Renewing Comm unity ........................................................................................... 310Ancestral Tempor alities ........................................................................................ 318The Life and Death of Settlem ents ....................................................................... 329Transposing Histories ........................................................................................... 3388 THE CONSTRUCTION OF MOUNT TAYLOR COMMUNI TIES ........................... 352Inhabiting the S t. Johns ........................................................................................ 352Post-Inscrip tion ..................................................................................................... 365Crafting Histories ac ross the Div ide ...................................................................... 369 APPENDIX RADIOCARBON DATA ............................................................................................... 373


8 LIST OF REFE RENCES ............................................................................................. 379BIOGRAPHICAL SK ETCH .......................................................................................... 428


9 LIST OF TABLES Table page 1-1 Culture-history of northeast Fl orida .................................................................... 623-1 Minimum and maximum flood stage fr equencies for the St. Johns at Deland USGS gaging station ........................................................................................ 1475-1 Metric attributes of shell matrix sites based on typologies of Wyman (1875) and Wheeler et al (2000) ................................................................................. 2295-2 Metric attributes of shell ri dges ......................................................................... 2305-3 Metric attributes of multi-mound complexes. .................................................... 2316-1 Stratigraphic units of Test Un its 3 and 3A, Node 1, 8VO 215 ............................ 2776-2 Cultural materials recovered from Test Units 3 and 3A, Node 1, 8VO215 ........ 2776-3 Stratigraphic units of Test Unit 2, Node 2, 8VO215 .......................................... 2786-4 Cultural materials recovered from Test Unit 2, Node 2, 8VO215 ...................... 2796-5 Stratigraphic units of Test Unit 1, Node 3, 8VO215 .......................................... 2796-6 Cultural materials recovered from Test Unit 1, Node 3, 8VO215 ...................... 2806-7 Stratigraphic units of Test Unit 4, Node 4, 8VO215 .......................................... 2816-8 Cultural materials recovered from Test Unit 4, Node 4, 8VO215 ...................... 2816-9 Stratigraphic units of Test Unit s 5, 6, and 7, N ode 5, 8VO215 ......................... 2826-10 Cultural materials recovered from Test Unit 5, N ode-5, 8VO 215...................... 2836-11 Cultural materials recovered from Test Unit 6, N ode-5, 8VO 215...................... 2836-12 Cultural materials recovered from Test Unit 7, N ode-5, 8VO 215...................... 2836-13 Comparison of the size, spacing, and culture-historical associations of discrete shell deposits at 8VO214 and 8V O215 ............................................... 284


10 LIST OF FIGURES Figure page 1-1 Preceramic Archaic Mount Taylor period shell ma trix si tes ................................ 631-2 A naturally er oded shell m ound .......................................................................... 643-1 Late nineteenth and early twentieth c entury impacts to the St. Johns River region. .............................................................................................................. 1483-2 Distribution of known Mount Taylor and contemporaneous Preceramic Archaic shell matrix sites in northeast Fl orida................................................... 1493-3 Hillshaded topographic m ap of the eastern Flor ida Platfo rm ............................ 1503-4 Environmental processes in relation to cultural periods along the Middle St. Johns River. ...................................................................................................... 1513-5 Physiographic provinces of the St. Johns River Region ................................... 1523-6 Monthly rainfall and temperature av erages. ..................................................... 1533-7 Springs, lakes, and islands of the Middle St. Johns River Valley ...................... 1543-8 Daily streamflow-dur ation hydrog raphs. ........................................................... 1553-9 Monthly streamflowand st age-duration hydr ographs. ..................................... 1563-10 Channels and wetlands along the Mi ddle St. Johns River Valley. .................... 1573-11 Distribution of wetlands in relationship to channeled surface water and elevation along the Middle St. Johns Riv er ....................................................... 1584-1 Calibrated radiocarbon assays from Mount Taylor and Orange period components ...................................................................................................... 1924-2 Median calibrated age probabilities within shell matr ix sites ............................. 1934-3 Object and raw material source areas in relation to the Mount Taylor region ... 1945-1 Shell matrix sites cons idered in Chapt er 5. ...................................................... 2325-2 Comparison of site plans for the Old Town/St. Francis Mound (8LA29) ........... 2335-3 Scatter plot of measured height compared against the product of measured length and width for shell matrix site types. .................................................... 2345-4 LiDAR-derived cross se ctions of ridges ............................................................ 235


11 5-5 Comparison of shell ridge hei ght and distance to water. .................................. 2365-6 False-color and hillshaded topographic m aps of intact s hell ridges ................. 2375-7 False-color and hillshaded topographic m aps of intact s hell ridges .................. 2385-8 Footprints of ridges and multi-mound comp lexes. ............................................ 2395-9 False-color and hillshaded topographic map of the Thornhill Lake Complex (8VO58). ..................................................................................................... 2405-10 False-color and hillshaded topogr aphic maps of mined multi-mound complexes ........................................................................................................ 2415-11 False-color and hillshaded topographic maps of the mined Bluffton Complex (8VO22, 8VO 23). .............................................................................................. 2425-12 False-color and hillshaded topographic map of the mined Silver Glen Run Complex. .......................................................................................................... 2435-13 Old Enterprise Comple x ................................................................................... 2445-14 Sites with documented or suspected Mount Taylor components on Hontoon Island. ............................................................................................................... 2455-15 Median calibrated age probabilities and elevations of radiocarbon assays from terrestrial and su baqueous contex ts. ........................................................ 2466-1 Hontoon Dead Creek Comple x topographic base map. ................................... 2856-2 Hontoon Dead Creek Village to pographic base map. ....................................... 2866-3 Cross section profile of the Hontoon Dead Creek Village ................................. 2876-4 Distribution and density of shell matr ix at the Hontoon Dead Creek Village site. ................................................................................................................... 2886-5 Shell node delimitati on at the Hontoon Dead Cr eek Villa ge. ............................ 2896-6 Topography and boundarie s of Node 1 ............................................................ 2906-7 Node 1 stratigraphy and composit ion ............................................................... 2916-8 Topography and boundarie s of Node 2 ............................................................ 2926-9 Photograph of Node 2 excavati on .................................................................... 2936-10 Stratigraphy of Test Unit 2, Node 2 .................................................................. 294


12 6-11 Objects recovered dur ing Node 2 te sting ......................................................... 2956-12 Topography and boundarie s of Node 3 ............................................................ 2966-13 Stratigraphy of Test Unit 1, N ode 3. ................................................................. 2976-14 Objects recovered dur ing Node 3 te sting ......................................................... 2986-15 Topography and boundarie s of Node 4 ............................................................ 2996-16 Stratigraphy of Test Unit 4, Node 4 .................................................................. 3006-17 Topography and boundarie s of Node 5 ............................................................ 3016-18 Stratigraphy of Test Units 5, N ode 5. ............................................................ 3026-19 Southern profile of Test Unit 7, N ode 5. ........................................................... 3036-20 Cross section of the H ontoon Dead Creek Complex ........................................ 3046-21 Paleohydrological context of t he Hontoon Dead Creek Village. ....................... 3056-22 Areas of the Hontoon Dead Creek Comple x submerged during flood events .. 3067-1 Stratigraphic profile and stages of the Hontoon Dead Cr eek Mound ................ 3407-2 Composite photograph of St age 4, Hontoon Dead Creek Mound ..................... 3417-3 Crows Bluff (Osceol a) mound prof ile ............................................................... 3427-4 Harris Creek Mortuary base map. ..................................................................... 3437-5 Simplified profiles of the Harris Creek Mortuary. .............................................. 3447-6 Comparison of resident ial and mortuary mounds. ............................................ 3457-7 Bluffton Burial Mound shaded relief topographic map and cross-sections. ...... 3467-8 False-color shaded relief topographic map of the Silver Glen Spring Run Ridge (8LA1W -A). ............................................................................................ 3477-9 Simplified composite profile of Test Units 5 and 8, Silver Glen Spring Run Ridge (8LA1W -A). ............................................................................................ 3487-10 Macro-stratigraphic units wit hin the Trench (TU 9, 10, 15) at the Silver Glen Spring Run Ridge (8 LA1W-A)........................................................................... 3497-11 Simplified composite profile of Test Unit 8, Silver Glen Spring Run Ridge (8LA1W-A). ....................................................................................................... 350


13 7-12 Topography and excavations at the Hontoon Island No rth site ........................ 3518-1 Reconstructed Silver Glen Run Co mplex ......................................................... 372


14 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy REMAPPING HISTORIES: ARCHAIC PERIOD COMMUNITY CONSTRUCTION ALONG THE MIDDLE ST. JOHNS RIVER, FLORIDA By Asa Robert Randall August 2010 Chair: Kenneth E. Sassaman Major: Anthropology This study examines the significance of hunter-gatherer built landscapes through an analysis of Mount Taylor period (ca. 7300 cal B.P.) shell mounds along the Middle St. Johns River in northeast Florida. Mount Taylor communities are best known for the incipient exploitation of shellfish an d other aquatic resources in the region. In Florida and elsewhere, shell mounds are routinely interpreted as refuse accumulations, while their repeated occupation is taken to re present long-term continuities enabled by abundant wetland resources. Through a historiography of local research, I show that this model of shell mound growth reflec ts widely held anthropological assumptions regarding hunter-gatherer social simplicity and stasis, and obscures evidence for change through time. I develop an alternative framework based on practice theory to detail how shell mounds and associated plac es emerged as a historical process in which communities inscribed and politicized social memories through the deposition of shellfish and other material s within landscapes. In order to problematize t he contexts of shell mound inhabitation, I examine the paleohydrology of the Middle St. Johns River I also consider the evidence for


15 continuities in subsistence, settlement, ex change relationships, and mortuary traditions throughout the Mount Taylor per iod. The social and ecolog ical contexts provide a spatial and chronological fr amework for examining how Mount Taylor communities inhabited places. Because mo st shell mounds were destr oyed by twentieth century land-use practices, I use historic observations and modern remote sensing data to develop a geospatial database detailing the loca tion and organization of Mount Taylor places. The results of topographic and stra tigraphic testing of non-mounded shell sites are also reported. This analysis details how Mount Taylor communities established and renewed settlements as witnessed in small-sca le depositional practices. Finally, I reconstruct the histories of six Mount Ta ylor shell mounds based on stratigraphic testing. The results demonstrate that shellfishing wa s initiated at a time of considerable landscape instability by arguably diverse r egional populations. After being established as places to dwell, some preexisting settlem ents were reconfigured as platform mounds upon which shellfish was deposited in rit ualized sequences, while others were converted into foundations for mortuary mounds. Through time, other shell mounds experienced complex histories of abandonmen t, renewal, and transformation as well. I argue that along the St. Johns Mount Taylor communities routinely referenced past places in order to construct new social histories that accommodat ed or denied social and ecological change.


16 CHAPTER 1 SHELL MOUNDS AND OTHER NONEVENTS This study examines the ways in which so-called prehistoric hunter-gatherer communities wrote their own histories through inscriptiv e practices that modified their landscapes. By histories here I am explicit ly referring to the unfolding of daily and extraordinary events, as we ll as the practices, historical consciousness, and cosmologies to make those moments and proc esses significant with respect to an acknowledged past. Followi ng the arguments of Barrett (1994:169), Bradley (2002), Pauketat (2001b), and other s, it is my contention t hat these histories matter when archaeologists attempt to explain different arrangements of past communities and how they were reproduced and trans formed in time and space. Within the vast and complex literature on landscape, this is a line of argument that has significant currency with respect to past farming communities. In the absence of written narratives, such communities rout inely constructed and politicized the past as social memories. Histories in this sense were materialized through practical traditions of inhabiting settlements, constructi ng and transforming monuments, venerating ancestors, manufacturing and exchanging object s, and interacting with others (Barrett 1999; Edmonds 1999; Pollard 1999; Tilley 1994). They were the means by which diverse communities were integrated (C ameron and Duff 2008; Pauketat and Alt 2003; Yaeger and Canuto 2000), and by which physi cal environments were appropriated and made meaningful (Dwyer 1996; I ngold 1992; Ruiz-Ballesteros et al. 2009). In some cases social memories and their materializat ion provided a framework for interpreting catastrophic or unforeseen events (Pauketat and Emerson 2008; Sahlins 1985:138). In others, they structured practices in the c ourse of social reproduction, frequently to


17 interfere with perceived change (Shils 19 81; Wobst 2000). Above all, however, communities creatively generated histories as both a medium and outcome of social change (Pauketat 2001b:80). As such, the shifting arrangements of peoples, places, and things that archaeologists recover are the process of social becoming that we describe as societies (Barrett 2001:157), and are not reducible to generalized patterns of social evolution (Pauketat 2001b). One example of hunter-gatherer landsc ape inscription as history can be found amongst the communities who inhabited the Mi ddle St. Johns River Valley in northeast Florida during the preceram ic Archaic Mount Taylor Period (ca. 7300 cal B.P.1) (Figure 1-1). The hallmarks of preceramic Archaic inhabitation are highly visible mounds of freshwater shellf ish remains and earth in addition to subtle heaps and piles of shell distributed throughout the wetlands and terraces of the river valley (Randall 2008). At a minimum, these shell matrix sites2 register the consequences of intensive shellfishing by hunter-gatherer communities. Internally, such sites are characterized by (seemingly innumerable) depositional episodes varying in scale, composition, intensity, arrangement and social context. As I will detail in this study, over the course of three millennia preceramic Archaic communities generated places for daily habitation or inhumation of the dead through depositional practices involv ing shellfish and earth. Although many Mount Taylor places are composed primarily of shellfish remains and objects of residential life, some contain dedicated mortuaries, while others still were 1All radiocarbon assay estimates and age ranges are reported in calibrated years before present (cal B.P.) unless otherwise noted. 2The term shell matrix site will be us ed as a general reference to sites with shellfish remains. This term precludes static and normative typologies that view shellfish remains as proxies for behaviorally redundant refuse (i.e. middens); see Claassen (199 8:10-12). The term shell mound is reserved for those shell matrix sites characterized by significant relief and surficial extent.


18 constructed of sand primarily for the inhumation of the dead. Many of these places underwent rapid histories of construction, yet also show complex patterns of abandonment and reuse through which they were transformed in highly visible ways. Throughout this era communities variously engaged in local and regional interactions, signaled materially by the ex change and production of objects. As dynamic as community engagements were thr oughout the preceramic Archaic, so too was the river. Alternate periods of fl ooding and regression of the St. Johns obscured and resituated places of habitation, all t he while generating new ecologies that undermined or made more significant preexisti ng places. My argument is that shell matrix sites, and the practices occurring through them, figured pr ominently in the ongoing production of preceramic Archaic hist ories by providing a frame of reference against which contemporary social realities, su ch as regional interaction or hydrologic change, could be made meaningful. To draw a phrase from Gosden and Lock (1998:6), shell matrix sites were engines for the creat ion of time. As places of repeated habitation they framed the unfolding of daily life through the deposition of shellfish and other materials in places. At the same ti me, shell matrix sites provided historical resources as a past that could be made monum ental and commemorated, politicized, or neglected. My thesis presupposes that hunter-gatherer s have histories in the sense outlined above. Yet with few exceptions, such hist ories have not been readily forthcoming. Instead, Mount Taylor communities, and thei r hunter-gatherer equivalents elsewhere, have been treated as decidedly prehistoric phenom ena. In Florida, archaeologists have long recognized distinct associations between site location and assemblage


19 composition: sites composed pr incipally of freshwater she llfish remains and associated cultural materials are preferentially situat ed within or adjacent to wetlands and channel segments (Goggin 1952:67; Wym an 1875:10). The incipient exploitation and deposition of shellfish 7000 years ago appears roughly c oeval with a near-modern hydrological regime that established partially as a consequence of decelerating middle Holocene sea level rise (Miller 1992). Taking these two observations together, it has been presumed that the organization and distribution of s hell matrix sites directly reflects the development of wetland resources and subsequent pressures brought on by population increase (Milanich 1994:87). As a consequence, Mount Tayl or traditions are routinely characterized as a response to large-scale environmental phenomena (Goggin 1952; Miller 1992, 1998). The over-arching presum ption is that the environment provided resources to which populations variously adapted through a combination of increased settlement permanence and intensiv e resource extraction. In a related vein of thought, shell mounds are routinely interpreted merely as refusetrue middensand thereby unintentional and socially insignificant. While burials have been widely recognized in sand and shell deposits within mounds, their presence is presumed to be incidental to other biologically important processes such as consumption and discard (see also Milner and Jefferies 1998). Questions regarding Archaic lifeways in northeast Florida are dom inated by issues of ecology, diet, and paleoenvironmental reconstruction (Cumbaa 1976; Miller 1992, 1998; Russo et al. 1992). As one local researcher recently articulated, the Archaic period shell middens of central-east Florida represent a series of outdoor laboratories where the study of human adaptation to the early and middle Holocene Epoch of peninsular Florida can


20 take place (Quitmyer 2001:25) This sentiment is echoe d throughout the literature on shellfishing (e.g., Ambrose 1967; Mar quardt and Watson 2005a; Morey and Crothers 1998; Stein 1992). Mount Taylor communities are part of a larger assemblage of early to middle Holocene hunter-gatherer soci eties who engaged in intensive aquatic exploitation in the post-glacial era in the southeastern Unit ed States (Claassen 1996; Dye 1996; Russo 1996b). Shellfishing emerged as early as t he eighth millennium in the Midsouth (Dye 1996) and persisted as a widespread phenomena into the Woodland period 1000 years ago (Peacock 2002). It should be noted that earlier coastal exploitation is likely. The most widely recognized tradition is the so-called Shell Mound Archaic, originally applied to all shell bearing sites dating to the middle Holocene, but the term is now preferentially used to refer to shell matrix sites in the Midsouth (Sassaman 2004a). Elsewhere shell matrix sites have been identi fied along the Atlantic and Gulf coasts. The timeframe encompassed by incipient shellf ishing in the greater Southeast coincides with the Middle and Late Archaic periods, often portrayed as a time of increasing economic specialization and technological innovation, decreasing mobility, and widespread and elaborate exchange (Bender 1985; Brown and Vierra 1983; Jefferies 1996; Marquardt 1985; Sassaman 1996). Worldwide there is a dramatic increase in riverine and coastal settlement centered on aquatic exploitation roughly 7000 years ago (Bailey and Parkington 1988; Nicholas 1998). In many cases, this apparent int ensification emerges with other aspects considered emblematic of so cial complexity: status i nequality, relative settlement permanence, storage, and technological innov ation (Price and Brown 1985). A number


21 of arguments have been offered to explain th is chronological and social pattern, including the irresistibility of newly established aquatic resources (Brown and Vierra 1983), competition over scarce terrestrial resources (Charles and Buikstra 1983), appropriation of surpluses and competitiv e feasting in the context of abundant resources (Hayden 1994, 1995), or gr oup fragmentation, increased settlement permanence, and technological s pecialization and efficiency in tension zones (Binford 1968, 1990). Such explanations of shellfishing and its cons equences are fundamentally based on the role of subsistenc e economies shaping social c hange, particularly as they would have affected population dynamics, provided new opportunities, or introduced new crises to long-term f ood-getting strategies. The orthodox approaches outlined above have perpetuated a particular kind of history for ancient huntergatherer societies. By fo cusing on long-term social and environmental processes, what Braudel (1980:27) referred to as la longue dure, archaeologists have composed structural historie s at the expense of investigating shortterm moments of social r eproduction, interaction, and transformation. In these uneventful histories, the primary agents delimiting past societies are impersonal processes such as economy, climat e, and demography operat ing over multigenerational time scales. Although particula r events, practices, and beliefs no doubt existed within these structures, they have no bearing on the long-term persistence of any one pattern. What is left are decontex tualized and gradualist perspectives on past hunter-gatherer social change and stability. Little room is le ft for the histories of past communities, who are presumed to operate in ignorance (or denial) of their own pasts.


22 This is not to say that structural histor ies are entirely uneventful. As noted by Gamble (2007:24), archaeologists since t he time of V. Gordon Childe have tended to view the past as a series of revolu tions (e.g., Neolithic, Urban, Industrial), characterized by rapid change, interspers ed with long periods of gradual change. Although most archaeologists today do not per ceive such watersheds as events per se, revolutions are given the rhetorical force of events. Gamble ar gues that revolutions continue to serve two functions in archaeologica l interpretation. The first is that they provide a container for organizing a series of concepts and observations into a manageable network (Gamble 2007: 19). As a collection of related issues, such problems can be studied without reference to past conditions. Yet revolutions also provide locomotives of change, as explan ations for how the Western world as we know it came into being by successively transcending tradition and primitiveness (Gamble 2007:13). Hunter-gather ers, defined as they are in terms of subsistence, and occurring chronologically prior to those re volutions that would eventually lead to contemporary nation-states, re main subject to the whims of dehumanized, long-term processes. For a more thorough examinat ion of the implication for Gambles perspective on hunter-gatherer s and the Archaic period in the southeastern United States see Sassaman (2010:Chapter 6). Within structural histories, practice s such as shell mound inhabitation and construction equate to what Fogelson (1989) has referred to as nonevents. Fogelson observed that many processes, practices, and beliefs important to nonwestern peoples are ignored or obscured by Western sensibilities of what constitutes proper or legitimate history (see also Echo-Hawk 2000; Nabokov 2002). Such accepted histories invoke


23 linear concepts of time, emphasize concrete facts or persons, and require verification via accepted standards. In narrative histories, events are isolatable moments around which plots turn, such as a series of so cial and technological revolutions discussed above. In contrast to an event, a nonevent does not carry empirical weight or rhetorical force in the reconstruction of the past. A nonevent, then, is process or moment that does not fit dominant ontologies. Fogelson envisioned an array of nonevents, which he termed pseudo events, imagined events, and epitomizing events. All of these constructions of the past may have been cent ral to past community identity, but were nonetheless ignored in syntheses and grand narratives. As an example, Fogelson described the difficulty with which Native Americans experience gaining recognition under Federal law, particularly when armed solely with oral traditions It is precisely because these groups often resisted intrusive documentation or intervention that their history is a nonevent: by law and decr ee they never happened (Fogelson 1989:141 142). This is, in fact, a problem for many hunter-gatherer communities whose land use practices often do not match criteria set by t he State to justify owner ship or control of traditional lands (Pinkoski and Asch 2004; Zv elebil 2003). Fogelsons larger point, however, is that there can be differential recognition of w hat is considered an event depending on the history and histor ical consciousness of the co mmunities involved (see also Pauketat and Emerson [2008] and Sahlins [1985]). Fogelsons (1989) discussion emphasized the importance of oral traditions amongst non-lettered societies in the construction of social histories. I would note that Fogelson (1989) himself perpetuat ed a view of ancient native North American history as uneventful by invoking a revolution of sorts. In addition to referring to the Archaic period


24 as characterized by monotonous sameness (Fogelson 1989:139), he suggests that Native American communities were largely ahist orical and slowly changing, or cold in Levi-Strauss (1966:233) terms, only becoming historical in the centuries leading up to, and following, contact with the West (Fogelson 1989:139). The implication being that it was not until later in time, and then into the Colonial era, that Native Americans experienced sufficient change and social comp lexity that they would reflect upon and transform their own histories to in corporate contempor ary concerns. Yet written or oral hist ories are only one means through which societies make history. Unfortunately, bec ause archaeologists do not have access to the verbal or written discourse of past societies, it is even easier to perceive them as ahistorical nonevents. Bradley (2002:3) makes a similar point in his discussion of the invention and dual meaning of prehistory. When c oupled with a model of unilineal social evolution, prehistory became synonymous with ancient or di stant peoples, who could be used as a foil against which the success of Western Civilization could be measured (see also Fabian 1983). Like Bradley, Kehoe (1998:1) situates the emergence of prehistoric archaeology, closely allied with geology (the study of long-term process), in the nineteenth century as social scientists attempted to ex amine the development of humanity untrammeled by the authority of texts. Bradley (2002:5) continues, noting that another meaning of prehistor ic refers to those societie s whose historical traditions are not treated as history, including monument construction, seasonal movement, object curation, and gift giving. Because such prac tices are not routinely recognized as being produced with respect to an indigenous historical consciousness, they are reduced to epiphenomena of past economic or social systems.


25 Archaic communities have become nonevents, and their history prehistoric, by the imposition of orthodox concept ions of linear time that foreground gradual and stepwise change. In this noneventful view, Archaic shell mounds provide a record of food consumption practices that were optimally selected against a range of alternatives, or serve as paleoproxies of environmental proce sses operating at centennial or millennial time scales. Hiding beneath dietary or pal eoenvironmental reconstructions is the belief that shell matrix sites are significant for what they contain (i.e shellfish and other materials), and not the social processes ultimately responsible for their generation and final dispositions. To be sure, human-environm ental interactions are one of a myriad of important issues surrounding a shellfishing way of life. Moreover, long-term processes were certainly at work throughout the St. Johns. Yet emphasis on shell matrix sites as a proxy record has precluded investigating how the creation of shell matrix sites may have been central to ongoing Archaic history-maki ng. Indeed, many events of Archaic significance may have been obscured by imposit ions of western hi storiography that emphasized gradual change through time and the prim acy of diet over contingent social reproduction. In this study, I will develop the notion of shell matrix si tes as more than unintentional heaps, and certainly more than w hat they contain. Central to this discussion is a need to rethink the relati onships between huntergatherers, social change, and space. Drawing on the work of Barrett (1999, 2000, 2001) and Pauketat (2001a, b) I argue that landsca pe inhabitation, including the creation of settlements and sacred places, was a historically conti ngent process of becoming, through which communities were reproduced and transformed. As such, placemaking emerges as the


26 central means through which past hunter-g atherers along the St. Johns and elsewhere wrote and contested their own histor ies. In the sections that follow in this chapter, I will outline some of the epistemological and ontological barriers to hunter-gatherer landscape use as anything but a nonevent, and describe some of the unnecessary contradictions that emerge w hen we fail to take into acc ount social histories. How is it then, that Archaic shell m ounds specifically emerged as nonevents? Whose historiographies are we reproducing in our own research? More importantly, is it necessary that we continue to emphasize impersonal ecologies and nat ural histories at the expense of historical alternatives? A co mplete answer to how we came this far will take us back to the origins of archaeologi cal thought in Florida and elsewhere, and necessitates examining the under lying assumptions that have dominated discussions of past and contemporary hunt er-gatherers. Inconceivable Agencies al ong the St. Johns River It was in the nineteenth century that prehist oric time and the significance of shell mounds were invented along the St. Johns, as they were throughout the Westernized world (Bradley 2002; Kehoe 1998: 1). Early st udies were concerned with establishing the anthropogenic or geologic origins of mounds along the St. Johns, but quickly turned towards questions of culture-history by the e nd of the century. During these embryonic years the dichotomies between natural and cult ural, chronologically early and late, and simple and complex were established. These dichotomies have hence served as frames of reference for understanding societie s along the St. Johns. In particular, a suite of normative cultural traits were isolated and arranged along an evolutionary continuum. These traits included mortuary treatment, object exchange and production, subsistence, and intentional mound constructi on. As I will document here, the image of


27 ahistorical Archaic communities that c ontinues to be produced in archaeological discourse rests on the uncritical acceptance of assumptions and observations dating to these early years. We can begin this story in medias res as the first concerted efforts to document and explain the shell mounds of the St. Johns were well underway. Much of our current understanding of Archaic shell mounds can be traced back to February 8, 1871. Jeffries Wyman, first curator of Harvards Peabody Museum of Ethnology and Archaeology, spent that morn ing examining the er oded profile of a massive freshwater shell mound on the western s hore of the Middle St. Johns River in northeast Florida, known today as the Silver Glen Springs Comp lex (Figure 1-1). This was one of many winter trips he made to the region betw een 1860, 1867, and 1871 to study the shell mounds along the river. These trips organized in part to collect specimens for Harvards new museum (Wyman 1875:15). More than collecting, how ever, his goal was to determine whether human or natural actions were responsible for these geomorphologically anomalous, yet widespre ad topographic features. In addition to Wymans (1867) earlier explorations in New England his interest was piqued by the discoveries of shell matrix kjkkenmdding sites in Denmark (Morlot 1861) and along the coast of Maine (Chadbourne 1859). Heaps of shell thought to be naturally deposited shellfish beds were determined to be anthropogenic upon systematic scrutiny. The term kjkkenmdding is the basis for our well-worn term midden, which in its original context referred expressl y to material that had been discarded in the course of food preparation and consumption (Claassen 1998: 11; Trigger 198 6:xiixiii).


28 In his own travels, Wyman observed scores of shell mounds distributed throughout the Middle and Upper St. Johns basin similar to that at Silver Glen Springs. He identified many variations on a theme: isolated linear or crescent-shaped ridges, immense multi-mound complexes, and even sm all heaps that were all primarily composed of freshwater shellfish remains. He also noted smaller mounds of sand, sometimes associated with shell ridges but o ften set apart. Centuries of wave action and bankward erosion at many shell mounds pr ovided stratigraphic profiles for him to examine large-scale features and small-scale objects in situ (Figure 1-2). In most he found incontrovertible evidence for the mounds artificial origins in the form of pottery, stone and bone tools, and even human remains. He also identified variations in the thickness and composition of shell deposits, not ing that many strata lacked anything but shellfish remains. No doubt struck by the extent of the mound facade at Silver Glen Springs, estimated at 8 m high and 300 m long, Wyman re tired that afternoon to an orange grove upon the mound and contemplated the apparent contradict ion between the overwhelming scale of the shell mounds on one hand, and their mundane content on the other. Wymans journal entries (1872) and publish ed monograph (1875) provide a sense of the questions he was struggling with: How did these landscape features come into being? What agents (anthropogenic or otherwise) could account for their deposition and final configuration? What role, if any, did the St. Johns River have in creating such features? If humans were res ponsible, what was their intent in amassing materials in these locations? If shellfish a ccumulations were monumental in scale, were


29 they monumental in significance? What re lationship, if any, do shell mounds have with either the sand burial m ounds or low-lying shell fields identified throughout the region? Based on these ruminations, Wyman ( 1875:11) would later pen the following statement that at once deftl y summarized his empirical observations, and arguably set the agenda for over a centurys worth of archaeological investigations: Any one who for the first time views the larger ones [shell mounds]might well be excused for doubting that such immense quantities of small shells could have been brought together by human laborIt is, however, absolutely impossible that such quantities of shells and such combinations of objects could have been the result of natural causes. It is impossible too to escape the conviction that these deposits of shells, and the remains of the various animals they enclose, were collected in any other way than as the refuse portions of articles of f ood. If the object of the Indians had been merely to construct a mound, materials for this in all cases could be had, as with the burial mounds, by scraping t ogether sand from the neighborhood, and which even if brought from a consi derable distance, would involve far less time and labor than the slow gathering of shells from beneath of the waters of creeks and lagoons. Drawing on the cumulative forc e of his data, he correctly conc luded that the shell matrix sites along the St. Johns were ultimately the product of human agency. Yet Wyman also offers a pragmatic resolution to t he shell mound contradiction by introducing a dichotomy: in contrast to the sand bur ial mounds, which manifestly demonstrate intentionality and forethought, she ll mounds singularly represent the cumulative result of food collection and refuse deposition. He ev en extended this deductive logic to the fragmentary human remains he uncovered amidst the shell mounds, arguing that their disposition was direct evidence for cannibal ism (Wyman 1874; 1875:6771). Taking the data at face value, how could the scale, final configurations, and apparently mundane composition of shell mounds be otherwise explained? At the time, Wymans insistence on the i ndigenous origins of the shell mounds, in addition to his recognition of change through time, was a progressive stance for a North


30 American scientist. Speculators, government officials, and naturalists alike were then entrenched in debating the various non-indigenous identities of mound builders across the Eastern Woodlands (see Trigger 1989:104). It woul d be another twenty years before the debate would be resolved through archaeol ogical research that demonstrated, in fact, Native Americans ha d produced them (i.e. Thomas 1894). Only after the turn of the century would archaeol ogists seriously consider change through time (Trigger 1986:xxii). Alt hough others had come to this conclusion for other shell mounds in Florida (Brinton 1859:177), it was the insightfulness, rigorous attention to detail, and skepticism that m ade Wymans observations endure. Wymans contributions to Florida archaeology cannot be overestimated. Many of the she ll mounds he recorded have since been destroyed or significantly co mpromised. As wit h anthropogenic shell deposits the world over (Ceci 1984), shell mounds were targeted as sources of construction fill and razed by bulldozers and draglines in the twentieth century. Wymans own student, Clarence Bloomfield (C. B.) Moore, was responsible for much destruction as well. Cannibalism aside, Wymans empirical observations have withstood the test of time. Any account of s hell mounds in the region must consider his monograph and notes, as his descriptions form an irreplaceable record of places that no longer exist. Wyman did not simply provide a record of observations, he also established prehistoric time along the St. Johns. Time emerged naturally fr om the stratigraphic profiles he examined. In many mounds, he recognized basal strata without pottery, which were superseded by successive strata bearing fiber-tempered pottery and check stamp decorated pottery. Many mounds lacked pottery altoget her, or it was expressed


31 only on the surface, and he argued that these m ounds were the earliest. This tripartite scheme not only forms the basis for our current culture histories, but was one of the earliest recognitions of change in the past by an American archaeologist (Trigger 1986:xvii). Less recognized, however, was the lasting impact of Wymans dualism between the intentional construction of sand mounds for the purposes of burial and the unintentional accumulation of shellfish refuse on the other. Wyman (1875:47) held open the possibility that burial mounds were as old as the earliest shell mounds. But this delineation was appropriated by later res earchers. The differences between a monument and a midden were used to co nstruct culture histories, and make behavioral interpretations as well. His dichotomy continues to shape the contours of archaeological theory and practice, and as I will detail below, has done much to obscure the details of past practices. Wyman was neither the first nor last to in vestigate the St. Johns shell mounds. In his synthesis of northeast Floridas Coloni al period, Goggin (1952:31) argued that no significant mention of indigenous use of shell matrix sites was made by either the early Spanish and French settlers. The earliest collected references are known from the late eighteenth naturalist William Bartram (1791), who described shell mounds as shelly bluffs, which were frequently used as ca mpsites by foreigners and Native Americans alike. A second wave of observations em erged in the later portion of the nineteenth century. Louis de Pourtals, a surveyor and student of the naturalist Louis Agassiz at Harvard, discovered human bones in shell d eposits at Stone Island/Doctors Island on Lake Monroe in 1848, and considered the depos it to be intentionally placed there (Agassiz 1881; Wyman 1875:15).


32 Brinton wrote several accounts on Flor ida in general, and shell deposits in particular for popular (1859, 1869) and prof essional (1867) audiences. Brinton recognized correctly that many of the s and mounds were construc ted as burial mounds (1859:172). He only recognized shell m ounds occurring on or near the coast as anthropogenic in origin. Some he considered to be mere refuse heaps, the debris of villages of an icthyophagous population, showing no indication of their having been designedly collected in heaps (Brinton 1867:356) Others, such as Turtle Mound on the Atlantic Coast and Crystal River near the Gulf of Mexico, he considered to be designed as watchtowers. In contrast he supposed that shell mounds along the St. Johns were geological in origin and were reused as places of burial by indigenous groups (Brinton 1867:357). Other contempora ry guidebooks lack the historical depth and consideration of Brinton, and describe shell mounds as s hell bluffs or coquina and do not ascribe them to human agents (e.g., Barbour 1882). William Dall (1885) briefly made mention of the 5-m high Old Enterprise Complex, and was convinced that it was intentionally constructed due to the general lack of anything but shell. Finally, Le Baron (1884) provided an account of mounds along the river that was, by his own admission, a reconnaissance and not a thor ough examination. Although Dall and Le Baron were aware of Wymans resear ch, it would was one of his last students, the wealth y and prodigious C. B. Moor e, who formalized and tested his observations through extensive excavati ons (Mitchem 1999). Moore revisited all sites observed by Wyman, found many ot hers, and attempted to verify Wymans conclusions. Unlike Wyman, Moores work is notable for his geographically expansive survey, extensive excavations, and, relative to the times, detailed and well illustrated


33 publications. His earliest explorations targeted shell mounds, which he identified throughout the Upper and Middle St. Johns basin (Moore 1892, 1893a, b, c, 1894c). Moores excavations led him to uphold an d expand upon Wymans conclusions: the shell heaps of the St. Johns (sic) are refu se heaps simply, and in them refuse alone can be expected under ordinary circumstances and any stratification within shell mounds was of course a matter of acci dent (Moore 1892:913). Despite this opinion he suggested in his earlie st synthetic paper of shell mounds that variation in shell composition reflected diffe rent modes and intensities of refuse disposal and living, although he never continued this line of thought. For example, he suggested that crushed shell lenses were likely the result of trampling, while whole shell strata were the result of dumping episodes and lacked sign ificant post-depositiona l trampling (Moore 1892:915). Following this initial focus, Moore expan ded his efforts to sand mounds (1894a, b), and largely stopped excavating shell mounds once he recognized the greater likelihood of locating objects elsewhere (1894c:26). Almost all sand mounds were found to contain human burials, and most contained pottery. However, he identified some mounds, such as those at Bluffton, Thornhill Lake, and Orange Mound in the Upper St. Johns, that lacked pottery except for surf ace exposures. Thornhill was particularly interesting for him because at least one i ndividual was interred with numerous marine shell beads, in addition to extralocal objects such as groundstone beads and bannerstones (Moore 1894b:168). Bannerstones were reported in a non-pottery bearing sand mound at Tomoka on the coas t as well (Douglass 1882). Elsewhere Moore had identified near-basal sand burial lenses within shell mound strata lacking


34 pottery altogether, or upon which pottery wa s emplaced. These inhumations presented a problem. Most evidently they violated Wy mans shell/sand dichotomy. Because the human remains were largely unbroken, they could not be easily dismissed as evidence for cannibalism. More problematic, however was that the lack of pottery in mounds violated his progressive evolutionary schema. Most shell mounds lacked pottery except in their upper stra ta, they were demonstrably older and by extension less evolved socially. In discussing the relative age of sand mounds and shell mounds, Moore (1894b:210) rescued Wyman and social evolution by concluding that save with burials, no incentive existed for the inhumation of pottery, whole or in sherds, in the sand mounds, and that even in the burial mounds it was simply a matter of custom, almost universally followed, it is true, but still not without exceptions. With the conclusion of Moores investigat ions, interest in archaeology within the region waned considerably (Goggin 1952:34 ), although Nelson (1918) did conduct excavations at Oak Hill on the Atlantic coast where he identified a preceramic sequence. Nelson in fact suggested the mound was constructed rapidly, but this observation has never been seriously considered (although it should have been). Beginning at the turn of the c entury, archaeological investi gations shifted towards triage, as one site after another disappeared at t he end of draglines an d steam shovels, accelerating with the growth of population in the region. During the 1930s work was conducted by the Civilian Conservation Co rps (CCC) along the Lake George segment of the basin (Abshire et al. 1935). Beginning in the 1950s Ripley Bullen and others at the Florida Museum of Natural History (then the Florida State Museum) initiated more systematic investigations at intact sites (Bullen and Bry ant 1965), and others as they


35 were being destroyed (Bulle n 1955b; Jahn and Bullen 1978) Aided by publication through The Florida Anthropologist avocational archaeologists made numerous contributions, searching through mined backfill and as shell was washed prior to final transport (e.g., Gut and Neil l 1953; Neill 1954). It was also in the middle of the twentieth century that the culture history of the region was literally transcr ibed from shell mound prof iles and systematized by John Goggin (1952) (Table 1-1). Fr om the beginning of concerted investigations in northern Florida archaeologists recognized a pervasive non-association between ceramics and the basal deposits of shell mounds. This comp lex was initially recognized as part of the Southeastern Archaic Stage, and defined simply by J. W. Griffin (1952:323) and Goggin (1947:123) as a Pre-pottery period or Non-ceramic horiz on respectively. After a detailed analysis of extant collections and historic observations, Goggin (1952:40) defined the preceramic Archaic Mount Tayl or Period, named for the shell mound for which there was the most contextual data, as the earliest period of occupation on the St. Johns River. This period has now been securely dated between 7300 and 4600 cal B.P. Today we recognize earlier traditions of the Late Paleoind ian and Early Archaic periods, but they are poorly understood. Fo llowing the systematics of the times, his definition of the Mount Taylor period was based on the asso ciation of a host of objects, but also coupled with behavioral interpretation. Although recognizing that burials could occur within shell matrix sites, he (1952:66) argued that the indi fference towards the dead is emphasized by the lack of burial offerings. It is also at this time that we see a movement toward shell matrices as nothi ng more than a container for objects and an aid for chronological ordering, as he noted (1952:41) there is some stratification in the


36 deposits but it apparently has little cultural significance. Bullen and Bryant (1965:21) would echo this sentiment later, noting t hat the presence of shellfish layers without abundant objects made chronological consider ations and reconstruction of past lifeways difficult. The regional chronology continues wit h the appearance of fiber-tempered pottery, which signals occupation during the ceramic Archaic Orange period (ca. 4600 cal B.P.). Bullen (1972) w ould later subdivide the Orange period based on what he perceived to be changing decorative styles Finally, superimposed upon many shell mounds are the residues of t he post-Archaic St. Johns I and II periods, evidenced by the presence of spicule-tempered St. Johns pottery (ca. 3600 cal B.P.), often found in association with objects and inhumations in low, truncated cone sand mounds. Although the orthodox culture-historical sequence of the middle St. Johns was internally consistent, in truth it belied seve ral empirical contradictions. Most notably, burials in mounds or other elaborate contexts were not recognized as Archaic in origin, precisely because they did not fit the expec tations of unilineal social change through time. For example, Goggin attempted to use changes in burial treatment to define subperiods within the St. Johns period, al though with admittedly limited success. He did, however, define an Unclassified Complex which included sand mounds, bannerstones, and stone beads. As Endonino (2008) has recognized, this is the same complex that C.B. Moore struggled with in his investigat ions at Thornhill Lake. Goggin did not give it a precise cultur al attribution, but suggested it was of the St. Johns period, arguing that the concept of sand burial mounds is apparently post-Archaic or at least only recognized in this part of Florida in S t. Johns I times. Of the bannerstones which


37 Goggin (1952:53) knew to be of Archaic age, he could onl y say that there was some sort of a carry-over from very early times. Other culture-historical problems with burial mounds we re encountered by Sears (1960) at Bluffton, where an individual was interred beneath rapidly deposited lenses of shellfish, sand, and muck. Although the individual was not interred with pottery, Sears concluded that it was St. Johns in origin Perhaps the most egregious cultural misascription was by Jahn and Bullen (1978). While performing a salvage excavation at the multi-component Harris Creek mound they documented a mortuary containing over 160 individuals interred in a mound constructed of white sand and shell. This had been covered over by meters of later strata. Although the mate rial culture was consistent with Mount Taylor assemblages, and no pottery was found within undi sturbed strata in the mortuary, they were loathe to ascribe it to anything but the later St. Johns period. While holding out the possib ility that the mortuary was Ar chaic in age, they (Jahn and Bullen 1978:21) suggested t hat St. Johns era people dug a tremendous hole in a preexisting preceramic Archaic midden, built some type of charnel house therein, introduced vast quantities of white sand in which they serially interred hundreds of burials which had been collecting in the charnel house (or hous es), covered the complex with the Archaic midden material previously removed. This is a prime example of the constr uction of a nonevent, wherein prevailing assumptions regarding the worldview and abilities of Archaic communities precluded an objective assessment of the data at hand. Beginning in the 1980s treatments of she ll matrix sites consisted of targeted research-based excavations and cultural re source management (CRM) mitigative work. In particular, Barbara Purdy (1987, 1994b) l ed several investigations into inundated wetsite deposits at Groves Orange Midden, an extension of the Old Enterprise


38 Complex, and the Hontoon Island North site, one of the largest shell complexes in the region before it was mined for shell in 1935. Her research interests lay solely in recovering well preserved organics. At Groves Orange Midden she and her colleagues identified a saturated shell component of Mount Taylor age in which abundant wooden objects and botanical remains were recovered. Analyses focused on material culture (Wheeler and McGee 1994b) and subsistence remains (Newsom 1994; Russo et al. 1992; Wheeler and McGee 1994a). At the Hont oon Island North site, excavations by Purdy (1987; 1991:102) and colleagues (Newsom 1987; Nodine 1987; Wing and McKean 1987) focused on wetsite deposits adjac ent to what had been the eastern apex of the largest shell ridge. They reco vered well-preserved faunal and botanical assemblages in addition to a ceramic inventor y of over 70,000 St. Johns sherds in a shell matrix. Although the presence of preceramic deposits were surmised (Purdy 1991:130), they were thought to only be of limited significance and extent. Later excavations by Sassaman, myself, and colleagues (2005) would actua lly prove that the majority of the 300 m long shell ridge dat ed to the Mount Taylor period. The late twentieth century was also a period of renewed synthesis, which principally involved reducing Mount Taylor to its most essential qualities. Following Goggin, Miller (1992, 1998) sugge sted that Mount Taylor lifeways were enabled by the emergence of productive wetland locales. He developed a hydrological model that equated the onset of near-moder n hydrology with the est ablishment of productive ecosystems, in large part enabled by increased flow from freshwater springs. In his synthesis of Florida archaeology, Milani ch (1994:87) offer ed that the apparent emergence of shellfishing during the Mount Taylor period was part of a restructuring of


39 regional populations, largely because it was much less difficult to harvest fish and mollusks than to maintain a hunter-gathering way of life in the interior forests. He envisioned Mount Taylor populations initiating what would be a very long running record of sameness enabled by these new economic re sources, as he not ed that there are changes apparent in the archaeological reco rd of the Late Arc haic periodBut these changes are related to demography rather than changes in basic lifeways (Milanich 1994:86). As revealed in this short historiography Wymans original dichotomy has emerged as both a temporal and ontol ogical duality. The initial distinction between shell mound deposition and sand burial mound constructi on has been tacitly accepted and coupled with a model of unilineal, progressive social evolution that remain s the default frame of reference. Despite the clear evidence fo r Mount Taylor mortuar y mounds, they were routinely ascribed to a later period (Endonino 2003b). Beyond mortuaries, however, it is inconceivable that Archai c shell deposits reflect anything but the unintended consequences of mundane and unreflective behavior by chronologically early huntergatherers. In fact, the St. Johns River is allotted a gr eater role in creating and structuring Archaic period regi onal inhabitation than the comm unities themselves (e.g., Miller 1998). It is equally inconceivable t hat mounds and associated places could tell us anything about Archaic societie s beyond subsistence, or alter natively, that such places were significant to Archaic communities bey ond trash heaps. As a result, research into the structure, organization, and significance of Archaic practices at shell mounds has largely been precluded.


40 Interpretations of the signi ficance of the first and last vestiges of occupations are animated by opposed narratives and operationaliz ed by different research agendas. Archaic-period components, produced by chronologically early and presumably socially simple hunter-gatherers, are viewed as a proxy for past environmental conditions. Change in site location or size and vari ations in depositional episode content are reduced to either ecological change or demographic processes. In contrast, the largest and most structurally comp lex sites are attributed a priori to the post-Archaic St. Johns period (e.g. Purdy 1987), where site structure and location serves as a proxy for social organization and ritual processes (G oggin 1952:76; Milanich 1994:254). Depositional acts at mound sites are cons idered to be truly monumental, reflecting intentional acts constitutive of the politi cs and histories of St. Johns communities at sacred venues. Higher population densities, se ttled village life, the presumed use of cultigens, and social stratification are model ed as coincident with a step-wise increase in the scale of monumental construction (Go ggin 1952; Milanich 19 94:256; Miller 1998; Smith 1986; Steponaitis 1986). A Persistent Other That a noneventful Archaic has its roots in the nineteenth century is evident by now. If this was a parochial problem, it woul d be simple to reset the research agenda in northeast Florida and carry on. Unfortunately, the interpre tations along the St. Johns are part of a much larger ont ological handcuff. Among hunt er-gatherer specialists there has been a long-running discourse on the existence and significance of the social category hunter-gatherer (Barnard 1999, 2002; Ingold 1996; Kent 1992; Lee 1992; Shott 2001; Solway and Lee 1990; Wilmsen and Denbow 1990). Much of this discussion has been in regards to whether t he contemporary forager s that frame our


41 ethnographic sources of analogy (including communities in the African Kalahari, Australia, and elsewhere in the tropics) repres ent social configurations unaffected by cultural contact, or are a pr oduct of complex histories of in teraction with more powerful others (Sassaman 2004a:229; Sassaman and Heckenberger 2004). The discussion has great significance for the identity and soci al status of those labeled hunter-gatherers by the State (Buntman 2002; Lee 2006; Pinkoski and Asch 2004). Moreover, it is highly relevant to how we understand contempor ary hunter-gatherer societies, and the potential power relationships underlying many culture contact experiences in the past (Sassaman 2001; Wolf 1982). In mo st cases, however, this noti on of history is forensic. The emphasis is on determining which aspects of forager lifeways may be modern, and which may be considered a more persist ent form of a general condition. Although widespread social, historical, and st ructural variation is recognized within the category (Ames 2004; Kelly 1995), most who have considered the subject conclude that there are enough similariti es amongst present-day for agers to continue using the construct (Barnard 2002; Lee 2006; Politis 2007:29). A prominent component of this discourse is the structural relationsh ip between hunting and gathering as a mode of production and landscape use (Ingold 1988). How communities use space at their disposal is part of a core of attributes that delimits the categor y hunter-gatherer in opposition to non-foraging peoples (Kelly 1995: 111; Lee and Daly 1999). To be sure, Lee and DeVores (1968:11) famous dictum that hunter-gatherers move around a lot has been undermined by a greater recogniti on of sedentary or infrequent annual mobility (Kelly 1992; Ma rquardt 1985; Price and Brow n 1985; Whitelaw 1991). However, most researchers view movement through or engagement with landscapes as


42 restricting the possible range of variation in hunter-gatherer societ ies (following in the steps of Stewar d [1955]). Central to this emphasis on landscape use are other aspects defining huntergatherer lifeways, most notably subsis tence economies based entirely on wild foodstuffs, and concomitant worldviews and so cial organization that emphasize sharing and egalitarianism (Bird-David 1990; Ingold 1999). Numerous typologies within the class of hunter-gatherer have been construct ed to describe the relationship between landscapes and alternative strategies of food getting, settlement arrangement, technological organization, modes of so cial interaction, and socioeconomic differentiation (e.g. Bettinger and Baum hoff 1982; Binford 1980; Testart 1982; Woodburn 1982). Archaeologists have also cons tructed sophisticated models and sets of expectations based on material culture signatures to exam ine how systems of landscape use and settlement organization were structured in the past (Binford 1979, 1980, 1990; Chatters 1987; Gamble 1991; Habu and Fitzhugh 2002; Kent 1991). Yet landscape use has a dual identity in anthropological conc eptualizations of hunter-gatherers. On one hand, spacevariously conceived of as the physical distribution of social or economic res ourcesis accorded a fundamental role in structuring variation in for ager lifeways (e.g. Binford 1980; Kelly 1995:111; Lee and Daly 1999). On the other hand, the ways in whic h hunter-gatherers are perceived to inhabit landscapes forms the basis of a chronol ogical and ontological chasm between those who subsist by hunting and gathering and those who do not (Ingold 2000a:56; Pluciennik 2002). As argued by Ingold (1995:67) foragers are perceived as collecting from the environment, wit h lifeways determined ultimately by the physical


43 distribution of social or economic resour ces (e.g. Bettinger 1991; Binford 1980; Keen 2006; Kelly 1995). Any modification of the envi ronment in this view is incidental and inconsequential (Ingold 1995:68), with the structure of subs istence pursuits undermining socially significant long-term engagements in place (e.g., Meillassoux 1972:99). Non-foragers, to the contrary, produce their food and surroundings through domestication and landscape appropriation as manifested through the construction of buildings, monuments, and other features. Or, as Bradley (2003a:xvi) has put it, the productive labors of farmers in modifying the environment for either mundane or ritual facilities are often considered to result fr om the strategic invest ment of agricultural surplus. Non-forager landscape use is r outinely understood as social reproduction, through which communities construct, politicize, and inscribe their social and physical worlds. In so doing communities write and m anipulate their own histories by variously referencing, destroying, or co-opting memori es and materiality of the past (Barrett 1999; Meskell 2003). Devoid of their own agency, and without the capacity or recourse to a past (mythical or otherwise), hunter-gatherer s and the places they create still remain fixed as static entities. This simple division has been the foundation for much social theorizing. In this framework foragers remain as passive objects, subject to nat ure, whereas non-foragers emerge as active participants in their own cu ltural productions because they are not limited by the demands of hunt ing or gathering. When c oupled with evolutionary models of social change, the posited differenc es in landscape alteration are the basis for further distantiation bet ween foragers and others. Alt hough many researchers allow that present-day hunter-gather ers do not represent evolutionary hold-overs in and of


44 themselves (Lee 1992), the notion of hunting and gathering as a way of life that is both distinctive and inherently more ancient (read: primitive) persists. This dichotomy of social types along primarily economic and ch ronological axes has deep antiquity within social thought (Pluciennik 2001). As early as the seventeenth century, social theorists postulated the notion of mobile and property -less peoples as the antithesis to thenemerging economies and civil society based on property ownership, capital, and progress (Barnard 2004; Pluciennik 2002). By the late nineteenth century, social scientists such as Morgan (1877:6) and others began propping up and justifying stadial evolutionary schemes. Emerging ethnographic observations and ar chaeological research positioned nonwestern societies as spatial and temporal others (Fabian 1983:144; Pluciennik 2002). Shell matrix sites were situated prominently in the construction of categories of simplicity and savagery at the ti me. One of the most influent ial volumes in this line of thought was Lubbocks (1865) Pre-historic Times whose overtly racist sentiments found a home amongst many western readers (Kehoe 1998:58). Lubbocks method was to link archaeological finds with traveler accounts of nonwestern people, and in doing so he was able to assert that contemporary communities were still living in the Stone Age (1865:3). Amongst the mix he drew on ethnographic observations of Fuegians (inhabitants of Tierra del Fuego) to anima te accounts of shell mounds recently discovered in Denmark. His primary sources were Charle s Darwin (1860) and Robert Fitz-Roy (1839), captain of the Beagle (Darwi ns vessel). The shellfishing Fuegians provided the perfect other for Lubbock (1865:556): If not the lowest, the Fuegians cert ainly appear to be among the most miserable specimens of the human ra ce, and the habits of this people are


45 of especial interest from their probabl e similarity to those of the ancient Danish shellmound builders, who, however were in some respects rather more advanced, being acquainted with the art of making pottery Not surprisingly hunter-gatherers were left on the bottom rung as savages, with no land, no law, and arguably no society. While anthropology has distanced itself from the racism and ethnocentrism of nineteenth-century evolutionary thought, the division between forager and farmer is still reproduced in method and theory today (Bradley 2003a; Strassburg 2003; Zvelebil 2003; Zvelebil and Jordan 1999). This dich otomy has implications for how we understand past social organization and change amongst hunter-gather er communities. Theories and models that privilege biological process such as evolutionary ecology or neo-Darwinian theory are widely used to ex amine hunter-gatherer social variation because foragers are interpreted as somehow more connected to and subjects of independent environmental process (Chatters and Prentiss 2005; Kelly 1995; Winterhalder 2001; Winterhalder and Smith 1992) As a result, all hunter-gatherer actions are reduced to adaptations or cost-benef it solutions (Bettinger 1999; Hawkes et al. 1982; McGuire and Hildebrandt 2005). Change through time equates to an independent process of evol ution, in which past communities were not active participants (Ingold 2000b). The reproduction of hunter-gatherer as a socially-simple and hi storically ignorant category can most readily be seen in contem porary discussions on the significance of mounds, monuments or other mate rially impressive edifices. Indeed, it is current orthodoxy that monumental construction is incompatible with hunt ing and gathering. Over fifty years ago V. Gordon Childe coul d write that, along with agriculture, the emergence of monumental construction was one of the major events in the emergence


46 of civilization (i.e. complex societies) (Sherratt 1990:148). More recently, Kchler (1999:53) argued that a culture without monuments appears to us like a ship lost to the seaOnly hunter gatherer societies, lacking institutionalized author ity, are genuinely bereft of the need to memorialize the lives of their dead. Burial mounds and other monuments are ostensibly repr esentative of labor coordination, territoriality, and attendant power structures (Dunham 1999; Gibson and Carr 2004a; Shennan 1983; Trigger 1990). As often massi ve constructions or accumula tions of material, they are conspicuous evidence of modification of the environment (Denevan 1992:377). They also presuppose complex cosmologi es and senses of history. These impressions of the social signific ance of monuments stand in opposition to the abundance of ethnographicall y-documented hunter-gatherer societies who did not engage in monumental construction, and who we re characterized by low population densities, egalitarian social relations, and settlement im permanence (Kelly 1995:15; Lee and Daly 1999). The emergence of monumental construction is presumed to represent a fundamental shift away from foraging towards food production, sedentism, high population densities, and inevitable social complexity (Sherratt 1990; Trigger 1990; Yoffee 1993). Only societies who have met t hese preconditions for social complexity would have the labor, central authority, and worldview to construct monuments. Any society lacking these preconditions (i.e. hunter-gatherers) would be incapable of monumental construction, and any massive edi fices attributable to hunter-gatherers must represent fundamentally di fferent processes (cf. Sassa man [2004a:254]). In this frame of reference, it is easy to see how simplicity and noneventful status could be imposed upon shel lfishers.


47 Monumental Discord As with any grand narrative, however, there ar e not-so-subtle contradictions to the monumental axiom. Just as Wyman was faced with reconciling the natural and anthropogenic components of s hell mounds, we are faced with emerging data that demonstrates hunter-gatherers were quite capable of inscribing landscapes for apparently non-functional purpose s. Consider, for example, the recent discovery of the Pre-Pottery Neolithic (PPN) Gbekli Tepe site in southeast Turkey. As described by Schmidt (2000:46), Gbekli Tepe is composed of artificial earthen knolls, larger than, and in an unusual topographic setting for, other Neolithic sites in the region. What distinguishes the site furt her are the cellar-like arcuat e depressions constructed in antiquity by first excavating large holes in the ground and then lining them with masonry walls and terrazzo floors. These walls supported and framed large T-topped stone pillars, measuring between 3 and 5 m in height, and weighing upwards of 10 tons. Carved in relief on many pillars are a host of animals, including snakes, boars, leopards, aurochs and other wild animals (Peters and Schmidt 2004). Lacking at the site are objects or features typically associated with da ily living. In fact the excavators argue that there is very little evidence fo r any day-to-day activity on site. What sets Gbekli Tepe apar t the most, however, is the lack of domesticated fauna or flora associated with the earliest construction events. Dating the site has been problematic, and some of the f ill within the structures and possibly one or two structures are associated with later PPN communities who had access to domesticated species. Yet the current data indicate that the site was erected 12,000 years ago by regional hunter-gatherer communities (Scham 2008). Because of its early age and association with hunter-gatherers, the excavators conservatively suggest the site served a


48 ceremonial purpose possibly associated with hunt ing rituals. In the popular press the site has gained celebrity as the worlds firs t temple (Scham 2008), its construction by chronologically early hunt er-gatherers made even more significant because of its location alongside the nominal cradle of civilization. Yet another contradiction is the Povert y Point site, situated along the Bayou Maon in northeast Louisiana. The site dat es towards the end of the Archaic period, and was constructed between roughly 3600 years ago by hunter-gatherer communities. What makes Poverty Point so enigmatically prec ocious (Gibson 1996; Smith 1986) is its scale and implications fo r labor organization and worldview. At the time of European contact, Poverty Poin t was second only to the much later Mississippian-period Cahokia co mplex (in the American Bottom region of Illinois) in terms of both areal extent and earth moving in pre-contact North America (Kidder et al. 2008:9). Unlike Cahokia, whic h was orchestrated by chiefly elite and underwritten by agricultural production, Poverty Point emerged from the efforts of seasonally mobile communities (Jackson 1991). The main feat ures of the comple x include six nested earthen ridges that range in length from 600 to 1200 m, the 22-m high earthen Mound A and associated platform mound, and other feat ures (Gibson 1996; Kidder 2002). Dates on the sequence of construction across the entire expanse of the site are still forthcoming. Yet early indications are that Mound A went up rapidly, and Kidder et al. (2008) have recently argued that the 10-m earthen platform of Mound A was constructed in less than 3 months. The staged construction of this particular mound appears to them to represent the material telling of a cosmogonic narrative. A consideration of the scale of imported material culture s uggests that like Gbekli Tepe,


49 Poverty Point was at the cent er of a wide network of far-flung communities (Sassaman 2005b). Material culture from the site is diverse, and indicates that either large quantities of objects or materi als were being imported and purposefully discarded at the site from as far away as Florida and the Great Lakes region (Gibson 1996). In contrast to these cosmopolitan assemblages, the re sident population appears to have been quite small. The implication is that communiti es from far away were involved in the construction of Poverty Point (Sassaman 2005b). Places such as Gbekli Tepe and Po verty Point are poignant for their singularity (Kidder et al. 2008; Sassaman 2005b). They at once draw our attention to moments in which histories were materialized and reproduced through creative acts by past huntergatherer communities in vastly different ti mes, environments, and social contexts. At the time of their creation, arguably no other places on the regional landscape encapsulated so much symbolic capital. They also underscore the fallacy of the forager/non-forager dichotomy as regards the construction of massive facilities and intentionality. Clearly l andscape modification was not outside the worldview, labor organization, or subsistence economies of those who engaged in practices there. These places indicate that seasonally mob ile hunter-gatherer soci eties could amass the labor necessary to construct mounds, thus decoupling mound construction from sedentism, food production, or stratifi ed society (Gibson and Carr 2004a). Because of their singularity, however, it would be easy to treat Poverty Point and Gbekli Tepe as anomalous, interesting but ulti mately inconsequent ial in the broader scheme of regional histories and social devel opments through time. This is particularly the case given the lack of consonance between the observed levels of social complexity


50 and the scale of the effort in each case. Yet from a social-evolutionary perspective neither site was the necessary catalyst for in creasingly complex societies. In the case of Gbekli Tepe, it preceded the emergence of monumental construction elsewhere in the region by many millennia. In fact, afte r its construction the si te was intentionally filled in and obscured by later occupants, lik ely through visible and public acts. Poverty Point was abandoned as well, possibly in connection with a major avulsion in the associated river system (Kidder 2008). The hi storical significance of Poverty Points collapse and its associated so cial networks is unknown, although it may have heralded in a new age of more localized politics and interaction at the onset of the Woodland period in the Southeast (Sassaman 2010). Ea ch reminds us that past hunter-gatherer communities did materialize their worldv iews in unanticipated ways that are not reducible to rational choice alone, but ar e outside the expectations of social theory today. Recent data, however, from the Southeastern United St ates, including Florida, indicates that Gbekli Tepe and Poverty Poin t are not alone. Since the early 1990s, evidence that mounds of either earth or shell date to t he Archaic period has emerged (Gibson and Carr 2004b; Russo 1994a, b, 1996a). The most widely recognized amongst these are the mound complexes of the Lower Mississippi River Valley (Saunders and Allen 1995; Saunders et al. 2000; Saunders et al. 2005). These mound centers preceded Poverty Point, and were li kely referenced in its final design (Sassaman 2005b). They are characterized by earthen mounds, many of which have numerous construction phases and conform to shared geometric principles (Clark 2004; Sassaman and Heckenberger 2004). Another well documented mound tradition


51 emerged along the western Gulf of Mexico and South Atlantic coasts, where Russo and others have identified numerous Late Archaic shell rings (Russo 1996a; Russo and Heide 2002). These sites are frequently composed of massive lenses of shell arranged in an arcuate fashion surrounding a cleaned central plaza. Shell mounds of the St. Johns arguably present the most compelling case of construction, if only because there is a cl ear link between inhumation of the deceased and sand (Sassaman 2004a). As detailed a bove, much of the evidence for the intentional construction of shell mounds was identified by Wy man and C.B. Moore over a century ago, but was dismissed as anom alous. Most notable are places such as the Thornhill Lake Complex, which Endonino (2008) has dem onstrated was built in part as a mortuary mound towards the end of the Mount Taylor period. The chronology, sequence of construction, and presence of bannerstones is cons istent with what Douglass (1882) and Piatek (1994) documented at the preceramic Tomoka Mound complex adjacent to the At lantic Coast. Aten (1999) reviewed Jahn and Bullens interpretation of the Harris Creek mortuary, and found that it is ent irely a preceramic Archaic facility which may date as early as 7000 years ago. Using Moores excavation notes, Beasley (2008) has documented that Ha rris Creek was not alone. Palmer-Taylor and Orange Mound in the Upper S t. Johns share similar sequences and layouts. Shell mound inhabitation would appea r to be more complex than typically portrayed. Stratigraphic testing in Archaic period expos ures routinely encounters shellfish remains lacking the expected output of residential activities (Sassaman 2003a, 2005a; Wyman 1875). As Nelson (1918) suggested long ag o, other contexts appear to have been deposited rapidly, countering arguments for their gradual, c ontinuous accumulation.


52 Excavations also suggest that preexisting she ll matrix sites were actively mined for use as mound construction fill, in many cases to create summit platfo rms upon shell mounds (Piatek 1994; Sassaman 2005a; S ears 1960). Collectively, su ch observations indicate that shell matrix sites cannot be simply reduced to refuse disposal alone. The St. Johns shell mounds and the comm unities who participated in their construction present marked contradictions to established axioms and categories. They join an ever-accumulating list of chronologically early hunter-gatherer societies who engaged in practices and reproduced institutions not anticipated by cultural evolutionary models. What are we to make of thes e apparently precocious societies? One possibility would be to simply downstream accepted monumental significance, and test whether presumed attendant dimensions of social hierarchy (large populations, economic inequalities, permanent leadership, etc.) are mate rially manifested among Archaic communities (e.g., J. Saunders 2004) Taking cues from such thought, Archaic mounds along the St. Johns and elsewhere th roughout the Southeast ern United States may be significant for the soci al relations, and potential hi erarchies, they represent (Russo 2004; Sassaman and Heckenberger 2004). The demonstrated intensive shellfishing economies could be readily in terpreted as a more-than-adequate foundation for large populations, while mounds could simp ly represent materialized contestations and politics regarding territorial rights to resources between any number of social categories (individuals, lineages, or communiti es). Russo (2004, 2008), for example, has identified recurrent variations in the height and placement of shellfish at coastal shell rings, and argues that these variations c an be used to infer differences in status between groups within a community. Or, we could follow Claassen (1996), who has


53 argued that at least some Shell Mound Arch aic sites in the Midsouth should be considered ceremonial mounds akin to late r period mortuary mounds, in this case with shell as the primary m edium of interment. Such approaches mirror recent attempts to evaluate archaeologically identified hunter-gatherer societies characterized by relative settlement permanence, constructed facilities, and social ranking. Inspiration is typically drawn from those few historically documented groups who defy traditional forager definitions, such as the Pacific Northwest coastal communities and ethnohis toric Calusa of southwest Florida (Marquardt 1985, 1988; Renouf 1984). In archaeological contexts, such nonegalitarian yet non-stratified socialities are routinely conceptualized as emergently complex, transegalitarian, and even affluent (Arnold 19 95; Hayden 1994; Price and Brown 1985; Sasaki 1981). Complexity is reduced to the degree to which enduring economic inequalities can arise through economic intensification, such as in the context of varying resource abundance, wherein soci al trajectories are preordained by ecology or technological innovation (Arnold 1995; Hayden 1994, 1995; Keen 2006). Prestigebased power structures have also been advan ced as transformative, yet tenuous, alternatives to social complexification ( Clark and Blake 1994; Hayden 1995). Current models invoking emergent complexity attempt to resolve the cont radiction of huntergatherers and monumental construction by situat ing such communities in a liminal state between simplicity and complexity. Barring ecological catastr ophes or truncation through contact with technologicallyadvanced others, early complex foragers are significant principally because they provided the foundation for later developments. Simplicity is thus preserved for the sake of complexity.


54 Despite the logic surrounding a model of intensification, we can ask whether conceptualizing complexity amongst hunter -gatherers along purely economic lines obscures significant social processes. As discussed by Flanagan (1989), egalitarian social relations presumed to be original to the human condition, and ubiquitous amongst most ethnographically document ed small-scale hunter-gather er societies, are also present within states. Such relations ar e perhaps better understood as ideologies of autonomy or as traditions of resistance (S assaman 2001), and have more to do with the historical relationships and contestations between communities than a natural condition characteristic of early societies. In the Southeast, the link between mounds as emblematic of complex societies has met with resistance, precisely because the scale of mound construction is often incongruous with expectations of economically differentiated individuals. Some such as Hamilton (1999), have argued that Middle Archaic mounds can simply be explained as wasteful behavior, which was selected for (in Neo-Darwinian terms) in a fluctuating environment. In th is case, mound construction diverted efforts away from the reproduction of offspring, and ensured that population pressure would not undermine continued biological success of mound building populations. More frequently, however, Archaic societies are shown to not match standard criteria for social comp lexity. In this case, prior simplicities are presumed as a standard against which complexities can be tested (e.g., J. Saunders 2004; White 2004). The emergence of shellfishing may also not be as transformative as has been portrayed in the literature. Bailey and M ilner (2002) have recently reviewed the divergent interpretations and archaeological contradictions relating to the Holocene


55 emergence of coastal and riverine settlements. They note that these positions vary primarily as to whether aquatic resources should be modeled as either first-line food sources that would have pu lled populations into a region, or secondary food sources targeted in times of stress. These are, in Erlandsons (1994:273) words, the Garden of Eden and Gates of Hell models. Each position presumes that post-Pleistocene archaeological records accurately reflect the distribution of past social practices. In fact, the use of aquatic resources before the Holocene is likely considerably underestimated (Bailey and Craighead 2003; Erlandson 2001; Rowley-Conwy 2001), and the role of terrestrial resources in the past has been greatly overemphasized. Sea-level rise and isostatic landform variance have contributed to a spotty record of early coastal or riverine inhabitation. Nor is there a clear temporal or geographic trend in the use of coastal or riverine resources ( contra Binford [1990]). Thus, the social developments regarding shellfishing and wetlands use cannot simply be reduced to economics alone (Claassen 1998:229). Explanations that attempt to discr iminate between monum ents and middens (e.g., Beasley 2008; Thompson 2006) further obf uscate the issue. At the extreme end, perspectives such as Claassens require that mounds of either earth or shell had a single purpose and meaning. Not only does this recapitulate a binary opposition between sacred and secular that may only be rele vant in Western worldviews, it also reifies the typological di stinctions between unintent ional middens and purposeful monuments that have confounded interpretati ons since Wymans ti me. For example, Milner and Jefferies (1998) attempted to c ounter Claassens mortuary mound view in their analysis of the Read site in Kentucky. They cited the presence of abundant stone


56 tools, fire cracked rock, and other debris consistent with de facto refuse disposal as evidence that it was a midden in the trues t sense of the word. Although recognizing that burials occurred prefer entially on the valley-facing si de of the mound, they argued that because of the long time span across which burials occurred, and the fact that many burials were not made within shell, that there was no intentionality behind the burial patterns. As Sassaman (2004a) notes Milner and Jefferies had to ignore nonrandom patterning in the location of interments to make their case. Hunter-gather communities were not all that orthodox thought have lead us to presume, and engaged in practice s not predicted by theory. A review of Archaic shell mound research indicates that instead of inve stigating how such places emerged, much theorizing and research has targeted how well they fit long held expectations of simplicity. Our explanator y frameworks are mired in the same dialogue with the continued use of gradualist metaphors for hunter-gatherer landscape use and social change. The current discussion regardi ng the archaeological delineation of monuments versus middens is a prime example (e.g., Marquardt and Watson 2005b). Lacking appropriate frameworks, we ar e left reproducing the very stereotypes and misconceptions clearly in need of being overcome. Chapter Overview The deficiencies of either a monument al or noneventful approach to Archaic histories leads to a new way of thinking abou t shell mounds. In Chapter 2 I review two different paradigms for investigating the emergence and long-term histories of shell mound dwelling communities. T he first category is referr ed to as mapping, which traces its most direct lineage to Lewis Binf ord (1980). Although the primary frame of reference for contemporary studies, mapping approaches have essentialized the


57 relationship between places, subsistence practices, and past social change. The second is the concept of landscape inhabitati on as history. As articulated by John Barrett (1999, 2000, 2001) an archaeology of inhabitation emphasizes how community histories emerge as an ongoing process in which contemporary worldviews are reproduced and transformed in the course of the mundane and the extraordinary. The past, as written in modifications of the env ironment, provides a frame of reference for these actions. I develop a framework and ex pectations of these concepts to use in examining shell matrix sites. In particular the practical act of creating places through the deposition of shellfish and other materials provides an access window into Archaic history-making. I suggest that we can trac e how communities continually reconstructed places with significance by examining t he location, organization, social context, and tempo of depositional events. Thus, we move from simply testing whether sites were monuments or middens, and move towards ex amining how different communities emerged through time, and how social memo ries and histories were produced through actions therein. Interpreting shell matrix sites as inhabited requires detail ed knowledge of the ecological and social contexts surrounding their creation and reproduction. The St. Johns has been allotted a significant role in the long-term development of Mount Taylor communities. To be sure, the river figured pr ominently in not only providing subsistence resources, but by defining the temporality and viability of inhabitation in any one physical location. In Chapter 3, I explore the varying material conditions of inhabitation during the Holocene epoch, and focus on the many different scales and tempos of resources that would have provided frames of reference for Mount Taylor communities.


58 Contrary to traditional thought, the earliest shell matrix site inhabitation occurred at a time characterized by significant ecological and fluvial variability, while successive generations experienced changing rhythms in re source structure a nd distribution. In Chapter 4, I consider the genealogies of Mount Taylor traditions as they were reproduced and transformed within the context of a variable world. Although the Mount Taylor period is defined on the basis of objects and patterns within the St. Johns, I consider how Mount Taylor lifeways can only be understood by examining multiple scales of interaction and movement that inhabitants engaged in across the Florida Peninsula and beyond. In this sense, I reframe the St. Johns as a domain through which social interactions and identities coul d be asserted and contested. I focus on the available evidence for settlement and mobility, exchange, subsistence, and mortuary practices. Although many traditions continued unabated throughout the Mount Taylor period, social interaction at local and regi onal scales was a significant factor in the emergence and changing signif icance of places. The discussion of the contexts surr ounding Mount Taylor inhabitation turns towards shell matrix sites themselves in Ch apter 5. Although I may have vilified Wyman for his pragmatic impositions, he is one of th e heroes of Florida archaeology. Much of our knowledge of Mount Taylor places is based upon the observations of Wyman, Moore, and others who saw them before they were destroyed. Complicating matters is the output of millennia of la ter inhabitants who have obscured Mount Taylor places. These data have never been properly synthesized, and as a result we know very little about the specific location, a rrangements, or chronology of Mount Taylor shell matrix sites. Drawing on prior observations, hist oric aerial photographs, recent excavations,


59 and recently acquired high resolution LiDAR data, I develop a GIS spatial database to reconstruct the configuration and location of many Mount Taylor places. Using the hydrological model developed in Chapter 3, I also evaluate the relationship between shell mound inhabitation and ri sing surface water levels. It is in detailed life histories of Mount Taylor places, as observed through depositional strategies, that we can trace how communities emerged and were reproduced through time. Unfortunately, ther e have been few attempts to document the variability of Mount Taylor depos itional practices. In orde r to examine the details of Mount Taylor placemaking, I engaged in a series of excavations at mounded and nonmounded Mount Taylor shell matrix sites. T he technical details of these excavations have been published as reports (i.e., R andall and Sassaman 2005, Randall 2007, Sassaman et al. 2010). A synthesis of these efforts, coupled with observations made at other Archaic shell matrix sites in the regi on, forms the core of Chapters 6 and 7. Chapter 6 presents the results of strati graphic testing of the Hontoon Dead Creek Complex. This site is composed of a la rge Mount Taylor ridge and associated shell field. Excavations into the mound were reported by Sassaman (2005a), and I save a discussion of those results for Ch apter 7. The shell field is known as the Hontoon Dead Creek Village, and was targeted because it has not been comp romised by millennia of later shell deposits. Close-interval co ring and topographic mapping was used to delineate discrete shell matrix heaps. Those of Mount Taylor age provide a minimalist viewshed into what constitutes the arrangement and composition of shell matrix sites. The local hydrological context of the comple x is reconstructed through the examination of sediments extracted in the adjacent swam p. Collectively, the results of testing


60 indicate that there were long-term tradi tions regarding the proper maintenance and reproduction of places. These traditions belie considerable change and transformation at the regional scale, and suggest that re sidential settlements provided a model of history and place-making along the St. Johns in microcosm. The patterns identified at the Hontoon Dead Creek Village provide a framework for examining the biographies of shell mounds distributed thr oughout the study region, and which span the length of the Mount Taylor peri od. In Chapter 7 I trace the history of transformations at the Hontoon Dead Creek Mound, Live Oak Mound, Silver Glen Springs Complex, Bluffton, the Harris Creek Complex, and the Hontoon Island North site using published descriptions and recent investigations. Although each place had a complex history of abandonment and reuse, acts of capping a place of prior inhabitation and replacing it with a new kind of soci al domain was a dominant theme running throughout the Mount Taylor period. Places of prior residence were frequently converted into mortuary mounds. Even shel l mounds that were places of residence underwent complex transformations as communities were reworked. Finally, in Chapter 8 I present Archaic community histories as seen through patterns of landscape inhabitati on. I focus on major transf ormations evident in the content, scale, and organization of depositional practices that include the construction of mortuary mounds, the replacement of smallscale settlements with platform mounds for the living, and the reorganiza tion of mound complexes. These practices intersect complex ecological processes and increasing scales of social connectivity. Such events and trends no doubt shaped the contours of structural reproduction, yet Archaic communities routinely mobilized the past as a means of accommodating or interfering


61 with such change. These transformations in traditional practices resulted in a continuous reworking of relationships bet ween people and places, the final outcome of which are the shell matrix sites we encounter today.


62 Table 1-1. Culture-hist ory of northeast Florida Period Years 14 C B.P. Years cal B.P. Paleoindian 11, 500,000 13,500,700 Early Archaic 10,000 11,700 Mount Taylor 6200 7300 Orange 4100 4600 St. Johns I 3500 3600 St. Johns II 1250 1250


63 Figure 1-1. Preceramic Archai c Mount Taylor period shell ma trix sites within the Middle St. Johns River Valley in Northeast Florida, with emphasis on those sites discussed in Chapter 1.


64 Figure 1-2. A naturally eroded shell mound that Jeffries Wyman would have observed, in this case the outer shell ridge of the Hontoon Island North site (8VO202). Photograph titled Shell heap deposits at Hontoon Island attributed to C.B. Moore, dated 4/18/1893. C ourtesy of the State Arch ives of Florida.


65 CHAPTER 2 HUNTER-GATHERERS, LANDSCAPES, AND HISTORIES Mount Taylor shell matrix sites were reduced in conventional thought to prehistoric nonevents on the basis of their composition, early age, transient occupation, and produc tion by hunter-gatherer communities. Despite the normative model that has been developed, data indicative of contrast ing patterns of continuity and change have emerged. Long-term patterns of c ontinuity can be seen in the repeated occupation of places, as well as the exploi tation of shellfish and aquatic resources. Yet short term and radical transformations are also eviden t. New practices were innovated in the context of environmental and social change, in cluding the initiation of shellfishing and the production of mortuary mounds. As arg ued in Chapter 1, these patterns suggest that Mount Taylor lifeways were not unchanging, and that shellfish collection and deposition cannot be disregarde d as nothing more than unreflexive behaviors. In this chapter, I examine how the r epeated use of places, materialized through mundane practices such as shellfishing, are inextricably bound with larger scale social processes. In order to refr ame the practices of Mount Tayl or communities, however, it is necessary to first consi der in more detail how hunte r-gatherers are projected in archaeological thought today. C ontemporary interpretations of hunter-gatherer lifeways are characterized by what I called the m apping perspective. In this view, huntergatherer settlement systems and social ch ange are modeled as dependent variables, and subject to evolutionary processes. This is the orthodox framework for explaining hunter-gatherer lifeways, and it s products are the structural histories I discussed in Chapter 1.


66 An alternative perspective that I will develop in this chapter I have labeled landscape inhabitation as histor y. In this perspective, soci al histories are constructed by inhabiting landscapes filled with significanc e. While I will explore history and historymaking throughout the latter portions of this chapt er, it is important t hat I state up front what I mean by history. One basic definit ion of history would simply be what happened in the past. As I illustrated with referenc e to nonevents in Chapter 1, however, the significance of what happened (or did not happen) depends on with who is talking about it. As Boone (2000:13) has argued, the past bec omes history when it is organized. If history is not what happened, then what is it? Noting the ambivalence with which the te rm history can be employed, Trouillot (1995:2) suggested that histor y can best be understood along two axes, a position that I follow here. On one hand, humans participate as actors in history. Social actors reproduce practices learned in collective social contexts, and creatively generate new traditions as an ongoing and contingent process (following Pauketat 2001a,b). Actors also modify their surroundings, leaving traces of past prac tices. On the other hand, humans participate in history-ma king through the construction of narratives. It is in the talking about, mythologizing, or referencing prior events (real or imagined) that the past is continually remade as collective social memories. Yet these narratives, and the elements of the past that are selected fo r telling, are informed by the cultural taxonomies and historicity (or historical cons ciousness) of the narrator(s). These two axes (sociohistorical process and narrative pr oduction) are recursively linked, such that each is structured by the other (Trouillot 1995:29). In consi dering history, then, I will be referencing both the contingency of co mmunity reproduction through practice and


67 tradition, as well as the historicities that make those practices and their consequences meaningful. Mapping In orthodox anthropological thought, the relationship among hunter-gatherer society, change, and space falls under the mapping perspective. At the core of this perspective is an assertion that hunter-gather ers are defi ned first and foremost by the distribution and availability of re sources. Central here is the belief in a direct, causal association between environment, the organi zation of subsistence practices, and population dynamics. There is a formal relationship between the physical distribution of economic resources and social resources that is quantifiable. Furthermore, this relationship is regarded as a determinant factor of social process and variability. The relationship between hunter-gather ers and space was formalized by Julian Steward, first in his salvage ethnographi es of the Numic Shoshone and Paiute communities of the Great Basin (S teward 1933, 1938), and later in his Theory of Culture Change (Steward 1955). Steward developed t he dual concepts of cultural ecology and multilinear evolution as a m eans of transcending what he saw as the deficiencies of unilineal social evolution of the nineteenth century and the particularistic approach of culture-historians in the early twentieth cent ury. Steward held that variations in trends towards social complexity between societ ies were primarily structured by local environments. The crucial step for Steward (1955:37) was defining the culture core the constellation of features which are most closely related to subsistence activities and economic arrangements, notably technology and settlement patterns. Although religion and belief systems could be consi dered part of the core, this was only the case if they were empirically demonstrated to be functiona lly related to food getting procedures.


68 Steward envisioned a methodology which emphasized tiered analyse s of environment and exploitative patterns, the relationship between food-getting and behavior patterns, and finally between behavior and other aspects of the culture co re (1955:40). These methods would enable the determination of different levels of sociocultural integration, Stewards basis for making cross-cultural analyses. Levels of sociocultural variation differed in the relationshi p between food distribution, economic interdependence, and cooperat ion between family and other kin groups. As an example, Steward discusses the Western Shoshone as at the family level of sociocultural integration; they were characterized by disparate nuclear families who were integrated at the regiona l level only during times of resource abundance, but even then there was little cooperati on or need for coordination ( 1955:114). That is, food was so scarce that small family units were forced to move to areas where food was available. As there was no functional need for ceremonialism to integrate families, interaction and communal performances were important mostly for recreational purposes (1955:114). According to Steward, it was only when white settlers provided both horses and resources to be exploited (i.e. raided) that higher levels of sociocultural integration were possibl e amongst the Western Shoshone (1955:121). Increasing levels of sociocultural in tegration include the patrilineal band, composite band, and clan. Variations between levels were inherently caused by ecological phenomena, such as whether food stuffs were migratory or scattered, population density, and large-scale interaction. Steward allow ed for historical causes in bringing more people to a region, such that the evolution of a patrilineal band to a composite band could be instigated by population movements. The role of religion, in


69 the sense of rituals, commemorative cere monies, and public performances were, in his words, independent variables due to thei r diversity and apparent lack of consonance with ecological phenomena (Steward 1955:141). While Steward injected studies of hunt er-gatherer diversity and space into anthropology, it was Lewis Binf ord to whom we can directly attribute the systematics of the mapping perspective. The term m apping was canonized in Binfords (1980) foragercollector subsistence-settlement pattern model. Using ethnographic data to develop middle-range theories, Binford inferred tw o alternative strat egies of settlement organization based on the different ial distribution of resources in time and space. In this model, collector strategies are expected in areas where there are significant incongruities (his words) in the timing and a bundance of resources. Collector strategies are typified by logistical settlement organi zation, wherein long-term settlements in centralized locations are provisioned by task gr oups. In contrast, fo rager strategies are expected in areas where resource patches ar e more evenly distributed within the local environment but are nonetheless s patially discrete. In this strategy resource patches varying in space-time are mapped onto (B inford 1980:10) through mobility and group membership flux. To sum up each alternative, foraging strat egies bring consumers to the food, while collector strategies bring food to the consumers. For each strategy, Binford posited ways in which the location, structure, and di stribution of sites on the landscape could be expected to vary. In particu lar, differences in assemblage composition are used to define different functional site types. Because he presented expectations for their archaeological correlates, the forager-collect or dichotomy has been widely exploited by


70 archaeologists. Finally, Binfor d attempted to correlate these different strategies with changes in effective temperature, such that collector strategi es typified groups living in regions with short growing seasons in higher latitudes, while forager strategies would be more typical of tropical foragers. Al though Binfords scheme may have been the most general, and thus the most wi dely used, there have been many similar turns of phrase. These include tethering [Taylor 1964 cited in Binford (1980)], residentially constrained mobility (Graham and Roberts 1986), central place wandering (Beardsley et al. 1955), traveler versus processor (Bettinger 1999) and even territorial defense just to name a few. We could also add divisions between simple and complex foragers based on the degree of residential permanence. Binford was primarily interested in det ailing process within a stable system, as such the forager-collector continuum was nev er intended in his original work as an evolutionary model (Binford and Johnson 2002) In a series of papers, however, he outlined related thoughts which have had sign ificant impact on how archaeologists have approached shell matrix sites and their significance. Underlying each of these arguments was a model of soci ety in which culture is t he extra-somatic means of adaptation (Binford 1962:218), which he adopt ed from Leslie White (1959:8), whereby cultural systems can be modeled as tendi ng towards optimum efficiency through a process akin to natural selection. In Post-Pleistocene Adaptations Binford (1968) noted that there was an increase in riverine and coastal exploita tion after the close of the Pleistocene, apparently coeval with but spatially segregated from regions with evidence for plant and animal domestication. He ar gued that the onset of aquatic exploitation was, in part, the result of a greater trend towards sedentism in which populations


71 targeted seasonally productive but migratory resources such as waterfowl and anadromous fish. Sedentism was enabled by tec hnological innovations relating to food storage, but also had the effect of incr easing populations. As a consequence, he posited that there should be an increase in em igration of family groups out of productive localities into regions of lower environment al productivity. The net effect was that daughter groups moving out of productive zones would need to increase their technological efficiency, which included plant and animal husbandry, particularly when these daughter groups appropriated landscapes already inhabited. Population increase and sedentism were the expected and necessary out comes of the regiona l availability of aquatic resources. This line of thought presaged Binfords (1990) later argument that aquaticand terrestrial-oriented hunter-gatherer s exhibited graded differences by latitude in terms of mobility, housing permanence, and overall land-use patterns. He noted that ethnographically documented aquat ic-oriented hunter-gatherers tended to be located at higher latitudes and employed a logistical pattern, while terrestrial groups tended towards a foraging pattern. Although recogn izing that certain aspects of social complexity were present among aquatically-o riented foragers, he suggested that this was more likely in areas with low productivi ty of terrestrial resources, and highly productive aquatic biomes. He critiqued the not ion that either thick midden deposits or robust housing could be taken as evidence for increasing social complexity. To the contrary, he suggested that because foragers could not move members to the direct source of aquatic biomes (i.e. they could not live directly where fish and other resources were available), that we should expect thi ck middens to occur at access points. Thick


72 midden sequences, in his view, represent resi dential tethering to habitable areas that are in closest proximity to exploitable aquat ic resources. Redundant site use follows preferentially, regardless of whether or not groups were sedentary or seasonally mobile. In the literature on Floridas sh ell mounds, Binford is rarely ci ted directly. However, his influence can be seen in the widely held interpretation of shell mounds as proxy for past environmental productivity and redundant unreflexive deposition of food remains (see Chapters 1 and 4). A related strain of thought within hunt er-gatherer studies is behavioral or evolutionary ecology (Kelly 1995:50; Wi nterhalder and Smith 1992). Evolutionary ecologys central premise is that human acti ons are subject to natural selection, and that the diversity of human behavior thr ough time is the product of differential reproductive success at the level of the i ndividual. Underlying the basis for this framework are the dual concept s of methodological individualism and optimality (Kelly 1995:53). Evolutionary ecology assumes that i ndividuals will choose to act in ways that maximize the likelihood of meeting a se t goal. Secondly, behaviors will be oriented towards optimizing their reproductive fitness relative to a currency, such as caloric intake or prestige (e.g., McGuire and Hildebrandt 2005). Individuals will attempt to either maximize their currency, or at the very least, mini mize risk. At the heart of evolutionary ecology is a suit e of models initially borrow ed from ecology to describe nonhuman foragers. These incl ude the diet breadth, patch-choice, and central place foraging models. In each case, mathematical models are used to predict at what point a behavior will either be engaged in or avoided, given a set of optimizing priorities.


73 In a mapping perspective, then, hunte r-gatherer lifeways are reduced to an algebra in which space and time are independent variables. The diversity inherent in hunter-gatherer societies can be modeled as an equation, whereby time = temporal availability of social and economic resource s, and space = those resources distributed across physical landscapes. To solve a par ticular question regarding hunter-gatherer diversity, one simply needs to plug in time and space. Patterns of site location, architecture, social interaction, and even the construction of ceremonial facilities follow dependently (Binford 1980, 1982, 1990; Whallon 2006; Whitelaw 1983). That is, the arrangements of physical settlements in this vi ew take their particular form due to the physical constraints of mobility and settlement as directed by environmental processes. Mapping Away History Although a pervasive framework for examining hunter-gatherer diversity and social change, we can be critical of mappi ng on a number of empirical and theoretical points. Wiessner (1982), for example, immediately criticized Binfords collector-forager model for not taking into account social interacti on, and suggested t hat the distribution of sites and their assemblage structure was as much a product of reciprocity and risk reduction amongst communities as it was environm ental organization. Soon after optimal foraging theory was put into practice, cont radictions to the optimizing assumption emerged in fieldwork. Dwyer (1985) noted how the prac tical act of hunting was inexorably linked to a web of socially cons tituted logic and a host of other actors. Others (e.g., Clastres 1998) documented how food-getting strategies of foragers, thought to represent ideal opt imizing behaviors (e.g., Hawkes et al. 1982) (and thus reflect a natural condition), should not be disart iculated from their ow n historical context involving contacts with more powerful others.


74 Empirical problems abound as well. As I not ed in Chapter 1, there is increasing evidence for the exploitation of aquatic resour ces worldwide. Just like other aspects of hunter-gatherer lifeways, shellfish use mu st be problematized and not presumed to represent a particular evolutionary trend (Erlandson 2001). Claassen (1996; 1998:231 233) has been a vocal critic of viewing shell matr ix site locations as direct proxies for the distribution of non-mobile riverine resources such as shellfish. In her own work on shellfishing in the Southeastern United States, she has shown t hat there is limited association between the location of Archaic period shell heaps and areas targeted by commercial shell fishing industries in the hist oric era. She suggested that the presence of shellfish at archaeological sites cannot be necessarily ascribed to local contexts, and that their inclusion in sites may have mo re to do with their us e for non-food purposes (although see Morey and Crothers [1998]). At a fundamental level, however, the mapping perspective suffers from an underlying model of spatial repr esentation which is ahistorical and decontextualized. In the mapping perspective, hunter-gatherers ar e portrayed as making economic decisions with reference to a wide range of alternatives that are separated in physical space. For example, the decision to mainta in a campsite or to move to another locality is informed by expectations of food-getting success and the predictability of resources elsewhere. All places of similar economic viability have the same value, such that any given locality that shares an isomorphic ecological configuration is accorded the same status given equal transportation costs. Secondly, movement between localities is presumed to take the least-cost pathway. When viewed as change through time, prior activities at places on the landscape have no significance, unles s there has been an economic investment


75 in their long-term utility. The ongoing reproduction or change in lifeways emerges as non-recursive and unreflexive. Foragers ar e modeled as methodological individuals attempting to maximize food intake or minimize energy loss through differentially adaptive means. Conceived in this way, hunter-gatherers lit erally mapped their landscapes in reference to optimizing priori ties, much like a modern-day land surveyor would plot the geometric and finite relations hips between buildings, roads, and other landscape features. This conception of maps and mapping as an objective and rational (in a microeconomic sense) response to the physical distri bution of resources is itself a product of a particular phase of Western history (Cosgrove 1985a; de Ce rteau 1984:120-121; Thomas 1993). Cosgrove (1985a,b) has tr aced the development of a distanced, quantifiable landscape perspective to fift eenth century realist Italian Renaissance painters. Alberti, for example, formalized a method for transforming three-dimensional landscapes into two dimensional representat ions. Based in geometry, Albertis linear perspective allowed artists to produce images with the illusion of depth. By fixing a scene from an observers point in physical spac e, artists could calculate physical spatial relationships between elements in the scene with reference to distance and angle (Cosgrove 1985a:48). The geometric properties of things would later be defined in Cartesian space. The innovat ion and elaboration of linear per spectivism resulted in the production of static and quantif iable representations of real world phenomena. In addition to privileging vision over other sensory experiences, linear perspectivism maintained distance and separation between t he observer and subjects of a scene.


76 Cosgrove (1985a:55) argues that su ch landscape perspectives were fundamentally an act of appropriation and objectification. That is, elements of the landscape scene could be could be ordered, stru ctured, and controlled according to the observer. It is significant, then, that the geomet ric fundamentals of linear perspectivism were translated to cartography and quantificat ion of real world property during the transition from feudalism to capitalism and early nation states. In this historical context, maps were produced as representations of political boundarie s, which could be conquered and exploited, or as land which could be alienated and tr ansferred just as any other commodity. Tilley (1994:21) furt her suggests that Western divisions of culture and nature, and an em phasis on demarcated spaces of economic interest, are the product of increasingly expansive nations (see also Pluciennik 2001). Mapping in this sense reflects an increasing need to surveil and control resources and communities. In either case landscapes could be removed from their social context and valued and controlled according to the observer. In practice, the mapping perspective is inscribed through our own cartographic conventions that precisely locate the ma terial output of past human activity among various ecological alternatives or economic t opographies. The prioritization of what is included in a map is based on preconceive d notions of what are relevant and significance processes (Thomas 1993). This model of spatial di stributions has been referred to as map-making by Ingold ( 2000a:233), who suggests that these spatial models have been reduced from proce sses that were constantly in motion. In this view, mapping problematically reduces practices, worldviews, and experiences into totalizing and static frameworks (Chapin et al. 2005; de Certeau 1984:120; Thomas 1993;


77 Wood 1993). At the same time, mapping inserts foreign assumptions and categories upon what were once historically contingent and diverse social productions (Kchler 1993). Maps and mapping, as defined here, are inher ently ahistorical representations in which the relationships between places and peoples are held cons tant and separated by geometric distance. Implicitly, then, maps also distance past societies from experiences in, and knowledge of, their own landscapes. From the perspective of archaeological interpretation, all past agency is removed from the cont exts of historical production. Like Thomas (1993:27), I am no t suggesting that the heuristic of creating spatial models of past processes be abandoned; fa r from it. In fac t, by entitling this dissertation Remapping Histories, I am su ggesting that a rapprochement between the science of spatial representati on and the investigation of pas t, contingent histories is entirely possible. But I am suggesting the underlying model on which the mapping perspective is based is opposed to historical re construction. That is, the map, as an essentialization of diverse and dynamic proc esses, obscures the diverse practices and world views through which its production was made possible (de Certeau 1984:120; Thomas 1993). Inhabitation as History There are other ways to investigate hunter-gatherer social process. In this section, I introduce an alternative set of theories based on a different paradigm. Archaeologists have become increasingly inte rested in understanding how cultural change and stability at the micr o-scale of everyday life was reproduced and transformed through human interaction in the world at larger and longer scales (Dobres and Robb 2000; Dornan 2002; Joyce and Lopiparo 2005; Or tner 1984; Pauketat 2001b). Practice-


78 theories, initially developed out side of archaeology, such as those of Bourdieu (1977, 1990), Giddens (1979, 1984), and de Certeau (1984), provide an means to break away from normative descriptions of archaeological cultures, ec o-functionalist models that reduce the past to behaviors ordered by ulti mate causes, and ahistorical structural accounts that explain past variability in mate rial culture by reference to generalized or vague notions of social systems (Pauketat 2001b; Shennan 1993). As a body of interrelated concepts, practice theory probl ematizes how historically contingent practices and structures are recursively impl icated in social construction (Dobres and Robb 2000; Joyce and Lopiparo 2005, Ortner 1984). This is not a particularistic pursuit, as communities of varying modes of or ganization can be confronted with moments of radical environmental change, incursions of other peoples, or other unforeseen events. It is these contexts of interaction and cont ingent social reproducti on, not generalizable patterns of social evolution, in which we can make comparisons and trace alternative pathways between different communiti es (Pauketat and Emerson 2008). The framework that I find has promise for historicizin g Archaic communities is derived most immediately from Barretts (2000, 2001) archaeol ogy of inhabitation and Pauketats (2001a, b) historical process ualism, who in turn have developed concepts from the practice and phenomen ological theories of Bourdieu, de Certeau, Giddens, Heidegger (1977), Sahlins (1985), and others. The central premise of inhabitation is that communities live in, experience, and value their worlds in distinctive ways (Barrett 1999; Mills and Walker 2008). These ways of inhabiting are always in a state of becoming, and emerge from a constant interplay between people, environmental processes, material conditions, and a rec ognition of a past through social memory.


79 The goal of investigating inhabitation is not to find proximate and ultimate causes for why something may have happened in the past. Instead, the focus is on how communities were able to i nhabit a world, how they were able to effect social continuities and generate moments of fundamental change, intentionally or otherwise (Pauketat 2001b:87). Because such approaches problematize social practices as constitutive, contingent, generating, and cons training they allow us to transcend the long-held dualism between hunt er-gatherer and others (as discussed in Chapter 1), and instead focus on how past communities (irrespec tive of mode of subsistence) situated themselves in reference to a wider world. This situatedness was always framed by material conditions which were ever changing (Barrett 2000:67). I will highlight three aspects of inhabitation that have significance for shell matrix sites and history on the St. Johns: (1) the relationship between practices and communities; (2) the link between landscapes and places in social reproduction; and (3) social histories as viewed through biographies of place. Practice and Community Shell matrix sites have been dis missed as nonevents because they appear, largely on stratigraphic evidence, to have been occ upied repeatedly over long stretches of chronological time. In order to investigate in what ways such repetition could be construed as social history, it is necessary to broadly consider the role of practice in the formation of community. How is it that prac tical acts such as shellfish collection and deposition or the (re)cr eation of settlements framed communities? At the heart of an archaeology of inhabitation are models recursiv ely linking the embodied dispositions, or practices, of human agents (individuals or coll ectives) and the structures within which those practices were enacted. These frame how we can conceptualize social


80 reproduction and transformation as a continuum that is grounded in practices made with respect to material or social co nditions (Barrett 2001; Johnson 1989). Giddens structuration theor y, in particular his duality of structure in which structure and practice are recursively implicated in thei r mutual reproduction, is relevant here. For Giddens (1979:64) structure is composed of rules and resources. The relationship between these components of st ructure are critical, but are poorly defined by Giddens (Sewell 1992). Barrett (2000:65) makes the following di stinctions in which both components of structure can be reconfigured. Structural conditions are the material matrix of everyday life, the so-called givens which would include t he material results of practice, techniques and practical skills, as well as cultural logics of symbolic order. Such conditions exist and can be studied in isolation, but as Barrett notes, they have no significance without practice. Structuring principles on the other hand, are the ways through which agents act with reference to t hese conditions. These are in Sewells (1992) terms, transposable schem ata, ultimately derived from Bourdieus (1977) notion of the habitus (see below). Structuring prin ciples accord with what Pauketat (2001b) refers to as traditions, they are the ways in which agencies make structural conditions relevant. The other side of the duality of structur e is agency, defined by Giddens (1984:15) as the ability of agents to effect outcomes or the organization of the material world through practice. Broadly defined, all indi viduals, collectives, and even things have agency, the power to do. Because agents in this paradigm are relational, agency should not be equated with specific indivi duals or class of people (Barnes 2001; Gillespie 2001; Pauketat 2000). Human agents are assumed to have some knowledge


81 about both the setting and timing of conduct (Giddens 1984:90), yet it is important to analytically separate intent or rationalization and the unintended consequences of practice (Giddens 1984:10). Important ar e those moments when the deployment of stocks of knowledge were not effectiv e, or had unintended consequences (Barrett 2001:154; Sahlins 1985:152). Regar dless of motivation, all pr actices have the potential to reproduce the unacknowledged c onditions of action. Practi cal acts also transform, even in subtle ways, the material conditi ons of their deployment. They open up new possibilities through which their significanc e can be transformed (Joyce 2004; Pauketat 2000). As social principles, structure is reproduced through practi ce, and as such it only exists in a virtual state. That is, pr actices only have significance in that they are performed (Giddens 1984:25; Pauketat 2001a:10). We can extend the structuring process to examining how traditions of embodied dispositions are reproduced through Bour dieus conception of the habitus. For Bourdieu, the habitus is a system of learned and enduring practical, bodily dispositions, which are structured by the organization of material conditions of a particular collective (1977:71). As structured by those conditi ons, the habitus provides for action in the world and generates thoughts, perceptions, and meaning associated with practice (Bourdieu 1977:97). Bourdieus primary interest lay in understanding the reproduction and routinization of non-discursive, practica l forms of knowledge which are immune to self-reflection except in extreme cases (1977:94). Experience and perception are products of an individuals corporeal and sensory body being in the world, while practices generated on the basis of experience can be shared and reproduced between individuals and larger collectives (1990:54). Practices are not infinitely variable, nor are


82 they relevant only at micro-scales (Thomas 1996:97). The tempor ality of practice differentially incorporates individuals and communities through time and across space. Because individual expectations of events ar e inherently based on the actions of others, practices across space are orchestrated or coordinated without necessary institutional control or containment (Bourdieu 1977:80, 163) The organization of practices of collectives occurs because agents to a certain extent share mutual stocks of knowledge about how to act or go on in places and in in teractions with others (Giddens 1984:90). Institutionalized and routine feat ures of that organization provide for the differential spacing and timing of fields of practices and resources. The logic of practices are schemata, wh ich include taxonomies for organizing the world, and interpreting and imbuing the acts of others with significance. There are potentially infinite responses that can be generated within the historically situated structure of the habitus (Bourdieu 1990:59) However, the enactment of bodily dispositions (i.e. practice) inherently r eproduces both the meaning and structure of dispositions in the habitus, and by extens ion the relationships between individuals and collectives (Owoc 2005). Through routinization, the taxonomies through which the world is understood and their material referents become doxa, the unquestioned takenfor-granteds of social life. That is, they appear as naturalized, without history, and are generally unassailable no matter how arbitrary the distin ctions may be (Bourdieu 1977:164). Practices have context, and take their signi ficance in relationship to other doxic referents, the organization of people and things. As individuals perform practices they are situated in time and space (Bourdieu 1990:81). Practices have both a tempo and a


83 sequence, such that each practical act implies past and future actions (Ingold 2000:Chapter 11; Joyce and Lopiparo 2005). Pr actices are situated within a spatial context in the world. The tempo and situat edness of a practice is defined by its own sequence in relationship to other practices, and thus any given practice is embedded in fields or webs of relationships to other pr actices (see Barrett 2001). Together these form a system of reference, in which any given practice implicitly refers to many others (Gosden 1994:16). Enchained as they are bot h temporally and spatially, each practice gains its significance in relation to other times and places (Barrett 1999), and has the potential to reference other objectified social or natural rhythms. Practices also reference and act upon objects in the worl d, and as such people and materials are relational (Gosden 1994:82; Pauketat 2003). The significance of materials within a system of reference, or their materiality, is thus inherently interwoven with their context of deployment, as well as underlying syst ems of value (Gosden 1999:120; Mills and Walker 2008; Owoc 2005; Pauketat 2003) For example, Bourdieu (1990:250) discusses how the meanings embedded within subs istence practices are fundamentally linked with the cyclicity of the seasons as well as the practices themselves. Alternatively, the arrangement and orientation of agents with respect to others at the micro-scale, such as in the context of a settlement, can become a model for how the larger world is encoded. Because schemata are transposable (Bourdieu 1977, Sewell 1993), new meanings and strategies can be generated by deploying stocks of practical and discursive knowledge into alternative fields of practice. Such shifting practical deployments have genealogies in a sense of their accepted sequence, but also in


84 histories of use and transformation (Pauketat and Alt 2005). As a base example in a contemporary American frame of referenc e, dumping your garbage in a predetermined place such as a trash receptacle reproduces the (in)significance of the material, making reference back to the cleanliness of the house and more general views on the separation of clean and not-cl ean. At the level of a neighborhood, improper disposal may incur some intervention by concerned citizens but likely will not result in sanctions. Dumping your garbage on the steps of the Wh ite House, however, would be understood within a fundamentally different frame of reference. The significance of practices and their re ferents is not inherent in things and people, but in the intersections when practices are enacted (Robb 1998; Thomas 1996:97). As traditions or stru cturing principles they have contexts which reference other times and places. They are also a process of negotiation, subject to acceptance, modification, resistance, and accommodation (P auketat 2001a). It is in the deployment of dispositions that their significance is open for negotiation, and thus they present opportunities for reproduction and transformation, and there is no necessary guarantee of their intended success. Communities emerge as ongoing, yet histor ically contingent intersections of peoples, places, and things that are defined in part by their co-presence in the act of inhabitation, yet they are relationally contrasted and under stood with other communities (Appadurai 1996:178; Yaeger and C anuto 2000). It is in the continual reworking of enchained practices (Robb 1998) that comm unities take shape through continued interactions at local and regional scales, and are differentiated and incorporated via


85 traditions ranging from the organization of residences, to food getting strategies, to the recognition of who is, and w ho is not an other. In contrast to the subtex t of Bourdieus habitus, we should not conceive of community formation and reproduction as si mply non-discursive, and need to consider how communities discursively and actively c onstructed themselves (e.g., Barrett 2001; Basso 1996). All practices embody within t hem acts of memory by drawing links between what has happened and what should o ccur next in relation to materiality (Gosden 1994:16). Social memory, as a co llective understanding of an acknowledged past, is also central to the construction of communities (Bradley 2002; Connerton 1989; Halbwachs 1925, 1992; Van Dyke and Alco ck 2003). Following Halbwachs (1925, 1992), Connerton (1989) made an important movement to wards investigating how practices were central to the process of remembering. Conner ton suggested that individual frames of reference can be collect ivized in shared experiences. So-called commemorative ceremonies afford a domi nant perspective on how the present is a continuous extension with t he past (Connerton 1989:48). T hese are formalized, what anthropologists typically refer to as rites or rituals, and make ex plicit reference to historical and mythic founding event s and persons (Connerton 1989:61). For Connerton, commemorative practices can invo lve either incorporative acts which are habituated exchanges and interacti ons, or inscriptive acts, those which actively inscribe and mark out the material world (see also Rowlands [1993]). The latter example has implications for archaeology, as they are ma nifested in a tangible world that we can investigate (Knapp and Ashmore 1999; Van Dyke and Alcock 2003).


86 Subsequent working through the implicati ons of commemoration indicates that social memories are literally everywhere, and that the distincti on between incorporative acts and inscriptive acts are not so clear ly defined (Berliner 2005; Mills and Walker 2008; Van Dyke and Alcock 2003). Nor is the distinction between ritual and mundane necessarily discrete or analytically useful. Amongst archaeologists, ritual activities are normally equated with practices that are non-functional or appear irrational because they do not fit economic models of behavior (Brck 1999; Walker 2002). One example is the concept of structured deposit s, in which various materials were thought to be brought together to creat e a new kinds of meaning and interpretation of the world (Richards and Thomas 1984). These were contrasted with unstruc tured or incidental deposits (Hill 1995), thought to represent day-t o-day residential activities, and thus not representative of social memory constructi on or a wider sense of myth and religion. Recent treatments of structured deposition have pointed out that by definition all practices are structured, in accordance with a system of referents that likely does not accord with contemporary Western sensibi lities (Pollard 2008). Amongst communities many rituals inhere to every day life, and are not necessarily separable as events in the larger flow of inhabitation (Bradley 2003b) Brck (1999) argues that the continued separation or ritual and mundane practice needlessly recapitulates Western notions of rational and irrational, as well as secular and sacred It is thus paramount to consider how pr actices can be drawn from the everyday to generate social memories in particular c ontexts. Bell (1992: 88) suggests that we focus on how practices were strategically combined to create social memories in a process of ritualization For Bell (1992:90), ritualizatio n does not involve the enactment


87 of fundamentally different practices that ar e outside the bounds of routine activities. Instead, it is the process of emphasizin g certain practices and performing them in contrast to the routine flow of inhabitation (Gosden 1999: 129). In this view, some elements of practice and materiality are drawn together to create and recreate continuity, or rupture, with the past at multiple scales (Bell 1992; Bradley 2003b; Stahl 2008). Landscape and Place In charting out the relationship between pr actice, community, an d social memory I have left the problem of landscape on hold. However, it is through inhabiting landscapes that histories are lived and social memories are constructed. All practices are spatial and make reference to a material world, and thus traditions and communities are inseparable from the landscapes they i nhabit (Soja 1985). In a relativist paradigm, landscapes are not the im position of either social structur es or sets of meanings onto an inert physical backdrop. Instead, they are the historically generated structuring principles, the system of re ference, within and through wh ich practices and communities are spatially organized and made significant (Barrett 1999). Landscapes emerge as the relationship between places of human activity and inactivity resulting from practices enacted through time and space (Hirsch 199 5; Ingold 2000a:189; T illey 1994:34). In one sense landscape is virtual, in that it is lived, experienced, and subject to structuration. Without human agents, ther e would be not landscape. With this starting point we can begin to erode the distinction between hunter-gatherers and others based on settlement types, and consider how comm unities could inhabit the structural conditions of their time (Barrett 2000).


88 The notion of an inhabited landscape can be traced to phenomenological literature, in particular Hei deggers (1977) conception of t he dwelling perspective. For Heidegger dwelling is the proce ss of experiencing the world. As bodies are situated in space, knowledge, biographies, and experien ce in the world is necessarily spatial (Casey 1996; Thomas 1996:89). The implication is that communities do not generate cognitive maps onto a tabula rasa Instead, the perceived cons traints, significance, and potentiality within landscapes are gener ated as historical process. De Certeau (1984) and Ingold (2000) provid e useful insights for considering the complexity of landscape inhabitation and its relationship to social memory. De Certeau was interested in examining how inhabit ants appropriate and make physical spaces significant through practice. He explicitly addressed the relationship between being, practice, and the physicality of the world. De Certeau (1984:Chapter 9) argued that it is through spatial stories or narratives that inhabitants create links and pathwaysliterally and figurativelybetween elements on the land scape. Memories created in one space necessarily refer to others. De Certeau uses the image of a pedes trian walking on the street, linking different places together thro ugh movements. Stor ies may mark out the boundaries between places, define regions in which practices are interwoven, and delimit non-places through avoidance (see Munn 1996 on avoidance). These spaces, boundaries, and paths links are created not by natur al features, but by the practices and webs of significance drawn bet ween them. With referenc e to spatial stories, De Certeau makes a distinction between the m ap and the tour. For de Certeau (1984:121) the map represents how the world should be organized based on tradition (de Certeau 1984:121), but it is continually modified by exper ience. In contrast, the


89 tour, or itineraries, ar e created as spatial stories, in wh ich the biographies of individuals, as well as the significance of places, emerge as connections are drawn between them in a network. De Certeau argues that the m ap, an idealized arrangem ent of places, and itineraries, the links drawn betwe en spaces in everyday life, were mutually constitutive. Ingold (2000a:Chap 11) also recognized t hat landscapes are made possible by many acts of remembrance as generations inhabited or dwelled wi thin them. These practices, and their many expressions withi n a landscape may comp rise taskscapes, an embodied form of landscape. The primary factor in this view is that landscapes have temporalities of experience that are embedded in the many practices distributed on the landscape. Making tools, hunting animals or collecting particular plants, moving between settlements, and venerat ing the dead all have their own rhythms. Such ways of being comprise the many st ructuring principles at work (Barrett 2001). These rhythms are informed as much by the cycles of the natural world as they are interactivity between human agents. The te mporality of the landscape is found not in different rhythms per se, but where fields of practice are formed at intersections of complex interweaving of very many concurrent cycl es (Ingold 2000a:197). A ll activities that occur within the flow of inhabitation such t hat histories are wov en, along the with lifecycles of plants and animals, into the textur e of the surface itse lf (Ingold 2000a:198). In this sense, places gather meaning (follo wing Heidegger 1977) fr om the experiences of individuals in them, and from their relationships to othe r places through space (Basso 1996). As interpreted by Ingold (2000a:199) landscape is never complete: neither built nor unbuilt, it is perpetually under construction.


90 Landscape inhabitation can be also be understood in terms of economies of memory. As described by Ahearn (1995:53), de Certeau employed a broad definition of economy, not necessarily as it related to financal or material transactions, but as a way in which a society orders, manages and distributes its material and intellectual resources. Indeed, the production of m aps and itineraries verge on what de Certeau (1984: Chapter 10) called economies of writi ng. De Certeau saw the act of writing in contemporary Western society as one way in which social space was divided and through which relationships between communi ties were prescribed (Ahearne 1995:53). Likewise, spatial stories were another way of writing, in that they imbued places with social memories, and thus redirected hum an experience and history. Through the creation of maps, the systems of reference which compose landscapes are reproduced and resignified as a historical process. This is not to say that landscape use was real writing, or that landscapes can be read like a text. Instead, landscapes are deeply implicated in the process of social remembering (Kchler 1993), and they are in separable from the politics and worldviews of their inhabitants. As argued by Sant os-Granero (1998) with respect to Yanesha communities in the Amazon, elements of the landscape can be imbued with social memories through ritual performances or practices in place. In his description of Yanesha landscape inhabitation, Santos-G ranero makes a distinction between topograms and topographs. Amongs t the Yanesha, topograms are places which are thought to have been inscribed by humans or supernatural beings, and effectively stand as reminders of isolated events or persons of the past. Topogr aphs, however, are narratives that are constructed by combin ing different topograms in discourse or


91 movement. By moving between, and acting wit hin, particular places the Yanesha can reproduce or create new social memories to contest, explain contemporary concerns, or actively resist incursions by outsiders. That movement and inscription is synony mous with history-making among huntergatherer societies, is illustrated by several examples. Morphy (1995) describes how the landscape forms a complex system of referents in which time is subordinate to space. For the Yolngu people of Australia, so cial identities are deeply enmeshed as cosmogonic and biographical narratives. Kinship and clan affiliation are associated with ancestral beings whose acts of creation are experienced in places distributed across the landscape. Ceremonies are conducted to retr ace these ancestral tracks, which in turn reproduce and legitimate sacred objects and laws of the clan. In contrast, mortuary rituals for individuals trace a path through a landscape which cuts across the ancestral places, yet ties together the dead individual with the living. These individual and community biographies are enact ed as practical and discursive spatial stories that reference the movement of ancestral acto rs upon the landscape. Yolngu historicity effectively subverts social change to myth yet also provides a powerful method for creating social memories across communities. This mode of historical consciousness is typical of so-called cold societies (Chapter 1). A similar kind of historicity has been documented by Wilson (2005), who suggests that mobile Native American communities of the Great Plains understood time as a cycle related to the annual migration of target animals. Persistent places were created in which ritualized deposition of deer antler led to the accumulation of monumental piles. Adding to the piles was seen as an act of worl d renewal, replicating the first practices in


92 that place. According the historic documentation, monuments were not seen as territorial markers, nor as the agency of a pa rticular individual. Instead, they were added to by diverse communities and provid ed one avenue in which a sense of place could be shared between groups. The Siberian Khanty frequently in scribe their landscapes. Traditions of settlement and the commemoration of ancestors and nonhuman spirits form a complex web of mundane and ritualized practices that Jordan (2003) refers to as enculturation, as opposed to domestication (see also Zvelebil 2003; Zvelebil and Jordan 1999). These practices are embedded in the temporality and s patiality of mobility and subsistence. For example, sacred settlements are parsed out and modifi ed through the creation of shrines and the deposition of feasting remain s. Such sites can only be visited during certain times of the year, which roughly correspond with mobility practices through the area. The Khanty also have cemeteries, whic h are largely avoided except in times of burial. Jordan argues that these landscape features roughly correspond with kin groups, but that they overlap with others on the landscape as we ll. In this case, social memories and histories of interaction are understood in relation to traditional lands. These examples highlight how places and the paths between them generate provide a series of material referents for commemoration. Far from epiphenomenal, such traditions have a real world impact on how, when, and under what contexts huntergatherers engaged or avoided people and places. It is not without some irony that I note that different approaches towards t he origins of monumentality in Europe developed from alternatively tracing the spatial stories of hunter-gatherers and farmers. Many of the aspects of taskscapes and landscape temporality have been developed by


93 Ingold (2000a), a hunter-gatherer specialis t. One can trace in his writing the development of a line of thought attempti ng to break the bounds of the ontological divide that has dominated discussions of hunter-gatherer societies from others (as discussed in Chapter 1). Authors coming from the perspective of the European Neol ithic, however, frequently use hunter-gatherers as a foil to explain the emergence of monumental construction in the context of an incursi on of agriculturalists, or the adoption of agriculture (Bradley 1998; Hodder 1990; Thomas 1991). Bradley (1998:34), for example, bases much of his model for ear ly monumentality in Europe on a simplistic division between Mesolithic hunter-gatherer s as emphasizing continuity between humans and the animal kingdom. Such idea s have little place among farmers. This argument is based in part on two burials, a pair of Mesolithic inhumations and one Neolithic found within 2 m (hor izontally) of each other. Dr awing on the substantive differences between burials, he suggests that the Mesolithic burial was set in a natureoriented frame of reference, completely diffe rent from that of t he Neolithic burial. Unfortunately for Bradleys argument, the burials were likely coeval based on recent chronological refinements (Strassburg 2003, Meiklejohn et al. 1998). Whatever led to this juxtaposition, it does not appear that it was gradual erosion of a Mesolithic way of life with farming winning out. It is in this context of the Mesolithic/Neol ithic burial example that we can consider how particular places enable social memo ries. From a routinized perspective, inhabitation becomes meaningful when it is situated between different frames of referencethe inhabited place is known wi th reference to past experiences and by


94 actions at that place which are played off agains t a wider reality of social continuity and order (Barrett 1999:259). Similarly, Hir sch (1995) contrasted background and foreground in landscapes. The significance of places is recursively derived from a dialectic of the foregr ound of daily life (life as it is), and a background (the way the world was in the past or could be in the fu ture), each of which is constantly being reformulated. To resituate this within Barretts discourse, the deployment of structuring principles is made meaningful as juxtapos ed by the background of expectations in particular social fields. These frames of reference constitute different temporalities, and as such, moving between frames of refer ence can be considered as moving between different times. The inhabitation of settlem ents implies acts of remembrance that draw together the relationships of community members, which in turn structures how such relationships are remembered and marked out (Pollard 1999:88). Social memories, in terms of sense of place, may be expressed as a routinized renewal of the settlement, which like Wilsons (2005) example of deer antler mounds recapitulates the founding myth of the community (see also Appadurai 1996:183). Although backgrounds are a lways latent and can be referenced, certain places serve as focal points through which dominant discourses, or models of how the world should work, can be made that relate the pres ent and the past (Barrett 1999). It is in this sense that the construction of monument s is increasingly recognized as processes in which peoples with distinct biographi es, and possibly even distinct social and cosmological histories, were integrated through shared yet likely contested ritualized performances (Adler and Wilshusen 1990; Barre tt 1999; Bernardini 2004; Bradley 1998; Pauketat and Alt 2003; Van Dyke 2003). In m any cases, the authority of experience


95 within a monument is enabled by referencing t he past as a means of projecting a future, both of which are projections of a backgroun d into the foreground (Bradley 2002:86). Regardless of their relationshi p with the past, ritualized performances in focal places provide a moment of transformation as in dividuals and materials are shifted between fields of practice and revalued. For example, this would occur in the process of burying the recently deceased and transforming them into ancestors (Barrett 1999). Just as the deceased are transformed, the liv ing are unified (if only for a moment). It is through such practices that places can become imbued with ancestral presence, thus solidifying their significance to the community. Biographies of Place From the perspective of inhabitation, so cial memories emerge amongst the many spatial stories distributed thr oughout landscapes, from the routinized to the ritualiz ed. These memories enabled inhabitation by providing ways in which referents could be creatively combined or neglected according to schemata. Because dwelling perspectives based on phenomenol ogy emphasize being in the here and now, they tend to be synchronic (Gosden 1999:128) Indeed, one flaw with Ingolds formulation of the dwelling perspective is t hat there can be no breaks in the taskscape. In one sense this is true, structuring principles will be reproduced in practice, and will themselves be structured by the ongoing flow of inhabitation. Ingolds view, however, is hard to reconcile with the process of inhabitati on and structuration in t he long term. It is just as important to consider those mo ments when the system of reference changes (e.g., Gillespie 2008). As Barrett (1999:257) is fond of illustrating, all societies must live in the accumulated materiality of previous generations. Communities are thus routinely confronted with material remi nders of the past, the so-calle d surviving remains of


96 antiquity (Bradley 2002:14). While thes e elements may have provided a basis for societies to create a sense of continuity in place (Tringham 2000), they could also be appropriated, politicized, and set apart fo r discursive action (Barrett 1999). One way to examine how communities incorporated the past to accommodate change is through a consideration of biographi es of place. Biographies emerge as inhabitants rework the materiality and sign ificance of places, and are recursively structured by those acts through time. Such ac tivities provide the potential for continuity and change as histories are negotiated am ongst practioners (Gillespie 2008). A consideration of the burgeoning literature on the past in the past indicates that discriminating between continuity as l ong-term reproduction and the assertion of continuity by referencing the past is not a straight forward task (Blake 1998; Bradley 1998, 2002; Gosden and Lock 1998; Van D yke 2004, 2009). To the point, the generation of social memory is a selective process by which agents actively resignify, reference, or discard past practices and materialities as a means of negotiating contemporary social concerns (Bradley 2002:12; Connerton 1989; Shils 1981:45; Van Dyke and Alcock 2003:3). As structured by tradition, selective memories provide the medium for ongoing culture making. Ther e are, however, potentially diverse and dissonant reasons for referencing the past. As the basis for social reproduction, monuments, narratives, landscapes, heirloo ms, ceremonial perfo rmances, or habitual practices all afford dominant discursive and experiential versions of history which structure potential action (Barrett 1999; Bourdieu 1977; Mills 2004). At the smallest scale, unintentional consequences may forever modify the physicality of a place and open up new ways of experience (Joyce 2004). That is,


97 despite intentions of agents to create enduri ng histories that aut henticate their own identities and naturalize social differences, the contexts in which traditions are enacted are inherently dynamic (Edmonds 1999:134; Pauketat and Alt 2005; Rumsey 1994). Throughout the lifetime of an individual, di fferentiated biographi es and networks of relationships can accumulate In such cases, the pot ential discordance between practice and structure can be denied in commemorative ceremonies that reproduce shared ancestral pasts (Morphy 1995), or through institutionalized and politicized forgetting as a means of relinquishing change to history (Kchler 1993). Maintaining continuity in the face of constant change requires a significant investment (see also Shils 1981:166). This is perhaps no more evident than in the history of atalhyk, as related by Hodder and Cessford (2004). atalhyk is a tell site in Turkey primarily inhabited dur ing the Neolithic period between 9400 cal B.P. Hodder and Cessfords analysis points to t he complexities of multiple layers of memory work enabled by repeated practices. For example, in their analysis of household organization and longev ity, they argued that social memories were embodied by inhabitants in the temporality and spatiality of daily house life. They examined the repetitiveness of house forms through time, the te mpo of interior wall plastering, and the spatiality of intraand extra-household cleaning and refuse disposal. Hodder and Cessford found that houses tended to be built to a very strict set of organizational principles. An analysis of the microartifacts found recurr ent patterns of segregated activities within houses. They also determined that the interior wall s were plastered at least annually, and likely much more frequent ly. Outside of houses, middens are significantly less segregated. Taking these stru cturing principles t ogether, it could be


98 argued that the repetition of cleaning and constructing acts formed the basis for the ongoing constitution of community through embodied dispositions. Yet remembering also took the form of more public even ts of less frequency. Most houses lasted between 50 to 100 years before being replaced. At the ti me of rebuilding, new walls were emplaced direct ly on top of the wall foundati ons of old houses. Hodder and Cessford argue that in creating new houses, a variety of acts ai med at cleaning and preparing the foundation for t he new house were deployed. In many cases foundational deposits were interred. In some cases thes e deposits included human remains, while in others inhabitants deposited what may hav e been the remains of feasts. Other evidence for the operation of more specif ic memories is accorded by the apparent excavation of objects for use in shrines, including human crania. The archaeologists argue that the house, in their estimate, wa s the main mechanism for creating social rules (2004:22), given the complexities of aggregated inhabitation. More important for the present discussion, however, is the am ount of concern placed on maintaining houses, as well as the settlement, through general and specific moments of memory construction (see also Tringham 2000). Cont inuity was not a given as social life unfolded, nor was it taken for granted. The inhabitants of atalhyk were able to manufacture continuities over the long term. Yet there are other examples where selected memories were drawn together to make new social histories. In some cases, these new histories were forged during during moments of radical soci al change, such as culture contact or political upheaval. In others, the taskscapes of old no longer had the same references. In such cases the conspicuous manipulation of the past was likely as histories were asserted and


99 negotiated across diverse social fields (Bender 1993; Sahlins 1985; Shils 1981). Expected are the rapid reconfi guration or inventions of tr adition by which agents can assert a timeless ancestry, real or imagine d (Hobsbawm 1983; Pauket at and Alt 2003). Indeed, places like Cahokia, Chaco Canyon, and Poverty Point in North America provide dramatic examples of the (re)construction of history. In these cases, landscapes were transformed through inscriptive practices as new mythic histories were forged from communities of diverse traditi ons, according to recent interpretations (Pauketat 2000; Pauketat and Alt 2003; Sass aman 2005b; Van Dyke 2003). No doubt, it was through participation and negotiations in founding events that a new world order emerged. Equally impor tant however, its that during t he construction of these places, the system of landscape refe rents was fundamentally alte red. Unless these places were subsequently erased from the land scape or abandoned, they would have to be incorporated into future systems of referenc e. Indeed, that may have been the reason for their construction, as has been argued for early European monuments by Bradley (2002:82). The strategy fo r archaeologists, and no doubt agent s of the past, is to discern the alternative co mmemorative tactics embedded in the performance of past practices, and the referencing of prior materia lities, selectively draw n from tradition. Towards Archaic Histories I began this chapter with a two-fold definiti on of history. In one sense, histories are created through the practices of people in the course of routine and public living. At the same time, communities construct soci al memories as narratives about the past. Building on insights of practice theory, I have argued that conceiving landscapes as inhabited provides one avenue in which archaeologists can access the two sides of history-making among past hunte r-gatherer societies. Cent ral to an archaeology of

PAGE 100

100 inhabitation is the social constitution of landscapes, in which places are imbued with significance in the form of social memori es, woven together by experiences engaging with or avoiding places. These experiences are informed as much by the structure of natural processes as they are the tempo and dur ation of residential lif e, interactions with others, intentional and unintentional modi fications of the physical landscape, and treatment and venerati on of the dead. Following de Certeau (1984), I have suggest ed that social memories (as spatial stories) are written as lin ks are drawn between different places and peoples on the landscape. While historical referents within landscapes cannot be treated as true writing, they nonetheless are principle sources for reme mbering and contesting the past. Inscriptive practices and movement between places constitute economies of memory through which the world is ordered and understood. Communities can create a sense of continuity through disciplined commemorative strategies, such as house building and other ritualized practices, that ma ke reference to extant materiality. Yet it is also in the construction of narratives about the past that the potential for social transformation exists. This is particularly the case when hist orical referents change, or communities are faced with unanticipated experi ences that bring the taken-for-granteds of life to the level of discourse. From an inhabitation perspecti ve, biographies of place, and community histories, persist in the deposits of shellfish, earth, and other materials laid down by Mount Taylor communities along the St. Johns. Becaus e of their durability and archaeological visibility, these materials provide a unique opportunity to examine how places were created and transformed through time. It is also likely that the continued presence of

PAGE 101

101 these material traces likely provided Mount Ta ylor communities with a past that could be referenced or obscured to create social memori es in the context of ecological and social change. The question remains, of course, is how does one go about delineating the histories and social memories of Archaic comm unities? In this regard, I follow Pauketat and Alt (2005:230) who suggest three proc edural fundamentals that recursively inform the final analysis. The first procedure is to document practical variability through time and across space, with the goal of delimiting the primary structuring principles (in the sense of Barrett [2001]) or practical traditions (in t he sense of Pauketat [2001b]) which Mount Taylor communities deployed. In this st udy, my primary concer n are those depositional practices involving shellfish or earth. Unfortunately, alon g the St. Johns, shell matrix sites are still, to this day, considered pr imarily refuse, and as a consequence, little attention has been paid to variations in depositional practices within sites or between them. I focus on macro-stra tigraphic reconstructions of depositional surfaces and associated matrices. Because no site has had comparable data recovery, it is impossible to compare specific frequenc ies or abundances of shellfish and other materials between sites. As discussed in Ch apters 5 and 6, however, it is possible to assess the variability in depositional practices across the valley in terms of their scale and location through excavations undertaken fo r this study and a consideration of historic records. The second component of the methodology consists of detailing genealogies or histories of practices, in order to underst and how practices were innovated, deployed, revalued, ritualized, or extended. As r egards the deposition of shellfish and other

PAGE 102

102 materials, particular attenti on must be paid to preexisting material referents and the situatedness of communities. At a mi nimum this procedure requires detailed chronologies and reconstructions of depositional surfaces to isolate the organization of space within shell matrix sites (Chapter 6). Yet any consideration of depositional histories must take into account how stru cturing principles were translated between alternate social fields, and how places were tr ansformed. In this regard, I consider how deposition in residential places, ritualized she ll ridges, and mortuaries varied in terms of tempo and sequence as registered in the biographies of shell matrix sites. This analysis is effectively a process of remapping, wher ein I trace the accumu lating materiality of Mount Taylor communities thr ough time and across space. The third component of the analysis consis ts of tacking back and forth between many lines of evidence at multiple scales. These multiple sources include non-shellfish related practices, temporalities of the land scape, and material structural conditions. I lay the groundwork for these comparisons in Chapters 3 and 4 by establishing the ecological and social contexts of inhabitation. The St. Johns is distinct in that both the natural and social realms were in considerable flux throughout the Mount Taylor period. As a basis for delimiting the changing or ganization of Mount Taylor taskscapes, I present a detailed reconstruc tion of the ecologies and hydr ologies of the St. Johns through time. In Chapter 4 I consider the genealogies of Mount Taylor traditions that framed the deposition of material s in place. At the regional scale, this discussion emphasizes how new and old inhabitants of the St. Johns had distinct biographies which were negotiated throughout the landscape. I expand this comparison to the biographies of peoples and places outside of Peninsular Florida as seen through the

PAGE 103

103 deposition and curation of objects. As t he basis for further discussion, I will first consider the history of the physical landscape in which Mount Taylor inhabitation unfolded.

PAGE 104

104 CHAPTER 3 THE ECOLOGIES OF PRECERAM IC ARCHAIC COMMUNITIES In popular references and imagination, the St. Johns River basin exists as a timeless and enduring landscape (Bellev ille 2000:xvxxxi; Mohan 1990). For many, the region represented the last vestige of a r eal and unadulterated Flor idian wilderness. Together with the sluggish black waters of the river, the seemingly unending marshes and swamps teeming with ecological diversity present a picture of exotic, primordial, and even undisciplined nature. The river-as-relict concept was first superimposed upon the landscape with the initial Spanish entrada, and was reinforced by successive waves of naturalists (e.g., Bartram 1791) and cons umers of those works (e.g., Porter 1974) who perceived the landscape as largely devoi d of significant anthropogenic impact. Together with this notion of an undevel oped world, explorers, migrants, government agents, escapist aut hors, recreational promoters, and property speculators valorized the St. Johns in terms of its unfulfilled economic potent ial (e.g., Barbour 1882:294; Brinton 1869). All resources were targeted for exploitation, resulting in the rapid transformation of both cultural and natural resour ces by the middle twentieth century (Figure 3-1). Chann el dredging (Figure 3-1C), cypress logging (Figure 3-1B), clearing and draining for agricultural use (notably orange groves and cattle ranches), and the continued construction of buildings and impoundments hav e all had lasting effects on the distribution and dynamics of resources seen today (Brenner et al. 1999; Kroening 2004; Sassaman et al 2003; Sterling and Padera 1998). Whether recognized as anthropogenic or not, shell-bearing archa eological sites were commodified and actively sought after as sources of cons truction fill, foundations for buildings, and treated as curiosities for week end explorers (Figure 3-1A).

PAGE 105

105 This pristine ecological narrative is dually the product of an emergent American objectification of land as property to be freed from nature or wilderness (Cosgrove 1985b:169), and denial of the existence or significance of preexisting indigenous traditions (Bale 1998; Cronon 1983:33; Denevan 1992; Miller 1998:1). Archaeologists and geologists alike have demonstrated that the configuration of both the cultural and natural heritage of the St. Johns today reflec ts a series of complex processes dating back at least to the Pleistocene. The St. Johns basin was the locus of nearly 13,000 years of human settlement which has le ft the region rife with the accumulated materiality of pre-contact practices (G oggin 1952; Milanich 1994; Miller 1998). At certain times and places, archaeological evidenc e for inhabitation is limited in its extent and visibility. This is the case for traditions of the late Pleistocene and early Holocene (ca. 13,000 cal B.P.), when t he surviving output of practi ce is limited to dispersed lithic scatters, often in localities now inun dated by higher sea level and inland surface waters. By the middle Holocene, however widespread and intensive inhabitation is registered by shell-bearing co mponents at the onset of the Mount Taylor period (Figure 3-2), some of which were r epeatedly inhabited up through to the Colonial era. While these patterns undermine any notion of a vacant landscape bereft of anthropogenic input, the river is given more agency than th e communities themselves (e.g., Miller 1992). The St. Johns is thus preserved as a bounda ry, an inhospitable frontier prior to the onset of shellfishing, and alternatively as a center once shellfishing is established. Shellfishing is thus repres entative of a stable ecosystem, and by extension shellfishing cultures of the St. Johns ar e necessarily defined by unreflexive practices in socially

PAGE 106

106 undifferentiated places. From the perspective of inhabita tion, however, the veil of sameness is not something to be accepted, but must be interrogated. In this chapter I consider the ecologies and material structur al conditions experienced by Mount Taylor communities. Although separated in my discussi on, I want to make it clear that the environment and practical traditions of M ount Taylor communities should not be decoupled. If we accept that the temporality of the landscape is in part generated by environmental structural conditions, then it is necessary to consider the potential range of variation within such processes. Clim atic events and hydrological responses can be seen as potentially disrupting systems of reference that would have provided opportunity and constraint on the kinds of spatial stories possible. All ecological change is understood through hist orically contingent traditions, such that perception of and response to change is always understood against the background of cosmology, history, and comm unity identity. Indeed, constant change could have been the system of reference as well Differences at multiple temporal and spatial scales were likely recognized and addr essed in varying ways depending on how those changes were perceived (Bailey 2007; Hu ll 2005; Robb 2008). At the coarsest or longest scales, such alterations in the rhyt hms of the taskscape could have resulted in new intersections of biographies. Some changes would have potentially occurred within the observational scale of a generation or two, and may have necessitated the excision of places from memory. Others, however, potentially required the reconstitution of communities. The data I consider in this chapter are multi-scalar and mu lti-temporal in nature. Mount Taylor communities were nothing if not regional in the scale of their seasonal

PAGE 107

107 movements, exchange relationships, and by extension their biographies. As will be detailed in Chapter 4, there is a long-running relati onship between communities who inhabited Floridas interior and coasts, with the St. Johns River mediating the relationship between the two. The implicatio n is that the St. Johns was important not solely as an economic resource as it has been modeled, but as a co-inhabited region that provided the contexts for diverse ancestries and biographies to be negotiated through place. In this vein, I first consider the large-scale structural conditions of the Florida Peninsula, whose geomorphologic al configuration has had long-term significance in terms of the the arrangement of the coasts, the availability of surface waters and raw materials for stone tools. In turn, these different domains have framed the situatedness and interactions of inhabitants through the course of the Archaic. As a historical process, varying local conditi ons across the peninsula in the course of environmental change resituated communities and inserted new or di fferent biographies and temporalities into the St. Johns. From the Florida Platform, I shift perspec tives to the study area, the Middle St. Johns River Valley, and consider the evidence for the many different rhythms that would have framed the ongoing reproduction and trans formation of taskscapes along the St. Johns. Detailing change and process in pas t hydrological systems, the purview of paleohydrology (Schumm 1967), is crucial fo r understanding recursive humanenvironmental relationships and long-term land scape histories (Kidder 2008:1268). On a practical level, how fluvial systems hav e responded to climatic change has a realworld impact on the long-term preservation and identification of cultural resources (Wells 2001). Any account of Holocene fluvia l history must take into consideration a

PAGE 108

108 host of phenomena. In low-lying areas of Fl orida, and particularly along the St. Johns River, archaeologists routinely reference eustatic sea-level variation, or base-level change with respect to fluvial systems, as the primary medi ator of localized environmental change throughout the Holo cene (Miller 1992, 1998). Although no doubt an important component of regi onal ecological change, base-level change must also be integrated with geomorphology, atmospheric circ ulation data, and rela ted precipitation regimes (Bloom 1983; Knox 1983; Schumm 1993). After reviewing the evidence for local and regional rhythms, I conclude the chapter with a consideration of possible scenarios of landscape stability and change. Geomorphic Context of the St. Johns River The St. Johns River is a dynamic and unique blackwater fluvial system that defines nor theast Florida (Figure 3-3). In contemporary geography the river emerges from the St. Johns marsh near Vero Beach, and traverses north to Jacksonville where it flows into the Atlantic Ocean. At roughly 500 km in length it is the largest river exclusively in Florida, and the longest river wholly within t he coterminous United States that flows northward. The river is expansive and drains approximately 24,000 km2 (Environmental Consulting & Technology 2004), which accounts for roughly 21 per cent of Floridas area (Kroening 2004). The system is characterized by a very low gradient (0.02 m/km) and concomitantly low daily discharge and average channel velocity compared with other large ri vers (Miller 1992). As a consequence, the expansive floodplains and terraces have remained in their characteristic form (Knighton 1998:283) in recent times as evidenced by extensive freshwater wetlands. However, the overall low gradient and maximum floodplain elevation (roughly 8 m at the headw aters, but less than a meter at its mid length) has also m ade the region potentially susceptible to small

PAGE 109

109 perturbations in precipitation regime or sea level (Guccione 1995; Schumm 1993; Woodroffe et al. 1989). The study area is the Middle St. Johns River, or St. Johns Offset, which is emplaced along a roughly 125 km segment midway within the valley (Figure 3-3). This region was the heartland of Mount Taylor inhabitation, and arguably contains the highest density of shell matrix sites in the re gion (Figure 3-2). I consider some of the more important physiological, climatic, hydr ological, and ecological trends that define the region. The divisions of the river va lley employed throughout this work reference geomorphological distinctions that do not neatly confor m to hydrological drainage basins. Geomorphological divisions be tter describe the resources, ecological distinctions, and human settlement between river segments. Hydrogeology The study region is loca ted within Penins ular Florida, which occupies less than one-half of the Florida Platform, a linear landmass extending southward from the Southern Coastal Plain. The terrestrial porti on of the Platform is characterized by low relief ranging from an approximate maximu m elevation of 110 m above sea level1 in Northwest Florida where it transitions to th e Southern Gulf Coastal Plain (Figure 3-3), with elevations decreasing to the east, west and south. The subsurface Peninsular Arch forms the spine of the Platform and appears to have influenced deposition of materials on the Platform throughout its hi story (Green et al. 2007). The Platform further extends down to 200 m below sea level in the Atlantic and Gulf of Mexico basins 1 Above sea level is a generic term used to refer to elevations in excess of a determined surface level of the ocean averaged over a stated period of time. I have converted elevation values from the earlier National Geodetic Vertical Datum of 1929 (NGVD 1 929) to be relative to the North American Vertical Datum of 1988 (NAVD88) whenever possible. Base d on a limited sample of spot elevations the difference between these data are typically less than a meter.

PAGE 110

110 (Scott et al. 2004:14), with much of the elevated portion of the platform offset to the east. The subaqueous extension has a gradual sl ope extending 240 km into the Gulf of Mexico, and a more abrupt slope extending so me 70 km to the east of the Peninsulas coast. Because of the low gradient, the Pla tform is vulnerable to rapid inundation in accord with sea level change (Fau ght and Donoghue 1997; Scott 1997). The Platforms geomorphic history was wr itten by water: its contemporary configuration is ultimately the product of ma rine processes (Schmidt 1997). Much of the Platform is composed of carbonate-rich sedim entary lithologies, such as the Avon Park, Ocala Limestone, and Suwannee Limestone forma tions, that are intercalated with siliclastic-rich formations (Randazzo 1997; Scott 2001). The carbonate composition of many of the Platforms sedimentary units are particularly susceptible to chemical weathering which results in karst topography and hydrogeology (Lane 1986). The karst configuration of Floridas geom orphology controls the distri bution and availability of surface water, which is directly relat ed to subsurface aquifers (Miller 1997). The Surficial Aquifer System (SAS) is present throughout the Peninsula. This is an unconfined system that re tains precipitation not immediately evaporated or transported as runoff. The other aquifers are confined, meaning that water is contained under hydrostatic pressure within permeable bedrock, typically carbonates, below formations with limited permeability, typically siliclastic bedrock that serves as an aquiclude. The Floridan Aquifer System (FAS) is the most extensive in the Platform, and is confined within the Avon, Ocala, and Suwannee Limestone formations (Miller 1997). Like all aquifers, t he FAS is recharged by rainfall which permeates into the system through surface openings or via interchange with the surficial system.

PAGE 111

111 Approximately 55 per cent of the Florida Peninsula is a recharge area for the FAS, which also receives input from Georgia, Alabama, and South Carolina (Miller 1997; Scott et al. 2004:15). Generally the FAS is restricted to carbonate rocks of Tertiary Age and remains confined well below the ground su rface. The aquifer is unconfined or debouches in regions where carbonate rocks ar e thin or have been penetrated. The surficial hydrogeology of karst landsc apes is typified by sinkholes, sinking rivers, disappearing lakes, and springs (Whi te 1988). Springs have had long-running significance for Floridas ancient inhab itants (Dunbar and Waller 1983; Miller 1998; Thulman 2009). As noted by Scott et al. (2 004:15), springs are most likely to occur in areas with well developed karst, and where aqui fer pressures are high enough to allow flow from the vent. Artesi an flow from vents in sinkho les or caves connected to confined aquifers occurs when the potentiometr ic surface of the aquifer is higher than the elevation of the vent. Lar ge-scale impacts to hydrological characteristics are related to the hydrogeology of the aquifer. The elevation of the potentiometric surface varies in relationship to the base of the confining uni t, the level of water within the aquifer system, and sea level (Miller 1992). While the elevation of the confining unit remains relatively constant, lowered sea level and/or rainfa ll rates will decrease the potentiometric surface, and concomitantly lower the volume of water flowing from the spring. Most artesian springs are associated with the Oc ala Karst District, an arcuate zone that encompasses much of northwest Peninsular Florida (Figure 3-3). This area is part of the Ocala Platform which is characterized by ka rst plains and valleys. The FAS is thinly confined here, and spring discharge occurs from vents in sinks and stream beds, but has also been identified offshor e on the platform. An impor tant anomaly in the context

PAGE 112

112 of the present study is the presence of spri ngs outside the Ocala Platform located in the Middle St. Johns River Valley to the east. The Middle St. Johns is the only riverine segment in eastern Florida to receive significant flow from spring sources. Economic Geology The develop ment of karst has influenced the distribution of chert and other silicate rocks. These materials were heavily expl oited by Archaic communities for the manufacture of stone tools (Austin and Estabrook 2000; Purdy 1981; Ste. Claire 1987). In Florida, chert and silicified corals we re formed as carbonates in bedrock were replaced by silicates from overlying formati ons (Austin and Estabrook 2000). Siliceous stone can potentially occur throughout limestone bedrock, and is accessible wherever chert-bearing formations are exposed on the su rface. Exploitable rock suitable for toolstone is restricted to West Central and Northwest Florida (Figure 3-3). The distribution of chert is gener ally conformant with the Oc ala Platform and well-developed karst related to Floridas springs where limestone bedrock is exposed. Unlike the distribution of karst, however, chert is not available in the st udy area. In fact, the closest source of chert from the Middle St. Johns region is approximately 50 km to the west. The implication is that any stone raw materials recovered within the study area were necessarily transported from outside the region. Floridas cherts can be assigned to at least 19 quarry clusters based on the visual identification of diagn ostic index fossils and other in clusions and determination of their relative abundance (Austin 1997; Au stin and Estabrook 2000; Endonino 2007; Upchurch et al. 1982). As argued by Endon ino (2007), many of the current quarry cluster boundaries are based on the expected o ccurrence of chert within their parent formations, and do not necessarily reflect disc rete distributions. The within-cluster

PAGE 113

113 distribution of outcrops is not ubiquitous, but is discontinuous and patchy. Regardless, both the distribution of quarry clusters and t he identification of raw material source provenance is a crucial for understanding larger scale dynamics of Archaic settlement and regional interaction. Relative Sea Level One of the defining c haracteristics of Holocene landscape variability is relative sea level (RSL) change. Along the St. Johns RSL is potentially signific ant for two reasons. A change in river base-level in step with RSL could have affected the hydrological regime of the river, in part due to the ac tivation of springs debouching from the FAS and a general increase in local water tables (Miller 1992). A second RSL factor is the potential impact on coastal-dwel ling populations duri ng the Archaic. A number of RSL curves have been published for the Atlantic (Colquhoun and Brooks 1986; Horton et al. 2009), Gulf of Mexico (Balsille and D onoghue 2004; Otvos 2004; Simms et al. 2007; Tornqvist et al. 2004; Wright et al. 2005), and globe (Siddall et al. 2003), although none are available for the Atlant ic seaboard adjacent to the St. Johns. There is general agreement as to the shape and pace of sea level change. All models depict a rapid rise in sea level from -125 m at the height of the Last Glacia l Maximum (LGM) 18,000 years ago, with a noted deceleration during the middl e Holocene. There is considerable divergence, however, as to the presence of still-stands or hi gher-than-present stands during the middle Holocene. This is not the place to engage in a discussion on the merits of either model. The choice of paleoenvironmental proxies, sampling methodologies, and local conditions have a signi ficant impact on how to best construct and interpret the curves (Otvos 2004, 2005; Tornqvist et al. 2004).

PAGE 114

114 Presented in Figure 3-4G are three of the most recent high resolution curves for the Gulf of Mexico, which extend from between 7300000 cal B.P. to the present. Balsille and Donoghue (2004) generated two cu rves post-dating 7000 cal B.P. based on the geomorphic location of samples. T he Younger A curve represents 77 samples located seaward from the present shoreline, while the Younger B curve is derived from 108 samples collected landward of the present shoreline. After culling outliers, the authors applied a 7-point floating average to smooth variance between points. Their Younger B curve does suggest high stands duri ng the Mount Taylor period, while the Younger A curve presents a much more gradual curve without high stands. Also depicted (Figure 3-4G) is a smoothed curve der ived from a detailed reconstruction of the depositional history of t he Suwannee River delta in Florida (Wright et al. 2005), which also matches corrected curves presented by Tornqvist et al. (2004). In all curves, RSL is approximately -7 m at 7000 cal B.P., and rises to roughly -4 m by 6000 cal B.P. Wright et al. note that this is a period of peat formation and aggradation along the coast, which may explain why some curves identify hi gh stands at this time. This rise equates to between 3 and 4 mm of RSL increase per y ear. At that rate, approximately 12 m of coastline would be transgressed annually along the Gulf Coast (Faught and Donoghue 1997). Beginning ca. 6000 cal B.P. there is a deceleration in RSL rise to roughly 1 to 2 mm per year, which equates to roughly 1 m of coastline transgression a year on the Gulf Coast. Regional Physiography A number of physiographic divi sions in the St. Johns region are recognized (e.g., Cooke 1939; Schmidt 1997; White 1970). The St. Johns River is located in the Atlantic Coastal Lowlands, a relatively low lying zone typified by coast-parallel features (Figure

PAGE 115

115 3-5). The valley is com posed of three segments (U pper, Middle, Lower) whose characteristics relate to a complex geomorph ic history (Adamus et al. 1997; Schmidt 1997; White 1970). Like many of the large river systems in Florida, the St. Johns is situated in a swale between elevated, upland ridges (Figure 3-5). The headwaters (Upper) and mouth (Lower) of the river are situated within portions of the Eastern Valley. The most prominent positive feature to the east is the Atlantic Coastal Ridge, which represents beach ridges formed during the Pleistocene and earlier (Scott 1997). The Upper St. Johns flows between Indian Rive r and Brevard County near Vero Beach, to Sanford, Florida in Vo lusia County. The western boundary of this segment is composed of the Osceola Plain which is characterized by undifferentiated and beach ridge and dune formations of Pleistocene and Ho locene age (Scott 2001). The Osceola Plain is drained by the Econolockahatchee River, a tributary of the S t. Johns. Within the Upper segment, the St. Johns emerges from marshlands characterized by poorly integrated stream segments. Channeled surface water fl ows between shallow basins, including lakes Hell n Blazes, Sawgrass, Washington, Winder and Poinsett. Wetlands associated with the floodplain in this r each accounted for as much as 1800 km2 prior to the turn of the century (Brenner et al. 1999). The study area is comprised of the Middle St. Johns, between Sanford and Lake George. It is oft en referred to as the St. Johns Offset, and is distinct in both hydrology and geomorphology (Figure 3-5). In a headwardconsequent course, the river would be expected to flow from the headwaters to Jacksonv ille in a straight line following the late Pleistocene beach ridges of the Eastern Valley. However, at Sanford the St. Johns jogs to the west, and north of Lake Ge orge, the river jogs back to t he east. It is believed that

PAGE 116

116 this segment of the river formed during t he early Pleistocene in a period of low sea level, when the offset portion of the rive r captured the headwater s south of Sanford (White 1970:107). As a result of this offs et, the river is confined between uplands and ridges on either side of the valley. T he eastern boundary is formed by the Crescent City-Deland Ridge, which rises over 20 m amsl above the floodplain. This ridge is an outcrop of the Cypresshead formation (Scott 2001). To the west are the Marion Upland at elevations below 10 m rising up to the M ount Dora Ridge. The Marion Upland is characterized by extensive sand dunes, wh ile the Cypresshead and Coosawhatchie Formations are expressed near surface al ong the Mount Dora Ridge (Scott 2001). Although the presence of thes e ridges may ultimately be due to relatively insoluble clastic lithologies (White 1970:128), their current surficial expression is typical of a mantled karst landscape. Of particular note are the numerous sinkholes that have dissected the uplands and ridges, and freshwater springs at the contact between the St. Johns floodplain and t he upland zones. The northern, or Lower St. Johns is situated between the eastward jog of the main river channel north of Lake George to the mouth at Jacksonville. This course is parallel with Crescent Lake, a relict channel of the St. Johns likely abandoned when the Middle St. Johns switched to its current location duri ng the Pleistocene. The western border of the St. Johns is formed by the Northern Hi ghlands, part of the Cypresshead formation, a Pliocene aged deposit of unconsolidated marine sands and clays (Scott 2001). This section of the river is essentially a drowned estuary, and is characterized by a broad channel, averaging over 1 km in width. In shore marine and brackish habitats dominate the local ecology, resulting in a lack of fr eshwater shell mounds along most stretches of

PAGE 117

117 this portion of the river. Shell matrix si tes composed of marine species near the mouth of the St. Johns register intensive exploitation of marshlands here beginning in the late preceramic Archaic per iod (Russo 1993). Holocene History of the Middle St. Johns As Miller (1998:28) notes, the dominant fact or in the study regions landscape is water, which is concentrated along the St. Johns River drainage. Atmospheric circulation variation, particula rly temperature and rainfall, is a key variable relating to the dynamics and long-term histories of floodplai ns (Brown and Quine 1999). The annual cycle plays an important role in defining the hydroperiod, or duration in which ground is saturated or flooded, which directly impacts ecological community structure (Ganopolski et al. 1998; Hupp 2000). Changes to the hydroperiod and floodplain structure can amplify hydrological responses to climate change including aggradation, incision, and channel form reorganization (Brown 2003). Thus annual and long-term availability of water no doubt affected ecological structure and diversity, the location of channels and water bodies, and the kinds of responses the river may have initiated (Brown 1996; Brown and Quine 1999; Knox 1984; Thomas 2001). Impacts of Historic Land Use The hydrological and climatic data available for the St. Johns has been influen ced by several recent impacts to the study regi on. I will make specific mention throughout the subsequent discussion, but there are seve ral caveats associated with recent trends and their utility in reconstructing past environments along the St. Johns. The Upper St. Johns marsh serves as catchment for annual precipitation of the St. Johns system upstream from the study area, and is a critical source of natu ral floodwater retention. Sterling and Pandera (1998) have noted that after 1900, upwards of 70 per cent of the

PAGE 118

118 Upper St. Johns was drained and converted to pasture or orange groves. They suggest that the loss of this natural floodwater storage resulted in several large magnitude floods in the 1920s and 1930s. One response was the widespread modification of the Upper Basin for flood control by the United Stat es Army Corps of Engineers (USACE), approved in 1954. This project invo lved the construction of impoundments and diversion canals that decreased the amp litude of flood events but was severely detrimental to the wetlands of the region. Sedimentation rates within lakes increased because runoff of improved lands was dire ctly injected into the river channels circumventing the marshlands, but channel ve locity had decreased due to flood control measures (Brenner et al. 1999). By 1980, the basin downstream2 from the Upper St. Johns may have received less than 55 per cent of the pre-di sturbance flow (Tai and Rao 1982, cited by Brenner et al. 1999). Recognizing the negative effects on wildlif e and flora the project was ceased in the 1970s and replaced with a program of floodplain maintenance aimed at restoring pre-impoundment hydrology (Sterling and Pa dera 1998). The St. Johns River Water Management District (SJRWMD) is now char ged with maintaining the ecological and economic viability of the river, and continues to enact flood prevention measures today. Water is also routinely withdrawn from bot h the SAS and FAS in the study region for agricultural, commercial, and domestic consumption (Marella 2004). Equally problematic, the main channel from the mouth at Jacksonville as far south as Lake Harney has been dredged to enable deep draught vessels. Congressional documents 2 The terms upstream and downstream refer to the di rection of water flow, thus downstream or below refers to areas generally to the north of the discuss ed feature, while upstream or above refers to areas generally south of the discussed feature.

PAGE 119

119 indicate that the river was first dredged in the 1880s within the study area (House of Representatives [HR] 1111, 1910). Subs equent plans were authorized in 1910 to dredge and maintain the entire length of Middle St. Johns. The plan was amended in 1930 to allow for the creation of meander cut-offs and the easing of channel bends to facilitate navigation by longer vessels ( HR 691, 1930). Since 1945, the USACE has been tasked with maintaining a channel at least 3.65 m deep and 30.48 m wide within the study area (Senate 208, 1945). Dredging activities are manifested today in artificial cut-offs and dredge spoil piles adjacent channel segments. Dredging cutoff channels for navigation purposes decreased the total length and increased the gradient of t he main channel. Dredging channel beds also effectively lowered the basel evel of the local channel reach. These two processes have likely increased the veloci ty of the system relative to pre-dredging times (Brooker 1985), while the higher c apacity of the channel segment may have increased the magnitude and impa ct of flood events (Brookes et al. 1983; Brookes 1987). Secondary channels likely began to receive sediment immediately after channelization, leading to a decrease in storage capacity and water trapping (Hupp 2000). Because secondary channels associ ated with the main channel receive less flow on average today than in the past, the main channel is likely less capable of transporting higher-magnitude flood waters wit hout erosion, and wetland communities experience decreased hydroperiods.

PAGE 120

120 Climate Ultimately, all of Floridas freshwater is derived from precipitation, with the stage3 height and discharge of rivers in Florida corre lated to surficial and deep aquifer levels (Miller 1997). Today the Middle St. Johns ex periences a humid subt ropical climate with a dichotomous warm and wet summer season and dry and cool winter season (Figure 3-6). During a a typical weather year based on averages between 1971 at Orlando International Airport, the annual me an temperature was 23 degrees Celsius, while each year averaged 123 cm of rainfall. Evapotranspiration equals or exceeds rainfall rates (Kroening 2004; Robison 2004) and temperature and precipitation are roughly correlated. The beginning of the dry season, typically around October 15, occurs when the Peninsula beings to be influenced by continental air masses and the first cold front drops temperatures below 15.5 degrees Celsius (Lascody 2002). Dry season conditions are characterized by lower average temperatures and decreased rainfall frequency and intensity. Monthly rainfall averages are at their lowest from October through May, below 10 cm. There is a tendency in March for slightly higher rainfall rates, but April is as dry as t he months preceding March. The wet season corresponds to the onset of minimum average temperatures of 19 to 21 degree Celsius, starting between May 25, with average tem peratures from J une through October ranging between 32 and 33 degrees Celsius ( Lascody 2002). Although the wet season accounts for little more a third of the annual cycl e, it experiences more than 60 per cent of annual rainfall. 3 The term stage refers to the elevation of surfac e waters in reference to an arbitrary datum (Kennedy 1984). Unless otherwise noted, all stage values are presented in meters relative to the North American Vertical Datum of 1988 (NAVD88).

PAGE 121

121 In central Florida the uptick in atmospheric water during the summer is partially the result of drenching tropical cyclones (Kr oening 2004). For exampl e, from a 52-year record of rainfall at Deland in the study region, Rao (1988) determined that one-day rainfall totals of 23 cm could be expected at least once every 10 years, which is only slightly less than total monthly averages. Over the course of four days, rainfall totals over 37 inches could be expected at least on ce every 100 years. These rainfall patterns are no doubt related to tropical cyclone activity. Analysis of historic hurricane records has demonstrated that they have not occurred with a const ant frequency over the past 100 years (Landsea et al. 1999). Periods of hurricane intensity were highest in the 1940s through 1960s and again towards the end of the century. Although no clear periodicity has been determined for the mode rn era, the frequen cy and intensity of tropical cyclones in the Atlantic basin is infl uenced by periods of increased rainfall in the western Africa (Gray 1990). One of the primary mo derators of local interannual vari ation in Florida is the onset of the El Nio-Southern Oscillation (ENS O) phenomenon. ENSO is most directly associated with sea surface te mperature (SST) anomalies in the eastern Pacific basin, but is robustly expressed through ocean-atmo spheric teleconnections globally (Dunbar 2000). Although routinely referred to as El Nio, the ENSO phenomenon is composed of two phases: warm (El Nio) phases are char acterized by positive SST anomalies in the Pacific, while cold (La Nia) phases ar e characterized by negative SST anomalies (Tudhope et al. 2001). During warm phases, Flor ida receives increased rainfall during the winter months, which correlates with higher surface water levels and lower temperatures (Beckage et al. 2003). Tropica l cyclone development is significantly less

PAGE 122

122 likely to occur during warm phases (Landsea et al. 1999). In contrast, during cold phases Florida receives less-than-average dry season rainfall, increased temperatures, and a greater likelihood of tropical cyclone impacts. A review of current paleoclimate proxies indi cates that the current climatic regime changed considerably over the course of the M ount Taylor period. It was not long ago that the early to middle Holocene was broadly modeled as a period of post-glacial climate amelioration, with gl obal climate becoming increas ingly like the modern era (Schulderein 1996). The one divergence from this model was the so-called middle Holocene Climatic Optimum or Hypsitherma l episode, characterized by above modern temperatures (Deevey and Flint 1957). Howe ver, this generalized model is being replaced by reconstructions and simulati ons that demonstrate climatic phenomena operating at varying temporal and spatial sca les (Braconnot et al. 2007; Dunbar 2000; Wanner et al. 2008). These in turn have likely lead to considerable diversity in Holocene-period ecological and hydrol ogical responses (K nox 2000). Mayewski et al. (2004) have recently argued that climat ic change during the Holocene was more variable than has been tr aditionally accepted. While not of the same magnitude as Pleistocene perturbations, changes would have occurred at a pace within the perception of past communities (s ee Bard 2002). In thei r review of widelydistributed high-resolution paleoproxies (incl uding ice cores, varve deposits, and glacial extent models) Mayewski et al. identified at least six episodes of rapid climatic change (RCC) events at a global scale during the Holocene (Figure 3-4H). The first RCC was likely induced by atmospheric reorganization at the close of t he Pleistocene, with one sub-event at 8200 cal B.P. inst igated by a massive release of freshwater into the North

PAGE 123

123 Atlantic (Alley et al. 1997). For the rest, they postulate that the timing of each RCC was linked to orbital and solar insolation variation. The high chronological resolution of the results indicates that many of these RCCs would have unfol ded over the course of a few hundred years, and likely less (Mayewski et al. 2004:245), although these events are not necessarily equally distributed, synch ronous, or expressed lo cally (i.e. Kidder 2008). For the present study, the RCC of 60000 cal B.P. is the most significant as it occurs in the midst and end of the M ount Taylor period. Globally, this RCC is generally characterized by lower average tem peratures in polar regions and increased aridity at lower latitudes, although its specific expression in the Middle St. Johns is unclear. As noted by Beasley (2008:90) in his c onsideration of Mount Taylor period paleoclimate, the recent reconstruction of Ho locene temperature variations across North America by Viau et al. (2006) holds promise fo r contextualizing middle Holocene climate change. Viau et al. compared 752 pollen re cords with over 2500 radiocarbon dates, and calibrated resultant pollen spectr a with modern analogues to generate mean continental and regional July temperatures (Viau et al. 2006: 2). The pollen time series were sampled at 100-year intervals using linea r interpolation. A fter a rapid rise in temperature between 10,000 cal B.P. the continent witnessed a decrease in mean temperature between 80 00 cal B.P. Beginning at roughly 6000 cal B.P., and corresponding with Mayewski et al.s scheme, there is a rapid rise in temperature that persists until roughly 3000 ca l B.P. As noted by Viau et al. (2006:4), however, these trends are often expressed inversely or asynchronously across regions.

PAGE 124

124 The reconstruction of temperature anom alies within the Southeastern United States reveals trends that are at times inversely related to the continental-scale temperatures (Figure 3-4G). In a previous application of this methodology, Sawada et al. (2004) found that Florida had the highest rate of error due to its position on a boundary in the model. Thus, these reconstruc tions should be taken more as guides than absolute data points. There is a rapid rise in te mperature between 7300 cal B.P., coeval with the earliest radiocarbon determinations on shell bearing deposits within the Middle St. Johns and onset of the Mount Taylor period. At this scale, temperatures remain constant until roughly 6500, at which ti me there is a decrease in temperature, followed by a rapid cycle betw een higher and lower mean temperatures at 6000 cal B.P. Temperature remains relative ly constant until anot her temperature cycle at 4800 cal B.P., near the end of the Mount Taylor period. More locally, Leigh and Feeney (1995) reported higher-than-present flood regimes on the Ogeechee River in the Coastal Plain of Northern Georgia. Using paleoflood estimates, they suggest that large pal eomeanders dating betw een 8500 cal B.P. are indicative of a 10 to 30 per cent in crease in annual precipit ation and twice the modern bankfull discharge. Goman and Leigh (2004) have reported the results of a stratigraphic and palynological study of a pal eochannel of the Little River in the upper Coastal Plain of North Carolina. They recognized two different depositional regimes after the segment was orphaned from the main channel of the river. The first dated between 9000 cal B.P., and registered five flooding ev ents, evidenced by sand overbank deposits. Based on a grain-size analysis they determined that the floods would have been of moderate to large magnit ude. The upper zone, dated 6100 cal B.P.

PAGE 125

125 to present, was characterized by essentia lly modern assemblages indicating slightly drier conditions and significantly less chance of a major storm event. Although increased snow melt or other factors may a ccount for this increased discharge (Leigh 2006), there is some support for an in crease in storminess during the middle Holocene along the eastern seaboard (LaMoreaux et al. 2009). Liu and Fearn (2000) analyzed overwash storm deposits in Northwest Florida along the Gulf Coast. On the basis of analogy, they argued that th e overwash sediments represent the landfall of major hurricanes, and thus can be used as a proxy for hurricane landfall frequency in that region. They found t hat the period of record bet ween 7800 cal B.P. lacked evidence for significant storm events, while the millennia afterwards were characterized by very high frequencies of large storms. Th is shift in landfall frequency is likely the product of the location of the Bermuda High, a high pressure system that can provide steering currents for tropica l cyclones. Goman and Leigh (2004) and LaMoreaux et al. (2009) argue that the Bermuda High was likely to the north and east of its present location before 6000 cal B.P. When it is in this position, significant rainfall events are likely along the Florida Peninsula and eastern seaboard. Because central Florida is situated within a transition zone between t he Caribbean and Lower Southeast, the direct implications of these studies is not clear. It would seem likely, however, that increased precipitation, and thus major flood events, was one consequence. LaMoreaux et al. also ar gue that the decrease in storminess and above-average rainfall after 6000 cal B.P. was likely the resu lt of the onset of ENSO. The onset of ENSO conditions, particularly a warm El Nio phase, dramatically decreases the likelihood of major storm events. Dating the onset of ENSO conditions is controversial,

PAGE 126

126 but independent paleoproxy records outside of central Florida suggest an increase in ENSO activity between 7000 cal B.P., with higher intensity signals occurring thereafter (Donnelly and Woodr uff 2007; Moy et al. 2002). Although a decrease in storm frequency is likely, Donders et al. ( 2005) also demonstrate that the onset of ENSO after 5200 cal B.P. enabled the establis hment of long hydroperiod wetlands in the everglades as a product of increased winter rainfall. Hydrology With the regional geographic characterist ics assessed, we can shift scales now to the study area (Figure 3-7). The souther n boundary is defined by the junction of Lake Monroe and Lake Jessup. The northern segm ent of the study ar ea terminates at Buffalo Bluff. At this segment the St. J ohns is joined by the Ocklawaha River, which drains much of the Central Valley. In the past century t he Ocklawaha was impounded by construction of the Rodman Dam, with surface water rerouted in part through the Cross Florida Barge Canal (Noll and Tegeder 2009). Ch anneled surface water along the study area is confined to a floodplain wit h elevations typically less than a meter above mean sea level. This floodplain is characterized by a series of geomorphic basins that are constrained at their upper and lower segments (Figure 3-7). As a consequence, the channel of the St. Johns alternates between several kinds of configurations. In part, surface water fl ows through large lake systems, most notably Lakes Monroe, Beresford, Woodruff, and Geor ge, which in tandem with the low velocity of the river have inspired the ni ckname River of Lakes for t he St. Johns (e.g., Belleville 2000). Between lakes, the river can take on di fferent characteristics based on the relative proximity of uplands. In areas where the uplands are in close proximity, such as

PAGE 127

127 the outlet of Lake Monroe, or downstream from Lake Beresford, the river is confined to a relatively narrow floodplain. In contra st, several areas, particularly between the Wekiva River outlet and below Lake Beresford, and in the vicinity of Lake Woodruff, surface water flows through multiple channels of varying sinuosity, lagoons, and relict channel segments that course through extensive swamp and wetlands. In some cases there are no clear geomorphic controls that would act to separate channels. In others, channel segments flow around low islands of sand, clay, and marine marl such as Hontoon Island and Tick Island. In this r egard the Middle St. Johns matches Nanson and Knightons (1996:218) definit ion of an anabranching pattern: a system of multiple channels characterized by vegetated or other wise stable alluvial islands that divide flows at discharges up to nearly bankfull. Although the geomorphology and hydrology of anabranching rivers is still poorly known, recent studies indicate that they are characteristic of many large rivers with low gradients, with multiple channels allowing the system to effectively manage sediment and water discharge when increasing the depth of channels is not possible (Huang an d Nanson 2007; Latrubesse 2008). As an adjunct to precipitation, freshwater springs provide an important source of water for streamflow. The majority of wa ter is provided by discharge from the SAS where water is largely unconfined in Plei stocene and Holocene sediments averaging 15 m in thickness, and is present at or just below the ground surface. The Middle St. Johns is situated within the easternmost ext ension of karst related to spring discharge (Figure 3-3). Of the 71 springs recorded within the St. Johns basin, 59 have expression within the Middle St. Johns (Figure 3-7). All but two springs are fed by the FAS which is unconfined and discharges along the Crescent City-Deland Ridge, Mount Dora Ridge,

PAGE 128

128 and Marion Uplands. This total includes three first-order magnitude springs (greater than 2.8 m3/s or more) at Silver Glen Spring, Al exander Spring, and Blue Spring. Other springs of second or lower magnitude occur throughout the study region. Channels for any given spring vary in length from 0.3 km at Blue Spring to over 12 km at Alexander Spring. Annual input from springs is estima ted to be roughly 13 per c ent in the vicinity of Blue Spring, and up to 22 per cent downstream from Lake George (Robison 2004:61). Variables relating to streamflow include discharge, the amount of water flowing through a channel; stage, the height of the water surfac e; and sediment load, the amount of sediment carried either on the bed or suspended. Detailed records are available for discharge and stage, but limited for sediment load. Based on a review of water quality characteristics Kroening (2004:89) suggested that suspended sediment loads were highest in periods of low-flow, and likely resulted from resuspension of bed material and not erosion of land surfaces. Anecdotal accounts in periods after major floods suggest that there is not a significant amount of sediment that is transported within the water column during flood events, nor are there welldeveloped point bar formations. This is likely a factor of a re latively low stream power and a prevalence of organic and very fine bed materials. The characteristics of annual surface water flow can be explored through streamflowand stage-duration hydrographs using data collected by the USGS (2009) at stream gages. Hydrographs show the c hange in streamflow characteristics over a particular period of time. The shape of the hy drograph is controlled by variety of factors including size of the catchment area, geomorphology, land use, and climatic season, for

PAGE 129

129 example. The duration of stream disc harge and stage, as well as the slope of the hydrograph show how the fluvial system responds to water input in relation to these variables (Knighton 1998:67). The caveat t hat the current hydrological parameters likely do not reflect pre-disturbance regimes must be remembered. However, we can glean some insight into what kinds of ri ver conditions may have been experienced. There are several gaging stations within the study area, with thos e employed in this discussion highlighted in Figure 3-7. Although hydrographs are traditionally depicted by water years, beginning in October and ending in November, the start of the annual Julian calendar was used in t he hydrographs discussed below in order to emphasize the how the wet seasons effects ex tend into the dry season. Interannual variation in streamflow is appar ent in daily streamflow values at the three gaging stations for the years 2005 as an example (Figur e 3-8). The data points were standardized by the drainage area of the gaging station to facilitate comparison. As seen in the hydrograph, the year 2005 was characterized by relatively high discharge values, while 2006 and 2007 were significantly less peaked. In 2005 the Southeast experienced one of the highest levels of tropical storm activity on record, and concomitantly the hydrograph r egisters two peak discharge events from June to July (198 m3/s) and October to November (263 m3/s). The latter event approximates the expected 5-year flood (Table 3-1). For each event, high discharge values were maintained for durations between one and three months. During 2006 and 2007 two major discharge peaks are also registered in the summer months, even though they are of significantly lower magnitude. In contra st with the wet season, the dry season is less

PAGE 130

130 likely to register peak discharges. Quite to t he contrary, low flow or negative values are recorded at all stations between February and May. The longer the period of record for a par ticular gaging station, the more likely the data will be representativ e of longer term conditions. Unfortunately, the only gage with an extended record is the St. Johns near Deland gage. As illustrated in Figure 3-8, however, hydrographs throughout the river segment are quite similar, lending confidence to the use of one gagi ng station for the purposes of long-term generalization. Although data for the De land station is available from 1934 until present, a start date of 1957 was chosen. In that year a recorder was installed that could read both positive and negative values (Kroening 2004; Robison 2004). A comparison of median, 25th, and 75th percentile monthly discharge hydrogr aphs with rainfall for the Deland station reveal the normal discharge conditions (Fig ure 3-9A). The St. Johns discharge is at its lowest in May, just before the beginning of the wet season and the low average rainfall in April. This is also the period of least variance. Discharge remains relatively low until JuneAugust, where t he leading limb ascends rapidl y. There is a clear lag between rainfall and discharge, which is likely the result of channels and the broader floodplain storing floodwaters, in addition to a slower rate of groundwater flow. The falling limb after each peak is more steep, as water stored on the floodplain is discharged reflecting the onset of the dry seas on and a decrease in precipitation. It is notable, however, that the cumulative increa se in precipitation is reflected in high discharge levels into November and December. The increased rainfall in February and March can contribute to higher variance in discharge, but due to a lack of abundant flow and rainfall in April there is a rapi d decline of discharge into May.

PAGE 131

131 Discharge is a useful metric for mode ling and engineering purposes, but in terms of human experience, stage is a better measure of environm ental impact. Since stage is the height of the water surface, it def ines how saturated or flooded landforms will be and for how long. As seen in Figure 3-9B, c hanges in stage are similar but not entirely synchronous with discharge. In a system such as the St. Johns with multiple channels and broad floodplains, increases in discharge are distributed across more channels as discharge increases. Thus, the increase in rainfall experienced during the summer is not registered in a gradual increase in stage height. Although both peak discharge and stage can be expected in October, stage height lags behind in June and July. As indicated by the steep ascending limb from July through October, higher stages will occur rapidly once a threshold is met, but will also dissipate rapidly without continued rainfall. Throughout a typical year, the St. Johns can be expected to range less than a meter, between -0.18 m and 0.65 m NAVD88 (Figure 3-9B). Variance is greatest in the second half of the rainy season, and least in the tail end of the dry season. Unlike discharge values, stage does not appear to be affe cted by the increase in rainfall during March, likely due to an abundance of floodplai nor within-bank storage potential. Although the central tendencies of fluvia l systems can tell us about the standard state of a system at any given time, it is the extreme events that define the boundaries of the system, and changes in the fr equency and amplitude withi n these boundary conditions can lead to rapid floodplain transformations (Knox 2000). As with the precipitation regime discuss ed earlier, the use of aggregate metrics to describe normal events obscures extremes. In blackwater ri vers of the Southeaste rn Coastal Plain the hydroperiod, or annual period of inundation, is a primary factor that defines the

PAGE 132

132 distribution and composition of wetlands (H upp 2000). Decreases in the hydroperiod can have rapid consequences to ecological structure. Recently the SJRWMD has conducted extensive modeling and field studi es to determine the minimum flows and associated stage levels (MFLs) necessary to maintain the hydroperiod and current wetland biodiversity (Mace 2006, 2007; Robis on 2004). These are reported in Table 31. Low flow conditions are also widely experienced throughout t he region. Negative flow is routinely recorded at the Buffalo Bl uff station because tidal influence is robust there, and sporadic stream fl ow reversals can occur as far upstream as Lake Harney (Kroening 2004). Reversals are more frequent below Sanford, up to 4 percent of the time annually and strongly correlated with per iods of low flow, and reflect tidal influences as well as wind. Less well understood are the temporality, magnitude, and pot ential impact of extreme floods. Extreme floods are events that occur in frequencies beyond the bounds of modern instrumentation, but likely play an important role in defining channel morphology and location. The frequency an d magnitude of extreme events can be extrapolated based on extant stream gage data, but gi ven inherent nonstationarity within fluvial systems extrapolati on has a high error associated with it. An alternative is interpolating high magnitude events from pa leoflood records (Baker 1987, 2008), but there are none available for the St. Johns. In his analysis of flood frequencies using 52 years of data, Rao (1986:Table 13) computed 5through 500-year flood discharges (Table 3-1). The highest stage on record was 1.55 m, recorded on October 12, 1953, and this is the effective height of a 50-year flood. Because the relationship between discharge and stage are unknown beyond this range, the st age of 100and 500-year

PAGE 133

133 flood discharges are unknown. There is a need for programs aimed at developing paleoflood indicators al ong the St. Johns. Channel and Floodplain Organization Archaeologists working on the St. Johns routinely cite its productivity as one of the reasons for its long-term significance to Ar chaic populations. Rarely, however, is the actual distribution of productivity in time and space cons idered. Rivers and floodplains have a memory which is written in the a rrangement of wetlands, channels, and water bodies (Knighton 1998:261). In turn, the distribution of th ese features impacts the location, predictability, and vulnerability of resources that could be exploited. The dynamics between channel organization, ecology, and hydrological change are best examined at a sub-basin leve l and can be revealed through floodplain plan and cross section. Depicted in Figure 310 is a plan view of a reach of the river from below the Wekiva outlet to Hontoon Island, just above the St. Johns at Deland gaging station. Also provided is a series of cross-sections of the valley floor showing the relationship between topography, relict and dredged channels, and floodplain vegetation (Figure 311). At this point the floodplain reaches a maximum width of 4 km before it contracts below Hontoon Island. This reach encapsulat es much of the fluvial and ecological diversity exhibited elsewhere, includi ng first magnitude Blue Spring, islands, and multiple active and relict channel segments. Shell matrix sites including at least five shell mounds are distributed throughout this reach (see Chapters 4). The effects of recent land use patterns are quite evi dent. The main channel has been dredged along the eastern margin of the reach. At l east four meander cutoffs were created during dredgework, and can be seen to above and below Pine Island. Dredge spoil piles are present on the margins of the main channe l and are expressed as circular mounds

PAGE 134

134 (Figure 3-11B,C). Other rec ent effects can be seen in the east-west channels that were likely cut for cypress logging at t he turn of the century. An examination of river planform reveal s a number of import ant features that relate to recent and long-term fl uvial histories. At the la rgest scale, there is a clear division in geomorphic and surface water expr ession. The western basin is largely devoid of topographic surface anomalies or channeled surface water, and is composed of undifferentiated bottomland hardwood swamp, likely dominated by cypress. It is unfortunate we do not know when this cyp ress swamp began developing. The lack of channels within the western bas in may be a factor of age, such that evidence for channel processes has been obscured at the surface by aggradation and the development of cypress swamps. Alternat ively, the pre-inundat ion substrate may be situated at slightly hi gher elevations than the eastern floodplain. In contrast to the western expression, there is considerable diversity in the eastern reach that may relate to more recent fluvial and geomorphic processes. Topographic highs up to 5 m above the floodplain consist of sand/marl islands including Hontoon Island and Pine Island, in addition to the smaller Scrub, Live Oak, and Bush Islands. Disconnected channel segments in the southern aspect appear unrelated to subaerial exposures (Figure 3-11C, D). Connected wate r bodies include lagoons and seasonally flooded wetlands. Lagoons are char acterized by shallower depths and tend to not receive streamflow directly. Ho wever, they are frequent ly situated on the downstream side of meander bars, suggesting that the creatio n of these features is directly related to c hannel location and age.

PAGE 135

135 There are two major channel segments t hat are largely unc onnected from the primary flow of the main river, in addition to many shorter and narro wer linear features. The largest relict channels include the Banana River, which may follow the contour of a low lying terrace as indicated by the pres ence of islands and the inflection of the channel. Shell deposits hav e been reported on this land form, but their age and configuration is unknown. Situated just down stream is Hontoon Dead Creek, which as its name suggests is not a primary vector of surface water. It is however, connected to the primary channel via the sinuous, narrow, and shallow Snake Creek at its junction near Hontoon Island. Whether there was a much larger channel here or if Snake Creek is a recent response to dredging is unknow n. Government Land Office (GLO) survey maps compiled in 1853 suggest that there was a main channel here. However, given that much of the basin was characterized as i mpracticable to survey, it is likely that the channel segment illustrated in 1853 was an extr apolation from the visible junctions near Pine and Hontoon Islands based on surveyors instructions at the time (see Bourdo 1956). Both the Banana River and Hontoon Dead Cr eek were once active channels, and their current relict status has resulted from avulsions at a date currently unknown. Avulsions in anabranching systems can occur preferentially during periods of increased flooding, channel reorganization, or aggradation of the floodplain associated with baselevel change (Slingerland and Smith 2004). More mundane and unpredictable processes such as a tree falls and vegetative cl ogging are also possible. In either case, flow is preferentially distributed to new segments or older paleochannels at lower elevations on the floodplain. Unlike river systems where avulsions in wide floodplains

PAGE 136

136 could resituate the main channel many kilome ters away, such as the Mississippi (e.g., Aslan et al. 2005; Kidder 2008), the effect of an avulsion along the St. Johns would only be locally felt due to the relatively narrow floodplains that are cons tricted at geomorphic basin boundaries. The Banana River and H ontoon Dead Creek channel wi dths are equal to or greater than the current main channel, su ggesting similar amount of flow passed through them at some point in time (Drury 1976). Hont oon Dead Creek in particular contains a number of meander scars suggesti ve of higher-than-present stream power. That a channel or other kind of water body was situated closer to Hontoon Island is evidenced by the presence of the Hont oon Dead Creek Mound some 200 m distant from the main channel (Fi gure 3-11B). The 7000-year-old anthropogenic shell deposits beneath organic sediment off the western edge of the mound atte sts to at least a meter of floodplain aggradation since de position first occurred there. Live Oak mound on the eastern terrace of the main channel is simila rly positioned, indicating that either the river was closer to the mound or the floodplain has aggraded. A variety of floodplain ecological comm unities are situated along the main channel today (Figure 3-10). These communiti es reflect different hydroperiods (Mace 2007). Slight variations in floodplain elev ation derived from substrate anomalies, alluvial features, and in filled channels of vary ing maturity can result in very different resource distributions (Hupp 2000). Thus, the distribution of vegetation communities represented here likely reflect more recent fluv ial activity. During a recent survey of ecological community structur e at sites along this reach, Mace (2006) found that elevation and vegetation were correlated. Most widespread are bottomland hardwood

PAGE 137

137 swamps characterized by abundant red maple ( Acer rubrum ), bald cypress ( Taxodium distichum ), pop ash ( Fraxinus caroliniana), and American elm (Ulmus americana ) at elevations below 0.46 m. These grade into hydric forests dominated by water hickory ( Carya aquatica ) and other species particular along the margins of islands and terraces. These bottomland forests eventually give way to upland forests with live oak (Quercus virginiana ) and other mesic species. Typically adjacent to the main channel or in backwater environments are found zones of emergent vegetati on and low-lying scrubshrub. Vegetation consists of water tolerant plants, such as cattails ( Typha spp.) or sawgrass ( Cladium jamaicense ) and common reed ( Phragmites communis ). The wetlands, lagoons, and channel segments provide habitat for a diverse array of fauna. Aquatic vertebrates such as alligator ( Alligator mississippiensis) otter ( Lutra canadensis) a variety of turtles, and upwards of 40 species of fish of economic importance to humans ar e present. Manatees ( Trichechus manatus) annually congregate at spring runs in the winter as wa ter temperatures decline, but these do not appear to have been exploited during the Arc haic period (Cumbaa 1980). While there are clear habitat preferences for many fish species, including open channel versus littoral zones, most can be found in any body of freshwater connected to the main channel and are available y ear round. Anadromous fish such as American shad (Alosa sapidissima ) and Hickory shad (Alosa mediocris) spawn up to Lakes Monroe and Harney, typically between November and Ma rch (Harris and McBride 2004). Striped mullet (Mugil cephalus) are also found in permanent re sident populations in some springs, and other saltwater species such as Atlantic stingray ( Dasyatis sabina ) and blue crab ( Callinectes sapidus) can also be found within spring runs (Walsh 2001). The

PAGE 138

138 flatwoods and hardwood hammocks of the uplands and islands provide habitat for numerous terrestrial fauna. Those of ec onomic importance to humans include whitetailed deer ( Odocoileus virginianus ), black bear ( Ursus americanus ), raccoon ( Procyon lotor ), opossum ( Didelphis virginiana ), gopher tortoise ( Gopherus polyphemus ), and turkey ( Meleagris gallopavo ). Numerous species of birds, mammals, reptiles, amphibians, and gastropods also inhabit these zones. Channels and springs are habitat for many molluscs (Walsh 2001). Species most frequently exploited by Archaic inhabitants include the gastropods Viviparus georgianus (banded mystery snail) and Pomacea paludosa (Florida apple snail), as well as freshwater bivalves (Unionidae) Smaller gastropods such as Elimia sp. (rasp Elimia), and the rams horn and mesa-rams horn ( Planorbella sp. ) can be found with these other species. In general, all species prefer shallow near-shore environments, such as grassy marshes and shallow lagoons (Qui tmyer 2001). It is unknown in what frequencies these species normally co-occur, or whether there is predictable ecological variation in their availabili ty. Unfortunately, little deta iled information on the habitat preferences, habit, and seasonal lif e histories of many of these species is available. Given that Viviparus frequently constitute the majority of shell mounds by volume (Cumbaa 1976), it is particularly vexing and curious that Viviparus colonies are difficult to find today, and infrequently occur in macrobenthic invertebrate surveys (e.g., Phelps et al. 2006). Moore (1894c:123) indicated that Viviparus were widely available before the turn of the century, stating that whol e lake bottoms are covered with them, while they are found in abundance in numer ous creeks and lagoons. Molluscs are particularly susceptible to groundwater contamination and in creased suspended

PAGE 139

139 sediment loads, and historic land use prac tices have no doubt caused significant harm (Bogan 1993; Walsh 2001). Viviparus prefer soft, muddy subs trates with slack water, such as lagoons, creek edges, lakes, and sp rings (Clench and Turner 1956). Apple snails are better known, due to their status as a primary food source for the Florida Snail Kite. Apple snails are known to prefer marshes with emergent vegetation, typically with at least 50 cm of water (Dar by et al. 2002). Annual draw downs in stage level are necessary for successful breeding, and they can survive exposed for up to 12 weeks during extended droughts by burrowing and aestivating (Darby et al. 2009). A consideration of nearby wetland histories indicates the contemporary hydroperiod within the St. Johns did not emerge until after 7000 years ago, and was not fully established until even later (Figure 3-4C). Peat deposits preserved in the lakes and wetlands of central Florida suggest a significant transformation in atmospheric circulation between 7000 cal B.P. Data from pollen analysis supports a general trend of dry to wet conditions evidenced by oak-dominated assemblages replaced by wetter pine dominated vegetation communiti es during the middle Holocene. These changes are indicative of longer hydroperio ds and likely the establishment of higher water tables (Watts et al. 1996). Several pollen cores have been analyzed in the vicinity of the Middle St. Johns. Watts (1969) reported an analysis of a sediment core from Mud Lake. This is a shallow open water la ke which drains into the Ocklawaha River today, at higher elevations to the east of t he study area. Two sedimentary zones dating to the Holocene were recognized, with a basal date between 9520 cal B.P. (2 ) for Zone M2 and 6190 cal B.P. (2 ) for the stratigraphically superior Zone M1. The lower M2 zone was dominated by oak and herbaceous Chenopodiaceae-

PAGE 140

140 Amaranthaceae pollen (hereafter Cheno-ams), indicating drier-than-present conditions. Although Watts could not find a modern analog for this pollen community, he suggests that it represents a dry oak forest or scrub, with marsh grasses and occasional open water patches. The transition to Zone M1 was characterized by an increase in pine, cypress, hickory, and other taxa typical of the pine-dominated uplands and bottomland hardwood forests present in that area today. Similar results have been reported from pollen and peat analyses at the Windover site (8BR246) in the Upper S t. Johns basin. The Windover site is a small pond that was utilized as a mortuary between roughly 9000 cal B.P. (Doran 2002c), the social significance of which I will examine in C hapter 4. Sediments in the pond were composed of at least five distinctive peat deposits that span the Holocene. The Upper Red Brown Peat has a near-basal date between 9550 cal B.P. (2 ), and spans the length of mortuary use at the site. Pollen is dominated by Cheno-ams, almost to the exclusion of other taxa, while pine and oak are present. Holloway (2002) interprets this trend as reflecting increased aridity, and possibly indicative of grassy salt flat vegetation communities and low water tables. Analysis of the peat itself revealed the development of a hardwood swamp community that was seasonally flooded with less than 30 cm of water (Stout and Spackman 2002:235). The hi gher Black Peat has a near basal date ranging between 7170 cal B.P. (2 ). This assay places the beginning of peat deposition in this regime roughly coeval with the onset of the Mount Taylor period in the Middle St. Johns. Palynologically the transition is abrupt, with pine, oak, bayberry and grasses increasing in visibility at this time in concert with a dramatic decrease in Chenoams. Holloway interprets this vegetation shift as reflecting the onset of cooler and

PAGE 141

141 moister conditions. Analysis of the peat determined that the sedge-grass community that typifies the area today emerged at roughl y this time (Stout and Spackman 2002). No peat deposits have been subjected to pollen analysis within the Middle St. Johns. However, their presence is widel y attested adjacent to shell mounds, where they typically are found superimposed upon or intercalated with subaqueous anthropogenic shell deposits (Aten 1999; McGee and Wheeler 1994; Sassaman 2005a; Wheeler et al. 2000). Groves Orange Midd en (8VO2601) is the subaqueous extension of the Old Enterprise (8VO55) site, and is char acterized by saturated shell matrix layers, intercalated within a well-developed peat. Ba sed on the preservation of organic objects, the excavators have argued that the anthropogenic deposits were deposited subaqueously, with materials likely discarded at the shore of Lake Monroe. Dates from throughout the deposit demonstrat e that occupation occurred there near the onset of the Mount Taylor period, as early as 6960 cal B.P. Excavations within this wetsite component identified a well developed peat layer situated between shell matrix, with dates above and below the peat indicating it developed between 5900 540 cal B.P. at the maximum 2 range of the assays. Because the peat was not subjected to further analysis, little can be said about what kind of vegetative community was responsible for its formation, and thus what kind of hydroper iod it may represent. Regardless, the initiation of peat development is relatively coeval with the 6000 ca l B.P. RCC identified by Mayewski et al. (2004), and consonant with the temperature variation recognized by Viau et al. (2006). A Dynamic Land There is no master solution or climatic ev ent to which shellfishing and associated traditions can be attributed to. Instead, a re view of paleoproxy da ta demonstrates that a

PAGE 142

142 variety of interwoven processes were at work through time (Figure 3-4). These processes influenced the organization and devel opment of the St. J ohns floodplain and associated terraces. Given the lack of ch ronologically sensitive paleohydrological indicators along the St. Johns, it is difficult to detail exactly how the St. Johns would have registered changing cond itions. Since most paleohy drological studies have emphasized interior drainages, models for detailing hydrological responses in lowgradient systems are poorly dev eloped. For example, bas e-level change associated with a relative increase in sea level can resu lt in the drowning of low-lying reaches of river segments and the reduction in co mpetency and development of backwater marshes. In turn, such transformations have implications for aquatic resource distribution, diversity, and sustainabili ty (Brown 1996, 2003; Morey et al. 2002). However, such changes may be relatively isolated to near coastal regions and have little effect on up-stream hydrological r egimes (Koss et al. 1994; Leigh and Feeney 1995). Consequently, only portions of lowland ri vers may significantly register base gradient changes. Similarly, perturbations in the fre quency of higher-magnitude rainfall events or evapotranspiration can alter channel gradien ts, morphology, and associated vegetation regimes (Starkel 2002). Modeling the paleohydrology of large-basins and complex fluvial systems such as the St. Johns River can also be hampered by non-stationarity, where hydrological response to differing am plitudes of climate change is unpredictable or non-scalar (Coulibaly and Baldwin 2005; Knox 1993, 2000; Koutsoyiannis 2006; Redmond et al. 2002). Typically, large river systems are less likely to record subtle short term events or low am plitude processes than longer-t erm larger scale climatic

PAGE 143

143 trends (Kidder 2008; Knox and Daniels 2002). Fluvial histories along the St. Johns are further complicated by the relationship between surface flow, sea-level, and the potentiometric surface of t he FAS (Faure et al. 2002; Mill er 1992; Thulman 2009). Although there is not enough data to spec ifically model how the St. Johns and associated regions were organi zed at any given time, it is possible to suggest some potential scenarios of landscape change and st ability that can be examined through a more detailed consideration of Mount Tayl or places and traditions in subsequent chapters. The paleoclimate data are in accord with earlier interpretations of near modern conditions established at the onset of shellfishing. RSL rates were in the process of decelerating from the terminal Pleistocene, and there is evidence from Windover that a wetter and longer hydroperiod wa s on the verge of being established. However, a closer inspection of the condi tions surrounding shellfishing suggests that there was the pot ential for considerable variability operating at the level of the floodplain. Within the Middle St. Johns, lowe r than present sea leve ls still meant that water tables, and likely the springs that fed the Middle St. Johns, experienced decreased flow relative to modern times. However, spring discharge would increase relative to sea level rise between the period of 7300 and roughly 6000 cal B.P. The data from Hontoon Dead Creek Mound and Groves Orange Midden suggest that the floodplain was situated at least a meter, if not more, below its present elevation at 7300 cal B.P. However, there does not seem to be a corresponding aggradation of the floodplain with organic matter until much later. What we do not know at the moment is the relationship between this lowered floodpla in elevation, channel organization, and the vegetation communities associated with a lon ger hydroperiod. The presence of tree

PAGE 144

144 stumps at the base of the Groves Or ange sequence (McGee and Wheeler 1994) would suggest that hardwood communities at lower elevations were effectively drowned out roughly 7300 years ago. Whether this was abrupt or gradual is unknown, although the temperature fluctuations recorded by Viau et al. around 7300 cal B.P., and the synchronous development of t he black peat at Windover would indicate that it occurred rapidly. The transformation of entire ecologica l communities throughout the floodplain would likely open up large expanses of once-forested floodplain prior to the onset of a more stable hydroperiod, much like the lateral sections of the flood plain today that are characterized by emergent or scrub-shrub v egetation. The suggestive evidence for frequent flooding may mean that despite a lower floodplain and lower-than-present annual flow, the region woul d have experienced periodic flood events that raised the river stage to above normal bankfull conditions, flooding large expanses of the low floodplain. In this scenario, despite the lowe r water tables, there may have been more expansive open water conditions at lower elev ations during the summer and early fall in comparison with today. It is also likely that it was in this context that the anabranching pattern we see today emerged. With respec t to the higher storm frequencies, then we might expect that the avulsi ons would have been more likely to occur. As a result of these processes, we would expect that local embayments that were the habitat for shellfish and other species of economic importance would be subject to potentially frequent relocations. The ongoing cycles of flood ev ents, rising water tables and potential temperature fluctuation seems to have persisted until roughly 6000 cal B.P., and coeval with

PAGE 145

145 Mayewski et al.s RCC. It is at this ti me that sea level decelerated, there was an increase in the frequency of ENSO events, and a decrease in major storm events. There was a moderate temperature increase as well, although whether or not this is reflected in the southeastern United States is unclear. It may also be the case that flow from the FAS attained its present regime around this interval. The local consequences of these changes appear to be manifested in floodplain aggradation, as witnessed in the development of peat deposits at Groves Orange Midden. This process could be interpreted in two ways. The excava tors of Groves Orange Midden (McGee and Wheeler 1994) felt that the peat reflected a lower-than-present sea level stand, because they assumed the materials that had a ccumulated subaqueously were deposited in relatively deep water. The peat then wo uld have formed as a consequence of a marsh developing in what had once been open water. The scenario that I find more likely, given what we know now, is that the peat was enabled to form by a persistent and stable hydroperiod. Although relating to lower-than-present water tables, the normalized flow regimes allowed the marshland to be established, leading to the development of stable vegetation communities in areas that had once been open water. In this scenario, the floodplain began to aggrade through the devel opment of organic deposits throughout the basin. Eventually, t he floodplain would stabilize to a higher water budget. The end of the peat sequence at Groves, roughly 5000 years ago, would again appear to signal a regime change along the St. J ohns. Of particular inte rest is the fact that the Groves peat is superimposed with highly fragm ented shell and sand deposits which McGee and Wheeler (1994) interpreted as evidence for storm deposits. Given

PAGE 146

146 the lack of storm deposits elsewhere, th is may signal a one time event, that was perhaps a significantly higher magnitude th an has been recorded along the St. Johns. The long-term significance of th is storm event is less clear, other than the fact that it does correlate stratigraphically with the end of peat formation and a return to shell deposition locally. Given that both well-preserved organic remains and terrestrial gastropods were recovered in these upper units, this zone was alternatively flooded and dried out on an annual basis (Russo et al. 1992). If this is the case, then the region may have experienced lower stage levels. This period is consonant with the end of the postulated RCC as well. Correlating the end of peat formation to larger scale processes is less clear than before. However, if we take the evidence from the Everglades for example, this would have been a time in wh ich essentially modern hydroperiods were established, and roughly consonant with the onset of the near-modern frequency and amplitude of ENSO events gl obally. As a consequence, channels along the Middle St. Johns may have finally become entrenched, and subject to processes of lateral migration. Contrary to orthodox models, the present review highlights local and regional variability in landscape stability. The alte rnative scenarios of hydrological development point to different temporalities in the physical landscape through time. Although only a broad outline, they can can be used to examine concomitant processes in landscape inhabitation. It is in this milieu of a complex and dynamic env ironment that Mount Taylor communities emerged.

PAGE 147

147 Table 3-1. Minimum and maximum flood sta ge frequencies for the St. Johns at Deland USGS gaging station Event Return Frequency (Years) Discharge (m3/s) Stage (m NAVD88) Hydroperiod Source Minimum Average 1.5 58 -0.05 Typically Saturated Robinson (2004:81) Minimum Frequent High 3 130 0.29 Seasonally Flooded Robinson (2004:81) Minimum Frequent Low 5 31 -0.20 Semipermanently Flooded Robinson (2004:81) Flood 5 294 0.89 n/a Rao (1986:Table 13) Flood 10 354 1.23 n/a Rao (1986:Table 13) Flood 25 433 1.35 n/a Rao (1986:Table 13) Flood 501 479 1.55 n/a Rao (1986:Table 13) Flood 1002 535 n/a n/a Rao (1986:Table 13) Flood 5002 663 n/a n/a Rao (1986:Table 13) 1The 50-year stage was averaged from discharge measurem ents of 453 and 484 m3/s between 10/12/1953 and 10/14/1953. A dischar ge of 484 is the highest on record. 2Rao extrapolated discharges for these infre quent events. There are no corresponding stage values.

PAGE 148

148 Figure 3-1. Late nineteenth and early twentieth century impacts to the St. Johns River region, photographs and captions courtesy of the State Ar chives of Florida. A) Group of people digging in shell mound: Lake Monroe, Florida, 1896, most likely the Old Enterprise site (8VO55). B) Cypress trees ready for logging, 1929. C) Government dredge on the Ocklawaha River, 1892.

PAGE 149

149 Figure 3-2. Distribution of known Mount Taylor and contemporaneous Preceramic Archaic shell matrix sites in north east Florida (after Wheeler et al. 2000:Figure 11), with emphasis on those sites discussed in Chapters 3 and 4. Also included is the Windover site, an ear lier non-shell matrix mortuary pond.

PAGE 150

150 Figure 3-3. Hillshaded topographi c map of the eastern Fl orida Platform emphasizing surface hydrology, karst areas (after Sc ott et al. [2004:Figure 6]), and the distribution of chert quarry clusters (after Endonino [2007:Figure 13]) in relation to the study area.

PAGE 151

151 Figure 3-4. Environmental processes in relation to cultural periods along the Middle St. Johns River. A) Culture-historical periods. B) Onset and duration of shellf ishing (this study). C) Hypothetical St. Johns floodplain hydroperiod change inferred from paleoproxy data (this study). D) Rela tive strength of ENSO signal (Donders et al. 2005; Moy et al. 2002). E) Relative flood frequency and stor m magnitude (Goman and Leigh 2004; Liu and Fearn 2000). F) Sea level curves for the Suwannee River delta (modified from Wright et al. 2005) and Gulf of Mexico (modified from Balsille and Donog hue 2004). G) Mean July temperature difference from modern based on pollen analysis (Viau et al. 2006). H) Rapid Climate Change events (Mayewski et al. 2004).

PAGE 152

152 Figure 3-5. Physiographic provinces of the St. Johns River Region in northeast Florida, data derived from the Florida Geographic Data Library (2005).

PAGE 153

153 Figure 3-6. Monthly rainfall and temperature averages for the years 1971 at Orlando International Airport, Flori da (Florida Climate Center 2009).

PAGE 154

154 Figure 3-7. Springs, lakes, and islands of t he Middle St. Johns River Valley. Elevation and stream gage available from the U.S. Geological Survey (2009), all other geographic data provided by the Florida Geographic Data Libr ary (2005).

PAGE 155

155 Figure 3-8. Daily streamfl ow-duration hydrographs st andardized by drainage area for the years 2005. Years begin in January. Stream gage data from the USGS (2009).

PAGE 156

156 Figure 3-9. Monthly str eamflowand stage-duration hy drographs (USGS 2009) and rainfall (National Oceanic and Atmospheric Administration 2006) for the years 1957 at the St. Johns near Deland, Florida gaging station. A) discharge. B) stage height.

PAGE 157

157 Figure 3-10. Channels and wetlands along the Middle St. Johns River Valley, from the Wekiva River outlet (south) to Hontoon Island (north). The lowest contour elevation is 1 m amsl.

PAGE 158

158 Figure 3-11. Distribution of wetlands (U.S. Fish and Wildlife Service 1992) in relationship to channeled surface water and elevation (Volusia County Public Works Department 2006) along the Middle St. Johns River, from the Wekiva River outlet (south) to Hont oon Island (north). A) plan view. B-D) extr acted transects depicting elevation and vegetation type, vertical ex aggeration approximately 230x.

PAGE 159

159 CHAPTER 4 A SHORT GENEAOLOGY OF MO UNT TAYLOR TRADITIONS The Mount Taylor period wa s defined on the basis of sh ellfish-dominated matrices, a suite of artifact types, and notably a lack of pottery (Goggin 1952). This definition has been further refined (e.g., Wheeler et al. 2000), and some have suggested new culturehistorical divisions within the Mount Tayl or period on the basis of mortuary traditions and exchange relations (e.g., Beasley 2008; Endonino 2008). While increased attention to chronological precision is an important turn in regional res earch, the taxonomic approach has the potential to obscure geneal ogies of practice and histories of communities at multiple scales. The prec eding review of the ecologies of the middle Holocene St. Johns has outlined processes operating at multiple scales that defy easy separation. Instead of a taxonomic approach, in this chapt er I take a multiscalar and historical perspective to trace the hist ory of Mount Taylor communi ty traditions. I emphasize some of the better known el ements of Mount Taylor lif eways, including regional settlement and mobility, subsistence, technol ogy, exchange, and mortuary traditions. However, I want to draw attention away from Mount Taylor as a typological entity, and instead examine what we define as Mount Taylor along the St. Johns should be understood as a historical product of comm unities engaging in relationships between central Florida, the Atlantic and Gulf of Me xico coasts, and beyond. To this end, I consider in brief the lifeways of the regions inhabitants prior to the onset of shellfishing and those on the margins of the St. Johns. I will conclude the chapter with a brief statement on tradition s of the Orange period. Many of the processes supposed to characterize the Mount Taylor period hav e actually been derived from analyses of

PAGE 160

160 Orange period lifeways. In rethinking the M ount Taylor period, ho wever, our current interpretations of their descendants require considerable revision. Amidst the Rising Waters The organization and history of early occupa tion prior to shellfishing in the St. Johns is largely a black box. Few sites dating before the Mount Taylor period have been discovered. The extent to which this is a factor of preservati on bias or represents a true pattern of limited occupation is cu rrently unknown. This is an unfortunate situation, as it is difficult to track what kinds of innovations communities may have made across the shellfis hing divide. The deep history of the region was written by communities of the late Plei stocene Paleoindian traditions including Clovis, SuwanneeSimpson, and Dalton and various Early Holocene traditions identified by Side-Notched and Corner-Notched Bolen hafted bifaces (Bullen 1975). Given the drier atmospheric conditions and lower water tables, much of in terior Florida was characterized by xeric vegetation. Outside of the Middle St. Johns River Ba sin, water was a precious economic and symbolic resource. Sites of these time per iods are typically restricted to inundated contexts such as drowned river segments (Dunbar et al. 1988; Faught 2004; Thulman 2009), sinkholes (Clausen et al. 1979), or perched basins and depressions (Daniel and Wisenbaker 1987; Neill 1964; Sassaman 2003b). Noting the co-occurrence of Paleoindian artifacts and ka rst topography in northwest Florida, Dunbar and Waller (1983) posited the Oasis hypothesis, that in effect Paleoindian populations were tethered to karst regions, abundant in toolst one and reliable surface water. In the St. Johns Basin, early sites are expected to occu r adjacent to first-magni tude springs fed by the FAS (Miller 1998:84; Thulman 2009). The few known sites and isolated finds seem

PAGE 161

161 to fit this overall pattern (Sassaman et al. 2000). A survey of Crescent Lake, a perched water source, has demonstrated that ther e is great potential for recovering early assemblages in the region (Sassaman 2003b) Thulman (2009) has also mentioned a large Paleoindian encampment on a paleote rrace or dune beneath Lake George. The extent to which occupation was tethered to these water sources is unknown. Regardless, it should not be presumed that the St. Johns was vacant prior to the initiation of shellfishing. Between 9000 and 7000 rcybp Floridas Ea rly/Middle Archaic traditions also remain poorly defined (Austin 2004; Milanich 1994). As before, our knowledge of this period is largely derived from hafted bifa ce technology. Stemmed hafted bifaces, consistent with the Kirk Stemmed type and locally referred to as Kirk, Wacissa, Hamilton, and Arredondo (Bulle n 1975) are distributed throughout the North, Central, and Gulf Central portions of the state, often in similar localit ies as early forms (Milanich 1994). This period also witnesses the est ablishment of a longstanding tradition of aquatic mortuaries, wherein individuals were in terred in shallow bodies of water such as ponds or sinkhole margins. Windover Pond in the Upper St. Johns basin represents the earliest so far identified, and is the most thoroughly investigated mortuary pond in the region (Doran 2002c), the significance of which I will discuss below. Following these poorly documented periods, across much of Peninsular Florida researchers have recognized the Middle Arc haic Newnan Horizon, characterized by short, narrow stemmed, broad bladed chipped stone hafted bifaces including Newnan and others forms (Bullen 1975; Mila nich 1994:76). There is si gnificant variation in the form of stemmed hafted bifaces from this pe riod, leading to a less formal designation of

PAGE 162

162 the Florida Archaic Stemmed type, which includes any broad-bladed stemmed hafted biface. Lithic artifacts during this period were typically manufactured from thermally altered chert or silicified coral (Ste. Claire 1987). Radiocarbon assays place Newnan sites as early as 7000 cal B.P. and likely earlier (Clausen 1964; Milanich 1994:77), although similar forms were likely produced throughout the Archaic. Settlement in interior Florida, particularly the Central Highl ands, which contains much of the available chert and silicified coral for the production of stone tools, is characterized by a dichotomy between large sites with diverse a ssemblages and small lithic scatters. The large sites have been interpreted by Milani ch (1994:79) as indicative of reduced seasonal mobility. Austin (2001) suggests however, that the larger sites likely represent more intensive short-term reducti on episodes near raw material outcrops. Several quarries have also been identified in the Central Highl ands, including the Senator Edwards site (Purdy 1975). In contrast to the widespread, if poor ly understood, evidence for inhabitation within interior Florida on the cusp of the Mount Tayl or period, direct evidence for occupation or interaction on the Atlantic coast lags by se veral millennia. The lack of early coastal populations was once ascribed to the paucity of marine resources during the preceramic era (e.g., Goggin 2952:67). Ste. Claire (1990) argued that Archaic coastal occupation has been significantly underestimated, due in large part to the s ubmergence of the coastal strand during the early Holocene. So far, the earliest shell matrix sites on the coast date no earlier than 6150 cal B.P., identified at Spencers Midden in the Timucuan Preserve in the Lower St. Johns Basin (Rus so 1992). There is abundant, albeit indirect, evidence for coastal contact or seasonal mobili ty involving coastal resources. In the

PAGE 163

163 millennia prior to the emergence of shellfishing along the St. Johns, the communities associated with the Windover mortuary were acquiring obj ects of marine origin, including shark teeth, whelk ( Busycon ) shell, and Marginella shell beads (Dickel 2002; Penders 2002). Although these may have been acquired through exchange, analysis of stable isotopes from the i nhumations at Windover has dem onstrated that individuals were also consuming marine resources (Tu cker 2009; Tuross et al. 1994). Today the site is some 8 km from the coast, and so the presence of marine resources should not come as a surprise. Although Windovers lo cation in relation to the early Holocene shoreline is unknown, it was a greater distance back then. Windover communities also traveled to, or had contact with, communities farther afield. Amongst the grave inclusions were four bifaces, one biface fragment, and one lithic flake. All were manufactured from raw material sourced to the Panasoffkee cluster, 130 km away to the west (Dicke l 2002). Although small in quantity, these objects hint at biographies that ranged across the Florida Peninsula. I would suggest that we need to consider that the coastal resources docum ented at Windover may have come from the west coast as well. Evi dence for early marine subsistence is also emerging on the northwest Gulf Coast. Al though coastal occupation no doubt occurred on the Atlantic and Gulf margins, the earliest preserved ev idence comes from the Gulf Coast which experienced a less destructive, low-energy wave environment. Along the PaleoAucilla River, Faught (2004) has identified diagnostic Middle Archaic stemmed hafted bifaces submerged to a depth between 4 and 5 m. Also found at the site were beds of disarticulated oyster shell, dated as early as 7000 cal B.P., which Faught

PAGE 164

164 suggests may be indicative of marine procur ement. Equivalent coastal settlement at this time was likely in Tampa Bay and points south as well (Russo 1996b). Mount Taylor Chronology Until the last decade, little was known of the chronology of Mount Taylor practices due to a lack of abundant or widely spaced radiocarbon determinations. Recent research and mitigative projects have gener ated a large enough sample to contextualize practices through time. The absolute chronology of Mount Taylor practices within the Middle St. Johns is defin ed by a total of 52 radiocarbon assays from 11 sites (Appendix A). This total incl udes those previously published and those reported in this study. Site s represented in this analysis span the the study area, and include (from north to sout h) Silver Glen Run Complex (8LA1), Bowyers Bluff 2 (8VO88B), Harris Creek Complex (8VO24) Hontoon Island North (8VO202), Hontoon Dead Creek Mound (8VO214) and Village (8 VO215), Live Oak Mound (8VO41), Blue Spring Midden B (8VO43), Lake Monroe Ou tlet Midden (8VO53), Groves Orange Midden (8VO2601), and the Thornhill Lake Comp lex (8VO58). One other date from the Tomoka Mound Complex (8VO81) on the in tercoastal waterway is also included for comparison. All samples were recovered from within shell and/or sand matrix strata. I have also provided selected assays (n=15) from Orange period components as a point of comparison. These include samples from shell matrix strata and vessel soot. I have excluded all assays on materials other than charcoal, nutshell, marine shell, or human bone. Throughout the years authors have attempted to directly date freshwater shell (e.g., Bullen and Bryant 1965; Johnson 2005), or use chronometric techniques other than radiocarbon (Bullen an d Sackett 1958). However, all freshwater shell dates are affected by the old carbon problem typical of regions downstream of

PAGE 165

165 karst bedrock. As a result all determinations directly on freshwater shellfish remains return significantly earlier results than their ac tual age. Although it is possible that a calibration curve could be developed, none has been forthcoming as of yet. All radiocarbon assays were calibrated in Calib 5.0 (Stuiver and Reimer 1993). For samples from terrest rial materials, the IntCal04: No rthern Hemisphere (Reimer et al. 2004) calibration curve was used. For samp les from marine materials, the Marine04 (Hughen et al. 2004) marine calibration curve was used (Reimer et al. 2002; Reimer et al. 2006). A local correction of 380 years was added to uncorrected 14C determinations prior to calibration, and a marine reservoir correction factor of R33 16 was applied to marine samples based on the nearest source (14CHRONO Centre 2010). No modeling was conducted during calibration, and each assa y was considered independently. One final caveat regarding the calibration of human bone is required here. Stable isotope analysis by Tucker (2009) of human remains from Harris Creek determined that individuals were consuming marine foods as part of their annual die t. The consumption of marine resources introduces depleted carbon into human bone, as it does for other consumers (Meiklejohn et al. 1998:204). In this analysis, no marine reservoir correct factor was applied on human bone sample s. However, there are discrepancies between Tuckers (2009) assays on human bone fr om Harris Creek, and those acquired by the original excavators who sampled c harcoal from the mound fill. The charcoal assay intercepts clustered within a twohundred year span (ca. 6200 cal B.P.), suggesting a relatively short use life (Aten 1999). Because of large sigmas, however, these charcoal assays place construction of the mortuary sometime between 6900 5600 cal B.P. Assays of hum an bone from the mortuary laye rs acquired by Bryan

PAGE 166

166 Tucker (2009) yielded a similar range between in tercepts, but which were much earlier and restricted in time, between 7240 ca l B.P. The human bone assays place mortuary construction roughly coeval with the regional onset of shellfishing. The reality likely lies somewhere in between. The Harris Creek assays on human bone may be younger on the order of mo re than a century. The radiocarbon assays will be referred to th roughout this study. However, there are several patterns worthy of note up front. At the scale of t he region, the most compelling argument for continuity is the di stribution of assays through time. Presented in Figure 4-1 are the 2 -calibrated assays from Archaic components, ordered chronologically by the median probability intercept. As a general rule, the median intercept is a poor measure of centrality because probability distributions can have multiple intercepts (Telford et al. 2004). Thus, the median should not be used for point sampling or estimates. However, the m edian can be used to order assays for general comparisons. In this gr aph there are no temporal gaps over the 3700 year time span that includes both Mount Taylor and Orange per iod components. While the sample size is relatively small, a stepwise comparison of assays within 2 intervals suggests that there were no obvious regional hiatuses. If the temporal di stribution of mirrors human occupation, then any social transformations identified can not necessarily be explained by regional abandonment. I would also note that the initiation of shellfishing is geographi cally extensive and event-like from the perspective of the assa ys. Shellfishing in the region appears abruptly, as early as 7300 cal B.P. Although the earliest assays from shell matrix are situated at two nearby locations, the H ontoon Dead Creek Complex and Live Oak

PAGE 167

167 mound, assays falling within the 2 range of this early date are found as far south as Groves Orange Midden and potentially as far north as Harris Creek. The implication is that the region was rapidly inscribed with the deposition of shellfish in localities beginning 7300 cal B.P. Finally, a consideration of samples ordered by latitude from north to south reveals several temporal and spatial trends (Fi gure 4-2). The only long-term sequence spanning the entire Mount Taylor period was reported for Groves Orange Midden (McGee and Wheeler 1994). As noted before, this site is a subaqueous extension of the Old Enterprise Complex. All radiocar bon assays were from subaqueous strata and present a largely continuous sequence. However, at least one shift in deposition occurred between 5900 and 4540 ca l B.P., possibly reflecti ng floodplain aggradation (McGee and Wheeler 1994). Despite the longterm reoccupation of this place, assays collected from elsewhere in the region reve al much more punctuated inhabitation. Assays from 8LA1W-A, 8VO202, 8VO214, and 8VO53 indicate that habitation was of variable duration, with ranges between 200 and 400 years. There are, however, two important points that are rev ealed by the disparate chronolog ies of shell matrix sites. First, we should not expect that every shell matrix site in the region was active at any one point in time. The other point is mo re mundane. Our knowledge of Mount Taylor traditions is tied to poorly contextualized r egional surveys and more intensive, localized research. As a result, we are left with a series of snapshots that do not necessarily reflect change through time at the regional level. Settlement Mobility The annual movements of Mount Taylor communities loomed large in their experiences engaging in residential and ritualized fi elds of practice. Currently, there is

PAGE 168

168 no clear answer as to what the annual mobili ty or residential permanence of inhabitants may have been throughout the precer amic Archaic (Wheeler et al. 2000). A review of the current data sugge sts there were a number of pathways that communities took throughout the duration of the period, and even between years. The current state of our knowledge regarding these patterns reflects as much the preconceived notions of researchers as it does processes of the past. In this respect, the settlement patterns of Mount Taylor communities has been the subj ect of some debate, with the seasonality, range of mobility, and long-term stability of occupation in any one place the primary points of concern. Prior to the 1970s, there was little intere st in determining the seasonality of regional habitation, although Wyman (1875:79) did note when some species were most likely to be captured. Goggin as well ( 1952:66) does not address m obility directly, and only offers that the Archai c Tradition was characterized by a semi-sedentary group of people obtaining their food by hunting, fishin g, and gathering. The importance of the coast was considered minimal. Cumbaa ( 1976) was the first to address mobility patterns directly. His model framed Mount Taylor populations on an evolutionary pathway somewhere between mobile big ga me hunters and sedentary villagers. His primary assumption was that Mount Tayl or communities were oriented towards increasing their energy captur e efficiency, and used the ex tant ethnographic analogies of low latitude foragers as a guide. He s uggested that individual shell mounds (he was expressly interested in the largest sites) reflected a pattern of central-placed nomadism (1976:52), wherein one band composed of approxim ately 50 individuals would inhabit a location until daily collecti on and hunting forays resulted in diminishing

PAGE 169

169 returns. At the regional scale, this w ould result in an isolated band moving between shell mounds in a serial fashion. Underly ing the model was the assumption that local Viviparus colonies could be easily overexploited locally, but would rebound within 1 to 5 years. Thus, each mound would be occupied on a return interval of several years. Milanich and Fairbanks (1980:61,150) also deemphasized aquatic resources in their model of Mount Taylor and Orange period settlement mobility. They proposed that the Mount Taylor period refl ects gradual and increasing us e of aquatic resources by populations of the central hi ghlands. In their model, th e increased use of the river eventually gave way to more sedentary occ upation and marshaled in an era of village life by the Orange period. They envisi oned groups seasonally mi grating between the central highlands and the St. Johns River. By the onset of t he Orange period, they supposed that coastal resources would be exploi ted in the winter (on the assertion that shellfish would be more likely to spoil in the hotter months), with populations migrating into the St. Johns during the winter. S purring the change to freshwater shellfish collection was, in this model, a degradati on of oak-dominated forests in the Central Highlands that were replaced by pine-domin ated forests. Miller (1992,1998) also argued that environmental conditions were cent ral to the development of Mount Taylor lifeways. He asserted that the establishment of stable wetlands around 7000 years ago was enabled by higher relative sea level and a concomitant rise in the water table and spring discharge. Together, these enabled pe rmanent or at least seasonal occupation of the St. Johns Basin. Like Milanich and Fairbanks, he did not believe that coastal habitation was an important component of the annual round.

PAGE 170

170 In the 1990s, Russo and colleagues leveled several critiques at these earlier models, particularly as they regarded terminal Mount Taylor and early Orange period settlement. Instead of migrat ing populations, they envisioned different ecological zones inhabited by distinct and permanently settl ed populations. Along the St. Johns, the model was based on an analysis of flor a and fauna from Groves Orange Midden (Russo et al. 1992) (see below). At Groves they noted that the botanical assemblages indicated settlement between late summer th rough fall. Although the faunal remains could not be used to determine seasonality, they argued that dry-season occupation was also a possibility. The basis for t he argument was that they recovered both freshwater and terrestrial commensal gas tropod species in strata thought to be deposited underwater. They suggested that the terrestrial gastropods were incorporated into the matrix during annual dry-downs in the winter and spring. Although not decisive in and of itself, they noted t he presence of multi-seasonal occupations elsewhere, including the Upper St. Johns Basin (Sigler-Eisenberg et al. 1985), mouth of the St. Johns (Russo 1992, 1993), and the Atlantic coast (Ru sso and Ste. Claire 1992). What was perhaps most revolutionary in these analyses was the determination that coastal shellfish, particularly coquina ( Donax variablis ) were exploited during the summer months. Although these studies were based on Orange period components, Russo later (1996b) extended the model into the preceramic Archaic. As he noted, many of the sites in the Timucuan Preser ve, as well as Spencers Midden on the south side of the mouth of the St. Johns, contained precer amic components. Seasonality studies at Spencers Midden in particular dem onstrated that coquina were collected in the summer, oysters were collected in the winter, and estuarine fishes were captured in

PAGE 171

171 the spring through fall (Russo 1996b:190). T hese data accord well with considerable evidence for intensive and often multi-seas onal coastal occupations in Southwest Florida dating as early as 6000 cal B.P. (Russo 1996b). Tucker (2009) tested these alternative se ttlement models through a stable isotope analysis of skeletal remains recovered from Windover Pond and the Harris Creek mortuary. This sample selection allowed him to track changes between pre-shellfishing (ca. 9000 cal B.P.) and early Mount Taylor communities (ca. 7000 cal B.P.). In his analysis, he compared carbon and oxygen ratios of serially sampled teeth. This methodology allowed him to obt ain sub-annular resolution of isotopic variation that occurs during the development of mature dentition. Annual temperature and precipitation variation is regi stered by variations in the ra tio of oxygen isotopes, wherein enriched samples are indicative of warmer months. The consumption of marine foods are tracked by variations in carbon isotopes, such that enriched samples are typical of marine food consumption. In general, Tuckers analysis showed that the Windover population and the Harris Creek population were consuming marine foods during the wa rmer (i.e. summer) months, but were not routi nely consuming marine resources during cooler (i.e. winter) months. The correlations were stronges t amongst the Windover population, suggesting that Windover communities were moving from the interior to the coast during the summer months, and then returning to the interi or during the cooler months. The Harris Creek sample was more variable. Some individuals clearly followed the Windover seasonal pattern, while in others the relati onship was less robust. Although this may suggest more diverse mobility practices dur ing early Mount Taylor times, Tucker

PAGE 172

172 indicates that sampling errors may have c ontributed to the variance. Regardless, Tuckers research demonstrates that m obility patterns during the middle Holocene spanned the St. Johns region and the coastal z one, and this pattern had great antiquity. This is a reversal of Milanich and Fairbank ss model. The extent to which it is a challenge to Russos work is less clear. If the new radiocar bon assays from Harris Creek can be trusted, then Tuckers sample predates much of the preserved coastal record by up to 1000 years. In this view, t he onset of sedentary, or at least seasonally restricted settlements along the St. J ohns may have been a later phenomenon. Subsistence As the defining characteristic of Mount Taylor practices, surprisingly little is known of the temporality, diversity, or changes in subsistence prac tices through time. Prior to the 1970s, most approaches to Mount Taylor subsistence were defined by speculation as to the significanc e of shellfishing in gener al, and fishing specifically. Since Brintons time, archaeologists have recognized that shellfishing was a component of lifeways along the St. Johns. Wyman (1 875) presented taxonomic lists of species he identified amongst the shell matrix sites he observed, arguably one of the first zooarchaeological analyses in the United States (Trigger 1986). At the height of salvage excavations, Neill et al. (1956) reported on faunal remains from four Mount Tayl or sites, although they explicitly excluded shellfish and fi sh from their quantifications (Russo et al. 1992). They did not systematically sample, nor were they concerned with time. They did note that a total of 47 species were ident ified, all of which are found within the region today. Current debate and research centers on the importance of terrestrial, aquatic, and marine resources and the implications for seasonal settlement. The first concerted effort to model Mount Tayl or subsistence was Cumbaa (1976), who analyzed a small

PAGE 173

173 shell matrix sample. Because he could not find an adequately excavated sample from Mount Taylor contexts, he used an Orange peri od assemblage from the Colby Site. In his now discredited paper (i.e. Russo et al 1992), Cumbaa argued that shellfish would have contributed roughly 24 per cent of the daily intake of a Mount Taylor individual, vertebrates would consist of 43 per cent of t he diet, while fish would account for only 1.7 per cent. This interpretation was bas ed on a small sample of shell matrix, approximately 1 m3 in volume, from an Orange period component. The only size grade analyzed was greater than 1/4-inch mesh. T he results of this analysis were combined with total caloric values for the minimum num ber of individuals he identified in the sample, as well as an estimate of 55 per cent plants that would have been a be present in the sample as well. The plant estimate was derived from his survey of extant huntergatherer literature, although t here is no clear reason why he chose that particular contribution. For Cumbaa, Mount Taylor communities were primarily terrestriallyoriented hunters and gatherers, and in this sense were more like Kalahari San groups than comparative temperate zone hunter-gatherer s. Cumbaas analysis has been overturned by re cent zooarchaeological studies that have demonstrated the greater importance of aquatic resources. In a much needed paper, Russo et al. (1992) systematically co llected and analyzed a column sample from a saturated Late Mount Taylor /Early Orange period stratum at Groves Orange Midden. They noted that Cumbaas approach was flaw ed due to small sample size, the use of large size grades which would deemphas ize small vertebrates, and incorrect calculations of meat weight used to calculat e dietary contribution. In their analysis they examined all faunal remains greater than 1/16-inch, and used ratio and allometry to

PAGE 174

174 determine the contribution of species by meat weight. Based on their calculations, the column sample was dominated by aquatic species. Shellfish composed a total of 96 per cent of the sample by meat weig ht. The sample was dominated by Viviparus (58 per cent), followed by bivalve (27 per cent) and apple sna il (11 per cent). In contrast, fish and turtles contributed less than 2 per cent and mammals le ss than 1 per cent of the total sample. A subsequent preliminary stud y of samples spanning the Mount Taylor period at Groves Orange Midden by Wheeler and McGee (1994a) replicated the earlier results of Russo and colleagues in broad outline. One point of variance, however, was the relative contribution of shellfish through time. Although there was no temporal trend, they found that contribution of invertebr ates varied between 33 and 85 per cent. The dominance of aquatic species has been further demonstrated by zooarchaeological studies of later Mount Ta ylor samples from the Lake Monroe Outlet Midden (Quitmyer 2001) and late Mount Taylor and Orange period assemblages at Blue Springs Midden B (Sassaman 2003a: 127). In general, broad similarities in species captured and exploited ecological zones are evident. Zooarchaeological samples have recognized that a diverse array of species were collected. Fishes include catfishes (Ictaluridae), sunfishes ( Lepomis sp. ), gars (Lepisoteidae), bowfin (Amia calva), largemouth bass (Micropterus salmoides) striped mullet, American shad, chubsucker ( Erimyzon sucetta ), and American eel ( Anguilla rostrata ). Most of these species prefer shallower, near shore environments (Quitmyer 2001). In many cases, freshwater turtle is relatively abundant, including the soft shelled turtle ( Apalone ferox ), pond turtles ( PseudemysITrachemys sp .), and mud/musk turtles (Kinos ternidae). Other reptiles include alligator and various snakes. Alt hough rarer, birds such as turkey, various

PAGE 175

175 ducks ( Anatidae ), and other species have been recove red. Similarly, the typical mammalian assemblage includes opossum, rabbit ( Sylvilagus spp.), raccoon, otter, and white-tailed deer. Invert ebrate species include Viviparus, apple snail, and bivalve in varying abundance. The broad similarities in species composition mask subtle but likely important changes in local ecological conditions and exploitation patterns (Sassaman 2003a:153). For example, eel appears more pr evalent in Mount Taylor assemblages than in Orange period assemblages, which may relate to a decrease in river velocity through time (Sassaman 2003a:133). At Lak e Monroe Outlet Midden Quitmyer (2001) suggested that there was decreasing diversity in the exploitation of species, particularly shellfish. Although often overlooked, and difficult to qu antify in terms of relative importance, a wide array of plant remains were exploi ted by Mount Taylor communities. Where waterlogged conditions have enabled the preser vation of plant matter, such as at Groves Orange Midden (Newsom 1994; Russo et al. 1992) and Windover Pond (Newsom 2002), a consistent pattern characte rized by high diversity is established by no later than 9000 cal B.P. Mast resources such as hickory and likely acorn were exploited. Pulpy fruits such as black gum ( Nyssa sylvatica Marsh), prickly pear ( Opuntia humifusa ), saw palmetto (Serenoa repens), maypop ( Passiflora incarnata), wild plum ( Prunus sp.), blackberry ( Rubus sp.), persimmon (Diospyrus virginiana ), red mulberry ( Morus rubra), elderberry ( Sambucus canadensis ) and grape ( Vitis sp.) appear to have been the most important (Newsom 2002) These fruits were supplemented with starchy seeds such as amar anth, pigweed, and knotweed, as well as the greens from these and other species. Numerous tuber s were potentially eaten. Cabbage palm

PAGE 176

176 hearts and shelf fungi have also been i dentified (Newsom 2002) Undomesticated varieties of both bottle gourd and Cucurbita pepo gourd have been identified. The former may have been used as net floats, while the seeds of the latter could have been consumed. All of these species could be located within the floodplain or in the immediately adjacent mesic uplands (Newsom 1994). There are a number of caveats and questi ons regarding Mount Taylor resource exploitation. Because so few samples have been analyzed, there is currently no way to assess whether there were significant changes through ti me, or whether there are differences between social contexts of deposit ion. Despite the presence of shellfish within matrices, the degree to which shellfis h were collected for food or other purposes such as construction material is largely unknown. Tucker (2009) attempted to model the dietary contribut ion of shellfish by comparing the Windover and Harris Creek populations. He found that t here was no isotopic model solution that could account for the proportion of shellfish found within zooarchaeological samples, presuming that shellfish were a primary contributor to diet. He was able to identify a slight increase in the use of riverine resources between Windover and Mount Taylor times. However, the increase could not account for the relati ve abundance of shellfish documented within the shell matrix sites. T hat is, on isotopic grounds we cannot assume that shellfish were being consumed in the quantities sugges ted by zooarchaeological analyses. This is an important finding. It suggests Mount Taylor communities were collecting and depositing shellfish for purposes other than c onsumption. Technology The broad outlines of Mount Taylor te chnolog y have been described by Wheeler and McGee (1994b) based on the analysis of t he Groves Orange Midden assemblages,

PAGE 177

177 and further summarized by Wheeler et al. (2000). As demonstr ated at Groves, much of the toolkit of Mount Taylor communities wa s in place from the beginning of regional settlement, and shares similarities with the Windover material cult ure (e.g., Penders 2002). Mount Taylor period assemblages are typified by mundane and decorative material culture manufactured from locally available bone, fired clay, and wood, in addition to marine shell and various chipped st one tools (Wheeler et al. 2000). In most contexts, formal and informal chipped stone tools are rare. The hafted biface assemblages are consistent with the Newnan horizon. Aside fr om hafted bifaces, Mount Taylor lithic assemblages are dominated by unifacial tools that appear to have been used for a wide range of applications includ ing perforating, scraping, cutting, and wood splitting (Archaeological Consultants, Inc. [ACI] and Janus Research [JR] 2001). The rarity, fragmentary disposition and apparent rates of attrition on working edges would suggest that stone tools were of limited availability or utility (Purdy 1994a). The discovery of a dense lithic workshop at the Lake Monroe Outlet Midden, separate from the shell matrix (Scudder 2001), suggests t hat a sampling strategy focused on shell matrix may under represent activities that were traditionally spatially segregated away from loci of shell deposition. Marine shell tools and objects are rout inely documented in Mount Taylor assemblages. Most widely exploit ed were varieties of the whelk, Busycon sp. which are available on the Atlantic Coast. Many whelk shells we re modified to use as wood cutting tools. Adzes and axes were manufactured by pierci ng the whorl for insertion of an organic handle, while the sipho nal canal was cut or ground to a working edge. Celts or gouges were also manufactured out of whelk, in this case by removing the columella

PAGE 178

178 and sharpening the distal edge of the whorl. Additionally, ce lts were made from the lip of the Strombus gigas shell, only available in southern Florida (see below). Marine shell was also used to make containers. That many were used for cooking purposes is attested to by residue adhering to the inte rior surfaces, and many have thermally fractured bases. Decorative shell objects are also typical, and include marine shell beads and plummets made from large whelk columella. Oliva sp. shells were also modified, but their function is unknown. Fossilized and non-fossilized shark teeth are often recovered, and many have been drilled to fa cilitate hafting for use as a tool or as personal adornment. Aside from the ornamental objects, much of the inorganic toolkit was apparently geared towards manufacturing tools from loca lly available bone and wood. Bones from deer and other terrestrial animals were used to make a variety of tools. Many of the bone tools were likely associated with textile ma king, some of which was likely oriented towards fishing technology including gou ges, awls, needles, fids, net gauges (Wheeler and McGee 1994b). Interestingly enough, ther e are no known examples of fishhooks anywhere during the Archaic al ong the St. Johns, suggesting t hat many of the capturing techniques involved nets, weirs, or baskets. Socketed antler projectile points have been documented in many contexts, as have antler handles. Bone beads and pins have also been recovered. One of the remarkabl e finds at Groves Orange midden was an assemblage of wooden objects that had been preserved in the anaerobic conditions there. As detailed by W heeler and McGee (1994c), t he wooden inventor y of Mount Taylor assemblages was diverse, and included tool handles, net floats, stirring paddles, and canoe paddles, and canoes. Many of the w ood objects were made from either

PAGE 179

179 radial branches or split planks. Copious amounts of wood debitage attests to significant cutting and chopping as well. The importance of canoes framing and enabling community interaction and mobility probably cannot be overstated. Due to t he abundance of wetlands with anaerobic conditions, Florida is home to one of the highest dens ities of preserved ancient canoes in the world (Newsom and Purdy 1990). They have been recovered from throughout much of the state, and are often found in either springs or the margins of ponds. In the study region, the earlie st known canoe was recovered from De Leon Springs, and dates as early as 7000 cal B.P. Within the St. Johns, canoes typically occur as isolated finds, and on this basis it would be hard to quantify their presence in the past. However, Wheeler et al. (2003) documented 95 canoes on the margin of Newnans Lake in the highlands region. Of the 55 canoes subjected to radiocarbon dating, 37 were contemporaneous with the later Mount Taylor and Orange periods along the St. Johns (ca. 6000 to 3600 cal B. P.). All Archaic canoes were likely manufactured out of pine, using fire to hollo w out the interior. They averaged roughly 7 m in length, although one example likely exceeded 9 m. In contrast, they were narrow, averaging .58 m, and shallow, typically .30 m or less in height. Wheeler et al. suggest that the canoes would have been best suited for local transportation of people and minimal onboard cargo, as t hey had limited internal volume. Although they were not apparently designed for stability under weight, they were well-suited for fast travel amongst shallow lake s and streams. Exchange One of the defining c haracteristics of Middle and Late Archaic lifeways at the scale of the Southeastern United States was exchange involving a variety of objects

PAGE 180

180 (Claassen 1996; Jefferies 1996; Sassaman 1995). Mount Taylor communities were no different (Wheeler et al. 2000). The seemi ngly mundane and local content of much of the Mount Taylor assemblages belies consider able diversity in social interaction at increasing scales of Florida and the Great Southeastern United States. We have already seen how much of the inorganic assemb lages from Mount Taylor contexts were derived from the coasts (marine shell and other marine s pecies) and the interior (chipped stone). Limited sourcing of lithics from the Lake Monr oe Outlet Midden (ca. 6000 cal B.P.) dem onstrates that materials were being acquired from sources ranging from Tampa Bay in the southwest as far north as the Suwannee River Valley to the northeast (Endonino 2007) (Figur e 4-3). While this assemb lage occurs later in the Mount Taylor era, and may not be representative of earlie r pathways of exchange, the presence of lithic resources from the Lake Panasoffkee clus ter as early as Windover times would suggest a long-term pattern. The extent to which procurement was accomplished via exchange or embedded within mobility patterns has yet to be adequately assessed, however. Much of the marine shell inventory at M ount Taylor sites could have been directly acquired from the nearby Atlantic Coast. This is certainly true of many of the shell beads. In contrast, celts manufactur ed from the lip of the queen conch, Strombus gigas had to be acquired from communities inhabiting southern, and likely southeast Florida (Wheeler and McGee 1994b; Wheeler et al. 2000). Strombus celts have been recovered from places throughout the Middle St. Johns region. However, they occur later in the settlement history of the region. For example, all examples recovered by Bullen (1955b) at Bluffton were located in the upper portion of the sequence. At

PAGE 181

181 Groves Orange Midden, the celts were found only in layers post-dating 5000 cal B.P., and at Lake Monroe Outlet Midden they appear no earlier than 5600 cal B.P. These objects were no doubt utilized in woodworki ng activities, presumably for the production of canoes. They are frequently recovered fr om general matrix contexts, and so their local significance is ha rd to trace. Other objects point to connections farther afield, which in many cases resulted in complex transformations of value and biographies within the St. Johns. Of particular note are two classes of objects: st one beads and bannerstones. Groundstone beads have been recovered from several, chronolog ically young contexts, including Lake Monroe Outlet Midden (ACI and JR 2001), Silv er Glen Run Locus-A (Sassaman et al. 2010), and Thornhill Lake (Moore 1894b:277). Although their or igins are unknown, they are quite similar to tubular beads produced in Mississippi and the Mid south during the Middle Archaic (Brookes 2004; Crawford 2003). Fragments or beads of soapstone (steatite) have also been recovered from Groves Orange Mi dden and the Lake Monroe Outlet Midden, all of which had to originat e in the Piedmont region of the Lower Southeastern United States. The other revealing object class is bannerstones. Bannerstones are elongated groundstone objects perforated along their midaxis, and are associated with atlatl or spear-thrower technology (Sassaman 1996). Fr om a functional perspective, atlatls are characterized by an elongated haft, which frequently has a hooked inset at the distal end to receive a spear-shaft. In many cont exts beginning in the Early Holocene if not earlier, atlatls were constructed out of wood or antler. Begi nning in the middle Holocene, however, organic atlatls were embellished with groundstone bannerstones

PAGE 182

182 that would be attached along the end of the shaft or would replace the distal hook (Sassaman 1996). Throughout the Middle and Late Archaic there is considerable diversity in the form and final disposition of bannerstones. Some are incredibly ornate, and even hypertrophic, while others appear in a normative view as functional (Sassaman 1996). Bannerstones also appear to have had dual lives in the sense that they occur in both mortuary and non-mortuary contexts, although this pattern varies considerably with culture-historical context (Sassaman and Randall 2007). Bannerstones of various shape and raw material type have been recovered from a number of contexts along the St. Johns. All were in finished form. A few have been recovered from apparently non-mortuary contex ts, including Dillards Grove (Goggin 1952), Salt Springs Run (8MR1/2) (Neill 1954), Bluffton (Bullen 1955b; Neill 1954), Silver Glen Springs (8MR123) (Neill 1954) Lake Monroe Outlet Midden (Neill 1954), and Groves Orange Midden (Wheeler and Mc Gee 1994b), and one manufactured from the lip of a Strombus gigas shell was identified at Mount Taylor (Moore 1898). All others, however, appear to have been deposited in either burial mounds with individuals, such as at Thornhill La ke (Moore 1894b) and McDonalds Mound (Jones 1995), or in caches (with or without inhumat ions) as is the case at Coontie Island (Clausen 1964) and Tomoka Mounds (Dougla ss 1882) on the Atlantic coast. Some stone pendants also appear to be the latera l wings of bannerst ones that have been perforated, arguably for adornment Because there is no local stone suitable for their production, bannerstones were by necessity exchanged from the lower southeast. Those bannerstones recovered from Florida appear to have originated from the Savannah River Valley, one of several nodes

PAGE 183

183 of bannerstone innovation and production in the Southeast (Sassaman 2004b; Sassaman and Randall 2007). Most of the forms are typical of Sassaman and Randalls (2007) Phase II types In the source area they are associated with the Paris Island tradition, and dated between ca. 5200 ca l B.P. (Figure 4-3). Within the Savannah River Valley, bannerstones we re being manufactured and exchanged, possibly for shell beads, with communities in the lower Coastal Plain (Sassaman 2004b). In these contexts bannerstones do not appear to have been associated with mortuary practices. Dates of mortuary moun d construction at Thornhill Lake (Endonino 2008) and Tomoka (Piatek 1994) are roughly coeval with the Paris Island phase as well. Excluding the mortuary data, evidence for active local participation in the exchange networks outlined above has been difficult to gather. The recent excavations at Lake Monroe Outlet Midden site provide a possible glimpse into local production for exchange (ACI and JR 2001). This site is act ually composed of a large shell matrix component that fronts the St. Johns River (although now severely reduced), in addition to shell and non-shell activity areas higher up on the landform. In addition to a preponderance of debitage and other chipped st one tools, the excavators documented a large (n=236) assemblage of microlithic tools (ACI and JR 2001). These microliths are characterized by narrow distal working edges. Their function is not entirely clear, although the excavators argue on macrosco pic grounds alone that many may have been used as scrapers. I would like to s uggest, however, that these tools may have been involved in the production of shell beads. Elsewhere at the site, a large sample (n=266) beads of unknown shell were re covered. Although no associated shell debitage was recovered, it ma y be that the shell bead blanks were prepared elsewhere

PAGE 184

184 onsite. Lacking detailed microscopic examination of either t he tool working edges or the shell beads themselves, it is impossible to know if either was part of a larger exchange economy. I suspect that futu re research elsewhere in the study region may begin to unravel these patterns of local production. Mortuaries and Biographies If there is one attribute aside from shellfish that defines M ount Taylor it is traditions of reverence and respect paid to the dead. Unlike many shellfishing communities throughout the southeastern United States, where inhumations were made within the proximity of, or even underneath, comm unal areas, Mount Taylor communities separated the dead through inhumation into pr epared mounds of sand and shell. While I will examine the details of mo rtuary mound construction in C hapter 7, I want to briefly follow the thread of the discussion I have star ted here with regards to the composition of communities and their temporality surrounding ritualized sequences for the dead. The construction of mortuary mounds in one vein may appear like a radical transformation from the perspective of landscape inscription. Yet even before inhabitants of the St. Johns collected s hellfish, the region was already deeply imbued with ancestral presence. Ponds and springs se rved as the locus of interment. Mortuary ponds have been identified at sites widely dist ributed throughout Florida, including Bay West (Beriault et al. 1981), Little Salt Springs, Republic Grove (Wharton et al. 1981), and Windover. The majority of these mortuaries date to the middle Holocene, ca. 7600 cal B.P., and are loca ted in the western and sout hern portions of the state (Doran 2002a). A notable exception, however, is the Windover site lo cated in the upper reaches of the St. Johns basin. Be tween 9000 and 7900 cal B.P. at least 168 individuals were interred in peat deposits along the margins of this pond (Doran 2002b).

PAGE 185

185 Analysis of the burial patterns indicates a highly structure sequence (Dickel 2002). The deceased were interred rapidly after death, likel y on the order a day or two. They were wrapped in textiles and buried in shallow depressions excavated in the underlying peat and sand, frequently flexed on their left sides Included with some individuals were inclusions of bone, wood, or the occasional stone tool. Frequently wood was piled on top of the burial below the water level. Stakes were placed upright near the bodies, although it is unclear if thes e would have been visible above the surface of the water. Whether or not a residential area was a ssociated with the Windover mortuary is unknown, although at the later Little Salt Sp rings mortuary an unexcavated residential area was apparently situated close by (Clausen et al. 1979). From a historical perspective the si gnificance of place, and arguably its relationship to communities wa s maintained, if not materially transformed. Based on recent radiocarbon assays detailed above, the transformation to mortuary mound inhumations may be effectively synchronous wit h the onset of shellfishing, some 7000 years ago. So far, the best evidence for early mortuary practices has been revealed at Harris Creek, where Aten (1999) has reconstr ucted Jahn and Bullens (1978) salvage excavations into an early Mount Taylor mort uary. The mortuary it self was situated at the base of a much larger and multicomponent shell mound. In their excavations of the relict portion of the mound, Jahn and Bullen recovered a minimum of 175 individuals, which was likely less than 25 per cent of the total burial population. The interment of the dead involved arguably smal l-scale depositional acts, the media of which were materials such as white sand and shell. The presence of both secondary and primary burials, in addition to a possible mortuary stru cture indicates that the preparation of the

PAGE 186

186 dead was an extended performance, which st ands in contrast to prior mortuary programs in which the deceased were rapidly interred after death. Recent stable-isotope studies of Harris Creek burials in dicates that those interred in the mound had distinct spatial stories of their own. As reconstructed by Tucker (2009), the majority of individuals interred at Harris Creek were born and spent their childhood years somewhere near the St. Johns although a few individuals (n=4) were from as far away as Lake Okeechobee to t he south and Virginia or Tennessee to the north. Although individuals may have been mostly from the Florida peninsula, a finergrained isotopic analysis suggests that the bi ographies of individuals interred in the mound were referenced in burial treatments. Those individuals who spent their childhood years closest to Harris Creek t ended to be interred as bundle burials after extended processing, while those individuals fr om farther afield were interred rapidly after death. The implication is that the mo rtuary was not construct ed for the purposes of territorial marking, but was in stead an integrative place. This shifting temporality in the timing of interment, as well as the widespread distribution of these mortuaries suggests that commemorative performances at buria l mounds incorporated individuals from throughout the valley, creating collective memo ries across diverse social fields. Harris Creek remains the only well-documented and radiocarbon-dated ex ample of this mortuary practice. Endonino (2003b) and Beas ley (2008) have identified many other likely candidates for early mortuary mounds, including Persimmon, Mulberry, PalmerTaylor, and Orange Mound. These smaller sand mounds eventually gave way to new traditions involving the construction of large mounds, typically of brown sand and shell. Based on Endoninos

PAGE 187

187 work at Thornhill Lake, this transformati on occurred sometime after 5600 cal B.P. Beasely (2008) has argued that in addition to the use of brown sand as a material of internment, later Mount Taylor burials tended to be extended, unlike the previous traditions of flexed or bundled burials at Windover and Harris Creek. Other mounds likely of this era include Tomoka, McDonal ds Mound, and the Bluffton Burial Mound based on stratigraphic similarities. The relic t mound at DeLeon Springs may also date to this era. Unfortunately, little is known of the life hist ories of specific individuals interred within later Mount Taylor mortuaries, as no bioarchaeological studies have been conducted. Further, it is unclear whether the density of inhumations identified at Harris Creek is replicated in later mortuary m ounds. For example, in his excavations at Thornhill Lake, Moore (1894b) recorded only 50 or so buria ls. Because he did not excavate either of the mounds in their entirety, we cannot make a direct comparison with the Harris Creek mortuary. Yet places like the Blu ffton Burial Mound appear to have been constructed over an isol ated individual as well. While we may never be able to fully reconstruct the differences in burial temporality, a consideration of grave inclus ions does provide a sense of the kinds of biographies that were negotiate d through mortuary places. Almost all later sand/shell mounds are linked by the inclus ion of extralocal objects, including bannerstones, stone and marine shell beads, and stone beads. Al though they could occur isolated or in small quantities individuals, there is some evidence to sugges t that objects with distinct stories, ranging from throughout the greater Southeast, were interred as bundles, either by themselves or with individuals. Fo r example, Moore ( 1894b:168) discovered one individual at Thornhill Lake who was interred with a bannerst one on either wrist,

PAGE 188

188 and many stone beads and small marine s hell beads. A similar bundle of a bannerstone and stone beads may have been uneart hed at McDonald Mound, interred with an individual (Jones 1995). At Tomo ka Mounds, Douglass (1882) discovered a cache of five bannerstones, which were si tuated approximately a meter above another cache of three bannerstones. No human rema ins were encountered. Further to the north on the coast at the Coontie Island site a cache of three bannerstones, red jasper bead and other hard stone beads, as well as large Archaic Stemmed hafted bifaces was discovered (Goggin 1952). In this case the objects were found eroding from a beach, and so it is unclear what their asso ciation with death or mounds in general may have been. Taken together, these caches or bundles suggest that the combination of objects provided a means through which diverse biographies of exchange and exchangers could be brought together and framed within the context of communal mortuaries (cf. Pauketat 2008). In the Wake of the Mound Builders The appearance of pottery in shell matrix si tes of the St. Johns River and Atlantic coastal lagoon signals the end of the preceramic traditi ons and the beginning of the Archaic period pottery producing traditions. Orange tradition fiber-tempered pottery has been dated as early as 4800 cal B.P. in the Lower St. Johns, although pottery does not appear in the Middle St. Johns until a century or so later (Sassaman 2003c). By 3600 cal B.P. fiber-tempered pottery ceases to be manufactured at the end of the Orange period, and is wholly replaced by spiculatepasted wares of the St. Johns period. As I have noted, many Orange traditions have their origins amongst Mount Taylor communities. As evidenced by subsis tence data from Blue Spring Midden B (Sassaman 2003a) and Groves Orange Midden (Russo et al. 1992), communities

PAGE 189

189 continued to exploit aquatic habitats (Quitmyer 2001; Russo et al. 1993; Russo et al. 1992), routinely collecting from local shellf ish beds, and capturing fish and turtles. Given the emergent evidence for incr eased exchange and the possibility for sedentism towards the end of the Mount Tayl or period, one may expect these patterns to be exacerbated or amplifi ed in the Orange period. Qu ite to the contrary, the appearance of fiber-tempered pottery in shell matr ix sites ca. 4700 cal B.P. along the St. Johns resulted in a fundamental change in the structure of extra-r egional relationships and local mortuary practice. Unlike the preceding Mount Taylor era, extralocal objects are exceedingly rare. There is an apparent dec rease in the use of marine shell along the St. Johns as well (Sassaman 2003a). Perhaps more striking, however, is the lack of credible evidence for the inhumation of the dead in mounds of sand or shell, at least within the Middle St. Johns River region. Moreover, diagnostic Orange period ceramics are rarely recovered from the surfaces of many shell mounds throughout the region, nor are they found in any great frequency upon sand mounds (Endonino 2008; Piatek 1994; Sears 1960). Despite this fact, evi dence for Orange occupation is widespread throughout the region, with some researchers suggesting this represents an increase in regional population density (Miller 1998). A full examination of the underlying reasons for this transformation awai t another study, although I will return to the issue in the conclusion of Chapter 8. For now, however, it is important to reme mber that there was no inevitable trend towards more centraliz ed or elaborated societ ies along the St. Johns, and that we need to consider more explicitly the situat edness of traditions and biographies in the ongoing pr oduction of place.

PAGE 190

190 A Long Running Record? If we were to examine Mount Taylor as a taxonomic entity, static and situated along the St. Johns, we might be able to make a strong argument for the long-term reproduction of sameness through time. Many of the practi ces and techniques of Mount Taylor communities ha d their origins cent uries or even millennia later. Between 7300 and 4700 cal B.P. we see some striking similari ties as regards subsistence, particularly the importance of both marine and riverine re sources. There are of course some transformations in food procurement that may relate directly to the hydrological processes at work described in Chapter 3. In particular, it is unclear whether the seasonal mobility practices of early Mount Taylor communities persisted until 4700 cal B.P. Lacking detailed seasonality data or zooarchaeological analyse s from multiple contexts we cannot know for certain. Despite the appearance of unreflexive subsis tence-related practices, however, a consideration of Mount Taylor traditions in a multi-scalar and historical framework reveals transformations, discontinuities, and contradictions in the temporality of experiences and interactions. The incipient use of shellfish did not occur during a period of landscape stability, but likely in t he context of landscape instability. Although not necessarily reflected in annual mobility practices or subsistence per se, shellfish consumption and deposition sli ghtly predates the earliest mortuaries, themselves a transformation of an earlier mortuary tradition. By 6000 cal B.P. there is evidence for yet another transition at least as regards increased connectivity to regions farther afield than the adjacent coasts or uplands. These increased distances of exchange may have first extended down to southeast Florida, but eventually would incor porate objects from

PAGE 191

191 throughout the lower southeastern United States Here too there is evidence for a new tradition of mortuary practice that enmeshed the biogr aphies of objects and people. To reiterate a statement I made at the beginning of Chap ter 3, the St. Johns has alternatively been framed as a boundary or a c enter. In this chapter, I have emphasized instead its relational characteristics. Sinc e at least 9000 years ago, the St. Johns was important not necessarily because of the ec onomic resources that it provided, but instead as an intersection of many different individual and community biographies. What we do not know is how the different temporalities of the taskscape and natural landscape were framed from t he perspective of places along the St. Johns. It is to these spatial stories, and the central role t hat placemaking played in the creation and reproduction of social histor ies that I turn to next.

PAGE 192

192 Figure 4-1. Calibrated r adiocarbon assays from Mount Taylor and Orange period components. All Mount Taylor samples recovered from shell matrix. Orange period components include shell matrix and vessel soot. Data for each assay provided in Appendix A.

PAGE 193

193 Figure 4-2. Median calibrated age probabilities within shell matrix sites, ordered by latitude. Also included is one date from Tomoka Mounds (8VO81) on the Atlantic coast.

PAGE 194

194 Figure 4-3. Object and raw material source ar eas in relation to the Mount Taylor region, showing idealized routes of exchange and/or movement. Bannerstone routes adapted from Sassaman (2004b).

PAGE 195

195 CHAPTER 5 DISCERNING MOUNT TAYLOR PLACES Until now I have avoid ed discussing the places created and maintained by preceramic Archaic communities, and for good r eason. Although shell matrix sites have figured prominently in discussions of M ount Taylor lifeways, there have been few attempts to characterize them. T erms such as midden, mound, or heap are generic descriptions that are used to describe shell-b earing deposits. In th is study, I have used the term shell matrix site bec ause it avoids functional implications. Yet it begs the question, just what do we mean by a Mount Taylor shell ma trix site, and are there any differences between Mount Taylor places and those of later inhabitants? The reasons why we do not have readily av ailable answers to these questions are many, but a few stand out. One is epistemological: if all shell matrix sites are truly palimpsests of long-term occupation, then there is little to gain from comparison (e.g., Cumbaa 1976). Another is the long held be lief that hunter-gatherer settlements are unstructured affairs, and that the physical arrangement of space is haphazard. Through time randomized depositional episodes will meld into a complex series of stratigraphic facies (e.g. May 2005; Nodine 1987). The earliest research was also broadly comprehensive. As hubristically argued by C. B. Moore at the end of his own treatise on sand mounds (1894b:246): We are of the opinion that no extended notice of the river mounds can ever again be written, and we sincerely hope t hat others may be induced to take up and to publish reports of the mounds of the east coast, of the west coast, and of the interior, that the archaeol ogy of Florida may be redeemed from the obscurity that has hi therto characterized it. Moores ultimate goal of elevating the visibility of Floridas cultural resources is admirable, and one with which I agree. Howe ver, neither he nor Wyman before him

PAGE 196

196 observed all sites within the r egion, nor did they submit each to the same standards of data recovery. We can point to taxonomic i ssues as well: many shell matrix sites with Archaic components have been ascribed to later periods simply on the basis of surface finds (e.g., Purdy 1987). Pe rhaps the most influential reason for a lack of synthesis is also the most mundane. Most s hell matrix sites of substant ial size were significantly altered or erased and remain faint vestiges of their final configuratio ns. C. B. Moore can be blamed for some of this destruction, as he was fond of noting that the sites he examined were completely dem olished in the process (M oore 1894b:158). However, it was the pervasive and industrialized shell mining that began in the late nineteenth century and extended into the 1960s that des troyed many shell matrix sites. Preservation aside, regularities in pl acemaking traditions should be expected amongst Mount Taylor communities. Amongst hunter-gatherer s in general, the arrangement and reproduction of spaces within places is central to community organization. Settlement structure routinely re ferences cosmological principles of how the world is ordered (Grn 1991; Grn and Kuznetsov 2003; Jordan 2003; Whitelaw 1983, 1991, 1994; Wilson 2005). For example, Gr n (1991) argued that social structure and ranking can be inferred from the distance between households and their arrangment in space using concepts derived from proxemics. In his model, circular or linear settlements typify egalitarian social relationships, while U-shaped settlements can represent more centralized or hiearchical so cieties, with the meeting point of the U being the highest status place. In a cross-cultural analysis of hunter-gat herers, Whitelaw (1991, 1994) identified trends in the arrangment of hunter-gatherer settlements. The physical spacing between

PAGE 197

197 co-resident households references the re latedness between household members, the degree of economic cooperation, and the leng th of residence in any one place. Whitelaw found that random, clustered, and ci rcular arrangments were typical (in his typology) of extended families through multiband aggregates, while linear settlements typified band and multi-band a rrangements. The duration of occupation was an important variable, with extended settlem ent expressed most robustly in linear settlements. Structuring principles regarding the proper location of disposal, the presence of ancestral beings, resources of economic and symbolic significance, and the situatedness of residences with respect to previous settlements are all incorporated into the physical arrangements of places (Grn and Kuznetsov 2003; Jordan 2003; Whitelaw 1994). There is abundant evidence for structuring principles regarding the reproduction of community in place through the deposition of shellfish amongst Late Archaic coastal dwellers who interacted with Mount Taylor communities. Michael Russo (2004, 2008) has used Grns model to examine the organi zation and history of coastal shell rings. Russo has shown that there are two variatio ns on a theme, closed rings and U-shaped rings. Using detailed topographic maps and stratigraphic interpolation he has demonstrated that there are differences in the height and arr angment of depositional episodes. The final forms of the shell rings appear to have resulted from communities either constructing rings en masse or routinely renewing the su rface. He argues that areas with the highest density of shell represent places of hi gher status. Alternatively, the arrangement of connected rings may point to the association of multiple lineages. Whether variations in shell density represent hi earcharchy in a socioeconomic sense, or

PAGE 198

198 community composition divided along ancestral or internal divisions, is another question (e.g., Sassaman and Heckenberger 2004). Knowing how past places were physically arranged and where they were located does not provide an unadulterated viewshed in to the intents and concerns of past communities. Thomas (1996:88) has argued t hat archaeologists cannot infer the ways in which places or regions were inhabited so lely by reference to the physicality of places, their arrangement in geometric space (see also Barrett and Ko 2009; Whitelaw 1994). Thomas was citing the disconnect between the intents and beliefs of those societies who inhabited places and an archaeologists contingent and inculcated perspective. All places have social and hist orical context, and they must be situated with respect to ongoing local and regional inhabitation (Appadurai 1996). This is particularly the case for places inhabited repeatedly. Giv en long stretches of time, the likelihood of social transformations, and di sjointed social memories, the earliest practices in place may have had little in common with those that generated the final disposition of a place (Barrett 1999; Gosden and Lock 1998; Holtorf 1998; Joyce 2004; Meskell 2003). Places of prior practice can also serve as ritualized models for the creation of new social memori es, either as a foundation or physical model, such that we must be circumspect when comp aring architectural or practical commonalities (e.g., Bradley 1998, 2002). Shell matrix sites of the Mi ddle St. Johns register these two intercalated aspects of community histories: historically cont ingent traditions regarding placemaking, and ruptures and disjunctures in frames of reference. If we want to understand how preceramic Archaic communities materialized history as place, then it is necessary to

PAGE 199

199 consider what traditions may have been at work throughout the era of shell matrix site inhabitation. In this chapter I delineate some of the primar y structuring principles that Mount Taylor communities reproduced in inhabiti ng places. Of course it is impossible to not consider changing frames of reference eit her. Some continued to receive shellfish, objects and even inhumations during the subsequent Orange and St. Johns periods. I draw together historic observations, extant geospatial data, and excavation records in order to discriminate the primar y structuring principles of preceramic Archaic places. My goal is not to generate a comprehensive tally of preceramic Archaic sites, but to explore variability within a well-documented dataset and dev elop a working spatial and chronological framework for the analysis of detailed site historie s presented in Chapters 6 and 7. I emphasize two axes of variat ion: site organization and placement. Modeling Shell Matrix Si tes: Data and Methods The only site s included in this discussion were those that (a) fell within the study area defined in Chapter 3, (b) had documented or strongly suspected preceramic Archaic components, and (c) had sufficient documentation to derive metric and qualitative data. Excluding si tes with radiocarbon determinati ons, the primary source of chronological inference is either the lack of pottery within basal strata or objects of known preceramic Archaic age, including cert ain projectile point forms, bannerstones, beads, and other diagnostics. A to tal of 34 sites were select ed for further consideration (Figure 5-1). Archaeological Data Some of the most important sources of data are observations made prior to widespread mining. For sites that no longer exist, these are the sol e source of information for the purposes of reconstructi on. Wyman, for example, paid special

PAGE 200

200 attention to the size, location, shape and composit ion of shell matrix sites. As as result we have detailed discussions on the scale and organization of sites in the region, which frequently includes actual on-ground measurem ents of the shape and depth of deposits. When possible, he would also make notes re garding the depth of pottery or other finds, and the extent of strata. However, he did not c onsistently report all axes of variation for the sites he observed. Although Wyman di d not publish any base maps, his field journals (Wyman 1872) housed at Harvards Countway Library do contain several hand drawn maps (Figure 5-2A). Recent topographic mapping at Live Oak Mound (Sassaman 2003a) and Hontoon Dead Creek Mound (Sassaman 2005a) have verified the relative accuracy of his measurement s. Like Wyman, Moore also provided descriptions of specific localities and the occasional base maps. In this analysis only Moores published articles were consulted, but his unpublished notes do include a wealth of information. The period of salvage excavations in t he middle twentieth century produced a range of variable quality information. T he only detailed topographic maps published include the Bluffton Burial Mound (Sears 1960), which is partially intact, and the Kimball Island Mound (Benson and Green 1962), which re mains undisturbed today. Plan view sketch maps, some of which illustrated distri butions of surface finds, were published for Harris Creek (Bushnell 1960; Jahn and Bullen 1978), Salt Springs Run-2 and Silver Glen Springs Boil (Abshire et al. 1935). These sites have been mined to varying degrees. There have been large-scale reconnaiss ance projects consisting of site visits (e.g., Miller and Griffin 1978; Willis 1995, 2004, 2006; Willis et al. 1977), and full coverage survey (Randall and Sassaman 2005).

PAGE 201

201 Recent projects have produced a w ealth of plan maps and topographic reconstructions. Most have tested shell matrix sites that were already mined, such as Bluffton and Mount Taylor (Wheeler et al. 2000), Old Enterprise (McGee and Wheeler 1994; Piatek 1984), Lake Monroe Outlet Mi dden (ACI and JR 2001), Hontoon Island North (Purdy 1987; Sassaman et al. 2005), and the Silver Glen Run Complex (Chapter 7). Investigations have also taken place at shell matrix sites that have only been minimally compromised, including Blue Spring Midden B and Live Oak Mound (Sassaman 2003a), Groves Orange Midden, Thornhill Lake (Endonino 2008), and Hontoon Dead Creek Mound, Village, and associ ated shell matrix sites (Randall 2007; Sassaman 2005a). Wheeler et al. (2000) also published reconstructed base maps for Old Enterprise and Bluffton (McGee and Wheeler 1994). Geospatial Data A wide array of geospatial data are useful for estimating the length, width, height, and location of shell matrix sites. Ni neteenth century GLO maps and Spanish Land Grant survey data may be used to locate unk nown sites, but have limited utility for reconstructing the specific arrangement of pl aces. USGS topographic quadrangles are also available for the study regi on, but are likewise of restrict ed utility. Many features of interest (i.e. shell mounds) were not visible, or they could not be resolved at the coarse 5-foot-contour interval. Large shell mounds such as Hontoon Dead Creek and Live Oak, which still exist today in mostly unadulterated formsand eclipse 5 m in heightare not visible on 1:24,000 topographic quadrangles. Remotely sensed data provide a useful al ternative. Traditional remote sensing data record visible or invisible light wavelengths. Variation in light intensity and spectra illuminate features on t he ground that can be precis ely mapped given the proper

PAGE 202

202 resolution. Archaeologists have made extensive use of aerial photography to discover and map large architectural features the are clearly visible in open or parkland vegetation (Barnes 2004; Bewley 2003). When obscured by ground cover, variations in vegetation or moisture content can reveal f eatures of interest. Both aerial photography and multispectral satellite data have been used to infer or delineate features in obscured zones based on changes in v egetation compositi on and measures of disturbance (Heckenberger et al. 2003; Heckenberger et al 2008; Neubauer 2001). Because of the horizontal resolution of many datasets (typically between 1 and 10 m pixel size), these are not necessarily us eful for identifying or mapping shell matrix sites. This is not to say this method is inf easible, as shell matrix sites are often hosts to distinctive vegetative communities due to calc ium in the shell (Claassen 1998). This possibility has not yet been ex plored along the St. Johns. Although the precise shape and contours of intact s hell mounds cannot be discern ed from aerial photography, mounds are highly visible when they have been disturbed and appear as very lightcolored non-vegetated zones (Figure 5-2B). Fl orida is privileged by having a series of aerial photographs dating back to 1937, which we re acquired primarily for agricultural purposes (United States Department of Agri culture 2010). Because many were taken during or directly after shell mining oper ations were underway, they record the horizontal extent of shell mini ng activities. In this sense, the more recently acquired (and higher resolution) aerial photographs are of less utility for the present study. A recent and revolutionary development in remote sensing is LiDAR (Li ght D etection a nd R anging), from which high resolution topographic maps can be produced without manual survey. LiDAR datasets consist of elevation values of surfaces that are

PAGE 203

203 recorded in high densities and with high precis ion. As reviewed by Ackermann (1999) this technique uses rapidly fired infrared laser pulses, directed between a stable or moving platform to a surface, to record range values. These range values are then corrected for the distance between the platform and the target area to generate XY and Z values that can be used to create digi tal terrain models (D TM) that represent topographic surfaces. LiDAR points are sampled at close interv als, up to 20 points per m2 depending on the platform and system used. As a result of this very high sampling rate and precision, DTMs can have horizontal errors of less than 1 m, and vertical accuracy within .40 m. These high rates of accuracy and density are typical in areas with low or sparse vegetation, much like image-base photogrammetric methods. Unlike traditional methods, however, LiDAR has the advantage in areas like the St. Johns, where forest cover typically obscures point s of interest. LiDAR pulses can penetrate forest canopies, with up to 70 per cent return rates (Ackermann 1999). Because of variations in the timing and amplitude of laser pulse returns, those points that record surfaces (including low-lying vegetation) can be used to generate bare earth DTMs that represent the ground surface uncluttered by the vegetation canopy. As a result, very detailed topographic maps can be generated of places t hat are far removed from access, or are so large that it w ould be prohibitive to map by hand. There are some caveats with LiDAR data. Because it is a random sample, areas of interest may not have adequate sampling, particularly in wooded areas (Crow et al. 2007; Devereux et al. 2005). Also, because of the standardized rate and geometric footprint of laser scanning, slopes are typically over exaggerated due to under sampling (Aguilar et al. 2005; Harmon et al. 2006). Along the St. Johns, LiDAR coverage is

PAGE 204

204 available for Volusia and Lake Counties, although the sampli ng density of the Volusia coverage is higher. In this study, my use of LiDAR is restricted to the reconstruction of site plans, and the derivation of some basic measurements (Figure 5-2C). However, it also holds potential as a prospecting tool (H umme et al. 2006), in addition to a platform for visualizing and quantifying past landsc apes in 3D (Corns and Shaw 2009). Methods The geospatial data and archaeologica l re cords were integrated through the process of heads up digitizing and georeferencing (Figure 5-2) I first assembled the relevant basemaps and digitally scanned them. Using the locational data provided in the FMSF and publish ed reports, I then generated DTM topographic basemaps using the Volusia County and Lake County LiDAR data. Location data from publication records and the Florida Master Site Files (FMSF) were then checked against the visible surface topography and adjusted to the DTM basemaps. Of the 34 selected sites, 22 of them could be accurately located by visual inspection based on surface expressions. For intact sites, no further map work was necessary. For those sites in which there were basem aps or descriptions available, it was necessary to georeference and then heads-up-digiti ze several sources of information. This process is depicted in Figure 5-2 in refer ence to the Old Town site. In some ways, this is a best-case scenario. The plan m ap in 5-2A is one of the few base maps produced by Wyman (1872). It is remarkable in that the distances along the river and its orientation closely matches the river channel today (Figur e 5-2B). This sketch map is also proof that Wyman was very careful w hen measuring shell matrix sites, such that his recorded observations should be trusted. Figure 5-2B represents the USDA 1941 aerial photograph which shows a large, white vo id that I have outlined. This void is

PAGE 205

205 typical of sites that have been mined or were in the process of being cleared for mining, and thus the white clearing indicates an ar ea that likely had shellfish remains to be targeted for removal. The last image provides the loca l topography based on the LiDAR generated DTM. In it a ridge can be seen that correspo nds to the southern extent of the 1941 shell distribution. This is a diagnostic feature of industrialized mining. For reasons unknown to us now, the shell miners would typically le ave an escarpment or apron of shell which provides a ghostly outline of the original plan. Typically this escarpment faces the water, but in this case it is on the backside, away from the channel. When coupled with Wymans descriptions we can derive more specif ic basic metric attributes. Once sites were incorporated into the geospatial databas e they could be queried to derive other metric attributes. Although not perfect, it do es provide a framework for exploring ranges of variation. Shell Matrix Site Variation Since archaeological investigations were initiated along the St. Johns there have been severa l attempts at creating shell ma trix classification schemes. I am not interested in creating a new typology, or te sting the extent to which past typologies match current data. However, those ty pologies that have been created provide avenue through which preceramic structuring principles can be discerned from those of later traditions. The most appropriate starting point is Wyman, who aside from Moore had the most intimate knowledge of the range of variation amongst shell matrix sites of the St. Johns. Wymans classifi cation scheme can be broadly split into sand/shell mounds, and shell-only sites. He located sand mounds throughout the valley, sometimes associated with shell matrix sites, although in many cases they were not. He also

PAGE 206

206 identified two primary shell matrix site types, including mounds and fields, although he used the term heap to describe them incl usively. He found them generally on the rivers edge, although many could be found a hundred or more meters within wetlands. He described shell mo unds (1875:10) as: in almost every case built on the banks of the river, resting either on one of the ridges of sand and river mud already mentioned, or on land slightly raisedThe most of the mounds are in the form of long ridges parallel to the shore, though a few, as Bartram s and Brysons mounds, are, or were before the current encroached upon them, nearly circul ar. The limits of all are sharply defined, and at a few feet from the base shells cease to be found. Wyman identified a few select locations that contained multiple ridges and associated sand mounds and shell fields, including Bluffton (8VO22), Hontoon Island North (8VO202), Bluffton (8VO55), and Silver Glen Run East (8LA1E). In contrast to the elevated shell deposits at mounds, he also discriminated shell fields. These he found to be distinct by their lack of elevation or relief (1875:11): The shell fields differ from the m ounds in having their materials more evenly distributed and in a comparativel y thin layer, varying from a few inches to two or three feet in thi ckness. The distinction is, however, arbitrary. By an arbitrary distinction, Wyman thought th at shell fields were simply less well developed mounds. Throughout his monograph Wyman describes shell fields as alternately isolated from mounds, such as th e Snake Creek Village site, or associated with mounds, such as at Blu ffton. Excluding the occasi onal mention by Moore, and some recent mitigative work in the past decade, shell fields have largely gone unexamined. In their review of the Mount Taylor period, Wheeler et al. (2000: 145) expanded upon Wymans typology arguing that similar featur es occur regularly at different sites, suggesting some intentional construction and shared ideas on village

PAGE 207

207 layout, refuse disposal, and mound constructi on. They suggested a typology of ovoid mounds, linear ridges, shell fiel ds, and multicomponent mounds. A consideration of the basic typology of Ridge, Multi-mound complex, and Shell Field espoused by Wyman and Wheeler et al. indicates that there are some distinctions between places based on overall size (Figure 5-3, Table 5-1). Ridges in this classification are any mounded place with o ne primary surface feature, Multi-Mounds include all sites with at least two distin ct mounded features, i ndependent of culturehistorical association, and fields are thos e sites with limited surface expression. A comparison of the recorded height compar ed with an approximati on of their surface area indicates some separation between the different classes. Multi-mounds are consistently the largest in terms of height and they also tend to be situated closest to water. Shell fields were the most discret e of the group in terms of elevation although they vary considerably in terms of their di stance to water and total footprint. Starting with this framework, we can examine so me of the within-class variation. Ridges There were 17 ridges for which basic measurements could be obtained, and they are distributed widely throughout the region (Tabl e 5-1). Because this analysis was not oriented towards examining the finite distri bution of places throughout the valley, the absolute distribution and frequency of shell ri dges cannot be assessed. I would note that the stretch of river between Hontoon Island and below Blue Spring contains a minimum of five similarly configured ridges which equates with roughly a ridge per 2 km. However, Bowyers Bluff-1 and -2 ar e situated roughly 450 m apart, and the ridges at Silver Glen Run are only 400 m distant. There was a significant range between where ridges would be located in relati onship to one another. Without radiocarbon

PAGE 208

208 dates from both Bowyer Bluff sites it is unknow n if they were occupied at the same time, although both Silver Glen Run mounds were apparently coeval. At a smaller scale, most ridges appear isolated without any other s hell matrix sites attached to or situated adjacent to them. In the sample considered here, only 3 have known shell fields in association: Blue Spring Oxbow, the M ound at Blue Spring, and the Hontoon Dead Creek Mound. This may be a factor of visibi lity, however. A reconnaissance survey of Hontoon Dead Creek Mound isolated two shell fi elds in association withbut spatially segregated from the mound that would not necessarily be noticed during pedestrian survey (Chapter 6). Another pot ential source of variation that is not directly related to their original shape is that some places, such as Horse Landing or Blue Spring Oxbow, had been eroded prior to Wymans observati ons. As such, these measurements represent minima. Shell ridges appear to date primar ily to the preceramic Archaic. This is suggested by two chronological indicators. Radioc arbon dates at the Hontoon Dead Creek Mound, Silver Glen Run-Locus A, Bowyers Bluff-2, and Live Oak Mound all fall within the range of the Mount Taylor period (A ppendix A). However, these det erminations do not provide a holistic indicator. Another measure is whether or not pottery has been observed in strata. Of the 17 shell ridges in this samp le, 15 are recorded as having pottery only near the surface, within a maximum of 2 f eet; and one (Crows Bluff) has had no pottery recorded at all. The ridge at the Thursby Complex has had pottery found in deep levels. Given that there was a St. Johns period sand mound on the summit, it is not surprising that St. Johns pottery was noted below the surf ace, particularly along the water front.

PAGE 209

209 Ridges show considerable diversity in scale. They ranged between 61 and 300 m in length, and between 15 and 80 m wide, whil e their median tendency is 140 m long and 55 m wide. Height, measured as the difference between the ground surface and the top of the mound, varied fr om a low of 1.8 m at the T hursby Complex to a high of 8.3 m at Mount Taylor. Mi nimum volume was derived for the five sites which were intact, and was calculated between the ma ximum surface and the minimum contour interval that bounded a ridge. Volume co-var ies with the height and area of of ridges, and ranges between only 376 m3 at Blue Spring Oxbow to 7602 m3 at Hontoon Dead Creek. Despite variations in scale, there are sim ilarities in how ridges are configured. All ridges are longer than they are wide, with a m edian length:width ratio of 2.7. This should not be surprising as this configuration is the basis for the typological separation. However, some other basic prin ciples regarding the reproduction of localities is hinted at by a comparison of ridge height and cross sect ion. In Figure 5-4 I have plotted the current cross-sections of Live Oak Mound, Hontoon Dead Creek Mound, the basal remnant of Mount Taylor, and a projection of Mount Taylors pre-mining cross section based on Moores (1893a:13) description. In a ll cases, similar angles of repose are evident, with steep slopes present on the upland and swamp facing sides. The upland terrace-facing side tends to have a slight ly lower slope, and this may be due to the higher elevation on that side of the ri dges. If ridges accumulated from random deposition, then one would expect low, att enuated slopes. Yet Mount Taylor and Live Oak share roughly similar basal widths, but M ount Taylor is 3 m higher. This pattern suggests that an emphasis was placed on maintaining the summit throughout the

PAGE 210

210 process of deposition. Although the footprint may have been expanded to allow summits to attain greater elevation (Chapter 7), in general an original basal configuration was kept intact. This accord s well with Wymans comment that shell was rarely found away from the base of she ll ridges, and not randomly scattered about. Another intriguing point of variation is f ound between the relationship of distance to water and the height of a ridge (Figure 5-5). In the dataset I could assemble there are two clusters of ridges, those that are close to water today and those that are at least 150 m away from an active channel. It s hould be noted that almost all shell ridges occur at roughly the same elevation today (T able 5-1). LiDAR-deriv ed elevation values show a central tendency towards 0.0 m NAVD88. It would have been ideal to know the minimum elevation of basal components at any given locality. Because many of these places were not subjected to testing the depth of their deposits below ground surface is unknown. There is some deviation from th is pattern, as the Silver Glen Run-Locus A ridge is situated at roughly 1.61 m above NAVD 88. If the location of different ridges was simply a matter of increased water levels through time, we would expect the earliest places to be surrounded by swamp, away from upland terraces. Yet those ridges that are now situated the farthest distance from a channel are also emplaced on terrace edges on the east and west sides of the basin. The implication is that surface waters, or flood waters at a minimum, were actually higher during early ridge formation as suggested in Chapter 3. Those ridges si tuated away from the main channel today are statistically larger, t (16) = -3.70, p < .001, (see also Kimbal l Island in the following section). Assuming that those sites situat ed away from the main channel are also the earliest, this would suggest that the earliest mounds are also the largest. This certainly

PAGE 211

211 holds for Hontoon Dead Creek Mound and Live Oak Mound, both inhabited 7000 years ago. More patterning is revealed at the level of individual ridges, in particular the plan and final surface expression. Detailed topography could be developed for the five intact ridges (Figure 5-6, Figure5-7AB ), and footprints could be reconstructed from LiDAR for another four mined ridges (Figure 5-8). Hontoon Dead Creek Mound is characterized by an arcuate or crescentic ridge, orient ed parallel to the topogr aphy of the eastern terrace edge, which increases in width from nor th to south (Figure 5-6A). It has abrupt slopes, particularly on the swamp-facing wester n edge. The summit is not flat, but has an asymmetrical apex at the southern end. It has an almost an inverse twin in the nearby Live Oak Mound (Figure 5-6B), which is characterized by an arcuate ridge that corresponds to the western terrace edge. Unlike Hontoon Dead Creek Mound, however, the asymmetrical apex is situated on the northern aspect of the mound. The Blue Spring Oxbow Mound (Figure 5-6C) is a much smaller linear ridge. The mound is too small to resolve surface features, but a visit there confirmed that the apex is effectively flat (Randall and Sassaman 2005). Blue Spring Oxbow is the smallest ridge recorded. Although the possibility exists that this was it s original form, Wyman suggests that it has in fact been eroded by the river. In plan form it is situated adjacent to a lagoon that may correspond with an old channel meander. It is likely then that this is but a faint vestige of its original configuration. The other two intact shell ridges are sit uated farther to the nor th. In comparison with Live Oak and Hontoon Dead Creek both Bowy ers Bluff-1 and -2 are much smaller, but show similar principles (Figure 5-7A,B). Their bases are relatively ovoid, but this

PAGE 212

212 footprint is surmounted by li near or arcuate ridges. Both Bowyers Bluff ridges have slightly asymmetrical summits, in this case pointing towards the north. Taking this small sample at face value, there does not appear to be a necessary cardinal orientation for the higher apex. Moreover, the chronological association of the apexes are not clear. For example, Sassaman (2003a) identified what may be a St. Johns period mortuary mound at the apex of Live Oak Mound, which suggests that later St. Johns communities appropriated the preceramic Archaic ridges. Although structurally similar, a number of visits to Hontoon Dead Creek Mound has failed to identify any significant surficial St. Johns assemblages or human remains, at leas t in the quantities identified at Live Oak Mound. Wyman did record a the articulated remains of an individual somewhere on the mound summit. For his part, Moore (1893a:13) indicates that the summit of Mount Taylor, which was 8.3 m above the floodplai n, was effectively flat. Thus, the significance and pervasiveness of ridge asymmetry is unknown. Multi-Mound Complexes The multi-mound complex category is compos ed of seven sites, all of which likely eclipse the scale of shell ridges prior to s hell mining (Figure 5-8, Table 5-3). Of the sample, only two remain intact today (T hornhill Lake and Kimball Island), likely because of their locations away from the main c hannel. Four also ha ve sand and shell burial mounds attached to them, or situated near, and I have include what metric information is available for them in Table 5-3 under thei r corresponding site. Shell fields are likely associated with many of them. Because of the extensive nature of mining there has been little known about the the extent of shared layouts, and whether they were produced by Mount Taylor or later communities As I will review here, there are several interrelated patterns of placemaking woven am ongst them. That is, this type does not

PAGE 213

213 register a single process, nor does it r epresent a gradual increase in the scale of depositional events through time. Excluding Kimball Island, all are located on the waters edge, with deposits exis ting at or below 0.0 m NAVD88 and are extensive in terms of surface area and height. The Kimball Island site is one of the fe w intact mounds left in the river valley (Figure 5-7C), and is also one of the least we ll known. The site is located 130 m away from a primary channel today although there are ponds nearby, and is completely surrounded by swamp. As seen in the discussion of ridges earlier, it is more typical for sites away from main channels to be locat ed directly on the terrace edge. The complex has an intriguing plan, and consists of two shell loci: a 5-m high arcuate ridge (roughly the same size as other isolated Mount Ta ylor ridges) and a 3.5 m high ovoid shell dome to the northwest connected by an apparent ramp. The main ridge has at least two high points on its summit, and western dome has an asymmetrical summit mount as well. The site was first documented by Benson and Green (1962), who produced a topographic map and conducted limit ed testing. They recovered only three St. Johns sherds in the upper 6 inches of the mound, and continued their test pits down a total of 5 feet. It would seem then that this is likel y a Mount Taylor production, but little else can be said of it at this time. The other intact mound complex is Thornhi ll Lake, originally documented by Moore (1894b) and recently tested extensively by Endonino (2008) (Figure 5-9). Today the site consists of a 3-m high arcuate ridge, a western ridge or ramp extension, and two lakeside depressions. There are also two truncated-cone sand and shell burial mounds that surmount the center (Mound B) and eas tern (Mound A) end of the basal ridge.

PAGE 214

214 Although largely intact, I should say that shell mining may have produced the depressions on the lake side and Moore dug extensively into Mounds A and B (Endonino 2008). As noted by E ndonino, Moore reported that the height of Mound A prior to excavation was 3.4 m, while Mound B was 2.7 m high, today (and in Figure 5-9) they both have the same relative height of appr oximately 1.8 m. This site has provided pivotal information the demonstrates Mount Taylor communities constructed sand and shell mounds, and that individu als and extralocal objects were interred in them (Chapter 4). This complex is also primarily preceram ic Archaic, based on the fact that ceramic sherds of the Orange and St. Johns period were limited to near-surface exposures. As reported by Endonino, the basal shell ridge dates at least as early as 6200 cal B.P. Basal dates on the mounds place construction of Mound B as early as 5860 cal B.P., while Mound A was constructed no later than 4600 cal B.P. Because these assays were acquired from the sub-mortuary surf ace, it is unknown if the mounds were contemporaneous, or constructed serially. Regar dless, the data here indicate that the preexisting shell ridge was used as a platform for the construction of paired and structurally similar mortuary mounds (between 27 and 35 m in diameter). Taking the observations we have of intact and fragmented shell ridges, we can assess the relationship of Mount Taylor comm unities to the four largest, and mostly destroyed, multi-mound complexes. Each of these places has preceramic Archaic components that have been reconfigured by la ter communities. We can begin with one of the better known examples, Hontoon Island North (Figure 5-10A). As seen today, the mound complex is characterized by an arcuate ridge, measuring some 300 m in length. The ridge is highlighted by a corresponding interior escarpment and linear depression

PAGE 215

215 that terminates in a large depression to the west. This escarpm ent maps the southern ridge of what was once a complex composed of two ridges and two burial mounds offset on the backside. Wyman (1875:27) describes the site as consisting of a large southern ridge with a 7.6-m high asymmetrical summit to the west, and a smaller ridge that rose approximately 4 m high. Thes e were separated by a large valley in the middle, but may have been conn ected on the eastern end. I have superimposed the likely location of these ridges on Figure 510A. Today, the northern ridge has been mined away and is now a harbor for Hontoon Island State Park. Wyman noted that pottery was to be found at all depths in t he southern ridge exposed profile, indicating that it was likely an Orange or St. Johns construction. That this was a place of particular import ance to St. Johns communities is attested to by the recovery of wooden effigies from the channel (Bullen 195 5a), as well as a St. Johns period sand mound on the opposing bank (8 VO2600). In addition, research by Barbara Purdy (1987) on the wetsite deposits on the eastern margin also confirmed that the apex was likely constructed by St. J ohns communities. However, reconnaissance and test excavations in the southern ri dge escarpment by Sassaman and colleagues (2005) demonstrated that much of that ridge is preceramic Archaic, dating as early as 5200 cal B.P. and likely earlier (see Chapter 7). Finally, Wyman also described two paired and conical mounds on the backside of the souther n ridge, the larger was estimated at 7.6 m, while the smaller at 6 m high. The placement on Figure 5-10A is based on McGees reconstruction. There ar e no discernable remnants of either mound today. Wyman reported finding pottery and hum an remains to a depth of 1 m in one of the mounds. Whether these ac tually date to the preceramic Archaic, however, is

PAGE 216

216 unknown. They are consistent with later St. Johns period mortuary mounds elsewhere. Yet based on their paired status (see Bluffton below), and their comparison at least in their basal widths of Thornhill Lake, we s hould consider a Mount Taylor origin as a possibility. Regardless, as reconstruc ted today (red line on the map), the mound complex approximated a U-shape, which apparently post-dates the Mount Taylor inhabitation there. The Harris Creek site is best known for t he early Mount Taylor mortuary that was exposed during shell mining and initially do cumented by Jahn and Bullen. Like Hontoon Island North, the center of the comple x was mined out, leaving a ca. 150 m long and 90 m wide depression, in addition to miscellaneous shell extensi ons (Figure 5-10B). This ridge is consistent with the Mount Taylor shell ridges de scribed above. It was at the base of this shell ridge that Jahn and Bullen discovered the Mount Taylor mortuary. The specific configuration of the site prior to destruction wa s never properly recorded. Moore (1893b:606) describes the complex solely as acres of shell ridges. He excavated a 3 m deep test pit in one ridge, and recovered abundant Orange fibertempered Incised sherds. In all subsequent tests he noted Orange pottery, in addition to St. Johns pottery, across the complexs su rface. Bushnell (1960) noted that Orange sherds were largely absent from on top of the main ridge, but present on the western and southern shell extension. More recently Aten (1999) reconstructed t he history of excavations. He provided an illustration showing the likely configuratio n of the ridges prior to mining. This reconstruction was based on a 1961 ske tch map and 1941 aerial, which I have superimposed on the LiDAR DTM in Figure 3-10B. Although not entirely clear, the

PAGE 217

217 imagery and topography suggests t hat there were two ridges that joined at the Mount Taylor shell ridge. These extensions appear to have formed a U-shaped shell ring, with the Mount Taylor ridge forming much of the northern half of the ring. Abundant Orange sherds were noted everywhere except for on t op of the Mount Taylor ridge. On these grounds it would appear that the multiple ridge configur ation post-dates the Mount Taylor ridge. A similar sequence, consisti ng of a Mount Taylor mortuary beneath strata containing abundant Orange Incised sherds, was identified by Moore at the Orange Mound. In this case there were apparently no shell ridge extensions approximating a Ushape. Just 7 km to the northeast lies Bluffton (F igure 3-11). Much of the site was mined below grade, and so all that is left is an el ongated and low-lying ridge of shell. Wheeler et al. (2000) have reconstructed the topogr aphy of plan of the site, which I have superimposed over the DTM. The site was described by Moore (1894a) and Wyman (1875:37) as consisting of a great shell dome, 7.6 m high, and an associated shell ridge. It is unclear if the dome and ridge were attached at one time. No work was conducted on the shell ridge fronting the wa ter, but Bullen (1955b) performed salvage excavations in the dome to t he north as it was being mined. In a 5-m deep, 7.6-m long trench he recorded a sequence spanning a basal Mount Taylor component (over 4 m thick) and an Orange component. To the sout hern and landward side of the southern ridge Moore (1894a:48) identif ied abundant sherds of unknow n type in a large shell field. In addition, Moore (1894a:46), Wyman (1875:37), and Sears (1960) identified a conical sand and shell mound, and an elongated sand and shell mound on the

PAGE 218

218 eastern terrace edge of the complex. Excavations in the southern mounds demonstrated that it was constructed rapidly of shell, sand, and muck over an individual (Chapter 7). Although there are no radi ocarbon assays published for this mortuary mound, it is consistent with those identified at Thornhill Lake and Tomoka in terms of structure, content, and size. Thus it is more than likely Mount Taylor in age. As to the sand mound to the north, little is known about its content, but the fact that there are two mortuary mounds, much like Thornhill, suggests that they represent paired mortuaries. The extent to which Bluffton was configured like a U is unknown, but it would appear that an extensive Mount Taylor ridge under lies much of the river side complex. The northernmost multi-mound complex is situated at the confluence of Silver Glen Springs Run (Figure 3-12), and is known as 8LA1-East. It is part of a much larger assemblage of ridges and shell fields that are situated to the we st along the spring run (Chapter 7). The Mount Taylor shell ridge (Locus A) is visible to the west in Figure 312, and is separated from 8LA1-East by a large seep spring. Today, 8LA1-East has almost no surface expression. However, Wyman described it as one of the largest mound complexes on the St. Johns. This site was completely mined in 1923. Probate documents show that a contra ct of $17,000 was made for the removal of the shell, with a fixed price of $0.15 per yard. Assuming this was a cubic yard, it would translate into 113,333 yards3 of shell, or 86,649 m3. This estimate may also have included the Locus A shell ridge that was also mined. Regardles s, it was a massive quantity of material that was removed. It was upon this mound that Wyman contemplated the significance of origins and significance of shell mounds as discussed in Chapter 1. Wyman (1875:38) described a large U-shaped complex at the m outh of the Spring Run,

PAGE 219

219 roughly 300 m long on a side, and upwards of 10 m high. Although he did not publish a map of the site, he did sketch one in his jour nal, which together with recent excavations and surface surveys conducted by Sassaman and colleagues (2010) served as the basis for the boundaries reconstructed in Figure 3-12. Although there is little left on the site today, excavations on the southern ridge identified basal sh ell deposits dating to the late Orange period. Beneath the northwest corner of the m ound, and fronting Lake George, many decorated and plain Or ange sherds have been recovered. There can be no doubt that much of th is place was constructed during Orange times in a shape approximating a U-shaped she ll ring. However, there is anecdotal evidence that suggests the fr ont ridge, particularly the por tion near the seep spring to the west, may have been inhabited during the Mount Taylor period, and a portion of the site may have been a mortuary. The basis fo r this inference is the combination of skeletal remains found eroding from the surface, and concreted shell. Concreted shell is shell-bearing matrix that has been cemented together. Although the precise formation process is unknown, Palmer and W illiams (1977, cited in Wheeler et al. [2000]) have argued that concreted shell forms as calcium carbonate in superimposed strata percolates downward and eventually precipitates. M any of the Mount Taylor ridges have basal strata composed of concreted shell, although it does occur in nonbasal contexts as well (Wheeler et al. 2000: 145). I will return to the significance of concretion in the subsequent chapters. Regardl ess, there may be a Mount Taylor ridge here as well. The final multi-mound complex under consideration is the Old Enterprise complex. Today, virtually nothing is left of the terrest rial component (Figure 3-13A). Wheeler et

PAGE 220

220 al. (2000) and Piatek (1984) have attempted to reconstruct its size and shape based on historic observations. Towards the end of the nineteenth century, Old Enterprise was very much a tourist destinat ion, there was a hotel on its summit, and perhaps one of the most famous shell mounds along the St. Johns (Figure 3-1A). When Wyman visited the site he noted that it was mostly eroded on the lake side, but was characterized by a high central dome with two ridges, one extending perpendicular to the shore, and another irregular one along Lake Monroe. Surrounding t he complex were shell fields. Virtually nothing is known of the complexs culture-historical association. However, Wyman (1875:20) does indicate that in his ex amination of the er oded mound profile: Pottery was found in considerable quantities, mostly ornamented with tracings in straight lines; but few of the pieces were stamped. Palmetto fibre [sic] was largely used, mixed with the cl ay, in the making of the vessels. This description of recovered pottery is consonant with the Orange Incised variety. While it is unclear how the complexs basal components were related to Mount Taylor communities, it would appear that as wit h Harris Creek, Hontoon Island North, and Silver Glen Run-East, a multi-ridge comp lex, possibly a U-shaped shell ring, was constructed in this locality. The Old Enterprise complex is the terrestrial component of the Groves Orange Midden wet site that spans the entirety of the Mount Taylor period. While not included in the traditional classification scheme, wet sites are prominentalbeit typically obscuredaspects of regional shell mounds (Wheeler et al. 2000). A minimal definition of wet site deposits are those in which ant hropogenic matrices are situated below the current water table. Largely owing to its pervasive wetlands and low relief, the state of Florida is home to some of the most abundant and chronolog ically diverse wet site deposits in the world (Purdy 1991). Along the Middle St. Johns River, wetsite deposits

PAGE 221

221 have been identified that span t he preceramic Archaic through St. Johns II periods. As discussed in Chapter 4, excavation t hese permanently inundated resources have provided crucial details on rarely-preserved botanical and other perishable assemblages. At Groves Orange Midden the researchers also used a piston core rig to map out the distribution and thickness of shell matrix that extended ou t into lake Monroe (Figure 3-13B). They f ound that anthropogenic deposits could be located up to 30 m away beneath the lake, which was likely a func tion of the bathymetry. However, shell matrix and other materials were actually thickest within a linear depression running perpendicular to the shoreline (h ighlighted in Figure 3-13B at the -0.90 m isopach). The excavators argue that this was a paleochann el that was infilled by shell and other materials. Although it is unknown if Mount Taylor communities filled this channel in during periods of low water, the repeated deposition of ma terials within a wetland may have harkened back to the pond burials of old. At the very least, Mount Taylor communities apparently inscribed landscapes in ways that would be obscured or forgotten as much as they materialized social histories in shell matrix places. Shell Fields Ridges and multi-mound complexes were clearly places of intensive deposition. But there were many other places through wh ich spatial stories we re written. Shell fields remain the least understood of all she ll matrix sites along the St. Johns, but they are arguably one of the most pervasive manifest ations of inhabitation. They show an incredible range of variation, save for the lack of dense or thickly accumulated shell (Table 5-1). They are found at all elevations, from the floodplain up to 13 m high on terraces. They can be isolat ed, or adjacent to ridges or multi-mound complexes. Some appear to have been residential loci. For example, Sassaman (2003a) isolated an

PAGE 222

222 Orange period semi-circular compound at Blue Springs Midden B. The deposits there actually spanned the Mount Taylor/Orange interf ace. The compound was identified on the basis of a discrete distri bution of crushed shell patches. Others appear to register discrete loci of inhabitation. At the Lake Monroe Outl et Midden, the excavators identified discrete depos itional episodes on the upper terrace as well, some of which only involved the production of lithic t ools (ACI and JR 2001; Scudder 2001). The absolute distribution of shell fields is also unknown. However, a recent reconnaissance survey of the margins of Hont oon Island suggests that shell fields were literally everywhere (Figure 5-14). Over the course of five seasons, the University of Florida St. Johns Archaeological Field School surveyed the interior and margins of the island (Endonino 2003a; Randall and Hallman 2005; Randall and Wallis 2007). Methods were designed to identify, characte rize, and delimit site boundaries. A total of 10 shell matrix sites were identified, eight of which contained, or were suspected to contain, preceramic Archaic components. T he most visible features on the island are the Hontoon Dead Creek Mound to the south and the Hontoon Island North complex to the north. However, thes e are situated amongst a diverse non-mound-scape composed of shell matrix sites found at low elevations and at the intersec tion with wetlands. The western and eastern flanks of the island are characterized by two elongated sites (Indian Mound Trail and Hontoon Hammo ck). They share some internal similarities, including discrete shell-bearing deposits, some of which yielded concreted shell. In contrast to the shell, the majori ty of each site is composed of more diffuse anthropogenic deposits which cont ained only traces of vertebrate faunal remains. Neither site yielded substantial pottery inv entories. Although quite similar, there are

PAGE 223

223 important differences. Tests at Indian Mound Trail recovered a relatively large lithic assemblage, composed of stemmed hafted bifaces, modified flakes, and lithic waste flakes. At least two bone pins and other mo dified bone artifacts were recovered. The limited presence of Orange and St. Johns sher ds suggests that these sites were occupied only intermittently during these later periods. On the balance their use occurred during the prec eramic Archaic. In stark contrast to the eastern and we stern margins, the three bounded sites to the south (Hontoon Dead Creek Village, Sn ake Creek Village, and South Hontoon Midden), reflect regularized plans. Each is characterized by one or more centralized nodes of elevated shell. At least two of the sites, Hontoon Dead Creek Village and Snake Creek Village, have varied surfac e topography suggestive of discrete, and potentially routinized depositio nal practices (Chapter 6). Hontoon Dead Creek Village in particular has been demonstrated to span the entire culture-historical sequence starting with Mount Taylor. Evidence for all cerami c periods was present among these three sites, with St. Johns Plain and Orange Plain sher ds dominating. This pattern is actually quite surprising. The prev alence of Orange habitation c ontrasts with Hontoon Island North and Hontoon Dead Creek Mound where very limited evidence for Orange components has been documented. While surp rising at Hontoon Island, these sites actually share some similarities with Blue Sp ring Midden B at Blue Sp ring State Park. At a minimum the patterns identifi ed at Hontoon Island suggest that any consideration of regional inhabitation must take into account the collective assemblage of spatial practices. The smallest, or l east obvious, archaeological sites today were likely the means through which the broader patterns of landscape inhabitation were

PAGE 224

224 articulated as spatial storie s. We should also be suspic ious of viewing non-mounded places as fundamentally opposed to those occurring at s hell ridges or multi-mound complexes. That the social contexts may have been different does not mean that practices, and their significances, could not be transposed and transformed across fields of practice. The Structure of Regional Inhabitation The reconstruction of Mount Taylor places enables us to consider some of the larger principles at work t hat structured the distribution and disposition of inhabitation throughout the Middle St. Johns. This analys is was not based on a complete population sample. However, it should be obvious by now that the Middle St. Johns was rife with the accumulated materiality of preceramic Archaic practice. In a consideration of placemaking, several patterns have emerged (Figur e 5-8). Most notab ly, Mount Taylor communities routinely created linear places. The significance of the linearity is not evident from a regional sample, but it is fu ndamentally distinct fr om the later U-shaped multi-mound complexes. The limited data at t he regional scale suggests that once ridge ground plan were established they were r eproduced through depositio nal practices that emphasized the ridge summit. This is not to say that bases did not expand, but they appear to have done so only to allow higher summits. Contradicting the presumption that these places were unstructured, the GISbased analysis has shown that in fact there was a basic plan and footprint that was reproduced many times over. In terms of situatedness, most Mount Taylor ridges were located nearby sources of water, includi ng channels, lagoons, and springs as well. Given that ridges are compos ed of shellfish derived from the water, this should not come as a surprise. However, I would note that no one has yet determined whether

PAGE 225

225 shellfish deposited on ridges were derived from local contexts. Other patterns regarding the placement and relationality of later-Mount Taylor mortuary mounds are also evident. In at least two cases paired sand and shell m ounds were emplaced either on top of or adjacent to preexisting shell ridge s. Whether or not these were constructed in a serial fashion, their duality is contradi ctory to the isolated linear places they reference. These data suggest the importanc e of ridges, and likely shell fields as well, lay not only in their economic potential, but in the signific ance woven through practices in place. One might be lulled into thinking that linear ridges all emerged in the same social and ecological contexts, much like Cum baa, and that each place was repeatedly inhabited over the course of millennia. Alternatively, one might suppose that shell mounds were constructed for t he purposes of elevating a resi dential space out of rising water. Both of these assertions can be test ed with a combination of the LiDAR elevation data and radiocarbon assays. During the pr ocess of georeferencing site plans, I calculated absolute elevations relative to NAVD88 for a series of assays distributed throughout the river basin (Figure 5-15). Thes e assays are derived from terrestrial and subaqueous contexts. In Figure 5-15A I have graphed the median intercept and absolute elevation of each assay. Also shown on the graph is the maximum span of peat deposition at Groves Orange Midden, argued to represent high water tables and a stable hydroperiod. I also gr aphed the elevation range of the minimum flows and levels (MFLs) for the Deland gaging stat ion (Table 3-1). MFLs are ty pical flow conditions of the river, from the frequent low to the minimum frequent high. They represent the river stages that inhabitants could be expected to experience along the contemporary St. Johns within a 3 year time span. In order to examine potential differences in stage

PAGE 226

226 values throughout the basin, in Figure 5-15B I normalized the elevation values by the maximum recorded flood stage at the nearest gaging station. For sites near Deland this equates with the 50-year flood. Elsewhere, the local gages had less than 50 years of data, and thus the highest magnitude flood ev ent does not necessarily have a similar return probability. Regardless, these values help show the relationship of places to infrequent but high m agnitude events. A consideration of the elevations of occupation through time reveals several important patterns. From the inception of shellfishing 7300 years ago through the onset of the Orange period 4600 year s ago, no terrestrially deposited contexts are below current MFLs. As discussed in Chapter 3, th e current MFLs are likel y at or near their highest stages in the history of the St. Johns Shell matrix site minimum elevations have thus remained effectivel y unchanged over the course of the Mount Taylor period. The location of terrestrial deposition is in contrast to pattern of the subaqueous contexts, which show an overall increase in elevation through time. I would note that except for one assay, all s ubaqueous dates are from the Groves Orange Midden site. Thus, their higher elevation through time refl ects a process of deposition, and does not necessarily reflect increased water levels through time. Regardless, the lowest subaqueous deposits are also the earliest. There is no clear trend in an increase in elevation until ca. 6000 cal B. P. when assays are cluster ed between roughly -0.5 and 1.5 m NAVD88. This cluster of assays is coeval with the depositio n of peat at Groves Orange Midden, and appears to reflect a period of aggradation and higher surface waters.

PAGE 227

227 Despite the potential increase in stage ranges through time, there is not a concomitant increase in minimum terrestrial el evations. This point is driven home when one considers the flood-normaliz ed distributions. Some terre strial contexts would be submerged during infrequent flood events. That these differences in location (by elevation) do not simply reflect rising water levels through time, however, is evidenced by terrestrial deposits at the chronological ly young Hontoon Island North site (8VO202) at relatively low elevations. The terrestri al pattern suggests that throughout the Mount Taylor period places were always mainta ined above todays typical flood stages, and thus the deposition of shellfis h and other materials on top of ridges cannot be explained by the need to elevate places out of rising water. As to the long-term rhythm of inhabitation, the clusters of assays for specific localities imply differing intensities of local inhabitation, contra Cum baa. Most places, if not all, appear to have emerged rapidly, and were subsequently decommissioned or abandoned. This is certainly the case with Hontoon Dead Creek Mound, where upwards of three meters of deposition occurred over the course of several centuries. The similarity in overall site plans though time cannot be reduced to a long-standing pattern of unreflexive site occupation. As demonstrated in Chapter 3, Mount Taylor communities experienced widely differing ecologi es and hydrological conditions over the course of the Mount Taylor period. Thus, even though Mount Taylor places superficially share similar organizational plans, the experiences of the inhabiting communities would have been framed by different tem poralities of the landscape. For example, the earliest shell ridges were produced at a time of lo wer water, but more frequent flood events. The distribution of shellfish may have also varied considerably from year to year.

PAGE 228

228 In contrast, by 6000 cal B.P. many places that were once directly associated with channels, such as the Hontoon Dead Cr eek Mound, were orphaned by floodplain aggradation and infilling. New places were apparently established elsewhere, but with respect to fundamentally different hy droperiods and concom itant ecological distributions. In this c ontext of change, the continued production of localities with shared structuring principles over the long term emerges as a central question to be addressed. It is the histories of inhabitati on at the local level that will enable us to untangle the different ways in which the reproduction of places through time enabled social memories and spatial stories as a historical process. In Chapters 6 and 7 I will examine the specifics of place-making, and trace out the genealogies of these practices.

PAGE 229

229 Table 5-1. Metric attributes of shell ma trix sites based on typologies of Wyman (1875) and Wheeler et al. (2000) Type Mean Median Std. Dev. Min Max Ridge (n=15)1 Length (m) 144.4 140.0 71.6 45.7 300.0 Width (m) 52.1 54.9 19.5 15.0 80.0 Height (m) 4.3 4.0 1.5 2.1 8.3 Distance to Water (m) 87.5 29.4 102.1 0.0 288.0 Multi-Mound Complex (n=7) Length (m) 316.8 330.0 154.4 134.1 589.0 Width (m) 192.4 200.0 108.0 48.8 349.0 Height (m) 6.6 7.6 1.5 4.6 8.2 Distance to Water (m) 30.6 0.0 53.9 0.0 130.4 Field (n=10) Length (m) 268.4 173.5 229.6 58.0 725.0 Width (m) 111.1 69.5 110.9 38.0 400.0 Height (m) 1.1 1.0 0.3 0.8 1.9 Distance to Water (m) 67.6 11.2 98.9 0.0 241.5 1There were two ridges that lacked one dimension of variation, and were excluded from this summary.

PAGE 230

230 Table 5-2. Metric attributes of shell ridges Number Name Length (m) Width (m) Height (m) Volume (m3) Min. Elev.1 (m) Water2 (m) Source3 Status 8LA1W-A Silver Glen Run-Locus A 200.0 75.0 3.0 1.61 29.0 Ex, LiDAR Mined 8LA21 Above Manhattan Landing 61.0 3.7 0.0 37.0 Ex Mined 8LA29 St Francis/Old Town 274.3 61.0 4.0 0.1 0.0 Ex, LiDAR Mined 8LA32 Crow's Bluff (Osceola) 76.2 30.5 4.0 0.0 Ex Mined 8LA88A Bowyer's Bluff 1 45.7 24.4 4.2 2011.0 0.5 288.0 Ex, LiDAR Intact 8LA88B Bowyer's Bluff 2 115.8 30.5 5.4 4613.0 0.5 246.0 Ex, LiDAR Intact 8MR123 Silver Glen Boil 300.0 75.0 3.7 0.0 0.0 Ex Mined 8MR2 Salt Springs Run 2 131.0 67.1 2.1 0.1 0.0 Ex, LiDAR Mined 8PU27 Horse Landing 114.3 53.3 4.6 56.0 Ex Mined 8VO7 Morrison's Creek 106.7 54.9 3.6 0.6 28.0 Ex, LiDAR Mined 8VO19 Mount Taylor 152.4 53.3 8.3 -0.2 171.0 Ex, LiDAR Mined 8VO41 Live Oak 140.0 80.0 5.3 6730.0 0.5 180.0 Ex, LiDAR Intact 8VO43A Blue Springs 152.4 45.7 4.0 0.0 Ex Mined 8VO44 Blue Spring Oxbow 47.0 15.0 3.4 376.0 0 9.0 Ex, LiDAR Intact 8VO162 McDonald (Paramore) 155.5 61.0 3.1 -0.4 68.0 Ex Mined 8VO214 Hontoon Dead Creek 155.0 55.0 5.5 7602.0 0.0 214.0 Ex, LiDAR Intact 8VO2600 Thursby Complex 182.9 1.8 0.0 0.0 Ex Mined 1Minimum elevation observable fr om LiDAR coverage (NAVD88) 2 Distance to water, either major lake, main channel, or tributary 3Source of metric observa tions. Ex = published accoun t or excavation record.

PAGE 231

231 Table 5-3. Metric attribut es of multi-mound complexes. Number Site Length (m) Width (m) Height (m) Volume (m3) Min. Elev.1 (m) Water2 (m) Source3 Status 8LA1-East Silver Glen Run East 400.0 250.08.2 0.0 0.0 Ex, LiDAR Mined 8LA89 Kimball Island 213.4 109.75.2 25707.0 0.0 130.0 Ex, LiDAR Intact 8VO22 Bluffton 589.0 349.07.6 0.0 0.0 Ex, LiDAR Mined (8VO23) Burial Mound A (attached) 30.5 30.54.9 Ex Mined Burial Mound B (attached) 32.0 20.0Ex Mined 8VO24 Harris Creek 330.0 280.07.6 0.0 0.0 Ex, LiDAR Mined 8VO55 Old Enterprise 134.1 48.84.6 0.0 0.0 Ex Mined 8VO58 Thornhill Lake 191.0 109.05.3 8366.0 0.0 0.0 Ex, LiDAR Intact Burial Mound A (attached) 35.0 35.03.4 Ex, LiDAR Disturbed Burial Mound B (attached) 27.0 27.02.7 Ex, LiDAR Mined 8VO202 Hontoon Island North 360.0 200.07.6 0.5 0.0 Ex, LiDAR Mined Burial Mound A (attached) 30.0 30.07.6 Ex Mined Burial Mound B (attached) 20.0 20.06.0 Ex Mined 1Minimum absolute elevation observable from LiDAR coverage (NAVD88) 2Distance to water, either major lake, main channel, or tributary 3Source of metric observa tions. Ex = published accoun t or excavation record.

PAGE 232

232 Figure 5-1. Shell matrix si tes considered in Chapter 5.

PAGE 233

233 Figure 5-2. Comparison of site plans for the Old Town/St. Francis Mound (8LA29 ). A) Sketch composed by Jeffries Wyman, scaled approximately (redrawn from Wymans Unpublished Journal, courtesy of Harvards Countway Library of Medicine). B) USDA aerial photograph from 1941 with mined ar ea highlighted. C) LiDAR-derived false-color and hillshaded topography with 1941 cl earing and Wymans sketch map superimposed.

PAGE 234

234 Figure 5-3. Scatter plot of measured height compared against the product of measured length and width for shell ma trix site types.

PAGE 235

235 Figure 5-4. LiDAR-derived cross secti ons of ridges. The projected height and configuration of 8VO19 is based on Moores (1893a:13) description and the remnant basal escarpment. The vertical exaggeration is approximately 5x.

PAGE 236

236 Figure 5-5. Comparison of shell ri dge height and distance to water.

PAGE 237

237 Figure 5-6. False-color and hillshaded topographic maps of intact shell ridges. A) Hontoon Dead Creek Mound. B) Live Oak Mound. C) Blue Spring Oxbow Mound.

PAGE 238

238 Figure 5-7. False-color and hillshaded topographic maps of intact shell ridges. A) Bowers Bluff 1. B) Bowers Blu ff 2. C) Kimball Island Mound.

PAGE 239

239 Figure 5-8. Footprints of ri dges and multi-mound complexes.

PAGE 240

240 Figure 5-9. False-color and hillshaded topographic map of the Thornhill Lake Complex (8VO58).

PAGE 241

241 Figure 5-10. False-color and hillshaded t opographic maps of mined multi-mound complexes. A) Hontoon Island No rth (8VO202) and the Thursby Mound/Hontoon Island Parking Lot Comp lex (8VO2600) (modified from Purdy [1991:Figure 35B]). B) Harris Creek (8 VO24) (after Jahn and Bullen [1978]) and Aten (1999).

PAGE 242

242 Figure 5-11. False-color and hillshaded t opographic maps of the mined Bluffton Complex (8VO22, 8VO23) (after Wheeler et al. [2000]).

PAGE 243

243 Figure 5-12. False-color and hillshaded topo graphic map of the mined Silver Glen Run Complex, showing the linear ridge (8LA1-A) and ring (8LA1-East).

PAGE 244

244 Figure 5-13. Old Enterprise Complex. A) False-color h illshaded topographic map (after Wheeler et al. [2000]), showing shell complex in relationship to the Groves Orange Mi dden wetsite component. B) ba thymetric contour of the surface of the wetsite component with the paleochannel highlig hted. Entire region to the east of the shoreline is saturated shell matrix deposits (afte r McGee and Wheeler [1994]).

PAGE 245

245 Figure 5-14. Sites with documented or suspected Mount Taylor components on Hontoon Island.

PAGE 246

246 Figure 5-15. Median calibrat ed age probabilities and elevati ons of radiocarbon assays from terrestrial and subaqueous contexts. A) Absolute elevation. B) Elevation relative to maximum flood recor ded at nearest USGS stream gage.

PAGE 247

247 CHAPTER 6 MOUNT TAYLOR IN MICROCOSM The world of Mount Taylor communities wa s inscribed at the regional scale in a series of shell ridges and fields linked by the flow of the river. The distribution and layout of shell matrix sites and their rela tionship to hydrology sheds light on long-term structuring principles. Without furt her context, this pattern remains a map (see Chapter 2), a totalized representation of the cumulative actions of M ount Taylor inhabitants (e.g., Barrett 1999:255; de Certeau 1984:119). In order to understand how communities emerged as a historical process, it is necessa ry to shift from the macroscale of regional inhabitation to the microscale of everyday paths and traces. In this regard, three related questions are of interest: (1) what traditions (sensu Pauketat [2001b]) did Mount Taylor communities generate and reproduce to appropriate existing places and define fields of practice; (2) how did the temporality and si tuatedness of these tr aditions frame the ongoing flow of inhabitation; and (3) how were these various practices and their referents reconfigured, transposed, and reva lued to either accommodate or effect change? In this chapter I focus on the first question, while Chapters 7 and 8 will explicitly examine placemaki ng as a historical process. How is it then, that M ount Taylor communities were reproduced through the routinization of shellfish collection and depos ition in place? Although the details of place-making are often treated as epiphe nomena, day-to-day practices and their material referents form the basis for commu nity identity, often in unconscious ways (Lightfoot et al. 1998; Pauket at and Alt 2005). It is th rough small-scale, seemingly inconsequential depositional acts and their organization as spaces that fields of practice are defined. The resultant physicality not only structur ed subsequent inhabitation but

PAGE 248

248 provided material referents that could be manipulated (de Cert eau 1984:118; Robin 2002:249). Although many traditions are enacted non-discursively, acts of settlement (re)creation inherently invoke memories of past places and peoples (Ingold 2000a:189; Pollard 1999). Thus, the (re)produc tion of place provides a model for how the world composed of different peoples and pl aces should be organized. Along the St. Johns, the deposition of shellf ish and other materials are the most visible evidence of Mount Taylor community tradi tions. Minimally, acts of deposition were the final movement of materials in the flow of dwelling. However, the emplacement of materials implies a chain of prior actions involving collection, processing, and in some cases consumption. After deposition, these materi als and their distributions provided material referents that structured how inhabitants subsequently experienced and interpreted biographies in place. As I will detail here and in the subsequent chapter, the system of reference established in the construction of local biographies formed the basis of regional economies of memory. These economies were variously reconfigured and deployed as the system of local and regi onal referents changed. Thus, the only way to understand large-scale regional inhabitati on, ritualization, and transfo rmation amongst Mount Taylor communities is to detail the routinization and reproduction of communities in place. To date, there is no better c ontext than the Hontoon Dead Cr eek Complex to examinein microcosm traditions of Mount Taylor plac e-making. The complex is situated on the southwest corner of Hontoon Island, midway along the Middle St. Johns valley. The locality is composed of a shell field (Hontoon Dead Creek Village, site 8VO215) and Mount Taylor ridge (Hontoon Dead Creek Mound, site 8VO214). The ridge figures

PAGE 249

249 prominently in the history of Mount Taylor communities as a memory place, and I will discuss its history in Chapter 7. Here, however, I will focus on the Hontoon Dead Creek Village, the details of which will serve as the basis for an explicitly diachronic and historical examination of Mount Taylor places in the next chapter. In brief, my field research demonstrates that the Hontoon Dead Creek Village is characterized by discrete shell deposits (her ein referred to as shell nodes) registering nearly 7000 years of repeated inhabitation spanning the Mount Taylor, Orange, and St. Johns periods. Internal divisions within nodes are indicative of differentiated activity areas, and in some cases may reflect coeval and equally spaced residential compounds. Time-transgressive trends ar e also evident. T he earliest deposits are coeval with Mount Taylor basal strata at the adjacent Hont oon Dead Creek Mound. Later Orange and St. Johns per iod inhabitation is situat ed away from the mound, a pattern that reflects the cessa tion of activities at the mound and localized hydrologic change. Despite the time span involved, and the regional transformations in community histories, the patterning between discrete s hell nodes indicates that there were longreproduced traditions regarding the proper partitioning of spaces to maintain places. History of Investigations Details of the Hontoon Dead Creek Complex emerged first from Wymans observations, and then from reconnaissance and stratigraphic exca vations by the University of Florida St. Johns Arc haeological Field School. Wyman (1875:26) described two shell mounds and two shell fields on Hontoon Island. The shell mounds include Hontoon Island North and the Hont oon Dead Creek Mound, situated on the northern and southwestern aspec ts of the island respecti vely. Adjacent to the southwestern shell rid ge Wyman (1875:26) noted:

PAGE 250

250 At the point where Huntoon Creek enters this lagoon is the remnant of a small shell field, and a second was f ound a quarter of a mile higher up. Both show signs of having been largely destroyed; from each, pottery, bones of animals, worked bones and shell tools were obtained, and from the heap last mentioned an arrowhead. Leaving the first mentioned shell field and following the edge of the swam p in a northwesterly direction for about a quarter of a mile, a large and conspicuous mound is reached. While frustratingly short, this passage rev eals several key aspects of the villages location and disposition. The lagoon Wyman mentions is today situated just south of the intersection of Hontoon Dead Creek and Snake Creek (Figure 6-1). Collectively these water bodies shape the western and s outhern boundaries of Hontoon Island. On the basis of this description, John Goggin enter ed the two shell fields into the FMSF as 8VO215, the Middle Midden situated on the lagoon, and 8VO 216, the Southern Midden situated above or upstream on Snake Cr eek, to the south. Of the structure of site 8VO215, Wyman suggests that 8VO215 was largely destroyed. Although he does not offer any further detail, a consideration of his descriptions of other sites along the St. Johns indicates he frequently referred to si tes as destroyed when there was evidence for river-bank erosion. It would appear from his description that he considered the site to not be horizontally extensive, and confined to the southern terrace edge approximately one quarter m ile south of the Hont oon Dead Creek Mound. The extent and nature of s ubsurface deposits at the complex were initially characterized during reconnaissance survey by the St. Johns Archaeological Field School (Endonino 2003a:102; Randall and Hallman 2005). The site was relocated and bounded with shovel test pits (STPs). Cult ural deposits were confined to an area measuring 50 m east-west and 100 m north-south along Hontoon Islands southwest terrace edge. Orange and St. Johns sherds, so me marine shell, and vertebrate faunal bone were found concentrated in the southern two-thirds of the site. Shell was

PAGE 251

251 encountered across much of the bounded site area and consisted of strata dominated by Viviparus shell, ranging from 30 to 50 cm in thickness. The field crew also identified shell and c oncreted sands eroding from the southern edge of the terrace that fronts Snake Creek. It is presumed that it was this same cutbank that suggested to Jeffries Wyman the si te was mostly destroyed, and is the likely location from which he collected materials. In contrast, shovel testing on the eastern back side of the ridge determined that shell does not extend past the visible topographic margins of the mound. This is consistent with Wymans observati on that the boundaries of ridges are discrete and shell was not randomly scattered. A casual surface survey of the site was also conducted after the acti ve 2004 hurricane season. Shell was noted on the surface, but did not appear to be a recent disturbance. However, three discrete and subtle topographic anomalies, higher than the surrounding terrain, and deviating from the slope of the terrace were observed. Because of dense ground cover, it was not possible to determine their orientation. They appeared to be arranged in a linear or curvilinear fashion along the terrace edge, spaced approximately 30 m apart. Their location was generally conform ant with the distribution of shell identified through previous shovel testing. Additional field work was conducted by the field school at the adjacent Hontoon Dead Creek Mound. A comprehens ive review of this work is provided by Sassaman (2005a), and I will present the implications of stratigraphy and chronology there in Chapter 7. However, three observations do warrant mention. Stratigraphic excavations provided evidence that the mound emerged r apidly, over no more than several centuries (7170 cal B.P., 2 ). In addition to massive lenses of freshwater

PAGE 252

252 shellfishlacking abundant vertebrate f aunal remainsthe mound is composed of routinized couplets of burned and unburned shell matrices. Collectively, these patterns attest to moments of rapid, staged construction of the platform. There is little evidence to support the conclusion that the ridge was a pl ace of habitation. Most of the deposits are characterized by shell lacking abundant ve rtebrate fauna or objects indicative of daily life. Because excavations did not in tercept the base of t he mound, such deposits may be internally present at lower elevat ions. Regardless, the bulk of the mound appears to date to the Mount Tayl or period. While sherds of the later Orange and St. Johns period were recovered in the trench, they were restricted to surface exposures and disturbances. A final key detail was revealed through bucket augering. Today 8VO214 is located some 200 m from permanently standing water, and is thus a significant distance from shellfish habitat. However, bucket augerin g in the cypress swamp west of the mound demonstrated that 8VO214 was once situated immediately adjacent a stream channel or lagoon. Wetsite deposits extending 30 m out from the base of the mound were identified under upwards of 1 m of muck (Figure 6-1). The di stribution of shell matrix was found to mirror that of the mound from north to south. These deposits became increasingly thinner away from the mound, and terminated abruptly in a 2-m thick sequence of shell-free muck 30-m east of the mound. This sequence likely reflects an in-filled paleochannel or lagoon of Hontoon Dead Creek. An AMS date on uncharred hickory from the saturated matrix yiel ded a conventional age estimate of 6040 14C yr (7150 cal B.P.). Similar patterning was identified at Groves Orange Midden on Lake Monroe (McGee and Wheeler 1994).

PAGE 253

253 Site Organization A primary goal of my field research was to map the surface and subsurface extent of shell-bearing deposits at 8VO215. In or der to determine the surface expression of shell, a detailed topogr aphic map of 8VO215 was generated. Hereafter, all measurements will be referenced to the local arbitrary datum.1 The resultant topographic map including sites 8VO214 and 8VO215 is presented in Figure 6-1, projected with 25-cm contour intervals. Mo st obvious on this map is the 5-m high Hontoon Dead Creek Mound to the north. To the west of the mound is the cypress swamp, and to the east is a low zone leadin g up to the pine/palme tto flatwoods above 8 m in elevation. This latter area lacks any subsurface shell as determined by reconnaissance survey. South of the mound, the terrace edge extends some 220 m in a southeasterly direction until its abrupt intersection with Snake Creek. This zone encompasses the known boundaries of site 8VO215. The locality is generally characterized by a noticeable slope trending 0-5%, down from 8 m in the east to 6.5 m to the west. This slope is typical for H ontoon Island and similarly configured landforms along the St. Johns However, between elevations of 7 and 7.75 m there are deviations from the general slope of the terrace edge, and are evident as high-points, typically 25 to 50 cm above the surrounding terrace. A higher-resolution topographic map of onl y 8VO215 is presented in Figure 6-2, projected with 10-cm cont ours. This resolution provides more detail on the structure of surface features. At least five nodes are evi dent. Just to the south of the mound is a small node, approximately 30-cm high and 10-m wi de. Roughly 20 m to the south is a 1 Topographic mapping of the complex was accomplished with a Nikon DTM-310 total station prior to the local availability of LiDAR.

PAGE 254

254 larger, ovoid area. It is characterized by a central elongated node, approximately 40-cm high, with subtle extensions to the north and east. A more extens ive circular node is evident 30-m to the south of this point, again about 40-cm higher than the surrounding terrain, and 20-m in maximum ex tent. Like the area just to the north, it has an attenuated slope to the west along the swamp margin, but extends further to the east. Some 40 m to the south of this area is a smaller and lower node, 20-cm higher than the surrounding terrain. Moving south again anot her 30 m is an extensive area of higher topography. This locale is composed of at least two domes 1 m higher than the surrounding terrain, and is characterized by hi ghly variable boundaries to the north and east. The southern edge is abrupt, ending in a near-vertical slope into Snake Creek. Topographic mapping confirmed the presence of discrete nodes at site 8VO215. They are aligned in a linear array along the terrace edge, spaced 30 to 60 m apart, and situated 7 m to 8 m above datum. Each node is characterized by a central elongated dome, although there is significant variation in the overall size, height, and structure of each. The distribution of shellfish relative to topographic variation was revealed through a subsurface survey. Testing was conduct ed with a bucket auger to gather baseline data on shell and non-shell matrices. Subsequently, a close-interval soil core survey was performed to test whether surface feat ures co-varied with shell deposits, and to determine shell density and thickness when po ssible. Similar procedures have been used by Russo et al. (2002) and Heide (2002) at coastal shell rings to delimit the extent of shell matrices relative to topographic expression.

PAGE 255

255 Limited testing was conducted with a 4-in ch bucket auger. A total of 17 auger tests were completed along four east-west transects perpendicular to the terrace. Within transects, auger-tests were placed at roughly 10-m intervals. Grid coordinates and elevations above datum were recorded for only augers extracted along Transect-2 (Figure 6-2). However, this transects profile can be generalized because other augers yielded similar results. Bu cket augering along Transect-2 rev ealed five distinct matrices (Figure 6-3). Higher elevations on the eas tern terrace edge yielded sand, underlain by concreted hard pan. At lower elevations to the west a thin lens of organic muck overlay shell or sand. Shell matrix of varying densit y was identified in Augers 6-8. Because of the limited scope of the bucket auger survey, li ttle can be said about the distribution of shell across the site. Bucket augering does, however, provide a key insight into the broader geomorphic context of anthropogenic deposits. Higher elevations are characterized by sand with an underlying hardpan, typical of Florida soils s ubject to frequent water table fluctuations within flatwoods. A surprising result of this survey is that neither shell or muck was encountered in appreciable quantities to the west of 8VO215. At 8VO214, thick muck deposits underlain by dense shell deposits were encountered upwards of 30 m to the west of the mound. At 8VO215, however, lower elevations are characterized by a 20 to 25-cm thick deposit of muck with limited shell. This suggests that there is not an early, inundated component to the west of 8VO215. By extension, the lagoon or channel that is implied by the stratigraphic sequence we st of Hontoon Dead Creek Mound was not present to the west of 8VO215.

PAGE 256

256 The results of topographi c mapping and bucket augering guided a close-interval core survey of 8VO215. The survey was conducted with a 1-inch Oakfield soil probe, which enabled rapid a ssessment of subsurface matrices. Cores were extracted at 2-m intervals across the site. Notes were re corded for each extracted core, including a description of matrices and thei r depth below surface. Matrices were broadly classified by the presence or absence of shell. Non-s hell matrices were typically categorized as terrace sand. Occasionally, non-shell anthro pogenic matrices were identified by the presence of bone or pottery frag ments. Shell-bearing matric es were divided into two categories based on relative density. Matric es dominated by soil interspersed with shell were classified as sparse or low-density shell. Matrices dominated by shell were characterized as dense shell. The distribution of shell-bearing deposit s superimposed on the contemporary topography of the site is present ed in Figure 6-4. Shell matrices are tightly restricted to a swath expanding from 10-m wide in the north to 40-m wide in the south, and follows the curvature of the terrac e edge. Shell-free anthropogeni c deposits were encountered in isolated cores throughout the site, al though they tended to cluster on the lower western edge of the shell depos its. In general, anthropogenic deposits closely overlap the surface features identified through topogr aphic mapping. There is a distinct northsouth trend of increasing surfac e elevations associated with shell. Adjacent to the mound, shell is found only at elevations 2030 cm above the terrace surface. In contrast, the highest elevations associated with shell are found in the southern aspect of the site, and rise upwards of 50 cm abov e the terrace. As noted through surface

PAGE 257

257 survey, shell was encountered up to the edge of Snake Creek, where it was visibly eroding out of the bank. Variability in the density and presence of shell is also evident. The eastern edge of shell deposits is the most distinct, where the contact between shell and shell-free cores is closely associated with abrupt topographic breaks with few exceptions. Isolated shell deposits were rarely encountered along this margin. Cores were most frequently characterized by dense shell, fu rther suggesting a difference between shell and non-shell deposits. This pattern contra sts starkly with the western edge, which generally conforms to the slope of the terrace. Diffuse an d isolated shell and non-shell deposits are more frequent and widely distributed at lower elevations on this western trailing edge. Coring did not follow all west ern deposits to their ma ximum extent, and it is unknown how far to the west the trend of isolated deposits may continue. Judging by the distribution of such deposits at the south end of the site, it is unlikely that they continue below elevations of 6.5 m. T he western edge is also dominated by sparse shell, characterized by a low abundance of shell in a sand or muck matrix. The underlying cause of the differential disposition of the eastern and western margins of shell is not directly evident from the core surv ey. This will be more fully discussed in the following section on the results of test uni t excavation. Shell free zones are also present between shell clusters. These zones tend to correlate with small depressions or gullies. From north to south ther e is also a general trend for decreasing distinctiveness between she ll and shell-free zones. At the northern edge of the site, there is a clearly defined zone separating the toe of t he Hontoon Dead Creek Mound and 8VO215.

PAGE 258

258 Shell deposit thickness values were der ived by subtracting the depth below surface at which shell was first identified by the maximum depth of shell encountered. These data should be considered minimum thickness values because cores could be extracted to a maximum depth of only 40 cm bel ow surface. Moreover, cores frequently could not penetrate through dense shell deposits, resulting in only the top of the surface identified. The resultant interpolated shell deposit th icknesses greater than 5 cm are plotted in Figure 6-5B. For comparative purposes the distribution of all identified deposits is presented adjacent in Figure 6-5A. As suggested by topography and the distribut ion of shell, discrete shell deposits are evident within the site boundaries. The smallest and most discrete is the northernmost, just south of the mound. South of this cluster the next cluster is composed of both thick shell to the east, and shallower shell on the downslope western edge. A small patch of shell is evident on t he southern edge of this cluster, but is not evident on the surface of the site. Another discrete cluster centering at N825.00/E1060.00 is characterized by th ick deposits to the east with less dense deposits to the west. As indicated by the dist ribution of shell, the southern half of the site contains generally diffuse (less than 15cm thick) shell, terminating in a large elongated dome of thick shell deposits. Subsurface survey determined that she ll deposits within 8VO215 are generally conformant with elevated surface topography. The density and distribution of shell within the site boundaries further indicates that shell is present in discrete clusters that are typically separated by lowdensity or shell-free zones. In some cases, internal divisions within these clusters may be presen t. The eastern, terrace margins of clusters

PAGE 259

259 tend to be characterized by abrupt changes in elevation associated with dense shell. Western edges are typified by diffuse and thin shell deposits trailing into the wetlands. Based on these patterns, at least five s hell nodes can be identified within the site, presented in Figure 6-5C. The term node is used to refer to discrete clusters of elevated shell. Surface topography and shel l presence, thickness, and density were considered when delimiting the boundaries of each node. While the boundaries of Nodes 1-3 are distinct, the division of Nodes 4 and 5 are less clear. In this case, the distinctiveness of the surf ace topography was given more weight in making such a division. These nodes serve to organize the results of test unit excavations discussed in the following section. Node 1 Node 1 is situated approximately 10 m sout h of 8VO214 (Figure 6-6). It is the smallest and most discrete shell deposit within 8VO215, and is separated from both 8VO214 and Node 2 by shell-free matrices. This is an early Mount Taylor component, based on stratigraphy, a lack of pottery, and a radiocarbon assay dating as early as 7300 cal B.P. The node is ovoid in shape, and measures 13 x 10-m in maximum dimension, and is only 50 cm high on the surf ace. The margins are characterized by sparse shell, between 5 and 10-cm thick, as determined through coring. Cores within the apex of the node consistently encounter ed dense and impenetrable concreted shell below a 5 to 10 cm thick organic mantle. Inve stigations initially targeted the relatively flat apex with TU3, a 1 x 2-m unit. Excava tion of this unit ceased after two levels because concreted shell was encountered, approximately 15 to 20 cm BS. Subsequently TU3A, a 1 x 1-m unit, was placed to the west of TU3. Descriptions of

PAGE 260

260 identified stratigraphic units are presented in Table 6-1, and a tabulation of recovered material culture is presented in Table 6-2. Three distinct strata were revealed in TUs 3 and 3A. Stratum I is a black/dark brown silty fine sand encountered immediately in Level A across both test units. This is a unique stratum, the characteristics of which were not encountered anywhere else during excavations. It is possible that the upper 5 cm are non-anthropogenic, as few cultural materials were encountered in this upper matrix. However, as excavation proceeded towards the base of Level A, incr easing quantities of vertebrate fauna and large fragments of charred wood were recovered. A 14C assay of this charred wood showed that at least the wood fragments were related to a recent surface fire. The trend of increased particulate organic matter coupled with abundant vertebrate fauna and large charred wood fragments continued wit h the excavation Stratum I in Level B. Material culture recovered from Stratum I included three lithic waste flakes and four fragments of modified bone, two of which may be a bone awl. Lying unconformably below Stratum I is co ncreted shell, designated Stratum II. Based on the core survey results, this dome-lik e shell deposit forms th e core of Node 1. Excluding a few pockets of partially disaggr egated shell at the cont act with Stratum I, the shell was too concreted to excavate with trowels or shovels. In order to determine the thickness and composition of the matrix a test pit measuring approximately 30 x 30cm was excavated in the center of TU3 wit h the aid of chisels, hammers, and a large pick-axe. This test pit enc ountered approximately 30-cm of w holly concreted shell lying on top of culturally sterile s and (Stratum III). In order to further characterize the

PAGE 261

261 structure of the deposit, and collect materials for a radiocarbon assay, a large block of material was removed from the pr ofile of the test-pit and retur ned to the lab (Figure 6-7). Closer examination of the sample reveals it is composed of tightly packed crushed and whole Viviparus and bivalve shell in an ashy fine sand matrix. Small fragments of vertebrate fauna and charred material are ev ident throughout. There is also a trend towards increasing bivalve shell with depth, evident in the photograph as the lighter lower half of the block. The base of the bl ock is dominated by mostly whole bivalve with abundant fragments of charred material dist ributed throughout. Whether or not this represents a discrete lens or feature cannot be determined fr om the available data. A sample of charred material from the base of the block (ca. 44 cm BS) was submitted for AMS radiocarbon assay, and returned a corrected age estimate of 6280 40 14C yr (2 calibrated range of 7310 cal B.P. [Beta-219933]). This age estimate is the earliest published assay for freshwater shell deposit withi n the St. Johns region. However, this assay falls within the 2 range of assays derived from the wet site deposits to the west of the adjacent Hontoon Dead Creek Mound, as well as the base of the nearby Live Oak Mound (8VO41) on the eastern terrace of the St. Johns (Sassaman 2003a). Excavations of TU3 and TU3A revealed a sequence co mposed of dense concreted shell capped with an organicallyenriched shell-free matrix. These deposits suggest at least two different depositional proce sses are responsible fo r their formation. In many respects Stratum II shares commonalities with deposits of similar age at nearby shell mounds, characterized by concret ed and fragmented shell associated with ash and other materials. Varying hypotheses hav e been offered on the origins of concreted

PAGE 262

262 shell, although it is generally thought that a combination of bur nt shell, ash, and fluctuating water levels are needed for concretion to form (Wheeler et al. 2000). Regardless, concreted shell is typically rest ricted to the base of shell mounds, although it can also occur as lenses within sites. Stra tum II is unique in that it is lying relatively unprotected on the surface. Node 2 Node 2 is evident on the surface as an el ongated ridge that ri ses 40 cm above the surrounding terrain (Figure 6-8, 6-9). It measures 25-m l ong and 8-m wide along a north/south axis, and is situated 20 m south of Node 1. The ridge has a generally flat summit, except for an apex abov e 7.7 m on the northern aspect. Based on surface survey, this apex likely re flects upheaval of underlying matrix from a large tree positioned on the ridge, and not t he differential deposition of shell across the surface. The downslope edge west of the ridge is evident only as a subtle bulge topographically. Coring indicated that the central ridge is composed of dense shell that terminates abruptly to the east in organi cally enriched non-shell terrace sands, and trails off to lower elevations in the west as low density shell. Test Unit 2 was positioned on the eastern edge of the central ridge with the goal of penetrating thick shell deposits and documenting the relationship bet ween shell and non-shell matric es. Eight stratigraphic units were identified during excavation. All strata are Mount Taylor in age based on a near-surface radiocarbon assay and recover ed objects. Composit e profile drawings and photographs of TU2 are presented in Fi gure 6-10, and descriptions of identified stratigraphic units are provided in Table 5-3. Tabulation of re covered material culture is presented in Table 5-4, and photographs of sele cted artifacts are presented in Figure 611.

PAGE 263

263 The most significant result of TU2 excavations was the identific ation of a stacked sequence of shell situated adjacent to unconsolidated shell-free deposits above culturally-sterile basement sands. These st rata indicate that shell was routinely emplaced on top of a preexisting shell surface, while at the same time a shell-free zone was maintained at higher elevations on the te rrace side. Lying below Stratum I in the western 180 cm of TU2 is a stacked sequence of shell lenses lying more or less horizontal, excluding the basal Strata V and V I. In profile, no less than five ca. 10-cm thick shell-matrix strata are discernable based on the relative abundance of whole and crushed shell. These strata alternate bet ween primarily whole and primarily crushed shell. The mechanical taphonomic process res ponsible for crushed shell lenses is not known. They are considered as a proxy fo r post-depositional tramp ling, and indicative of heavily used activity areas (Randall and Sassaman 2005; Sassaman 2003a). Others have suggested that they result mostly from bioturbation (Beaton 1985). While it is currently not possible to rule out either of th ese factors, both are indicative of a hiatus between depositional events and post-depositional mechanical alteration. The implication is that inhabitant s routinely emplaced shell in the same location after a period of abandonment. In the case of TU2, at least three such surfaces are present (Strata II, IV, IV). In each case a roughly flat lying crushed shell lens is superimposed on a stratum composed primarily of whole Viviparus and some crushed and burned bivalve shell. A fragment of marine shell re covered at ca. 9 cm BS was submitted for a radiometric assay which returned a conventional age estimate of 5950 14C yr (6480 cal B.P. [Beta217769]). This age estimate fa lls well within the accepted

PAGE 264

264 range of the preceramic Archaic Mount Taylor period, and provides a useful termination date for preexisting deposits. Shell tapers out into a shell-free component to the east. In t he eastern edge of the unit, approximately 15 cm BS, Stratum I grades into VIII, a dark grayish brown fine silty sand that appears to be an organically enriched pedon lacking any clear vertical divisions in profile. Stratum VIII was vi rtually shell free and contained occasional vertebrate fauna that decreased in abundance with depth. At approximately 40 cm BS Stratum VIII grades into Stratum VII, a dark brown loamy fine sand with occasional mottling and mineral concretions throughout. This was excavated as Zone C. Excluding the contact between t he overlying shell in Strata V and VI, this basal sand deposit is free of shell and vert ebrate fauna. Excavation ceased at the base of Level H (80 cm BS) due to water. Diverse, if low density, artifact assemb lages suggest a wide range of activities occurred on site. Relatively abundant faunal remains were recovered from throughout the sequence, although there is a trend of decreasing faunal remains by depth, particularly below 40 cm BS (Level D). Paleofec es were also recovered, although it is unknown if these are of human origin. Excl uding a biface fragment from Stratum I (Level A/Zone A), all material culture was re covered from the she ll deposits. The lithic assemblage includes two chert flakes and a small medial fragment of a biface. Three modified bone fragments were recovered throughout the strata. These all appear to be portions of cut and ground bone awls (Figure 6-11A). Marine shell was recovered, including fragments and tools. Modified fragments in the assemblage include a columella with a bitted siphonal canal (Figure 6-11B), an Oliva sp. shell with a fractured

PAGE 265

265 apex (Figure 6-11C) and a battered fragmented apex of a Busycon sp. Figure 6-11D). This latter fragment was the marine shell submitted for radiometric assay from Stratum II. In summary, excavations of TU2 identifi ed a preceramic Archaic Mount Taylor sequence of shell deposits in primary context, associated wit h a shell-free, organically enriched matrix. Judging from stratigraphic s uperposition, Stratum V II is the original, sloping terrace surface upon which shell was init ially deposited. Over the course of at least three depositional events, shell was successively added to this surface in ca. 20cm thick lenses, the surfaces of which were either trampled or bioturbated resulting in crushed shell. A separate prepared area to the east of the shell is implied by the presence of vertebrate faunal remains in a sh ell-free, organically enriched matrix. The sharp contact between shell and non-shell matrices suggests the eastern edge of the node was routinely kept clean of shell debris. Further te sting would be necessary to determine whether there are domestic features such as posts or hearths associated with the shell. Finally, limited sediment deposition is implied by the lack of interstitial sand within the shell, and the relative el evation of eastern non-shell deposits. Node 3 The center of Node 3 is located approximat ely 50 m south of Node 2. Evident as a sub-triangular dome on the surface, Node 3 measures 25-m long and 20-m wide, and is centered between 7.5 and 7.9 m in elevation (F igure 6-12). Coring within this central dome encountered both dense an d low-density shell on the terrace. Coring als o identified a low-density shell deposit extending along the western slope, above 6.6 m in elevation. Bucket augeri ng along the southern edge of Node 3 encountered a 10 to 20cm thick low-density shell stratum at lowe r elevations. Like Node 2, the eastern

PAGE 266

266 boundary was discrete. Test Unit 1 was situated on a noticeable ridge that extended out from the southern edge of Node 3. As the first test unit of the field season, placement was guided by the presence of s hell on the surface, derived from a nearby uprooted palm. The unit was oriented parallel to the slope of the ridge. A total of seven shell and non-shell stratigraphic units were identified. Based on the distribution of diagnostic ceramic sherds, the ceramic Archaic Orange period is well represented. A Mount Taylor component may be present based on the lack of pottery at depth. Profile drawings and photographs are presented in Figure 6-13, wit h stratigraphic descriptions presented in Table 6-5. Tabulations of recovered material culture by level are presented in Table 6-6, while photographs of selected objects are presented in Figure 6-14. Test Unit 1 revealed a sequence reminiscent of TU2 in Node 2. The northern, higher half of the unit is characterized by den se shell which decreases in thickness to the west, and terminates in an organically enriched shell-free sand. Despite general similarities with TU2, there ar e a number of important differences. Most notable is that Orange period fiber-tempered sherds were re covered throughout the shell and non-shell matrices. Secondly, only one or two depositional events are suggested by the sequence in TU1. Stratum I and II developed post-occupation. A small amount of vertebrate fauna and one Orange crumb sherd were recovered; these were derived from the contact with t he underlying Stratum III shell. Below this sterile overburden the unit is characterized by shell in the north half of the unit, and shell-free matrix in the south. In the northern half of the unit is Stratum III, a 20 -cm thick shell deposit (Levels A, B, and C). The shell matrix is composed primarily of whole and crushed Viviparus

PAGE 267

267 shell, with bivalve and Pomacea shell occurring in small quantities throughout. The matrix is relatively homogeneous, although sli ghtly more crushed shell was evident at the contact with the overlying St ratum II. The base of the Stratum III is essentially lying flat above shell (Stratum IV) to the north and sand to the south (Stratum VI). Beneath Stratum III in the north half of the unit is Stratum IV, a fi ne gritty sand with moderate amounts of whole and crushed Viviparus bivalve, and Pomacea shell. Shell abundance is considerably lower than the overlying stratum. The cu ltural contents of this stratum (excavated as Level D) were sim ilar to Stratum III. In the southern half of TU1, excavations encountered a buried A horizon contemporaneou s with the Strata III/IV shell deposits in the nor thern half of the unit. The recovered cultural assemblages were diverse and consistent with Late Archaic Orange period residential occupation, like that docum ented at the nearby Blue Spring Midden B (Sassaman 2003a). Within s hell matrices (Zone B), vertebrate fauna was found in moderate abundance, with density increasing with depth. Two small fragments of unidentified mari ne gastropod shell were also recovered. Orange plain sherds were present throughout the deposi t, but were found in greatest abundance towards the base of the shell strata. Two fragmentary hafted bifaces manufactured from chert were recovered within the subsistence column, lying adjacent to each other (Figure 6-14D,E). Both are consistent with Late Archaic stemmed va rieties. Vertebrate fauna was recovered in abundance throughout the non-shell anthropogenic strata at lower elevations. In addition, a large Archaic stemmed chert hafted biface was recovered (Figure 6-14G). The distal end of the biface has evidence of doublepatination, suggesting it may have been scavenged from a preexisting deposit and

PAGE 268

268 subsequently resharpened. Also recovered from the shell-free matrix were two distal bone tool tips (Figure 6-14AC). Test Unit 9 was emplaced on the western edge of Node 3 to intercept the contact between shell and non-shell matrices. Coring in this upslope component encountered an uneven pattern of lowand high-density shell deposits interspersed with the occasional core lacking shell altogether (Fi gure 6-5). Contrary to expectations based on coring, excavation of TU9 exposed a complex sequence characterized by a dichotomy between flat-lying shell and non-shell deposits in the northern half of the unit, and dense shell in the southern half of the unit. In contrast, the east and west profiles contain discontinuous and interbedded lens es of shell and non-shell matrices. The structure of this unit registers at least one shell-filled pit associated with a non-shell matrix. An Orange period component is indicated by th e presence of plain and engraved fibertempered pottery. A lower preceramic Archaic component may also be present, but could not be delineated. Inhabitation of Node 3 occurred principally during the Orange period. Excavations of TU1, TU9, and associated coring suggest that Node 3 is composed of discrete activity areas. Two depositional events and a possible living surface were exposed along the western, swamp-facing shell escarpm ent in TU1. The shell matrix contained two macrostratigraphic deposits of shell likel y representing two peri ods of deposition. The lower shell-free surface of TU1 appeared to contain a buried A horizon with lithic artifacts and Orange sherds. Whether this horizon represents an occupational surface is unclear. Coring identified a flat surface with low-density shell a few meters to the southwest. This extension of shell may repres ent yet another activity surface, but this

PAGE 269

269 cannot be verified with the current data. Upslope on the terrace edge, TU9 encountered at least one large shell pit truncating a shel l and shell-free sequence. The lower strata may actually date to the Mount Taylor period. Like TU 1 the upper shell matrix was situated immediately adjacent to an or ganically enriched horizon that contains vertebrate fauna in addition to Orange pottery. Other features in the vicinity are suggested by the generally hit-or-miss results of coring in this upslope component of Node 3. It is likely that other features are present on this landform, and may provide evidence for a residential structure. Node 4 Node 4 is a diffuse shell deposit located im mediately south of Node 3. Coring in this locality identified widespread high and lo w-density shell across the terrace. In maximum dimensions the node is 48-m long and 20-m wide (Figure 6-15). Like other shell nodes it is characterized by a centralized dome. Within Node 4 this dome is relatively small at 15-m long and 6-m wi de, and rises 20 cm above the surrounding terrace surface. Excavation of Test Unit 4 yielded a convoluted pict ure of deposition. One component dating to t he Orange period was identifi ed by a few Orange crumb sherds. A sequence composed of developed soil horizons superimposed on shell was documented. Profile drawings and photogr aphs are presented in Figure 6-16, with stratigraphic descriptions presented in Table 57. Tabulations of recovered material culture by level are presented in Table 5-8. As seen in profile, Stratum III is a moderat ely dense shell matrix that decreases in thickness, from 24 cm to 10 cm towards the western, downs lope edge of the unit. This stratum is characterized by whole and crushed Viviparus shell, with bivalve and Pomacea occurring in lower frequencies throughout a tan/brown fine sand. Possible

PAGE 270

270 animal burrows were noted during excavation, and are the likely source of shell matrix disturbance. Moderately dense vertebrate faunal bone was recovered throughout this stratum, in addition to two fiber-tempered crumb sherds and one no n-diagnostic biface fragment. Underlying all shell deposits is a gritty fine sand (Stratum IV) that graded into a concreted sand (Stratum IVa). These stra ta represent the original, pre-habitation terrace surface. Stratum IV was first encountered in Level D, where it was treated as Zone C. The contact between this gritty sand and the overlying shell undulated, and pockets of concreted gritty sand were found within the shell. Excavation in Level E was restricted to a pocket of non-concreted Stra tum IV sand in the northwest component of the unit, as seen in profile. As was the case with Stratum III, it appears that there has been significant biogenic rework ing of these deposits. Node 4 shell matrix strata were generally th in and without clear internal structure. Moreover, widespread evidence for post-depos itional disturbances, including animal burrows and possible tree-throws in antiq uity were evident. However, the characteristics of this component of Node 4 provide some perspective on thin and diffuse downslope deposits elsewhere at 8VO215. Test Unit 4 documented that at least in this vicinity, western edges of shell nodes do not appear to have been intensively used, as is evident in the lack of well-defined crushed shell lenses. Node 5 Node 5 is coterminous with the shallow s outhern deposits of Node 4, as defined through coring (Figure 6-17). At a maxi mum the node measures 50-m long and 40-m wide. Like other locales within 8VO215, the eas tern margin is marked by an abrupt shift between topographic r elief and non-shell de posits. The southern margin is characterized by a steep escarpment, refl ecting bank-erosion from Snake Creek. The

PAGE 271

271 western margin is characterized by low-den sity shell deposits that are diffused within the swamp. Sub-areas differentiated by mi cro-topographic variations are evident within the boundaries of this node. These are labe led areas A for the north and B for the south in Figure 6-17. The division between areas is formed by the 7.9 m contour interval. Locus A is the center of Node 5, composed of a central dom e 20 m in diameter that rises 50 cm above the terrace. A semi -circular depression is also evident on the northern aspect of Area A. This depression is bordered on the east and west by arcuate extensions of shell matr ix. Area B to the south is a more typical central dome of shell, measuring 15 m in diam eter, and separated from Area A by a slight depression. Four test units excavated within Areas A and B in Node 5 demonstrate that it is principally shell matrix depos ited during the St. Johns I per iod. The structure of deposition as best revealed in a 6-m long trench (TUs 5), where shell was emplaced, apparently in several deposit ional episodes, to extend the upslope node (Figure 6-18, Tables 6-9, 6-10, 6-11, 6-12 ). In profile, shell dips down to the west and north, suggesting deposition occurred from near the top of Area A (Figure 6-19). This sequence is different from other nodes in that Node 5 is has a wide summit. It may be that any residential structures were emplaced on top of t he apex, and not adjacent to the node as is likely the case in Nodes 1, 2, and possibly 3. Alternatively, such features, if they exist at a ll, may be present in the non-s hell upslope component of the terrace where shovel testing recovered St. Johns sherds. Testing between the apexes in TU8 failed to identify any clear evidence fo r habitation features such as differentiated shell strata, post holes or pit features. Thos e features identified were apparently small discontinuous lenses of charcoal or crus hed shell. Finally, a preceramic component

PAGE 272

272 may be present at the base of Node 5 as suggested by low density shell below the concreted sand zone. Radiocarbon determinations may help resolve this issue in the future. Microscale Histories The methods used to delineate the hist ory of the Hontoon Dead Creek Village demonstrated that it is com posed of discrete shell depos its registering nearly 7000 years of repeated inhabitation, spanning t he Mount Taylor, Orange, and St. Johns periods. In terms of size and apparent significance, t he Hontoon Dead Creek Village is at first glance dwarfed by the adjacent shell ridge (Table 6-18). The seemingly mundane c haracter of deposits within the confines of 8VO215 is overshadowed by the quantity of shell and implied si gnificance of large-scale acti vities at the ridge. The details of inhabitation in the microscale, however, reveal how communities reproduced traditions of settlement through the routinized deposition of shellfish. These principles were not limited to the establishment of new places, but were enacted to renew preexisting referents. The em phasis on partitions of places and the renewal of surfaces appear to have been the central structuring principles through which communities reproduced and transformed social histories. Not only were the practices of moving through and inscribing habitation sites the likely the way in which day-to-day senses of place were created, but practi ces and their material refer ents served as metaphors for community and history. Disregarding intrasite temporal patterning, the shell nodes at 8VO215 share several characteristics in structure and plac ement. Comparative metric data on shell nodes are presented in Table 6-13. The west ern, swamp-facing aspects of nodes are characterized by low-density shell along the slope, indicative of disposal activities

PAGE 273

273 oriented preferentially towards the water, and away from other upslope components. Shell nodes all terminate on t he western edge at elevations between 6.65 m and 6.8 m above datum, and there is only limited evidence for deposition within the swamp itself. In most cases, the northern and southern lateral margins were diffuse, with little evidence for an organized use of space based solely on shell deposition. The eastern, upslope components are defined by abrupt topographic relief and a discontinuation of shell. Shell was routinely deposited on t he gently sloping terrace edge, which was visible at the base of most test units. In each case the net effect was to create a linear or dome-shaped ridge of higher ground, and in some cases may have provided a foundation for architectural features. Fina lly, shell nodes are roughly equally spaced at intervals between 27 and 57 m, measured betw een the geographic center (centroid) of each node. Taken together, the shell nodes likely represent discrete residential localities differentiated into multiple, prepar ed activity areas. While no architectural features such as post-holes were observed, the totality of the evidence points towards shell nodes representing organized resident ial compounds of unknown duration, but which were nonetheless returned to repeatedly. Today the village is arranged in a linear arra y, but there is not enough evidence to demonstrate the contemporaneity of Mount Taylor nodes or the presence of Mount Taylor components at the bases of nodes to the south. A much more complex perspective on the long-term history of i nhabitation emerges when the shell nodes are considered in spatial and te mporal contexts. Illustrat ed in Figure 6-20 are two crosssectional profiles of sites 8VO214 and 8VO215 with the distribution of culture-historical components detailed at the botto m of the figure. The uppe r profile depicts the surface

PAGE 274

274 topography, with the presence of near-surface shell depos its shown. As has been detailed in this report, elevated surface topography is the result of shell deposition. Also depicted in the lower profile are the swamp surface and subs urface stratigraphy. Most notable in figure 6-20 are satura ted shell matrix deposits fronting the Hontoon Dead Creek Mound and r adiocarbon dated to 7150 cal B.P. The matrix is thickest along the western edge of the mound, and trails o ff to the north and south. More importantly, the contac t with the underlying sand can be seen dipping in front of the mound, and increasing in elevation to t he north and south. Muck deposits are again thickest in front of the mound, and thinnest in the north and south. This distribution of muck, wet site deposits, and sand suggests that there was once a lagoon or channel that has been infilled since 7100 cal B.P. (Figure 6-21). In the LiDAR DTM a series of weakly developed linear formations are s een at the confluence of Snake Creek and Hontoon Dead Creek. These suggest that t he sand body in front of Hontoon Dead Creek Village has migrated to the south through time. W hen these features began to form is unknown, but their shape and amplitude ar e similar to others that have formed around embayments and lagoons elsewhere al ong the St. Johns. Through time, the lagoon or channel filled in through the aggradati on of organic sediments, likely from the formation and repeated oxidati on of peat deposits, similar to Groves Orange Midden. The body of evidence indicates that inhabitation was established as early as 7300 cal B.P., adjacent to a lagoon or river channe l. The incipient occupation involved localized deposition at Node 1, which creat ed a low mound around which activities were centered. Within a century or two, the area to the immediate north (i.e. the Hontoon Dead Creek Mound) became the fo cus of community activities. A suite of radiocarbon

PAGE 275

275 assays indicate that as much as 3 m of shell were emplaced on top of the mound between 7150 and 6640 cal B.P. We do not know when deposition on the mound surface was terminated, nor do we know if places like Node 1 were subsumed under the mound (see Chapter 7). However, by 6400 ca l B.P. a new area of inhabitation was established to the south of the mound. Inhabitation continued, and reproduced the same depositional and spatial principles, but on a much smaller scale. Inhabitants may have referenced the mound in the plan of Node 2, although it is likely that the significance of this configuration has its or igins at the beginning of the Mount Taylor period. Importantly, the existing footprint of shell was ma intained as a model around which subsequent inhabitation co uld be ordered. These small-scale practices draw our attention towards those moments when practi ces were decidedly of a larger scale, potentially for a larger number of people, and referenced differ ent times. Within these subtle acts we can see the possibilities of how communities were structured with reference to existing material conditions, and how inhabitation was framed by material referents and soci al context. Like ridges at the regional scale, explanation for the location and scale of nodes cannot be reduced to changing water levels thr ough time. That is, the height of shell deposition cannot be used as a proxy for risi ng water levels. In Figure 6-22 I have digitally flooded the LiDAR DTM based on t he elevation of the minimum frequent high flood (which can be expected to occur every 3 years) (Figure 622A), and the 50-year flood (Figure 6-22B). These are the same hydrological patterns examined at the regional scale in Figure 5-15. Today all nodes and the ridge are above the elevation of the three year flood, suggesting that inhabit ants were well aware of the hydrologic

PAGE 276

276 regime and emplaced their nodes accordingly. During a 50-year flood event, portions of 8VO215 are submerged, and the ridge is partially an island. Although one might take this as evidence for the need to build up floor s to rise out of the water (thus looking at shell deposits solely in functional terms), I woul d note that it is only a short 50 m walk to be outside the danger of an extreme flood event. What was important was the nearby lagoon and the social histories that c ould be emplaced and negotiated in ridge-top performances. It is in this fr ame of reference that the stru cture, scale, and histories of practices at ridges need to be problematized.

PAGE 277

277 Table 6-1. Stratigraphic units of Test Units 3 and 3A, Node 1, 8VO215 Stratum Max. Depth (cm BS)1 Munsell Color Description I 15 10YR2/1 Organically enriched very silty fine sand; abundant palm roots throughout; large and small charcoal clasts throughout; increasing amount of vertebrate fauna with depth, some degraded shell fragments at undulating contact with Stratum II. (80 50 [BETA244053]) II 45 10YR6/2 Concreted whole and crushed Viviparus and bivalve, with very ashy fine sand; charcoal flecks and peasized clasts are apparent throughout; increasing bivalve with depth; base is situated above unconsolidated terrace sands (not encountered in profile) III 45+ n/a Fine sand; culturally sterile 1maximum depth in east profile below surf ace at southeast cor ner (N904.29 E1037.57 7.36) Table 6-2. Cultural materials recover ed from Test Units 3 and 3A, Node 1, 8VO215 Level Lithic Flake Modified Bone Vertebrate Fauna (g) Test Unit 3 A 1 1 176.4 B Zone A 2 1 849.2 B Zone B 7.8 Test Unit 3A B 2 125.8 C 245.7 Total 3 4 1404.9

PAGE 278

278 Table 6-3. Stratigraphic units of Test Unit 2, Node 2, 8VO215 Stratum Max. Depth (cm BS) 1 Munsell Color Description I 4 10YR2/1 Organically enriched loamy fine sand with abundant palm roots; trace amounts of crushed shell. II 16 10YR3/1 Very abundant crushed and some whole Viviparus and bivalve, occasional Pomacea ; crushed at contact with Stratum I; limited non-shell ashy very fine sand in shell matrix. (5950 60 [Beta-217769]) III 35 10YR3/1 Abundant whole and crushed Viviparus and Pomacea with some crushed bivalve; notable concentrations of whole Pomacea and other areas of crushed bivalve; slightly more sand in matrix than overlying Stratum II; zones of concreted shell throughout IV 48 10YR4/1 Very abundant crushed and whole Viviparus bivalve and some Pomacea ; more crushed shell and very fine sand than Stratum III; some concretion in horizontal layers V 61 10YR3/2 Abundant whole and crushed Viviparus and bivalve; some crushed Pomacea ; some bivalve appears to be burned; increasing sand in matrix with depth; concreted in some portions VI 55 10YR3/1 Very abundant crushed and whole bivalve, in addition to abundant whole and crushed Viviparus ; some Pomacea VII 75+ 10YR4/2 Fine silty sand; occasional palm toots throughout; shell only occurs in trace amounts VIII 75+ 7.5YR3/2 Loamy fine sand, some roots throughout, shell free 1maximum depth in west profile below surface at southwest corner (N870.757 E1047.19 7.69 m).

PAGE 279

279 Table 6-4. Cultural materials recover ed from Test Unit 2, Node 2, 8VO215 Level Lithic Marine Shell Modified Bone Paleofeces Vertebrate Fauna (g) Flake Biface Fragment Modified A 15.9 A Zone A 1 120.0 B 1 3 288.2 C 1 461.2 D 2 1 2 294.6 E Zone A 31.8 E Zone B 1 1 46.9 F 1 F Zone A 30.7 F Zone B 1 3.4 G 20.4 H Zone C 2.6 Total 2 1 5 2 3 2 1315.7 Table 6-5. Stratigraphic units of Test Unit 1, Node 3, 8VO215 Stratum Max. Depth (cm BS) 1 Munsell Color Description I 10 10YR5/1, 10YR2/1 Organically enriched fine loamy sand with abundant palm roots and detritus II 21 10YR5/3 Fine silty sand with moderate amount of palm roots; small flecks of charcoal throughout III 38 10YR3/1 Fine silty sand with abundant whole and crushed Viviparus moderate whole and crushed bivalve, occasional Pomacea IV 57 10YR5/1 Fine silty/gritty sand with moderate amount of whole and crushed Viviparus occasional bivalve and Pomacea ; shell appears to decrease with depth, no lenses of crushed shell apparent V 62 10YR2/2 Organic very fine silty sand with occasional roots throughout; buried Organic A horizon associated with Stratum III/IV deposits VI 62 10YR4/2 Sterile gritty silty fine sand; very rare shell fragments throughout VIa 75+ 10YR4/2 Sterile concreted silty fine sand with occasional calcreted root casts and gleying throughout basal depths; contact with parent Stratum VI undulates 1 maximum depth in west profile below surf ace at northwest co rner (N820.87 E1060.69 7.85 m)

PAGE 280

280 Table 6-6. Cultural materials recover ed from Test Unit 1, Node 3, 8VO215 Level Orange Sherd Lithic Marine Shell rag. Mod. Bone Paleofeces Vertebrate Fauna (g) Plain Crumb1 Flake Hafted Biface A Zone B 1 11 1 77.1 B Zone A 20.9 B Zone B 1 26 22 1 178.4 C Zone A 1 24.4 C Zone B 7 22 1 290.1 D Zone A 7 33 1 195.9 D Zone B 26 1 2 376.2 D Zone C 6.5 E Zone A 1 11 3 2 251.1 E Zone C 22.7 F Zone C 5.2 Total 17 119 3 3 3 2 2 1448.5 1 Any sherd that is less than 1/2-inch in minimum dimension 2Recovered in adjacent subsistence column (Stratum III)

PAGE 281

281 Table 6-7. Stratigraphic units of Test Unit 4, Node 4, 8VO215 Stratum Max. Depth (cm BS) 1 Munsell Color Description I 14 10YR5/1 Organically enriched fine sand; abundant palm roots and detritus II 25 10YR5/3 Fine sand; occasional palm roots and charcoal flecks; contact with Stratum III undulates III 40 10YR4/2 Abundant whole and crushed Viviparus some crushed bivalve and Pomacea ; shell appears more crushed at the contact with Stratum II; matrix is a fine ashy sand IIIa 48 10YR3/2 Fine loamy moist sand with moderate amounts of whole and crushed Viviparus occasional crushed bivalve and Pomacea ; appears to be organically enriched low shell density zone associated with Stratum III, possibly a buried A horizon; only present in south wall IV 54 10YR5/3, 10YR7/3 Moist gritty fine sand with occasional shell fragment; occurs as either unconsolidated matrix or as small clasts IVa 54+ 10YR5/3, 10YR7/3 Concreted gritty sand; surface undulates with Stratum IV; concreted terrace sands 1maximum depth in north profile below surf ace at northeast co rner (N774.64 E1082.65 7.59 m) Table 6-8. Cultural materials recover ed from Test Unit 4, Node 4, 8VO215 Level Orange Crumb Sherd Biface Fragment Vertebrate Fauna (g) A 65.6 B Zone A 1 59.1 B Zone B 1 288.2 C Zone A 1 116.5 C Zone B 187.6 D Zone A 52.1 D Zone B 1 1 106.4 D Zone C 7.6 D Zone D 2.5 E Zone C 0.2

PAGE 282

282 Table 6-9. Stratigraphic units of Te st Units 5, 6, and 7, Node 5, 8VO215 Stratum Max. Depth (cm BS) 1 Munsell Color Description Ia 42 10YR6/1, 10YR2/2 Organically enriched fine sand; abundant palm roots and organic detritus Ib 60 10YR4/2 Organically enriched fine sand; moderate palm roots Ic 52 10YR3/2 Organically enriched fine sand with few roots II 78 10YR2/2 Abundant whole and crushed Viviparus some bivalve and Pomacea in greasy fine sand IIa 30 10YR4/1 Abundant whole and crushed Viviparus ; some Pomacea and bivalve; in TU7 west profile only IIb 16 10YR4/2 Abundant crushed Viviparus ; occasional crushed bivalve; significantly less sand in matrix than Stratum II; localized lens in TU7 south profile. III 40 10YR4/2 Silty fine sand with some shell fragments and occasional fauna; appears intrusive in east, west, and south profiles in TU5; likely animal burrow IV 78 10YR4/1 Moderate amount of whole and crushed Viviparus bivalve and Pomacea in silty/loamy fine sandy matrix; significantly more sand in matrix than Stratum II IVa 58 10YR3/2 Moderate amount of crushed and whole Viviparus some bivalve and Pomacea in fine ashy sand; occasional concreted clasts IVb 68 10YR2/2 Moderate amount of crushed and whole Viviparus some bivalve and Pomacea in fine ashy sand; occasional concreted clasts IVc 78 10YR3/2 Moderate amount of crushed and whole Viviparus some bivalve and Pomacea in fine ashy sand; occasional concreted clasts; more silt than Strata IVa and IVb; present only in north profile. V 55 10YR4/2 Gritty fine sand, grit consistent with Stratum VIIII (concreted sand); appears mostly devoid of shell and fauna; likely disturbance VI 50 10YR3/1 Loamy fine sand; no shell; localized zone below Stratum I in north profile of TU6 VII 82 10YR5/2, 10YR5/3 Gritty fine sand; concreted and mottled; upper contact with Strata II and IV contains occasional shell fragments; increasingly concreted with depth VIII 82+ 10YR7/1, 10YR5/1, 10YR5/4 Concreted gritty fine sand, highly mottled; root casts present throughout; contact with overlying shell strata contains cemented shell

PAGE 283

283 Table 6-10. Cultural materials recovered from Test Unit 5, Node-5, 8VO215 Level Ceramic Sherds Lithic Marine Shell Bone Tool Vertebrate Fauna (g) St. Johns Crumb Flake Uniface Hafted Biface Frag. Tool A 85.6 B 2 1 1 281.7 C 5 1 1 2 519.8 D 20.3 D Zone A 5 2 3 1 462.6 D Zone B 26.3 E Zone A 2 1 222.2 E Zone B 26.8 E Zone C 96.6 F Zone A 1 116.6 F Zone C 16.9 G Zone A 45.4 Table 6-11. Cultural materials recovered from Test Unit 6, Node-5, 8VO215 Stratum Ceramic Sherds Marine Shell Fragment Vertebrate Fauna (g) St. Johns Plain St. Johns Crumb Orange Plain I 6 16 2 170.0 II 2 7 3 2 302.7 VII 2 4 1 42.3 Table 6-12. Cultural materials recovered from Test Unit 7, Node-5, 8VO215 Level Ceramic Sherds Lithic Marine Shell Vertebrate Fauna (g) St. Johns Plain Orange Incised Crumb Flake Hafted Biface Frag. Tool A 1 73.8 B 1 1 182.0 C 2 3* 1 272.4 D 8 428.9 E 1 1 2 148.7 F 140.2 G Zone A 1 1 82.8 G Zone B 34.2 *1 crumb Orange Plain, 2 St. Johns Plain

PAGE 284

284 Table 6-13. Comparison of the size, spacin g, and culture-historical associations of discrete shell deposit s at 8VO214 and 8VO215 8VO214 Shell Node 1 2 3 4 5 Minimum Elevation (m)a 6.17 6.87 6.80 6.65 6.86 6.67 Maximum Elevation (m) a 12.21 7.37 7.78 7.96 7.82 8.25 Thickness (m) b 7.34 0.40 0.60 0.50 0.20 1.00 Length (m) 155.00 14.00 25.00 38.00 46.00 50.00 Width (m) 60.00 8.00 18.00 30.00 22.00 41.00 Area (m2) 6607.00 90.00 342.00669.00 883.00 1594.00 Distance to Previous (m) d 50.00 27.00 50.00 57.00 45.00 Distance from Mound (m) 16.00 43.00 93.00 150.00 195.00 Culture-Historical Association e MT/O?/SJ?MT MT O/MT? O MT?/O/SJI/ SJII a measured surface elevation b based on test unit data c minimum estimate based on core and test unit data d distance to nearest northern node center e MT=Mount Taylor, O=Orange, SJ=St. Johns, SJI=St. Johns I, SJII=St. Johns II

PAGE 285

285 Figure 6-1. Hontoon Dead Creek Complex topographic base map.

PAGE 286

286 Figure 6-2. Hontoon Dead Creek Village topographic base map.

PAGE 287

287 Figure 6-3. Cross section profile of the Hontoon Dead Cr eek Village based on bucket auger testing.

PAGE 288

288 Figure 6-4. Distribution and density of she ll matrix at the Hontoon Dead Creek Village site.

PAGE 289

289 Figure 6-5. Shell node delimit ation at the Hontoon Dead Creek Village. A) presence of shell. B) minimum shell thickness. C) shell node boundaries.

PAGE 290

290 Figure 6-6. Topography and boundaries of N ode 1 at the Hontoon Dead Creek Village.

PAGE 291

291 Figure 6-7. Node 1 stratigraphy and compos ition. A) north profile. B) plan view of excavations. C) extracted concreted shell block.

PAGE 292

292 Figure 6-8. Topography and boundaries of N ode 2 at the Hontoon Dead Creek Village.

PAGE 293

293 Figure 6-9. Photograph of Node 2 excavation facing southwest. Pin flags denote the boundary of the shell node.

PAGE 294

294 Figure 6-10. Stratigraphy of Test Unit 2, Node 2, t he Hontoon Dead Creek Village. A) Profile drawing, and B) photograph.

PAGE 295

295 Figure 6-11. Objects recover ed during Node 2 testing. A) modified bone (possible awl). B) marine shell gouge/chisel. C) modified Olivella sp. shell. D) Busycon sp. battered apex fragment.

PAGE 296

296 Figure 6-12. Topography and boundaries of Node 3 at the Hontoon Dead Creek Village.

PAGE 297

297 Figure 6-13. Stratigraphy of Test Unit 1, Node 3, the Hontoon Dead Cr eek Village. A) profile dr awing. B) photograph.

PAGE 298

298 Figure 6-14. Objects recovered during Node 3 testing. A-C) modified bone (possible awls). D-E) hafted biface fragm ents. F) reconstructed Orange Incised/Engraved rim sherd. G) hafted biface.

PAGE 299

299 Figure 6-15. Topography and boundaries of Node 4 at the Hontoon Dead Creek Village.

PAGE 300

300 Figure 6-16. Stratigraphy of Test Unit 4, Node 4, the Hontoon Dead Cr eek Village. A) profile dr awing. B) photograph.

PAGE 301

301 Figure 6-17. Topography and boundaries of Node 5 at the Hontoon Dead Creek Village.

PAGE 302

302 Figure 6-18. Stratigraphy of Test Units 5, Node 5, the Hontoon Dead Creek Village. A) profile drawing. B) photograph.

PAGE 303

303 Figure 6-19. Southern profile of Test Unit 7, Node 5, t he Hontoon Dead Creek Village. A) profile drawing. B) photograph.

PAGE 304

304 Figure 6-20. Cross section of the Hontoon Dead Creek Complex. A) location of transects used to extract elevation profiles. B) elevatio n profiles of surface shell deposits, swamp su rface, muck surface, and wet site component. C) culture-historical periods represented in each of the shell ri dge and nodes.

PAGE 305

305 Figure 6-21. Paleohydrological context of the H ontoon Dead Creek Village.

PAGE 306

306 Figure 6-22. Areas of the Hontoon D ead Creek Complex submerged during flood events. A) minimum frequent high flood (recurring every 3 years). B) 50-year flood event.

PAGE 307

307 CHAPTER 7 DIVERGENT HISTORIES IN PLACE At the microscale, as the Hontoon D ead Creek Village revealed, Mount Taylor communities were reproduced through partitioned and discrete spatial practices. These material and practical referents im ply memories of place and person as communities returned to the St. Johns. The lack of si gnificant shell deposit ion atop nodes at the Hontoon Dead Creek Village enabled the discri mination of Mount Taylor traditions in microcosm. However, the sites final disposition makes it unlike most documented Mount Taylor places that are characteri zed by deep stratigraphic sequences composed of shell and earth. As discuss ed in Chapter 1, the reoccupation of places was taken in local orthodox thought as the gradual and unreflexive deposition of shellfish and other materials at ridges. To be sure similarities in site layout and depositional content at the largest and smallest scales imply that comm unities generated a sense f continuity in place through reoccupation. However, Mount Taylor inhabita tion occurred in the context of changing structural conditions: the river and associated ecologies were variable, the networks of peoples and places outside of the St. Johns were constantly in motion, and inhabitants within the St. J ohns were faced with an ever-emergent materiality of their own making as they deposit ed shellfish and other materials in place. We must seriously consider the possibility that in spite of continuity in site (re)occupation, communities manufactured alternative social memories through practical traditions to make new social and ecological contexts meaningful. In this chapter, I problemat ize the repeated occ upation and modification of places during the Mount Taylor period by considering two interrelated issues. The first regards the situatedness of Mount Taylor communi ties as they would have been understood

PAGE 308

308 within the larger flow of regional inhabitation. As Appadurai (1996:186) notes, processes of locality-making seem paradoxical because they both constitute and require contexts. As discussed in Chapter 2, practices and material referents in particular social contexts gather their significance in relation to other times, places, and ways of being (Barrett 1999, 2001; de Ce rteau 1984). Other times include the temporalities of experience within a plac e as well as the sedimented ancestral or mythical pasts referenced in their inhabita tion. As suggested by Barrett (1999:260), moving spatially between different frames of reference is equivalent to moving between alternative temporalities, from the biographies of co mmunities in the routinization of the day-to-day to the social rhythm s of institutional activity and the timeless continuities of cultural order. As argued in Chapter 2, these different tem poralities comprise economies of memory, in which peoples, places, and things are drawn together and histories are asserted or subv erted. These various practices can become doxic through reproduction, as commemorative economies ar e routinized, and may remain unreflected upon because they work within the structural c onditions of the time (Barrett 2001:154). However, it would be a mistake to r educe social reproduction to a passive process. As discussed by Weiner (1992:150) and Morphy (1995), social reproduction is not something that just happens. Amongst soci eties with a cold historicity, the paradox of reproduction is how a sense of cont inuity in place is maintained in terms of significance and meaning, while at the same ti me social relationships and the material conditions of life change. In this mode of hist orical consciousness, change is effectively denied by the assertion of the way it has always been (Pauketat 2001a). At atalhyk ,for example, the obsessive cleaning and wall painti ng within houses, as

PAGE 309

309 well as the emplacement of buildings on top of defunct stru ctures may reflect such an economy of memory (Hodder and Cessford 2004) However, Gillespie (2008:135) and Barrett have suggested that we pay attention to those moments when practioners stood apart from the world of their actions and looked in upon that world discursively (Barrett 2001:154). Such moments would be expected when the system of referents changed, and routine and rit ualized inhabitation as it had been known was no longer possible (Gosden 1994:18; Sahlins 1985:xiv). It is on these occasions that longreproduced schemata and refer ents could be revalued as they they were transposed into new social fields to acco mmodate or interfere with social or ecological variation. In the following sections I examine the evidence for the changing situatedness and temporalities of Mount Taylor places thr ough a consideration of biographies of shell ridges. These biographies are reconstructed by detailing sequences of depositional practices. Using the Hont oon Dead Creek Village as a m odel for the organization and reproduction of places, I trace how co mmunities established and transformed shell ridges as a process. I pay particular att ention to macro-stratigraphic units which represent changes in the overall intens ity and temporality of inhabitation. The following discussion is grouped around shell ridges that evince different kinds of temporalities. The first section consider s those ridges that were constructed primarily out of shell. The second considers how some places were transformed from residential contexts to mortuaries. The final sect ion examines those places that were characterized by residential social fields, but whose histories of transformation indicate that the traditions of ancestr al temporalities were extended to settlements as well. In

PAGE 310

310 each case, I will emphasize how communities appropriated material traces and practices and reconfigured them to produce alternative pasts in place. Renewing Community A consideration of chronologically early ridge biographies indicates they were rapidly conf igured as platforms within a few centuries of shellfish exploitation. Once emplaced, ridges were the locus of s hellfish deposition that renewed the mound surface. Based on the paleohydrological reconstruction at Hontoon Dead Creek Mound, these places were situated with refer ence to wetland localities and inhabitation there was enabled by local emba yments and shellfish colonies. However, communities mobilized places as historical resources that could be referenced in the course of largescale commemorative events. This pr ocess has been documented by Sassamans stratigraphic excavations at the Hont oon Dead Creek Mound (Sassaman 2005a) and Live Oak Mound (Sassaman 2003a). At Hontoon Dead Creek Mound, the testing program involved excavation of a 9 m-long, 1 m-wide trench on the west side of the ridge (Figure 7-1A). The upper trench was located approximately 2 m below the apex, and so the structure and culturehistorical associations of the most recent depositional episodes are unknown. At the lower end, th e basal stratigraphy was only observed in cores due to the water table. Because of the unconsolidated and loos e shell within the mound, excavation in the upper test units wa s terminated 2.5 m below surface for safety concerns. In total, a profile spanning 4.3 m of vertical relief and 9 m in horizontal extent was exposed. Four stages that differ in terms of temporality, s hell composition, and fragmentation were documented withi n the trench (Figure 7-1B). Stage 1 was observed only in cores extr acted from the lower basal units. Underlying thin shell matrix deposits was a dev eloped soil that dips gently down to the

PAGE 311

311 west, and is generally conformant with the sl ope of the terrace. This is the pre-mound surface upon which shell, vertebrate faunal bone, and charcoal was deposited. In its composition it is similar to the shell matrix recorded for N ode 1 and Node 2. This initial depositional unit is laterally extensive, and increas es in thickness to the east. It may be isomorphic with the saturated s hell matrix within the swamp (Sassaman 2005a). It may also be analogous to the diffuse downslope deposits identified in shell nodes to the south. No developed A horiz on was identified above this basal unit. There was little time between the cessation of small-scale deposition and the creati on of a large shell ridge and platform. Sometime prior to 7170 cal B.P. comm unities began amassing shellfish adjacent to the lagoon. Where inhabitants had once maintained vertically and horizontally discrete places they constructed a platform mound of shell. This process is manifest in the structure and composition of Stage 2 (F igure 7-1B), which rises 1.6 m above the preexisting shell. Based on the angle of the upper depositional unit of the Stage 2 ridge, the summit was relatively flat, although its apex was not exposed. Unlike Stage 1 which contained abundant vertebrate fauna and charcoal, stages 2 through 4 are characterized by low frequencies of non-shell materials. For exam ple, the vertebrate fauna is one-third to one-sixth the re lative abundance documented at residential contexts elsewhere. The implication is that a very restricted range of practices occurred here, and appear to have been limited to the pr ocessing and mounding of shell on the surface. The constituent stratigraphic units that differ in composition and fragmentation are: (1) whole Viviparus shell, and (2) whole and crushed shell lenses with observable sediment. The whole shell is indicative of rapid deposition, while the crushed shell and

PAGE 312

312 sediment are characteristic of surfac es that may have been exposed and walked upon over a period of time. Sometime after the establishment and enlarge ment of the Stage 2 shell platform, a large quantity of crushed shell interspers ed with fragments of concreted shell was deposited on the western margins of the platform. In Figure 71B this is Stage 3, which is stratigraphically superior to Stage 2. There are multiple lines of evidence that indicate this lens was placed here as an expansion of the preexisting mound. The constituent shell is finely crushed, implying a me chanical process for its formation. Large sometimes greater than 30 cm in diameterfragment s of concreted shell are situated throughout. As discussed before, concreted s hell forms either at the base of shell matrix sites through percolation of calcium ca rbonate, or in lenses composed of burned shell. In this case, the fragments are dis aggregated and of a differ ent composition than the underlying surface, suggesting they are allocthonous. T hey are not only distributed throughout the layer, but are found lying flat against the underlying mound. They were likely positioned to keep shell in place. Lacki ng a natural origin for shell and concretion in this quantity and configuration, it would seem that this entire macrostratigraphic unit was emplaced adjacent to and on top of the lower shell ri dge to expand the footprint of the mound. Where these materials came from is another question. Given that concreted shell was encountered in Node 1 at 8VO215, inhabitants may have mined preexisting shell matrix for construction materi al. This practice has been identified in mortuary contexts at Tomoka (Piatek 1994), Harris Creek (Aten 1999), and the Bluffton Burial Mound (Sears 1960). Having used picks and chisels to excavate concreted shell at other localities, I can attest that this would have been a labor-intensive activity.

PAGE 313

313 The expansion of the basal platform ma rgins established a wider summit upon which a new depositional regime was depl oyed (Figure 7-1B, 7-2). Based on a radiocarbon assay at the contact between the Stage 2 and Stage 4, this process occurred between 7170 cal B. P. As a macro unit, Stage 4 is characterized by alternating layers that vary in terms of s pecies composition, the application of fire, and fragmentation. One type of load was composed of whole gastropod shell, typically Viviparus but occasionally Pomacea These lenses are unburned and rarely fragmented. They have very little non-shell ma trix, making them susceptible to collapse when exposed in trenching. The alternat ive depositional unit is composed of burned and crushed bivalve frequently conc reted together. In this case, concretion is likely due to the burning process and the addition of ash to the matrix. Some of these burning events may have occurred on the summit, but the origins of much of the burned shell is unknown. The two depositional loads are not emplaced haphazardly, but occur in pairs or couplets (Figure 7-2B). Whether the burned or whole shell was deposited first is unknown, and not self evident from the documented stratigraphy. Covering the Stage 2 ridge is an extensive lens of burned, concreted, and crushed Viviparus and bivalve shell. This layer was emplaced above whole, unburned, and very loose shell. If the whole shell was the initiatory even t, then inhabitants would have accumulated and emplaced whole shell atop the mound summit. Following this depositional event, bivalve was burned and spread ov er the whole shell. In addition to the act of burning and covering the platform, t he bivalve likely stabilized the surface. Based on stratigraphy alone this sequence of couplet deposition cannot be untangled. Over time, however, this stru cture of whole shell and bur ned shell is repeated at least

PAGE 314

314 eight times within the trench. The short-term temporality of deposit ion is unknown. The lack of sand or humic development on any surfaces suggests any abandonments were of short duration. In the longer term, however, the summit of the mound was constructed rapidly. A radiocarbon assay on the fifth burned shell lens was statistically contemporaneous with the first burned shel l event, and an assay from the seventh burned shell layer falls just outside of the 2 range of the lower event. Hontoon Dead Creek Mound was not alone. Ju st a few kilometers away lies the similarly configured Live Oak Mound, which was the subject of stratigraphic testing by Sassaman (2003a). Unlike Hontoon Dead Creek, the Live Oak Mound has been significantly damaged by looting, likely due to the presence of St. Johns interments with grave inclusions at its ape x. Sassaman targeted looter pits, which provided an opportunity to examine basal de posits within the core of t he mound. Two test units placed 90 m apart yielded similar basal profiles. In both cases low density shell was emplaced upon the underlying terrace. A nearbasal radiocarbon assay was obtained from one of these lenses and returned an age estimate between 7300 cal B.P., coeval with Node 1 at 8V O215 and predating the Stage 2 platform at Hontoon Dead Creek Mound. The basal sequence in both test units at Live Oak were characterized by lenses of whole shell alternating with crushed, burned, and charcoal-rich layers. Unlike Hontoon Dead Creek, there wa s not enough stratigraphic visibility to discern an analogous platform towards the base. Thus, it is unknown if the burned-shell events were laterally extensive. The sequence of stages at Hontoon Dead Creek and Live Oak mounds demonstrate that depositional events at some ridges were not small in scale (relative to

PAGE 315

315 the site). Moreover, they o ccurred over a relatively shor t time span (relative to the Mount Taylor period). After being establishe d as a residence ca. 7300 cal B.P., a place of prior inhabitation was rapidly transformed into a platform through the bulk loading of shell. This platform was then laterally fortif ied with mined matrix to expand the summit. At Hontoon Dead Creek Mound, the time between the first occupation and its transformation into the 1.6 m high Stage-2 ridge has not been established. The base of Node 1 adjacent to the mound, the first burned shell stratum, and the fifth burned shell stratum all fall within a 2 range, although the Node 1 medi an probability intercept is earlier. While the Stage 2 construction was rapid, involving mostly loads of whole and unburned shell, it was replaced with a new tradi tion of couplet deposition. Unlike the lower stage, these new depositional episodes were scattered across the surface of the mound. With the details of early mound construction at hand, we can consider how the deposition of shellfish at early she ll ridges framed the ongoing production of communities. From a practical standpoint, t he net effect of shell deposition was the accretion of the mound summit. One might be tempted to argue that increasing the height of the mound, say for competitive or prestige-gaining purpose s, was the ultimate goal of deposition. However, such an ar gument would place undue weight on the final configuration of the place, and negate the importance of the actual history of deposition. While the mound may have been constructed rapi dly with respect to the Mount Taylor period, it nonetheless emerged th rough the labor of several generations. Instead of the final form being the intent of deposition, I w ould suggest that the prac tices of collection,

PAGE 316

316 processing, and deposition of shell atop the mound were acts of renewal in which communities negotiated and reasserted social memories in place. As Pauketat (2000:118) has noted with re spect to Mississippian communities, the staged construction of earthen mounds follo wed a variety of tempos and scales. Major construction episodes at mounds likely signal the death or ascendance or particular persons, or other moments of variable or irregular timing. Yet mounds and other spaces provide evidence for regular construction cycles, perhaps even annual in some cases. Frequent communal depositio nal practices include the addition of sediment mantles to mounds or the reconstruction of public buildings. In these cases, Pauketat argues that communities came toge ther to ritually pur ify places prior to construction or to renew places in accor dance with a ritual cycle. To be sure, the specific meaning of these Mississippian acts of renewal, or their associated political economy, have no relationship to Mount Taylor practices. Where these cases are similar, however, is that the act of renewi ng mounds provided a context in which diverse participants could (re)create regional community i dentity. It is thro ugh such collective practices that orthodox systems of m eaning could be constructed and reproduced (following Pauketat 2000:122). Indeed, one possible temporal analog is provided by Morrison (2003) in his discussion of shell mound construction along Au stralias Cape York Peninsula. This coastal region contains a number of small and large (over 13 m high) mounds composed of Andara shellfish remains, a marine species. Like shell mounds of the St. Johns they have typically been interpreted in terms of function. In par ticular, it has been argued that mounds would have provided dr y ground during the wet season, although

PAGE 317

317 many of the mounds are situated on dry l and or well above annual flood ranges. Internally the shell mounds are composed of thick lenses of Andara shell, separated by lenses of crushed shell and organic matter suggestive of periods of abandonment. Andara can be found in very large quantities but the distribution varies annually. Morrison suggests that local abundance of the species enabled la rge gatherings in which marriages and various ceremonies co uld be held. In the Cape York case, shellfish were consumed as part of feasti ng, and the deposition of shell on mounds was encompassed within a larger tr adition that incorporated diverse communities across the country. Morrison argues that mounds and the practices occurring at them likely provided a framework for the politics of lineage alliance-building. Today, the mounds have obtained significance wit hin larger cosmogonic narratives on the landscape. Although shellfishing is not a component of contemporary community lifeways, inhabitants have actively incorporated mounds as ancestral traces into spatial stories (Hiscock and Faulkner 2006). With specific reference to the Mount Ta ylor period, we have seen in microcosm that communities renewed residential spaces through the deposition of shellfish and other materials in residential middens. These small-scale practices provide an analogue for the routine resurfacing of she ll ridges, which were conducted at a much larger scale and likely with more participants. If it is the case that multiple communities were responsible for the deposition of couplets at Live Oak and the Hontoon Dead Creek mound, then resurfacing shell ridg es may have been a central component to social commemoration and integration as communities returned to the St. Johns during annual rounds. In this early context, the depositional practices at work along the St.

PAGE 318

318 Johns appear to have been transposed from resi dential contexts and elevated to large scale commemorative ceremonies aimed at renewing mound surfaces, and by extension, the social relationships involved in mound maintenance. Further research will be necessary to deta il the tempo of these practices (e.g., Blessing 2009), although a leadi ng hypothesis is that rid ge-top deposition coincided at times of periodic resource abundance. Thes e ridges emerged at a time of greatest ecological instability and social diversity, as documented in Chapter 3. In contrast to the regional variability, however, ridges provide a pict ure of stability. In this context, ridges may have been renewed as acts to interv ene with change on the broader horizon of social experience, and to create orthodox sys tems of reference in an otherwise diverse and changing world. OyuelaCaycedo (2004) has described a process in which dances are performed during periods of resource abundance. These dances take the form of mythical narratives using ecological metaphors that bring outsiders into the local community as process of difference recogni tion and subversion. If a similar process was afoot along the St. Johns at this time, periodic collapses and abundance of symbolic and subsistence resources would have provided a powerful political economy that at a minimum would have sedimented communal identities throughout the basin. Ancestral Temporalities Many of the depositional practices at early shell ridges reference an economy of memory oriented towards the living. The preparation and potent ial consumption of shellfis h, and the deposition of shell atop sh ell ridges provided an arena for gatherings that referenced the physical arrangement of communities. The temporality of these events was no doubt tied to local availability of shellfish. As suggested previously, however, depositional practices may have emphasized the renewal of peoples and

PAGE 319

319 places as an ongoing process. At both Live Oak and H ontoon Dead Creek, no evidence for mortuary practices has been observed in basal strata, although sub-mound coverage is admittedly limited. Other places, however, demonstrate that the commemoration of death and community integr ation was yet another rhythm interwoven through landscape inhabitation. Wyman and Moore noted the pres ence of inhumations at the base of shell ridges. For example, at the eroded Crows Bluff (Osceola) in the Middle St. Johns, Wyman (1875:33) documented two mingled skeletons at the base of t he shell ridge within consolidated shells and mud (Figure 73). These were si tuated beneath ca. 2.4 meters shell. Wyman described massive, whole Viviparus layers alternating with thin bivalve/ Pomacea lenses. His description suggests that a mortuary was eventually transformed into a platform fo r shellfish deposition. A si milar sequence was identified by Moore (1893b:622) at the base of Orange Mound in the U pper St. Johns. The most vivid details of early Mount Ta ylor mortuary practice come to us from the Harris Creek mortuary. A consideration of the construction history of this place suggests that communities appropriated a preexisting resident ial locality and transformed it into a mortuary mound. Like the ridges just discussed, the depositional emphasis was directed towards referencing the physical arrangement of communities. At mortuary mounds, however, shellfish and sand deposi tion was transposed and revalued across social fields of practice to enable the co mmemoration of the dead and the integration of the living through place. As described in Chapter 5, the Harris Creek mortuary was part of a U-shaped complex of ridges and shell fields that covere d five acres prior to its destruction.

PAGE 320

320 Salvage excavations were conducted by Jahn and Bullen (1978), who documented a Mount Taylor mortuary at t he base of shell ridge (Figure 5-10B, 7-4). Because of extensive mining, they were only able to excavate the northern and southwestern portions of the basal mortuary. However, the mining also left a long escarpment which cross-cut the mortuary (Figure 7-4, 7-5A). Jahn and Bullen we re unable to accept that a prepared mortuary mound could be Archaic in age, even though they recovered Mount Taylor objects in association with the inhum ations. However, Aten (1999) has recently reconstructed the history of the mortuary based on Jahn and Bu llens field records. He demonstrated that it is Mount Taylor in age, and was built in several stages. My discussion here is based on his reconstruction and observations. The foundation of the mortuar y (Layer 1) was saturated and not excavated. It appears to have been between 1 and 1.5-m thick, and composed of heterogeneous shell matrix. Because it was not consistent ly observed it is unknown if the surface was flat or node-like. After a period of indet erminate abandonment, inhabitants returned and constructed a low shell ridge (L ayer 2) (Figure 7-5). Th e base was composed of shell and brown sand matrices (Layer 2a), but the central core of the mound was constructed of clean shell, likely all Viviparus (Layer 2b). These materi als were emplaced on top of the preexisting shell floor. Collectively, these two depositional matrices comprise Layer 2. In plan view this structure is a linear ridge, measuring approxim ately 1-m high, 15-m wide and 30-m long (Figure 7-5, 7-6B). At en suggests that these materials may have been emplaced to accentuate a pr eexisting ridge in this location, but that could not be verified from the field records. In both structure and shape Layer 2 is similar to the lowlying ridge identified at the Hontoon Dead Creek Village site (Figure 7-6A). It is

PAGE 321

321 unknown what the temporal relationship of the Layer 2 event was to either earlier habitation or later monumentality, but these actions do not appear to be associated directly with mortuary activities. The implic ation is that the mortuary was literally and figuratively founded on a preexis ting habitation space. The Layer 2 surface was then used as a foundation for the construction of two successive mortuaries, into which over 150 i ndividuals were interred (Figure 7-6C,D). As discussed in Chapter 4, there is some uncer tainty over the precise chronology of the mortuary. If Tuckers (2009) assays are co rrect, then the Harris Creek Mortuary is coeval with events at the Hontoon Dead Creek Mound and Live Oak. If not, then the mortuary likely dated somewhere between 7000 and 6000 cal B.P. Regardless, the tightly clustered radiocarbon assays on hum an bone and charcoal suggest that this mortuary was short lived and active for a period of a few centuries at most. The lower mortuary (A) corresponds with Layer 3, and lie s unconformably on Layer 2. Mortuary A is characterized by allocthonous white s and deposits interspersed with clean shell and dark soil (Figure 7-5A). The geologic origin of the white sand is currently unknown, but it had to be brought in from outside the boundaries of the mortuary. The depositional structure of inhumations in Mortuary A were of two types. In many cases, tightly flexed single and multiple interments were placed into piles of sand less than a meter high and several meters wide. Alternatively, pits were dug into the basal Layer 2, and white sand was used as bur ial pit fill. Eventually this mortuary covered the entire basal sub-mound. A series of superimposed post holes were located near the burial area, which may reflect the repeated erection of a structure. Jahn and Bullen argued that the posts ma y be related to a charnel house. A lens of black soil

PAGE 322

322 adjacent to the post holes was also identif ied, and has been referred to as the Black Zone (Figure 7-5B). Tak en together, the post holes and organically enriched sediment suggest that suggest that at least some of the deceased were processed at the mortuary for some time prior to their inhumation. Regardl ess of the function, this feature remains the only known non-shell architecture of Mount Taylor age in the Middle St. Johns. Above this layer was Mortuary B (Layer 4 [Figure 7-5A, 7-6D]). Like Mortuary A, deposition was oriented towards accentuating the underlying ridge, and maintaining a stable surface. Practices continued to emphasize the original st ructure of the mound, but did not appear to include white sand as a burial medium. Instead, the burial matrix and mound fill was composed of brown sa nd and shell. One deposit with abundant vertebrate fauna was also noted, which Aten argues may register feasting associated with mortuary activities. The possible charnel house or structure that was emplaced on top of the mound in Mortuary A was no lon ger reproduced, and was covered over with the Layer 4 sand and shell. Aten notes that the transition to this new depositional mode was accompanied by a reduction in grave inclusion diversity as well. This process of decreasing diversity may regist er new categories of person which were still referenced with respect to the past in place. Communities ceased routinely interring t he dead atop the mortuary after a century or two. Although occasional burials were noted by Jahn and Bullen emanating from these upper layers, most activities at the mound in later periods do not appear to have been mortuary related. This is not to say t hat ancestral presence was not recognized in this place, but only that burials were not routinely emplaced her e. Based on the

PAGE 323

323 composition and distribution of sand and shell layers, Aten argues that efforts were oriented towards increasing the width of t he summit platform, typically by emplacing shell and other materials on the east and west margins of the mortuary (Figure 7-6E). At least one layer (8) involved the excavati on of existing Layers 36. Some of the excavated fill may have been emplaced on the su mmit. However, massive lenses of whole, clean shell attest to shell mounding here as well. Otherwise, many circular patches of shell and vertebrat e faunal remains atte st to small-scale primary deposition on the mound surface. The Mount Taylor era ridge was eventually capped with organic muck at the onset of the Orange period (Layer 9). The mound platform continued to receive the occasional shell and sediment load throughout the subsequent millennia, likely up until the co lonial period. The sequence and structure of construction at Harris Creek demonstrates that Mount Taylor communities replicated a ridge conformant with residential spaces elsewhere. Practices within the mortuary may have been similarly partitioned. Burials were restricted to the northern and eastern aspec t of the ridge. Be cause of the recent extensive shell mining, it is unknown whet her burials were present along the western edge, while anecdotal evidence indicates many bur ials were present in the core of the mound. Aten argues that because no burials were observed in the dragline pond, that it is likely they were spatially segregated. I would note that the easte rn side of the ridge today faces Lake Woodruff, and a similar body of water was likely present there as well. In contrast, the western aspect faces the te rrace. As document ed at the Hontoon Dead Creek Village, the western/terrace side of shell nodes appears to have been where activities not involving the deposition of s hellfish were occurring. It is tempting to

PAGE 324

324 suggest that in the case of the Harris Creek Mortuary, the microc osm of the residential shell node was also a model for the living a nd the dead. That is, the mortuary mound was effectively a model of community structure. At the scale of individual bur ials, there is considerable di versity in the matrices and objects included. Some individuals were placed in pits and then covered in sand, while others were placed on the mound surface a nd covered with sand or shell. Some interments involved fire, wherein a fire was st arted within a burial pit or on top of the surface. The deceased (single or multiple individuals) were empl aced on the hot coals, and then covered in white sand. These are not true cremations, but demonstrate fire was at times integral to the construction of memory surrounding t he transformation of the deceased in place. This practice has ana logs to the burning of bivalve shell and its emplacement above the summit at the Hontoon Dead Creek Mount. Although there was significant diversity in the specific materials and methods of inhumation at Harris Creek, interments were structured along temporal axes that suggest distinctions in commemorative events we re made based on origins. It is worth restating an observation I made in Chapter 4. As discussed by Aten, burials were of two types. The majority (60 per cent) were vertical and tight ly flexed, suggesting bodies were processed for an extended period of time prior to inhu mation. In contrast, another 38 per cent were loosely flexed, suggesting that the bodies were emplaced within the mound soon after death. As demonstrated by Tucker (2009), di fferent modes of interment are closely associated with either loca l or non-local points of origin. So-called delayed or flexed burials tend to be a ssociated with individuals who spent their childhood years in the vicinity of the mound, while immediate or primary burials are

PAGE 325

325 associated with individuals who spent their childhood years away from the St. Johns. The implication is that the timing of burials, and by extension, the co-presencing of living individuals, was central to the ordering, ju xtaposition, and subversion of regional differences. Delayed burials would have provided the opportunity for multiple gatherings, likely drawing in kin and allied gr oups from throughout the basin, possibly in the context of recently deceased. Early Mount Taylor mortuaries appear to be open and integrating affairs that incorporated inhabitants from throughout Peninsular Florida. The materials of shell and earth were transformative media by whic h biographies of the deceased and the living could be worked out and sedimented in place. Based on the temporality and structure of interments and depositional episodes, t hese mortuaries were maintained over generational time. Whether similar mortuar ies were maintained throughout the Mount Taylor period is not clear. Late mortuary mounds such as Thornhill Lake contained a smaller number of individuals, but this lower interm ent frequency may reflect the absolute amount of time in which the mortuary was active. However, there is evidence that at least some later mortuaries were cons tructed for individuals or objects during the later Mount Taylor period. A consideration of the Bluffton Burial Mound demonstrates that mortuary mound building events fixed places on the landsca pe through new traditi ons of mortuary practice. The Bluffton Burial Mound (8 VO23) is one of two paired mortuary mounds situated behind the Bluffton dome and ridge comp lex (Chapter 5). Nothing is known of the other mound, other than it appears to have been constructed of sand. The details of the construction of 8VO23 are provided by Wyman (1875:37) Moore (1894a:44),

PAGE 326

326 and Sears (1960). There are no published radiocarbon assa ys for this mortuary. However, no pottery was found within the co re, and the mound is similar in composition and structure to other late Mount Taylor mortuaries else where, such as Tomoka and Thornhill Lake (Endonino 2008; Mitchem 1999; Piatek 1994; Wheeler et al. 2000). The mound is described by Moore as a s lightly elongated dome with a flat summit, measuring 4.2-m high and 30-m wide. This general configuration was mapped by Sears (1960) as part of his investigations there (Figure 7-7C). The topography in 1959 reflects the destruction of the southern marg in by the excavation s of Moore and other parties. Moore (1894b) placed an 11.5-m l ong trench in the eastern aspect of the mound (Figure 7-7C), and documented a well stratified construction sequence. The upper mantle of the mound had been disturbed by previous excavators, including Wyman, and so he could not prof ile the extent of layers ac ross his excavation unit. He noted that the base of the mound was a shell ma trix deposit. At the center of the base was a shell ridge, roughly 3-m wide and 1-m high. This shell ridge (in his terms) shares a similar size and shape to Node 2 at the H ontoon Dead Creek Village. Above this was a 1-m thick layer of sand with some shell mix ed in, with successive layers of shell, black organic muck, sand, and brown sand with shell at the surface. Moor e did not encounter burials within the basal component, but did identify several intrusive burials. Pottery was discovered only on the surface. Sears (1960) made a more complete inve stigation of the mound as the complex was being mined for shell in 1959 (Figure 77B). His results mostly replicated and expanded upon Moores work. Un like Moore, Sears did not encounter a central shell ridge underlying the mound, but did identify an individual interred on top of the

PAGE 327

327 preexisting shell floor. Sears suggested the followin g sequence of construction. First, a fire was started on top of the existing shell, resulting in the thermal alteration of the underlying shell floor. The individual was pl aced on the old shell midden, and covered with an organic material that may have been allig ator feces. Thereafter, a series of layers were added atop the mortuary. First, basket loads of sand were emplaced above the individual, followed by alternating s and, shell midden, sand, a layer of black gumbo or organic sediment, sand, and then a lar ge deposit of shell again. Prior to this last deposit another individual was emplaced on the edge of the mound. Sears argues that the layers were emplaced rapidly, as there was no internal evidence for erosion. Finally, several individuals of the later St. Johns period were interred within surface deposits. These intrusive grave pits were also noted by Wyman in his superficial examinations there. A review of Mount Taylor mortuary places suggests several commonalities and divergences as regards temporalities and lar ger economies. In both cases, the past as a preexisting settlement node wa s appropriated as a model for ancestral transformation. It is unknown if the basal shell nodes were tr uly preexisting on the landscape, or if they were recreated for the act of interment. Bradley (1998:46; 2002:Chap 2) has discussed how Linearbandkeramic communities in Eu rope constructed long houses which may have been referenced in the architectural principl es of long barrow mo rtuaries. In this case, abandoned houses and the biographies of t heir communities we re memorialized through the construction of permanent hous es for the dead. Lacking radiocarbon assays from basal, pre-mortuary layers at th ese places along the St. Johns, we cannot currently decipher the chronology of mortua ry foundation and subsequent interment. At

PAGE 328

328 Thornhill Lake, two mortuary mounds were cons tructed directly on top of a preexisting shell ridge, lending credence to the idea t hat these basal nodes were already present on the landscape. Regardless, communiti es appropriated and transformed the past in place by bringing together sand, shell, and other materials drawn from the surrounding landscape to emplace the dead within th is new frame of reference. However, a consideration of the tem porality of commemorating the dead, and the life histories of the mo rtuaries themselves, indicates diffe rent kinds of social memories may have been constructed in these places. At Harris Creek, the mortuaries were experienced for extended perio ds of time. Although memories of the deceased may have been completed during t he extended mortuary treatm ent, the spaces created through interments maintained an active ancestral presence. After Harris Creek was discontinued as a mortuary, it was reconfigured as a locus of large-scale shellfish deposition and remained within the spatial stories of Mount Taylor communities. In contrast, the construction of the Bluffton Burial Mound was an event. The mortuary constructed rapidly and for a single per son. More importantly, however, the emplacement of sand, shell, and muck reca pitulated the sequence of the Harris Creek mortuary in terms of deposition and the materials used. In the case of Bluffton, however, all layers were emplaced at the same time. The specific sequence of this depositional narrative may have recreated a cosmogonic myth whose specific meaning is unknown to us today. I would point out again that the Harris Creek mortuary wa s actually excavated by Mount Taylor inhabitants during the construction of Layer 8. This moment may have afforded inhabitants an opportunity to see the sequence of construction there. Whether

PAGE 329

329 or not this sequence was translated into the mortuary at Bluffton is unknown, but its an idea worthy of testing through chronological correlation. Regardless, Bluffton was not only built as an event, but the memories and meaning inherent in its construction were apparently restricted, such that limited modification could occur there. While an individual was interred on top of the mound (in addition to the founding burial) prior to the completion of the mound, no further individuals were interred during the Archaic period. This restriction no doubt reorganiz ed the system of references that framed community commemorative practices in that pl ace. Such finishing work may have also created a new spatial and histor ical distinctions between ancestral or mythic time (Gosden and Lock 1998; Rumsey 1994). Moreov er, the new temporality surrounding death would have framed new ways in which inhabitants within and outside of the region could be dr awn together. The Life and Death of Settlements Through depositional acts involving she llfish and sand, I argue that the deceased were transformed into ancestors in some places. In this sense, a structural link was established between the communal performanc e of venerating the dead and the act of capping or emplacing materials on places of everyday practice. These acts were truly monumental, in that they revalued places with new dominant narratives and restricted the range of practices that could occu r on them. These ritualized modes of transformation, however, were not restricted to monumental places. As illuminated in the biographies of two shell ridges, Silver Glen Run-Locus A and Hontoon Island North, the act of resettling or abandoning a locality r equired similar reconfigurations. That is, new temporalities and domains of practica l experience were po ssible only when the

PAGE 330

330 underlying system of referents was transform ed. These transformations were enabled by the transposition of depositional practi ces between mortuary and residential fields. Locus A is a preceramic Archaic single ri dge that was mined for shell, presumably with the rest of the 8LA1 complex, starting in 1923. The curr ent surface configuration of the mound reflects the mining operations (F igure 7-8). There ar e no known historic descriptions of the ridge, and it remained undocumented until the 2007 season of the University of Florida St. Johns Archaeologi cal Field School (Sassaman et al. 2010). Today the ridge is evident topographically as a tear-drop or elliptical feature 200-m long and 75-m wide. The ridge is ori ented parallel with the spring run, and is separated from the run by a hydric swamp to the north. A relict seep spri ng frames the eastern border. It is unknown if there are we tsite deposits along either margin, but they would not be unexpected if found. The most defining feature of the ridge is an intact escarp ment that at its highest point is approximately 4 meters above the floodplain along its northern border. The mounds height prior to mining is unknown, although a total of 3 m of intact deposits remain (1.5 m above and below the current interior escarpment surface). Based on the relict slope, and the fact that the ridge was not identified in the ninet eenth century, the ridge may have only risen a meter above its current surface (Figure 7-8). Although destructive, shell mining had the unintended effect of exposing basal deposits that would otherwise be inaccessible. Stratigraphic te sting in three loci (Figure 7-8) involved cleaning exposed profiles and testing the subsurface extent of intact deposits along this interior mining escarpment. Test Unit s 5 and 8 were emplaced on the highest relict escarpment, while Test Units 9, 10, and 15 (hereafter the trenc h) were positioned

PAGE 331

331 perpendicular to the ridges longitudinal axis. Test Unit 6 was situated on a lower swamp-facing margin, and will only be discussed in brief below. The ridge underwent at least two transfo rmations over its documentable lifetime, and these are expressed in all stratigraphic te sts (Figure 7-9, 7-10, 7-11). Locus A was established as a place to dwell through the depos ition of thin layers of shell across a ca. 90-m long segment of the landform. For exam ple, the basal 80 cm of TU5 to the east contained anywhere from three, four, or five occupational episodes, while at least two were evident at the base of TU6 approx imately 90 m to the west. These are characterized by highly fragmented bivalve s hell, charcoal, ash, and other constituent shell fragments. The basal unit in TU5 is covered by a st erile sand that may exhibit some soil development, suggesting a period of abandonment in this locus. It is unknown what kinds of activities were o ccurring when the settlement was established, but they apparently involved the processing of mussels and the cleaning out of hearths. This last inference is based on the abundant ash and semi-concreted nature of the strata. An assay of 6190 cal B.P. wa s obtained from TU5 in near basal strata (Figure 7-9). Unlike the eastern and western aspects of the ridge, these surfaces were not documented at the base of the trench. Instead the basal deposits here are characterized by concreted shell filled pits (Figure 7-10). These may have been related to early occupation, but may have also truncated the earliest depositional units. Elsewhere, basal surfaces were exposed for extended periods of time. This inference is based on the extremely fragmented nature of the shellfish. In addition, root casts indicative of post-abandonment vegetation growth were identified at the base of TU6, while incipient soil development between crushed shell surfaces at the base of TU5

PAGE 332

332 attest to a period of non-use. Collectively, these data indica te that initial occupation was sporadic but routinized. Sometime between 6300 and 5940 cal B.P. inhabitants emplac ed a mantle of tan/brown sand across the entire ridge. This transformation is visible in all test units, although is most evident in the trench and TU6. Roughly 10 to 20 cm of tan brown sand were emplaced on top of the pree xisting surface. It is uncl ear if this cap was truly a homogeneous layer across the landf orm, or was deposited in smaller scale depositional events. Much of this sand cap was subs equently altered by extensive burning, as indicated by thermally-altered or oxidized sand in the trench and lenses of charcoal impregnated sand in TU5. Numerous overla pping pits were identified in the TU6 and TU5 surface of the sand mantle, often associated with shellfish remains. It would seem then that this transforma tive event was coincident with the mass processing of shellfish and other materials in place. That these capping events across the ridge were coeval is suggested by statis tically overlapping radiocarbon assays from sub-sand lenses at TU5 (6190 cal B.P.) and a sample from just superior to the sand in the trench (6290 ca l B.P.). Once this cap had been emplaced upon the ground surface, a layer of crushed shell devel oped above it. How the shell was crushed is unknown, although post-deposit ional trampling by the si tes inhabitants is one possibility. Alternatively, the shell may have been crushed pr ior to deposition. Unfortunately, it is not possible to articula te the order of events here. While the sand may have been emplaced as a capping event, af ter which time the site was abandoned, the sand may have also served as an initiatory cleansing of the surface. Regardless, the emplacement of sand reconfigured the surficial re ferents of practice.

PAGE 333

333 Given the preexisting link between sand, death, and community renewal it is not out of the realm of possibility that this stratum marks the death of the settlement, while at the same time opened the space up for new traditions. Once the sand was emplaced it apparently provided an arena for intensive habitation activities. Although the viewshed is limited, data from the trench and TU 5/8 suggest that a seri es of shell nodes were emplaced across the ridge surface, and t hat they were routi nely reproduced over the succeeding centuries. For example, the trench above the sand mantle is characterized by two different, but s uperimposed stratigraphic sequences. The southern (to the left in Figure 7-10A) is char acterized by successive lenses of whole shell, interspersed with crushed shell layers. In contrast, the northern (to the right in Figure 7-10A) unit is characterized by more homogenized sand and shell. Unfortunately, there is a large disturbance in the middle of the trench, likely a treethrow, and so the specific stratigraphic relationship between the two is unknown. However, based on the Hontoon Dead Creek Village model, I would argue that earlier living surfaces were replaced by depositi onal practices that effectively created new micro-mounds on the surface. The recreati on of shell nodes is best seen in the TU5/8 sequence (Figure 7-9, 7-11). Overlying the sand mantle in this unit are a series of depositional episodes of shell and sand, interspersed by surfaces that contained burned and fragmented crushed shell, in addition to ash. The slope and inflection of these deposits is reminiscent of the household cluste rs identified at the Hontoon Dead Creek Village. At Silver Glen, once the lower shell nodes were created, their form was maintained through time as new loads of s hell were deposited on the surface, and then covered with burned materials.

PAGE 334

334 It is important to note, however, that i nhabitation here was not continuous, in the sense that the ridge underwent a pparently numerous abandonments and reoccupations. For example, at least one of the lower shell units in TU5/8 contains large root casts, indicative of long-te rm abandonment. Whether th e entire settlement was abandoned is not known. However, a similar sequence of abandonment and surface renewal is evident in the southern end of the trench (Figure 7-10). This area is characterized by abundant whole, uncrushed she ll, which is superfici ally similar to the depositional couplets at the H ontoon Dead Creek Mound. Also similar are interstitial lenses of crushed shell and charcoal that appear to represent burning events separating periods of deposition. Yet unlike the Hontoon Dead Creek Mound, soil is present between the whole shell, and t here are root casts emanating from surfaces throughout the sequence. The development of such layers has been i dentified in shell mounds in Australia, where Beaton (1985) has suggested they r epresent periods of abandonment (see also Morrison [2003]). The implicat ion is that the ridge at Silv er Glen Run was framed by a fundamentally different temporality of experience than the earlier ridges composed primarily of shellfish remains. In this place, communities routinely reproduced the spatiality of shell nodes, but they did so over extended and interrupted sequences of time. Taken together, the stru cture and composition of deposit ional histories at Silver Glen Run appear to represent new temporalities in daily practice, which may indicate an increasing occupational durati on at these places. At the same time, however, the presence of root casts indicates that even these established places could be abandoned and reestablished over extended durations with f eelings of continuity.

PAGE 335

335 When the Silver Glen ridge was ultima tely abandoned is unknown, because the apex was destroyed. During testing the occa sional Orange or St. Johns sherd was encountered, but only in disturbed contexts Given the widespread mining across the landform, these may have come from another locality. A brief consideration of the biography of the Hontoon Island North site indicates that the abandonment of some residential localit ies may have been enabled by the capping and obscuring of prior material referents as well. As detailed in Chapter 5, the Hontoon Island North site is a multi-mound complex located on the northern boundary of the Hontoon Island. The site was mined for shell in 1935, leaving only an escarpment and basal componen ts from the southern ridge (Figure 7-12A). Due to historical circumstances surrounding t he location and scale of archaeological investigations, Hontoon Island North has been considered a St. Johns period construction (e.g. Purdy 1987). As docum ented by Sassaman et al. (2005), however, much of the southern ridge was laid down by Mount Taylor communities. Like the ridge at Locus A, shell mining removed and distur bed a large amount of material, but also provided an unrivaled opportunity to examine basal deposits that would have typically been obscured beneath meters of later deposition. Excavations in four places by the Univ ersity of Florida St. Johns Archaeological field school targeted the interior of the surviving deposits and demonstrated that upwards of 2 m of intact shell matrices dating to the preceramic Archaic remain (Sassaman et al. 2005). We argued that t he different stratigraphic sequences revealed throughout suggest that a settlement composed of spatially segregated zones, including a mortuary, was capped off with shellf ish upon its abandonment. That the ridge

PAGE 336

336 included a number of distinct areas is revealed through a consideration first of the easternmost test units. For example, the bas e of Test Unit 1 was characterized by a homogenous shell matrix the contained abu ndant vertebrate fauna, paleofeces, and bone and stone tools. This deposit would appear to be true secondary midden, deposited outside of activity ar eas. It may be that this co mponent is analogous to the water facing slope of residential units at Silver Glen and the Hontoon Dead Creek Village. In contrast, the base of Test Unit 3 to t he west was composed of alternating lenses of whole apple snail, Viviparus and bivalve shell that were separated by crushed shell lenses. As seen in profile (Figure 7-12C) th ese lenses share formal properties with the trench excavations at Silver Glen as well, suggesting a similar routinized rhythm to surface creation and renewal. Radiocarb on assays from the base and top of each sequence within Test Units 1 and 3 demonstrate that they are contemporaneous, and that deposition was rapid. All as says statistically overlap at 2 Elsewhere, in TU2, excavations intercepted a partially articulat ed inhumation. This burial was within a brown sand and shell matrix consistent with the Harris Creek Layer B mortuary. Shovel testing and augering in the vicinity of the unit isolated sand/shell deposits within the vicinity of the burial. Abiding with legal protection for burials, all excavation in the area was stopped, and so it is unknown if more burials are present. However, given the sand, human remains, and even concreted shell in the area, it is likely that a mortuary was emplaced here. Lacking radiocarbon assays from the mortuar y, it is unknown whether it is contemporaneous with the residential area to the east. Regardless, stratigraphic

PAGE 337

337 evidence from nearby Test Unit 4, in additi on to the upper strata in Test Units 1 and 3 suggests that some time after being establ ished as a locus of se ttlement, the entire shell ridge was capped with shell. This shell cap is best seen in Test Unit 4, which exposed over a meter of shell (Figure 712B). Thick strata here were composed primarily of whole Viviparus shell, and lacked abundant vertebrate fauna or sand within the shell matrix. As argued by Sassaman et al. (2005), the dip of the deposits indicates they were down slope from a higher summi t which has since been removed (Figure 712B). The implication is t hat a large cap was emplaced over this area, which likely included a mortuary. Stratigr aphic evidence from the eastern units is also consistent with a shell cap. The diversity of sec ondary midden and prepared surfaces were both replaced sometime after 4800 cal B.P. Therea fter, only undifferentiated lenses of shell were emplaced on the surfac e. Although vert ebrate fauna and some objects were recovered from these strata, the lack of abundant ash and restricted evidence for a wide range of activities suggests that the shell was emplaced on the surface to cap it off. A similar pattern of capping was seen at the earlier Hontoon Dead Creek Mound, as well as the post-mortuary surfaces at Ha rris Creek. Taken together, the evidence for transformation at the scale of the ridge sugg ests communities literally replaced the diversity of material referent s that comprised the settlement with a layer of shell. This action not only obscured prior traces, but effe ctively provided a completely new frame of reference upon which new social memories c ould be constructed. Because we do not know how the post-capping surface was conf igured, it is unknown what kinds of practices may have occurred there. On one hand, a new depositional regime that only renewed the surface, much like that seen at the Hontoon Dead Creek Mound, may have

PAGE 338

338 been instituted there. Alter natively, the entire locali ty may have been abandoned. The sub-capping radiocarbon assays in Test Units 1 and 3 place them at the cusp of pottery production. However, little evidence for Orange period practices have been recovered from the site. It may be that Hontoon Island North was effectively killed as a memory place towards the end of the Mount Taylor period. Future research there, including radiocarbon assays of the she ll cap, will be needed to further articulate these practices. Transposing Histories In this chapter I have examined the biographies of six ridges. All are nominally composed of shellfish, earth, and other mate rials stratigraphically emplaced upon and in association with one another. It is this s uperposition upon which an inference of continuity through ti me has been based in pr ior analyses (e.g., Goggin 1952, Milanich 1994). Many places were repeatedly inhabited, with depositional practices highly routinized. Yet neither places nor deposit ional events were created or experienced equally. That is, shell ridges emerged as differ ent temporalities and fields of practice were woven together and trans lated between places. What did remain constant, however, was the ways in which Mount Taylor communities referenced existing places. In my discussion here, I initially em phasized those places that emerged as distinctive foci of community operations. At places such as Hontoon Dead Creek Mound, Live Oak, Harris Creek, and Blu ffton, communities emphasized the background or horizon of how the world should be, and as such were not social fields that constituted day-to-day residential life. My reasons for starting with these places were two-fold. The first was to demonstrate that seemingly mundane practical acts such as shellfish deposition were not gradual or unchanging in their deployment, but varied by social context and the underlyin g commemorative stra tegy. My second reason was to

PAGE 339

339 examine how Mount Taylor communities woul d transform places in order to allow new temporalities and histories to be experienced in place. At Hontoon Dead Creek Mound and Live Oak, the construction of basal pl atforms provided a new system of reference for the renewal of communities in times of widespread ecological and social change. In the case of the construction of mortuar ies, Mount Taylor communities situated the dead with respect to the living by em placing them on top of, and in the same configuration as, prior settl ements. Mortuary practices were not unchanging either. While some mortuaries could be continually experienced and reworked, others were fixed on the landscape during Mount Taylor time s. These ways of transforming fields of practice were transposed between ancestral or community monuments to places of residence. At both Hontoon Island North and Silver Glen-Locus A, communities reorganized existing referents in order to make new ways of being in place meaningful. Arguably, foundational and transformative acts were routinized as inhabitants returned to these ridges. Taken together, these di fferent ways of being in place underwrote alternative economies of memory through time. It is to these histories that I now turn.

PAGE 340

340 Figure 7-1. Stratigraphic profile a nd stages of the Hontoon Dead Creek Mound (modified from Sassaman [2005a:Figure 4-5]). A) topographic map showing location of the excavated trench. B) west trench profile.

PAGE 341

341 Figure 7-2. Composite phot ograph of Stage 4, Hontoon Dead Creek Mound, showing the location of radiocarbon assays.

PAGE 342

342 Figure 7-3. Crows Bluff (Osceola) m ound profile redrawn from Jeffries Wymans journal (1872).

PAGE 343

343 Figure 7-4. Harris Creek Mortuary base map, showing the location of profiles (after Aten [1999]).

PAGE 344

344 Figure 7-5. Simplified profiles of the Harris Creek Mortuary (modified from Aten [1999]). A) profile bb. C) profile e-e.

PAGE 345

345 Figure 7-6. Comparison of resident ial and mortuary mounds (all Harris Creek data modified from Aten [1999]). A) Hontoon Dead Creek Village, Node-2. B) Harris Creek, pre-mortuary Layer 2. C) Harris Creek Mortuary A, Layer 3. D) Harris Creek Mortuary B, Layer 4. E) Harris Creek Mortuary Layer 6.

PAGE 346

346 Figure 7-7. Bluffton Burial Mound shaded relief topographic map and cross-sections. A) C.B. Moores (1894a:47) excavation profile. B) Sears (1960:56) excavation profile. C) Sears (1960:55) topographic map.

PAGE 347

347 Figure 7-8. False-color shaded relief topogra phic map of the Silver Glen Spring Run Ridge (8LA1W-A).

PAGE 348

348 Figure 7-9. Simplified composite pr ofile of Test Units 5 and 8, Silv er Glen Spring Run Ridge (8LA1W-A).

PAGE 349

349 Figure 7-10. Macro-stra tigraphic units within the Trench (TU 9, 10, 15) at the Silver Glen Spring Run Ridge (8LA1W-A). A) simplified west prof ile. B) composite photograph of TU 9A (approximate scale).

PAGE 350

350 Figure 7-11. Simplified composite profile of Test Unit 8, Silver Glen Spring Run Ridge (8LA1W-A).

PAGE 351

351 Figure 7-12. Topography and excavations at the Hontoon Is land North site. A) hill shaded topographic map. B) photograph of Test Unit 4 west profile (from Sassaman et al. [2005:Figure 3-24] ). C) Photograph of Test Unit 3 south profile (from Sassaman et al. [2005:Figure 3-13]). D) Photograph of Test Unit 1 south profile (from Sassaman et al. [ 2005:Figure 3-10]).

PAGE 352

352 CHAPTER 8 THE CONSTRUCTION OF MOUNT TAYLOR COMMUNITIES I began this study with the premise that shell matrix sites and the practices occurring within them were a central com ponent of Mount Taylor history-making. I contrasted this possibility with conventional approaches to hunter-gatherer settlement and social change that foreground biological reproduction and gradual so cial evolution. As revealed through a consideration of the distribution, structure, and biographies of Mount Taylor places, shell matrix sites along the St. Johns cannot simply be reduced to refuse or unreflective sameness. If ther e was one aspect of Mount Taylor lifeways that remained unaltered over the course of three millennia, it was that communities routinely emplaced shellfish and other materials in or with reference to places of prior inhabitation. As a generalized sequence, these inscriptive practices constituted technologies of remembrance (Jones 2003). With their resu lting materiality distributed across the landscape, such spatial stories (following de Certeau 1984:Chap 9) provided a framework for the production of social hist ories. Mount Taylor depositional practices were not static, but were actively transpos ed and reproduced across fields as structural conditions varied. In this chapter, I revi ew how Mount Taylor communities inhabited and transformed the St. Johns River through depositional practices that gathered the histories of persons, places, and things and reordered them within emergent economies of memory. I conclude with some observati ons regarding inhabitat ion in the post-Mount Taylor world. Inhabiting the St. Johns Seen with the advantage of hi ndsight, the incipient exploitation of shellfish by Mount Taylor communities 7300 years ago appears as a fundamental transformation in

PAGE 353

353 the structure of regional i nhabitation. Within the bounds of current chronological precision the onset of shellf ishing along the St. Johns was eventful, as registered by coeval radiocarbon assays at widely separated s hell matrix sites. The significance of this tradition and its implications for subs equent history-making along the St. Johns would be obscured if we focused on shellfishing as an economic indicator. This has been the approach of previous analyses of t he Mount Taylor period, which emphasize the importance of shellfishing in subsist ence terms alone (e.g., Cumbaa 1976, Goggin 1952; Milanich 1994; Miller 1992, 1998). T hese models are based on an apparent lack of intensive occupation of the St. Johns prio r to the Mount Taylor period. The reduced archaeological visibility may be due to rising water tables, or a true lack of on-theground presence. Economic models also re ly on the establishment of near-modern hydrological conditions that would have provided new opportunities for exploiting aquatic resources. While new water budgets clearly influenced the distribution of wetland resources, it is unknown to what exte nt shellfish were available prior to their earliest exploitation. Were we to take t he local evidence at face value, we might suppose in an ahistorical framew ork that the first Mount Tayl or communities colonized a largely vacant landscape, bereft of significance aside from new subsistence resources (e.g., Binford 1968; Miller 1992). In such a view, Mount Taylor communities targeted the St. Johns River for its economic potentia l, which provided the basis for gradual social evolution through time. Quite to the contrary, on theor etical and empirical grounds it is unlikely that Mount Taylor communities entered an empty landsc ape. All communities inhabit worlds that are filled with habitual and mythical significance that make reference to a background of

PAGE 354

354 possibilities of how the world works (Ba rrett 1999; Bourdieu 1977:Chap 3; Gosden 1994:16). As Barrett and Ko ( 2009:289) argue: people have to be in the world before they can objectify it. That is, communities understand the world as a system of naturalized (i.e. objectified) relationshi ps between humans, things, and places (Bourdieu 1977:164). It is in the deployment of practices in a range of contexts that this order is reproduced and subject to transfo rmation. As discussed in C hapter 2, such systems of reference are inculcated at birth (Bour dieu 1977:85), and emerge through experiences of dwelling, or being in the world (G osden 1994:16; Ingold 20 00a:186). Thus, communities know a landscape prior to it s inhabitation (Ingold 2000a:208), and it is through inhabitation that material referents and collective biographies structure and are structured by rhythms of taskscapes At the level of discourse the order of the world is reproduced through cosmogonic and ancestral narratives that interanimate (sensu Basso 1996:55) ecological and material re ferents with the movements and histories of persons, communities, and other beings (Howey and O'Shea 2006; Jordan 2003; Morphy 1995; Oetelaar and Oetelaar 2008; Santos-Granero 1998; Wilson 2005). We should expect that what we re fer to as early Mount Taylor communities entered the St. Johns with a preexisting system of refe rence in which the world was ordered and valued. It would only be through the deployment of extant practices within different structural conditions that new traditions emerged. Thus, the significance of the shift towards shellfishing must be contextualized wit hin the set of spatial stories that were constructed prior to the M ount Taylor period. I argued in Chapters 3 and 4 that Floridas inhabitants during the late Pleistocene and early Holocene experienced landscapes t hat were in some ways fundamentally

PAGE 355

355 different than those of the middle Holocene. Interior Florida was characterized by drierthan-present conditions ,which may have partially constrained inhabitation to well watered localities. Along the St. Johns, firs t and second order springs or perched water sources were likely important localities that would be revisited (Thulman 2009). That such water sources cannot be reduced to ec onomic tethering is demonstrated quite well by the details of the Windover mort uary pond. Over the course of several centuries, communities returned to this loca lity in the Upper St. Johns. The repeated inhumation of the recently deceased indicates that water, ancestors, and the temporality of death were closely interwoven within the bi ographies and social histories of at least some of these communities. Given the ev idence that the dead were interred rapidly after death (Dickel 2002), it is possible t hat many other ponds of similar age exist elsewhere within the annual r ange of Windovers inhabitant s. That this mortuary tradition persisted outside of the St. Johns after shellfishing was initiated underscores the importance of water as a symbolic reso urce. That is, the new mortuary mounds in the St. Johns did not result in a Flori da-wide abandonment of ear lier traditions. At the regional scale communi ty biographies were not nece ssarily tethered either. While evidence within the St. Johns is depauperate, late Pl eistocene and early Holocene communities outside of the St. Johns wrote diverse spatial stories that ranged geographically throughout Florida. Movement between or social interaction throughout the interior is evidenced by chipped stone tools far from their geological sources. Details from the Windover mortuary also highlight the fact that even before Mount Taylor times, Floridas inhabitants were maki ng annual excursions to the coast. These practices are demonstrated by the presence of marine shell objects in the mortuary, in

PAGE 356

356 addition to Tuckers (2009) dietary reconstruc tions. We currently do not know what kinds of marine resources were exploited. However, Faughts (2004) discovery of a possible inundated coastal shell matrix site in the PaleoAucilla indicates the ancestors of early Mount Taylor communities already had long-reproduced traditions of exploiting and depositing shellfish, just not along the St. Johns. When situated within unfolding social hist ories, the incipient exploitation of shellfish along the St. Johns at places such as the Hontoon Dead Cr eek Complex, Live Oak Mound, and Groves Orange Midden could hav e initially been a nonevent. Given the prior evidence for marine, and possibly s hellfish exploitation, early Mount Taylor communities may have attached no otherworl dly or transformative significance to processing freshwater shellf ish. Following the example developed by Joyce (2004), the creation of new material referents in prev iously uninhabited places by Mount Taylor communities was likely an unintended c onsequence of communities extending traditional practices to new fields. Al though an unintended conse quence, the deposition of shellfish fundamentally alte red the material referents that could be objectified by subsequent communities (see also Gillespie 2008; Pauketat 2000). All future generations who experienced these places w ould encounter the materiality of past agencies distributed throughout the emerging taskscape (Barrett 1999:258). No doubt, these referents structured the tempo of the landscape, as inhabitants incorporated their persistence and reproduction (Joyce 2004:20). On this basis new kinds of histories could be asserted through place, and w hat may have been an unintentional act (creating a shell field) could subsequently have become an intentional practice.

PAGE 357

357 This line of thought needs to be expanded wit h the observations of Appadurai and de Certeau regarding the establis hment or colonization of place. In particular, the creation of new places necessitates a founding spatial story. For de Certeau (1984:125), founding stories involve ritualized, embodied sequences that reenact the initial colonizing acts of a place. Such s patial stories open a fiel d, or enable continued social practices in those places. At t he same time, they provide a basis for the construction of genealogies of place and per son as communities structure and are structured by the materiality of their su rroundings. Appadurai (1996:183) extends de Certeaus discussion, and suggests that: All locality building has a moment of colonizationThe transformation of spaces into places requires a conscious moment, which may be subsequently remembered as relatively routineThe anxiety that attends many rituals of habitation, occupation, or settlement is a recognition of the implicit violence of all such acts of colonization. The establishment of any new place invo lves the habituation and routinization of communities with respect to a potential failure of inhabitation. These foundational practices are ritualized and draw n from the many acts that constitute placemaking (e.g., Bell 1992). The colonization of place is thus an appropriation and manifestation of what were once values previously immanent in the world (Barrett 1999:255). It is important to note, however, that the es tablishment of new places not only rearranges local social fields, but by necessity opens the opportunity fo r other places to be revalued. In this frame of reference, both We stern and non-Western patterns of land use are colonizing processes in which pre-established traditions recursively impose order and are altered through experience (cf. de Certeau 1984:121). Where the earliest expressi ons of Mount Taylor placemaking have been exposed, they were organized, discret e, and repeatedly inhabited. Upon close inspection, the

PAGE 358

358 earliest practices at would-be mound si tes appear mundane and indicative of daily, routinized tasks that occurred in the course of resettlement. For ex ample, at both the saturated components at the Hontoon Dead Creek Co mplex and Groves Orange Midden, the excavators uncovered objects, f ood remains, and shell that could easily be interpreted as true midden deposits representing a wide array of activi ties. The material culture inventory at the Hontoon Dead Creek Village did not match Groves in terms of abundance or diversity, but an ex amination of the concreted shell at Node 1 identified episodes of shell and vertebrate fauna deposit ion. The Hontoon Dead Creek Complex also offers tantalizing evidence for multiple residential nodes or house mounds oriented in a linear array. As discussed in Chapter 5, linear settlements tend to be characteristic of longer-duration encampment s by multiple family gr oups. Whether linear, multihousehold settlements are situat ed beneath later shell ridges is a problem that needs to be addressed in subsequent research. As detailed in Chapter 6, these initial depositional episodes had the net effect of transforming anthropogenically sterile localities into practiced places. Shell nodes were rapidly incorporated into struct ure as inhabitants dr ew on traditional notions of place in reorganizing their movement through the landscape. From the beginning, the deposition of shell was refer entially important. That th ese places were rapidly incorporated into memory is evidenced by repeated deposition that appears to have been similarly partitioned. Resettling places collecting and preferentially depositing materials, and the eventual abandonment of encampments toge ther framed a body of traditions within which biographie s and memories of daily life were intricately enmeshed. While not monumental in scale, the cyclical renewal of past places was the basis for

PAGE 359

359 their ongoing significance. Mount Taylor sett lements were no less domesticated than those of farmers. Although experienced through different temporalities, and likely different recognition of the significance of subsistence resources, Mount Taylor communities inscribed and renewed social memories through place. One might surmise that once established as places to dwell, early Mount Taylor places were reproduced in an unending cycle. Yet one of the unexpected finds of this study is that within a century or so of init ial colonization, inhab itants rapidly rearranged the social and ancestral geograp hy of the St. Johns. This transformation is evidenced by the mounding of shell at ridges such as Hontoon Dead Creek and Live Oak, in addition to the construction of mortuarie s such as Harris Creek. In each case, communities asserted temporalities that enabled new economies of memory. They repositioned inhabitants within multiple, overlapping fi elds of practice that drew connections between biographies, communities, and ancestors. These traditions did not emerge as a c onsequence of stable social or ecological structural conditions. As reviewed in Chapter 3, the St. Johns had begun to experience high water tables and longer hydroperiods, and in this sense was near modern. The extent to which pre-Mount Taylor places may have been inundated in this process is unknown. It is important to remember that what was once a region characterized by low flow and shallow ponds was being repl aced by deeper and more abundant waters. Springs were still not flowing at a modern rate, and were likely more vulnerable to periodic droughts. The entire floodplain was si tuated at least a meter, if not more, below its current configuration, and may have had a significantly different channel arrangements and vegetation stru cture. Higher magnitude flood events would not have

PAGE 360

360 submerged inhabitable landforms, but could have reconfigured or displaced channels, oxbows, and lagoons. At the scale of Peninsul ar Florida, sea levels were still rising rapidly and swamping the coas t. As demonstrated by Tucke rs (2009) isotopic studies at Harris Creek, many Mount Ta ylor inhabitants were still ma king an annual trip to the coast. However, new biographies from throughout Florida and the southeast were also incorporated into the St. Johns. Despite the r egional variability in social composition of communities, the spatial stories that were generated in these contexts referenced an earlier historical mode and created new doma ins for the incorporation of inhabitants across diverse social fields. In particular, Mount Taylor inhabi tants projected a new background that stressed a sens e of continuity and an invariant horizon (e.g., Barrett 1999). Some places appear oriented towards the living. At Hontoon Dead Creek Mound and Live Oak Mound inhabitants capped existi ng settlements. Currently, these two places are the only two shell ridges that have been dated to the period between 7100 and 6900 cal B.P. This new configuration wa s structured by the materiality of the preexisting settlement, as well as the social relationships that were enmeshed in its reproduction. Inhabitants opened up this place for ritualized performances involving the preparation and deposition of shellfish and ot her materials in massive quantities. However, I would argue that the founding acts of place-making were transposed into strategic acts that renewed t he surface of the mound. That is, routinized depositional practices were intentionally directed at increasing the height of the mound summit. Mounds themselves may have become imbued with agency that e ffectively enmeshed practioners with the identitie s sedimented in the mound. The underlying politics of

PAGE 361

361 memory remain unknown to us, although follo wing Morrison (2003) I find it likely that ridge-top performances involved marriages or ot her alliance-building practices. They would have been enabled by and coincident with locally abundant resour ces. Given the volume of shellfish, however, we should not be surprised if freshwater shell was brought in from elsewhere. We also do not know the short-term frequen cy and temporality of mound summit maintenance. Based on the lack of soil development, resurfacing occurred with enough pace that vegetat ion could not take hold. It is notable that these two early ridges are situated at leas t 200 m from running water today. As argued in Chapter 5, the Bo wyers Bluff ridges, Mount Taylor ridge, and others that are also positi oned away from the main channel may have been initiated during this initial period. A future av enue of research should be directed towards establishing the contemporaneity of these places on the landscape. At any point in time more than one ridge may have been actively accret ing. The final forms of these places today may reflect different communities within the valley hosting commemorative ceremonies. At the microscale of every day life shell nodes were likely residential spaces situated with respect to others in linear encampments. These microscale oppositions may have been turned outwards onto the entire basin through ridge-top ritualization. Not all places of this age were oriented to wards the living, however. As detailed in Chapters 4 and 7, early Mount Taylor communi ties constructed mortuaries of sand and shell that appear to have been modeled on the spatiality of residential spaces. If Appadurai and de Certeau are correct regardi ng the importance of reproducing founding events, then the construction of mortuaries on top of, and in the same arrangement as,

PAGE 362

362 shell nodes suggests that the dead were si tuated with respect to community origin myths. Atens reconstruction of the Harris Creek mortuary demonstrates that depositional practices at early Mount Taylor mortuaries were markedly conservative and small-scale once the burial surface was est ablished. Interments may have altered the surfaces, but throughout the lifespan of mortuar ies A and B the overall arrangement and size of the mortuary was maintained. This persistence in mortuary mound architecture stands in marked contrast to the active performance and resurfacing of shell ridges. Taking Hontoon Dead Creek Mound as an example, inhabitants r eengineered them to allow for an increase in overall height. Regardless of the specific int ent of referencing the past in these ways, it is clear that mortuary practice provided yet another temporality of experience for Mount Taylor inhabitants. The presence of both secondar y and primary burials, in addition to a possible mortuary structure, i ndicates that the preparation of the dead was an extended performance, which stands in contrast to prior mortuary programs in which the deceased were rapidly interred after death. Th is shifting temporality in the timing of interment, as well as the widespread distribu tion of these mortuaries, suggests that commemorative performances at burial mo unds incorporated individuals from throughout the valley, creating collective memo ries. In this sense, neither mortuaries nor ridges can be reduced to territorial markers. Community movements between places would have constituted a repeated tr anscendence of times, from the permanent and ancestral to the routine of public and private lives. The landscape composed of mortuaries, ridges, and settlements became routine over the course of several centuries. However, beginning sometime after ca. 6200 cal

PAGE 363

363 B.P. (based on assays at Silver Glen Run), the region under went yet another transformation in the structur e of Mount Taylor place-ma king. From a hydrological perspective, this period is characterized by a deceleration in sea level rise, the onset of ENSO events, and the decreased probability of major storm even ts. The modern hydroperiod was established wit hin the Middle St. Johns around this time as well. One of the consequences of this fl oodplain stability, however, was t he infilling of lagoons and channel segments that once fronted active she ll ridges. From a purely economical standpoint, wherein sites are import ant only as entry points to riverine resources (e.g., Binford 1990), an aggrading floodplain would be expected to have little effect on social reproduction. Given what we know about the kinds of political and commemorative economies that were enabled by collectiv e ridge-top performances, I find it probable that abandonments of ridges due to infilling we re major events. Lacking assays from early ridge tops, we do not know when these early ridges were abandoned. Not only were the material referents within the S t. Johns changing, but new exchange networks would have resulted in the emergence of di fferentiated biographie s locally (following Gosden and Marshall 1999). These networks are evidenced by extralocal objects such as bannerstones, Strombus celts, and beads. Writing about the early est ablishment of states, Yoff ee (2005: 91) notes that one of the most fundamental sh ifts in the organization of politics and ritual was the reduction or simplification of numerous identities and histories through centralized laws and commemorative practices. While there is no evidence for kings, chiefs, or other like social categories, the St. Johns provi des evidence for simplified commemorative strategies after 6200 cal B.P. Variations in practice that emerged amongst these new

PAGE 364

364 structural conditions register new temporalities of experi ence that contradicted earlier naturalized orders. The depositional narratives created at this time appear to have been strategically deployed to make novel sc enarios meaningful in ways that at once obscured, and in some cases eradicated, past pl aces to make them newly relevant. For example, residential domains were either crea ted anew, or maintained at loci that were less vulnerable to floodplain aggradation. Ho wever, these residential places were recreated on a scale and intensity not seen before. In ridges such as at Silver Glen Run and Hontoon Island North, formalized settlements were reproduced above more ephemeral habitation traces. As argued in Chapter 7, inhab itants treated old settlements with the same veneration as they did for the dead. In the case of Silver Glen Run, the community appears to have capped and obscured all traces of preexisting material referents. In re-colonizing the place, al ternative spatial stories were enabled on the ridge. Once the sand cap had been laid down, a linear settlement composed of multiple households was constructed there. Although it was abandoned for extended periods of time, communities did return and remake the settlement in ways that conformed with the new layout. Communities also capped the Hontoon Island North site, apparently to open up the space for alternative practices. The extent to which communal mortuaries continued to be maintained is unknown. The body of evidence, such as that documented at Bluffton and Thornhill Lake, indicates that later Mount Taylor inhabitants may have fundamentally altered the commemorative and political significance of anc estry. In the case of Bluffton, an individual was interred within a depositional sequence that arguably recapitulated the founding myths of earlier communities. Ye t in this case, the mortuary mound was

PAGE 365

365 excluded from further modifica tion. At Thornhill Lake, burial mounds were emplaced upon what had been an active shell ridge. Although habitation near the ridge may have continued after the mounds were constr ucted, they changed the physicality and significance of the place. The generati on of these new mortuary mounds, which are paired in some cases, suggests that t he singularity of linear places that had been reproduced throughout the Mount Taylor period was ruptur ed. The dual mortuary mounds at Thornhill Lake and Bl uffton, as well as the possi bility of them at Hontoon Island North, suggests a fundamental transfo rmation in how regional identities were inscribed at the local level. That is, ther e would appear to be a new coming together of distinct identities, possibly based on ancestr y but also based on the biographies that could be traced to the coast through exchange networks. Post-Inscription I have restricted my examination of social histories along the St. Johns from the first visible manifestation of shellfishing until the end of the preceramic Archaic period. As detailed, I argue that shell mounds and thei r significance were routinely constructed over the course of this period, although the act of shellfishing and the medium of placemaking (i.e. depositing shellfish remains and other materials) remained constant. Just as Mount Taylor communities had to incorpor ate the past into co ntemporary frames of reference, so too did their descendants of the Orange and later St. Johns periods. Mount Taylor communities made available the conditions (Barrett 1999:264) through which Orange and St. Johns communities could politicize their own histories. As a consequence, much of the presumed continuity and increase in social complexity that has been attributed to later communities on the basis of site size and evidence (or lack thereof) of monumentality has to be reframed. These communities were living in a

PAGE 366

366 landscape already sedimented with ancestral pr esence. In the cour se of investigating Mount Taylor traditions, in fact, several patterns have emerged that warrant a brief mention. We have seen that by 5000 years ago, the M ount Taylor period culminates in the production of a number of loci inscribed wit h mortuary mounds into which persons and objects with arguably diverse biographies were interred. In this landscape, community identity was apparently negotiat ed through performances at mortuary mounds, wherein ancestors and the threads linking them with s patial stories outside of the St. Johns were incorporated. Contrary to the expectations of social evolution arguments, all of the diversity of monumentality, ex change, and mortuary treatment collapses at the close of the Mount Taylor period. As briefly discussed at the conclusion of Chapter 4, diagnostic artifacts of the Orange period are typically found situated aw ay from shell ridges, even though Orange settlements were emplaced near by. Radiocarbon dates from below near-surface mounded shell at the Hontoon Is land North site indicate the site was capped and abandoned sometime after 4800 years ago, and apparently remained unoccupied until the later St. Johns peri od. Orange period ceramics are rarely recovered from the surfaces of shell ridges throughout the region, nor are they found in any great frequency upon sand mounds (Endoni no 2008; Piatek 1994; Sears 1960). One caveat, of course, is that many ridges have had their apexes removed. Regardless, evidence for mortuaries anywhere in the interior during the Orange period is spotty at best. Yet, as discussed in Chapters 3 and 4, there is no evidence for a regional population collapse.

PAGE 367

367 What happened? We can start with the observation that Orange period pottery has its origins a century earlier on the coas t of Florida, and its introduction was likely facilitated by preexisting exchange netwo rks (Sassaman 2004b). Although this process is currently modeled as an in situ evolutionary progression, pottery production in Florida in fact signals a major restructuring of domai ns of practice, the scale of which is only coming to light now. For example, at the Blue Springs Midden B site (8VO43), Sassaman (2003a) documented the presence of a semi-circular Orange Period compound, characterized by four crushed shell floors, measuring 4 to 6 m in diameter, and spaced approximately 8 m apart. These clusters were oriented in an arc, approximately 34 m in diameter, facing the S t. Johns and adjacent to Blue Spring Run. This organization of space is arguably new to the region. However, circular villages and monuments are well documented in coastal gr oups arguably historically affiliated with those of the interior (Russo and Heide 2001; Sassaman et al. 2006). As a further contrast to Mount Taylor inhabitation, Orange compounds were routinely sited near, but noticeably away from preexisting shell ridges, which were apparently avoided. As noted above, this observation has to be tempered by the fact that many of the summits of the largest sites have been removed prior to modern investigations. Regardless, this pattern is evident at Blue Spring Midden B, the Hontoon Dead Creek Village, and Locus A at t he Silver Glen Run Complex. It would be a mistake, however, to conc lude that Orange communities did not engage in large scale gatherings and monumental cons truction. In this regard, the distribution of Orange period pottery wares provides some clues Most Orange residential spaces in the middle St. Johns are dominated by plain wares. In contrast, incised and technologically

PAGE 368

368 distinct Orange pottery is largely restricted to four locations in the valley: the Silver Glen Run Complex, Harris Creek, Old Enterprise and Orange Mound. These localities are situated some 20 to 30 km apart along the St. Johns River. A similar spatial division in the deposition of plain pottery at small sca le residential sites and decorated pottery at large scale aggregation places (such as shel l rings) has been identified along the coast. This pattern arguably represents a new division between commemorative and residential spaces (Sassam an 2003c; R. Saunders 2004). Recent excavations at the Silver Glen R un Complex East site are now providing a context for interpreting this emergent pattern. As discussed in Chapter 5, prior to its destruction in 1923, Wyman described the site as a U-shaped amphitheater, upwards of 300 m long on a side, rising to a height of up to 8 m, with a deep valley between the ridges. In Figure 8-1 I have reconstructed the surface topography of this shell mound, in addition to the primarily Mount Taylor-aged Locus A shell ridge (situated to the west) as Wyman would have seen it. I have s uperimposed Wymans description (based in part on an unpublished sketch map he made) over the existi ng surface topography. The mound is scaled to the distribution of s hell present at the si te identified during subsurface testing, which is largely conf ormant with Wymans description (Sassaman et al. 2010). If these dimensions are even close to correct, this mound is the largest in northeast Florida. Although the mound shares design principl es with Orange period rings elsewhere, the multi-mound complex at the mouth of Silver Glen was not constructed de novo Not only was the Orange period mound situated in opposition to a large shell ridge to the west (Locus A), but it appears Orange communi ties grafted a new monument on top of

PAGE 369

369 a preexisting Mount Taylor ridge in a way that referenced both coastal and interior ancestries. This configuration shares a pattern with the Harris Creek complex, where Orange pottery is found in shell ridges extend ing from the Mount Taylor shell ridge. Moreover, at Silver Glen there is some suggestive evidence fo r within-site differentiation in the production and disposal of pottery. Incised wares, found in great abundance, are largely restricted to the peri phery of the mounds water-facing sides. In the terrestrial portion of the rear ridge, however, excava tions uncovered Orange plain wares. Yet even in these contexts, there is little to suggest that shell accumulated here through acts of daily settlement. To the contrary in the Orange period inhabitants dug large pits (upwards of 2-m wide) and filled with clean, mounded shell. Future research will examine whether the two ridges ac tually represent the constructi ve efforts of two distinct ancestries. Regardless, the current data suggest that Orange communities, like their Mount Taylor ancestors, constructed their own histories by creatively combining and reworking the past in place through depositional practices. Crafting Histories across the Divide When viewed through the lens of subsis tence and repeated use of place, the practices that resulted in shel l matrix sites of the Middle S t. Johns River could easily be mistaken for unreflexiv e continuities. There is little evidence for significant variations in subsistence throughout either the Mount Tayl or or Orange periods. Where shifts in settlement location are present, they rare ly involved the whol esale abandonment of large swaths of territory. Yet, we must be circumspect when examining continuities, as they may be more apparent than real (Bradley 2003; Gosden and Lock 1998). As shown along the St. Johns, over the course of three thousand years, communities were variously reproduced through an ongoing historical production of biographical and

PAGE 370

370 ancestral spatial stories in place. While altered through the recognition and incorporation of different kinds of events and others, the past as place and process was recursively referenced and drawn upon in creative ways. The results of my investigations into s hell matrix site histories indicate that traditions of routine and commemorative pr actice emerged in the context of new ecologies and social interactions. Such places of past practice rapidly emerged as central foci for return visits and local interaction as inhabitants returned repeatedly from throughout the state. This pot ential diversity appears to have been negotiated through veneration of the dead in highly public acts. The primary inscriptive practice here was not necessarily the interment of the deceas ed and construction of the mound. Instead, it was the striking replacement of se ttlements with sacred places emplaced and modeled on those structures, as is commo nly found throughout the world (e.g., Bradley 1998:Chapters 3; Johansen 2004; Jordan 2003; Joyce 2004). These central metaphors for referencing comm unity, and for negotiating contem porary social relations, were continuously reworked throughout su ccessive millennia. During times of considerable ecological and social upheaval, some settlements were abandoned, others became intensive loci of daily practice, and still others were maintained as commemorative venues. Yet throughout, communities mobilized the past to write histories that either acco mmodated contemporary concerns or countered change. In this frame of reference, landsca pe use as traditions of ecolog ical exploitati on, residential spaces, commemoration of ancestral and mythic orders, and interactions with others emerges as the central method through which these hunter-gatherer communities wrote

PAGE 371

371 their own histories. In fact this historiographic method w ould seem to be present in most societies, irrespective of t heir particular subsistence economy. Writing over a century ago, Jeffries Wyman (1875:11) commented that anyone seeing the St. Johns shell mounds for the fi rst time might well be excused for doubting that such immense quantities of small shells could have been brought together by human labor. While Wyman accurate ly recognized mounds as anthropogenic, questions of agency still surround their origins and histories. Decontextualized and repositioned along a complexity continuum, shell matrix sites and the practices with which they were reproduced rapidly fade as epiphenomena of a mode of production or an ecological adaptation. However, a consi deration of the social contexts of their construction indicates that these places, wh ether they were early settlements, dedicated mortuaries, or ritualized platforms structured and were structured by the complexities of community history-making in place. In this frame of reference, the density and distribution of Archaic shell matrix sites along the St. Johns is not a reflection of abundant aquatic resources. Instead, the St. Johns was relationally ranked as the center of the world because of the histories that could be negotiated through place.

PAGE 372

372 Figure 8-1. Reconstructed Silver Glen Run Complex.

PAGE 373


PAGE 374

374 Table A-1. Radiocarbon assays fr om Mount Taylor and Orange components Site Number Site Name Sample Method Material Measured 14C Age B.P. 12C/13C Ratio Conventional 14C Age B.P. 2 cal B.P. Provenience Source 8LA1-E Silver Glen Run Complex East BETA255903 AMS Charcoal 3610 40 -25.8 3600 40 4080 Feature 1 Sassaman et al. (2010) 8LA1-E Silver Glen Run Complex East BETA166672 AMS Soot 4020 60 -25.2 4020 60 4810 Vessel 6 Sassaman (2003c) 8LA1-E Silver Glen Run Complex East BETA166673 AMS Soot 4060 40 -24.4 4070 40 4810 Vessel 27 Sassaman (2003c) 8LA1W-B Silver Glen Run Complex Locus B BETA255904 AMS Charcoal 3820 40 -24.3 3830 40 4410 Feature 15 Sassaman et al. (2010) 8LA1W-A Silver Glen Run Complex Locus A BETA248528 AMS Charcoal 5160 50 -25.4 5151 50 6000 TU10A, (C144) This Report 8LA1W-A Silver Glen Run Complex Locus A BETA236137 AMS Charcoal 5290 40 -24.7 5290 40 6190 TU5E, Stratum XXII This Report 8LA1W-A Silver Glen Run Complex Locus A BETA248527 AMS Charcoal 5390 40 -24.5 5400 40 6290 TU10A, (C143) This Report 8LA1W-E Silver Glen Run Complex East BETA166671 AMS Soot 3690 60 -25.8 3680 60 4220 Vessel 12 Sassaman (2003c) 8LA28 Mosquito Inlet BETA166674 AMS Soot 3600 40 -24.6 3610 40 4080 Vessel 2 Sassaman (2003c) 8LA88B Bowyers Bluff 2 I-563 Radiometric Charcoal 4725 180 -25.0* 4725 180 5890 33-inch depth Bullen and Bryant (1965) 8V02601 Groves' Orange Midden AMS Charcoal 3850 70 -25.0* 3850 70 4500 Stratum II, Level 3 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65865 AMS Hickory nut 4080 60 -25.0* 4080 60 4820 Stratum II, Level 7 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden AMS Charcoal 4115 75 -25.0* 4115 75 4830 Stratum II, Level 5 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65864 AMS Hickory nut 4190 60 -25.0* 4190 60 4850 Stratum II, Level 13 McGee and Wheeler (1994)

PAGE 375

375 Table A-1. continued Site Number Site Name Sample Method Material Measured 14C Age B.P. 12C/13C Ratio Conventional 14C Age B.P. 2 cal B.P. Provenience Source 8V02601 Groves' Orange Midden AMS Charcoal 4400 125 -25.0* 4400 125 5450 Stratum II, Level 8 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65859 AMS Wood Charcoal 4810 70 -25.0* 4810 70 5660 Stratum IV, Level 10 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65860 AMS Hickory nut 4930 80 -25.0* 4930 80 5900 Stratum IV, Level 14 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65862 AMS Hickory nut 5040 60 -25.0* 5040 60 5910 Stratum IV, Level 16 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA59801 AMS Hickory nut 5100 70 -25.0* 5100 70 5990 Stratum IV, Level 11 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA59802 AMS Hickory nut 5160 80 -25.0* 5160 80 6180 Stratum IV, Level 14 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA59803 AMS Hickory nut 5580 80 -25.0* 5580 80 6550 Stratum IV, Level 19 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA59804 AMS Hickory nut 5930 80 -25.0* 5930 80 6960 Stratum IV, Level 21 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65863 AMS HICKORY NUT 6200 70 -25.0* 6200 70 7270-6920 Stratum V, Level 20 McGee and Wheeler (1994) 8V02601 Groves' Orange Midden BETA65861 AMS HICKORY NUT 6210 -25.0* 6210 7260-6970 Stratum V, Lev 22 McGee and Wheeler (1994) 8VO202 Hontoon Island North BETA255905 AMS Charcoal 4480 40 -28.1 4430 40 5280 TU1, Stratum VII, Level D This Report 8VO202 Hontoon Island North BETA244050 AMS Charcoal 4430 40 -23.9 4450 40 5290 TU1, Stratum V, Level B This Report 8VO202 Hontoon Island North BETA244049 AMS Charcoal 4500 40 -25.6 4490 40 5300 TU3, Stratum IX Level A This Report 8VO202 Hontoon Island North BETA241924 AMS Charcoal 4440 40 -20.7 4510 40 5310 TU3, Stratum V, Level C This Report 8VO214 Hontoon Dead Creek Mound BETA255907 AMS Charcoal 5900 50 -24.3 5910 50 6880 TU5, Stratum V (Bulk 8) This Report

PAGE 376

376 Table A-1. continued Site Number Site Name Sample Method Material Measured 14C Age B.P. 12C/13C Ratio Conventional 14C Age B.P. 2 cal B.P. Provenience Source 8VO214 Hontoon Dead Creek Mound BETA202281 AMS Hickory nut 6070 70 -26.4 6040 70 7160 Auger 1 Randall and Sassaman (2005) 8VO214 Hontoon Dead Creek Mound BETA202281 AMS Nutshell 6100 40 -24.6 6110 40 7160 TU5, Stratum XXII (Bulk 1) This Report 8VO214 Hontoon Dead Creek Mound BETA255906 AMS Charcoal 6090 50 -22.0 6140 50 7170 TU5, Stratum XII (Bulk 3) This Report 8VO215 Hontoon Dead Creek Village BETA244052 AMS Charcoal 3730 40 -27.0 3700 40 4150 Feature 2 This Report 8VO215 Hontoon Dead Creek Village BETA217769 Radiom etric Marine shell 5670 60 -7.9 5950 60 6510 TU2, Level B Randall (2007) 8VO215 Hontoon Dead Creek Village BETA219933 AMS Charcoal 6320 40 -27.6 6280 40 7310 TU3, Base of Block Randall (2007) 8VO24 Harris Creek BETA16675 AMS Soot 3600 40 -23.3 3630 40 4080 Vessel 21 Sassaman (2003c) 8VO24 Harris Creek BETA166676 AMS Soot 3730 40 -24.4 3740 40 4230 Vessel 1 Sassaman (2003c) 8VO24 Harris Creek BETA166677 AMS Soot 3920 40 -24.1 3930 40 4510 Vessel 252 Sassaman (2003c) 8VO24 Harris Creek M-1265 Radiometric Charcoal 5320 204 -25.0* 5320 204 6500 Layer 3, Mortuary A Jahn and Bullen (1978) 8VO24 Harris Creek M-1270 Radiometric Marine shell 5030 120 0.0 5432 204 6270 Layer 7 or 8 Jahn and Bullen (1978) 8VO24 Harris Creek M-1268 Radiometric Charcoal 5450 184 -25.0* 5450 184 6660 Layer 8 Jahn and Bullen (1978) 8VO24 Harris Creek M-1264 Radiometric Charcoal 5450 303 -25.0* 5450 303 6930 Layer 3 Base, Bottom of Mortuary A Jahn and Bullen (1978) 8VO24 Harris Creek X9111A AMS Human bone 5825 62 -17.6 5825 62 6780 Burial 9, Layer 4 Tucker (2009) 8VO24 Harris Creek X9109A AMS Human bone 5904 62 -16.1 5904 62 6890 Burial 3, Layer 4 Tucker (2009) 8VO24 Harris Creek X9112RA AMS Human bone 6053 65 -16.2 6053 65 7160 Burial 31, Layer 3 Tucker (2009) 8VO24 Harris Creek X9110 AMS Human bone 6125 83 -17.8 6125 83 7240 Burial 7, Layer 3 Tucker (2009) 8VO41 Live Oak Mound BETA164961 AMS Charcoal 6260 50 -24.9 6260 50 7280 LPIA, Stratum XIII Sassaman (2003a)

PAGE 377

377 Table A-1. continued Site Number Site Name Sample Method Material Measured 14C Age B.P. 12C/13C Ratio Conventional 14C Age B.P. 2 cal B.P. Provenience Source 8VO43 Blue Spring Midden B BETA145694 AMS Nutshell 3500 70 -24.9 3510 70 3980 TU2, Level E Sassaman (2003a) 8VO43 Blue Spring Midden B BETA145695 AMS Charcoal 3720 40 -24.3 3730 40 4230 TU2, Level K Sassaman (2003a) 8VO43 Blue Spring Midden B BETA145692 AMS Charcoal 3790 50 -25.4 3780 50 4380 TU4, Feature 7 Sassaman (2003a) 8VO43 Blue Spring Midden B BETA164963 AMS Charcoal 4210 50 -25.3 4210 50 4860 TU5, 140-150 CM BS Sassaman (2003a) 8VO43 Blue Spring Midden B BETA145693 AMS Charcoal 4360 120 -24.9 4360 120 5310 TU1, Level M Sassaman (2003a) 8VO53 Lake Monroe Outlet Midden BETA146752 AMS Charcoal 4640 50 -25.0* 4640 50 5580 TUA, Level 19 ACI, Inc. and Janus Research, Inc. (2001) 8VO53 Lake Monroe Outlet Midden BETA146748 Radiometric Charcoal 4650 110 -25.0* 4660 110 5600 TUA, Level 9 ACI and JR (2001) 8VO53 Lake Monroe Outlet Midden BETA146751 AMS Charcoal 4710 40 -25.0* 4710 40 5580 TUA, Level 14 ACI and JR (2001) 8VO53 Lake Monroe Outlet Midden BETA146753 AMS Charcoal 4760 40 -25.0* 4760 40 5590 TUC, Level 9 ACI and JR (2001) 8VO53 Lake Monroe Outlet Midden BETA146756 AMS Charcoal 4760 40 -25.0* 4780 40 5600 TUC, Level 15 ACI and JR (2001) 8VO53 Lake Monroe Outlet Midden BETA146749 Radiometric Charcoal 5080 80 -25.0* 5090 80 5990 TUA, Level 13 ACI and JR (2001) 8VO58-60 Thornhill Lake Complex BETA231047 AMS Charcoal -22.8 4170 50 4840 Mound A, shell at mound base Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231054 AMS Charcoal -26.1 4430 40 5280 TU1, Stratum V Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231049 AMS Charcoal -25.1 4950 90 5910 South ridge, TUF, Feature 7 Endonino (2008)

PAGE 378

378 Table A-1. continued Site Number Site Name Sample Method Material Measured 14C Age B.P. 12C/13C Ratio Conventional 14C Age B.P. 2 cal B.P. Provenience Source 8VO58-60 Thornhill Lake Complex BETA231048 AMS Charcoal -26.0 4970 40 5880 Mound B, TUD, Stratum VI, Basal Shell Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231053 AMS Charcoal -24.4 5130 40 5990 TU H, Stratum IV Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231050 AMS Charcoal -24.3 5170 40 6000 North Ridge, TU1, Stratum Va top Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231052 AMS Charcoal -22.7 5190 40 6170 TUH, Feature 2, Level 2 Endonino (2008) 8VO58-60 Thornhill Lake Complex BETA231051 AMS Charcoal -26.5 5420 40 6300 North Ridge, TU1, Stratum VI Endonino (2008) 8VO81 Tomoka Mounds BETA54622 AMS Marine shell 4060 70 0.0 4460 70 4810 Coquina Midden at Base of Mound 6 Piatek (1994) *Presumed corrected ratio

PAGE 379

379 LIST OF REFERENCES 14CHRONO Centre 2010 Marine Reservoir Correction. Electronic document, http://intcal.qub.ac.uk/marine/ accessed 10/1/2008. Abshire, A. E., Alden L. Potter, Allen R. Ta ylor, Clyde H. Neil, Walt er H. Anderson, John I. Rutledge and Stevenson B. Johnson 1935 Som e Further Papers on Aboriginal M an in the Neighborhood of the Ocala National Forest Civilian Conservation Corps, Company 1420, Ocala Camp, Florida F5. Ackermann, Friedrich 1999 Airborne Laser Scanning-Present Status and Future Expectations. ISPRS Journal of Photogramme try and Remote Sensing 54:64. Adamus, Christine, Davi d Clapp and Stephen Brown 1997 Surface Water Drainage Basin Boundaries St. Johns River Water Management District: A Reference Guide Technical Public ation SJ97-1. St. Johns River Water Management District, Palatka, Florida. Adler, Michael A. and Richard H. Wilshusen 1990 Large-Scale Integrative Facilities in Tribal Soci eties: Cross-Cultural and Southwestern US Examples. World Archaeology 22:133. Agassiz, Alexander 1881 Biographical Memoir of Louis Franois de Pourtals Read before the National Academy of Sciences, April 22, 1881, Washington. Aguilar, Fernando J., Francisco Agera, Manuel A. Aguilar and Fernando Carvajal 2005 Effects of Terrain Morphology, Samp ling Density, and Interpolation Methods on Grid DEM Accuracy. Photogrammetric Engineer ing & Remote Sensing 71:805. Ahearne, Jeremy 1995 Michel de Certeau: Interpretation and its Other Stanford University Press, Stanford, California. Alley, R. B., P. A. Mayewski, T. Sowers, M. Stuiver, K. C. Ta ylor and P. U. Clark 1997 Holocene Climatic Instability: A Pr ominent, Widespread Event 8200 Yr Ago. Geology 25:483. Ambrose, W. R. 1967 Archaeology and Shell Middens. Archaeology and Physical Anthropology in Oceania 2:169.

PAGE 380

380 Ames, Kenneth M. 2004 Supposing Hunter-Gatherer Variability. American Antiquity 69:364. Appadurai, Arjun 1996 Modernity at Large: Cultural Dimensions of Globalization University of Minnesota Press, Minneapolis. Archaeological Consultants, Inc. and Janus Research 2001 Phase III Mitigative Excavations at Lake Monroe Outlet Midden (8VO53), Volusia County, Florida Submitted to U.S. Department of Transportation Federal Highway Administration and Florida Depar tment of Transportation District Five. Aslan, Andres, Whitney J. Autin and Michael D. Blum 2005 Causes of River Avulsion: Insights from the Late Holocene Avulsion History of the Mississippi River, U.S.A. Journal of Sedimentary Research 75:650. Aten, Lawrence E. 1999 Middle Archaic Ceremonialism at Tick Is land, Florida: Ripley P. Bullen's 1961 Excavations at the Harris Creek Site. The Florida Anthropologist 52:131. Austin, Robert J 1997 The Economics of Lithic-Resource Use in South-Central Florida. Unpublished Ph.D. Dissertation, Department of Anthropology, University of Florida, Gainesville. 2001 Paleoindian and Archaic Ar chaeology in the Middle Hillsborough River Basin: A Synthetic Overview Southeastern Archaeological Re search, Inc. Submitted to Tampa Bay Water, Inc. 2004 Multidisciplinary Investigations at We st Williams, 8HI509: An Archaic Period Archaeological Site Locat ed within Florida Gas Transmission Company's Bayside Lateral Pipeline Corridor, Hillsborough County, Florida. Southeastern Archaeological Research, Inc. Submitt ed to Florida Gas Transmission Company, Inc. Austin, Robert J. and Richard W. Estabrook 2000 Chert Distribution and Exploitation in Peninsular Florida. The Florida Anthropologist 53:116. Bailey, Geoff 2007 Time Perspectives, Palimp sests and the Archaeology of Time. Journal of Anthropological Archaeology 26:198. Bailey, Geoff and Alan S. Craighead 2003 Late Pleistocene and Holocene Coastal Palaeoeconomies: A Reconsideration of the Molluscan Evidence from Northern Spain. Geoarchaeology 18:176.

PAGE 381

381 Bailey, Geoff and Nicky Milner 2002 Coastal Hunter-Gatherers and Social Evolution: Marginal or Central? Before Farming 3:1. Bailey, Geoff and John Parkington 1988 The Archaeology of Prehist oric Coastlines: An Introduction In The Archaeology of Prehistoric Coastlines edited by Geoff Bailey and John Parkington, pp. 1. Cambridge University Press, Cambridge. Baker, Victor R. 1987 Paleoflood Hydrology and Extraordinary Flood Events. Journal of Hydrology 96:79. 2008 Paleoflood Hydrology: Origin, Progress, Prospects. Geomorphology 101:1 13. Bale, William 1998 Historical Ecology: Premises and Postulates In Advances in Historical Ecology edited by William Bale, pp. 13. Columbia University Press, New York. Balsille, James H. and Joseph F. Donoghue 2004 High-Resolution Sea-Level History for the Gulf of Mexico since the Last Glacial Maximum Report of Investigations No. 103. Florida Geological Survey, Tallahassee. Barbour, George M. 1882 Florida for Tourists, Invalids, and Settl ers: Containing Practical Information Regarding Climate, Soil, and Productions; Cit ies, Towns, and People; the Culture of the Orange and Other Tropical Fruits ; Farming and Gardening; Scenery and Resorts; Sport; Routes of Travel, Etc., Etc D. Appleton, New York. Bard, Edouard 2002 Climate Shock: Abrupt Changes over Millennial Time Scales. Physics Today 55:32. Barnard, Alan 1999 Images of Hunters and Gat herers in European Social Thought In The Cambridge Encyclopedia of Hunters and Gatherers edited by Richard B. Lee and Richard Daly, pp. 375. Cambridge Un iversity Press, Cambridge. 2002 The Foraging Mode of Thought In Selfand Other-Images of HunterGatherers edited by Henry Stewart, Alan Ba rnard and Keiichi Omura, pp. 5. Senri Ethnological Studies No. 60. Na tional Museum of Ethnology, Osaka.

PAGE 382

382 Barnes, B. 2001 Practices as Collective Action In The Practice Turn in Contemporary Theory edited by T. R. Schatzki, K. KnorrCetina and E. von Savigny, pp. 17. Routledge, New York. Barnes, Ian 2004 Aerial Remote-Sensing Techniques Used in the Management of Archaeological Monuments on the British Army's Salisbury Plain Training Area, Wiltshire, UK. Archaeological Prospection 10:83. Barrett, John C. 1994 Fragments from Antiquity: An Archaeolog y of Social Life in Britain, 2900 1200 B.C. Blackwell, Oxford. 1999 The Mythical Landscapes of the British Iron Age In Archaeologies of Landscape: Contemporary Perspectives edited by Wendy Ashmore and A. Bernard Knapp, pp. 253. Blackw ell, Malden, Massachusetts. 2000 A Thesis on Agency In Agency in Archaeology edited by Marcia-Anne Dobres and John E. Robb, pp. 61. Routledge, London. 2001 Agency, the Duality of Structur e, and the Problem of the Archaeological Record In Archaeological Theory Today, edited by Ian Hodder, pp. 141. Polity Press, Cambridge. Barrett, John C. and Ilhong Ko 2009 A Phenomenology of Landscape: A Cr isis in British Landscape Archaeology? Journal of Social Archaeology 9:275. Bartram, William 1791 Travels through North & South Carolina, Georgia, East & West Florida, the Cherokee Country, the Extensive Territories of the Muscogulges or Creek Confederacy, and the Country of the Chac taws: Containing an Account of the Soil and Natural Productions of Those Regions, Together with Observations on the Manners of the Indians: Embellished with Copper-Plates James & Johnson, Philadelphia. Basso, Keith H. 1996 Wisdom Sits in Places: Notes on a Western Apache Landscape In Senses of Places edited by Steven Feld and Keith H. Basso, pp. 53. School of American Research, Santa Fe. Beardsley, R., P. Holder, A. Krieger, B. J. Meggers, J. Ri naldo and P. Kutsche 1955 Functional and Evolut ionary Implications of Community Patterning In Memoirs of the Society for American Archaeology No. 11, edited by R. Wauchope, pp. 130. Society for American Archaeology, Salt Lake City.

PAGE 383

383 Beasley, Virgil R. 2008 Monumentality During the Mid-Holocene in the Upper and Middle St. Johns River Basins, Florida. Unpublished Ph.D. Dissert ation, Department of Anthropology, Northwestern University, Evanston, Illinois. Beaton, J. M. 1985 Evidence for a Coastal Occupati on Time-Lag at Princess Charlotte Bay (North Queensland) and Implications fo r Coastal Colonization and Population Growth Theories for Aboriginal Australia. Archaeology in Oceania 20:1. Beckage, Brian, William J. Platt, Matthew G. Slocum and Bob Panko 2003 Influence of the El Nio Southern Oscillation on Fire Regimes in the Florida Everglades. Ecology 84:3124. Bell, Catherine M. 1992 Ritual Theory, Ritual Practice Oxford University Press, New York. Belleville, Bill 2000 River of Lakes: A Journey on Florida's St. Johns River University of Georgia Press, Athens. Bender, Barbara 1985 Emergent Tribal Formations in the American Midcontinent. American Antiquity 50:52. Bender, Barbara 1993 StonehengeConteste d Landscapes (Medieval to Present-Day). In Landscape: Politics and Perspectives edited by Barbara Bender, pp. 245. Berg, Oxford. Benson, Carl A. and Howard Bruce Green 1962 The Kimball Midden, Lake County. The Florida Anthropologist 15:113. Beriault, John, Robert Ca rr, Jerry Stipp, Richard Johnson and Jack Meeder 1981 The Archeological Salvage of the Ba y West Site, Collier County, Florida. The Florida Anthropologist 34:39. Berliner, David 2005 Social Thought & Co mmentary: The Abuses of Memory: Reflections on the Memory Boom in Anthropology. Anthropological Quarterly 78:197. Bernardini, Wesley 2004 Hopewell Geometric Earthworks: A Case Study in the Referential and Experiential Meani ng of Monuments. Journal of Anthropological Archaeology 23:331.

PAGE 384

384 Bettinger, Robert L 1991 Hunter-Gatherers: Archaeologi cal and Evolutionary Theory. Plenum Press, New York. 1999 From Traveler to Processor: R egional Trajectories of Hunter-Gatherer Sedentism in the Inyo-Mono Region, California In Settlement Pattern Studies in the Americas: Fifty Years since Vir edited by Brian R. Billman and Gary M. Feinman, pp. 39. Smithsonian In stitution Press, Washington. Bettinger, Robert L. and Martin A. Baumhoff 1982 The Numic Spread: Great Ba sin Cultures in Competition. American Antiquity 47:485. Bewley, Robert H. 2003 Aerial Survey for Archaeology. Photogrammetric Record 18:273. Binford, Lewis R. 1962 Archaeology as Anthropology. American Antiquity 28:217-225. 1968 Post-Pleistocene Adaptations In New Perspectives in Archeology edited by Sally R. Binford and Lewis R. Binf ord, pp. 313. Aldine, Chicago. 1979 Organization and Formation Processe s: Looking at Curated Technologies. Journal of Anthropological Research 35:255. 1980 Willow Smoke and Dog's Tails: Hunt er-Gatherer Settlement Systems and Archaeological Site Formation. American Antiquity 45:4. 1982 The Archaeology of Place. Journal of Anthropological Archaeology 1:5. 1990 Mobility, Housing, and En vironment: A Comparative Study. Journal of Anthropological Research 46:119. Binford, Lewis R. and Amber L. Johnson 2002 Foreword In Beyond Foraging and Collecting : Evolutionary Change in Hunter-Gatherer Settlement Systems edited by Ben Fitzhugh and Junko Habu, pp. viixiii. Kluwer Academic/Plenum Press, New York. Bird-David, Nurit H. 1990 The Giving Environment: Another Perspective on the Economic System of Gatherer-Hunters. Current Anthropology 31:189. Blake, Emma 1998 Sardinia's Nuraghi: Four Millennia of Becoming. World Archaeology 30:59 71. Blessing, Meggan E. 2009 The Snails' Tale: The Isotopic Record of Freshwater Gastropods in Archaic Shell Deposits of Northeast Florida Paper presented at the 66th Annual Southeastern Archaeological Conf erence, Mobile, Alabama.

PAGE 385

385 Bloom, Arthur L. 1983 Sea Level and Coastal Changes In Late Quaternary Envi ronments of the United States edited by H. E. Wright, pp. 42. Vol. 2, The Holocene. University of Minnesota Pre ss, Minneapolis. Bogan, Arthur E. 1993 Freshwater Bivalve Extinctions (Mollusca: Unionoida): A Search for Causes. American Zoologist 33:599. Bourdieu, Pierre 1977 Outline of a Theory of Practice Cambridge University Press, Cambridge. 1990 The Logic of Practice Stanford University Press, Stanford. Boone, Elizabeth Hill 2000 Stories in Red and Black: Pictorial Histories of the Aztecs and Mixtecs University of Texas Press, Austin. Bourdo, Eric A., Jr. 1956 A Review of the General Land Office Survey and of its Use in Quantitative Studies of Former Forests. Ecology 37:754. Braconnot, P. B. Otto-Bliesner S. Harrison, S. Joussaume, J.-Y. Peterchmitt, A. AbeOuchi, M. Crucifix, E. Dr iesschaert, Th. Fichefet, C. D. Hewitt, M. Kageyama, A. Kitoh, M.-F. Loutre, O. Marti, U. Merkel, G. Ramstein, P. Valdes, L. Weber, Y. Yu and Y. Zhao 2007 Results of PMIP2 Coupled Simulations of the Mid-Holocene and Last Glacial MaximumPart 2: Feedbacks with Emphasis on the Location of the ITCZ and Midand High Latitudes Heat Budget. Climate of the Past 3:261. Bradley, Richard 1998 The Significance of Monuments Routledge, London. 2002 The Past in Prehistoric Societies Routledge, London. 2003a Forward: Landscape Archaeology and Ethnographic Analogy In Matrerial Culture and Sacred Landscape: The Anthropol ogy of the Siberian Khanty,by Peter Jordan pp. xvixvii. AltaMira, New York. 2003b A Life Less Ordinary: The Ritualizat ion of the Domestic Sphere in Later Prehistoric Europe. Cambridge Archaeological Journal 13:5. Braudel, Fernand 1980 On History. University of Chicago Press, Chicago.

PAGE 386

386 Brenner, Mark, L. W. Keenan, S. J. Miller and C. L. Schelske 1999 Spatial and Temporal Patterns of Sediment and Nutrient Accumulation in Shallow Lakes of the Upper St. Johns River Basin, Florida. Wetlands Ecology and Management 6:221. Brinton, Daniel G. 1859 Notes on the Floridan Peni nsula, its Literary History, Indian Tribes and Antiquities Joseph Sabin, Philadelphia. 1867 Artificial Shell De posits of the United States In Annual Report of the Smithsonian Institution for 1866 pp. 356. Government Printing Office, Washington. 1869 A Guidebook of Florida and the South, for Tourists, Invalids and Emigrants, with a Map of the St. John River George Maclean, Philadelphia. Brooker, M. P. 1985 The Ecological Effects of Channelization. The Geographical Journal 151:63 69. Brookes, A., K. J. Gregory and F. H. Dawson 1983 An Assessment of River Channelization in England and Wales. The Science of the Total Environment 27:97. Brookes, Andrew 1987 River Channel Adjustments Down stream from Channelization Works in England and Wales. Earth Surface Processes and Landforms 12:337. Brookes, Samuel O. 2004 Cultural Complexity in the Middle Archaic of Mississippi. In Signs of Power: The Rise of Cultural Complexity in the Southeast, edited by Jon L. Gibson and Philip J. Carr, pp. 97. The Universi ty of Alabama Press, Tuscaloosa. Brown, A. G. 1996 Floodplain Paleoenvironments In Floodplain Processes, edited by Malcom G. Anderson, Des E. Walling and Paul D. Bates, pp. 95. John Wiley & Sons, New York. 2003 Time, Space and Causality in Floodplain Paleoecology In Alluvial Archaeology in Europe edited by Andy J. Howard, Mark G. Macklin and D. G. Passmore, pp. 15. Swet s & Zeitlinger, Lisse. Brown, A. G. and T. A. Quine 1999 Fluvial Processes and Environmental Change: An Overview In Fluvial Processes and Environmental Change edited by A. G. Brown and T. A. Quine, pp. 1. John Wiley & Sons, New York.

PAGE 387

387 Brown, James A. and Robert K. Vierra 1983 What Happened in the Middle Archai c? Introduction to an Ecological Approach to Koster Site Archaeology In Archaic Hunters and Gatherers in the American Midwest edited by J. L. Philips and J. A. Brown, pp. 165. Academic Press, New York. Brck, Joanna 1999 Ritual and Rationality: Some Problems of Interpretation in European Archaeology. European Journal of Archaeology 2:313. Bullen, Ripley P. 1955a Carved Owl Totem, Deland, Florida. The Florida Anthropologist 8:61. 1955b Stratigraphic Tests at Blu ffton, Volusia County, Florida. The Florida Anthropologist 8:1. 1975 A Guide to the Identification of Florida Projectile Points Revised ed. Kendall Books, Gainesville. Bullen, Ripley P. and William J. Bryant 1965 Three Archaic Sites in the Ocala Na tional Forest. American Studies, Report No. 6, Springfield, Vt. Bullen, Ripley P. and William M. Sackett 1958 Dates on Busycon Gouges at the Bluffton Site, Florida. The Florida Anthropologist 11:111. Bullen, Ripley P. and James B. Stoltman 1972 Fiber-Tempered Pottery in Southeastern United States and Northern Colombia: its Origins, Context, and Significance Florida Anthropological Society, Ft. Lauderdale. Buntman, Barbara 2002 Travels to Otherness: Whose Identity Do We Want to See? In Selfand Other-Images of Hunter-Gatherers edited by Henry Stewart, Alan Barnard and Keiichi Omura, pp. 65. Senri Ethnologic al Studies No. 60. National Museum of Ethnology, Osaka. Bushnell, Francis F. 1960 The Harris Creek Site, Tick Island, Volusia County. The Florida Anthropologist 13:25. Cameron, Catherine M. and Andrew I. Duff 2008 History and Process in Village Forma tion: Context and Contrasts from the Northern Southwest. American Antiquity 73:29.

PAGE 388

388 Casey, Edward S. 1996 How to Get from Space to Place in a Fairly Short Time: Phenomenological Prolegomena. In Senses of Place edited by S. Feld and Keith H. Basso, pp. 13 52. School of American Research, Santa Fe. Ceci, Lynn 1984 Shell Midden Deposits as Coastal Resources. World Archaeology 16:62. Chadbourne, P. A. 1859 Oyster Shell De posit in Damariscotta. Collections of the Maine Historical Society 6:347. Chapin, Mac, Zachary Lamb and Bill Threlkeld 2005 Mapping Indigenous Lands. Annual Review of Anthropology 34. Charles, Douglas K. and Jane E. Buikstra 1983 Archaic Mortuary Sites in the Cent ral Mississippi Drainage: Distribution, Structure, and Behavioral Implications In Archaic Hunters and Gatherers in the American Midwest edited by James L. Phillips and James A. Brown, pp. 117 144. Academic Press, New York. Chatters, James C. 1987 Hunter-Gatherer Adaptati ons and Assemblage Structure. Journal of Anthropological Archaeology 6:336. Chatters, James C. and William C. Prentiss 2005 A Darwinian Ma cro-Evolutionary Perspective on the Development of HunterGatherer Systems in Nort hwestern North America. World Archaeology 37:46. Claassen, Cheryl P. 1996 A Consideration of the Social Organization of the Shell Mound Archaic. In Archaeology of the Mi d-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 235. University of Florida Press, Gainesville. 1998 Shells Cambridge University Press, Cambridge. Clark, John E. 2004 Surrounding the Sacred: Geometry and Design of Early Mound Groups as Meaning and Function In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Philip J. Carr, pp. 162. The University of Alabama Press, Tuscaloosa. Clastres, Pierre 1998 Chronicle of the Guayaki Zone Books, New York.

PAGE 389

389 Clausen, C. J. 1964 The A-356 Site and the Florida Archaic. Unpublished Masters Thesis, Department of Anthropology, University of Florida, Gainesville. Clausen, C. J., A. D. Cohen, Cesare Emiliani, J. A. Holman and J. J. Stipp 1979 Little Salt Springs, Florida: A Unique Underwater Site. Science 203:609. Clench, William J. and Ruth D. Turner 1956 Freshwater Mollusks of Al abama, Georgia and Florida. Bulletin of the Florida State Museum, Biological Sciences 1:108. Colquhoun, Donald J. and Mark J. Brooks 1986 New Evidence from the Southeastern U.S. for Eustatic Components in the Late Holocene Sea Levels. Geoarchaeology 1:275. Connerton, Paul 1989 How Societies Remember Themes in the Social Sciences. Cambridge University Press, Cambridge. Cooke, C. Wythe 1939 Scenery of Florida Interpreted by a Geologist Florida Geological Survey Bulletin No. 17, Tallahassee. Corns, Anthony and Robert Shaw 2009 High Resolution 3-Dimensional Docum entation of Archaeological Monuments & Landscapes Using Airborne LiDAR. Journal of Cultural Heritage 10:e72e77. Cosgrove, Denis E. 1985a Prospect, Perspective and the Evolution of the Landscape Idea. Transactions of the Institute of British Geographers 10:45-62. 1985a Social Formation and Symbolic Landscape The University of Wisconsin Press, Madison. Coulibaly, Paulin and Connely K. Baldwin 2005 Nonstationary Hydrological Time Series Forecasting Using Nonlinear Dynamic Methods. Journal of Hydrology 307:164. Crawford, Jessica 2003 Archaic Effigy Beads: A New Look at Some Old Beads. Unpublished Masters Thesis, Department of Anthropology, University of Mississippi, Oxford. Cronon, William 1983 Changes in the Land: Indians, Colonist s, and the Ecology of New England Hill and Wang, New York.

PAGE 390

390 Crow, P., S. Benham, B. J. Devereux and G. S. Amable 2007 Woodland Vegetation and its Implications for Archaeological Survey Using LiDAR. Forestry 80:241. Cumbaa, Stephen L. 1976 A Reconsideration of Freshwater Shellfish Exploitati on in the Florida Archaic. The Florida Anthropologist 29:49. 1980 Aboriginal Use of Marine Mammals in the Southeastern United States. Southeastern Archaeological Conference Bulletin 17:6. Dall, William H. 1885 Memorandum on the Mounds at Satsuma and Enterprise, Florida. The American Journal of Archaeology and of the History of the Fine Arts 1:184. Daniel, I. Randolph and Michael Wisenbaker 1987 Harney Flats: A Florida Paleo-IndianSite Baywood, Farmingdale, New York. Darby, Philip C., Robert E. Bennetts, St even J. Miller and H. Franklin Percival 2002 Movements of Florida Apple Snails in Relation to Water Levels and Drying Events. Wetlands 22:489. Darby, Philip C., Robert E. Bennetts and H. Franklin Percival 2009 Dry Down Impacts on Apple Snail ( Pomacea paludosa ) Demography: Implications for Wetland Water Management. Wetlands 28:204. Darwin, Charles 1860 Journal of Researches into the Natura l History and Geology of the Countries Visited During the Voyage of H.M.S. Beagle Round the World, under the Command of Capt. Fitz Roy, R.N. J. Murray, London. de Certeau, Michel 1984 The Practice of Everyday Life University of California Press, Berkeley. Deevey, Edward S. and Richard Foster Flint 1957 Postglacial Hypsithermal Interval. Science 125:182. Denevan, William M. 1992 The Pristine Myth: The Land scape of the Americas in 1492. Annals of the Association of American Geographers 82:369. Devereux, B. J., G. S. Amabl e, P. Crow and A. D. Cliff 2005 The Potential of Airborne Lidar for Detection of Archaeological Features under Woodland Canopies. Antiquity 79:648.

PAGE 391

391 Dickel, David N. 2002 Analysis of Mortuary Patterns In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 73. University Press of Florida, Gainesville. Dobres, Marcia-Anne and John E. Robb 2000 Agency in Archaeology: Paradigm or Platitude? In Agency in Archaeology edited by Marcia-Anne Dobres and John E. Robb, pp. 3. Routledge, London. Donders, Timme H., Friederike Wagner, David L. Dilcher and Henk Visscher 2005 Midto Late Holocene El Nio-Southern Oscillation Dynamics Reflected in the Subtropical Terrestrial Realm. Proceedings of the National Academy of Sciences 102:10904. Donnelly, Jeffrey P. and Jonathan D. Woodruff 2007 Intense Hurricane Activity over th e Past 5,000 Years Controlled by El Nio and the West African Monsoon. Nature 447:465(4). Doran, Glen H. 2002a Introduction to Wet Sites and Windover (8BR246) Investigations. In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 1. Univer sity Press of Florida, Gainesville. 2002b The Windover Radiocarbon Chronology In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 59. University Press of Florida, Gainesville. 2002c Windover: Multidisciplinary Investigat ions of an Early Archaic Florida Cemetery. University of Florida Press, Gainesville. Dornan, Jennifer L. 2002 Agency and Archaeology: Pas t, Present, and Future Directions. Journal of Archaeological Method and Theory 9:303. Douglass, Andrew E. 1882 A Find of Ceremonial Axes in a Florida Mound. The American Antiquarian and Oriental Journal 4:100. Drury, G. H. 1976 Discharge Prediction, Present and Former, from Channel Dimensions. Journal of Hydrology 30:219. Dunbar, James S., Michael K. Faught and David S. Webb 1988 Page/Ladson (8je591): An Underwater Paleo-Indian Site in Northwestern Florida. The Florida Anthropologist 41:442.

PAGE 392

392 Dunbar, James S. and Ben Waller 1983 A Distribution Analysis of the Clovis/Suwannee Pale o-Indian Sites of Florida: A Geographic Approach. The Florida Anthropologist 36:18. Dunbar, Robert B. 2000 Climate Variability During the Holocene: An Update In The Way the Wind Blows: Climate, History, and Human Action edited by Roderick J. Mcintosh, Joseph A. Tainter and Susan Keech McIn tosh, pp. 45. Columbia University Press, New York. Dunham, Gary H. 1999 Marking Territory, Making Territory : Burial Mounds in Interior Virginia In Material Symbols : Culture and Economy in Prehistory edited by John E. Robb. Center for Archaeological Investigations Southern Illinois University Carbondale, Carbondale. Dwyer, Peter D. 1985 A Hunt in New Guinea: Some Di fficulties for Optimal Foraging Theory. Man 20:243. 1996 The Invention of Nature In Redefining Nature: Ecology, Culture and Domestication edited by Roy Ellen and Katsuyoshi Fukui, pp. 157. Berg, Oxford. Dye, David H. 1996 Riverine Adaptat ion in the Midsouth In Of Caves and Shell Mounds edited by Kenneth C. Carstens and Patty Jo Watson, pp. 140. University of Alabama Press, Tuscaloosa. Echo-Hawk, Roger C. 2000 Ancient History in the New World: Integrating Oral Traditions and the Archaeological Record in Deep Time. American Antiquity 65:267. Edmonds, Mark R. 1999 Ancestral Geographies of the Neolit hic: Landscape, Monuments and Memory Routledge, New York. Endonino, Jon C. 2003a Hontoon Island Reconnaissance Survey In St. Johns Archaeological Field School 2000-2001: Blue Springs and Hontoon Island State Parks edited by Kenneth E. Sassaman, pp. 91. Technical Report 4. Laboratory of Southeastern Archaeology, Department of An thropology, University of Florida, Gainesville.

PAGE 393

393 2003b Pre-Ceramic Archaic Burial Mounds Along the St. Johns River, Florida Paper presented at the 59th Annual Sout heastern Archaeological Conference, Charlotte, North Carolina. 2007 A Reevaluation of the Gainesville Ocala, and Lake Panasoffkee Quarry Clusters. The Florida Anthropologist 60:77. 2008 The Thornhill Lake Archaeologi cal Research Project: 2005-2008. The Florida Anthropologist 61:149. Environmental Consulting & Technology, Inc. 2004 Environmental Evaluati ons for the Development of Minimum Flows and Levels for the St. Johns River near Deland at State Road 44, Volusia County Special Publication SJ2004SP30. St. Johns River Wa ter Management District, Palatka, Florida. Erlandson, Jon M. 1994 Early Hunter-Gatherers of the California Coast Plenum Press, New York. 2001 The Archaeology of Aquatic Adapta tions: Paradigms for a New Millennium. Journal of Archaeological Research 9:287. Fabian, Johannes 1983 Time and the Other: How An thropology Makes its Object Columbia University Press, New York. Faught, Michael K. 2004 The Underwater Archaeology of Pale olandscapes, Apalachee Bay, Florida. American Antiquity 69:275. Faught, Michael K. and Joseph F. Donoghue 1997 Marine Inundated Archaeological Sites and Paleofluvial Systems: Examples from a Karst-Controlled Continental Shelf Setting in Apalachee Bay. Geoarchaeology: An International Journal 12:417. Faure, Hugues, Robert C. Wa lter and Dougla s R. Grant 2002 The Coastal Oasis: Ice Age Sp rings on Emerged Continental Shelves. Global and Planetary Change 33:47. Fitz-Roy, Robert 1839 Narrative of the Surveying Voyages of His Majesty's Ships Adventure and Beagle between the Years 1826 and 1836, De scribing Their Examination of the Southern Shores of South America, and the Beagle's Circum navigation of the Globe. Proceedings of the Second Expedition, 1831, under the Command of Captain Robert Fitz-Roy, R.N. Voyages of the Adventure and Beagle. Henry Colburn, London. Fogelson, Raymond D. 1989 The Ethnohistory of Events and Nonevents. Ethnohistory 36:133.

PAGE 394

394 Gamble, Clive 1991 An Introduction to the Livi ng Spaces of Mobile Peoples In Ethnoarchaeological Approac hes to Mobile Campsite s: Hunter-Gatherer and Pastoralist Case Studies edited by Clive Gamble and William A. Boismier, pp. 1 23. Ethnoarchaeological Series 1. International Monographs in Prehistory, Ann Arbor. 2007 Origins and Revolutions: Human I dentity in Earliest Prehistory Cambridge University Press, New York. Ganopolski, Andrey, Claudia Kubatzki, Martin Claussen, Victor Brovkin and Vladimir Petoukhov 1998 The Influence of Vegetation-Atmos phere-Ocean Interaction on Climate During the Mid-Holocene. Science 280:1916. Gibson, Jon L. 1996 Poverty Point and Greater Southeastern Prehistory: The Culture That Did Not Fit In Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 288. University of Florida Press, Gainesville. Gibson, Jon L. and Philip J. Carr 2004a Big Mounds, Big Rings, Big Power In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Philip J. Carr, pp. 1. The University of Alabama Press, Tuscaloosa. 2004b Signs of Power: The Rise of Cultural Complexity in the Southeast The University of Alabama Press, Tuscaloosa. Giddens, Anthony 1979 Central Problems in Social Theory: Ac tion, Structure, and Contradiction in Social Analysis University of California Press, Berkeley. 1984 The Constitution of Society: Outline of the Theory of Structuration University of California Press, Berkeley. Gillespie, Susan D. 2001 Personhood, Agency, and Mortuary Rit ual: A Case Study from the Ancient Maya. Journal of Anthropological Archaeology 20:73. 2008 History in Practice: Ritual Deposition at La Venta Complex A In Memory Work: Archaeologies of Material Practices edited by Barbara J. Mills and William H. Walker, pp. 109. School for Adv anced Research Press, Santa Fe. Goggin, John M. 1947 A Preliminary Definition of Arc haeological Areas and Periods in Florida. American Antiquity 13:114.

PAGE 395

395 1952 Space and Time Perspectives in Nort hern St. Johns Archaeology, Florida University Press of Florida, Gainesville. Goman, Michelle and David S. Leigh 2004 Wet Early to Middle Holocene Conditi ons on the Upper Coastal Plain of North Carolina, USA. Quaternary Research 61:256. Gosden, Chris 1994 Social Being and Time Social Archaeology. Blackwell. 1999 Anthropology and Archaeology : Changing Relationships Routledge, New York. Gosden, Chris and Gary Lock 1998 Prehistoric Histories. World Archaeology 30:2. Graham, Martha and Alexandra Roberts 1986 Residentially Constrained Mobility: A Pr eliminary Investigation of Variability in Settlement Organization. Haliksa'i 5:105. Gray, William M. 1990 Strong Association between West Af rican Rainfall and U.S. Landfall of Intense Hurricanes. Science 249:1251. Green, Richard C., David T. Paul and Thomas M. Scott 2007 Text to Accompany Geological Map of the Eastern Portion of the U.S.G.S. Perry 30 X 60 Minute Quadrangl e, Northern Florida (Open-File Map Series 98) Open-File Report 91. Florida Ge ological Survey, Tallahassee. Griffin, John W. 1952 Prehistoric Florida: A Review In Archeology of East ern United States, edited by James B. Griffin, pp. 322. The Un iversity of Chicago Press, Chicago. Grn, Ole 1991 A Method for Reconstruction of Social Structure in Prehistoric Societies and Examples of Practical Application. In Social Space: Human Spatial Behavior in Dwellings and Settlements edited by Ole Grn, Ericka Engelstad and Inge Lindblom, pp. 100.Odense University Press, Odense. Grn, Ole and Oleg Kuznetsov 2003 Ethno-Archaeology among Evenkian Forest Hunters. Preliminary Results and a Different Approach to Reality In Mesolithic on the Move: Papers Presented at the Sixth International Conference on t he Mesolithic in Europe, Stockholm 2000 pp. 216. Oxbow Books, Oxford.

PAGE 396

396 Guccione, Margaret J. 1995 Indirect Response of the Peace River Florida, to Episodic Sea-Level Change. Journal of Coastal Research 11:637. Gut, H. James and Wilfred T. Neill 1953 Bone Artifacts, Resemb ling Projectile Points, fr om Preceramic Sites in Volusia County, Florida. The Florida Anthropologist 6:93. Habu, Junko and Ben Fitzhugh 2002 Introduction-Beyond For aging and Collecting: Evolut ionary Change in HunterGatherer Settlement Systems In Beyond Foraging and Collecting: Evolutionary Change in Hunter-Gatherer Settlement Systems edited by Ben Fitzhugh and Junko Habu, pp. 1. Kluwer Academic/Plenum, New York. Halbwachs, Maurice 1925 Les cadres sociaux de la memoire. Les Travaux de L'Annee Sociologique. F. Alcan, Paris. 1992 On Collective Memory. Edited, trans lated, and with an introduction by Lewis Coser. University of Chicago Press, Chicago. Hamilton, Fran E. 1999 Southeastern Archaic Mounds: Exampl es of Elaboration in a Temporally Fluctuating Environment? Journal of Anthropological Archaeology 18:344. Harmon, James M., Mark P. Leone, Stephen D. Prince and Marcia Snyder 2006 LiDAR for Archaeolog ical Landscape Analysis: A Case Study of Two Eighteenth-Century Maryland Plantation Sites. American Antiquity 71:649. Harris, Julia and Richard McBride 2004 A Review of the Potential Effects of Water Level Fluctuation on Diadromous Fish Populations fo r MFL Determinations Special Publicat ion SJ2004-SP40. St. Johns River Water Management District, Palatka, Florida. Hawkes, Kristen, Kim Hil l and James F. O'Connell 1982 Why Hunters Gather: Optimal Foraging and the Ach of Eastern Paraguay. American Ethnologist 9:379. Hayden, Brian 1994 Competition, Labor, a nd Complex Hunter-Gatherers In Key Issues in HunterGatherer Research edited by E. S. Burch, Jr and L. J. Ellanna, pp. 223. Berg, Oxford. 1995 Pathways to Power: Principles for Creating Socioeconomic Inequalities In Foundations of Social Inequality edited by T. Douglas Price and Gary M. Feinman, pp. 15. Plenum Press, New York.

PAGE 397

397 Heckenberger, Michael J., Afukaka Kuikuro, Urissapa Tabata Kuikuro, J. Christian Russell, Morgan Schmidt, Carlos Fausto and Bruna Franchetto 2003 Amazonia 1492: Pristine Forest or Cultural Parkland? Science 301:1710 1714. Heckenberger, Michael J., J. Christian Ru ssell, Carlos Fausto, Joshua R. Toney, Morgan J. Schmidt, Edithe Pereira, Bruna Franchetto and Afukaka Kuikuro 2008 Pre-Columbian Urbanism, Anthropogen ic Landscapes, and the Future of the Amazon. Science 321:1214. Heide, Gregory 2002 Mapping of Fig Island Shell Ring In The Fig Island Ring Complex (38ch42): Coastal Adaptation and t he Question of Ring Function in the Late Archaic edited by Rebecca Saunders, pp. 67. Grant 45-01-16441, South Carolina Department of Archives and History, Columbia. Heidegger, Martin 1977 Building Dwelling Thinking In Martin Heidegger: Basic Writings edited by David Farrell Krell, pp. 323. Harper & Row, New York. Hill, J. D. 1995 Ritual and Rubbish in the Iron Age of Wessex: A Study of the Formation of a Specific Archaeological Record Bar British Series 242. Tempvs Reparatvm Archaeological and Historical A ssociates Limited, Oxford. Hirsch, Eric 1995 Landscape: Between Place and Space In The Anthropology of Landscape Perspectives on Place and Space edited by Eric Hirsch and Michael O'Hanlon, pp. 1. Clarendon Press, Oxford. Hiscock, Peter and Charles H. Faulkner 2006 Dating the Dreaming? Creation of Myths and Rituals for Mounds Along the Northern Australian Coastline. Cambridge Archaeological Journal 16:209. Hobsbawm, Eric J. 1983 Introduction: Inventing Tradition In The Invention of Tradition edited by Eric J. Hobsbawm and T. O. Ranger, pp. 1. Cambridge University Press, New York. Hodder, Ian 1990 The Domestication of Europe Blackwell, Oxford. Hodder, Ian and Craig Cessford 2004 Daily Practice and Social Memory at atalhyk. American Antiquity 69:17 40.

PAGE 398

398 Holtorf, Cornelius 1998 The Life-Histories of Megaliths in Mecklenburg-Vorpommern (Germany). World Archaeology 30:23. Horton, B. P., W. R. Peltier, S. J. Culver, R. Drummond, S. E. Engelhart, A. C. Kemp, D. Mallinson, E. R. Thieler, S. R. Ri ggs, D. V. Ames and K. H. Thomson 2009 Holocene Sea-Level Changes Along the North Carolina Coastline and Their Implications for Glacial Is ostatic Adjustment Models. Quaternary Science Reviews 28:1725. House of Representatives 1910 Document No. 1111. 60th Congress, 2nd Session. 1930 Document No. 691. 69th Congress, 2nd Session. Howey, Meghan C. L. and John M. O'Shea 2006 Bear's Journey and the St udy of Ritual in Archaeology. American Antiquity 71:261. Huang, He Qing and Gerald C. Nanson 2007 Why Some Alluvial Rivers Develop an Anabranching Pattern. Water Resource Research. 43:1. Hughen, K. A., M. G. L. Baillie, E. Bard, J. W. Beck, C. J. H. Bertrand, P. G. Blackwell, C. E. Buck, G. S. Burr, K. B. Cutler, P. E. Damon, R. L. Edwa rds, R. G. Fairbanks, M. Friedrich, T. P. Guilderson, B. Krom er, G. McCormac, S. Manning, Christopher Bronk Ramsey, P. J. Reimer, R. W. Re imer, S. Remmele, J. R. Southon, M. Stuiver, S. Talamo, F. W. Taylor, J. Van der Plicht and C. E. Weyhenmeyer 2004 Marine04 Marine Radiocarbon Age Calibration, 0 Cal Kyr BP. Radiocarbon 46:1059. Hull, Kathleen L. 2005 Process, Perception, and Practice: Ti me Perspectivism in Yosemite Native Demography. Journal of Anthropological Archaeology 24:354. Humme, Astrid, Roderik Lindenbergh and Chris Sueur 2006 Revealing Celtic Fields from LI DAR Data Using Kriging Based Filtering Proceedings of the ISPRS Commission V Symposium 'Image Engineering and Vision Metrology. Dresden, Germany. Hupp, Cliff R. 2000 Hydrology, Geomorphol ogy and Vegetation of Coas tal Plain Rivers in the South-Eastern USA. Hydrological Processes 14:2991.

PAGE 399

399 Ingold, Tim 1988 Notes on the Foraging Mode of Production In Hunters and Gatherers, Volume 1: History, Evolution, and Social Change edited by Tim Ingold, David Riches and James Woodburn, pp. 269. Berg, Washington. 1992 Culture and the Perc eption of the Environment In Bush Base:Forest Farm: Culture, Environment and Development edited by Elisabeth Croll and David Parkin, pp. 39. Routledge, London. 1995 Building, Dwelling, Living: How An imals and People Make Themselves at Home in the World In Shifting Contexts: Transformations in Anthropological Knowledge edited by Marilyn Strather n, pp. 57. Ro utledge, London. 1996 Hunting and Gathering as Ways of Perceiving the Environment In Redefining Nature: Ecology, Culture and Domestication edited by Roy Ellen and Katsuyoshi Fukui, pp. 117. Berg, Oxford. 1999 On the Social Relation s of the Hunter-Gatherer Band In The Cambridge Encyclopedia of Hunters and Gatherers edited by Richard Lee and Richard Daly, pp. 399. Cambridge University Press, Cambridge. 2000a The Perception of the Environment: Essays on Livelihood, Dwelling & Skill Routledge, New York. 2000b The Poverty of Selectionism. Anthropology Today 16:1. Jackson, Edwin H. 1991 Bottomland Resources and Exploitation Strategies During the Poverty Point Period: Implications of the Archaeobiological Record from the J. W. Copes Site. Geoscience and Man 29:131. Jahn, Otto and Ripley P. Bullen 1978 The Tick Island Site, St. Johns River, Florida Florida Anthropological Society Publications 10, Gainesville. Jefferies, Richard W. 1996 The Emergence of Long-Distance Exchange Networks in the Southeastern United States In Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 222. University Press of Florida, Gainesville. Johansen, Peter G. 2004 Landscape, Monumental Architecture, and Ritual: A Reconsideration of the South Indian Ashmounds. Journal of Anthropological Archaeology 23:309-330. Johnson, Robert E. 2005 Results of Radiocarbon and Faunal A nalysis for 8VO48, the Ft. Florida Midden on the Traderscove Projec t, Volusia County, Florida Manuscript on File, Bureau of Archaeological Research, Division of Historical Resources, Tallahassee.

PAGE 400

400 Jones, Andrew 2003 Technologies of Remembrance: Memo ry, Materiality and Identity in Early Bronze Age Scotland In Archaeologies of Remembrance: Death and Memory in Past Societies edited by Howard Williams, pp. 65. Kluwer Academic/Plenum, New York. Jones, Calvin 1995 McDonald (Paramore) Mound (8Vo162) Excavation in 1977 by Captain Robert C. McDonald of Astor, Florida Manuscript on File, Bureau of Archaeological Research, Division of Historical Resources, Tallahassee. Jordan, Peter 2003 Material Culture and Sacred Landscape: The Anthropology of the Siberian Khanty Rowman & Littlefield, Lanham, Maryland. Joyce, Rosemary A. 2004 Unintended Consequences? Monum entality as a Novel Experience in Formative Mesoamerica. Journal of Archaeological Method and Theory 11:5. Joyce, Rosemary A. and Jeanne Lopiparo 2005 Postscript: Doing Agency in Archaeology. Journal of Archaeological Method and Theory 12:365. Keen, Ian 2006 Constraints on the Development of Enduring Inequalities in Late Holocene Australia. Current Anthropology 47:7. Kehoe, Alice Beck 1998 The Land of Prehistory: A Critical History of American Archaeology Routledge, New York. Kelly, Robert L. 1992 Mobility/Sedentism: Concepts, Ar chaeological Measures, and Effects. Annual Review of Anthropology 21:45. 1995 The Foraging Spectrum Smithsonian Institution, Washington. Kennedy, E. J. 1984 Discharge Ratings at Gaging Stations. Techniques of WaterResources Investigations of the Unit ed States Geological Survey. United States Government Printing Office, Washington. Kent, Susan 1991 The Relationship between Mobili ty Strategies and Site Structure In The Interpretation of Archaeological Spatial Patterning edited by Ellen M. Knoll and T. Douglas Price, pp. 33. Plenum Press, New York.

PAGE 401

401 1992 The Current Forager Controversy: Real Versus Ideal Views of HunterGatherers. Man 27:45. Kidder, Tristam R. 2002 Mapping Poverty Point. American Antiquity 67:89. 2008 Basin-Scale Reconstruction of the Geological Context of Human Settlement: An Example from the Lower Mississippi Valley, USA. Quaternary Science Reviews 27:1255. Kidder, Tristam R., Anthony L. Ortman and Lee J. Arco 2008 Poverty Point and the Archaeology of Singularity. SAA Archaeological Record 8:9. Knapp, A. Bernard and Wendy Ashmore 1999 Archaeological Landscapes Cons tructed, Conceptualized, Ideational In Archaeologies of Landscape: Contemporary Perspectives edited by Wendy Ashmore and A. Bernard Knapp, pp. 1. Blackwell, London. Knighton, David 1998 Fluvial Forms and Processes: A New Perspective Arnold, London. Knox, James C. 1983 Responses of River Systems to Holocene Climates In Late Quaternary Environments of the United States edited by H. E. Wrigh t, pp. 26. Vol. 2, The Holocene. University of Minnesota Press, Minneapolis. 1984 Fluvial Responses to Small Scale Climate Changes In Developments in Applications of Geomorphology edited by J. E. Costa a nd P. J. Fleisher, pp. 318 342. Springer-Verlag, Heidelberg. 1993 Large Increases in Flood Magnitu de in Response to Modest Changes in Climate. Nature 361:430. 2000 Sensitivity of Modern and Ho locene Floods to Climate Change. Quaternary Science Reviews 19:439. Knox, James C. and J. M. Daniels 2002 Watershed Scale and the St ratigraphic Record of Large Floods In Ancient Floods, Modern Hazards: Principles and Applications of Paleoflood Hydrology edited by P. Kyle House, Robert H. Webb, Victor R. Baker and Daniel R. Levish, pp. 237. Water Science and Applic ation Series, Vol. 5. American Geophysical Union, Washington. Koss, J. E., F. G. Ethridge and S. A. Schumm 1994 An Experimental Study of the Ef fects of Base-Level Change on Fluvial, Coastal Plain Systems. Journal of Sedimentary Res earch, Section B-Stratigraphy and Global Studies 64:90.

PAGE 402

402 Koutsoyiannis, Demetris 2006 Nonstationarity Versus Scaling in Hydrology. Journal of Hydrology 324:239 254. Kroening, Sharon E. 2004 Streamflow and Water-Quality Characteri stics at Selected Sites of the St. Johns River in Central Florida, 1933 to 2002 U.S. Geological Survey Scientific Investigations Report 2004 177, Reston, Virginia. Kchler, Susanne 1993 Landscape as Memory: The Mapping of Process and its Representation in a Melanesian Society In Landscape: Politics and Perspectives edited by Barbara Bender, pp. 85. Berg, Oxford. 1999 The Place of Memory In The Art of Forgetting, edited by Adrian Forty and Susanne Kchler, pp. 53. Materia lizing Culture. Berg, New York. LaMoreaux, Heidi K., George A. Brook and John A. Knox 2009 Late Pleistocene and Holocene Envir onments of the Southeastern United States from the Stratigraphy and Pollen Cont ent of a Peat Deposit on the Georgia Coastal Plain. Palaeogeography, Palaeoclimatology, Palaeoecology 280:300 312. Landsea, Christopher W., Roger A. Jr. Pielke Alberto M. Mestas-Nunez and John A. Knaff 1999 Atlantic Basin Hurricanes: Indices of Climatic Changes. Climate Change 42:89. Lane, Ed 1986 Karst in Florida Special Publication No. 29. Florida Geological Survey, Tallahassee. Lascody, Randy 2002 The Onset of the Wet and Dry Seasons in East Central Florida a Subtropical Wet-Dry Climate? Electronic document, http://www.srh.noaa.gov/mlb/?n=wetdryseason accessed 12/08/2009. Latrubesse, Edgardo M. 2008 Patterns of Anabranchi ng Channels: The Ultimate E nd-Member Adjustment of Mega Rivers. Geomorphology 101:130. Le Baron, J. Francis 1884 Prehistoric Remains in Florida In Annual Report of the Smithsonian Institution pp. 771. Government Pr inting Office, Washington.

PAGE 403

403 Lee, Richard B. 1992 Art, Science, or Politics? The Crisis in Hunter-Gatherer Studies. American Anthropologist 94:31. 2006 Commonalities and Diversities in Contemporary Hunter -Gatherers: From Settlement Archaeology to Development Ethnography. Archeological Papers of the American Anthropological Association 16:157. Lee, Richard B. and Richard Daly 1999 Introduction: Foragers and Others In The Cambridge Encyclopedia of Hunters and Gatherers edited by Richard Lee and Richard Daly, pp. 1. Cambridge University Press, Cambridge. Leigh, David S. 2006 Terminal Pleistocene Braided to M eandering Transition in Rivers of the Southeastern USA. Catena 66:155. Leigh, David S. and Thomas P. Feeney 1995 Paleochannels Indicating Wet Climate and Lack of Response to Lower Sea Level, Southeast Georgia. Geology 23:687. Levi-Strauss, Claude 1966 The Savage Mind University of Chicago Press, Chicago. Lightfoot, Kent G., Antoinette Martinez and Ann M. Schiff 1998 Daily Practice and Material Cultur e in Pluralistic Social Settings: An Archaeological Study of Culture Change and Persis tence from Fort Ross, California. American Antiquity 63:199. Liu, Kam-biu and Miriam K. Fearn 2000 Reconstruction of Prehistoric Landfa ll Frequencies of Catastrophic Hurricanes in Northwestern Florida from Lake Sediment Records. Quaternary Research 54. Lubbock, John 1865 Pre-Historic Times, as Illustrated by Ancient Remains and the Manners and Customs of Modern Savages 4th ed., 1878 ed. D. Appleton, New York. Mace, Jane W. 2006 Minimum Levels Determination: St. Johns River at State Road 44 near Deland, Volusia County Technical Publication SJ20065. St. Johns River Water Management District, Palatka, Florida. 2007 Minimum Levels Determination: Lak e Monroe in Volusia and Seminole Counties, Florida Technical Publication SJ2007-2. St. Johns River Water Management District, Palatka, Florida.

PAGE 404

404 Marella, Richard L. 2004 Water Withdrawals, Use, Dishar ge, and Trends in Florida, 2000 Scientific Investigations Report 2004-5151. U.S. Geological Survey, Washington. Marquardt, William 1985 Complexity and Scale in the Study of Fisher-Gatherer-H unters: An Example from the Easter n United States In Prehistoric Hunter-Gat herers: The Emergence of Cultural Complexity edited by T. D. Price and J. A. Brown, pp. 59. Academic Press, Orlando. Marquardt, William H. and Patty Jo Watson 2005a The Green River She ll Mound Archaic: Conclusions In Archaeology of the Middle Green River Region, Kentucky edited by William H. Marquardt and Patty Jo Watson, pp. 629. Institute of Archaeology and Paleoenvironmental Studies, Gainesville. 2005b SMAP Investigations at the Carlston Annis Site, 15BT5 In Archaeology of the Middle Green River, Region, Kentucky edited by William H. Marquardt and Patty Jo Watson, pp. 87. Institute of Archaeology and Paleoenvironmental Studies, Gainesville. May, J. Alan 2005 Ethnographic and Ethnohi storic Suggestions Relevant to the Middle Green River Shell Middens In Archaeology of the Middle Gr een River Region, Kentucky edited by William H. Marquardt and Patty Jo Watson, pp. 71. Insitute of Archaeology and Paleoenvironm ental Studies, Mongraph Number 5, Florida Museum of Natural History, Gainesville. Mayewski, Paul A., Eelco E. Rohling, J. Curt Stager, Wibjrn Karln, Kirk A. Maasch, L. David Meeker, Eric A. Meyerson, Francoi se Gasse, Shirley van Kreveld, Karin Holmgren, Julia Lee-Thorp, Gunhild Rosq vist, Frank Rack, Michael Staubwasser, Ralph R. Schneider and Eric J. Steig 2004 Holocene Climate Variability. Quaternary Research 62:243. McGee, Ray M. and Ryan J. Wheeler 1994 Stratigraphic Excavations at Gr oves' Orange Midden, Lake Monroe, Volusia County, Florida: Meth odology and Results. The Florida Anthropologist 47:333 349. McGuire, Kelly R. and William R. Hildebrandt 2005 Re-Thinking Great Ba sin Foragers: Prestige Hunt ing and Costly Signaling During the Middle Archaic Period. American Antiquity 70:695. Meiklejohn, Christopher, Erik Brinch Petersen and Verner Alexandersen 1998 The Later Mesolithic Population of Sjlland, D enmark, and the Neolithic Transition. In Harvesting the Sea, Farming the Forest: The Emergence of Neolithic Societies in the Baltic Region, edited by Marek Zvelebil, R. Danell and Lucyna Doma ska, pp. 203. Sheffield Academic Press, Sheffield.

PAGE 405

405 Meillassoux, Claude 1972 From Reproduction to Production. Economy and Society 1:93. Meskell, Lynn 2003 Memory's Materiality: Ancestral Presence, Commemorative Practice and Disjunctive Locales In Archaeologies of Memory edited by Ruth M. van Dyke and Susan E. Alcock, pp. 34. Blackwell, Oxford. Milanich, Jerald T. 1994 Archaeology of Precolumbian Florida University Press of Florida, Gainesville. Milanich, Jerald T. and Charles H. Fairbanks 1980 Florida Archaeology Academic Press, New York. Miller, James A. 1997 Hydrogeology of Florida In The Geology of Florida edited by Anthony F. Randazzo and Douglas S. Jones, pp. 69. University of Florida Press, Gainesville. Miller, James J. 1992 Effects of Environmental Change on Late Archaic People of Northeast Florida. The Florida Anthropologist 45:100. 1998 An Environmental History of Northeast Florida University Press of Florida, Gainesville. Miller, James J. and John W. Griffin 1978 Cultural Resource Reconnaissance of Lake Woodruff National Wildlife Refuge. Interagency Archaeological Servic es. Submitted to the U.S. Fish and Wildlife Service. Mills, Barbara J. and William H. Walker 2008 Introduction: Memory, Materi ality, and Depositional Practice In Memory Work: Archaeologies of Material Practices edited by Barbara J. Mills and William H. Walker, pp. 3. School for Adv anced Research Press, Santa Fe. Milner, George R. and Richard W. Jefferies 1998 The Read Archaic Sh ell Midden in Kentucky. Southeastern Archaeology 17:119. Mitchem, Jeffrey M. 1999 Introduction In The East Florida Expeditions of Clarence Bloomfield Moore edited by Jeffrey M. Mitchem, pp. 1. The University of Alabama Press, Tuscaloosa.

PAGE 406

406 Mohan, Geoffrey 1990 St. Johns In The Rivers of Florida edited by Del Marth and Marty Marth, pp. 45. Pineapple Press, Sarasota, Florida. Moore, Clarence B. 1892 Certain Shell Heaps of the St. Jo hn's River, Florida, Hitherto Unexplored. (First Paper). The American Naturalist 26:912. 1893a Certain Shell Heaps of the St. J ohn's River, Florida, Hitherto Unexplored. (Second Paper). The American Naturalist 27:8, 113. 1893b Certain Shell Heaps of the St. J ohn's River, Florida, Hitherto Unexplored. (Third Paper). The American Naturalist 27:605. 1893c Certain Shell Heaps of the St. John's River, Florida, Hitherto Unexplored. (Fourth Paper). The American Naturalist 27:708. 1894a Certain Sand Mounds of the S t. John's River, Florida. Part I. Journal of the Academy of Natural Sciences of Philadelphia 10:4. 1894b Certain Sand Mounds of the St. John's River, Florida. Part II. Journal of the Academy of Natural Sciences of Philadelphia 10:12946. 1894c Certain Shell Heaps of the St. John's River, Florida, Hitherto Unexplored. (Fifth Paper). American Naturalist 28:15. 1898 Recent Acquisitions: A "Banner-Stone" Of Shell. Journal of the Academy of Natural Sciences of Philadelphia 11:1878. Morey, Darcy F. and George M. Crothers 1998 Clearing up Clouded Waters: Palaeoe nvironmental Analysis of Freshwater Mussel Assemblages from the Green River Shell Middens, Western Kentucky. Journal of Archaeological Science 25:907. Morey, Darcy F., George M. Crothers, Julie K. Stein, James P. Fenton and Nicholas P. Herrmann 2002 The Fluvial and Geomorphic Context of Indian Knoll, an Archaic Shell Midden in West-Central Kentucky. Geoarchaeology: An International Journal 17:521. Morgan, Lewis Henry 1877 Ancient Society Henry Holt and Company, New Brunswick, New Jersey. Morlot, A. 1861 General Views on Archaeology Annual Report of the Sm ithsonian Institution for 1860 pp. 284. Morphy, Howard 1995 Landscape and the Reproduction of the Ancestral Past In The Anthropology of Landscape: Perspectives on Place and Space, edited by Eric Hirsch and Michael O'Hanlon, pp. 184. Clarendon Press, Oxford.

PAGE 407

407 Morrison, Michael 2003 Old Boundaries and New Horizons: The Weipa Shell Mounds Reconsidered. Archaeology in Oceania 38:1. Moy, Christopher M., Geoffrey O. Seltzer, Donald T. Rodbell and David M. Anderson 2002 Variability of El Nio/Southern Oscill ation Activity at Millennial Timescales During the Holocene Epoch. Nature 420:162. Munn, Nancy 1996 Excluded Spaces: The Figure in the Australian Abor iginal Landscape. Critical Inquiry 22:446-465. Nabokov, Peter 2002 A Forest of Time: American Indian Ways of History Cambridge University Press, New York. Nanson, Gerald C. and David Knighton 1996 Anabranching Rivers: Their Cause, Character and Classification. Earth Surface Processes and Landforms 21:217. Neill, Wilfred T. 1954 Artifacts from the Bluffton Midden, Volusia County, Florida. The Florida Anthropologist 7:11. 1964 Trilisa Pond, an Early Site in Marion County, Florida. The Florida Anthropologist 17. Neill, Wilfred T., H. James Gut and Pierce Brodkorb 1956 Animal Remains from Four Preceramic Sites in Florida. American Antiquity 21:383. Nelson, Neils C. 1918 Chronology in Florida. Anthropological Papers of the American Museum of Natural History 22:75. Neubauer, Wolfgang 2001 Images of the Invisible-Prospecti on Methods for the Documentation of Threatened Archaeological Sites. Naturwissenschaften 88:13. Newsom, Lee A. 1987 Analysis of Botanical Remains from Hontoon Island (8VO202), Florida: 19801985 Excavations. The Florida Anthropologist 40:47. 1994 Archaeobotanical Data from Gr oves' Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:404.

PAGE 408

408 2002 The Paleoethnobotany of the Archaic Mortuary Pond In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 191. University Press of Florida, Gainesville. Newsom, Lee A. and Barbara A. Purdy 1990 Florida Canoes: A Mariti me Heritage from the Past. The Florida Anthropologist 43:164. Nicholas, George P. 1998 Wetlands and Hunter-Gather ers: A Global Perspective. Current Anthropology 39:720. Nodine, Bruce K. 1987 Refuse Disposal and Midden Formation at Hontoon Island. The Florida Anthropologist 40:19. Noll, Steven and David Tegeder 2009 Ditch of Dreams: The Cross Flori da Barge Canal and the Struggle for Florida's Future The Florida History and Culture Series. University Press of Florida, Gainesville. Oetelaar, Gerald A and D. Joy Oetelaar 2008 Indigenous Stewardship: Lessons fr om Yesterday for the Parks of Tomorrow C ontributed paper for the Canadian Par ks for Tomorrow: 40th Anniversary Conference University of Calgary, Calgary. Olick, Jeffrey K. and Joyce Robbins 1998 Social Memory Studies: From "Collective Memory" to the Historical Sociology of Mnemonic Practices. Annual Review of Sociology 24:105. Ortner, Sherry B. 1984 Theory in Anthropol ogy since the Sixties. Comparative Studies in Society and History 26:126. Otvos, Ervin G. 2004 Holocene Gulf Levels: Recognition Issues and an Updated Sea-Level Curve. Journal of Coastal Research 20:680. 2005 Validity of Sea-Level Indicators: A Comment On "A New Depositional Model for the Buried 4000 Yr BP New Orleans Ba rrier: Implications for Sea-Level Fluctuations and Onshore Transport from a Nearshore Shelf Source" By F.W. Stapor and G.W. Stone [Mari ne Geology 204 (2004) 215]. Marine Geology 217:177. Owoc, Mary Ann 2005 From the Ground Up: A gency, Practice, and Community in the Southwestern British Bronze Age. Journal of Archaeological Method and Theory 12:257.

PAGE 409

409 Pandion Systems, Inc. 2003 Carrying Capacity Study of Silver Glen Spring and Run DEP Contract Number SL 982. Submitted to the Fl orida Department of Environmental Protection. Pauketat, Timothy R. 2000 The Tragedy of the Commoners. In Agency in Archaeology edited by MarciaAnne Dobres and John E. Robb, pp. 113. Routledge, London. 2001a A New Tradition in Archaeology In The Archaeology of Traditions: Agency and History before and after Columbus edited by Timothy R. Pauketat, pp. 1. University Press of Florida, Gainesville. 2001b Practice and History in Archaeology: An Emerging Paradigm. Anthropological Theory 1:73. 2003 Materiality and the Immaterial in Historical-Processual Archaeology. In Essential Tensions in Arc haeological Method and Theory edited by T. L. VanPool and C. S. VanPool, pp. 41. The Universi ty of Utah Press, Salt Lake City. 2008 Founders' Cults and the Archaeology of Wa-Kan-Da. In Memory Work: Archaeologies of Material Practices edited by Barbara J. Mills and William H. Walker, pp. 61. School for Advanced Research Press, Santa Fe. Pauketat, Timothy R. and Susan M. Alt 2003 Mounds, Memory, and Contested Mississippian History In Archaeologies of Memory edited by Ruth M. Van Dyke pp. 151. Blackwell, Malden, Massachusetts. 2005 Agency in a Postmold? Physicality and the Archaeology of Culture-Making. Journal of Archaeological Method and Theory 12:213. Pauketat, Timothy R. and Thomas E. Emerson 2008 Star Performances and Cosmic Clutter. Cambridge Archaeological Journal 18:78. Peacock, Evan 2002 Shellfish Use During the W oodland Period in the Middle South In The Woodland Southeast edited by David G. Anderson, pp. 444. The University of Alabama Press, Tuscaloosa. Penders, Thomas E. 2002 Bone, Antler, Dentar y, and Lithic Artifacts In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 97. University of Flori da Press, Gainesville. Peters, Joris and Klaus Schmidt 2004 Animals in the Symbolic World of Pre-Pottery Neolithic Gbekli Tepe, SouthEastern Turkey: A Preliminary Assessment. Anthropozoologica 39:179.

PAGE 410

410 Phelps, G. G., Stephen J. Walsh, Rober t M. Gerwig and William B. Tate 2006 Characterization of the Hydrology, Water Chemistry, and Aquatic Communities of Selected Springs in t he St. Johns River Water Management District, Florida, 2004 U.S. Geological Survey Open-File Report 2006-1107, Reston, Virginia. Piatek, Bruce 1984 Preliminary Assessment of The "Enterprize Midden" Site: 8VO55 Enterprize, Florida. Submitted to the Historic St. Augustine Preservation Board. Piatek, Bruce John 1994 The Tomoka Mound Complex in Northeast Florida. Southeastern Archaeology 13:109. Pinkoski, Marc and Michael Asch 2004 Anthropology and Indigenous Rights in Canada and the United States: Implications in Steward's Theoretical Project In Hunter-Gatherers in History, Archaeology, and Anthropology edited by Alan Barnard, pp. 187. Berg, New York. Pluciennik, Mark 2001 Archaeology, Anthropology, and Subsistence. Journal of the Royal Anthropological Institute 7:741. 2002 The Invention of Hunter-Gather ers in Seventeenth-Century Europe. Archaeological Dialogues 9:98. Politis, Gustavo G. 2007 Nukak: Ethnoarcheology of an Amazonian People Left Coast Press, Walnut Creek, California. Pollard, Joshua 1999 'These Places Have Their Moments' : Thoughts on Settlement Practices in the British Neolithic In Making Places in the Prehistoric World: Themes in Settlement Archaeology edited by Joanna Brck and Melissa Goodman, pp. 76. University College London Press, London. 2008 Deposition and Material Agency in t he Early Neolithic of Southern Britain In Memory Work: Archaeologies of Material Practices edited by Barbara J. Mills and William H. Walker, pp. 41. School for Advanced Research Press, Santa Fe. Porter, William W. 1974 William Bartram's "Road to Xanadu" on Modern Florida Maps. Proceedings of the American Philos ophical Society 118:514.

PAGE 411

411 Price, T. D. and J. A. Brown 1985 Aspects of Hunt er-Gatherer Complexity In Prehistoric Hunter-Gatherers: The Emergence of Cultural Complexity edited by T. D. Price and J. A. Brown, pp. 3 20. Academic Press, Orlando, Fla. Purdy, Barbara A. 1975 The Senator Edwards Chipped Stone Workshop Site (8-Mr-122), Marion County, Florida: A Preliminar y Report of Investigations. The Florida Anthropologist 28:178. 1981 Investigations into t he Use of Chert Outcrops by Prehistoric Floridians: The Container Corporation of America Site. The Florida Anthropologist 34:90. 1987 Investigations at Hontoon Island (8-Vo-202), an Archaeological Wetsite in Volusia County, Florida: An Overview and Chronology. The Florida Anthropologist 40:4. 1991 The Art and Archaeology of Florida's Wetlands CRC Press, Boca Raton. 1994a The Chipped Stone Tool Industr y at Groves Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:390. 1994b Excavations in Water-Saturated Depo sits at Lake Monroe, Volusia County, Florida: An Overview. The Florida Anthropologist 47:326. Quitmyer, Irvy R. 2001 Zooarchaeological Analyses In Phase III Mitigative Excavations at Lake Monroe Outlet Midden (8VO53), Volusia County, Florida, pp. 1. Archaeological Consultant s, Inc. and Janus Resear ch. Submitted to U.S. Department of Transportation Federal Highway Administration and Florida Department of Transportation District Five. Randall, Asa R. 2007 St. Johns Archaeological Field Sc hool 2005: Hontoon Island State Park Technical Report 7. Laboratory of S outheastern Archaeology, Department of Anthropology, University of Florida. 2008 Archaic Shell Mounds of the St. Johns River, Florida. SAA Archaeological Record 8:13. Randall, Asa R. and Peter R. Hallman 2005 Reconnaissance Survey In St. Johns Archaeologic al Field School 20032004: Hontoon Island State Park edited by Asa R. Ra ndall and Kenneth E. Sassaman, pp. 155. Technical Repor t 6. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Randall, Asa R. and Kenneth E. Sassaman 2005 St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park Technical Report 6. Laboratory of S outheastern Archaeology, Department of Anthropology, University of Florida, Gainesville.

PAGE 412

412 Randall, Asa R. and Neill J. Wallis 2007 Reconnaissance Survey In St. Johns Archaeoligic al Field School 2005: Hontoon Island State Park edited by Asa R. Randall, pp. 89. Technical Report 7. Laboratory of Southeastern Ar chaeology, Department of Anthropology, the University of Florida, Gainesville. Randazzo, Anthony F. 1997 The Sedimentary Platform of Florida: Mesozoic to Cenozoic In The Geology of Florida edited by Anthony F. Randazzo and Douglas S. Jones, pp. 39. University Press of Florida, Gainesville. Rao, Donthamsetti V. 1986 Magnitude and Frequency of Flood Di scharges in Northeast Florida Technical Publication SJ86-2. St. Johns River Water Management District, Palatka, Florida. 1988 Rainfall Analysis for Northeast Florida, Part VI: 24-Hour to 96-Hour Maximum Rainfall for Return Periods 10 Years, 25 Years, and 100 Years Technical Publication SJ88-3. St. Johns River Wate r Management District, Palatka, Florida. Redmond, Kelly T., Yehouda Enzel, P. Kyle House and Franco Biondi 2002 Climate Variability and Flood Frequency at Decadal to Millennial Time Scales In Ancient Floods, Modern Hazards: Princi ples and Applicatio ns of Paleoflood Hydrology edited by P. Kyle House, Robert H. Webb, Victor R. Baker and Daniel R. Levish, pp. 21. Water Science and Ap plication Series, Vol. 5. American Geophysical Union, Washington DC. Reimer, P. J., M. G. L. Baillie, E. Bard, A. Bayliss, J. W. Be ck, C. J. H. Be rtrand, P. G. Blackwell, C. E. Buck, G. S. Burr, K. B. Cu tler, P. E. Damon, R. L. Edwards, R. G. Fairbanks, M. Friedrich, T. P. Guilderson, A. G. Hogg, K. A. Hughen, B. Kromer, G. McCormac, S. Manning, Christopher Bronk Ramsey, R. W. Reimer, S. Remmele, J. R. Southon, M. Stuiver, S. Talamo, F. W. Taylor, J. van der Plicht and C. E. Weyhenmeyer 2004 Intcal04 Terrestrial Radiocarbon Age Calibration, 0 Cal Kyr BP. Radiocarbon 46:1029. Reimer, P. J., G. McCormac, J. M oore, F. McCormick and E. V. Murray 2002 Marine Radiocarbon Reservoir Correcti ons for the Midto Late Holocene in the Eastern Subpolar North Atlantic. The Holocene 12:129. Reimer, Paula J., Mike G. L. Baillie, Edouar d Bard, J. Warren Beck, Paul G. Blackwell, Caitlin E. Buck, George S. Burr, R. Lawrence Edwa rds, Michael Friedrich, Thomas P. Guilderson, Alan G. Hogg, Konrad A. Hughen, Be rnd Kromer, Gerry McCormac, Sturt Manning, Ron W. Reim er, John R. Southon, Minze Stuiver, Johannes van der Plicht and C onstanze E. Weyhenmeyer

PAGE 413

413 2006 Comment on Radiocarbon Calibration Curve Spanning 0 to 50,000 Years BP Based on Paired 230th/234u/238u and 14c Da tes on Pristine Corals by R.G. Fairbanks et al. (Quaternary Science Re views 24 (2005) 1781) Star, Open and Extending the Radiocar bon Calibration Beyond 26,000 Years before Present Using Fossil Corals by T. C. Chiu et al. (Quaternary Science Reviews 24 (2005) 1797). Quaternary Science Reviews 25:855. Richards, Colin and Julian Thomas 1984 Ritual Activity and Structured Deposition in Later Neolithic Wessex In Neolithic Studies: A Review of Some Current Research pp. 189. BAR Monographs 133, Oxford. Robb, John E. 1998 The Archaeology of Symbols. Annual Review of Anthropology 27:329. 2008 Introduction: Time and Change in Archaeological Interpretation. Cambridge Archaeological Journal 18:57. Robin, Cynthia 2002 Outside of Houses: The Practices of Everyday Life at Chan Noohol, Belize. Journal of Social Archaeology 2:245. Robison, C. Price 2004 Middle St. Johns River Minimum Fl ows and Levels Hydrologic Methods Report Technical Publication SJ2004-2. St. Johns River Water Management District, Palatka, Florida. Rodman, Margaret C. 1992 Empowering Place: Mult ilocality and Multivocality. American Ethnologist 94:640. Rowley-Conwy, Peter 2001 Time, Change and the Archaeology of H unter-Gatherers: How Original Is the 'Original Affluent Society' In Hunter-Gatherers: An Inte rdisciplinary Perspective, edited by Catherine Pante r-Brick, Robert H. Layton and Peter Rowley-Conwy, pp. 39. Cambridge University Press, Cambridge. Ruiz-Ballesteros, Esteban, Jos Maria Valcuende, Victoria Quintero, Jos Antonio Cortes and Elena Rubio 2009 Naturalizing the Environment: Perc eptual Frames, Senses and Resistance. Journal of Material Culture 14:147. Rumsey, Alan 1994 The Dreaming, Human Agency and Inscriptive Practice. Oceania 65:116.

PAGE 414

414 Russo, Michael 1992 Chronologies and Cultures of the St. Marys Region of Northeast Florida and Southeast Georgia. The Florida Anthropologist 45:107. 1993 The Timucuan Ecological and Historic Preserve Phase III Final Report SEAC Accession Number 899. Florida Museum of Natural History, Gainesville. 1994a Archaic Mounds in the Southeast. Southeastern Archaeology 13:89. 1994b Why We Don't Believe in Archaic Ceremonial Mounds and Why We Should: The Case from Florida. Southeastern Archaeology 13:93. 1996a Southeastern Archaic Mounds In Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassam an and David G. Anderson, pp. 259 287. University Press of Florida, Gainesville. 1996b Southeastern Mid-Holocene Coastal Settlements In The Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 177. University Press of Florida, Gainesville. 2004 Measuring Shell Rin gs for Social Inequality In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Philip J Carr, pp. 26. The University of Alabama Press, Tuscaloosa. 2008 Late Archaic Shell Rings and Society in the Southeast U.S. SAA Archaeological Record 8:18. Russo, Michael and Gregory Heide 2001 Shell Rings of the Southeast US. Antiquity 75:491. 2002 The Joseph Reed Shell Ring. The Florida Anthropologist 55:67. Russo, Michael, Gregory Heide and Vicki Rolland 2002 The Guana Shell Ring. The Southeast Archaeological Center. Submitted to the Florida Department of State Di vision of Historical Resources. Russo, Michael, Barbara Purdy, L ee A. Newsom and Ray M. McGee 1992 A Reinterpretation of Late Archaic Adaptations in Central-East Florida: Groves Orange Midden (8VO2601). Southeastern Archaeology 11:95. Russo, Michael and Dana Ste. Claire 1992 Tomoka Stone: Archaic Period C oastal Settlement in East Florida. The Florida Anthropologist 45:336. Sahlins, Marshall 1985 Islands of History University of Chicago Press, Chicago. Santos-Granero, Fernando 1998 Writing History into the Landscape: Space, Myth, and Ritual in Contemporary Amazonia. American Ethnologist 25:128.

PAGE 415

415 Sassaman, Kenneth E. 1995 The Cultural Diversity of Interactions among Mid-Holocene Societies of the American Southeast In Native American Interactions : Multiscalar Analyses and Interpretations in the Eastern Woodlands edited by Michael S Nassaney and Kenneth E. Sassaman, pp. 174. Universi ty of Tennessee Press, Knoxville. 1996 Technological Innovations in Economic and Social Contexts In Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 57. University of Florida Press, Gainesville. 2001 Hunter-Gatherers and Traditions of Resistance In The Archaeology of Traditions: Agency and History before and after Columbus edited by T. R. Pauketat, pp. 218. University of Florida Press, Gainesville. 2003a St. Johns Archaeological Field Sc hool 2000-2001: Bl ue Spring and Hontoon Island State Parks Technical Report 4. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2003b Crescent Lake Archaeological Survey 2002: Putnam and Flagler Counties, Florida. Technical Report 5. Laboratory of Southeastern Ar chaeology, Department of Anthropology, the University of Florida, Gainesville. 2003c New AMS Dates on Orange Fiber-T empered Pottery from the Middle St. Johns Valley and Their Implications for Culture History in Northeast Florida. The Florida Anthropologist 56:5. 2004a Complex Hunter-Gat herers in Evolution and History: A North American Perspective. Journal of Archaeological Research 12:227. 2004b Common Origins and Divergent Histories in the Early Pottery Traditions of the American Southeast In Early Pottery: Technolog y, Function, Style and Interaction in the Lower Southeast edited by Rebecca S aunders and Christopher T. Hays, pp. 23. The University of Alabama Press, Tuscaloosa. 2005a Hontoon Dead Creek Mound (8VO214) In St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park edited by Asa R. Randall and Kenneth E. Sassaman, pp. 83. Technical Report 6. Laboratory of Southeastern Archaeology, Department of An thropology, University of Florida, Gainesville. 2005b Poverty Point as Structure, Event, Process. Journal of Archaeological Method and Theory 12:335. 2010 The Eastern Archaic, Historicized AltaMira Press, Lanham, Maryland. Sassaman, Kenneth E., Meggan E. Blessing and Asa R. Randall 2006 Stallings Island Revisited: New Evidence for Occupational History, Community Pattern, and Subsistence Technology. American Antiquity 71:539 565. Sassaman, Kenneth E. and Michael J. Heckenberger 2004 Crossing the Symbolic Rubicon in the Southeast In Signs of Power: The Rise of Cultural Comp lexity in the Southeast edited by Jon L. Gibson and Philip J. Carr, pp. 214. The University of Alabama Press, Tuscaloosa.

PAGE 416

416 Sassaman, Kenneth E. and Asa R. Randall 2007 The Cultural History of Banner stones in the Savannah River Valley. Southeastern Archaeology 26:196. Sassaman, Kenneth E., Asa R. Randall, Meggan E. Blessing and Peter R. Hallman 2005 Hontoon Island North (8VO202) In St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park, edited by Asa R. Randall and Kenneth E. Sassaman, pp. 27. Technical Repor t 6. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Sassaman, Kenneth E., Asa R. Randall, Zacka ry Gilmore, Meggan E. Blessing and Neill J. Wallis 2010 St. Johns Archaeological Field School 2007-2009: Silver Glen Run Technical Report 10, Laboratory of Sout heastern Archaeology, Department of Anthropology, the University of Florida, Gainesville. Sassaman, Kenneth E., J. Chri stian Russell and John Endonino 2000 St. Johns Archaeological Project P hase I: A GIS Approach to Regional Preservation Planning in Northeast Florida. Technical Report 3, Laboratory of Southeastern Archeology, Department of An thropology, University of Florida, Gainesville. Sassaman, Kenneth E., J. Christ ian Russell and Jon C. Endonino 2003 Archaeological Cons equences of Urban Land-Use Conversion in Northeast Florida, 1970-1995. Southeastern Archaeology 22:196. Saunders, Joe 2004 Are We Fixing to Make the Same Mistake Again? In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Clark C. Gibson and Philip J. Carr, pp. 146. The University of Alabama Press, Tuscaloosa. Saunders, Joe and Thurman Allen 1995 The Archaic Period. Louisiana Archaeology 22:1. Saunders, Joe, Thurm an Allen, Reca Jones and Gloria Swoveland 2000 Caney Mounds (16CT5). Louisiana Archaeological Society Newsletter 27:14 21. Saunders, Joe, Rolfe D. M andel, C. Garth Sampson, Char les M. Allen, E. Thurman Allen, Daniel A. Bush, James K. Feather Kristen J. Gremillion, C. T. Hallmark, Edwin H. Jackson, Jay K. Johnson, Reca Jones, Roger T. Saucier, Gary L. Stringer and Malcolm F. Vidrine 2005 Watson Brake, a Middle Archaic Mound Complex in Northeast Louisiana. American Antiquity 70:631.

PAGE 417

417 Saunders, Rebecca 2004 Spatial Variation in Orange Cult ure Pottery: Interaction and Function In Early Pottery: Technology, Function, Style and Interaction in the Lower Southeast edited by Rebecca Saunders and Christophe r T. Hays, pp. 40. University of Alabama Press, Tuscaloosa. Sawada, M., A. E. Viau, G. Vettore tti, W. R. Peltier and K. Gajewski 2004 Comparison of North-American Po llen-Based Temperature and Global LakeStatus with CCCMA AGCM2 Output at 6 Ka. Quaternary Science Reviews 23:225. Scham, Sandra 2008 The World's First Temple. Archaeology 61:22. Schmidt, Klaus 2000 Gbekli Tepe, Southeastern Turkey. A Preliminary Report on the 1995-1999 Excavations. Palorient 26:45. Schmidt, Walter 1997 Geomorphology and Ph ysiography of Florida In The Geology of Florida edited by Anthony F. Randazzo and Dougla s S. Jones, pp. 1. University of Florida Press, Gainesville. Schulderein, Joseph 1996 Geoarchaeology and the Mid-Holoc ene Landscape History of the Greater Southeast In Archaeology of the Mi d-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 3. University of Florida Press, Gainesville. Schumm, S. A. 1967 Paleohydrology: Application of Moder n Hydrologic Data to Problems of the Ancient Past In International Hydrology Symposium, Vol. 1 pp. 158. Colorado State University, Fort Collins, Co. 1993 River Response to Base-Level Change: Implications for Sequence Stratigraphy. Journal of Geology 101:279. Scott, Thomas M. 1997 Miocene to Holocene History of Florida In The Geology of Florida edited by Anthony F. Randazzo and Douglas S. J ones, pp. 57. University Press of Florida, Gainesville. 2001 Text to Accompany the Geologic Map of Florida Open-File Report 80. Florida Geological Survey, Tallahassee. Scott, Thomas M., Guy H. Means, Rebecca P. Meegan, Ryan C. Means, Sam B. Upchurch, R. E. Copeland, James Jones, Tina Roberts and Alan Willet 2004 Springs of Florida Bulletin No. 66. Florida Geological Survey, Tallahassee.

PAGE 418

418 Scudder, Sylvia 2001 Archaeopedological Analyses Phase III Mitigative Excavations at Lake Monroe Outlet Midden (8VO53), Volusi a County, Florida. Archaeological Consultants, Inc. and Janus Research Submitted to U.S. Department of Transportation Federal Highway Admini stration and Florida Department of Transportation District Five. Sears, William H. 1960 The Bluffton Burial Mound. The Florida Anthropologist 13:55. Senate 1945 Document No. 208. 79th Congress, 2nd Session. Shennan, Stephen 1983 Monuments: An Example of Arc haeologists' Approach to the Massively Material. RAIN 59:9. 1993 After Social Evolution: A New Archaeological Agenda In Archaeological Theory: Who Sets the Agenda? edited by Norman Yoffee and Andrew Sherratt, pp. 53. Cambridge University Press, Cambridge. Sherratt, Andrew 1990 The Genesis of Megaliths : Monumentality, Ethnicity and Social Complexity in Neolithic North-West Europe. World Archaeology 22:147. Shils, Edward 1981 Tradition. University of Chicago Press, Chicago. Shott, Michael J. 2001 Crises in Forager Studies Ethnographic and Archaeological. Reviews in Anthropology 29:211. Siddall, M., E. J. Rohling, A. Almogi-Labin, Ch. Hemleben, D. Meischner, I. Schmelzer and D. A. Smeed 2003 Sea-Level Fluctuations During the Last Glacial Cycle. Nature 423:853. Sigler-Eisenberg, Brenda, Ann S. Cordell, Richard Estabrook, Elizabeth Horvath, Lee A. Newsom and Michael Russo 1985 Archaeological Site Types, Distribution, and Preservation within the Upper St. Johns River Basin, Florida Florida State Museum Miscellaneous Project and Report Series, Number 27. Department of Anthropology, Florida State Museum, University of Florida, Gainesville, Florida.

PAGE 419

419 Simms, Alexander R., Kurt Lambeck, Anthony Purcell, John B. Anderson and Antonio B. Rodriguez 2007 Sea-Level History of the Gulf of Mexico since the Last Glacial Maximum with Implications for the Melting Hi story of the Laurentide Ice Sheet. Quaternary Science Reviews 26:920. Slingerland, Rudy and Norman D. Smith 2004 River Avulsions and Their Deposits. Annual Review of Earth and Planetary Sciences 32:257. Smith, Bruce D. 1986 The Archaeology of t he Southeastern United States: From Dalton to de Soto, 10,500 B.P. Advances in World Archaeology 5:1. Soja, Edward W. 1985 The Spatiality of Social Life: Towards a Transformative Retheorisation In Social Relations and Spatial Structures edited by Derek Gregory and John Urry, pp. 90. St. Martin's Press, New York. Solway, Jacqueline S and Richard B. Lee 1990 Foragers, Genuine or Spurious? Situating the Kalahari San in History. Current Anthropology 31:109. Stahl, Ann B. 2008 Dogs, Pythons, Pots and Beads: T he Dynamics of Shrines and Sacrificial Practices in Banda, Ghana, 1400 CE In Memory Work: Archaeologies of Material Practices edited by Barbara J. Mills and William H. Walker, pp. 159. School for Advanced Research Press, Santa Fe. Starkel, Leszek 2002 Change in the Frequency of Extreme Events as the Indicator of Climatic Change in the Holocene (in Fluvial Systems). Quaternary International 91:25. Ste. Claire, Dana 1987 The Development of Thermal Alterati on Technologies in Flor ida: Implications for the Study of Prehistoric Adaptations. The Florida Anthropologist 40:203. 1990 The Archaic in East Florida: Ar chaeological Evidence from Early Coastal Adaptations. The Florida Anthropologist 43:189. Stein, Julie K. 1992 The Analysis of Shell Middens. In Deciphering a Shell Midden edited by Julie K. Stein, pp. 1. Academic Press, New York. Steponaitis, Vincas P. 1986 Prehistoric Archaeology in t he Southeastern United States, 1970. Annual Review of Anthropology 15:363.

PAGE 420

420 Sterling, Maurice and Charles A. Padera 1998 The Upper St. Johns River Basin Proj ect: The Environmental Transformation of a Public Flood Control Project Professional Paper SJ 98-PP1. St. Johns River Water Management District, Palatka, Florida. Steward, Julian H. 1933 Ethnography of Owen's Valley Paiute. University of California Publications in American Archaeology and Ethnology 33:233. 1938 Basin-Plateau Aboriginal Sociopolitical Groups Bureau of American Ethnology, Bulletin 120. Smiths onian Institution, Washington. 1955 Theory of Culture Change: The Met hodology of Multilinear Evolution University of Illinois Press, Urbana. Stout, Scott A. and William Spackman 2002 Paleoecology Interpreted by Peat Pretology and Chemistry In Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery edited by Glen H. Doran, pp. 227. University Press of Florida, Gainesville. Strassburg, Jimmy 2003 Rituals at the Meso 2000 Conference and the Mesolithic-Neolithic Terminological Breakdown In Mesolithic on the Move: Papers Presented at the Sixth International Conference on the Mesolithic in Europe, Stockholm 2000 edited by Lars Larsson, Hans Kindgren, Kjel Knutsson, David Loeffler and Agneta kerlund, pp. 542. Ox bow Books, Oxford. Stuiver, M. and P. J. Reimer 1993 Calib 5.0. Electronic document, http://radiocarbon.pa.qub.ac.uk/calib/calib.html, accessed 1/15/2010. Telford, R. J., E. H eegaard and H. J.B. Birks 2004 The Intercept Is a Poor Estima te of a Calibrated Radiocarbon Age. The Holocene 14:296. Testart, Alain 1982 The Significance of Food Storage among Hunter-Gatherers: Residence Patterns, Population Densitie s, and Social Inequalities. Current Anthropology 23:523. Thomas, Cyrus 1894 Report on the Mound Explorations of the Bureau of Ethnology Annual Report 12. Bureau of Ethnology, Washington. Thomas, Julian 1991 Rethinking the Neolithic Cambridge University Press, Cambridge. 1993 The Politics of Vision and the Archaeologies of Landscape In Landscape: Politics and Perspectives edited by Barbara Bender, pp. 19. Berg, Oxford.

PAGE 421

421 1996 Time, Culture and Identity: An Interpretive Archaeology Routledge, London. Thomas, Michael F. 2001 Landscape Sensitivity in Time and Space-an Introduction. Catena 42:83. Thompson, Victor D. 2006 Questioning Complexity: The Prehistori c Hunter-Gatherers of Sapelo Island, Georgia. Unpublished Ph.D. Dissertation, Departm ent of Anthropology, University of Kentucky, Lexington, KY. Thulman, David K. 2009 Freshwater Availability as the Cons training Factor in the Middle Paleoindian Occupation of North-Central Florida. Geoarchaeology: An International Journal 24:243. Tilley, Christopher 1994 A Phenomenology of Landscape: Pl aces, Paths and Monuments Berg, Oxford. Tornqvist, Torbjorn E., Jess Emilio Gonzl ez Urquijo, Lee A. Newsom, Klaas Van der Borg, Arie F. M. De J ong and Charles W. Kurnik 2004 Deciphering Holocene Sea-Level History on the U.S. Gulf Coast: A HighResolution Record from the Mississippi Delta. Geological Society of America Bulletin 116:1026. Tringham, Ruth E. 2000 The Continuous House: A View from the Deep Past In Beyond Kinship: Social and Material Reproduc tion in House Societies edited by Rosemary A. Joyce and Susan D. Gillespie, pp. 115. University of Pennsylvania Press, Philadelphia. Trigger, Bruce G. 1986 Introduction In Native Shell Mounds of North America edited by Bruce G. Trigger, pp. xixxiv. Garland, New York. 1989 A History of Archaeological Thought Cambridge University Press, Cambridge. 1990 Monumental Architecture: A Ther modynamic Explanation of Symbolic Behavior. World Archaeology 22:119. Trouillot, Michel-Rolph 1995 Silencing the Past: Power and the Production of History Beacon Press, Boston.

PAGE 422

422 Tucker, Bryan D. 2009 Isotopic Investigations of Archaic Pe riod Subsistence and Settlement in the St. Johns River Dr ainage, Florida. Unpublished Ph.D. Dissert ation, Department of Anthropology, University of Florida, Gainesville. Tudhope, Alexander W., Colin P. Chilcott, Malcolm T. McCulloch, Edward R. Cook, John Chappell, Robert M. Ellam, Davi d W. Lea, Janice M. Lough and Graham B. Shimmield 2001 Variability in the El Nio-Southern Oscillation through a Glacial-Interglacial Cycle. Science 291:1511. Tuross, Noreen, Marilyn L. Fogel Lee A. Newsom and Glen H. Doran 1994 Subsistence in the Florida Arc haic: The Stable-Isot ope and Archaeobotanical Evidence from the Windover Site. American Antiquity 59:288. United States Departm ent of Agriculture 2010 Aerial Photography: Florida. Electronic document, http://ufdcweb1.uflib.ufl.edu/ufdc/?a=flap&m=hhh accessed 1/10/2010. Upchurch, Sam B., Richard N. Strom and Mark G. Nuckels 1982 Methods of Chert Provenance: Dete rm ination of Florida Cherts. Report prepared for the Florida Division of Ar chives, History, and Records Management, Bureau of Historic Sites and Properties, Tallahassee. Van Dyke, Ruth M. 2003 Memory and the Construction of Chacoan Society In Archaeologies of Memory edited by Ruth M. Van Dyke and Susan E. Alcock, pp. 180. Blackwell, Malden, Massachusetts. 2004 Memory, Meaning, and Masonry: The Late Bonito Chacoan Landscape. American Antiquity 69:413. 2009 Chaco Reloaded: Discursive So cial Memory on the Post-Chacoan Landscape. Journal of Social Archaeology 9:220. Van Dyke, Ruth M. and Susan E. Alcock 2003 Archaeologies of Memory: An Introduction In Archaeologies of Memory edited by Ruth M. Van Dyke and Susan E. Alcock, pp. 1. Blackwell, Malden, Massachusetts. Viau, A. E., K. Gajewski, M. C. Sawada and P. Fines 2006 Millennial-Scale Temperature Vari ations in North America During the Holocene. J. Geophysical Research 111. Walker, William H. 2002 Stratigraphy and Practical Reason. American Anthropologist 104:159.

PAGE 423

423 Walsh, Stephen J. 2001 Freshwater Macrofauna of Florida Karst Habitats In U.S. Geological Survey Karst Interest Group Proceedings, St. Petersburg, Florida, February 13, 2001 edited by E. Kuniansky, pp. 78. USGS Water-Resources Investigations Report 01-4011. Wanner, Heinz, Jrg Beer, Jonathan Btikofer, Thomas J. Crowley, Ulrich Cubasch, Jacqueline Flckiger, Hugues Goosse, Martin Grosjean, Fortunat Joos, Jed O. Kaplan, Marcel Kttel, Simon A. Mller, I. Colin Prentice, Olga Solomina, Thomas F. Stocker, Pavel Tarasov, Mayke Wagner and Martin Widmann 2008 Midto Late Holocene Climate Change: An Overview. Quaternary Science Reviews 27:1791. Watts, W. A 1969 A Pollen Diagram from Mud Lake, Marion County, North-Central Florida. Geological Society of America Bulletin 80:631. Watts, William A., Eric C. Grimm and T. C. Hussey 1996 Mid-Holocene Forest His tory of Florida and the Coas tal Plain of Georgia and South Carolina In Archaeology of the MidHolocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 28. University of Florida Press, Gainesville. Weiner, Annette B. 1992 Inalienable Possessions: The Pa radox of Keeping-While-Giving University of California Press, Berkeley. Wells, Lisa E. 2001 A Geomorphological Approach to Reconstructing Archeological Settlement Patterns Based on Surficial Artifact Distribution In Earth Sciences and Archaeology edited by Paul Goldberg, Vance T. Holliday and C. Reid Ferring, pp. 107. Kluwer Academic/Plenum, New York. Whallon, Robert 2006 Social Networks and Information: N on-"Utilitarian" Mobi lity among HunterGatherers. Journal of Anthropological Archaeology 25:259. Wharton, Barry, George Ballo and Mitchell Hope 1981 The Republic Groves Si te, Hardee County, Florida. The Florida Anthropologist 34:59. Wheeler, Ryan J. and Ray M. McGee 1994a Report of Preliminary Zooarchaeol ogical Analysis: Groves' Orange Midden. The Florida Anthropologist 47:393. 1994b Technology of Mount Taylor Pe riod Occupation, Groves' Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:350.

PAGE 424

424 1994c Wooden Artifacts from Groves' Orange Midden. The Florida Anthropologist 47:380. Wheeler, Ryan J., James J. Miller, Ray M. McGee, Donna Ruhl, Brenda Swan and Melissa Memory 2003 Archaic Period Canoes fr om Newnans Lake, Florida. American Antiquity 68:533. Wheeler, Ryan J., Christine L. Newman and Ray M. McGee 2000 A New Look at the Mo unt Taylor and Bluffton Sites, Volusia County, with an Outline of the Mount Taylor Culture. The Florida Anthropologist 53:133. White, Leslie 1959 The Evolution of Culture. McGraw-Hill, New York. White, Nancy Marie 2004 Late Archaic Fisher-Foragers in the Apalachicola-Lower Chattahoochee Valley, Northwest Florida-South Georgia/Alabama In Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Philip J Carr, pp. 10. The University of Alabama Press, Tuscaloosa. White, William Arthur 1970 The Geomorphology of the Florida Peninsula Bureau of Geology Division of Interior Resources Florida, No. 51, Tallahassee. White, William B. 1988 Geomorphology and Hydrol ogy of Karst Terrains Oxford University Press, New York. Whitelaw, Todd M. 1983 People and Space in Hunter-Gatherer Camps: A Genera lising Approach in Ethnoarchaeology. Archaeological Review from Cambridge 2:48. 1991 Some Dimensions of Variability in the Social Organization of Community Space among Foragers In Ethnoarchaeological Approaches to Mobile Campsites: Hunter-Gatherer and Pastoralist Case Studies edited by Clive Gamble and William A. Boismier, pp. 1398. International Monographs in Prehistory, Ann Arbor. 1994 Order without Architecture: Functi onal, Social and Symbolic Dimensions in Hunter-Gatherer Settl ement Organization In Architecture and Order: Approaches to Social Space edited by Michael Parker Pear son and Colin Richards, pp. 217 243. Routledge, New York. Wiessner, Polly 1982 Beyond Willow Smoke and Dog's Tails: A Comment on Binf ord's Analysis of Hunter-Gatherer Settlement Systems. American Antiquity 47:171.

PAGE 425

425 Willis, Raymond F. 1995 Management Summary: FY -1995 Heritage Resources Status Report 10, Alexander Springs Beach Reconstructi on, Seminole Ranger District, Ocala National Forest, Lake County (Acc#Semf00226) United States Forest Service, Tallahassee. 2004 Ocala NF Access Policy Development, Lake George & Seminole Districts USFS, ACC# LKGF00282 & ACC# SEMF00300, United States Forest Service Tallahassee. 2006 FY-06, Heritage Resources Status Report 13, St. Francis Timber Management EA Acc#Semf00407 United States Forest Service Tallahassee. Willis, Raymond F., Steven D. Ruple and Sharon Wells 1977 Archaeological and Historical Survey of the Southeastern Portion of the Ocala National Forest Miscellaneous Project Report Se ries Number 41. Bureau of Historic Sites and Properties, Florida Department of State, Tallahassee. Wilmsen, Edwin N. and James R. Denbow 1990 Paradigmatic History of the San-Speaking Peoples and t he Current Attempts at Revision. Current Anthropology 31:489. Wilson, Michael C. 2005 The "Placing" Of Identity in Nomadic Societies: Aboriginal Landscapes of the Northwestern Plains of North America. Material History Review 62:7. Wing, Elizabeth S. and L. McKean 1987 Preliminary Study of the Animal Remains Excavated from the Hontoon Island Site. The Florida Anthropologist 40:40. Winterhalder, Bruce 2001 The Behavioural Ecol ogy of Hunter-Gatherers In Hunter-Gatherers: An Interdisciplinary Perspective, edited by Catherine Pante r-Brick, Robert H. Layton and Peter Rowley-Conwy, pp. 12. Cambri dge University Press, Cambridge. Winterhalder, Bruce a nd Eric Alden Smith 1992 Evolutionary Ecology and the Social Sciences In Evolutionary Ecology and Human Behavior edited by Eric Alden Smith and Bruce Winterhalder, pp. 3. Aldine de Gruyter, Hawthorne, NY. Wobst, H. Martin 2000 Agency in (Spite of) Material Culture In Agency in Archaeology edited by Marcia-Anne Dobres and John E. Robb, pp. 40. Routledge, London. Wolf, Eric R. 1982 Europe and the People without History University of California Press, Berkeley.

PAGE 426

426 Wood, Denis 1993 The Fine Line between Mapping and Mapmaking. Cartographica 30:50. Woodburn, James 1982 Egalitarian Societies. Man 17:431. Woodroffe, C. D., J Chappell, B. G. Thom and E. Wallensky 1989 Depositional Model of a Macrotidal Estuary and Floodplain, South Alligator River, Northern Australia. Sedimentology 36:737. Wright, Eric E., Albert C. Hine, Stev en L. Goodbred, Jr. and Stanley D. Locker 2005 The Effect of SeaLevel and Climate Change on t he Development of a Mixed Siliciclastic-Carbonate, Deltaic Coast line: Suwannee River, Florida U.S.A. Journal of Sedimentary Research 75:621. Wyman, Jeffries 1867 An Account of Some Kjoekkenmoeddings, or Shell-Heaps, in Maine and Massachusetts. The American Naturalist 1:567. 1872 Journal and Field Notes. Manuscript on file, Harvard Countway Medical Library. (HMS B54.1). Cambridge, MA. 1874 Human Remains in the Shell Heaps of the St. John's River, East Florida. Cannibalism. The American Naturalist 8:403. 1875 Fresh-Water Shell Mounds of the St. John's River, Florida. Peabody Academy of Science Memoir 4. Yaeger, Jason and Marcello A. Canuto 2000 Introducing an Archaeology of Communities In The Archaeology of Communities: A New World Perspective edited by Marcello A. Canuto and Jason Yaeger, pp. 1. Routledge, New York. Yoffee, Norman 1993 Too Many Chiefs? (or, Safe Texts for the '90s) In Archaeological Theory: Who Sets the Agenda? edited by Norman Yoffee and Andrew Sherratt, pp. 60 78. Cambridge University Press, Cambridge. 2005 Myths of the Archaic State: Evolution of the Earliest Cit ies, States and Civilizations Cambridge University Press, Cambridge. Zvelebil, Marek 2002 The Invention of Hunter-Gather ers in Seventeenth-Century Europe? A Comment on Mark Pluciennik. Archaeological Dialogues 9:123. 2003 Enculturation of Mesolithic Landscapes In Mesolithic on the Move: Papers Presented at the Sixth International Conf erence on the Mesolithic in Europe, Stockholm 2000 edited by Lars Larsson, Hans Kindgren, Kjel Knutsson, David Loeffler and Agneta kerlund, pp 65. Oxbow Books, Oxford.

PAGE 427

427 Zvelebil, Marek and Peter Jordan 1999 Hunter Fisher Gatherer Ritual Landscapes Questions of Time, Space and Representation In Rock Art as Social Representation edited by Joakim Goldhahn, pp. 101. BAR International Series 794. Archaeopress, Oxford.

PAGE 428

428 BIOGRAPHICAL SKETCH Asa R. Randall is a native of Massac husetts. Asa began his college career at Boston University, where he earned his Bac helor of Arts in archaeological studies in 1999. He was accepted to the University of Florida in 1999, but deferred a year to work in the field of cultural resource m anagement. Asa began graduate studies in 2000 and received his Master of Arts in anthropology in 2002. Since 2003, Randall has examined the histories of shell mounds along the St. Johns River in northeast Florida, and this research forms the basis of his dissertati on. Asa earned his Doctor of Philosophy in anthropology in 2010, and is currently the Seni or Archaeologist of the Laboratory of Southeastern Archaeology at the University of Fl orida in Gainesville.