Plaquemine Archaeology

Plaquemine Archaeology

A Dan Josselyn Memorial Publication

Plaquemine Archaeology

Edited by Mark A. Rees and Patrick C. Livingood

the uni v ersit y of a l a ba m a press Tuscaloosa

Copyright © 2007 The University of Alabama Press Tuscaloosa, Alabama 35487-0380 All rights reserved Manufactured in the United States of America Typeface: AGaramond and Triplex ∞ The paper on which this book is printed meets the minimum requirements of American National Standard for Information Sciences-Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984. Library of Congress Cataloging-in-Publication Data Plaquemine archaeology / edited by Mark A. Rees and Patrick C. Livingood. p. cm. Includes bibliographical references and index. ISBN-13: 978-0-8173-1543-6 (cloth : alk. paper) ISBN-10: 0-8173-1543-8 (alk. paper) ISBN-13: 978-0-8173-5366-7 (pbk. : alk. paper) ISBN-10: 0-8173-5366-6 (alk. paper) 1. Plaquemine culture. 2. Mounds—Louisiana. 3. Mounds—Mississippi. 4. Plaquemine pottery—Louisiana. 5. Plaquemine pottery—Mississippi. 6. Excavations (Archaeology)— Louisiana. 7. Excavations (Archaeology)—Mississippi. 8. Louisiana—Antiquities. 9. Mississippi—Antiquities. I. Rees, Mark A. II. Livingood, Patrick C. E99.P635P57 2007 976′.01—dc22 2006016065

Contents Figures Tables Preface

vii xi xiii

1. Introduction and Historical Overview / Mark A. Rees and Patrick C. Livingood 1 2. Coles Creek Antecedents of Plaquemine Mound Construction: Evidence from the Raffman Site / Lori Roe 20 3. Extraregional Contact and Cultural Interaction at the Coles Creek–Plaquemine Transition: Recent Data from the Lake Providence Mounds, East Carroll Parish, Louisiana / Douglas C. Wells and Richard A. Weinstein 38 4. Plaquemine Mounds of the Western Atchafalaya Basin / Mark A. Rees 66 5. Transitional Coles Creek–Plaquemine Relationships on Northwest Lake Salvador, St. Charles Parish, Louisiana / Malcolm K. Shuman 94 6. Plaquemine Recipes: Using Computer-Assisted Petrographic Analysis to Investigate Plaquemine Ceramic Recipes / Patrick C. Livingood 108 7. Feasting on the Bluffs: Anna Site Excavations in the Natchez Bluffs of Mississippi / Virgil Roy Beasley III 127 8. Plaquemine Culture in the Natchez Bluffs Region of Mississippi / Ian W. Brown 145 9. The Outer Limits of Plaquemine Culture: A View from the Northerly Borderlands / Marvin D. Jeter 161

vi

contents

10. Contemplating Plaquemine Culture / Tristram R. Kidder References Cited Contributors Index

261

207 259

196

Figures 1.1. The Lower Mississippi Valley, showing the major drainages and regions discussed in the text 2 1.2. Areas discussed by individual authors

16

2.1. Core Coles Creek region and locations of most sites discussed in the chapter 22 2.2. Middle Woodland through Historic phase chronology for the central Lower Mississippi Valley 23 2.3. The Raffman site, showing locations of 1998 through 2004 excavations 27 2.4. Mound B at Raffman, showing locations of excavations and soil cores 33 3.1. Location of the Lake Providence Mounds

39

3.2. Contour map of the Lake Providence Mounds, showing the four extant mounds and the Corps’ right-of-way during the 1998–1999 datarecovery program 40 3.3. Pro¤les of north and east walls of TU N80W79 showing the three construction stages revealed during excavation 41 3.4. Plan of wall trenches and postholes associated with the Preston phase in Block 1, showing evidence of the three building episodes believed to have been present 43 3.5. Date ranges of the seven Coles Creek occupations recognized at Lake Providence 44 3.6. Proposed site-formation sequence at Lake Providence

45

3.7. Ceramics normally thought to be representative of the Balmoral and Routh phases but clearly part of the Preston phase component at Lake Providence 47

viii

figures

3.8. Preston phase ceramics from Lake Providence

48

3.9. Additional Preston phase ceramics from Lake Providence

49

3.10. Ceramic types and varieties of the Balmoral, Preston, and Routh phases 50 3.11. Examples of pottery associated with the “Preston ¤neware complex” 51 3.12. Sherds of the Powell and Coker sets recovered at Lake Providence 53 3.13. Additional artifacts suggestive of contact with the American Bottom region 54 3.14. Depiction of Early Mississippian in®uences on the Yazoo Basin during the Crippen Point phase 55 3.15. Chronology chart illustrating the temporal relationship of archaeological phases in the Tensas Basin and American Bottom regions since about a.d. 750 59 4.1. South-central Louisiana, showing the locations of major Plaquemine mound sites 73 4.2. Site 16SL3 4.3. Site 16SM38 4.4. Site 16SM5 4.5. Site 16SMY10

74 77 81 85

5.1. Lake Salvador and archaeological sites mentioned in the chapter 98 6.1. Plane-polarized scan of PRP27

115

6.2. Cross-polarized scan of PRP27

115

6.3. Example of a false-color enhanced image of the cross-polarized scan for PRP27 115 6.4. Boolean image of the shell and shell void identi¤cations for PRP27 115 6.5. Biplot of grog and shell percentages

120

ix

figures

6.6. Biplot of grog and shell percentages showing clusters determined by analysis of the petrographic data 122 7.1. Location of Anna site

128

7.2. Site map of Anna Mounds group

131

7.3. Neo-Indian chronology in the Lower Yazoo Basin and Natchez Bluffs region of the Lower Mississippi Valley 135 7.4. Plan view, Block 1 excavations 7.5. Chicot Red, var. Fairchild vessel 7.6. Addis Plain, var. Addis vessel

139 140 140

8.1. Selected Plaquemine sites in the Natchez Bluffs region of Mississippi 146 9.1. Southeast Arkansas, showing major streams, county boundaries, and locations of some key sites mentioned in the chapter 164 9.2. The Bellaire “serpent-cat” pipe

168

9.3. Addis Plain bowl with deeply notched rim from the Boydell site in southeast Arkansas 173

Tables 2.1. Radiocarbon dates from the Raffman site

29

3.1. Proveniences of samples and results of sourcing analysis on selected Lake Providence sherds 57 4.1. Radiocarbon dates from mound sites in the western Atchafalaya Basin 79 6.1. Samples used in the analysis

118

6.2. Description of the temper clusters

125

7.1. Faunal MNI and estimated meat yields from Foster phase surface 137

Preface

I wish I could be sure when I ¤rst encountered the term Plaquemine, but it cannot be less than 50 years ago. Probably it occurred when I was in Ann Arbor in 1949–1950 with Jimmy Grif¤n, although I trust that my earlier Yale mentor, Ben Rouse, could not have been ignorant of it either, due to his amazingly broad knowledge of world archaeology and his having been the editor of American Antiquity. In the text herein, I am cited as using the term in a 1956 article! Wow, how little one remembers of one’s own past. However, we Lower Valley scholars must not be too brazen in our knowledge and usage. I am sure that we cannot make Plaquemine a broadly known term in the eastern United States or even in the southeastern corner of our nation. But we can take heart from other quite well-known terminology and realize that every area in American archaeology has its own best-known words: “Clovis and Pecos” for New Mexico, “Cahokia” for Illinois, “Hopewellian” for Ohio, and so on. So why not “Plaquemine,” then, for a culture in the southern part of the Lower Mississippi Valley, named for a town in Louisiana? Of course, I do support large overviews from the elders. My own introduction to the Lower Mississippi Valley began even before Ann Arbor in the 1950s; it actually began with a December 1948 visit to the Sandy Woods site in southeast Missouri with my father and my twin brother, Philip. I had been working in the Yale Peabody Museum on an archaeological collection from that site for my Senior Honors paper. I grew up—one through ¤ve years—on the very banks of the Mississippi River in Minneapolis. But enough of the past. From southeast Missouri the whole Lower Mississippi Valley lay ahead for me: Nodena in northeast Arkansas, the Lake George site in the Yazoo Basin, and also much farther south still with the Lower Mis-

xiv

preface

sissippi Survey through Louisiana, only ending at Avery Island on the Gulf of Mexico. The Lower Mississippi Survey from the Harvard Peabody Museum has been there and done that, starting with Phil Phillips’s work in the late 1930s. So back to Plaquemine: this volume provides a very good coverage on the topic, and there is not too much more that I can add. When the readers ¤nish this lengthy discussion on Plaquemine I think they will have learned quite a lot of useful information on this perennially discussed topic. Stephen Williams, Professor Emeritus Harvard University

1 Introduction and Historical Overview Mark A. Rees and Patrick C. Livingood

The town of Plaquemine, Louisiana, seems an unassuming southern community for which to designate an entire culture. Of course names can be misleading and Plaquemine is no exception. Like Mississippian culture of southeastern North America, of which Plaquemine has been described as a regional focus or variant (Grif¤n 1946, 1967), Plaquemine was devised by archaeologists to describe the material remains and sites of pre-Columbian and protohistoric Native American societies for which there exists little or no written documentation. Its geographic position has been associated with the Lower Mississippi Valley (LMV), extending from the Mississippi delta on the Gulf coast to just south of the Arkansas River ( Jeter and Williams 1989:207–208; cf. Neuman 1984:259). This area spans the southern Yazoo Basin and Natchez Bluffs on the east and the lower Ouachita and Red river valleys on the west, encompassing a diverse range of valley topography and environments (Figure 1.1) (Autin et al. 1991; Kidder 2004b:545; Saucier 1994). So what’s the problem or, rather, what are the problems with Plaquemine? Like the town for which it was named, it has seemed at times on the verge of washing into the Mississippi River (Riffel 1985:38–41). Unlike its more wellknown and extroverted Mississippian cousin, Plaquemine seems to have been fraught with dif¤culties since early on. No sooner had the concept been de¤ned than the exigencies of archaeological research required that it be deconstructed and rede¤ned (Phillips 1970:950). Alternating descriptions of diagnostic pottery types, earthen mounds, and architecture in different regions have produced the appearance of an imprecise or “nebulous” tradition (Brown 1985b:252; Jeter and Williams 1989:205). Yet Plaquemine is today still regarded as a distinct cultural tradition on the frontier of the Mississippian world, with ties to both earlier Coles Creek and contemporaneous Mississippian tradi-

Figure 1.1. The Lower Mississippi Valley, showing the major drainages and regions discussed in the text.

introduction a nd historic a l ov erv iew

3

tions, beginning in the second or third century of the Mississippi period (ca. a.d. 1000–1700). As advances in research continue to make contributions to our understanding of the archaeology of the LMV, it is apropos to reconsider the concept of Plaquemine culture. An attempt has been made in this volume to draw together recent studies that call attention to variability as well as uniformity, including views from such disparate environments as the Mississippi River ®oodplain, the Tensas Basin, the Natchez Bluffs, and the “northerly borderlands” of southeastern Arkansas. Even so, much variability in the archaeological record has yet to be adequately investigated. There is little doubt that Plaquemine will continue to serve on one level as a valuable heuristic in examining and making sense of that variability. This chapter examines the origins and de¤nitions of Plaquemine, summarizes some of the history and themes of Plaquemine archaeology, and introduces some ongoing problems and directions for future research.

Historical Genesis and Adjustments Plaquemine ¤rst appeared in print in James Ford and Gordon Willey’s classic 1941 synthesis of eastern United States prehistory, although without further consideration in the text. Plaquemine was shown to be intermediate to the preceding Coles Creek period and subsequent Natchez and Bayou Goula cultures. The embryonic nature of the concept is illustrated in the atypical spelling (“Placquemine”) in two chronological pro¤les (Ford and Willey 1941:Figures 2 and 6). Ford and Willey (1941:359) also cited the Bayou Goula report as “in preparation” by Quimby. “Placquemine” was portrayed by Ford and Willey (1941:328, 330) as transitional between the Temple Mound I and II stages along the north–south axis of eastern North American cultures and associated with the Temple Mound II stage in the east–west axis. The Temple Mound stage terminology was eventually superseded by the Mississippi period and Plaquemine would come to be de¤ned by way of contrast (Grif¤n 1946, 1967; Williams 1956; cf. Willey 1966:292–304). The origins of Plaquemine can ultimately be traced to a series of distinctions made during the ¤rst half of the twentieth century, distinctions that at least initially had little to do with Plaquemine. Surveying the geographic distributions of Native American pottery in the eastern United States a century ago, William Henry Holmes (1903:21) predicted that the groups of pottery he laid out contained suf¤cient internal variability to eventually warrant further subdivision (see also Moore 1913). Holmes (1886, 1903) was instrumental in the creation of the Mississippian concept via his “Middle Mississippi Valley

4

rees a nd li v ingood

group,” which by mid-century had become ¤rmly established as the late preColumbian and protohistoric period in the Southeast, as well as a precocious and expansive culture subsequently thought to have emanated from the Central Mississippi Valley (DeJarnette and Wimberly 1941:99–110; Grif¤n 1943:257, 1946, 1952; Phillips et al. 1951; Williams 1956; see Grif¤n 1985 for an overview). Holmes (1903:21) suggested that ceramic traditions do not inevitably coincide with ethnicity or culture, nor are they always a reliable measure of cultural complexity (i.e., “important groups” versus “insigni¤cant communities”). He described the geographic distribution, function, stylistic variation, and temper of pottery in the Mississippi Valley, including the physical properties of shell temper and general similarities between Lower and Middle Mississippi wares (Holmes 1903:20–80, 101–104). Mississippian culture and the Mississippi period were derived from this and subsequent works in the context of an emergent culture historical archaeology in the Southeast (Grif¤n 1985:44–57). Plaquemine was developed by James Ford and George Quimby based on the results of WPA work conducted between 1938 and 1941 under the auspices of the Louisiana State Archaeological Survey. It represented a re¤nement of Ford’s (1935a, 1936, 1938) earlier chronology of ceramic-producing cultures in the LMV: Marksville, Coles Creek, and Natchez (Quimby 1942:256, 1951:87). In an article completed the same year Ford and Willey’s overview was published (1941), Quimby (1942:256) de¤ned Plaquemine as a “culture which preceded the Natchezan type.” Like Mississippian, Plaquemine had its roots in more fundamental pottery typologies and their chronological classi¤cation. More than any other culture trait, pottery types were seen as central in distinguishing Natchez from contemporaneous and earlier cultures. Among the types initially regarded by Quimby (1942:265–268) as characteristic of a “Plaquemine period” were Addis Plain (Baytown Plain, var. Addis), Manchac Incised (Mazique Incised, var. Manchac), Hardy Incised (Coles Creek Incised, var. Hardy), Plaquemine Brushed, Pocahontas Plain (Mississippi Plain, var. Pocahontas), and Pocahontas Punctated, the latter two types subsequently associated more strictly with the Natchez Bluffs (e.g., Neitzel 1983:95). Coles Creek was in effect split into thirds: Troyville, Coles Creek proper, and Plaquemine. Plaquemine was also envisioned as a spatially intermediate pottery complex, wedged between Caddo sites on the west and various Mississippian complexes farther north and east (Ford and Willey 1941:328, 330). In this sense, Plaquemine was initially conceptualized as “a logical correction for the hasty 1940 overextension of the Caddoan concept, rather than an outgrowth of it” ( Jeter and Williams 1989:205; cf. Phillips 1970:946). Plaquemine thus came into being as a somewhat vague designation for post–Coles Creek,

introduction a nd historic a l ov erv iew

5

pre-Natchez, non-Caddo, and non-Mississippian pottery assemblages in the Lower Valley (Brown 1985b:252; Jeter and Williams 1989:205; Quimby 1942:256). Plaquemine consequently received the dubious distinction early on of being “de¤ned more in terms of what it lacks than in what it possesses” ( Jennings 1952:267). Archaeological understanding of Plaquemine components and assemblages advanced slowly during the ensuing decades, despite the fact that it was never “adequately de¤ned” as a culture (Williams and Brain 1983:373). As it turns out, 1951 was a seminal year for publications on Plaquemine as well as Mississippian archaeology (e.g., Cotter 1951a; Ford 1951; Phillips et al. 1951; Quimby 1951). Sites such as Greenhouse, Medora, Emerald, and Anna yielded a wealth of information with which to ®esh out the concept of Plaquemine culture. With the addition of publications by Cotter (1952a) and Quimby (1957) on the Gordon and Bayou Goula sites, these works made up until the early 1970s, according to Phillips (1970:950), the “basic documentation for the Plaquemine culture period” (see also Belmont 1967; Bohannon 1963; Brain 1969; Hally 1967; Neitzel 1965). The rapid appearance of these groundbreaking publications, delayed by World War II, provided the impetus for examining internal variability in respect to pottery types and the rede¤nition of Plaquemine-related phases (Phillips 1970:950). Ironically, it would also in®uence consideration of similarities between regional variants of Plaquemine culture and Coles Creek, Mississippian, Natchez, and Caddo cultures, or precisely what had made it necessary to differentiate Plaquemine in the ¤rst place (e.g., Brain 1978; Brown 1985a, 1985b; Gregory 1969; Hally 1972; Neitzel 1965, 1983; Steponaitis 1981; Webb 1959, 1961). Quimby’s (1951) Medora site report is generally cited as establishing the formal, archaeological criteria for both Plaquemine culture and a Plaquemine period (e.g., Ford 1952:327; Jeter and Williams 1989:206; Phillips 1970:950). Quimby argued that speci¤c ceramic types such as Addis Plain (Baytown Plain, var. Addis), Plaquemine Brushed, L’Eau Noire Incised, Manchac Incised (Mazique Incised, var. Manchac), Hardy Incised (Coles Creek Incised, var. Hardy), and others are integral to the de¤nition of the Plaquemine concept. Certain ceramic vessel forms such as plates with interior incising and jars with brushed decoration remain key Plaquemine indicators. Because the report was based on excavation and not just surface collections, he was also able to suggest other traits that characterize Plaquemine (Quimby 1951:128). Some of these, such as the presence of truncated, pyramidal mounds around a plaza and post construction with and without wall trenches, have generally held up over time.

6

rees a nd li v ingood

Like most chronologies proposed before radiocarbon dating was available, the original dates for Plaquemine turned out to be too short and too recent. Originally thought to date not much earlier than a.d. 1400–1500 (Ford and Willey 1941:328, 330; Quimby 1951:130, 1957), it is now commonly considered to date from at least a.d. 1200. Phillips, Ford, and Grif¤n (1951:454, Table 17) suggested as much in the conclusion of their monumental Archaeological Survey in the Lower Mississippi Alluvial Valley, based on the correlation of sites with Mississippi River channel chronology. Plaquemine was preceded according to some by a Coles Creek transition (a.d. 1000–1200; Jeter and Williams 1989:172; Weinstein and Kelley 1992:31, 38). Medora, the type site for Plaquemine culture, in fact appears to date in large part from this transitional era (Brown 1985b:256; Weinstein 1987a:87–90). Plaquemine culture is now also usually placed within the Mississippi period (rather than Quimby’s Plaquemine period), re®ecting the greater in®uence of Mississippian culture and Mississippian archaeology (Brown 1998e).

Plaquemine Origins Reconsidered There remain substantial differences of opinion concerning Plaquemine cultural origins, particularly in relation to preceding Coles Creek and contemporaneous Mississippian traditions. At the risk of oversimplifying the present state of affairs, three lines of argument generally characterize these differences. The ¤rst is represented in the formative works of Ford (1951), Quimby (1942, 1951, 1957), and Phillips (1970) and is referred to here as “neither Mississippian nor Coles Creek.” In contrast, the second has been described as “Mississippianized Coles Creek” (Brain 1978:344–345; Brown 1985b:253; Williams and Brain 1983), while a third approach represents a middle ground of sorts and might be characterized as endogenous development or “continuity and change” (Kidder 1998b:131, this volume). Although there is certainly room for compromise (e.g., Kelley et al. 2000:17–18; Weinstein 1987a:87), we use these distinctions here as a means of highlighting the diversity of Plaquemine through time and space.

Neither Mississippian nor Coles Creek Phillips (1970) both emphasized the position of Plaquemine as a culture and rede¤ned the culture in terms of speci¤c phases. This allowed him to talk about the Mississippi period throughout the LMV but to distinguish between Plaquemine and Mississippian cultures. Phillips re¤ned and codi¤ed the typevariety system used for classifying ceramics in the Lower Valley. One of the hallmarks of this system is that it is hierarchical; it calls for investing more

introduction a nd historic a l ov erv iew

7

importance in certain distinctions by using them as the criteria to separate types, and it calls for subsuming other, presumably more minor differences under the divisions between varieties (e.g., Brown 1998b). The presence or absence of shell tempering was recognized as one of the ¤rst-order distinctions used in the descriptions of types. Since Phillips used ceramic evidence as the primary criteria for making cultural distinctions, the absence or presence of shell tempering became the axiomatic means for separating Plaquemine from Mississippian (Kidder 1998b:131; Williams and Brain 1983:337, 340). The de¤nition of Plaquemine we inherit from Phillips (1970:30, 34) thus hinges almost entirely on ceramics. He stipulated that Plaquemine is de¤ned by a cluster of ceramic types and forms, the absence of signi¤cant shell tempering being the most important characteristic that ties these complexes together. Plaquemine is distinguished by “long-standing typological criteria” developed out of Phillips, Ford, and Grif¤n’s (1951) earlier Archaeological Survey in the Lower Mississippi Alluvial Valley (Phillips 1970:923). Phases as well as cultures were distinguished on the basis of ceramics, according to Phillips (1970:9) representing “discontinuities of a minor order that are no less inherent in the material.” Phases were consequently emphasized as the central mechanism for understanding cultural dynamics, as embodied in regional designations such as Medora, Crippen Point, and Fitzhugh (Phillips 1970:558–560, 945, 950–951; cf. Phillips and Willey 1953). Phillips (1970:950–951) thus commented on the phase-based “dismemberment” of Plaquemine culture as a necessary analytical exercise, excising all except those phases that maintained Plaquemine as a cohesive and unadulterated complex. Of course, one result of this approach was to emphasize the need to determine whether a particular phase was either Plaquemine or Mississippian ( Jeter and Williams 1989:212). On a broader scale, Phillips (1970:13) described Mississippian in®uence in the Lower Yazoo Basin in terms of intrinsic cultural movements, falling somewhere between diffusion and invasion. He referred to the ability of Plaquemine culture bearers to repel, at least for a time, the seemingly indomitable Mississippian drive southward (“Drang nach Süden”; Phillips 1970:954), what a later generation of archaeologists would critique as the “Mississippian expansion” (i.e., Smith 1984). The idea of Plaquemine as neither Mississippian nor Coles Creek was thus integral to its origin as an archaeological culture, pottery complex, culture period, and phase. Ford (1951:85–89) found only minor evidence for an early Plaquemine component at Greenhouse, a major Coles Creek mound site in central Louisiana. Similarities between Plaquemine and Coles Creek were

8

rees a nd li v ingood

seen as strongest in this area and to the south, both in terms of pottery types and the continued use of burial mounds (e.g., Jennings 1952:267; Rees, this volume). In contrast, similarities between Plaquemine and Mississippian were more readily apparent to the north, particularly at sites with large platform mounds. Yet perceptions of difference and similarity had actually provided little explanation for Plaquemine cultural origins. Near the end of his Yazoo Basin report, Phillips (1970:967–968) amended an earlier suggestion that Plaquemine architecture and settlement patterns in the Yazoo Basin appeared to represent the “Mississippianization” of Plaquemine culture, arguing instead that “the hypothesis can only be tested by thorough excavation.”

Mississippianized Coles Creek Following Phillips’s prescient advice, Stephen Williams and Jeffrey Brain reformulated and advanced the concept of Plaquemine in several key ways from the vantage point of the Lower Yazoo Basin (Brain 1978, 1989, 1991; Williams and Brain 1983). They produced the most coherent and far-reaching interpretation of Plaquemine cultural origins and development by arguing that Plaquemine could be best viewed as “a hybridization of Coles Creek and Mississippian cultures” (Williams and Brain 1983:338). On the basis of the results of investigations at Lake George and Winterville, they argued that during the Coles Creek period LMV residents, especially in frontier zones, began to adopt new technologies with Mississippian origins, such as shell tempering, the Mississippian jar, and wall-trench houses (Brain 1969, 1989:114–120; Williams and Brain 1983:408–414). This was followed by the encroachment of small numbers of Mississippian communities into the northern part of the Yazoo Basin, at sites such as Buford. By a.d. 1200 direct person-to-person contacts were occurring between major Mississippian settlements to the north and major sites in the Lower Valley. Between a.d. 1300 and 1400 sites in the Lower Yazoo began to be acculturated to the Mississippian pattern and more closely resembled sites farther north (Brain 1989:125, 128). As a result, a Mississippian/Plaquemine frontier formed and moved south. Prior to a.d. 1300 it was somewhere north of Greenville, Mississippi, but by a.d. 1400–1500 the frontier had moved to the vicinity of Vicksburg. Plaquemine was consequently de¤ned as the Mississippianization of Coles Creek culture, a concept that has more recently undergone historical-processual revision (Brain 1989:122; e.g., Pauketat 2002, 2004). Williams and Brain further amended the de¤nition of Plaquemine by reclassifying the Crippen Point phase as terminal Coles Creek instead of early

introduction a nd historic a l ov erv iew

9

Plaquemine, thus moving the Coles Creek–Plaquemine transition from about a.d. 1000 to a.d. 1200. Phillips (1970:558–560) had originally classi¤ed Crippen Point as Plaquemine because it was associated with the appearance of Addis-paste ceramics. However, Williams and Brain (1983:318) argued that Crippen Point ceramics continued to exhibit linear decorative techniques focused on the vessel rim, like most previous Coles Creek forms. Beyond ceramics, they pointed out that the Lake George site underwent only small, gradual changes during the Crippen Point phase, placing it more comfortably within the preceding Coles Creek period. As Jeter and Williams (1989:208–209) indicate, the reformulation of the Coles Creek–Plaquemine transition has far-reaching consequences for how we characterize the late prehistory of the Lower Valley. If the Crippen Point phase is categorized as Coles Creek, then Williams and Brain (1983:405) are correct in describing Coles Creek as a relatively conservative period followed by dramatic changes, including the abrupt adoption of Mississippian technologies. Plaquemine is in effect the culmination of Mississippian cultural in®uences upon conservative Coles Creek societies. Alternatively, if the Crippen Point phase is categorized as Plaquemine, then the origins of Plaquemine culture predate major Mississippian in®uence and changes appear more gradual. Outside the Lower Yazoo Basin, researchers have faced related dif¤culties in classifying phases according to broader cultural categories. For example, there have been disagreements over whether to classify the Bayou Petre phase in the Mississippi delta as either Mississippian or Plaquemine (Brown 1985b:283; Phillips 1970:951–953; Weinstein 1987a:98). Differences of opinion have also arisen regarding the Preston phase in the Tensas Basin and whether or not to classify it as Coles Creek (Hally 1972:605–607; Jeter and Williams 1989:210– 212; Wells and Weinstein 2005). Such decisions ultimately transcend differences regarding which period or culture a certain phase should be associated and strike at the heart of how we conceive of Plaquemine. Faced with the ambiguities of drawing an exact boundary between Coles Creek and Plaquemine where none appears to exist, some scholars have instead advocated using “transitional Coles Creek–Plaquemine” to refer to intermediate phases such as Crippen Point, Gordon, Preston, and St. Gabriel (Belmont 1982a; Belmont and Williams 1981; Brown 1985b; Jeter and Williams 1989; Weinstein 1987a; Wells and Weinstein, this volume). In doing so, the question of Plaquemine origins is effectively rephrased as a problem of understanding continuity and change.

10

rees a nd li v ingood

Continuity and Change An alternative approach to Plaquemine origins is to describe it as an endogenous cultural tradition or outgrowth of preceding Coles Creek culture, with no less signi¤cant exogenous in®uences (Haag 1965:304–307, 1978, 1988; Hally 1972; Kidder 1998b; Neuman 1984:258; Sibley 1967:157). From this perspective, interpretations of Plaquemine culture as “Mississippianized Coles Creek” rely too heavily on the Lower Yazoo data and do not generalize well to other regions such as the Lower Ouachita, Tensas Basin, or Natchez Bluffs (Kidder 1998b:131). In fact, there is little evidence from Plaquemine components in areas far removed from the Lower Yazoo Basin for direct contacts or similar wide-ranging Mississippian in®uence (e.g., Quimby 1951; Rees 2003, this volume). Despite the fact that communities in regions closer to the Lower Yazoo Basin adopted similar site plans and ceramic forms, it seems unrealistic to consider all Plaquemine components throughout the LMV as Mississippianized Coles Creek (Brown 1985b; Hally 1972). When Mississippian contacts have been identi¤ed in other regions it has been in juxtaposition with a clear progression from Coles Creek to Plaquemine traditions (Brown 1999; Davis and Giardino 1980; Weinstein 1987a, 1987b; Weinstein and Kelley 1984; Wells and Weinstein, this volume). For example, Mississippian in®uences from the Mobile Bay region via the Bayou Petre phase have been identi¤ed for a coastal Plaquemine variant represented in the Barataria and Burk Hill phases (Brown 1985b; Brown et al. 1979; Davis 1984; Jeter and Williams 1989:218–220; Shuman, this volume; Weinstein 1987a). It should come as no surprise, then, that Caddo in®uence has likewise been discerned for Plaquemine phases on the western “frontier” ( Jeter and Williams 1989:213; Rolingson 1976). Similarities between Caddo and Plaquemine ceramic assemblages were recognized early on by Grif¤n (1946:83), Krieger (1946:253, 255), and Webb (1959:148, 150, 153, 202, 206), supposedly based on a common Coles Creek derivation (see also Sibley 1967:157–164). Despite differences of opinion, many archaeologists now recognize clear continuities between Coles Creek and Plaquemine traditions (Brown 1984:115, 118; Kelley 1984:43; Miller et al. 2000:47; Ryan, ed. 2004:23–24; Weinstein 1987a:87; Wells 2001:16). Regarding the “typological criteria” set forth by Phillips (1970:923), David Hally (1972) has suggested that archaeologists may be placing too much importance on the appearance of shell tempering. T. R. Kidder (1998b:132) has likewise suggested that an emphasis on shell tempering might be creating an “arti¤cial dichotomy between Mississippian and Plaquemine cultures and

introduction a nd historic a l ov erv iew

11

gloss over some obvious similarities” (see also Kidder 2004b:555). The appearance of shell tempering was likely a much more complex and varied phenomenon (Shuman, this volume). Shell tempering was adopted at some sites in the Tensas Basin, such as Transylvania, much earlier than at nearby sites (Hally 1972:423–426). At sites like Transylvania, where both Addis and shelltempered pastes were used, the majority of decorated forms were placed on shell-tempered ware while the majority of plain sherds were Addis (Hally 1972). At sites such as Pevey on the Pearl River, however, the opposite condition is true (Livingood 1999, this volume). Furthermore, several decorative techniques seem to have been exclusively reserved for Addis paste, while others were more likely to be executed on shell-tempered vessels or vessels containing shell in an otherwise Addis paste (Hally 1972:426–430; Kidder 1998b:132). Until some of these complex relationships are addressed, it is dif¤cult to fathom the precise meaning of the adoption of shell tempering. Whether it is regarded as an unambiguous technological innovation or part of a foreign cultural package, it would have involved the negotiation of new social meanings on a local level (Nassaney 2001:168–169; Pauketat 2001b:82). Investigating Plaquemine components in terms of both continuity and change is to allow for historical connections with Coles Creek, Caddoan, Natchezan, and Mississippian traditions, while at the same time recognizing essential differences. Unlike the Mississippianized Coles Creek point of view, this does not provide an intrinsic or straightforward explanation of cultural origins. From yet another perspective, the question of cultural origins seems misplaced. Following the example of Mississippian archaeology elsewhere in the Southeast, Plaquemine might be approached from a regional, political-economic perspective (e.g., Anderson 1996b, 1999; Barker and Pauketat, eds. 1992; Hally 1993; Pauketat 1997a, 1997b). Archaeologists are again confronted with the familiar challenge of working from the ground up, in hopes of re¤ning presentday knowledge of Plaquemine culture, one site and one assemblage at a time.

Future Directions and the Present Work Plaquemine, it seems, despite six decades of research, is still badly in need of de¤nition (cf. Jennings 1952:267; Kelley 1984:42; Williams and Brain 1983:373). Variability between phases in the Lower Yazoo Basin, Natchez Bluffs, Tensas Basin, and Baton Rouge area even now threaten to erode the conceptual foundation laid long ago by James Ford; yet these are the very regions that might be construed as a Plaquemine “heartland” ( Jeter and Williams 1989:207–212). Considerably less is known about the Lower Ouachita Valley or Felsenthal region of southeast Arkansas, the Bartholomew-Boeuf Basin, Catahoula Ba-

12

rees a nd li v ingood

sin, Lower Red River, Atchafalaya Basin, or Gulf coastal plain. Having de¤ned Plaquemine as a more or less cohesive cultural tradition in contrast to other spatially and temporally adjoining traditions, archaeologists are still in the process of ¤lling in the details on local and regional levels. Part of the problem, of course, is that the term Plaquemine has been so broadly applied (Jeter and Williams 1989:205–220). Consideration of a Plaquemine “heartland” implies Coles Creek antecedents more or less removed from Mississippian stimulus. In contrast, Plaquemine components along the western margin in the Catahoula Basin, Lower Ouachita, and Felsenthal region exhibit af¤liations with Caddo traditions (Fuller and Williams 1985; Gibson 1983, 1996:69–73; Gregory 1969; Gregory et al. 1987; Kidder 1990a, 1998b; Rolingson and Schambach 1981). A coastal Plaquemine culture can likewise be distinguished in regard to ceramics, settlement, and subsistence (Brown 1985b; Jeter and Williams 1989:212–220; Weinstein 1987a). Regardless of how we conceive of Plaquemine culture, greater attention to regional diversity promises to increase our understanding of those who were actively engaged in its production and dissemination (sensu Pauketat 2001a, 2001b). The intersection of practice theory and historical process has focused archaeological research on the political contexts of agency, material culture, and traditions (Dobres and Robb 2000:10–13; Pauketat 2000:114–117). At the same time, earlier explanatory models will continue to inform the direction of research. The presence or absence of shell tempering might be understood in this light not only as a technological choice or intrinsic cultural decision of potters but also in terms of long-distance contacts, migrations, intermarriages, ethnic identity, and resistance—as one facet of a historical process of Mississippianization (Livingood, this volume; Pauketat 2001b:81–83, 2004: 131–141). The majority of the debate over Plaquemine origins in the LMV has in retrospect been framed in the context of an older, essentialist culture historical approach, the same point of view that inspired James Ford to breathe life into the concept. Without undermining the culture historical foundation of Plaquemine, future research promises to enhance current knowledge in regard to historical ecology, subsistence economies, and political dynamics—the actual practices that produced and made up multiscalar cultural variability. Following in the footsteps of Mississippian archaeology, it is ¤tting that our understanding of Plaquemine diversity be more ¤rmly situated within a set of generalizations drawn from cultural ecology (e.g., Kottak 1999; Orlove 1980; Smith 1978). The hurricanes of 2005 that led to the evacuation of New Orleans and caused extensive damage to the Gulf Coast region are powerful recent reminders that the LMV and Gulf Coast region can be unpredictable

introduction a nd historic a l ov erv iew

13

environments with extremely volatile ecological processes (e.g., Saucier 1994: 321–322). A full understanding of Plaquemine culture must necessarily account for how people dealt with the abrupt changes and periodic threats of dynamic landforms. Hurricane Katrina, in particular, evokes anthropogenic factors and serves as a tragic reminder that the Mississippi River has become increasingly channelized since the ¤rst arti¤cial levees were built in the eighteenth century. Lower Valley residents took advantage of a dynamic riverine environment well before Mississippian times by hunting wetland game and harvesting ¤sh trapped in oxbow lakes replenished by annual ®oods. However, we have little understanding of how Plaquemine peoples or their progenitors coped with exceptional ®ooding events that on occasion must have affected thousands of square miles. Human–environmental relationships within any particular “riverine adaptation” (or variant thereof ) are negotiated in terms of socially and historically contingent landscapes (Crumley 1994:6, 11; Kidder 1998d; Patterson 1994; cf. Smith 1978). Just as recent anthropogenic processes are transforming the LMV (Saucier 1994:331–332), archaeologists should not assume indigenous populations were merely passive recipients of natural processes. Such regional and local-level cultural diversity and divergent historical trajectories defy straightforward eco-functionalist or systemic explanation (Anderson 1994b, 1999; Barker and Pauketat 1992; Cobb 1996; Pauketat 2001b). While it might be tempting to attribute such diversity at the macroscale to environmental differences, historical documentation hints at profound ethnic and linguistic variations within the LMV, distinctions undoubtedly transformed and trans¤xed by living peoples during the preceding millennium (Galloway 1998, 2002; Giardino 1984; Jeter 2002a; Kidder 1993b; Swanton 1911). One of the most valuable contributions in this regard involves problems of subsistence and subsistence economies (Fritz and Kidder 1993; Harmon and Rose 1989; Kidder 1992b; Kidder and Fritz 1993; Rose et al. 1984; Rose et al. 1991). In particular, the adoption of maize agriculture is no longer regarded as a causal mechanism in the evolution of Coles Creek social complexity (Kidder 1992a). Although maize has been recovered from Plaquemine components in the Lower Yazoo, Natchez Bluffs, and Tensas Basin, little is actually known about the relative dietary and economic signi¤cance of agriculture throughout the LMV. Furthermore, it is risky to generalize to other regions where data are sparser. The intensi¤cation of maize agriculture in the Tensas Basin was associated with the terminal Coles Creek–Plaquemine transition, raising additional questions regarding possible Mississippian contacts, the development of social inequalities, and political complexity (Kidder 1992a, 1992b; Kidder and Fritz 1993; Roberts 2004). Some Plaquemine groups may have continued in

14

rees a nd li v ingood

the Coles Creek tradition of focusing on an abundance of ¤sh, waterfowl, and shell¤sh (Fritz and Kidder 1993; Gregory 1969; Jeter and Williams 1989:215– 216, 219; Kidder et al. 1993; Miller et al. 2000:435–436; Rolingson and Schambach 1981; R. Smith 1996), while in some areas white-tailed deer and small mammals appear to have been relatively more important (Coxe and Kelley 2004; Springer 1980; Stevenson 1992). However tempting further generalization may be, it will be advantageous to situate new data on subsistence economies within the historical context of regional polities. One productive area of research will be to examine the evidence for feasting and food provisioning in such contexts (e.g., Beasley, this volume; Blitz 1993; Jackson and Scott 1995, 2003; Knight 2001; Pauketat et al. 2002; Welch and Scarry 1995). Finally, greater interest in political dynamics in the Mississippian Southeast during the past decade has paved the way for consideration of similar issues in the LMV (e.g., Anderson 1994a; Cobb 2003; Pauketat and Emerson, eds. 1997; Scarry, ed. 1996). Yet, in comparison, less research has focused speci¤cally on Plaquemine sociopolitical organization or dynamics. Much of what we know has been inferred from settlement patterns or extrapolated from documentary sources on early historic polities (e.g., Brain 1978; Brown 1985a, 1998c). The historic Natchez, with widely acknowledged Plaquemine associations, have informed current understanding of Mississippian polities (I. Brown 1990, this volume; Lorenz 1997; cf. Muller 1997:63–68). Recognizing the pitfalls of historic analogy (e.g., Galloway 1992, 1993), it would be bene¤cial to contrast the historical trajectories of regional Plaquemine polities with better-known Mississippian data sets (e.g., Knight 1997; Knight and Steponaitis 1998). A system of hereditary social ranking is thought by many to have emerged during the Coles Creek period in the LMV, subsequently elaborated upon by Plaquemine societies (Barker 1999; Kidder 1992a, 2002:79–90; Steponaitis 1986:385–387). Site hierarchies have been described for some polities, particularly in the Tensas Basin, Natchez Bluffs, and Lower Yazoo, suggesting multiple levels of political administration (Brain 1978; Kidder 1998b:143). Large mound sites appear to have become larger and less numerous from Coles Creek to Plaquemine times, possibly indicating a process of political centralization in which certain communities or factions were exerting political authority over larger territories (Brain 1978; Kidder 1992a, 1998b; Wells 1997, 1998). Kidder (1998b:137, 146) suggests that mound-and-plaza arrangements become increasingly exclusionary, indicating that elites were appropriating the symbols of public ceremony. Comparisons among Coles Creek, Plaquemine, and Mississippian polities stretch thin beyond this point. One of the greatest challenges in the LMV

introduction a nd historic a l ov erv iew

15

concerns the relative lack of archaeological evidence for preciosities or prestige goods and differential ranking in mortuary practices—the kinds of data that have proven helpful in past studies of ranked Mississippian societies (e.g., Pauketat 1997a, 1997b; Peebles and Kus 1977; Steponaitis 1991; Welch 1991). Another serious issue is the lack of comparable data sets in the LMV. Plaquemine as it is currently conceived may be too normative to encompass regional variation within the LMV. Part of the solution, then, will be to develop new explanatory models to account for this variation. Rather than approaching the Coles Creek–Plaquemine transition in terms of a Mississippian cultural expansion, it will be more productive to consider Plaquemine as a series of endogenous, local-level practices, in®uenced in some instances by agents of an analogous, macroscale tradition, what has been referred to as a historical process of Mississippianization (Pauketat 2001a, 2001b, 2004). This in turn raises the thorny question as to whether we might consider “Plaqueminization”— born of Coles Creek conservativism and traditions of resistance (see also Brown 1998e:657). The following chapters are arranged more or less geographically, with some re®ection of a chronological sequence (Figure 1.2). Beginning in the Tensas Basin of northeast Louisiana, a somewhat different scenario is emerging than across the river at Lake George and the Lower Yazoo Basin. Lori Roe addresses the Coles Creek antecedents of Plaquemine mound construction at the Raffman site. Greater attention to the Coles Creek–Plaquemine transition is particularly warranted in this region, where large mound centers were established during the Coles Creek period (Kidder 1998b:143). Roe argues for strong continuities in ceremony and ideology based on mound and plaza construction, architecture, and reuse. As a result, archaeologists working in the region must turn to Coles Creek rather than Mississippian societies in order to trace the antecedents of Plaquemine sociopolitical organization. At the Lake Providence Mounds, approximately 20 miles north of Raffman, Douglas Wells and Richard Weinstein have uncovered striking evidence for extraregional Mississippian contacts dating from the late Coles Creek transition. Presenting the results of a recent study by Coastal Environments, Inc., Wells and Weinstein demonstrate through sourcing analyses that ¤neware associated with the Preston phase (a.d. 1100–1250) included vessels that originated in the American Bottom region to the north. The expansion of interregional exchange in the development of a Lake Providence elite is suggested, although additional data are clearly needed. Subsequent Mississippian in®uence on Plaquemine society is regarded as largely indirect, via the Yazoo Basin, with Coles Creek society providing the basis for increased social inequalities and political centralization.

Figure 1.2. Areas discussed by individual authors.

introduction a nd historic a l ov erv iew

17

Moving south along the western edge of the LMV ®oodplain, Mark Rees investigates a series of little-known mound sites in the western Atchafalaya Basin, only 40 km (25 miles) west of the Medora site. Continuity between Coles Creek and Plaquemine is apparent in pottery types, a lack of shell tempering, and the spatial layout of mound groups. As this is far from resembling “Mississippianized Coles Creek,” an argument is made for endogenous political development in the context of regional conservativism. Possible connections between historic Chitimacha villages and ceremonial centers indicate that additional research has the potential to substantially expand on current understanding of regional political culture. A productive synthesis of culture historical and processual perspectives is advocated in this regard. Farther southeast in the Barataria Basin of coastal Louisiana, Malcolm Shuman draws attention to important yet poorly understood pottery assemblages on the north shore of Lake Salvador. From a vantage point deep in the Mississippi delta, the Plaquemine-Mississippian connection (divide?) seems more curious than ever, raising new questions concerning the apparent source, direction, and actual process of Mississippianization. As Shuman points out, Mississippian in®uence of some sort appears to have entered the LMV from the east, namely in the adoption of shell tempering at sites in the delta and Gulf coastal plain. Distinctions between Plaquemine and Mississippian are not so clear from the perspective of the Barataria phase, particularly if shell tempering is to serve as a diagnostic marker. Nor does there appear to have been a simple progression from grog-tempered to shell-tempered pottery types, raising fundamental questions regarding transitional Coles Creek. If such distinctions seem muddied this far south, it seems as though a careful reexamination of basic concepts is long overdue. Shuman also provides a compelling example of the dif¤culty in applying cultural generalizations about Plaquemine to its many diverse phases. Moving east across the Mississippi all the way to the Pearl River, Patrick Livingood examines the introduction of shell tempering as a basis for understanding Mississippianization. Using a digital image analysis technique, Livingood examines the paste recipes of Plaquemine potters at the Pevey and Lowe-Steen sites in south-central Mississippi. Pottery types based largely on grog or shell tempering do not separate into simple categories under digital image analysis but instead suggest the use of temper in paste recipes was a functional and aesthetic decision associated with household production. Conventional categories and assumptions must consequently be reexamined. On the Natchez Bluffs to the west, Virgil Beasley presents the results of the 1997 excavations at the Anna site, a major Plaquemine mound center and the

18

rees a nd li v ingood

type site for the Anna phase (a.d. 1200–1350). Long associated with ancestors of the historic Natchez, the Anna site has been described along with sites such as Emerald, Foster, and Fatherland as representative of the shifting political fortunes of Plaquemine chiefdoms on the Natchez Bluffs (Brain 1978; Brown 1985a, this volume; Neitzel 1965). Investigation of architectural remains in a mound precinct at Anna yielded associated ceramic and faunal deposits. Along with architectural evidence for a small scaffold or screen, Beasley argues, these may represent the remains of a ceremonial feast as described in early historic accounts of the Natchez. Drawing upon multiple lines of evidence, the archaeological interpretation of ritual feasting holds enormous potential for understanding the relationships between subsistence economy and regional political dynamics (e.g., Jackson and Scott 1995, 2003). Plaquemine culture in the Natchez Bluffs region of Mississippi is revisited by Ian Brown, who summarizes what is presently known about ethnohistoric connections with the Natchez, political development, and mound construction. He argues that large mound sites such as Anna and Emerald represent the culmination of historical trajectories that produced shifting regional consolidation, alliances, and hierarchies. Importantly, he points out that towns with mounds and dispersed households were the main components of Plaquemine settlement hierarchy on the Natchez Bluffs. Unlike previous Coles Creek or contemporaneous Mississippian societies, there do not appear to have been large villages without mounds. More controversial, the evidence for mound construction points toward Plaquemine rather than Coles Creek origins at single-component and multicomponent sites. Discontinuities in architecture and pottery lead Brown to further surmise a decisive break between Coles Creek and Plaquemine habitation in the Natchez Bluffs region. As to exactly what these discontinuities mean will require further study, yet comparison with other Plaquemine regions is striking, particularly in the context of a Coles Creek transition (e.g., Rees, this volume; Roe, this volume; Shuman, this volume; Wells and Weinstein, this volume). Moving boldly upriver to the “outer limits” of Plaquemine culture, Marvin Jeter reexamines Plaquemine from the perspective of southeast Arkansas or the “northerly borderlands.” The noticeable omission of recent research from the Lower Yazoo Basin, so prominent in the literature, may appear incongruous (see above). Yet in terms of exactly how Plaquemine is to be reconceptualized, the Lower Yazoo Basin looms large (e.g., Brain 1978, 1989; Phillips 1970; Williams and Brain 1983). In a wide-ranging essay, Jeter provides an introduction to Plaquemine in southeast Arkansas and examines its manifestation in the Bellaire, Bartholomew, and Gran Marais phases. On the basis of

introduction a nd historic a l ov erv iew

19

a careful reexamination of the evidence, Jeter concludes that the Bellaire phase in the Bartholomew-Macon region, tentatively identi¤ed by Phillips (1970), might be discarded or subsumed under Mississippian culture. The substantive evidence for the Bartholomew and Gran Marais phases is likewise summarized in terms of pottery, lithics, site and settlement data, and subsistence. Jeter addresses regional variation in Plaquemine societies in southeast Arkansas, including its antecedents and descendants, and concludes with a compelling argument for connections with ethnohistorically documented groups of Native Americans. The volume closes with T. R. Kidder “contemplating Plaquemine culture,” returning to several of the themes outlined in earlier chapters. Kidder summarizes by emphasizing the local Coles Creek antecedents of Plaquemine societies and drawing attention to political, economic, and religious practices in the LMV after a.d. 1000, historical variation that he argues cannot adequately be accounted for by “unspeci¤ed Mississippian in®uences.” A more nuanced understanding of regional political consolidation in the LMV after a.d. 1200 will likewise require greater attention to both the scale and source of interactions. Turning the familiar homology upside down, if the cultural heirs of Coles Creek in the LMV were in fact informed of the imposing Mississippian polities upriver and to the east, what conditions and events, what actions would have entailed their being or remaining Plaquemine? As the following authors demonstrate, the dif¤cult work needed to answer these and other questions has just begun.

Acknowledgments The chapters in this volume were originally presented in the symposium “Plaquemine Problems: Recent Investigations and Reappraisals” at the 60th annual meeting of the Southeastern Archaeological Conference in Charlotte, North Carolina, on November 12–15, 2003. We would like to thank the staff at the University of Alabama Press for their guidance, assistance, encouragement, and patience. We are indebted to Kathy Cummins, who edited the entire manuscript and corrected numerous mistakes. We would also like to extend a special thanks to those contributors who were personally affected by hurricane Katrina. The hurricane hit during the ¤nal revisions to the manuscript but did not prevent any authors from completing their revisions and submitting their chapters. We are also grateful for the support of our families, especially Johanna Rees and Susannah Livingood.

2 Coles Creek Antecedents of Plaquemine Mound Construction Evidence from the Raffman Site Lori Roe Plaquemine culture (ca. a.d. 1200–1700) developed in the Lower Mississippi Valley (LMV) at a time when the Mississippian cultural tradition dominated most of southeastern North America. Plaquemine culture shares traits such as maize horticulture and hierarchical social organization with Mississippian cultural variants throughout the Southeast (Brain 1989:133; Kidder 1993a). However, Plaquemine culture maintains a distinct LMV ®avor, and differences between Plaquemine and Mississippian iconography and technology continue in some areas into historic times (Kidder 2004b). The local ancestor of Plaquemine culture is the Late Woodland Coles Creek culture (ca. a.d. 700– 1200). While there is continuity between Coles Creek and Plaquemine culture, important changes in subsistence and sociopolitical developments take place during the transition from Coles Creek to Plaquemine. Evidence of interaction with Mississippian cultural groups, in the form of ceramic and lithic artifacts, is found at some Coles Creek sites in the LMV (e.g., Lake Providence, Lake George, and Winterville) beginning around a.d. 1000 (Brain 1989; Phillips 1970; Williams and Brain 1983; Wells and Weinstein, this volume). The impact this cross-cultural contact had on the origin of Plaquemine culture is debated by archaeologists and appears to have varied in different subregions of the LMV. Research in the Yazoo Basin, notably work by Williams and Brain (Brain 1989; Williams and Brain 1983), led to the idea that Mississippian contact was instrumental in the development of Plaquemine culture. However, some archaeologists working in the Tensas Basin (e.g., Hally 1972; Kidder 1993a) argue that evidence for Mississippian in®uence on the development of Plaquemine culture is equivocal. Disagreement over the nature of the preceding Coles Creek culture contributes to the debate over Plaquemine origins. Williams and Brain (1983:405) see the Coles

ev idence from the r affm a n site

21

Creek culture as relatively static, with early innovations followed by nearly 500 years of “conservative resistance to major change.” However, changes in settlement patterns, increases in the scale of earthen mound construction, and concomitant alterations of mound-and-plaza site plans have been offered as evidence that Coles Creek polities were becoming larger and more hierarchically organized over the span of the Coles Creek period (Kidder 1992a, 1998b, 2002:87–88; Wells 1997). In order to understand Plaquemine culture, we must examine the developments in the preceding Coles Creek period and reevaluate the nature of interactions between Coles Creek and external Mississippian groups. Earthen mound architecture has factored strongly in interpretations of both Coles Creek and Plaquemine sociopolitical organization, and continuities in Coles Creek and Plaquemine mound architecture are the focus of this chapter. In particular, results of research at the Raffman site (16MA20), an impressive Coles Creek mound-and-plaza center that exhibits strong similarities with later Plaquemine mound centers, are presented (Figure 2.1). Raffman is much larger than most contemporary mound centers in the Coles Creek region and does not ¤t the “typical” Coles Creek site plan of two to four modest-sized mounds (Williams and Brain 1983:405). The massive scale of construction at Raffman and a large and rapid expansion of the site around a.d. 1000 suggest that the Raffman vicinity may have been an early locus of political complexity in the region and support the idea that signi¤cant sociopolitical developments occurred during Coles Creek times. There is a long history of platform mound construction in the LMV. The ¤rst substructural platform mound-and-plaza precincts in the eastern United States were constructed by Coles Creek cultural groups (Knight 2001; Williams and Brain 1983:405). By Plaquemine culture times, platform moundand-plaza precincts clearly served as ceremonial and political centers (Brain 1978, 1988; Neitzel 1965; Quimby 1951). Continuities in this form of monumental architecture suggest that the ceremonial and, presumably, sociopolitical organization of Plaquemine began developing during the preceding Coles Creek period. The facts that Raffman mound construction and use predates Plaquemine culture and that it shows no direct evidence of Mississippian in®uences are signi¤cant for interpretations of indigenous cultural development and cross-cultural interactions in the LMV.

Coles Creek Culture The Coles Creek period (a.d. 700–1200/1250) marks an important transition in social organization in the LMV from the relatively egalitarian cultures of

Figure 2.1. Core Coles Creek region and locations of most sites discussed in the chapter.

ev idence from the r affm a n site

23

Figure 2.2. Middle Woodland through historic phase chronology for the central Lower Mississippi Valley (after Kidder 1992a:Figure 8.2).

the preceding Baytown period (a.d. 400–700) to the hierarchical chiefdoms that prevailed among the Mississippi period Plaquemine culture (a.d. 1200/1250–1450). Coles Creek culture is found throughout much of the LMV from the mouth of the Arkansas River to the Gulf of Mexico. It thrived in the Yazoo, Tensas, and Ouachita River basins, in the Natchez Bluffs region, and in coastal Louisiana (Figure 2.2). Coles Creek occupations are most commonly identi¤ed by their ceramics—grog-tempered wares with decorative motifs and temper characteristics

24

lori roe

that distinguish them from earlier types. Ceramics are by far the most abundant artifacts recovered from Coles Creek sites. Settlement types included camps, hamlets, villages, and mound centers (Kidder 1992a; Wells 1997). Ceremonial site architecture, consisting of ®at-topped mounds constructed around level plazas, is found throughout the Coles Creek region. Population in the region increased during the Coles Creek period, measured by increases in overall site size and density (Kidder 2002). Coles Creek subsistence was assumed to be based on farming of maize and other tropical and native cultigens because settlements are typically located on arable lands of major river valleys (Williams and Brain 1983:408). Research by Fritz and Kidder (Fritz and Kidder 1993; Kidder 1993a; Kidder and Fritz 1993), though, shows that maize was not a signi¤cant part of the diet until at least a.d. 1200. Even cultigens of native plants that were farmed elsewhere in the Mississippi River valley, such as maygrass and chenopod, are largely absent from Coles Creek sites south of the Arkansas River. Recent research indicates there is considerable plant food subsistence variability among Coles Creek peoples, with contemporary groups living in close proximity practicing different strategies (Roberts 2004). Evidence for increasing social complexity and the emergence of social elites in Coles Creek societies includes construction of platform mounds that may have served as the location for elite residences; mound architecture that suggests restriction of access to the mound-and-plaza precinct; a high degree of community planning; hierarchical settlement patterns; and, to some degree, mortuary patterns (Barker 1999; Belmont 1985; Kidder 1992a, 1998b, 2004b; Steponaitis 1986:385–386; Wells 1997; Williams and Brain 1983:370). However, status and wealth distinctions are not seen in accumulations of exotic trade goods or other prestige items or in elaborate tomb preparations and grave offerings. Coles Creek material culture is fairly mundane; ornate special-use or ceremonial objects are extremely rare. Few exotic items are found in Coles Creek contexts and the large majority of artifacts were made on local materials, suggesting that extraregional trade was uncommon (Kidder 2002:89). Excavation of Coles Creek mounds at sites such as Greenhouse, Lake George, and Morgan revealed that these features were often constructed in multiple stages. Perishable structures were erected on mound summits and burials were often interred in the mounds (Ford 1951; Fuller and Fuller 1987; Williams and Brain 1983). Mound summit structures have been interpreted as residences or charnel houses, on the basis of associated midden deposits or burials, though further research is needed to determine what functions these structures served (Steponaitis 1986:385). Steponaitis (1986:385–386) observes

ev idence from the r affm a n site

25

that the construction of multistage platform mounds that appear to have served as residences and/or burial locations is reminiscent of platform mounds known to have been used by generations of elites in Mississippian and protohistoric chiefdoms. He suggests that Coles Creek platform mounds may represent the emergence of hereditary social ranking, with each mound stage being constructed for and used by a new generation of chiefs (Steponaitis 1986:386). Excavations at sites such as Greenhouse, Mount Nebo, and Lake George have demonstrated that mound centers were used and expanded over hundreds of years. Greenhouse and Mount Nebo were mortuary ceremonial centers during the Baytown period and continued to be used for much of the Coles Creek period (Belmont 1967; Ford 1951; Giardino 1977; Kidder 1998b; Kidder and Fritz 1993). Mounds were constructed over existing structures and burial tumuli, eventually resulting in multiple mounds demarcating a central plaza. The Lake George site was also occupied during the Baytown and Coles Creek periods and had a remarkable expansion during the initial phases of Plaquemine culture in the Yazoo Basin (Williams and Brain 1983). The continued use or reuse of the same sites as ceremonial centers suggests continuity of cultural heritage and the importance of ancestral social and ceremonial centers. Coles Creek mound centers were often constructed over Baytown period mortuary centers. Steponaitis (1986:386) asserts that Coles Creek leaders began to transform these centers—which had previously been used for communal burial and ceremony—into symbols of elite power. Kidder (1992a) further contends that by constructing residential and burial mounds over ancestral burial locations, incipient Coles Creek elites symbolically and physically linked themselves with locations that were of the utmost social and ceremonial importance. The aggrandizement of some Coles Creek sites as the Plaquemine culture emerged may represent a similar legitimization of elite power and desire to perpetuate ancestral ceremonial ties. The growth of mound centers, through construction of new mounds and enlargement of existing mounds, led to mound-and-plaza precincts that were increasingly demarcated from surrounding areas (Kidder 1998b). As opposed to there being a single mound adjacent to an open plaza, multiple mounds were constructed and these more tightly delineated the boundaries of the mound-and-plaza precinct. Kidder (1998b:138–139) interprets this physical bounding as an indication that access to mound-and-plaza precincts was becoming more restricted and that the activities that took place at these precincts were becoming more clearly distinguished from activities that took place in the villages and hamlets where the large majority of the population lived. This

26

lori roe

architectural trend begins in the Coles Creek period, lending support to the idea that mound centers became associated with social elites during this time (Kidder 1992a, 1998b; Steponaitis 1986:385–386; Williams and Brain 1983). The increased scale of mound construction indicates that these centers played a more important role in surrounding communities and that social leaders were able to coordinate greater labor efforts for mound construction.

The Raffman Site Location and Description Unlike many Plaquemine and Mississippian mound centers, Raffman is not situated near a major river (see Brain 1978). The site is located near the west bank of Tensas Bayou, in Madison Parish, northeast Louisiana (Figure 2.1). The Raffman mounds were constructed at the edge of an elevated ®ood deposit created by a crevasse in Joes Bayou to the west. Floodwaters and sediments broke through the natural levee on Joes Bayou and formed an elevated, fan-shaped splay (Kidder 2004a:519; Saucier 1994:272–273). The great construction efforts expended to create the Raffman site suggest that it was an important ceremonial and possibly political center in its region. The mounds at Raffman are closely spaced around a central plaza that measures approximately 80–90 m wide by 100–110 m long (Figure 2.3). Seven of these mounds, Mounds A–G, delineate the plaza. The remaining two mounds, Mounds H and I, are situated to the southeast of the plaza. The largest mound (Mound A) is a ®at-topped, rectangular feature, approximately 12 m tall and 75 m by 100 m at its base. It is located at the southern edge of the plaza. Flanking the northwest and northeast corners of Mound A are two relatively large platform mounds, labeled B and G. Four more mounds (C, D, E, and F) complete the boundary of the central plaza. Mounds A, B, C, G, and I have well-de¤ned, ®at summits. Mounds E and F have low, slightly rounded summits, and Mounds D and H are conical. Because of the size, shape, and arrangement of earthworks at Raffman, the site was considered by some to be a Mississippi period site, associated with the Plaquemine culture (Kidder 1998b:143; Phillips 1970:Figure 447). Investigations at Raffman, however, revealed that the earthworks are primarily Coles Creek period constructions. These excavations also demonstrated that, in addition to mound construction, extensive landscaping was involved in the creation of the mound-and-plaza precinct (Kidder 2004a; Kidder and Roe 1999). Additionally, excavations in mound summits and slope midden deposits have uncovered cultural remains associated with mound activities.

ev idence from the r affm a n site

27

Figure 2.3. The Raffman site, showing locations of 1998 through 2004 excavations. Mound B excavations shown in Figure 2.4.

Research at Raffman Located in a large, dense tract of woods, the Raffman mounds are dif¤cult to access. Raffman’s isolation has made research at the site challenging but has also contributed to the site’s excellent preservation. Raffman was ¤rst recorded as an archaeological site by Phillips and Neitzel in 1954. Archaeological investigations at the site over the next few decades were extremely limited. In 1981, Williams, Belmont, and Kardish mapped the site with a tape and compass and excavated shovel tests (Belmont 1985:272).

28

lori roe

In the summer of 1998, Kidder initiated more extensive investigation of the Raffman site. These investigations were continued during the summers of 2000, 2002, and 2004. Initial research goals included establishing a basic occupation sequence for the site, uncovering evidence of how and when the mounds were constructed, and examining the geomorphology of the Raffman vicinity. During the initial ¤eld season, a single 1-by-1-m excavation unit was placed in the slope of each mound and two units were placed in the plaza. Solid soil cores extracted from the mounds and plaza provided further information about site stratigraphy. In addition, two backhoe trenches were excavated. The “north trench” was located at the edge of the raised landform on which Raffman is situated and the “south trench” was located at the southwest toe of Mound B (Figure 2.3) (see Kidder 2004a). Later ¤eldwork has been focused on excavation of mound summits and slopes to examine midden deposits and activity areas on the mounds. Additional soil cores were extracted from the mounds and from non-mound contexts; these cores have proved to be a valuable source of information about the construction of the site.

Construction Chronology and Mound Use The Raffman site is large and stratigraphically complex; mound construction and landscaping at the site are extensive. Additional excavations and analyses will be necessary to provide more details about the exact sequence of mound construction and the function of each mound at Raffman. However, ¤eld investigations and analyses of cultural remains to date have provided a framework for the chronology of construction and occupation at Raffman. Examination of cultural remains from Raffman, especially those associated with mound activity areas, is ongoing. Available data do not indicate how each mound at Raffman was used. It is clear, though, that not all the mounds were used for the same purpose or activities. Mounds at Raffman vary in terms of size, shape, stratigraphy, and the presence or absence of associated occupation debris. Following is a discussion of the current data on site chronology and mound function. Cultural remains at Raffman indicate occupation during the Tchula (ca. 500 b.c.–a.d. 1) and Marksville (ca. a.d. 1–400) periods, prior to construction of the mounds. There is a notable absence of cultural remains that can be con¤dently dated to the Baytown period (ca. a.d. 400–700). A few surfacecollected ceramics dating to the Fitzhugh phase of the Plaquemine culture (ca. a.d. 1400–1500) suggest an ephemeral occupation during the Mississippi period (Kidder 2004a:523, 525). Coles Creek cultural components at Raffman include early (ca. a.d. 700–1000) and late (ca. a.d. 1000–1250) phases. Re-

30

lori roe

search thus far indicates that all mound construction, with the possible exception of Mound I, dates to the Coles Creek period (Kidder 2004a:519–526). Excavations were conducted on the summit of Mound A, and a single excavation was located on the lower northwest slope (Figure 2.3). No structural remains or other occupation debris was uncovered on the mound summit. A mound construction stage that contained relatively few artifacts was situated immediately underneath the humus stratum. Hand augering on the summit revealed that this mound construction stage was at least 2.5 m thick. Likewise, excavation on the slope of the mound did not reveal any evidence of occupation debris associated with the mound (Kidder 2004a:519). Though three construction episodes were identi¤ed on the basis of color and texture differences, there was no evidence of occupation debris or erosion between these mound ¤ll strata (Kidder 2004a:519). Mound A is a very large earthwork, and excavations to date cannot rule out the possibility that earlier mound stages were used for residence, ritual, or burial. However, no evidence for activities associated with this mound has been uncovered. Excavations did uncover a sub-mound midden, though. A radiocarbon date from charcoal near the surface of this midden deposit places the ¤nal accumulation of the midden around a.d. 700–750 (Table 2.1). Ceramics from the midden are in line with this date, including early Coles Creek period varieties of Baytown Plain, with two early Coles Creek period decorated sherds (French Fork Incised, var. Larkin and Coles Creek Incised, var. unspeci¤ed ) (see Phillips 1970:69–70, 85). There is no indication that a substantial amount of time elapsed between the ¤nal deposition of this midden and the initial construction of Mound A (Kidder 2004a:519). Ceramics from sub-mound middens also suggest that construction of Mounds B and C was begun in the early Coles Creek period. Extensive excavations have been conducted in Mound B, and the results of these investigations are discussed in the following section. Mound C has had only limited testing, consisting of a single excavation on the slope of the mound, soil cores extracted from the mound, and probing for occupation debris on the summit using an Oak¤eld soil corer. Ceramics from midden underlying Mound C are similar to those found under Mound A and indicate a contemporary occupation (Kidder 2004a:519). A radiocarbon date from this sub-mound midden places the construction of Mound C around a.d. 800–900 (Table 2.1). Mound C appears to have been constructed in a single episode; no occupation debris has been located on the mound slopes or on the ®at summit. Mounds D, H, and I have also only been subjected to limited testing. Available data suggest that Mounds D and H were constructed during the Coles Creek period and that Mound I was constructed perhaps during termi-

ev idence from the r affm a n site

31

nal Coles Creek period or Plaquemine times. Mound D overlies a midden that contains a few Marksville decorated sherds and plainwares that appear to date from the Marksville, Baytown, and/or early Coles Creek periods. Mound D construction ¤ll contains early Coles Creek decorated sherds. A single stage of construction was identi¤ed during excavation, and no midden deposits were uncovered on the surface of this dome-shaped mound. Mound H overlies a Tchula and Marksville period midden, and the ¤ll of the mound contains Coles Creek ceramics. No occupation debris was uncovered on the mound surface. Mound I overlies a midden that contains ceramics associated with the terminal Coles Creek and Plaquemine periods (Baytown Plain, var. Addis, Baytown Plain, var. unspeci¤ed, and a single sherd of Plaquemine Brushed, var. unspeci¤ed ). It is unclear whether prehistoric occupation debris is associated with the mound surface. The upper surface of the mound was disturbed by the construction of an early nineteenth-century house. At approximately a.d. 1000, a large expansion of Raffman was undertaken. The elevated crevasse splay on which Raffman is situated was arti¤cially extended to the northeast. This landscape modi¤cation served to enlarge the plaza and created space for additional mound construction (Kidder 2004a). Stratigraphy in soil cores and the north backhoe trench revealed that an arti¤cial platform, measuring approximately 100 m long, 40 m wide, and 3–4 m thick, underlies Mounds E and F (Kidder 2004a:520–526). A radiocarbon date from charcoal near the base of the arti¤cial platform and another from midden overlying the slope of Mound E indicate that the platform was rapidly constructed (Table 2.1) (Kidder 2004a:519–520). Excavation and soil cores suggest that Mound E was constructed in two small stages and that occupation debris is associated with each stage (Kidder 2004a:520). A sub-mound midden was uncovered in excavations of Mound F, but no midden deposits were recovered from the surface of the mound (Kidder 2004a:520). The eastern portion of Mound G appears to overlie an arti¤cial surface that was constructed to extend the natural ground surface to the east. Excavations and soil cores have revealed that the western portion of the mound overlies a midden containing Tchula and Marksville period ceramics, and the eastern portion rests on over 2 m of loaded soil. It is not known whether this extension of the natural ground surface was constructed rapidly or whether it is part of the arti¤cial platform constructed on the northeast side of the plaza. A radiocarbon date from midden overlying the ¤rst construction stage of Mound G places the initial use of this mound in the Coles Creek period, but the date has a large standard deviation (Table 2.1). Mound G was constructed in at least two stages, and occupation debris has been recovered from the surface of each stage.

32

lori roe

Occupation at Raffman continued until the end of the Coles Creek period. Radiocarbon dates and ceramics from occupation debris on the summit and east slope of Mound B indicate that the mound was used until approximately a.d. 1100–1200 (Table 2.1). As discussed above, the erection of Mound I appears to be the ¤nal episode of mound construction at Raffman. Whether this mound was constructed after a hiatus of occupation at the Raffman site is not known.

Mound B Investigations Mound B is the most extensively excavated mound at Raffman (Figure 2.4). This platform mound is approximately 2.5 m tall, with a large square summit approximately 25 m on each side. The mound was constructed in two distinct stages, and occupation debris overlies each of these mound stages. The initial mound stage (stage B-1) was constructed and used during the early Coles Creek period, and the second stage (stage B-2) was used during the late to terminal Coles Creek period. Mound stage B-1 appears to be a small, low platform. The summit and one slope of this platform were encountered during excavations on Mound B. Construction of mound stage B-2 greatly increased the size of the mound summit. Most ceramics recovered from the surface of mound stage B-1 are Baytown Plain. Among the few decorated sherds recovered from the surface of this mound stage are examples of Coles Creek Incised, var. Coles Creek and Mulberry Creek Cord Marked, var. Smith Creek. The remains of a few small posts (10–13 cm in diameter) were uncovered on the surface of this ¤rst mound stage, but no pattern of posts was discernable. Midden deposits and structural remains have also been excavated from the summit and east slope of mound stage B-2, the ¤nal construction stage. Remains of posts, ranging in size from 15 to 30 cm in diameter, are concentrated in the southeast quarter of the summit. The size and shape of the structure or structures represented by these posts are not discernable. A possible hearth was uncovered in the southeast quarter of the summit as well. Midden deposits on the mound contained abundant ceramics and faunal remains and smaller amounts of lithics. Analysis of the cultural remains recovered from Mound B is in progress. Initial examination of these remains suggests that Mound B may have been a residential mound. Three radiocarbon dates from midden deposits on the east slope and summit of mound stage B-2 indicate that this ¤nal stage of the mound was used circa a.d. 1100–1200. A radiocarbon date from charred wood recovered near the base of one of the post molds on the Mound B summit indicates a some-

ev idence from the r affm a n site

33

Figure 2.4. Mound B at Raffman, showing locations of excavations and soil cores.

what earlier occupation, though the 2-sigma ranges of the dates overlap (Table 2.1). Ceramics recovered from midden deposits on the mound surface also indicate a late to terminal Coles Creek occupation of the mound. The majority of ceramics are markers of the Balmoral phase (a.d. 1000–1100), a late Coles Creek phase in the Tensas Basin. Commonly represented decorative types include Coles Creek Incised, vars. Coles Creek and Mott, Beldeau Incised, var. unspeci¤ed, and French Fork Incised decorative motifs on wares with ¤ne grog tempering. There are also some examples of decorative and paste varieties that are typically associated with the Preston phase (a.d. 1100– 1200), considered a transitional phase between the Coles Creek and Plaquemine cultures in the Tensas Basin. These include sherds classi¤ed as Carter Engraved, var. unspeci¤ed, Mazique Incised, var. Preston, and a few sherds of Baytown Plain, var. Addis. The radiocarbon dates from Mound B surface midden and the presence of some Preston phase ceramic markers indicate that Mound B was used during this culturally transitional phase. However, the ceramics from Mound B midden deposits do not resemble a typical Preston phase assemblage (see Kidder 1993a:75–109; Ryan 2004). The ceramic assem-

34

lori roe

blage from Mound B is weighted toward the earlier Balmoral phase. The midden does not appear to be strati¤ed, with earlier ceramic types more abundant in lower levels and later types more abundant in upper levels. In addition, dates from two different levels within the midden are essentially identical.

Social Implications of Raffman Mound Architecture There is a notable lack of direct evidence for contact with Mississippian groups at Raffman. There are no shell-tempered ceramics or imported ceramic styles. Rather, the combination of radiocarbon dates and ceramics from Mound B (stage B-2) suggests that the occupants at Raffman were somewhat conservative in terms of ceramic technology and style. Raffman demonstrates that large, impressive mound sites existed in the LMV without obvious in®uences from Mississippian groups. The scale of mound construction and landscaping at Raffman is far greater than expected for the petty chiefdoms thought to characterize the sociopolitical landscape of the Coles Creek period. Large Coles Creek sites like Raffman, and perhaps Mott (16FR11) as well (see Barker 1992:15, 42; Morgan 1999), may represent the centers of large hierarchically organized Coles Creek societies. The scale and arrangement of mound-and-plaza architecture at Raffman suggests an elite arena, physically separated from the community as a whole (Kidder 1998b). Substantial effort was expended to arti¤cially extend the natural ground surface in order to construct more mounds around the plaza, indicating that the con¤guration of the site was of considerable importance. Furthermore, elite leaders may have been living on some of the Raffman mounds, aligning themselves with ceremonial locations and setting themselves apart from the rest of the community. The large-scale, rapid expansion of the mound-and-plaza precinct circa a.d. 1000 suggests that Raffman’s position in the sociopolitical landscape had shifted. Raffman had presumably become an important ceremonial and political center. The grandeur of the site would have commanded respect for the leaders who coordinated its construction and perhaps for all communities interacting with the Raffman site.

Comparison of Mound Architecture The mounds at Raffman are similar in size and con¤guration to Plaquemine sites in the Tensas Basin, such as Routh and Fitzhugh (Kidder 1998b). Raffman, Routh, and Fitzhugh each have a preeminent truncated pyramidal mound, with other large, relatively closely spaced mounds arranged around a single plaza (see Gibson 1996; Hally 1972; Morgan 1999).

ev idence from the r affm a n site

35

While ceremonial activities presumably took place at Coles Creek mound centers, the centers do not appear to have supported large resident populations. Late Coles Creek mound sites are not necessarily “vacant” ceremonial centers; however, those sites that have been investigated do not have extensive occupation debris on or between the mounds (Kidder 2002; Williams and Brain 1983). During the Coles Creek period, occupation seems to have shifted away from the mound-and-plaza area (see Kidder 1992a, 1998b). While nearby settlements have been discovered at late Coles Creek mound sites such as Mott (Barker 1992:55–59), occupation on the mounds appears to be restricted to fewer individuals, presumably civic and/or religious leaders (Kidder 2002; Williams and Brain 1983:407). The relative scarcity of occupation at the mound-and-plaza precinct continues into Plaquemine cultural times. While Plaquemine mound sites often supported larger resident populations than their Coles Creek predecessors, most do not show evidence of intensive occupation, unlike Mississippian mound centers to the north (Brain 1978; Brown 1985b; Hally 1972). Relative to its size, the Raffman site does not have much occupation debris on or between the mounds. In addition, soil coring outside the mound-andplaza precinct combined with surface survey in plowed agricultural ¤elds has yet to reveal a residential area near Raffman. However, the entire landscape around Raffman was blanketed with between 30 and 70 cm of ¤ne-grained sediment as a consequence of postoccupational ®ooding and thus traditional site survey is not likely to provide good evidence of Coles Creek settlement patterning. The Routh site, a Tensas Basin mound site that has components spanning the late Coles Creek and Plaquemine cultures, also has few midden deposits in and around the mound-and-plaza precinct (Hally 1972:228, 243). The Fitzhugh mound site, another Plaquemine mound site in the Tensas Basin, has more occupation debris than Routh but still does not appear to have been a residential location for a large population (Hally 1972:137).

Contact with External Mississippian Groups Interaction between Coles Creek and Mississippian groups in the LMV appears to have varied between and within geographical subregions and was not necessarily correlated with the size or complexity of the Coles Creek polities involved in the interaction. Mississippian pottery and projectile points from the Central Mississippi Valley and American Bottom are found at Coles Creek settlements in the Yazoo Basin. Evidence of interaction with Mississippian groups is well known from the Lake George and Winterville mound sites in the Yazoo Basin (Brain 1989; Williams and Brain 1983). Large construction efforts and alterations to the site layout are evident at these sites after contact

36

lori roe

with Mississippian groups began and have been attributed to Mississippian cultural in®uences (Brain 1989; Williams and Brain 1983). There is less evidence for Mississippian contact in other regions of the Coles Creek culture area, and some archaeologists (Brain 1978; Hally 1972; Kidder 1993a) have argued that Mississippian in®uences on the development of Plaquemine culture in the Yazoo Basin were much more pronounced than in the Tensas Basin or Natchez Bluffs regions. Changes in mound-and-plaza architecture associated with the transition from Coles Creek to Plaquemine cultures are less dramatic in the Tensas Basin, where Coles Creek sites like Raffman and Mott are very similar architecturally to Plaquemine sites like Routh and Fitzhugh. Recent excavations led by Wells and Weinstein (this volume) at the Lake Providence mound site, however, have demonstrated that direct Mississippian contacts did occur in the Tensas Basin. These excavations uncovered ceramics from Cahokia as well as locally produced imitations of Mississippian pottery. The Lake Providence site is located approximately 20 miles north of Raffman. Trade goods and local copies of foreign ceramics are associated with a Preston phase, or terminal Coles Creek, occupation at Lake Providence (Weinstein 2005; Wells and Weinstein, this volume), indicating temporal overlap in Raffman and Lake Providence occupations. However, while extraregional interactions may have played a role in sociopolitical developments at Lake Providence (Wells and Weinstein, this volume), there is no indication that Raffman’s leaders were pursuing similar contacts and alliances.

Discussion Strong continuities are evident between the Coles Creek and Plaquemine cultures. Coles Creek antecedents to Plaquemine mound construction suggest hierarchical social organization and hereditary rights to leadership may have developed among Coles Creek societies. The great scale of mound construction and landscaping at the Raffman site and the presence of a possible residential structure on the Mound B summit support the claim that social elites commanded control of large communities in some Coles Creek societies. Continuities in ceremony and ideology are indicated by similarities in mound-and-plaza architecture and mound construction techniques and by reuse of the same locations for ceremonial precincts. The relative lack of occupation at Plaquemine mound centers is reminiscent of Coles Creek mound sites, not Mississippian sites. However, there is an apparent increase in population at some Plaquemine mound sites compared with Coles Creek sites, and this change may re®ect in®uences from Mississippian societies. Changes in subsistence and extraregional exchange are perhaps the core

ev idence from the r affm a n site

37

developments of the Plaquemine culture. Plaquemine settlement patterns in the Yazoo Basin appear to re®ect the increased importance of maize as well as the importance of access to and control of major water transportation routes (Brain 1978). However, few cultural developments and transitions are seen universally in the entire Plaquemine cultural area. Mississippian technologies such as shell-tempered pottery and wall-trench houses were not universally adopted in the LMV during Plaquemine times. Even societies that adopted these technologies often continued to use traditional LMV ceramic and structural technologies as well (Brown 1985b; Hally 1972; Williams and Brain 1983). The distribution of shell-tempered pottery and Mississippian projectile points at LMV sites demonstrates varying degrees of interaction and exchange between Coles Creek and Mississippian cultural groups. There is little doubt that contact with Mississippian groups in®uenced the development of Plaquemine culture. However, possible antecedents to Plaquemine social organization and ceremonial activities are indicated by the architecture and landscaping of large Coles Creek mound complexes like Raffman. Coles Creek groups were probably not passive recipients of Mississippian cultural in®uences. Changes in the Coles Creek political landscape may have led some groups to seek extraregional alliances. Additionally, subsistence changes may have both facilitated and required trade. As Hally (1972) has suggested, goods and presumably ideas were traveling up as well as down the Mississippi during this period of change in the LMV.

Acknowledgments My research at Raffman was conducted with grants from the Newcomb Foundation, the Tulane University Department of Anthropology, and the Anderson-Tully Corporation. T. R. Kidder provided me the opportunity to work at Raffman and has offered immeasurable support and advice. Dennis LaBatt, David Grif¤ng, and the rest of the staff at the Poverty Point State Historic Site provided housing, research facilities, support, and hospitality. I am grateful to them and the Louisiana Of¤ce of State Parks. E. Thurman Allen, Chip McGimsey, and Joe Saunders supplied soil coring equipment and worked hours extracting and interpreting soil cores from Raffman. T. R. Kidder and Anthony Ortmann volunteered their manual labor and expertise for ¤eld research and offered editorial comments for this chapter. Finally, I am indebted to all the students from Tulane University, Washington University, and the University of Alabama who assisted with ¤eld and lab work. Without their efforts, excavation of Raffman would not have been feasible.

3 Extraregional Contact and Cultural Interaction at the Coles Creek– Plaquemine Transition Recent Data from the Lake Providence Mounds, East Carroll Parish, Louisiana Douglas C. Wells and Richard A. Weinstein The Lake Providence Mounds site (16EC6) is located in the upper Tensas Basin, 4.5 miles north of the town of Lake Providence, East Carroll Parish, Louisiana (Figure 3.1). What remains of the site lies beneath and immediately adjacent to the west side of the modern Mississippi River levee (known in the area as the Wilson Point New Levee) that was constructed through the region in the 1930s. Unfortunately, extensive levee borrow pits removed most of the site’s occupation area to the east of the levee, and the levee itself buried or destroyed two of the mounds. Prior to levee construction, the site consisted of at least ¤ve mounds grouped around a central plaza, a man-made ridge located at the south edge of the plaza, and buried aboriginal midden deposits covering an area of about 116,280 m2 (28 acres) (Figure 3.2). In the 1990s, the Vicksburg District, U.S. Army Corps of Engineers, decided that additional levee construction in the area of the mounds needed to take place to prevent water seepage under the levee. Accordingly, the site was tested in 1996 and later subjected to data-recovery excavations in 1998–1999 by personnel from Coastal Environments, Inc. (Weinstein 2005). Three test units, a series of six “well units,” and two block excavations were the main means by which the site was examined (see Figure 3.2). Two of the test units were located in Mounds D and E and produced excellent views of mound construction. The unit placed atop Mound D showed that the mound consisted of at least three different construction stages, with the lowest stage presenting some of the best-preserved basket loading ever seen by the investigators (Figure 3.3). In fact, it appears that the basket loads actually consist of

Figure 3.1. Location of the Lake Providence Mounds. Note the modern mainline Mississippi River levee running through the site (after Weinstein 2005:Figure 1-2).

Figure 3.2. Contour map of the Lake Providence Mounds, showing the four extant mounds and the Corps’ right-of-way (ROW ) during the 1998–1999 data-recovery program. Note that Mound B was partially buried by the 1930s levee, while Mound C either was completely buried or entirely removed by the same levee construction. The relict channel situated immediately to the south probably was the source of the alluvium that buried much of the site. TU, Test unit; WU, well unit (modi¤ed from Weinstein 2005:Figure 5-32).

recent data from the l a k e prov idence mounds

41

Figure 3.3. Pro¤les of north and east walls of TU N80W79 showing the three construction stages (Stratum 3, Strata 6 through 8, and Stratum 9, respectively) revealed during excavation. Note that the lowest construction stage (Stratum 9) consists of well-de¤ned sod blocks that are believed to represent the northwestern end of a manmade ridge upon which the mound was constructed (modi¤ed from Weinstein 2005:Figure 5-37).

“sod blocks” (Van Nest et al. 2001) that had been acquired from a grassy area near the site and then deposited upside down, causing the old A horizon to appear at the base of each block. The unit in Mound E uncovered the remains of a mound-top midden deposit that was ¤lled with hundreds of artifacts, including a unique assemblage of aboriginal ceramics that will be discussed in more detail shortly. The block excavations examined areas southeast of Mound D and north-

42

w ells a nd w einstein

west of Mound B. Block 1 uncovered evidence of a low ridge that contained numerous wall trenches and postholes indicative of at least three building episodes that included extensive walls and small, square structures of unknown function (Figure 3.4). The ridge, in fact, likely extended to the northwest and actually formed the initial construction stage recognized in the test unit in Mound D. Block 2 revealed a sequence of midden layers separated by thin ®ood deposits and a large trash pit that contained organically rich layers of redeposited trash, including ash, charcoal, ¤red clay, and a wealth of artifactual, faunal, and botanical refuse. A variety of faunal remains were recorded, including squirrel, deer, turtle, ¤sh, waterfowl, opossum, and raccoon. Surprisingly, tree squirrels were dominant in terms of numbers of identi¤ed specimens present and second only to white-tailed deer in total weight of bone (Scott 2005). Of the botanical remains, corn was particularly abundant (ca. 70-percent ubiquity) along with arboreal resources such as persimmon, pecan, and acorn (Roberts 2005). Considering the proximity of the trash pit to Mound B, it was envisioned that much of the ¤ll represented material discarded by the inhabitants who lived atop that mound.

Occupation Sequence at Lake Providence In addition to the general stratigraphic sequence uncovered in each area of the site during the testing and data-recovery excavations, 22 radiocarbon dates were obtained from charred material (almost all of which was wood charcoal), organic sediment from bulk soil samples, charred seeds, or baked-on cooking residue scraped from the outer surface of several of the site’s sherds. Likewise, 27 samples of soil were submitted for oxidizable carbon ratio (OCR) dating. When all of these dates were coupled with the various artifacts recovered across the site, it was possible to develop what is believed to be a fairly accurate sequence of site occupation and mound and ridge construction, at least for those areas within the Corps’ right-of-way that were available for archaeological research. This sequence is shown graphically in Figures 3.5 and 3.6. A few scattered sherds related to the Tchefuncte and Marksville cultures, plus one dart point base, mark the ¤rst evidence of occupation in the area of the Lake Providence site, although all were found in secondary contexts, such as mound or pit ¤ll. These were very brief and ephemeral occupations that played no role in the construction of the earthworks at the site. Any midden once associated with these occupations probably was situated along the edge of the relict Mississippi River channel located to the northeast of the site, an area destroyed by the 1930s levee (see Figure 3.6a). The ¤rst occupation for which intact deposits still exist can be dated to the

Figure 3.4. Plan of wall trenches and postholes associated with the Preston phase in Block 1, showing evidence of the three building episodes believed to have been present (after Weinstein 2005:Figure 7-105).

44

w ells a nd w einstein

Figure 3.5. Date ranges of the seven Coles Creek occupations recognized at Lake Providence. Note that Occupation I falls within the Balmoral phase, while all others fall within the middle to late Preston phase (after Weinstein 2005:Figure 8-8).

late Balmoral phase of the late Coles Creek period (ca. a.d. 1065–1080). Identi¤ed simply as Occupation I, evidence for this comes from the pre-mound midden found beneath Mound E and the similar pre-ridge midden found beneath the ridge in Block 1 (see Figure 3.6b). Despite the fact that the ceramic sample associated with this occupation is somewhat limited in size, it contains a sherd of Coles Creek Incised, var. Mott and three Vicksburg rims. Although such sherds were found in subsequent occupations dating to the succeeding Preston phase, their early stratigraphic position, the lack of later ceramics, and several complementary chronometric dates indicate that occupation during the Balmoral phase was a reality. It is uncertain whether mound construction began during the phase, but the initial stages of Mounds A and C are shown to re®ect such a possibility (see Figure 3.6b). Regardless of the minor Balmoral component, the main occupation at Lake

recent data from the l a k e prov idence mounds

45

Figure 3.6. Proposed site-formation sequence at Lake Providence (after Weinstein 2005:Figure 14-1).

Providence can be attributed to the Preston phase of the terminal Coles Creek period. As shown in Figure 3.6c–f, a series of additional site occupations, identi¤ed as Occupations II through VII, all date to the middle to late Preston phase, circa a.d. 1150–1260. It was during this relatively short, roughly 100year period that most, if not all, mound and ridge construction took place

46

w ells a nd w einstein

and the trash pit in Block 2 was utilized. Certainly the extensive walls and wall-trench structures found in Block 1 date from this time. Given the relatively short Preston phase occupation identi¤ed at the site, plus its occurrence right at the tail end of the Coles Creek period, one would be hard pressed to ¤nd a ceramic assemblage with a more “transitional” look to it. Decorative styles normally attributable to both the Coles Creek and Plaquemine cultures are found in these deposits, plus a few sherds suggestive of Mississippian culture (Figure 3.7). Ceramics typically considered markers for the preceding Balmoral phase, such as Coles Creek Incised, vars. Mott, Greenhouse, and Blakely and Mazique Incised, var. Kings Point, are present in the Preston deposits, as are ceramics generally associated with the Routh phase: Plaquemine Brushed, var. Plaquemine, Coles Creek Incised, var. Hardy, Mazique Incised, var. Manchac, and a number of other early Plaquemine types and varieties that occur on a paste equivalent to the Addis variety of Baytown Plain. That these sherds do not represent separate occupations but rather a single cohesive unit is evident from the repeated and consistent associations of these markers across the site, both in sealed midden deposits and in the undisturbed ¤ll of trash pits, wall trenches, and postholes. Despite the presence of the above ceramics typically considered indicative of the Balmoral and Routh phases, the great majority of the Lake Providence collection can be assigned, without any hesitation, to the Preston phase, the terminal expression of Coles Creek culture in the Tensas Basin. Most important are those ceramics with a paste equivalent to the Little Tiger variety of Baytown Plain (Figures 3.8 and 3.9). These sherds provide the most consistent and identi¤able markers for the Preston phase. Well-established Preston varieties, such as Coles Creek Incised, var. Hilly Grove and Mazique Incised, var. Preston (Belmont 1983; Fuller and Kelley 1993; Hally 1972), are now joined by several newly proposed or rede¤ned varieties that are based on Joanne Ryan’s (2004) work on the ceramics from the Hedgeland site, Cherie Schwab’s (1996) master’s thesis at Louisiana State University, and the present investigations at Lake Providence. These include Plaquemine Brushed, var. Blackwater, Carter Engraved, var. Crawford, Anna Incised, vars. Hedgeland, Little Red, and Tassin, Rhinehart Punctated, var. Chatlin, Avoyelles Punctated, var. Stack Island, Hollyknowe Pinched, var. Rose Hill, Evansville Punctated, var. Jack Falls, and possibly French Fork Incised, var. Iberville (Figure 3.10). With the exception of the last variety, all are within the stylistic limits of the decorated varieties normally associated with the Routh phase, but they occur on a medium to ¤ne grog–tempered paste that does not meet the criteria of Addis. Also present within the Preston assemblage at Lake Providence is a group of ceramics identi¤ed by Wells (2005) as the “Preston ¤neware complex.” This

Figure 3.7. Ceramics normally thought to be representative of the Balmoral and Routh phases but clearly part of the Preston phase component at Lake Providence. (a-c) Coles Creek Incised, var. Mott; (d-e) Coles Creek Incised, var. Greenhouse; (f-g) Mazique Incised, var. Kings Point; (h-i) Plaquemine Brushed, var. Plaquemine; (j) Coles Creek Incised, var. Hardy; (k-l) Mazique Incised, var. Manchac; (m-n) Harrison Bayou Incised, var. Harrison Bayou; (o) Barton Incised, var. unspeci¤ed.

Figure 3.8. Preston phase ceramics from Lake Providence. (a-c) Coles Creek Incised, var. Hilly Grove; (d-e) Mazique Incised, var. Preston; (f-h) Plaquemine Brushed, var. Blackwater; (i) Anna Incised, var. Tassin; (j-k) French Fork Incised, var. Iberville; (l-m) Rhinehart Punctated, var. Chatlin.

recent data from the l a k e prov idence mounds

49

Figure 3.9. Additional Preston phase ceramics from Lake Providence. (a-c) Avoyelles Punctated, var. Stack Island; (d-f ) Hollyknowe Pinched, var. Rose Hill; (g-i) Evansville Punctated, var. Jack Falls.

group comprises more than 40 percent of the ceramics from the midden uncovered in the test unit atop Mound E. It contains several varieties of Carter Engraved, sherds of Baytown Plain, var. Vicksburg, a few decorated varieties that occur on Vicksburg paste, and numerous ¤ne, shell-tempered wares (Figure 3.11).

Figure 3.10. Ceramic types and varieties of the Balmoral, Preston, and Routh phases (after Wells and Weinstein 2005:Figure 13-2).

Figure 3.11. Examples of pottery associated with the “Preston ¤neware complex.” (a-g) Carter Engraved, var. Shell Bluff; (h) partial ¤sh-ef¤gy bowl of Carter Engraved, var. Carter; (i) Carter Engraved, var. Mud Lake.

52

w ells a nd w einstein

It was realized fairly quickly that many of these latter shell-tempered sherds did not ¤t comfortably into assemblages from the region, and a search of the literature and consultation with colleagues in neighboring regions led to the conclusion that a signi¤cant portion of the pottery might have been imported. Many of these exotic-looking sherds closely resembled pottery from the American Bottom region, in particular the Mississippian sites concentrated around the great mound center of Cahokia. Included were most of the classic elements of the so-called Powell set, a group of ceramic types and varieties established by Williams and Brain (1983:321) to re®ect the remains of vessels considered to be direct imports from the American Bottom—Powell Plain, var. Powell (n = 278), Ramey Incised, var. Ramey (n = 2), and Old Town Red, var. Cahokia (n = 70) (Figure 3.12)—together numbering 350 sherds. Of some surprise was that this assemblage represents the largest collection of American Bottom ceramics from any site in the Lower Mississippi Valley (LMV). Also included were 61 sherds of the related Coker set (Williams and Brain 1983:321–322), a group of locally made ceramics that are thought to represent copies of the exotic vessels associated with the Powell set—Mississippi Plain, var. Coker (n = 34) and Old Town Red, var. Sharbrough (n = 27) (see Figures 3.10 and 3.12)—which brought the total number of American Bottom sherds or copies of American Bottom sherds at Lake Providence to 411. To these can be added a partial vessel typed as Coleman Incised (Figure 3.13a) that was made on local paste but whose decoration is exceedingly similar to that found on Ramey Incised, plus a single Mississippi Triangular, var. Cahokia arrow point fashioned from Illinois kaolin chert (see Figure 3.13b). Although found in small amounts in most contexts at the site, these exotics or exotic-inspired ceramics came primarily from the midden atop Mound E, the ¤ll of the Block 2 trash pit just northwest of Mound B, and the midden and structures atop the ridge in Block 1 southeast of Mound D. Clearly, these locations suggest elite, high-status deposits.

The “Cahokia Horizon” in the LMV In the concluding chapter of their report on the Lake George site, Stephen Williams and Jeffrey Brain (1983:409–412) devoted several pages of text toward an attempt to identify the nature and extent of the so-called Cahokia horizon in the LMV, particularly at sites within the Yazoo Basin. They recognized this horizon as a distinct event that took place during terminal Coles Creek times and was marked by unmistakable evidence of direct contact with the great site of Cahokia and other locales in the American Bottom region of southern Illinois. Williams and Brain theorized that such contact occurred

Figure 3.12. Sherds of the Powell and Coker sets recovered at Lake Providence. (a-e) Powell Plain, var. Powell; (f ) Ramey Incised, var. unspeci¤ed; (g-k) Old Town Red, var. Cahokia; (l-n) Old Town Red, var. Sharbrough.

54

w ells a nd w einstein

Figure 3.13. Additional artifacts suggestive of contact with the American Bottom region. (a) Partial vessel of Coleman Incised, var. unspeci¤ed that is very similar in form and decoration to the type Ramey Incised; (b) Mississippi Triangular, var. Cahokia arrow point.

between circa a.d. 1150 and 1250 and, within the Yazoo Basin, was most noticeable at sites with components of the Crippen Point phase, especially the large mound centers of Winterville and Lake George (Figure 3.14). The main evidence indicative of contact with Cahokia came in the form of exotic ceramics, although at least one foreign arrow point also was noted. As discussed above, Williams and Brain (1983:321) classi¤ed the exotic pottery as elements within their Powell set, including Ramey, Cahokia, Powell and Tippets Incised, var. Tippets. They also recognized two closely related ceramic groups, labeled the Coker set and the Yazoo 1 subset (Williams and Brain 1983:321–322). As noted previously, these latter sets included locally made ceramics that were viewed as copies of the exotic vessels associated with the Powell set. Included were such varieties as Cahokia Cord Marked, var. Montrose and Old Town Red, var. Sharbrough for the Coker set and Cahokia Cord Marked, var. Buford and Old Town Red, var. Old Town for the Yazoo 1 subset. One arrow point, given the name Mississippi Triangular, var. Titterington, also was seen as excellent evidence of Cahokia contact (Williams and Brain 1983:236). Seven sites in the Yazoo Basin were identi¤ed as having produced examples of the Powell, Coker, and Yazoo 1 sets, thereby indicating direct ties to the American Bottom region (Williams and Brain 1983:Figure 12.15), while several

recent data from the l a k e prov idence mounds

55

Figure 3.14. Depiction of Early Mississippian in®uences on the Yazoo Basin during the Crippen Point phase, from Williams and Brain’s (1983:Figure 12.14) Lake George report. (Courtesy Peabody Museum Press. Copyright 1983 by the President and Fellows of Harvard College.)

other sites were known to have yielded only elements of the Yazoo 1 subset, thus implying some type of horizon in®uence, though not necessarily direct contact. In concluding their discussion, Williams and Brain (1983:411–412) noted: “The Yazoo remains the best laboratory for the consideration of the events outlined above. Artifactual diagnostics of Cahokia have not yet been reported from other Lower Mississippi regions, even where there has been intensive investigation.” Furthermore, they went on to say, “it is illogical to think that the Tensas Basin would have been completely ignored at the same

56

w ells a nd w einstein

time that the Yazoo was receiving considerable attention across the river. Nevertheless, no artifactual evidence relating to Cahokia was reported for the former during the course of two ¤eld seasons of survey and test excavation conducted by the LMS in 1963 and 1964” (Williams and Brain 1983:412). If con¤rmed, the presence of Cahokian imports at the Lake Providence Mounds would indicate that the in®uence of the American Bottom polities was also felt in the Tensas Basin of Louisiana.

Sourcing Analyses To con¤rm the likelihood that the apparent American Bottom sherds truly came from that area of the country, a small sample of both the local claytempered ware and the potentially exotic ceramics was submitted for sourcing analyses (Table 3.1). A petrographic analysis of 37 thin-sectioned sherds and two pieces of ¤red clay from Lake Providence was conducted by James B. Stoltman of the University of Wisconsin, and portions of these same samples then were sent to the University of Missouri’s Research Reactor Facility (MURR) where Enrique Rodríguez-Alegría and his colleagues conducted instrumental neutron activation analysis (INA A). Using point-counting techniques, Stoltman (2005) found that all of the clay-tempered sherds produced statistically identical indices, and he came to the conclusion, as suspected, that these probably represented locally produced pottery. Also as suspected, Stoltman found that the Powell Plain pottery identi¤ed at Lake Providence matched very well with examples from the American Bottom. Surprisingly, Stoltman further suggested that all of the other shelltempered sherds in the sample, including all specimens of Mississippi Plain, had a nonlocal but probable common origin. Thus, they also could have come from a far-off region, perhaps from Mississippian sites located in Arkansas, northern Mississippi, or Missouri, having traveled to Lake Providence by the same means that brought the American Bottom vessels to the site. The data collected by neutron activation allowed Rodríguez-Alegría et al. (2005) to group the sherds into three clusters on the basis of their chemical signatures. Groups 1 and 2 overlapped to a signi¤cant degree and were considered to represent locally made pottery. Signatures that fell outside these groups were usually classi¤ed as “unassigned.” A third group of Lake Providence sherds was assigned to MURR’s provisional Cahokia-Halliday group, a tentative formulation based on ongoing research on American Bottom collections. Two shell-tempered sherds were assigned to this chemical group. Petrographic and neutron activation analyses agreed that 19 of the 37 sherds submitted for analysis were locally made (see Table 3.1). They further agreed

Table 3.1. Proveniences of samples and results of sourcing analysis on selected Lake Providence sherds.

58

w ells a nd w einstein

that seven of the 37 sherds were nonlocal and that two of these imports were manufactured in the American Bottom area. Taken together, the two sourcing studies, particularly the thin-section analysis, suggest that the potential American Bottom sherds found at Lake Providence likely originated in that region, probably during the Stirling 2 (ca. a.d. 1150–1200) or Moorehead 1 (ca. a.d. 1200–1250) phase (Figure 3.15). However, the predominance of brown ¤lming on the sherds of Powell Plain from Lake Providence, as well as the absence of Cahokia Cord Marked, suggests that a Stirling 2 date may be more likely (Milner 1984:166). Recognizable rims at Lake Providence also include forms common to the Stirling 2 phase (see Pauketat 1998b:Figures 7.27, 7.28, 7.31, and 7.32). Thus, contact with the American Bottom region (presumably including the great mound center at Cahokia) probably occurred at Lake Providence during the earlier half of the Preston phase component at the site or roughly sometime between a.d. 1150 and 1200. This is in keeping with several other locales in the LMV, where sherds attributable to the so-called Cahokia horizon have been found (Brain 1969, 1989; McNutt 1996:237–238; Williams and Brain 1983:409–412).

Lake Providence and the Evolution of the Mississippi Period Elite The “conservative and introverted character” (Williams and Brain 1983:410) of Coles Creek society has often been noted (Kidder 1992a; Wells 1998). While some Weeden Island–related decorative styles and a number of Plum Bayou rim forms are found in Coles Creek sites in the Tensas Basin, Coles Creek ceramics prior to a.d. 1000 exhibit a very limited number of extraregional contacts and in®uences (Fuller et al. 1995:129–136; Hally 1972; Kidder 1990b, 1993a; Wells 1998; see also Williams and Brain 1983). This apparent parochialism has been interpreted as a sign that the incipient elite of Coles Creek societies preferred not to (or was not allowed to) participate in extraregional trade networks. This may re®ect the organization of early Coles Creek society; the low degree of social ranking probably prevented the incipient elite from exercising the prerogatives that later elites were to enjoy (Kidder 1992a:156; Nassaney 1991:195–196; Wells 1998). As the Coles Creek period approached its last two or three centuries (Balmoral and Preston phases), however, it seems evident that some changes were beginning to take place. Substantial programs of mound construction continued, often following the standard Coles Creek three-mound site plan. However, several sites with ¤ve or more mounds also appeared. This suggests a two- or three-tiered program of settlement in some areas, consisting of small hamlets and possibly two levels of mound settlement (Wells 1997). Limited

Figure 3.15. Chronology chart illustrating the temporal relationship of archaeological phases in the Tensas Basin and American Bottom regions since about a.d. 750. The Preston phase is highlighted, as it represents the principal occupation at Lake Providence (Tensas sequence after Wells and Weinstein 2005:Figure 13-1. American Bottom sequence after Bareis and Porter 1984:Figure 3; Pauketat 1998b:Figure 5.1).

60

w ells a nd w einstein

numbers of trade goods appear in Balmoral phase contexts (Kidder 1990b), and the ¤rst corn is found in the Tensas Basin during this phase. A distinctive ¤neware (Vicksburg paste) makes its ¤rst signi¤cant appearance, albeit in minor amounts (2 to 4 percent in Tensas Basin sites; Hally 1972:Table 28). Ceramically, this paste may be one of the strongest indicators of special-function, proto-elite activity (Ryan 2004:248). The bulk of pottery, however, is still made on a Percy Creek/Valley Park-like paste. Exotic designs are rare, and imported ceramics are virtually unknown from the phase. The terminal Coles Creek period dramatically ampli¤es this trend of social change. Larger villages appear that may suggest an extra tier in the settlement hierarchy below the level of mound site (Kidder 1993a; Wells 1997). Corn increased its presence in Preston phase contexts, although it had yet to reach the levels found in Plaquemine phases (Kidder 1993a:26). If the Lake Providence site is any indication of the constitution of elite activities in the Tensas Basin, however, one of the most dramatic changes in social differentiation and material culture may be seen in the ceramic containers used in high-status contexts. The presence of a “Preston ¤neware complex” has already been noted in the midden atop Mound E and in most other locations at the Lake Providence Mounds. This complex consists of Carter Engraved, Vicksburg, and related decorated varieties and the ¤ne shell-tempered wares Powell Plain, Ramey Incised, Mississippi Plain, var. Coker, and their red-¤lmed counterparts. Sherds identi¤ed as Powell Plain and Ramey Incised appear to be derived from the American Bottom. Their presence adds even more signi¤cance to this exotic-looking assemblage. Ryan (2004:248) notes an increase in ¤newares at Hedgeland and other sites during the Balmoral phase. While observing that ¤newares (in this case, represented by sherds with Baytown Plain, var. Little River paste) were used for a limited number of vessels in early Coles Creek contexts, she points out that the percentages of ¤ne clay–tempered pottery double during the Balmoral phase. She attributes this to the rise of an elite by the end of the Coles Creek period, a segment of society that required an expression of their status and social/ceremonial responsibilities through the use of well-made pottery. The numbers of imported goods evident in late Coles Creek contexts at Hedgeland indicate an increase in external contacts as well, although the Cahokian imports appear to be missing from this site. Instead, Ryan takes the presence of bone-tempered Morris Plain and possible limestone-tempered sherds (Unclassi¤ed Plain at Hedgeland) to indicate Caddoan and/or eastern contacts (Ryan 2004:248–249). It is tempting to view the sudden increase in external contacts and trade

recent data from the l a k e prov idence mounds

61

goods at terminal Coles Creek mound sites in the LMV as causal in the evolution of the Plaquemine-Mississippian elite. Certainly, Brain (1989) and Williams and Brain (1983:409–412) have cited Cahokian contact as one of the primary forces behind the “Mississippianization” of Coles Creek societies. However, the evolution of the incipient Coles Creek elite probably has its roots in the Baytown and early Coles Creek periods with the initial construction of platform mounds. These mounds were not intended for communal burials as they were in Baytown period (Troyville culture) sites. Instead, they represent an elaboration of mortuary ceremonialism (Kidder and Wells 1992). This change marked the gradual exclusion of the overall community from some aspects of ceremonial life and the beginning of an increasingly important role of a limited number of people in ritual activities. Steponaitis (1986:386), Kidder (1992a), Knight (1986), and Wells (1998) have postulated that it was primarily this co-option of ceremonial roles, not economic activities, that led to the rise of the Mississippi period elite. Certainly, the Coles Creek period saw a gradual but steady increase in the amount of mound construction at many Tensas Basin sites, a sign that central direction or control of labor was increasing from the Mount Nebo to the Preston phases (Wells 1997, 1998). This is an indication that the evolution of centralized power and the rise of the elite in Coles Creek society were taking place well before trade and external contact became factors. Therefore, these external contacts and the apparent monopolization of trade goods should be seen as signs that an existing elite segment of Coles Creek society was beginning to exercise its power more widely. These ongoing, indigenous processes probably have far more to do with the rise of the subsequent Plaquemine elite than the presence of a few trade goods in sites dating to the end of the Coles Creek sequence. As for the “Mississippianization” of the Tensas Basin, it does not seem likely that the succeeding Plaquemine phases of the area really derived much of their Mississippian leanings from Cahokian contact. The established Mississippi period phases within the Tensas Basin appear to be direct descendants of Coles Creek ceramic traditions (Kidder 1993a). Any Mississippian in®uences appear to be derived from sustained contact with middle Mississippi period (Winterville and Lake George phase) peoples in the Yazoo Basin and areas to the north. The lone exception is the Transylvania phase, believed to date largely to the late end of the Mississippi period sequence (Hally 1972). Transylvania does contain a large, truly Mississippian shell-tempered assemblage and may represent the intrusion of groups from outside the Tensas Basin. Surface collections from the Rose Hill (Panola Mounds) site just to the northwest of Lake Providence appear to be derived from true Routh phase

62

w ells a nd w einstein

deposits and contain very few shell-tempered sherds (Hally 1972:Table 25). Thus, it appears that the local sequence was largely unaffected by contact with the powerful American Bottom chiefdom (or chiefdoms). Although Cahokian contact may not have left a permanent impression on the local cultural sequence, there is clearly a Mississippian presence in the Preston phase assemblage at Lake Providence. Just over three percent (n = 625) of the sherds from this site are shell tempered and the wares they came from may have been produced by foreign potters. There are two primary explanations of how these sherds came to be part of the assemblage at Lake Providence. The ¤rst relies solely on trade, particularly from the north and especially with the American Bottom. Certainly, one of the strategies of an emerging elite is the manipulation of both exotic goods and the new ideologies they represent. There is often prestige attached to the exotic and mystical elements of other cultures, which can be translated to prestige for the manipulator of said elements (Bishop 1987; Earle 1991; Helms 1993; Wells 1998). On a less ethereal level, control over the exchange of essential raw materials, such as stone, and superior ¤nished goods may also play a role (Nassaney 1992; Redmond 1996). A second theory to explain the presence of exotic and exotic-looking ceramics involves a combination of trade for ¤nished goods and the physical presence of Mississippian potters. A certain number of these shell-tempered pots were probably imported into the site, particularly the examples of Ramey Incised, Powell Plain, and Old Town Red, var. Cahokia. However, many of the sherds of shell-tempered pottery may not be imported, as suggested by INA A. These sherds are very clearly not within the tradition of Coles Creek pottery. There are, however, sherds of var. Greenville and associated varieties (such as var. Carter) that exhibit both shell and grog tempering. This combination suggests that local potters were being in®uenced in their manufacturing techniques by Mississippian methods of pot construction, adding small amounts of shell to their grog-tempered repertoire. A likely explanation for this is the actual presence of a few Mississippian potters who were producing vessels in their own tradition. This would probably have involved the exchange of women (the most likely candidates for prehistoric Native American potters) or perhaps the presence of “war brides” or slaves taken in raids. Certainly, the appearance of wall-trench structures at Lake Providence argues for a more extensive Mississippian presence than could result from indirect, down-the-line trade contacts. Although Lake Providence has one of the largest samples of shell-tempered pottery from this time level in the LMV, it is not likely that the current data can be used to convincingly support either of the above hypotheses. The small

recent data from the l a k e prov idence mounds

63

percentage of these shell-tempered sherds suggests neither a “site-unit intrusion” of Mississippian peoples nor heavy contact between the cultures. Simple exchange might be an explanation, although the presence of shell-and-grog– tempered pottery in the terminal Coles Creek levels at Lake Providence and at sites in the Lower Yazoo Basin does imply more sustained communication between potters. Ceramic data alone are probably not adequate to evaluate these hypotheses; other lines of evidence, particularly mortuary data, are needed to help establish the presence or absence of Mississippian peoples, as well as the nature of the contact. Unfortunately, these analyses will have to await future projects, as there are no data currently at hand to answer these questions.

Conclusions Testing and data-recovery operations at the Lake Providence Mounds have yielded a wealth of information relating to the beginnings of Plaquemine culture in northeast Louisiana. In addition to feature and structure patterns, the site has produced a large number of artifacts that shed light on social development at the terminal end of the Coles Creek sequence. To date, it appears that the imported ceramics at Lake Providence are among the only examples of Cahokian pottery to be found in the southern LMV outside of the Yazoo Basin. Certainly, recent programs of similar intensity undertaken at mound centers such as Osceola (Kidder 1990b), Raffman (see Roe, this volume), and Hedgeland (Ryan 2004) have not encountered these imported shell-tempered types in Preston phase contexts. Despite this, Lake Providence shares a closely related assemblage of artifacts with these other sites, an assemblage that clearly belongs to the Preston phase. All four sites share a multiple-moundand-plaza plan and a robust program of mound construction. In all probability, it is Lake Providence’s unique position at the northern end of the Tensas Basin, close to the Mississippi River and mound centers in the nearby Yazoo Basin, that constitutes the primary difference. This proximity probably afforded the emergent Lake Providence elite access to the Cahokian goods. While this access may have given these elites a few quality exotics to bolster their positions within the local sphere of in®uence, shell-tempered types retained a minor position in the Tensas Basin until the late Mississippi period. It is dif¤cult to argue that contact with full-blown Mississippian societies to the north during terminal Coles Creek times had a substantial impact on succeeding generations. While Ian Brown, in this volume, has questioned the ancestral relationship

64

w ells a nd w einstein

of Coles Creek and Plaquemine culture bearers in the Natchez Bluffs region, there seems to be little doubt about this relationship in other areas. This is particularly true for the Tensas Basin, as expressed at Lake Providence. Traditional Coles Creek ceramic varieties appear alongside Plaquemine ceramics and have consistent associations with imported shell-tempered varieties. All of these, of course, are minority wares included in a larger Preston phase assemblage, which often features Plaquemine decorative techniques on a nonAddis ware. These associations suggest a strong continuity between the native Coles Creek potters and their Plaquemine successors. One thing is certain, however—the late Coles Creek residents of the upper Tensas Basin were not as parochial as many once believed. They clearly had contact with other groups farther to the north in the heartland of the Mississippian world. The terminal Coles Creek occupation at the Lake Providence site provides a clear break with the rather provincial tendencies of earlier Coles Creek phases, in which evidence of extraregional contact and interaction generally is lacking. The Preston ¤neware complex is predominantly associated with mound-top activities associated with the upper echelon of Preston society. The fact that much of the ¤neware at Lake Providence can be tied to sites in the American Bottom region adds even more signi¤cance to this complex. How and why these exotics came to Lake Providence, whether directly down the Mississippi River or through intermediate centers such as Winterville and Lake George, is still unknown.

Acknowledgments Several individuals aided in the preparation of this essay. James B. Stoltman, Professor Emeritus at the University of Wisconsin, conducted the petrographic thin-section analysis and produced the related data utilized herein. Enrique Rodríguez-Alegría, Robert J. Speakman, and Michael Glascock of the Research Reactor Facility, University of Missouri, conducted INA A and provided the relevant information. Curtis Latiolais and Cherie Schwab of Coastal Environments, Inc., produced and/or aided in the production of the maps and ¤gures used throughout the chapter. The Peabody Museum of Harvard University graciously permitted use of Williams and Brain’s (1983) Figure 12.14, reproduced here in Figure 3.14. In addition to Stoltman, several archaeologists examined examples of the unique and potentially exotic ceramic and lithic artifacts recovered from Lake Providence and provided insight into their possible points of origin. Included were T. R. Kidder, Washington University, St. Louis; Marvin Jeter, Station Archeologist with the Arkansas Archeological Survey at the University of Ar-

recent data from the l a k e prov idence mounds

65

kansas at Monticello; Timothy Pauketat, University of Illinois at UrbanaChampaign; John E. Kelly, Washington University, St. Louis; Frank F. Schambach, Station Archeologist with the Arkansas Archeological Survey at Southern Arkansas University; and Stephen Williams, Professor Emeritus, Peabody Museum of Archaeology and Ethnology, Harvard University. Others generously provided examples of pottery from their respective regions for the sourcing analysis. Included were Martha Rolingson, former Station Archeologist with the Arkansas Archeological Survey at Toltec Mounds; Mary Beth Trubitt, Station Archeologist with the Arkansas Archeological Survey at Henderson State University; and Timothy Perttula of Archaeological and Environmental Consultants, LLC, Austin, Texas. Lastly, Tad Britt, former archaeologist with the Environmental Resources Branch of the Vicksburg District, U.S. Army Corps of Engineers, and now with the Corps’ Construction Engineering Research Laboratory (CERL) in Champaign, Illinois, was instrumental in overseeing most of the testing and data-recovery investigations for the Corps. James Wojtala, present archaeologist with the Environmental Resources Branch of the Vicksburg District, provided the ¤nal contract needed to complete most of the analyses on the recovered data. Without their efforts and the concern of the Vicksburg District, none of the work at Lake Providence would have been possible.

4 Plaquemine Mounds of the Western Atchafalaya Basin Mark A. Rees

When the Louisiana State Archaeological Survey began digging into the Medora mounds on the west bank of the Mississippi River in the winter of 1939, Plaquemine was just a small river town around the next bend, thought to have been named for the wild persimmons that grew along the riverbanks (Riffel 1985:32). The cultural tradition of the people who built the mounds at Manchac Point was yet to be de¤ned, only vaguely conceived on the basis of surface collections as “either . . . late Coles Creek or . . . that of an unformulated period that followed Coles Creek but . . . older than Natchezan” (Quimby 1951:91). While James Ford (e.g., 1951:13) intended Plaquemine to serve alongside Coles Creek as a typological construct in his chronology of Lower Mississippi Valley (LMV) ceramics, it was not long before its identi¤cation as a period, a culture, and even a phase was recognized as problematic, particularly in relation to more expansive Coles Creek and Mississippian traditions (e.g., Cotter 1952a:124–125; Willey 1966:308). Phillips (1970:950) effectively isolated (or “reduced”) Plaquemine culture as represented by the Medora phase in his most in®uential work, while subsequent research upriver rede¤ned the problem of Plaquemine origins in terms of “hybridization”—an admixture of indigenous Coles Creek and subsequent exogenous Mississippian in®uence (Brain 1989:122; Rees and Livingood, this volume; Williams and Brain 1983:414). The Lower Yazoo Basin and Natchez Bluffs would thus be seen as focal points of Coles Creek–Plaquemine “transculturation” or at least the “geographical center of Plaquemine culture” (Brain 1978:332, 1989:122; Brown 1985b:251, this volume). The southern delta and vast ®ood basin to the west have consequently been regarded as “peripheral, or marginal” to mainstream Plaquemine development (Gibson 1975:28), a point of view ostensibly supported by relatively less-spectacular earthen

mounds of the w estern atch afa l aya basin

67

mound sites and a paucity of ceramic evidence for formative Mississippian in®uence (Duhe 1981; McIntire 1958; Weinstein 1987a). Of course, these different perspectives have in®uenced and in turn re®ect disparate theories of Plaquemine origins and development (Rees and Livingood, this volume). Setting aside questions of cultural origins and in®uence, it is also possible to approach Plaquemine peoples from a combined political and historical perspective, in terms of regional political culture. The most obvious comparisons to be made here are with contemporaneous Mississippian polities to the north and east, although apparent similarities with earlier Coles Creek polities should not be ignored. Acknowledging the pitfalls of typological constructs such as the chiefdom (e.g., Cobb 2003:65; Earle 1987:279–281), it is reasonable to conceive of Plaquemine culture in terms of regional polities of various geographic scales in the LMV, each interacting on different levels at different points in time between circa a.d. 1200 and 1700. With this in mind, we turn to the ®ood basin on the southwest edge of the LMV to examine the evidence for the political culture of Plaquemine peoples, as represented in numerous mound sites. It will then be possible to revisit the problem of cultural origins and consider whether it is accurate or even useful to conceive of polities in the western Atchafalaya Basin as peripheral to Plaquemine culture. First, however, a little background is in order.

Archaeology in the Atchafalaya Basin The Atchafalaya Basin in south-central Louisiana is a complex network of bayous, bottomlands, natural levees, swamplands, ®ood basins, and backwater lakes on the southwest edge of the LMV. From Old River north of Simmesport to Atchafalaya Bay west of the Terrebonne Basin, the Atchafalaya River is today a major distributary of the Mississippi and Red rivers. Covering approximately 3,800 square miles, the surrounding Atchafalaya Basin is the “largest major division of the deltaic plain” (Saucier 1994:30). While it has long been renowned for its proli¤c wildlife, archaeologists have only just begun to explore the history of human occupation in the region. The basin is made up of deltaic plain alluvium deposited during the mid to late Holocene (7000–3000 b.p.) and the river became a distributary of the Mississippi as recently as ¤ve centuries ago (Autin et al. 1991:561–564; Saucier 1994:30–31, 284–286; Smith et al. 1986:45, 49–56). The Atchafalaya Basin has been increasingly in®uenced by human modi¤cations, resulting in large-scale environmental transformations during the past century. Central to these efforts have been the system of levees and the Old River control station that restrict the volume of water entering the Atchafalaya from the Mississippi. Knowl-

68

m ark a. rees

edge of site geomorphology in such a dynamic and rapidly changing landscape is essential in understanding human habitations during the past millennium, particularly the Mississippi period (Autin et al. 1991:561–564; Gagliano 1984; Reuss 2004; Saucier 1981:16–17). The present study encompasses that portion of the basin west of the main channel of the Atchafalaya River and east of the Prairie Terrace and Chenier Plain of southwest Louisiana. It includes the Teche Ridge, a former Mississippi River meander belt (Saucier 1994:278–280). This is the traditional homeland of the Chitimacha, described by Swanton (1911:337–360) as living in villages along Bayou Lafourche, Bayou Teche, and Grand Lake by the early eighteenth century. Little has been published regarding the Chitimacha prior to the eighteenth century, yet they are generally recognized as among the descendants of Plaquemine populations ( Jeter 1989:242; Kniffen et al. 1987:53, 230; cf. Neuman 1984:278; Weinstein 1987a:101). Plaquemine components are far more numerous than Mississippian in the study area and the mound sites discussed here are culturally af¤liated with the present-day Chitimacha Tribe of Louisiana (Gibson 1990:105–107). The cultural landscape has remained somewhat obscure archaeologically, perhaps owing in part to characterizations of the natural environment as a backwater (Gibson 1975:28, 1976:20, 1978a:44). A brief overview of previous investigations serves to demonstrate an even more compelling reason that the region as a whole has been portrayed as culturally marginal. During the late winter and early spring of 1912–1913 the now legendary steamboat the Gopher plied the Atchafalaya River and Bayou Teche, carrying avocational archaeologist C. B. Moore to little-known Indian mound sites hidden along the muddy banks. Moore (1913:9–21) mentions seven mound sites in the study area, although he investigated only four in the most cursory fashion (16SL2, 16SMY2, 16SMY10, and possibly 16SM2). Describing his investigations as an “unproductive task” and ¤nding few of the spectacular artifacts he encountered at other mound sites, Moore’s attention was soon drawn elsewhere (Moore 1913:7; Weinstein et al. 2003:126–144, Appendix A-3). In the decades following Moore’s expedition some of the same mound sites were brie®y visited by Collins (1927), Kniffen (1938), and McIntire (1958). Groundbreaking research to the north and east meanwhile contributed to the formation of a culture historical framework and nomenclature that would in®uence subsequent investigations throughout the southeastern United States (e.g., Belmont 1967; Ford 1935a, 1936, 1951, 1952; Ford and Webb 1956; Quimby 1951, 1957). Of particular importance were excavations by the Louisiana State Archaeological Survey at Medora on a natural levee of the Mississippi River

mounds of the w estern atch afa l aya basin

69

immediately east of the Atchafalaya Basin (Quimby 1951; see also Quimby 1957). Although little work was attempted in the western Atchafalaya Basin during this time, surface collections from some sites were suf¤cient to ¤t within a broad culture chronology (e.g., Phillips 1970:920–923, 950–954). A new era of archaeology began in the basin in the 1970s, accompanied by the arrival of university archaeologists and the growth of cultural resource management (CRM). Jon Gibson (1976, 1978a, 1979, 1982, 1990), with the University of Southwestern Louisiana, now the University of Louisiana at Lafayette, surveyed portions of the basin and Teche Ridge (see Neuman 2002:90–91). Under the direction of Ian Brown (1978, 1979, 1980, 1981, 1982b, 1984), the Lower Mississippi Survey extended the scope of its investigations to Avery Island and the Petite Anse region of coastal Louisiana (Brown and Lambert-Brown 1978; Brown et al. 1979). Among the sites investigated were late Coles Creek, early Plaquemine, and Mississippian components at Salt Mine Valley (16IB23) on Avery Island (Brown and Lambert-Brown 1979) and at Morgan (16VM9), a Coles Creek mound site in coastal Vermilion Parish (Brown 1981; Fuller and Fuller 1987). The Atchafalaya Basin Archaeological Survey at Louisiana State University (LSU) picked up where C. B. Moore left off, documenting mound sites that had otherwise escaped the attention of archaeologists (Neuman and Servello 1976). Described by Neuman and Servello (1976:71) as the “¤rst systematic investigation of archaeological sites in the region,” their survey consisted of background research, brief site visits, and surface collections (see also Gagliano 1967, 1984; Gagliano et al. 1975; Neuman 1972, 1977). Also during the 1970s, James Springer (1973, 1977, 1980) produced detailed studies of Coles Creek ceramics and subsistence from excavations at the Bruly St. Martin (16IV6) mound site in the eastern Atchafalaya Basin south of Baton Rouge. Since the 1980s archaeologists employed in CRM have examined numerous mound sites in the basin (e.g., Gibson 1982; Goodwin et al. 1985; Goodwin et al. 1991b; Kelley et al. 2000; Manning et al. 1987; Santeford et al. 1995; Shuman 1985; Vigander and Maygarden 1994; Wells 2001). Many of these surveys dealt only indirectly with mound sites. Jones and Shuman (1987, 1991) mapped and performed surface collections at mound sites in a portion of the study area, producing some of the ¤rst maps of sites that had been known about for a century or more. Creation of a regional program within the Louisiana Division of Archaeology (LDA) provided stimulus for additional studies, including Mike Russo’s (1992, 1993) work at the Baker (16SM19) and Stelly (16SL1) sites (Russo and Fogleman 1996). Regional archaeologists continue to visit and record sites (e.g., McGimsey 1995, 1998, 2001; Mann 2001, 2002). By the late

70

m ark a. rees

1990s Plaquemine components were known for many of the mound sites in the western Atchafalaya Basin. Yet there had been no concerted program of research aimed speci¤cally at examining the timing of mound construction and site occupations, potential contemporaneousness of various sites, or regional political economy. In short, prior to the 1970s archaeological research in the Atchafalaya Basin as a whole was characterized more by general overviews and decades-long periods of inactivity than by any particular systematic or intensive program of study (Gibson 1978a:27, 1979:29; Kelley et al. 2000:9–10; Wells 2001:7–8). Since then the pace of research has escalated within the context of CRM and university-based investigations, yet it has involved few detailed studies of Plaquemine mound sites. Current understanding of Plaquemine societies of the Atchafalaya Basin as marginal to a Plaquemine “heartland” is consequently as much a re®ection of the data sets available to archaeologists as it is a measure of supposed cultural associations. The dearth of information ultimately raises questions concerning Plaquemine origins and the notion of a heartland or cultural center. The Medora site, after all, where Plaquemine was conceived, is located only 40 km (25 miles) east of the study area (Quimby 1951). Taking into account the ongoing risks of site destruction, the Plaquemine Mounds Archaeological Project (PMAP) was designed as a means of redressing the need for additional study of mound sites in the western Atchafalaya Basin. Earthen mounds are today still the most visible features of Native American sites and their study presents an opportunity to examine the timing and scale of regional political development and decline. During the ¤rst year of PMAP investigations (2003–2004) background research and reconnaissance surveys were conducted in order to identify mound sites with possible Plaquemine components. A total of 80 mound sites were identi¤ed through a search of the available literature and the LDA site ¤le database, encompassing portions of St. Landry (31 sites), Lafayette (6 sites), St. Martin (17 sites), Iberia (12 sites), and St. Mary (14 sites) parishes. These represent known or previously recorded mound sites; sites as yet unrecorded may potentially remain, and some sites may have been destroyed without ever having been recorded. Due to the vagaries of early site recording it is sometimes unclear whether an earthen mound, shell midden, or both mound and midden are present at a site. Terms such as shell mound and earth midden do not adequately distinguish intentionally constructed mounds from transposed, primary contexts resulting from accretional midden formation. Furthermore, insuf¤cient data are presently available from many sites to make such distinctions. Several of the above mound sites were recorded in only cursory fashion and

mounds of the w estern atch afa l aya basin

71

have subsequently been ®attened or destroyed. The number of obliterated mounds has grown steadily since the nineteenth century, including at least one presumably lost to coastal erosion and others having been leveled in agricultural ¤elds. The Pharr or Fairview Plantation site (16SMY148), near the town of Berwick, consisted of two mounds when visited by Collins in 1926 (McGimsey 2001:124–125). When revisited in July 2003 residential development had recently encroached where the mounds once stood. In the case of Berwick Mounds (16SMY184), across the Atchafalaya River from present-day Morgan City, four large mounds in a quadrilateral layout were described by James Cathcart in 1819 and subsequently destroyed before ever having been investigated or adequately documented (Prichard et al. 1945). Any investigation of mound sites in the region must consequently take into account the void created by site destruction, ¤lled in when possible by studies of intact deposits. On the basis of the preliminary results of the PMAP investigation, there are 23 mound sites with known or suspected Plaquemine components in the study area, including those with mounds partially or largely destroyed. This number includes multicomponent sites and re®ects a likelihood for mixed deposits or ephemeral Plaquemine occupations at mound sites dating from Coles Creek (ca. a.d. 700–1200) and earlier periods. Assessment of site components is in many instances problematic given the uneven nature of earlier investigations, often consisting of only brief surface inspections or unsystematic surface collections. More well-known sites such as Lafayette (16SM17) and Meche-Wilkes (16SL18), thought to date from the Tchefuncte period, can be excluded from further consideration (Ford and Quimby 1945; Gibson 1990:81–85; Melancon 1999). The majority of mound sites with Plaquemine components are thought to have one or two mounds, some of which are located in the vicinity of larger mound sites. Association of mound sites with historic Chitimacha villages in the LMV has received relatively little attention (but see Giardino 1984 for the Mississippi delta). Drawing upon Gatschet (1883, 1907) and Chitimacha sources, Swanton (1911:337–344) was instrumental in identifying village locations. In reviewing the archaeological and historical data, Gibson (1978b, 1980) assessed the possible locations of more than 36 villages dating from as early as 1702. The paucity of documentary sources has been exacerbated by a lack of reliable archaeological information, including investigations that might con¤rm whether speci¤c sites were occupied during the eighteenth century. Many of the mound sites in question may have initially been constructed ¤ve or more centuries prior to the earliest documentary evidence of Chitimacha vil-

72

m ark a. rees

lages. Reoccupation or reuse of mound sites is consequently a possibility, further complicating matters of cultural af¤liation at multicomponent sites (e.g., Mann 2005). While tentative at best, correlation of mound sites with known Chitimacha villages nonetheless represents an attempt at making historical connections between the archaeological record, descendant populations, and their living traditions (i.e., Lightfoot 1995; Trigger 1980). The remainder of this chapter summarizes what is presently known about six of the extant major Plaquemine mound sites in the western Atchafalaya Basin, drawing upon the results of recent PMAP investigations (Figure 4.1). Three of these sites can be associated with known Chitimacha villages, although it is uncertain at present whether the names referred to the mound complexes, nearby (or neighboring) villages, or both. This is followed by a discussion of regional chronology, settlement patterns, and political culture, including the implications of this study for Native American culture history in the LMV. Moving from north to south, the sites to be considered are 16SL3, 16SL20, 16SM38, 16SM5 (Hi′pinimtc na′mu), 16SMY2 (Okû′nkîskîn), and 16SMY10 (Qiteet Kuti′ngi na′mu).

16SL3 Site 16SL3 is located on a terrace overlooking the south bank of a bayou that ®ows in an easterly direction toward its former con®uence with the Atchafalaya River, approximately 8.9 km (5.5 miles) to the east. An abandoned course of the Mississippi-Teche meander lies 1 km (0.6 miles) to the west (Saucier 1994:Plate 11). The mounds at 16SL3 are among the most impressive and best preserved of any in the study area (Figure 4.2). Surprisingly, these appear to have escaped the attention of archaeologists until the early 1970s (LDA Site File, 16SL3). The PMAP crew visited the site in April 2004 in order to produce a contour map and collect core samples from the mounds. Four rectangular platform mounds are positioned in a linear arrangement along the bayou, covering an estimated 3.6 ha (8.8 acres). The site may be much larger, however, considering the potential areas of habitation around the mounds. Mound A, the largest in the group, lies farthest east. It is approximately 4.8 m (15.7 feet) high with basal dimensions covering approximately 3,600 m2 (38,750 square feet). Mounds B and C are approximately 3.6 and 3.3 m high, respectively, and are slightly smaller in size but appear to be multistage platforms. Possible ramps on the north slopes of Mounds A and B suggest an orientation toward the bayou rather than a centrally located plaza ( Jones and Shuman 1991:49–57). Mound C is distinguished from the other mounds in having what appears

mounds of the w estern atch afa l aya basin

73

Figure 4.1. South-central Louisiana, showing the locations of major Plaquemine mound sites.

to be a small conical mound on its northern summit. Depressions in the spaces between Mounds A, B, and C have been interpreted as borrow pits, although the mounds appear to have been constructed on low ridges separated by natural swales. Mound D is the westernmost mound and located farthest south of the bayou. In fact, the distance between this mound and the other three raises questions as to whether it was constructed contemporaneously and

Figure 4.2. Site 16SL3.

mounds of the w estern atch afa l aya basin

75

how it might relate to the rest of the mound complex. Mound D is a low-lying platform mound 1.8 m (6 feet) high, the smallest in height of the four mounds. The summit of Mound D is conspicuously level and larger than that of any of the other mounds, indicating its likely use as a substructural platform. A ravine has been dug into the west slope of Mound D and partially ¤lled in with debris. The mounds at 16SL3 are otherwise in good condition, with only slight evidence of pot hunting, damage from off-road vehicles, and erosion. Both Coles Creek and Plaquemine components have been identi¤ed at 16SL3 on the basis of unsystematic surface collections and a small-scale salvage excavation in Mound A. Unprovenienced pottery sherds reported from the site include Avoyelles Punctated, var. Dupree, Coles Creek Incised, vars. Hardy and Coles Creek, Evansville Punctated, var. Sharkey, French Fork Incised, var. French Fork, Harrison Bayou Incised, var. unspeci¤ed, Indian Bay Stamped, var. unspeci¤ed, Mazique Incised, var. Manchac, Plaquemine Brushed, var. Plaquemine, and Pontchartrain Check Stamped. Projectile point types include Alba, Bayou Goula, and Scallorn ( Jones and Shuman 1991:59). Excavation of 60 cm of disturbed soil within a 1-by-2-m unit placed over a tree fall on the summit of Mound A in 1998 produced a small collection of Coles Creek Incised, Harrison Bayou Incised, Pontchartrain Check Stamped, and undecorated pottery sherds (McGimsey 1998:87–89). Following Neuman and Servello (1976:23), Jones and Shuman (1991:62) suggest 16SL3 may represent a “classic Coles Creek occupation.” Mound construction might just as plausibly date from the Mississippi period given the pottery types recovered from Mound A. A village area and associated midden have yet to be identi¤ed at the site. Without radiometric dates or diagnostic artifacts from controlled, stratigraphic contexts it is dif¤cult to add much more.

16SL20 The second mound site of interest is located on the south bank of a southeasterly ®owing bayou approximately 800 m (0.5 miles) northwest of its con®uence with a smaller bayou in St. Landry Parish. Less is known about this site than any in the study area, perhaps in part because of its relatively remote location 5.5 km (3.4 miles) west of the main channel of the Atchafalaya River. The site was originally described as having three mounds arranged in a triangular pattern and a fourth located at some distance to the southeast (LDA Site File, 16SL20). Mounds A, B, and C have been described as rectangular platforms ranging in height from 1.2 to 3.7 m (4–12 feet). Mound D is a low rise approximately 0.6 m in height and 63 m southeast of Mound A. Jones and Shuman

76

m ark a. rees

(1991:124–128) described the site layout as a quadrilateral or diamond with the short axis (64 m or 210 feet) running between the two largest mounds, A and C, and the long axis (128 m or 420 feet) between Mounds B and D. The area enclosed by the mounds is estimated to cover 0.98 ha (2.4 acres). Extensive pot hunting has been reported at the site and it has never been systematically investigated (LDA Site File, 16SL20). Site 16SL20 has been described as a late Coles Creek center on the basis of the layout of the mounds and a few small surface collections that consisted mostly of undecorated and Pontchartrain Check Stamped sherds ( Jones and Shuman 1991:129). A surface collection by the Louisiana southwest regional archaeologist in 1992 included mostly Baytown Plain, var. unspeci¤ed sherds (n = 22) and one sherd of Plaquemine Brushed (LDA Site File, 16SL20). On the basis of this and earlier surface collections it has also been associated with Plaquemine culture (Neuman and Servello 1976:24; Prentice 2000:274). Little more can be said about 16SL20 without further investigation.

16SM38 In contrast, more is presently known about 16SM38 than any other mound site in the study area as a result of a series of ¤eld school investigations by the University of Louisiana at Lafayette (Everett et al. 2002; Gibson 1990; Rees 2002, 2003). The site is located on a natural levee of the Mississippi-Teche meander, on the south bank of a bayou that ®ows easterly into craw¤sh ponds and a modern canal (Saucier 1994:279). Prior to these landscape alterations and construction of the west Atchafalaya protection levee, the bayou would have ®owed more directly into the Atchafalaya River approximately 12 km (7.5 miles) to the east. The low-lying area immediately surrounding the site is subject to periodic ®ooding and does not appear to have been historically cultivated. Although it was not mentioned by Moore (1913) or Neuman and Servello (1976:27), Phillips (1970:920–923, 950–951:Figures 446, 447) associated 16SM38 with both the Coles Creek Bayou Cutler phase and the subsequent Medora phase. Six earthen mounds were initially identi¤ed, positioned around a centrally located plaza and surrounded by a series of shallow depressions, some of which have been interpreted as borrow pits (Figure 4.3). Mound A is the largest in both area and height, at approximately 4 m (13 feet) higher than the surrounding terrain. As a consequence of off-road vehicle traf¤c, pot hunting, and erosion it is estimated to have been reduced in height by at least 0.5 m. Mound A is conical in shape and connected to Mound B by a low causeway or ramp. Mounds C and D are similarly joined on the western side of the

mounds of the w estern atch afa l aya basin

77

Figure 4.3. Site 16SM38.

plaza. Mound D, the second largest in the group, is the only mound that appears to have been a rectangular platform. Two low rises, designated Mounds E and F, are positioned on the south edge of the plaza but may actually represent midden deposits, perhaps similar to two small mounds at the Greenhouse site (Ford 1951:101–102; Kidder 1998b:139). Additional similarities with Greenhouse include the paired layout and general orientation of four mounds (Rees 2002). A seventh small mound was recently identi¤ed approximately 200 m northeast of Mound A. Mound G lies in a wooded area outside of the main mound complex and its potential association with the other mounds is uncertain. Excavations at 16SM38 indicate that the site extends well beyond the mound-and-plaza complex, with midden accumulation around Mounds A, B, and C. The entire site is conservatively estimated to cover at least 3 ha (7.4 acres) on the basis of the recovery of artifacts from 50-by-50-cm test units during the winter of 2001–2002. Trench walls excavated into two erosional gullies in Mound A revealed intact mound deposits, basket loading, and a thick layer of slope wash. A core placed through the summit of Mound A revealed two distinct episodes of mound construction above a pre-mound surface. Pottery sherds recovered from excavation units during the 1998, 2001, and 2003 ¤eld schools are classi¤ed primarily as Baytown Plain (78 to 88 per-

78

m ark a. rees

cent). Among the decorated pottery, the most common types are Pontchartrain Check Stamped, Plaquemine Brushed, Mazique Incised, and several varieties of Coles Creek Incised. Other decorated types include Avoyelles Punctated, Evansville Punctated, Harrison Bayou Incised, and L’Eau Noire Incised. Two Scallorn or Colbert-like projectile points were recovered along with one Alba point during the 2003 ¤eld school, further indication of a multicomponent, Coles Creek–Plaquemine habitation. Site 16SM38 is at present the most precisely dated mound site in the study area (Table 4.1). Wood charcoal from a small hearth (Feature 1) northeast of Mound A ¤rst investigated during the 2001 ¤eld school returned a radiocarbon age of 620 ± 70 b.p. (UGa-10994). Accelerator mass spectrometry (AMS) analysis of two wood charcoal samples from midden excavated in the vicinity of Feature 1 yielded radiocarbon ages of 690 ± 40 b.p. (Beta-190916) and 860 ± 40 b.p. (Beta-190917). Calibrations for these samples and the one from Feature 1 are consistent with the overall stratigraphic sequence and produce calendar age ranges spanning circa a.d. 1040–1483 (Beta-190917, Beta-190916, and UGa-10994). AMS analysis of wood charcoal from a pre-mound surface beneath Mound A produced a radiocarbon age of 850 ± 40 b.p., indicating that construction of Mound A likely postdates circa a.d. 1212 (UGa-10995). This corresponds very closely with one of two previous radiocarbon ages obtained for charred materials from the lowest stratum of mound ¤ll in Mound B, at 850 ± 70 b.p. (Beta-127975). A second radiocarbon age of 560 ± 70 b.p. was more recent, but in the context of mound ¤ll provides no more reliable indication of the actual time of mound construction (Beta-132764; Everett et al. 2002:9; McGimsey and van der Koogh 2001:6–7). On the basis of these results the habitation northeast of Mound A appears to date from the early thirteenth through the late fourteenth centuries, with the construction of Mounds A and B likely dating sometime after a.d. 1200. This corresponds with the excavated assemblages of decorated pottery and con¤rms an early Plaquemine habitation. Given the substantial inclusion of Coles Creek pottery types in mound ¤ll, there also appears to be a sizable Coles Creek component.

16SM5 (Hi′pinimtc Na′mu) Like the aforementioned sites, 16SM5 is located on a bayou that ®ows easterly into the Atchafalaya Basin, entering Lake Fausse Pointe through Lake Dauterive. Association of this site with the Chitimacha village of Hi’pinimtc na’mu is far from certain but proposed here as one plausible scenario requiring further scrutiny. Drawing upon Gatschet (1883) and Chitimacha sources,

80

m ark a. rees

Swanton (1911:344) described the location of Hi′pinimtc na′mu or “Prairielanding village” as being “on the western part of Grand Lake, at Fausse Pointe, near Bayou Gosselin” (see also Brightman 2004:643). Swanton adds that another village of the same name was located “on Lac d’Autre Rive, between Charenton and St. Martinsville.” Site 16SM5 is 8.0 km (5 miles) from Little Gonsoulin Bayou but in fact near Lake Dauterive. Although this is hardly a ringing endorsement for the identi¤cation of Hi′pinimtc na′mu, a Chitimacha village was likely in the vicinity. The Chitimacha Tribe of Louisiana maintains oral histories regarding the use and symbolism of the mounds at 16SM5. It is believed to have been the location of a sacred tree marking the boundary of the Chitimacha homeland (LDA Site File, 16SM5; Kim Walden, personal communication, 2002; cf. Swanton 1911:354). Historic villages of the Chitimacha are also thought to be located across the bayou to the south (McGimsey 1995:8–10). An association between 16SM5 and one of two villages referred to as Hi′pinimtc na′mu is therefore one reasonable interpretation (cf. Gibson 1978b:14). Site 16SM5 is located on a natural levee dissected by several low-lying swales and a minor, intermittent drainage (Figure 4.4). Although these channels accentuate the height of the mounds, it is unknown whether they predate mound construction or represent intentional modi¤cations to the landscape. Three earthen mounds are laid out in a triangular arrangement around a bend in the bayou. Mound A is approximately 2.6 m (8.5 feet) high with a southeasterly sloping summit and roughly rectangular shape. It is bounded on the south by the bayou and on the west by an intermittent channel. Mound B is located 25 m (82 feet) northeast of Mound A and is approximately 1.2 m (3.9 feet) high. At around the same height, Mound C lies 50 m (164 feet) east of Mound B. The shapes of mounds B and C are dif¤cult to discern at present but they appear to have been platforms. A low ridge extends from the southeast slope of Mound C, turning westward along the bank of the bayou. The bayou demarcates the southern edge of what appears to be a centrally located plaza, enclosing an estimated area of 0.5 ha (1.2 acres). Two low-lying, intermittently water-¤lled areas east and west of Mound B may represent borrow pits. Connected by shallow channels and intermittent drainages to the bayou on the south, the mound-and-plaza complex seems to be nearly entirely surrounded. However, it is unknown whether these gullies and low-lying areas are cultural features associated with mound construction. An extensive shell midden lies outside of the mound-and-plaza arrangement, across a shallow gully northeast of Mound C. The midden extends for at least 50 m (164 feet) along a natural levee overlooking the bayou and exhibits high

mounds of the w estern atch afa l aya basin

81

Figure 4.4. Site 16SM5.

concentrations of pottery sherds, bone, and shell. The site is conservatively estimated to cover 1.3 ha (3.2 acres) but may have originally been much larger before having been impacted by nearby road construction. Each of the mounds and the midden exhibit signs of extensive pot hunting. Given the size and location of this site it is somewhat surprising that it was not mentioned by Moore (1913) or Neuman and Servello (1976). McIntire (1958:Plates 7a, 7b, 10, 12, and 13) described it as a Coles Creek site with a small collection of Pontchartrain Check Stamped and Mazique Incised pottery sherds. Phillips (1970:920–923, Figures 446 and 447) subsequently associated this site with the Coles Creek Bayou Cutler phase. More recent surface collections suggest both Coles Creek and Plaquemine components. McGimsey (1995:8–10) recorded decorated pottery types such as Coles Creek Incised, vars. Blakely, Coles Creek, Curtis, Hunt, Keogh, Manchac, and Mazique, Pontchartrain Check Stamped, L’Eau Noire Incised, Churupa Punctated, var. Churupa, and Hollyknowe Pinched (LDA Site File, 16SM5). Decorated ceramics noted on the surface of the midden and west of Mound A during the 2003 PMAP investigation included small sherds of Carter Engraved, Harrison Bayou Incised, Mazique Incised, Plaquemine Brushed, var. Plaquemine,

82

m ark a. rees

Pontchartrain Check Stamped, var. Pontchartrain, and French Fork Incised, var. Larkin. Two AMS dates were obtained from core samples taken from Mound A and the shell midden at 16SM5. The ¤rst consisted of wood charcoal in sediment from a sub-mound A horizon, at 2.66–2.73 m below surface (Beta190914, Table 4.1). The radiocarbon age was 1210 ± 40 b.p., merely indicating a terminus post quem for the construction of Mound A. The second sample consisted of one shell (Rangia cuneata) specimen from 10 cm below the surface of the midden. It produced a radiocarbon age of 730 ± 40 b.p. (Beta-190915). This generally corresponds with three radiocarbon dates more recently obtained for samples excavated from two 1-by-1-m units in the midden. Two samples of wood charcoal near the bottom of the midden yielded radiocarbon ages of 1260 ± 50 b.p. (Level 6, Beta-200908) and 1460 ± 40 b.p. (Level 7, Beta-200909). The third produced a similar radiocarbon age of 1270 ± 50 b.p. for charcoal from Level 3 (Beta-201513). Several small, shell-tempered pottery sherds were recovered from the upper levels of one test unit, lending support to the later radiocarbon age of 730 ± 40 b.p. for a more recent and perhaps terminal midden deposit (Beta-190915, 2-sigma cal a.d. 1240–1300). As at 16SM38, these dates indicate a multicomponent or transitional Coles Creek– Plaquemine habitation, in this case spanning seven centuries between circa a.d. 600 and 1300. Of course, this does not rule out the possibility that the historic Chitimacha village of Hi′pinimtc na′mu was subsequently located here or in the vicinity.

16SMY2 (Okû′nkîskîn) On the basis of Chitimacha sources, Swanton (1911:344, 352) described the village of Okû′nkîskîn (“deep-shoulder”) on Grand Lake, a short distance from the town of Charenton (Tc8t Kasi′tuncki). The Chitimacha Tribe of Louisiana recognizes 16SMY2 as the village of Okû′nkîskîn (LDA Site File, 16SMY2). Another possibility is that 16SMY2 was the nearby village known as Ama’tpan na’mu (Brightman 2004:643–644; Gibson 1978b:15, 1982:458– 459; cf. Swanton 1911:343; Weinstein et al. 2003:140–141). Surprisingly little is actually known about the mounds at 16SMY2 despite the site’s having drawn the interest of scholars throughout the past century (e.g., Collins 1927:201; Gibson 1982:450–459; Kniffen 1938:194; McIntire 1958:Plate 5b; Moore 1913: 19–21; Neuman and Servello 1976:29, 64; Phillips 1970:920–923, 950–951; Prentice 2000:274; Swanton 1911:344, 352). It was one of the few mound sites in the study area visited by C. B. Moore (1913:19–21), due no doubt to the

mounds of the w estern atch afa l aya basin

83

site’s high visibility and accessibility on Grand Lake. Moore (1913:19) described the site as “¤ve mounds composed largely of shell” but found little of interest there, perhaps because most of the mounds were planted in sugar cane. Henry Collins, assistant curator of ethnology at the U.S. National Museum, visited the site in 1926 at the request of the Bureau of American Ethnology and U.S. National Museum. He was accompanied to the site by Benjamin Paul, Chief of the Chitimacha. Collins described the site as consisting of three earthen mounds and an extensive shell midden. He excavated a test pit into one mound that he described as eight feet high (Collins 1927; McGimsey 2001:114, 125–126). McIntire (1958:Plates 5a, 5b, 7b, 8b, 13) subsequently described the site as having Troyville, Coles Creek, and Plaquemine components on the basis of surface-collected pottery. Phillips (1970:920–923, 950–951, Figures 446 and 447) associated 16SMY2 with the Coles Creek Bayou Cutler phase and Medora phase of the Mississippi period. Okû′nkîskîn may at one time have consisted of as many as six mounds and shell midden in linear arrangement along Grand Lake. It has been heavily impacted by road construction and recreational development. The shell midden had already been damaged by the time of Collins’s visit and is today nearly obliterated. Beginning on the western edge of the site, Mound A is a circular earthwork that has been damaged by modern landscape renovations. Mound B was destroyed by construction of a boat landing and access road. Mound C lies east of mounds A and B and may have been the focus of investigations by Moore (1913) and Collins (1927). Mound D is a large, rectangular platform approximately 3 m in height. It is the largest of the existing mounds at the site but has been impacted by a ditch that cuts across the northern slope. Mounds E and F, recorded during a site visit in 1997, are low rises adjacent to a remnant of the shell midden (LDA Site File, 16SMY2; McGimsey 2001:86, 125–126; Weinstein et al. 2003:141). Based on Collins’s brief visit, McGimsey (2001:125) suggests a Coles Creek component associated with Mound C. The pottery in this collection consists of a few sherds of Baytown Plain, Coles Creek Incised, vars. Blakely, Greenhouse, and Pecan, Plaquemine Brushed, var. Plaquemine, Mazique Incised, var. Manchac, and one sherd each of Pontchartrain Check Stamped, var. Pontchartrain, Evansville Punctated, var. Rhinehart, Carter Engraved, var. Shell Bluff, L’Eau Noire Incised, var. unspeci¤ed, French Fork Incised, var. McNutt, and Beldeau Incised, var. Bell Bayou. Plaquemine and historic Chitimacha components are also indicated on the basis of these pottery types, historic documentation, and oral histories (LDA Site File, 16SMY2).

84

m ark a. rees

16SMY10 (Qiteet Kuti′ngi Na′mu) The ¤nal site to be considered here is recognized by the Chitimacha Tribe of Louisiana as the village of Qiteet Kuti’ngi na’mu ( Jason Emery, personal communication, 2003; McGimsey and Cring 2003:184). It is located on the north bank of Bayou Teche near a former con®uence with the Atchafalaya River, subsequently diverted by levee construction. On the basis of conversations with Chief Benjamin Paul, Swanton (1911:344) referred to the village of “Tca′ti Kuti′ngi na′mu, at junction of Bayou Teche with Bayou Atchafalaya.” Gatschet (1883) mentioned this village as well, although since the con®uence of the Teche and Atchafalaya has changed over the years it is unclear whether Tca′ti Kuti′ngi na′mu was located at 16SMY10 or farther downstream near Morgan City (McGimsey and Cring 2003:208). Brightman (2004:643) refers to the same site as “Chati kutingi namu” but places it at the Atchafalaya River and Bayou Courtableu. While the issue is far from resolved, 16SMY10 is referred to here by the Chitimacha designation Qiteet Kuti′ngi na′mu. As discussed below, recent research con¤rms an eighteenth-century Native American presence at the site. C. B. Moore (1913:19) appears to have brie®y visited this location and described two large mounds, 13 feet and 10.5 feet in height, “in full view from the water [Bayou Teche]” (Weinstein et al. 2003:141–143). Moore had his crew dig into these mounds but found little of interest, except for the skeletal remains of one individual, which had previously been damaged by looters. Indiscriminate digging and erosion appear to have already severely truncated the larger of the two mounds by the time of Moore’s visit. It was this mound, according to Moore (1913:19), that was “connected with a shell ridge on the eastern side, formed almost entirely of a variety of clam-shell, Rangia cuneata, and midden deposit extending along the bank of the river.” In all, Moore spent less than three hours at the mounds “at Moro plantation.” Because the site was originally recorded in 1952 as a ceremonial center consisting of “four mounds and a shell midden,” the identi¤cation of 16SMY10 with the mounds at Moro plantation is not certain (LDA Site File, 16SMY10). However, it is possible that Moore missed two of the smaller mounds (Weinstein et al. 2003:142–143). If so, the larger mound described by Moore as attached to a shell ridge and midden may have been destroyed soon thereafter, perhaps as a result of continued bank erosion and/or the excavation of a narrow channel or boat slip. McIntire and Saucier recorded the site as a ceremonial center with Coles Creek, Plaquemine, and historic components (LDA Site File, 16SMY10). McIntire (1958:79) later referred to an initial Coles Creek

mounds of the w estern atch afa l aya basin

85

Figure 4.5. Site 16SMY10.

occupation on the basis of its location on the Teche Ridge. However, his pottery tabulation also included later types such as Fatherland Incised, Australia Interior Incised, and Plaquemine Brushed (McIntire 1958:128, Plate 13). Phillips (1970:920–923, 950–951, Figures 446 and 447) associated the site with the Bayou Cutler, Medora, and Delta Natchezan phases (cf. Neuman and Servello 1976:64–65). Collections in the LSU Museum of Natural Science include Anna Incised, vars. Anna and Australia, Coles Creek Incised, vars. Blakely, Hardy, and Pecan, French Fork Incised, vars. Larkin, McNutt, and Lafayette, Harrison Bayou Incised, var. Harrison Bayou, Mazique Incised, var. Mazique, Plaquemine Brushed, var. Plaquemine, and Pontchartrain Check Stamped, var. Pontchartrain (McGimsey and Cring 2003:187, Table 14.1). The site was revisited in 1991 as part of a larger survey of the lower Bayou Teche and again described as four earthen mounds with a shell midden (Goodwin et al. 1991a; Goodwin et al. 1991b:103–105; LDA Site File, 16SMY10). In December 2003 the PMAP investigation recorded three earthen mounds and a linear shell midden extending along the bank of the bayou (Figure 4.5). Mound A, the largest extant mound at the site, is approximately 2.8 m (9.2 feet) in height and 30 m (98 feet) in diameter. Mound B is located 25 m (82 feet) to the northeast and is approximately 2.4 m (7.9 feet) high. The south slope of Mound B appears to have been sharply truncated from the summit to the base. Although Moore (1913:19) noted similar damage at Moro plantation, it was in reference to the larger mound located “nearer the bayou.” Since this is

86

m ark a. rees

not an accurate description of Mound A, it once again raises the possibility that a fourth mound adjacent to the midden was subsequently obliterated. Mounds A and B appear to have included platforms, although the precise shapes are dif¤cult to establish as a result of sustained damage from pot hunting, off-road vehicles, and erosion. Mound C, located 30 m southeast of Mound B, is more noticeably a rectangular platform. It is approximately 0.8 m (2.6 feet) in height but may have been con®ated by years of cultivation and erosion. The shell midden extends along the bank of the bayou from 10 m (33 feet) southeast of Mound A nearly to the southwest slope of Mound C. The existence of a fourth mound could not be con¤rmed during the PMAP study, despite several attempts at coring an area of higher elevation on the western edge of the midden. The contour map indicates ramps may have extended from the northeast slope of Mound A and southwest slope of Mound B. Along with Mound C, the mounds appear to be positioned around a centrally located plaza, with the midden making up the southern boundary. The area enclosed by the mounds and midden is approximately 0.44 ha (1.09 acres) and the overall site size is conservatively estimated at 1.2 ha (2.97 acres). However, this does not include potentially inhabited areas outside of the mound group subsequently impacted by modern development. An associated village was reportedly destroyed by a large, modern borrow pit to the north (Goodwin et al. 1991b; Neuman and Servello 1976:29, 64). It is unknown whether this area or the mound group itself would have been the location of the Chitimacha village of Qiteet Kuti′ngi na′mu. The PMAP investigations support successive Coles Creek, Plaquemine, and historic habitations at 16SMY10. Cores were placed through each of the mounds and the midden, from which two samples were submitted for radiometric analysis. The ¤rst consisted of shell (Rangia cuneata) fragments from 27–50 cm beneath the surface of the midden (Beta-190918). The resulting radiocarbon age was 1210 ± 50 b.p. (2-sigma cal a.d. 1290–1450), in concurrence with the ceramic evidence for an early Plaquemine occupation. The second sample consisted of wood charcoal from beneath the lowest layer of ¤ll in Mound A. AMS analysis yielded a radiocarbon age of 1010 ± 40 b.p. (Beta-190919, 2-sigma cal a.d. 980–1050 and cal a.d. 1100–1140). It is therefore probable that construction of Mound A was most likely begun sometime after a.d. 1020, perhaps coinciding with a terminal Coles Creek or transitional Coles Creek–Plaquemine habitation. Recent investigations shed new light on the association of 16SMY10 with

mounds of the w estern atch afa l aya basin

87

the Chitimacha village of Qiteet Kuti′ngi na′mu, referred to by Swanton (1911:344) as Tca′ti Kuti′ngi na′mu. A site inspection in July 2003 yielded evidence of exposed human remains and artifacts eroding from the north slope of Mound A. A partially damaged burial in one of several off-road vehicle paths was in imminent danger of destruction. Personnel from the LDA, Chitimacha Tribe of Louisiana, and PMAP conducted a salvage excavation with permission from the landowner and a permit issued by the Louisiana Unmarked Human Burial Sites Board. Subsequent analysis indicated the primary interment of a 3-year-old child with numerous grave goods dating between a.d. 1763 and 1820. The burial offerings consisted of both indigenous items and trade goods, such as black, white, and blue glass beads (McGimsey and Cring 2003:200–205). Pottery sherds recovered from disturbed mound ¤ll consisted of Baytown Plain, var. unspeci¤ed, Carter Engraved, var. Sara, and Larto Red, var. unspeci¤ed. Although the burial likely postdates initial mound construction by several centuries or more, it supports either historic reuse of the mound precinct or continued inhabitation in this locale potentially associated with the village of Qiteet Kuti′ngi na′mu.

Discussion Several observations can be drawn from the preceding description of sites. First and foremost is the paucity of information on even the largest mound sites in the region. While some sites have been known about for a century or more (e.g., 16SMY2 and 16SMY10), others have until relatively recently gone virtually unnoticed by archaeologists (e.g., 16SL3 and 16SL20). Site destruction from indiscriminate digging, off-road vehicles, and erosion in many instances continues unabated, highlighting the critical need for additional information on extant mound sites. The PMAP has begun to document such poorly known sites by producing topographic maps, conducting systematic coring, and carrying out small-scale test excavations. Only two of the six sites discussed here have been systematically investigated in any detail (16SM38 and 16SM5). Second, greater temporal and spatial resolution regarding mound construction and use, site habitation, potential abandonment, and reuse are needed in order to provide tighter chronological controls for a more re¤ned culture history of the western Atchafalaya Basin. Third, combined with studies of material culture and subsistence, information on the construction, layout, and locations of mound sites has the potential to signi¤cantly advance our understanding of regional polities during the late Coles Creek and Mississippi pe-

88

m ark a. rees

riods (ca. a.d. 1000–1700). The present study has intriguing implications concerning the last two issues: regional chronology and Native American political culture. Chronology building does not in itself explain social or political dynamics, a point emphasized in earlier processualist critiques of a culture historical approach (e.g., Binford 1968; Flannery 1967). Yet archaeologists have more recently recognized that re¤nement of regional chronologies is fundamental to understanding political development and decline (e.g., Knight and Steponaitis 1998; Pauketat 1995). Advancement from cultural periods to phases and subphases in Mississippian archaeology has been characterized by more nuanced distinctions regarding regional political economy, allowing for consideration of human agency alongside long-term ecological factors (Pauketat 1997a, 1997b, 1998a). The gist of this approach has been to rearticulate the preColumbian, Native American past as a historical process (e.g., Pauketat 2001a, 2001b). The relevance of historical processualism, besides taking into account Native American cultural traditions, contradictions, diversity, and practices as agents of change, is in bridging the formidable protohistoric transformations of escalating contacts and colonialism (Lightfoot 2001; Wesson and Rees 2002). The PMAP has so far contributed only to the preliminary advancement of a regional chronology for the western Atchafalaya Basin. Earlier studies have hinted at a centuries-long hiatus in the occupancy of the western margin of the LMV prior to the Coles Creek period (Gibson 1990:25; see Phillips 1970: Figures 444 and 445). On the basis of the series of radiometric dates presented here, it appears that initial construction and habitation of major mound centers dates from the late Coles Creek and early Mississippi periods, or what has been called the Coles Creek–Plaquemine transition (i.e., Weinstein 1987a:87– 93). Eight of 13 available radiometric dates from three sites have calibrated calendar ranges within a.d. 1025–1483, with intercepts clustering between a.d. 1190 and a.d. 1405 (Table 4.1). The remaining dates are associated with sub-mound contexts, representing a terminus post quem for mound construction at 16SM5 and 16SMY10 (Beta-190914 and Beta-190919) and a Coles Creek context for the initial midden deposit at 16SM5. The argument for Plaquemine culture as representing Mississippianized Coles Creek ¤nds little support at 16SM5, where mound construction seems to have been preceded by two centuries of habitation and any evidence for interaction with Mississippian societies appears to have been late and ephemeral. While additional dates are urgently needed from these and other sites, the thirteenth and fourteenth centuries stand out as a time when Plaquemine peoples, including ancestors of the Chitimacha, were building large earthen

mounds of the w estern atch afa l aya basin

89

mounds that may have served as political-administrative and ceremonial centers. Arguments regarding the contemporaneous or successive establishment of these centers will require further re¤nement of a regional chronology. Issues regarding cultural af¤liation come into view at the other end of the chronology. While the importance of documentary evidence and oral histories has long been realized in establishing the locations and cultural af¤liations of sites in the LMV (e.g., Giardino 1984), there is much archaeology that remains to be done. In particular, compelling associations of the Chitimacha Tribe of Louisiana with three of the sites mentioned here indicate the potential for multidisciplinary research into mound building, reuse, and the persistence of traditions in the face of momentous demographic, political, and economic change (e.g., Mann 2005). The investigation of eighteenth-century villages and potential associations with mound centers represents a virtually untapped source of information on Chitimacha history spanning the Mississippi and early historic periods. It may be no coincidence that the best-known sites are those located farthest south, within the homeland of the Chitimacha Tribe of Louisiana. Archaeological and ethnographic evidence for protohistoric or early historic habitation, while sparse, is restricted at present to these sites (16SM5, 16SMY2, and 16SMY10). This raises questions regarding the construction, reoccupation, and reuse of mound sites that essentially demarcate the western boundary of the Chitimacha homeland. Given the relatively later habitations at these southernmost mound sites, perhaps sites to the north such as 16SL3 and 16SL20 correspond to the frontier of a yet earlier, ancestral Chitimacha homeland. A creation myth of the Chitimacha is purported to recount their origin in the vicinity of the Natchez (Swanton 1911:348, 356), with whom they have been compared and contrasted (e.g., Grif¤n 1952:363; Kniffen et al. 1987:53– 55). Additional re¤nement of a regional chronology is also necessary in order to evaluate such propositions and will undoubtedly assist in establishing clearer connections between archaeological, ethnographic, and historical sources. Information on the timing and scale of mound and village construction also has the potential to inform our understanding of regional political culture. Large political-administrative centers elsewhere in the Southeast were characterized by demographic nucleation and decentralization concurrent with the development and decline of regional polities. Six major multimound sites with known or suspected Plaquemine components were constructed at fairly regular intervals along the western margin of the Atchafalaya Basin, ranging in distance from 22.9 km (16SM38 to 16SM5) to 28.9 km (16SL3 to 16SL20), the mean straight-line distance being approximately 26.4 km

90

m ark a. rees

(16.4 miles). If distance or travel time between centers can be used to gauge the scale of political organization, as has been argued for Mississippian polities (e.g., Hally 1993, 1999), one might be tempted to draw inferences regarding the multimound sites discussed here. Of course, estimates of travel time must ultimately take into consideration the ef¤ciency of water transport, especially in the LMV. Arguments for a series of contemporaneous or consecutive polities will also ultimately require re¤nement of the regional chronology introduced here. The fact that four of the northernmost mound centers were located on major east–west drainages suggests political and economic interactions focused consistently around the regional transport and exchange of resources and information between the basin and Teche Ridge, rather than a single, regionally integrated polity. If the latter were the case, one might expect a strategic arrangement of mound centers along the Teche Ridge, taking advantage of both the higher elevation and the bayou as a major north–south transportation route. Yet surprisingly few mound centers are accordingly positioned (e.g., 16IB40 and 16SMY10). Instead, site placement suggests an overall east–west orientation in the cultural landscape, once more resembling the historic boundaries of the Chitimacha homeland and potentially indicating a borderland with ancestral Attakapa of the Prairie Terrace and perhaps other Coles Creek, Plaquemine, and Mississippian communities of the southwest Louisiana Chenier Plain (Swanton 1911:360–363). Did Plaquemine peoples, including ancestors of the Chitimacha, establish a half-dozen autonomous or loosely allied polities on the western edge of the Atchafalaya Basin prior to the fourteenth century? In order to answer such questions it will be necessary to collect additional information on the timing of mound construction and evidence for contemporaneous habitation. Whether these centers were built and inhabited in close succession or at different times, the data seem to indicate the shifting or movement of regional political culture. By examining the timing of mound construction and midden deposition it will be possible to more thoroughly address the pace and scale of regional political development. Lastly, ideas concerning dispersed settlement patterns and ceremonial centers must also be reexamined, particularly in light of the midden deposits noted at sites such as 16SM38, 16SM5, and 16SMY2 (Neuman 1984:260). Of the sites discussed here, none straightforwardly correspond to what might be thought of as a “typical” design for a Plaquemine ceremonial center (Kidder 1998b:141–148; cf. Neuman 1984:258–268; Quimby 1951). Rectangular platforms, conical mounds, causeways, shell middens, and ridges were incorpo-

mounds of the w estern atch afa l aya basin

91

rated in a variety of arrangements, including linear alignments along bodies of water (16SL3 and 16SMY2), quadrilaterals with centrally located plazas (16SL20, 16SM38, 16SMY184), and triangular patterns with the addition of a low ridge or midden (16SM5 and 16SMY10). Interpreting site layout and design as the cumulative result of intentional actions rather than merely accumulative deposition or accretional processes, it is possible to form general inferences regarding the social import of monumental architecture and ceremonial spaces (e.g., Knight 1998; Wesson 1998). The existence of social templates or archetypal patterns for mound groups and ceremonial spaces is suggested by similarities among the above sites, as well as between 16SM38 and the Greenhouse site in Avoyelles Parish (Ford 1951). The meanings of these archetypes are not entirely beyond the grasp of archaeologists, in that the built environment effectively represents the materialization and negotiation of social identities (Lewis et al. 1998). Similarities between sites 16SM5 and 16SMY10, ostensibly associated with the Chitimacha villages of Hi′pinimtc na′mu and Qiteet Kuti′ngi na′mu, might thus be viewed as diagrammatic of comparable community organizations. Substructural mounds corresponding to residential and mortuary precincts of the nobility, as noted among the Chitimacha, would have been designed, constructed, and administered by corporate groups or lineages (LDA Site File, 16SM5; Swanton 1911:348–351). The higher ranking or elite of Chitimacha society are reported to have comprised endogamous lineages. Marriages in such cases have been considered ethnographically to create “new alliances among equals” or maintain existing social inequalities, particularly in the context of land tenure (Comaroff and Comaroff 1991:138–140; Wolf 1959:228). Following exhumation and ceremonial bundling in the ha ′na katci ′ or “bone house,” the remains of a Chitimacha chief and his possessions were interred in a mound. According to Swanton (1911:351), “the mounds erected over chiefs are said to have been 4 or 5 feet high.” In view of this ethnographic analogy and the available archaeological evidence, it seems ill-advised to portray the western Atchafalaya Basin as a cultural backwater to the Plaquemine heartland of the Natchez Bluffs or Yazoo Basin. From the perspective of Coles Creek–Mississippian hybridization, inhabitants of the western Atchafalaya Basin may appear to have been left behind, with clearer connections to Coles Creek in terms of both mound complexes and pottery; shell-tempered sherds are nearly nonexistent in the assemblages from major mound sites discussed here. Even from the viewpoint of indigenous Coles Creek development, most of the mound sites discussed

92

m ark a. rees

here appear lacking in size and grandeur (except perhaps for 16SL3), particularly in comparison with large Coles Creek mound centers such as Raffman (Roe, this volume). From a regional, political perspective, the mound groups of the western Atchafalaya Basin take on new meanings. Blitz and Livingood (2004:299) point out that the size or volume of Mississippian platform mounds is a re®ection of both duration of use and political power or authority, as well as the mobilization of labor in different social contexts. To equate the magnitude of mound complexes with comparative cultural prominence seems ill advised, particularly for a region in which so little is known concerning the social contexts of mound construction and use. Mound centers are representative of historical processes of political consolidation, alliance formation, shifting authority, and ultimately the movement of peoples across a cultural landscape. In understanding the societies that constructed mounds in the western Atchafalaya Basin, it will be more productive to frame new ¤ndings not as marginal to Coles Creek or Plaquemine culture but in terms of regional political culture, as a historical process that produced living traditions such as the Chitimacha.

Acknowledgments The Plaquemine Mounds Archaeological Project was supported by a Research Competitiveness Subprogram grant from the Louisiana Board of Regents Support Fund. I would like to thank all those involved in this research at the University of Louisiana at Lafayette, including the many ¤eld school students, assistants, and volunteers who have participated in the project. Lance Blanchard, Rikki Dugas, Jamie Lee Grossie, Madelyn Hebert, Taylor Lasley, Rene “Chip” Lorio, and Rachel Moss were among the student assistants who contributed in the laboratory and ¤eld, despite the most inclement conditions. James Fogleman, avocational archaeologist with encyclopedic knowledge of St. Landry Parish sites, contributed valuable information on site reconnaissance. Essential equipment for the ¤eld schools was supplied by a grant from the Student Technology Enhancement Program at UL Lafayette. A special debt of gratitude is owed to all of the landowners who provided access to their property, especially Dr. Jean Kreamer and family, The Archaeological Conservancy, Mr. and Mrs. Silton Boudreaux, Allen Babineaux, Karen Broussard, Lou Ella Bonin, Anna Belle Bonin, and Suzanne Cocke. This research could not have been accomplished without their generosity and hospitality. At their request, an attempt has been made to safeguard the speci¤c

mounds of the w estern atch afa l aya basin

93

locations of mound sites in order to protect against further destruction from looting and off-road vehicles. At the University of Louisiana at Lafayette, C. Ray Brassieur continues to offer provocative insights on the cultural landscape of south Louisiana. Chip McGimsey, southwest Louisiana regional archaeologist, has been especially helpful in the ¤eld and laboratory. The Louisiana Division of Archaeology provided access to the state site ¤les and made the generous loan of a Giddings rig coring device. The late Thomas H. Eubanks, former Louisiana State Archaeologist, obtained ¤nancial support from the National Association of State Archaeologists at the beginning of the project for the emergency stabilization of Mound A at 16SM38. Kimberly Walden, Cultural Director with the Chitimacha Tribe of Louisiana, provided matching funds from the Chitimacha and along with John Paul Darden has assisted in the ¤eldwork and offered much appreciated advice. Patrick Livingood, co-editor and coorganizer of the Plaquemine Problems symposium, provided constructive remarks on this chapter. Jay Johnson and Martha Rolingson also offered helpful comments. Nonetheless, I am accountable for any errors. Finally, I am grateful for the personal support and devotion of the four most in®uential women in my life: Johanna Chautin, Sarah Caitlin, Jennifer Catherine, and Gwyneth Margaret Rees.

5 Transitional Coles Creek–Plaquemine Relationships on Northwest Lake Salvador, St. Charles Parish, Louisiana Malcolm K. Shuman

Despite the criticisms of the New Archaeologists about the use and implications of the taxonomic systems employed by culture historians (e.g., Binford and Binford 1966), the effective use of taxonomic terminology has continued to be a problem. There are several reasons for this. At least with regard to the Lower Mississippi Valley, Belmont (1982b:69–70) suggested the problem has been simple confusion among archaeologists, which has led them to use many taxa interchangeably. This confusion is at least partly the result of the inability of investigators to decide among themselves what the various terms mean. Just as important, however, are the limitations inherent in the way culture historians go about their work. Regardless of the criticisms leveled at them by processualists and their successors, the major ¤gures in southeastern archaeology have long been aware of the problems they faced with their approaches. Willey and Phillips attempted to resolve some of the issues with their evolutionary synthesis in 1958. The junior author, in the introduction to his monumental work on the archaeology of the Yazoo Basin, admitted to being unabashedly “old fashioned” in stating that such theory as he proposed to employ “is in fact little more than ‘method.’ ” Phillips gently criticized such theoreticians as Ford who, in attempting to derive laws of cultural change, “make certain assumptions . . . that are in effect the laws he is trying to deduce” (Phillips 1970:3). To Phillips, then, our tools are de¤cient, but at the present they are the only tools we have. This is especially problematic when we try to extend our terminology spatially: “So long as we are forced to operate with the mixed culturechronological criteria inherent in our ‘culture periods,’ it is more important that these be locally intelligible than that they be capable of expansion across the board” (Phillips 1970:8). This, then, highlights one of the problems this

north w est l a k e sa lva dor, st. ch arles parish

95

chapter will address. Let us now consider the meanings commonly ascribed to the terms Coles Creek, Plaquemine, and Mississippian. The term Coles Creek in archaeology derives from the landmark work of James A. Ford (1936), who analyzed collections from Louisiana and Mississippi. In that work Ford refers to Coles Creek as a ceramic complex. Thereafter, he and Willey, in one of the ¤rst syntheses of eastern archaeology, call it a period de¤ned on the basis of ceramic traits (Ford and Willey 1941:345). At that time the Coles Creek period was thought to have lasted until the middle of the sixteenth century (Ford and Willey 1941:346). Since that time, besides being described as a period (Phillips 1970), it has also been described as both a period and a culture (Williams and Brain 1983:369, 405), a hyphenated culture, linked to the preceding Troyville (Neuman 1984), and an interval (Weinstein and Kelley 1992:351). Central to all these is the notion of a set of ceramic characteristics, although since the time of Ford other diagnostic traits have been added, such as site features and plans and overall settlement patterns (Williams and Brain 1983:405). As regards ceramics, Coles Creek is recognizable by “the use of incised, stamped, and punctated pottery types in which the decorative zone is largely restricted to a band around the rim of the vessel” (Weinstein and Kelley 1992:37). Architecturally, Coles Creek sites often consist of small platform mounds centered on plazas (Weinstein and Kelley 1992:37). The heartland for Coles Creek is the mouth of the Red River and the Lower Yazoo Basin (Miller et al. 2000:44), though it eventually extended to the coastal areas. The dates conventionally accepted for Coles Creek in south Louisiana are a.d. 700–1200 (Weinstein 1987a:86). The concept of Plaquemine derives from the work of Quimby (1951) at the Medora site in west Baton Rouge Parish, Louisiana. As Neuman (1984:258) writes, little about Plaquemine is distinctive. However, he notes that, besides the decorative technique of brushing as applied to ceramics, other apparent characteristics are “the occasional use of shell to temper the clays; small projecting lug handles on the vessel rims; engraving; and the general absence of stamped designs” (Neuman 1984:259). Most Plaquemine sites contain at least two rectangular mounds, but the number of mounds may be as many as 24, centered on two plazas (Neuman 1984:259). It goes almost without saying that the term Plaquemine, like Coles Creek, has been ascribed to varying taxonomic units (see Miller et al. 2000:44), though probably culture has been most frequently employed (e.g., Neuman 1984). The Plaquemine phenomenon may be conceived as a triangle with coastal Louisiana as its base and Greenville, Mississippi, at its apex (Neuman 1984:259). Dates for Plaquemine are usually given as a.d. 1200–1700 (Weinstein 1987a:86).

96

m a lcolm k. shum a n

Finally, the term Mississippian derives from the pioneering work of W. H. Holmes, who employed it to describe the shell-tempered ceramics often found with burials in the Central Mississippi Valley (Holmes 1903). Thus, from its inception, the term was linked to shell tempering as a technique of ceramic manufacture. This term, like the others discussed, came to be used for the “entire cultural complex which usually accompanies this characteristic pottery” (Ford and Willey 1941:348). It, too, is variously called a culture (Neuman 1984), a period (Weinstein and Kelley 1992:37), and even a stage (Miller et al. 2000:47). The term has been so variously and confusingly used, in fact, that Brain, for this reason, refused to employ it in the title of a paper on the subject (Brain 1978:331). The Mississippian cultural complex is associated with the great mound site of Cahokia in East St. Louis, beginning about a.d. 800 (Neuman 1984:272). According to Neuman (1984:273), these in®uences radiated outward to other regional centers as far a¤eld as the Aztlan site in Wisconsin, the Etowah site in Georgia, and the Moundville site in west-central Alabama. In addition to the presence of shell-tempered pottery, fully developed Mississippian culture is de¤ned by a reliance on agriculture; the presence of extensive mound sites, sometimes consisting of 20 or more mounds; and a ceremonial complex variously referred to as the Death Cult, the Southern Cult, the Buzzard Cult, and the Eagle Warrior Complex (Neuman 1984:272– 278). Many centers of Mississippian culture, such as Cahokia and Moundville, seem to have declined in the fourteenth century into more dispersed and less centralized communities (Brown 2003:224, citing Wesson 1998:119–120; Young and Fowler 2000:311). So pervasive was the in®uence of the Mississippian culture concept that this term was also assigned to a period that began around a.d. 800 in the middle Mississippi Valley and reached the lower valley by about a.d. 1200 (Weinstein and Kelley 1992:31; Young and Fowler 2000). The Moundville in®uence stretched to the inhabitants of the Gulf coast, so that by a.d. 1200 we ¤nd elements of Mississippian culture at sites from Florida to Louisiana (Blitz and Mann 2000; Brown 2003). The earlier tendency to refer to Coles Creek as a period as well as a culture may be assigned to the literary styles of the various archaeologists who have written about the subject. The use of the term Mississippi for both culture and period, however, would appear to be a lot more self-conscious, in that there is an assumption that during the span a.d. 1200 to roughly a.d. 1700, Mississippian culture strongly in®uenced the other cultures of the Southeast. The nature of that in®uence is one of the preoccupations of specialists in the prehistory of the Southeast coast and one of the issues considered by this essay. On charts of culture chronology, the Coles Creek phenomenon is usually divided into a

north w est l a k e sa lva dor, st. ch arles parish

97

Coles Creek proper, lasting from a.d. 700 to 1000, and a transitional Coles Creek, lasting from a.d. 1000 to 1200 (Weinstein 1987a:86). What I ask here is what, precisely, did Coles Creek transition into? The two possibilities usually cited are the Plaquemine and the Mississippian cultures. Here there is a divergence of opinion. For Phillips (1970), Plaquemine was a development from the preceding Coles Creek culture, but over time interaction between Plaquemine and the external Mississippi in®uence produced changes in the former, so that as time passed there was considerable convergence between Plaquemine and Mississippian proper. Brain (1978), on the other hand, sees Coles Creek as the essential culture that, about a.d. 1200, under Mississippian in®uences, evolved into Plaquemine. As a result of Brain’s in®uence, “phases . . . which were formerly considered Plaquemine culture manifestations . . . are now placed late in the Coles Creek period and assigned to a transitional Coles Creek culture” (Weinstein and Kelley 1992:38). This essay does not provide a de¤nitive answer to the question posed above, but it does suggest the pitfalls of overreliance on techniques of pottery manufacture in assigning cultural categories. This problem was noted long ago by Phillips (1970:108, 152), who predicted that with the passage of time tempering would come to be de-emphasized as a primary typological criterion.

The 2002 Lake Salvador Shoreline Protection Project The sites of 16SC4 (Bois Chactas) and 16SC14 (Tabatiere Perdu) are middens composed primarily of Rangia shells, though occasional freshwater and saltwater bivalves are also found in the deposits. Each site is situated on the natural levee of a distributary that at one time emptied into Lake Salvador (Figure 5.1). Each site has suffered major erosion from the transgression of the lake over time, as can be seen from dead oak trees as much as 30 m offshore. The larger site is 16SC4 (Bois Chactas), which is also the site farther to the east. It consists of a shell beach approximately 210 m long from east to west, with approximately 75 m separating the lake on the south from the marsh to the north. Though the name implies the existence at some past time of Choctaw Indians at this location, no historical evidence of such an occupation could be found. The tribe most closely associated with this site and others in the area is the Washa. The site’s elevation is about 1 m above the lake surface in most places, but at its western extreme it rises to 2 m; it is said by locals that at one point this prominence was higher but that the lake has taken most of it.

Figure 5.1. Lake Salvador and archaeological sites mentioned in the chapter.

north w est l a k e sa lva dor, st. ch arles parish

99

The second site, 16SC14 (Tabatiere Perdu), is located about 5 km northeast of the point at which Bayou des Allemands empties into Lake Salvador and is the westernmost of the three sites examined in the Lake Salvador Shoreline Protection Project. It is 1.75 m high at the point overlooking the lake, where there are two camp structures, but this elevation, which is part of a relict distributary ridge, slopes down sharply to the lake on the south. The main portion of the site, which is wedged between the marsh on the north and the lake on the south, consists of about 1 acre, but on the northwest a shell ridge some 1.5 m high extends for 200 m north into the marsh. This ridge is part of the natural levee that continues into the main part of the site. These sites are in an area rich in archaeological remains. Several dugout canoes have been taken from the lake near them, and, indeed, one such canoe sat for years behind a camp house on 16SC4 until the craft deteriorated. The famous Grand Temple site (16LF4), a shell mound 15.2 m high, for years stood at the mouth of Bayou des Allemands where the bayou meets the lake. Because geomorphological reconstructions suggest that Bayou des Allemands did not begin to empty into Lake Salvador until a.d. 500, this impressive shell mound must have been no older than Baytown times. The mound was gradually destroyed by the Jahnke dredging company of New Orleans, which sold the shell, and what was left was ¤nished off by Hurricane Juan in 1985. Other shell middens are present up and down the lake and several signi¤cant archaeologically known sites are within approximately 20 km of the study area. These include Pump Canal, Bayou des Familles, the Sims site, the Discovery site, and the Peyregne site. The two sites described above have also received some archaeological attention over the years. McIntire visited and collected from both sites, as well as the Grand Temple, in the early 1950s. Neuman also went to the sites and made collections. Rebecca Saunders, while regional archaeologist for that area, visited the sites in the company of Forrest Traverica and also collected. Marco Giardino, now with NASA, and Michael Commardelle visited the sites at various times. Prior to 2001, however, the most intensive work at any of the sites was Traverica’s 1976–1977 excavations at 16SC14. Aided by a chapter of the Louisiana Archaeological Society and an Explorer Scout troop, Traverica excavated three test units and supervised an admirable analysis of the artifacts recovered. His collections and data are at the Louisiana State University Museum of Natural Science, along with his analyses, though a complete draft of his report could not be located (Traverica 1977). This, then, was the situation in 2001, when I was employed by St. Charles Parish to conduct a survey in connection with a shoreline protection project (Shuman 2002). The survey led

100

m a lcolm k. shum a n

to a second project in 2002, whereby test excavations were conducted at both sites (Shuman and Shuman 2003). The data presented herein derive from that project. A third site, 16SC8, was also tested during the project, but no data will be presented from 16SC8 because of the shallowness of the two test units and the extreme disturbance in portions of that site. It should be noted, however, that the 16SC8 data do not contradict the other data presented in this chapter.

Phases of Coles Creek, Plaquemine, and Mississippian The contemporary archaeological view is that in the Barataria area of Louisiana, Coles Creek culture (a.d. 800–1000) led into a transitional Coles Creek phase (a.d. 1000–1200) that, in turn, split off into two successor phases. In the western part of the Coastal Zone, it developed into a distinct Plaqueminetype culture referred to as the Barataria phase, distinct from the “Classic” Plaquemine farther to the north. In the eastern part of this area, it evolved into a synthetic Mississippian culture, represented by the Bayou Petre phase (Miller et al. 2000:343; Phillips 1970; Weinstein 1987a). Miller et al. (2000:347) have suggested dates for the Barataria phase as circa a.d. 1200–1450. They cite radiocarbon dates from 16JE218 that range from a.d. 1300 to 1450 and “limited radiocarbon dates” from other sites in the region ranging from a.d. 1200 to 1500 (Miller et al. 2000:345). They de¤ne the phase through the presence of Plaquemine ceramics, though there is an almost total absence of Plaquemine Brushed. On the other hand, Southern Cult motifs are present. The signi¤cant types they and others associate with the phase are Avoyelles Punctated, var. Dupree, L’Eau Noire Incised, vars. L’Eau Noire and Bayou Bourbe, Carter Engraved, Maddox Engraved, and Mazique Incised, var. Manchac (Miller et al. 2000:344). Buras Incised and Coles Creek Incised, var. Hardy are also apparently markers, though Miller and colleagues note that Holley and DeMarcay (1977) do not include them in their discussion of Plaquemine sites in the area. The Mississippian culture Bayou Petre phase, according to Kniffen (1936), contains a “high percentage of shell-tempered sherds” and is concentrated in northernmost St. Bernard Parish. Kniffen noted a similarity between Bayou Petre sites and sites in Mobile Bay on the Gulf coast to the east. Indeed, Blitz and Mann (2000:55) describe the use of shell tempering as being a Pinola phase (a.d. 1200–1350) innovation on the eastern Gulf coast and they relate this to Mississippian in®uence from the interior; it is clear, then, that there was considerable interaction among coastal peoples at this time (though, for a slightly different perspective, see Greenwell 1984:151). Until that time on the

north w est l a k e sa lva dor, st. ch arles parish

101

eastern coast, “most of the grog-tempered pottery consists of late Coles Creek/ early Plaquemine series type-varieties” (Blitz and Mann 2000:55). The divider between the western zone, with its dominant Barataria phase, and the eastern zone, with its dominant Bayou Petre manifestation, has been identi¤ed as the Barataria Basin (Miller et al. 2000:346–348). In the eastern zone, the Barataria phase is succeeded by the Bayou Petre phase: “Both of these phases are associated with the local Plaquemine culture, but the later phase shows strong in®uence from the east” (Miller et al. 2000:348). In the western zone, the late Plaquemine successor to the Barataria phase is “probably represented by the Delta Natchezan phase” (Miller et al. 2000:348). The demarcation between these two zones would pass very near the two sites considered herein. Thus, it should not be surprising that 16SC4 and 16SC14, during the transitional Coles Creek phase, seem to show characteristics of both zones. What is more interesting is what appears to be the early, extensive use of shell tempering. All the Lake Salvador sites have higher frequencies of shell than Sims (16SC2) (Davis and Giardino 1980). Bayou des Familles (16JE218) produced just three shell-tempered sherds, all evidently from the same vessel (Kidder 1995). At the Discovery site (16LF66), only one shell-tempered type, Pensacola Incised, is listed in the ceramic descriptions; the rest are grog tempered (Miller et al. 2000). Finally, the test units at the Pump Canal site (16SC27) yielded only three shell-tempered sherds out of a total of 3,359 (Giardino 1994). Winterville Incised, largely absent at the Sims, Pump Canal, Bayou des Familles, and Discovery sites, is present at both Lake Salvador sites. Mound Place Incised is also present as well as a small but signi¤cant amount of Moundville Incised. Interestingly, the only sherd of Maddox Engraved found during the current project was shell tempered and so was one sherd of Addis, attesting to the inclination of the inhabitants of these sites to use shell as much as possible.

Description of Excavations and Radiocarbon Dates 16SC4 (Bois Chactas) Five 1-by-1-m test units were excavated at 16SC4, the largest of the project sites. All units were as close as possible to the beach ridge (i.e., on the extreme south side of the site), approximately equidistant from one another. Of these units, four were productive and one showed disturbance from late historic activity. Test Unit No. 2 extended to a depth of 94 cm to the water table, at which point the cultural deposits had played out. The most complex of all the units excavated, this unit revealed an oval feature of burned shell at 20 cm

102

m a lcolm k. shum a n

below datum (bd) that extended to a depth of 40 cm bd. Included in the feature were animal bones and sherds. Extending from this feature on its east side was a crumbly matrix of solidi¤ed shell and ash. At 50 cm the shell ¤ll of the unit became dense, and at about 60 cm the shell became compacted. At about 80 cm was a sand lens 5 cm thick, sloping downward from east to west. Underlying this lens was a layer of loose Rangia shell. The unit produced 66 prehistoric sherds. Prehistoric ceramics were concentrated in the upper 50 cm of the unit. The single most frequent ceramic type found was Baytown Plain, var. Cataouatche, but there were also single examples of Coles Creek Incised, var. Hardy and Mazique Incised, var. North. There was also a sizable collection of shell-tempered material, which consisted largely of plainware, which I have typed as Bell Plain, Mississippi Plain, and Pinola Plain, the latter being a term I use (with apologies to Blitz and Mann) for a coarse grog–tempered ware identical to Baytown Plain, var. Cataouatche, except that it also includes shell. Decorated shell-tempered ceramics from this unit include Cracker Road Incised, Mound Place Incised, var. unspeci¤ed, and Moundville Incised, var. Singing River. The faunal complement of the unit consisted primarily of ¤sh bone (352 g), with small amounts of mammal (15 g) and reptile (9 g). The most frequent mammal identi¤able by species was the muskrat. This unit yielded three radiocarbon dates. A sample of charcoal from Feature A, at the 20–30 cm level, produced 2-sigma date ranges of cal a.d. 1050– 1095 and cal a.d. 1140–1300. A second date was obtained by accelerator mass spectrometry (AMS) from a fragment of charcoal associated with Feature B at a depth of 30–40 cm bd. The 2-sigma range for this date was a.d. 795–1000. A third date, from ¤ll charcoal, came from 94 cm bd and produced a 2-sigma range of cal a.d. 885–1020. It must be mentioned that a small fragment of bottle glass was recovered at the 50–60 cm level. The consistency of the radiocarbon dates and the fact that this was the only historic artifact from below 20 cm suggests that the glass fell from higher in the pro¤le and was inadvertently incorporated into the ¤ll when the ¤ll was water screened. The second test unit that was radiocarbon dated was Unit 5, which extended to a depth of 60 cm bd, at which point the water table was encountered. Prehistoric cultural materials did, however, continue below this point. The stratigraphy of this unit consisted of four layers of shell of varying density, with considerably more soil content than in other test units. The 10– 20 cm level was notable for producing oyster shells along with the Rangia shells and small quantities of historic and prehistoric artifacts. At 30 cm bd a stratum consisting of Rangia shell, quantities of animal bone, and prehistoric

north w est l a k e sa lva dor, st. ch arles parish

103

ceramics was encountered. The percentage of animal bones and ceramics increased at the 40–50 and 50–60 cm bd levels, and the Rangia shells became mixed with the shells of a freshwater clam, Quadrula sp. In all, the lower 20 cm of the unit produced 93.5 percent of the unit’s aboriginal ceramic content and 91 percent of the unit’s faunal complement. Two-thirds of the ceramics from this unit were grog tempered. This is the result of the presence of large quantities of Baytown Plain, var. Cataouatche, which made up 58.1 percent of the ceramics from the unit. Decorated grogtempered types included Coles Creek Incised, var. Hardy and Evansville Punctated, var. Braxton. Of the shell-tempered wares, the most frequent types were Bell Plain, Mound Place Incised, var. Walton’s Camp, and Pinola Plain, in that order. There were also two sherds of Moundville Incised, var. Snow’s Bend. In the Walton’s Camp category were four sherds, apparently from the same vessel, with a single external, parallel-incised line and a bright red interior slip. The red slip is not characteristic of var. Walton’s Camp and it would probably be equally justi¤ed to place these sherds in the var. unspeci¤ed category. The faunal array of this unit was composed predominantly of ¤sh, followed closely by mammals. Fully 83 percent of the mammals by weight were from one taxon, muskrat. A sample of muskrat bones from the 50–60 cm level produced an AMS 2-sigma range of a.d. 1040–1260. The other two productive test units yielded similar results, though no radiocarbon dates were obtained from them. In all, 68.8 percent of the ceramics from 16SC4 were grog tempered and 31.2 percent were shell tempered. The calibrated radiocarbon dates span the years from a.d. 795 to 1260.

16SC14 (Tabatiere Perdu) Three test units were excavated at this site. Two, near the beach ridge, were badly contaminated by recent materials, while the third, which extended to a depth of 90 cm bd, produced a large quantity of prehistoric ceramics and animal bones, as well as a radiocarbon date. Three-quarters (75.6 percent) of the aboriginal ceramics from this site were grog, grit, and sand tempered, while only a quarter (24.4 percent) were shell tempered. Again, the largest single variety was the grog-tempered Baytown Plain, var. Cataouatche, followed by the shell-tempered types, Mississippi Plain, Bell Plain, and Pinola Plain, in that order. Also represented were small amounts of Barton Incised, Moundville Incised, var. Snow’s Bend, and Winterville Incised, vars. Winterville and unspeci¤ed. The faunal remains consisted of almost equal amounts of ¤sh and mam-

104

m a lcolm k. shum a n

mal, plus a smaller amount of reptile. Of those mammalian remains that could be speciated, there was only one taxon, Odocoielus virginianus (whitetailed deer). A radiocarbon date from a deer femur taken from the north wall at 58 cm bd produced a 2-sigma range of cal a.d. 1290–1420. The ceramics and radiocarbon date from this unit strongly suggest a Plaquemine occupation. This is bolstered by Traverica’s excavations in 1976–1977. His test units, located atop the distributary ridge, were deeper than the 2002 units because they were located farther from the beach ridge. A tabulation of decorated types indicates that Traverica’s assemblage spans the Coles Creek and Mississippi periods, although nearly two-thirds are Mississippian in age, and almost all the Mississippi period sherds are grog-tempered types.

Summary of 16SC4 and 16SC14 To summarize, the ceramics from 16SC4 were concentrated in layers containing heavy accumulations of muskrat bones, this animal being particularly prevalent in Coles Creek sites in this area (Davis 1984; Misner and Reitz 1994). Four radiocarbon dates span the Coles Creek and transitional Coles Creek periods. On the other hand, 16SC4 produced a higher proportion of shelltempered, Mississippian ceramics than any other site in the immediate area, including Sims, Bayou des Familles, Pump Canal, and Discovery. Judging from the foregoing, the site seems poised to develop into a Bayou Petre phase Mississippian site. At nearby 16SC14 the ceramic assemblage is primarily Plaquemine with some Coles Creek sherds. The predominant mammal is deer, and the single radiocarbon date is Barataria phase Plaquemine. The differences in excavated assemblages, both my own and Traverica’s, might be explained by the greater protection of 16SC14 from storm surges as a result of higher elevation. If 16SC14 is what 16SC4 would have become (and may, in fact, have once been, if we assume that storm surges obliterated the upper strata of 16SC4), then we have a situation in which Coles Creek, heavily in®uenced by Gulf Coast Mississippian, evolves not into Mississippian but into Barataria phase Plaquemine. For if the Barataria phase is de¤ned by the presence of Anna Incised, Avoyelles Punctated, Carter Engraved, Coles Creek Incised, var. Hardy, Harrison Bayou Incised, Southern Cult motifs, and the near absence of Plaquemine Brushed, 16SC14 meets six of the seven criteria.

Questions One of the most intriguing questions raised by this project is the relationship between what I have called Pinola Plain and Kidder’s Baytown Plain, var.

north w est l a k e sa lva dor, st. ch arles parish

105

Cataouatche. The former refers to the grog-and-shell–tempered wares that Blitz and Mann (2000) describe for the Pinola phase (a.d. 1200–1350) of the eastern Gulf coast. The latter indicates the grog-tempered ware that Kidder (1995) found at the Mississippi period Bayou des Familles (16JE218) site. The only difference between these two classes from the Lake Salvador sites is the presence/absence of shell in the paste. If we accept the dating suggested by Blitz, Mann, and Kidder for these two classes of wares, we are faced with the situation of having a shell-tempered ware evolve into a grog-tempered one. Such a view, while not irrational, ascribes what I consider rather too much importance to paste characteristics. It is far more reasonable to view these as one single, contemporaneous group of wares, some of which had shell added (either by accident or design) and some of which did not. If the latter explanation is correct, then the question resolves into the issue of whether Pinola wares occur in the Lake Salvador area later than they do on the eastern Gulf coast or whether it is a matter of the var. Cataouatche wares occurring earlier on Lake Salvador (i.e., beginning in transitional Coles Creek times and lasting until Mississippi times). The available evidence suggests the latter is the case. The radiocarbon dates and faunal evidence for 16SC4 indicate a transitional Coles Creek occupation. If var. Cataouatche is late Mississippian, then we have a case in which the radiocarbon dates and faunal array indicate transitional Coles Creek even though few Coles Creek ceramics occur to support this assignment and in which Pinola phase shell-tempered ceramics are followed chronologically by an identical ceramic sans shell. If, on the other hand, var. Cataouatche is transitional Coles Creek and early Mississippian (a situation that does not exclude its also extending into late Mississippian times), then Pinola Plain is seen as contemporary with var. Cataouatche (i.e., the same ware with shell). This is especially likely if we adopt Davis’s (1981:68) view of some Mississippian designs (e.g., Mound Place Incised) as being simply Coles Creek ceramics with shell added. This view is bolstered by the statistical analysis of plainware versus decorated ware sorting at 16SC4, which suggested that plainwares tended to be sorted as being somewhat later than decorated ones. It is also supported by the dearth of decorated Coles Creek ceramics in the archaeological deposits: if var. Cataouatche is an all-purpose variety with a long lifespan, its high percentage in the ceramic inventory may indicate that it represents both plainware of the transitional Coles Creek and Mississippi periods and also the incised (and thus diagnostic) vessels of the same periods, which tended to have decorations only below the rims.

106

m a lcolm k. shum a n

In short, then, it seems probable that grog-tempered Baytown Plain, var. Cataouatche and grog-and-shell–tempered Pinola Plain are virtually the same ware and that the manufacture of this ware began no later than transitional Coles Creek times. Furthermore, it seems that this ware was used in transitional Coles Creek times as the vehicle for incised pottery as well as plainware. In the northwest Lake Salvador area, certain shell-tempered Mississippian wares based on Coles Creek designs also appear to occur with some frequency in the transitional Coles Creek period. This situation could provide partial validation for Brain’s (1978) suggestion that the Mississippian in®uence began in late Coles Creek times. If so, however, it would appear that this in®uence, at least on the ceramics, withered with time. Returning to Phillips, we must now ask whether, in applying the conventional terminology (transitional Coles Creek, Plaquemine, and Mississippian) we have, in the case of these sites, even a “locally intelligible” situation. If we analyze the data in terms of the conventional taxonomy, the following possibilities are evident: the use at these sites of shell tempering during Transitional Coles Creek and the withering of this use in the succeeding Plaquemine period represents an in®ux of Mississippian ideas/peoples during the former and their supercession by Plaquemine ideas/peoples during the latter. While these possibilities cannot be excluded, they tend to simplify what was in all probability a complex situation. The sites in question are situated at a crossroads of historic commerce (Shuman and Shuman 2003). The Barataria–Lake Salvador corridor was a gateway from the Gulf to the interior in historic times, perhaps most famously used by the pirate Jean La¤tte. In the nineteenth century groups from New Orleans held Sunday outings on 16SC4. It is clear that the same situation pertained prehistorically. Transport would have been from the Gulf to Barataria Bay to Lake Salvador and into Lake Cataouatche, or from Lake Salvador up Bayou des Allemands. In short, the sites described herein were at the nexus of an interaction sphere that linked groups along the east and west Gulf coast with the interior. While there may have been some migration of individuals from one area to another, it makes more sense to me to suggest that the major movement of individuals through this area was for the purpose of commerce and that the practice of shell tempering resulted from a combination of borrowing the concept from the eastern Gulf coast and the ready abundance of shell. The cessation of the use of shell may, indeed, be the result of a stronger in®uence (Plaquemine) from the interior and the west in the post–a.d. 1200 period, as the Mississippian interaction sphere shrank and the tribes as we know them today formed from the disintegration of Moundville (see Galloway 1995).

north w est l a k e sa lva dor, st. ch arles parish

107

These are, however, speculations, and in the end we must return to Phillips’s caution about not attributing too much signi¤cance to paste characteristics. These sites are located in an area of unusually high demographic activity, with frequent comings and goings, as others have long noted (i.e., Brown 1984:122). Sampling error is thus not out of the question. So, there are several possible explanations for the phenomena noted herein and it remains for future investigators to determine which explanation is most viable.

Acknowledgments I am grateful for the assistance of the following persons during the course of the project: Earl Matherne, Roy Madden, Rocky Sexton, and Duke Rivet. Dr. Becky Saunders of the LSU Museum of Natural Science kindly provided access to previous collections and documents. I am also grateful to my wife, Margaret, for assistance during the writing and analysis and to Mark Rees, Patrick Livingood, and the two reviewers, Jay Johnson and Martha Rolingson, for assistance during the preparation of this chapter. R. Hampton Peele, of the Louisiana Geological Survey, created Figure 5.1.

6 Plaquemine Recipes Using Computer-Assisted Petrographic Analysis to Investigate Plaquemine Ceramic Recipes Patrick C. Livingood

Archaeologists working in the Lower Mississippi Valley (LMV) have focused a great deal of energy on identifying, classifying, and explaining the Plaquemine-Mississippian dichotomy (Phillips 1970; Phillips et al. 1951; Williams and Brain 1983). The primary tool in this investigation has been ceramics, and the principal attribute is the presence or absence of shell tempering (see Rees and Livingood, this volume). Despite the axiomatic role that shell tempering plays in identifying the Plaquemine-Mississippian divide, there are many places and times in the Plaquemine world in which a signi¤cant percentage of the ceramic assemblage contains provocative mixtures of shell and grog tempering and in which decorative motifs span temper types. This has led some archaeologists (Hally 1972; Kidder 1998b) to question whether too much interpretive weight has been given to the presence and absence of shell tempering. One of the challenges faced by archaeologists attempting to study the Plaquemine phases in which shell and grog are both used as tempering agents is that it is not clear that the varieties speci¤ed for distinguishing ceramic fabric correspond well to real cultural and technological distinctions made by the potters. Part of the problem is that the type-variety system developed by Ford, Phillips, Williams, and Brain for the LMV is based on macroscopically observed criteria, whereas potters often ground the tempering agents into very small sizes and sometimes mixed tempering agents together in ways that make macroscopic classi¤cation challenging and subjective. A cursory glance at the intellectual history of the sorting criteria for plainwares in the Lower Yazoo and Natchez regions reveals how murky these classi¤cations can be. Addis Plain was originally documented by Quimby (1942:265–266, 1951: 107–109) as a “clay-tempered type” and a major diagnostic of Plaquemine culture. Phillips (1970:48–49) designated Addis as a variety of the type Bay-

pl aquemine cer a mic recipes

109

town Plain and de¤ned it as the “clay-tempered plainware of the Mississippi period from the Medora and Plaquemine phases in the Delta and Lower Red River regions to the Mayersville phase in [the Lower Yazoo].” Interestingly, Phillips (1970:60–61) also acknowledged the similarity between his Baytown Plain, var. Addis and his Bell Plain, vars. Holly Bluff and St. Catherine, which contain ¤nely pulverized shell sometimes in quantities so small that their inclusion seems to be “accidental.” Williams and Brain (1983:92) retained Addis as a variety of Baytown Plain but used Bell Plain, var. Greenville to describe Addis with the addition of shell temper to the paste. They also retained Bell Plain, var. Holly Bluff to describe heterogeneous Addis-like sherds in which shell tempering is slightly more prevalent than in Bell Plain, var. Greenville. Steponaitis (1974:116) proposed elevating Addis to the level of type and de¤ned it as having a heterogeneous organic grog-tempered paste but allowed for the presence of shell in some types. Steponaitis then relocated the Greenville and St. Catherine varieties from their position under Bell Plain to be varieties under Addis Plain. The elevation of Addis to the level of type has been made by some archaeologists, particularly those who have worked in the Natchez Bluffs (e.g., Brain 1989; Brain et al. 1994; Brown 1985a) but has not been embraced by others (e.g., Kidder 1993a; Ryan 2004). More recently, Ryan (2004:94) proposes keeping Addis as a variety of Baytown Plain for the analysis of Hedgeland ceramics because of the close relationship between Addis and Baytown Plain, var. Little Tiger. Little Tiger is proposed as a transitional ware between earlier Baytown pastes and Addis and is differentiated from Addis because it has less temper. There are two fundamental explanations for the tortured history of these types. First, Phillips (1970) and later Williams and Brain (1983) fundamentally viewed the Plaquemine-Mississippian phenomenon as a cultural divide and it was necessary to de¤ne phases as either belonging to Mississippian or Plaquemine ( Jeter and Williams 1989:212; Rees and Livingood, this volume). The primary tool in this enterprise was ceramic typology and in the hierarchical system of type-variety classi¤cation the presence or absence of shell was a ¤rst-order attribute. This is entirely understandable from a macroregional perspective, and it explains why Phillips would place the Addis variety under Baytown Plain and the closely related Holly Bluff and St. Catherine varieties under Bell Plain. However, for archaeologists working in phases in which these closely related types are common, this system of nomenclature can be unwieldy. Another explanation for the dif¤culties archaeologists have had in de¤ning and arranging these taxa is that in the phases in which potters were freely choosing between shell and grog temper, a larger number of permutations is possible; designing a system to accurately describe these permutations

110

patrick c. li v ingood

is problematic. Support for this argument can be found from merely counting the number of plainware varieties in use in the LMV. Despite large quantities of coarse shell–tempered ceramics, only three varieties of Mississippi Plain are in common use in the Plaquemine world: Coker, Mainfort, and Yazoo. However, there are eight varieties commonly used to categorize ¤ne-sized-grog– tempered and grog- and shell-tempered fabrics in the Plaquemine period: Addis Plain, vars. Addis, Greenville, Junkin, Ratcliffe, and St. Catherine, and Bell Plain, vars. Bell, Holly Bluff, and New Madrid. Michael Galaty (2006) has used a ceramic ecology approach to argue that on a regional scale differences in tempering choices in Mississippi might be related to differences in the types of clays and tempering agents available to potters. For example, he argues persuasively that pre-Mississippian period potters in west Mississippi may have lacked access to sand suitable for tempering and relied instead on grog, whereas potters in east Mississippi may have preferred the easily available sand. Although he does not speci¤cally address the regional differences in the use of grog and shell, his work raises the possibility that macroregional differences in temper might be related to ecology, not culture. However, within a site the variation being re®ected in the varieties of Addis Plain and other closely related types originates from the choices made by potters during clay preparation. It is possible that the potters could have been haphazardly adding grog and shell to certain vessels, which would lead to a random-looking distribution of temper frequency. However, it is much more likely that the Plaquemine potters were in fact very careful about clay preparation and the addition of tempering agents, like most ethnographically studied potters (Arnold 1985; Krause 1985). Potters often clean their clays to remove extraneous particles and then add carefully prepared tempering agents. Some potters mix temper with the prepared clays until the paste reaches a desired texture. For example, the Ibibio of Nigeria add grog or sand to the clay until it reaches the correct consistency called aduang nbibiot (Nicklin 1981:173; Rice 1987:121). Other potters follow a speci¤c recipe that designates the ratio of clay and temper (Rice 1987:121). For example, some Kavango potters in southern Africa mix two parts grog to three parts clay (Blandino 1997:26), whereas the Shipibo-Conibo of eastern Peru have a ratio of clay to temper of two to three (DeBoer and Lathrap 1979). The goal of this study is to test a small assemblage of Plaquemine ceramics to see whether it is possible to detect modes in the distribution of temper size and abundance. If there are modes, then these can be used to reconstruct the paste recipes used by the potters. Armed with knowledge of the paste recipes, we can evaluate the utility of the commonly used plainware varieties for class-

pl aquemine cer a mic recipes

111

ifying the ceramic assemblage. Furthermore, it seems likely that of all of the decisions a potter makes, that of temper is probably one of the most resistant to change over time. Vessel form and decoration can change over decades in pre-state societies, but temper choices typically persist for centuries. In most pre-state societies, pottery production is a household activity (Arnold 1985:100– 101; Sinopoli 1991:98–102; van der Leeuw 1977), and it is presumed that knowledge about pottery production is handed down through generations within the household. Whereas more visible aspects of pottery production such as vessel form and decoration might be subject to changing personal or group concepts of pottery construction, less visible and more technological decisions about temper are likely to be more resistant to change. Because of this, with suf¤cient information about paste recipes it might be possible to tease apart different communities of production, especially when multiple tempering agents encode greater information. Since the terms used in ceramic studies such as paste and temper are often de¤ned differently by different authors, it is important to be clear how they are used in this chapter. Following the connotations common to the discussion of LMV ceramics (which differ somewhat from the de¤nitions common to petrographic literature [e.g., Stoltman 1991:109–110]), temper refers exclusively to material intentionally added to a clay to improve its physical properties. Inclusion refers to all aplastics in a clay, whether they were added deliberately or were naturally occurring. Paste is de¤ned as the clay plus all inclusions. Therefore, a paste recipe refers to all of the rules a potter follows to create the paste used to form a vessel, including where the clay should be gathered, how it should be processed, what temper should be added and in what quantities, and how the clay should be handled and treated before and during vessel creation.

The Sample The sample under study comprises 29 sherds from the Pevey (22LW510) and Lowe-Steen (22LW511) sites located on the central Pearl River in Lawrence County, Mississippi. The Pevey site is a large mound site, with nine extant ®at-top mounds, located approximately 18 km south of the two-mound LoweSteen site. Both sites date primarily to the Winstead phase (Livingood 1999), which is temporally equivalent to the Anna or Winterville phases, and the Pevey site also has a small Pevey phase occupation, which is temporally equivalent to the early Foster or Lake George phases. The Pevey site was ¤rst investigated by Baxter Mann and John Blitz between 1982 and 1984 (Mann 1988). More extensive excavations were carried out at both sites by the University of North Carolina Field Schools in 1993

112

patrick c. li v ingood

and 1994 (Livingood 1999), including test units in every mound. In 2000, I returned with volunteers primarily from the University of Michigan to test the plaza area to the west and north of the Pevey site. The two most common decorated varieties found at the Pevey and LoweSteen sites are Anna Incised, var. Anna and Plaquemine Brushed, var. Plaquemine. Other common decorated types include Carter Engraved, var. Carter, Grace Brushed, var. Grace, D’Olive Incised, var. D’Olive, L’Eau Noire Incised, var. L’Eau Noire, Leland Incised, Mound Place Incised, and Parkin Punctated. Most of these types are most commonly found at sites to the west of the Pearl River along the Mississippi River, while other types, such as Mound Place Incised and D’Olive Incised, are more commonly associated with sites to the east such as Bottle Creek or Moundville. Several of the most abundant types represent decorative traditions that span multiple tempers. For example, Anna Incised, var. Anna is the type assigned to sherds from shallow bowls or plates with interior decoration if the temper is categorized as being Addis Plain, var. Addis or Addis Plain, var. Greenville. But if the exact same vessel were tempered more heavily with shell, it would be classi¤ed as D’Olive Incised, var. D’Olive. Likewise, jars with exterior brushing on Addis or Greenville pastes are called Plaquemine Brushed, var. Plaquemine, while jars with shell tempering are called Grace Brushed, var. Grace. The initial analysis and classi¤cation of ceramics indicates that middle Pearl potters were making interesting and complex choices with regard to temper. The most common fabric used at the Pevey and Lowe-Steen sites is coarse shell tempering (72 percent of all sherds). However, Pearl River potters assigned special importance to Addis paste sherds because they executed most of their decorative motifs on these wares. Fifty-two percent of all decorated sherds have Addis Plain, var. Addis paste, 17 percent of decorated sherds have a Greenville paste, and only 27 percent have a Mississippi Plain paste. Despite the preference for Addis pastes, several decorative techniques crosscut temper types. If we consider all of the interior decorated plates and bowls (Anna Incised and D’Olive Incised) together, 60 percent are executed on Addis paste, 25 percent on Greenville paste, and 14 percent on Mississippi Plain paste. On the basis of these initial observations, it is clear that middle Pearl potters were comfortable using a wide variety of temper combinations. Additionally, it is not obvious whether the plainware varieties developed primarily for the Lower Yazoo and Natchez regions are the most appropriate varieties to classify the ceramics from the middle Pearl River. All of the 29 sherds selected for this analysis are diagnostic of vessel shape, decoration, or both. The sherds were deliberately chosen to represent the dif-

pl aquemine cer a mic recipes

113

ferent temper combinations under investigation and to be representative of different vessel forms and functions. Of the 29 sherds in this analysis, 24 were excavated from a single 2-×-2-m unit at the Pevey site designated Unit M. This unit is located approximately 25 m northwest from the base of the site’s large mound, Mound E, in a small rise on the edge of the natural terrace. Only four of the sherds from Unit M come from the upper level, which probably has an early Pevey phase assignment while the remaining 20 sherds have a Winstead phase date. Of the sherds not from Unit M, two come from the Mound H excavation and one from the Mound I excavation at the Pevey site, while the ¤nal two come from the Feature 1 excavation at the Lowe-Steen site. All have a Winstead phase date.

Methodology The only technique appropriate to gather data on temper abundance and size from this sample is ceramic petrography, which is the practice of examining ceramics microscopically to study the clay characteristics and inclusions. Hitech chemically based approaches such as x-ray diffraction or neutron activation can provide only a partial picture since they are unable to chemically distinguish grog from the clay matrix. Ceramic petrography has been a part of American archaeology since the days of Anna Shepard (Shepard 1976), but most modern analysis owes much to Jim Stoltman, who systematized the use of point-counting techniques borrowed from geological petrography in order to bring a higher level of rigor and accuracy to the ¤eld (Stoltman 1989, 1991, 2000). Today, most ceramic petrographers use a point-counting technique to quantify inclusions, which involves overlaying the sample with a grid of points in order to obtain representative counts of constituent particles. This technique is excellent at measuring the abundance of constituent particles, and it remains the gold standard for measuring petrographic data (Cordell and Livingood 2004). For this study, I have employed computer-assisted petrographic analysis (CAPA) (Livingood 2002, 2004; Velde and Druc 1998). This procedure starts with a digital image of the thin section and uses digital image analysis software to help produce a map of the section identifying the constituent particles. For some particle types, the software can do most of the work. It can be scripted to automatically identify a class of particles with a high degree of accuracy and precision. In other cases, a human operator is required to map the particle types, but the software can help by producing false-color images that make identifying the particles much easier. CAPA has a few bene¤ts over traditional microscope-based petrographic

114

patrick c. li v ingood

analysis (Cordell and Livingood 2004). First, it is less expensive under some circumstances, since an inexpensive consumer-quality ®atbed scanner has suf¤cient resolution to identify temper particles. Second, because a complete map of the thin section is produced, far more information is produced. Since every particle is individually identi¤ed and measured, every possible metric related to particle count, size, shape, orientation, and location can be generated. Third, under some circumstances, this procedure can be faster than manual point-counting techniques. This is especially true if the samples are relatively homogenous and the features of interest are distinct. Fourth, the digital nature of the analysis makes it much easier to revise and correct analyses and to share results. For this study each thin section was scanned at 3,200 × 1,600 dpi using an Epson Perfection 1640 scanner with a transparency adapter and polarizing ¤lm. Two scans were produced from each thin section: the ¤rst scan was produced with plane-polarized light and the second with cross-polarized light (Figures 6.1–6.2). Next, the images were aligned as layers within Adobe Photoshop. Software from Reindeer Graphics called Image Analysis Toolkit (Russ 1999) was used to create derivative images from these two layers by manipulating the information in their color channels (Figure 6.3). If the image can be manipulated in such a way that the desired features are distinguished by color, intensity, or texture, it is possible to automate the process of identifying the pixels corresponding to the features. Under cross-polarized light certain crystals appear to have unusual or bright colors because they split the light into two rays with different refraction indices. This property is referred to as birefringence, and these particles are very easy to identify with the software. In general, the identi¤cation of birefringent particles and voids was almost entirely automated, the automatic identi¤cation of shell was fairly accurate but required some editing, and the identi¤cation of grog was primarily done by hand. The end result of each identi¤cation is a series of Boolean images for every type of feature of interest (Figure 6.4). Every pixel in a Boolean image is either black, indicating it is a part of the feature, or white, indicating it is not. A function in the Image Analysis Toolkit produces measurements of the features in the Boolean images for analysis in a spreadsheet or statistical analysis software package. At a scanning resolution of 1,600 dpi there are approximately 63 pixels per millimeter in the ¤nished scan. Based on the Wentworth scale (Rice 1987:38), silt particles would appear to be 0.2 to 3.9 pixels wide, very ¤ne sand would appear to be 3.9 to 7.9 pixels wide, ¤ne sand 7.9 to 15.7 pixels wide, and medium sand 15.7 to 31.5 pixels wide. Obviously, larger particles are easier to identify and map precisely. However, there are no easy rules to determine the

Figure 6.1. Plane-polarized scan of PRP27.

Figure 6.2. Cross-polarized scan of PRP27.

Figure 6.3. Example of a false-color enhanced image of the cross-polarized scan for PRP27.

Figure 6.4. Boolean image of the shell and shell void identi¤cations for PRP27.

116

patrick c. li v ingood

minimum size that a feature must be scanned at in order to accurately distinguish it. This size depends on the degree of visual contrast between the particle and the surrounding matrix. However, tests de¤nitively proved that 3,200 × 1,600 dpi is not suf¤cient to accurately identify particles the size of very ¤ne and ¤ne sand (Cordell and Livingood 2004). Since measuring particles of this size is not crucial to the research goals of this project, all measurements of birefringent particles with an area less than 0.2 mm2 were discarded from consideration in this study. Therefore, all measurements of birefringent particles in this report pertain only to larger particles. The majority of birefringent particles in these samples are sand, which appears to be a natural inclusion. However, traditional petrographic analysis by Ann Cordell (2004) also found in some of the samples a small number of naturally occurring constituents such as muscovite mica that also have high birefringent values. Regardless, because they are rare and because they made little difference to the study goals, no effort was made to differentiate different particles with high birefringent values. Furthermore, in order to accurately measure all of the birefringent particles in the sample, it is typically necessary to take two cross-polarized scans, with the sample rotated 90 degrees between scans, and together these will identify all of the birefringent particles. However, since the orientation of birefringent particles is assumed to be random, it is possible to estimate their total abundance by doubling the area measured from a single cross-polarized scan. This produces a reasonably accurate estimate, which is ¤ne for the research goals of this study. Therefore, the values reported in Table 6.1 to measure the abundance of birefringent particles in the sample are most accurately called an estimate of large birefringent particle abundance (ELBPA) and have been calculated by doubling the sum of the area of all birefringent particles at least 0.2 mm2 in area identi¤ed in the analysis of a single cross-polarized image. This estimate avoids making any claims about the abundance of birefringent particles of a smaller size. It can be effectively interpreted as a proxy for the abundance of medium to large sand particles or, more precisely, an estimate of the sum of the surface area of all sand with particle size greater than 0.2 mm2. In an earlier study, four samples were measured using CAPA and using traditional microscope-based petrographic point-counting (Cordell and Livingood 2004). Once the problems of identifying smaller birefringent particles using the scanning resolution of 3,200 × 1,600 dpi were controlled for, CAPA was deemed suf¤ciently accurate to proceed with additional petrographic analyses. In the test, there was one sample for which the grog measurements were signi¤cantly different. The problem was that I inaccurately identi¤ed

pl aquemine cer a mic recipes

117

some hematitic, ferric concretions, lumps, or stains as grog. Although I was able to ¤x the problem, it underscored the dif¤culty of distinguishing grog from other natural stains even when using traditional microscope-based analysis (Di Caprio and Vaughn 1993). Great care has been taken in this analysis to try to ensure that the category of grog measured in Table 6.1 only includes recycled pieces of pottery that were intentionally added as temper. However, these classi¤cations can sometimes be dif¤cult and I expect the range of error to be greater for this temper identi¤cation than for shell, voids, or ELBPA. Following the standards of petrographic point counting (Stoltman 1989, 1991), Table 6.1 reports the size of each thin section and the percentage of nontemper voids inside each sample. The percent abundance of each temper type is the ratio of temper area to matrix area, not including voids. Voids in the sample from leached shell are counted as shell temper.

Analysis Before presenting any results, it is important to mention a few caveats. First, there is the standard warning that this is a small sample size and any patterns that are discovered must be considered merely suggestive. Second, there is no expectation that any conclusions drawn from these Pearl River ceramics will generalize to the rest of the Plaquemine area or anywhere else. In fact, there are strong reasons to suspect that the Pearl River assemblage might be a unique re®ection of the social, historical, and ecological needs of the middle Pearl community. Figure 6.5 shows the biplot of the percentage of shell and grog with each sample coded by the original plainware variety classi¤cation. At ¤rst glance, a few important observations can be made. First, there is a single outlier that is a heavily shell-tempered sherd. Second, the graph has a general L shape. Eight of the samples have a relatively high abundance of shell (>10 percent) and low abundance of grog (<6 percent). Another eight of the samples have a relatively high abundance of grog (>8 percent) and a low abundance of shell (<1.2 percent). The remaining 13 samples have relatively small amounts of both grog (<8 percent) and shell (<10 percent). Another observation is that there is limited ¤t between the macroscopically observed temper and the microscopically measured categories. While the most heavily shell-tempered sherds were all correctly identi¤ed as Mississippi Plain and the most heavily grog-tempered sherds were classi¤ed as Addis Plain, var. Addis, there is a level of murkiness in the middle. Some sherds classi¤ed as Mississippi Plain appear to be nearly indistinguishable using abundance measurements from some Bell Plain and Addis, var. Greenville sherds. Also, some sherds classi¤ed as Addis

118

patrick c. li v ingood

Plain, var. Greenville are quantitatively similar to sherds classi¤ed as Addis Plain, var. Addis. Clearly the human eye is only moderately successful at assessing temper abundance at these scales when the edges of the categories are so close. This is not an unexpected ¤nding. In a summary of petrographic studies of Mississippi ceramics, Galaty (2006) found that almost every study noted a lack of correspondence between microscopic descriptions of fabric

pl aquemine cer a mic recipes

119

and ceramic types that were designed to describe macroscopic attributes such as decoration and form. Other than the abundantly grog-tempered and abundantly shell-tempered samples, it is not immediately clear whether there are any modes present in the data. It is entirely possible that if we increased the sample size we would see a continuous distribution of values and that the reason there are problems

120

patrick c. li v ingood

Figure 6.5. Biplot of grog and shell percentages. The shapes of the points correspond to the original temper assignments.

in applying the type-variety system to classifying fabric is that we are trying to apply discrete categories to a continuum. It is also possible that an increase in sample size would help to bring into focus the modes present in the data. Since the middle Pearl River potters took so much care with so many observable aspects of pottery manufacture such as ¤nely grinding temper particles, generally practicing careful incising and engraving techniques, and frequently polishing the ¤nished vessels, it is probably safe to assume that they were care-

pl aquemine cer a mic recipes

121

ful in their clay preparation and followed some form of paste recipe. If this is the case, here are the modes suggested by the data (Figure 6.6). The cluster boundaries described below were constructed by considering the results of various cluster analyses, visual inspection of the graphs, a consideration of homogeneity of vessel form and type within clusters, and the analytical utility of proposed cluster de¤nitions. Temper size was also used to construct these clusters, but abundance was found to be a much more useful metric. Generally, sherds with smaller temper sizes also had smaller abundances, while sherds with larger temper particles had higher abundances. Temper size was not as useful as a distinguishing variable for this sample, likely because the sherds were largely constructed during a relatively short period of time. In contrast, Ryan’s (2004:93) microscopic analysis of sherds from the Hedgeland site found that grog size was crucially important for differentiating between Percy Creek, Little Tiger, and Addis varieties. Finally, birefringent particles, which mostly represent quartz sand, are absent from the cluster descriptions because there is little correlation between ELBPA and any of the other major variables and all of the observed birefringent particles seemed to be rounded grains (Rice 1987:410), which suggests that the sand is a naturally occurring aplastic inclusion.

Cluster 1 If we exclude sample 23, which is a clear outlier, the ¤rst set of clusters contains sherds with a large quantity of shell and little or no grog. All of the sherds in Clusters 1A, 1B, and 1C have ratios of shell to grog between 5 to 1 and 10 to 1 and could be combined on this basis alone. However, since they also have signi¤cantly different amounts of shell and grog and tend to cluster according to vessel type, they are being subdivided as indicated. Cluster 1A. The sherds in this cluster have shell abundance between 10 percent and 18 percent and grog abundance less than 2 percent. This cluster contains four sherds of Mississippi Plain. Cluster 1B. This cluster is de¤ned as having a high abundance of shell (15 percent to 22 percent) and moderate amounts of grog (3 percent to 6 percent). The sherds in this sample include one piece of Mississippi Plain, a Grace Brushed, var. Grace sherd, and a D’Olive Incised, var. D’Olive vessel. When considering the total amount of temper added to the vessel (grog plus shell), this cluster contains the greatest amount of total temper of any of the clusters. Cluster 1C. The two sherds in this sample are both Grace Brushed, var. Grace sherds and have between 3 percent and 7 percent shell and almost no grog.

Figure 6.6. Biplot of grog and shell percentages showing clusters determined by analysis of the petrographic data.

pl aquemine cer a mic recipes

123

Cluster 2 This cluster is de¤ned as having a shell-to-grog ratio between 1 or 2 to 1. This cluster has abundances of grog between 3 percent and 5 percent and of shell between 5 percent and 8 percent. All of the sherds in this sample are highly decorated serving vessels, including one sherd of Anna Incised, var. Anna, two sherds of D’Olive Incised, var. D’Olive, and one sherd of Carter Engraved, var. Carter.

Cluster 3 This cluster contains three sherds that might have been called “untempered” in the nomenclature of Ford or Quimby. All three sherds in this cluster contain less than 2 percent shell and grog and all three sherds come from Anna Incised, var. Anna vessels.

Cluster 4 This cluster contains four sherds with moderate amounts of grog tempering (3 percent to 6 percent) and little shell tempering (<2 percent). The vessels in this cluster include examples of Bell Plain, Mound Place Incised, var. unspeci¤ed, Plaquemine Brushed, var. Plaquemine, and L’Eau Noire Incised, var. L’Eau Noire.

Cluster 5 This cluster contains all of the highly grog-tempered sherds. With a larger sample size it might be possible to subdivide this cluster, but for now it is left undifferentiated. The sherds in this cluster have between 8 percent and 21 percent grog and less than 1 percent shell. This cluster includes four examples of Plaquemine Brushed, var. Plaquemine, two examples of Anna Incised, var. Anna, one sherd of Addis Plain, var. Addis, and one sherd of Mazique Incised, var. unspeci¤ed. If we assume that these clusters are at least partially correct, what is the signi¤cance? First, these clay types correspond fairly well to variations in vessel function, which suggests that there may have been widely shared notions that different ratios of grog, shell, and clay were appropriate for different applications. The clays in clusters 2, 3, and 4 were used almost exclusively for serving vessels (plates and shallow bowls and decorated varieties such as Anna Incised, Carter Engraved, L’Eau Noire Incised, Mound Place Incised, D’Olive Incised) and the clays in clusters 1A and 1C were used exclusively for utilitarian vessels.

124

patrick c. li v ingood

The clays in clusters 1B, 4, and 5 were used for mixed purposes. Note that there is not a good correspondence between decorative type and cluster. For example, there were six Anna Incised, var. Anna sherds in the sample that were constructed using temper combinations from three clusters. Likewise, the ¤ve sherds of Plaquemine Brushed, var. Plaquemine span two clusters, the three sherds of Grace Brushed, var. Grace span two clusters, and the three sherds of D’Olive Incised, var. D’Olive span two clusters. Second, this study found a fairly strong association between high total temper abundance (grog plus shell) and utilitarian tasks while the majority of presumed serving wares have low total temper abundance. This may be explained functionally: coarsely tempered vessels may perform better in cooking and storage tasks. It may also be explained aesthetically: Pearl River potters seemed to prefer a more uniform, polished, “temperless” look for serving ware vessels. Finally, the data suggest that there were a greater number of temper combinations being used than were described by the original fabric types borrowed from the LMV. Furthermore, the clusters do not map very well to these existing plainware varieties, as shown in Table 6.2. This might indicate that there were different tempering practices being used in the middle Pearl because of cultural or ecological differences or it might indicate the expected lack of correspondence between macroscopic and microscopic categories. In trying to apply these established types, it is interesting to observe that very few samples are completely lacking in grog or shell and there appears to be no real analytical difference between samples that contain trace amounts and those for which a temper is absent. Therefore, the distinction between Addis Plain, var. Addis and Addis Plain, var. Greenville does not seem very important for this assemblage. However, the ratio between grog and shell and the relative abundance of both do seem to be important. Sherds with lots of shell or lots of grog get assigned to clusters 1 or 5, respectively. The remaining sherds get classi¤ed according to whether they are almost “temperless” (cluster 3) or contain slightly more shell than grog (cluster 1C), slightly more grog than shell (cluster 4), or about equal amounts of both (cluster 2).

Conclusion These results provide empirical evidence that the frustrated efforts to classify the Pearl River ceramics using the four paste categories of Addis Plain, var. Addis, Addis Plain, var. Greenville, Mississippi Plain, and Bell Plain were like trying to ¤t a square peg into a round hole. In reality, the categories I should be using rely more on observing the ratio between grog and shell temper.

pl aquemine cer a mic recipes

125

Like many studies, this one raises more questions than it answers. Many of the questions can probably be answered by increasing the sample size. For example, it would be interesting to see whether an expanded sample from the Pearl River would sharpen the focus of the plots and reveal several closely related temper recipes or whether an expanded sample would blur the rela-

126

patrick c. li v ingood

tionships and indicate that grog and shell were added haphazardly. It would also be interesting to apply these analyses to other regions in the Plaquemine world where grog and shell tempers are frequently mixed to ¤nd out whether potters there were making choices similar to those of the Pearl River potters.

Acknowledgments I would like to thank Ann Cordell, Michael Galaty, Chris Glew, Susannah Livingood, Mark Rees, Vin Steponaitis, and the reviewers for providing comments and feedback on various incarnations of this chapter.

7 Feasting on the Bluffs Anna Site Excavations in the Natchez Bluffs of Mississippi Virgil Roy Beasley III

Recent years have seen an explosion of archaeological theorizing regarding the nature and role of feasting in various cultures and societies (Dietler and Hayden 2001). Much of this literature has focused on obligation, power, and display, often as a correlate of commensal politics, chie®y in®uence, and dichotomies of superior/inferior (Blitz 1993; Costin and Earle 1989; Dietler 2001; Hayden 2001; Jackson and Scott 1995; Joffe 1998; Kelly 2001; Knight 2001; Kolb 1994; Pauketat et al. 2002; Perodie 2001; Phillips and Sebastian 2004; Smith and Williams 1994; VanDerwarker 1999; Welch and Scarry 1995). More recently, attention to the relationship between feasting behavior and identity has become an element of this discourse. Feasts provide an ideal arena for expressions of identity, being a locus of symbolically charged items such as food and ceramics (Weismantel 1988:7) and due to the ability of hosts to select those persons who are allowed to attend certain feasts. The intersection between material culture and practice associated with feasting behavior provides an opportunity for archaeologists to directly explore an essential dynamic of social interaction. Feasting occurs in diverse social contexts and the material signatures of feasts are recognizable archaeologically. Many of these contexts indicate power asymmetries, overt and subtle. However, apical relations are not a necessary condition of feasting behavior. Visibility is not a compulsory feature of hierarchical structures. Feasting occurs in a variety of political and economic organizations. It is this diversity in the kinds and contexts of feasting behavior that allows for varied and rich theoretical perspectives. In this chapter I present the evidence for a small feast at the Anna site in southwest Mississippi (Figure 7.1). The Anna site is one of the largest of the Plaquemine culture (a.d. 1200–1700) sites located in the Natchez Bluffs re-

128

v irgil roy be asley iii

Figure 7.1. Location of Anna site.

gion (Brown, this volume). Excavations in 1997 north of Mound 4 revealed a small mound that was no longer visible from the surface. On one summit of this mound an assemblage of animal bone and partial ceramic vessels was encountered, argued here to represent the remains of a small feast. A quick review of the literature and archaeological perspectives on feasting is presented, including how identity relates to feasting behavior. This is followed by a brief introduction to the Anna site and its place in Plaquemine culture, detailing archaeological evidence from one block of the 1997 excavations at the Anna site. It is proposed that these remains of a small feast are representative of local negotiations of identity.

Archaeological Perspectives on Feasting Feasting is a common feature throughout the world’s societies, with roots that may reach back as far as the Upper Paleolithic (Hayden 1998:22). Such a com-

a nna site exc avations in the natchez bluffs

129

mon trait has garnered the interest of anthropologists, with numerous references to feasting found in the ethnographic literature (e.g., Keesing 1983; Leach 1964; Malinowski 1922; Mauss 1990; Swanton 1931; Walens 1981; Weiner 1988), though only a few studies have examined feasting in a theoretical or cross-cultural light (Dalton 1977; Dietler and Hayden 2001; Mauss 1990; Rappaport 1968; Suttles 1968). Much more recently, archaeologists have begun to express an interest in this phenomenon and the insight such activity can provide toward questions of developing social complexity (Clark and Blake 1994), status (Kelly 2001; Smith and Williams 1994; VanDerwarker 1999; Welch and Scarry 1995), aboriginal economics and warfare (Costin and Earle 1989; Dalton 1977; Kolb 1994), performance and display (Costin and Earle 1989; Kelly 2001; Knight 2001; Kolb 1994; Pauketat et al. 2002), and cultural ecology (Hayden 1998). In concert with this growing interest, archaeologists have developed new and innovative techniques to investigate feasting in such areas as ceramic analysis (Blitz 1993; Welch and Scarry 1995), faunal studies ( Jackson and Scott 1995), and archaeobotanical research (Welch and Scarry 1995). Though often differing in considerations of scale and participants’ social distance, various attempts to de¤ne what is a feast tend to share the attribute of “communal consumption of food/drink” and feasts are distinguished from daily meals and exchange of food goods without associated consumption (Dietler and Hayden 2001:3). In the 2001 volume, both Dietler and Hayden provide speci¤c de¤nitions of feasting, with Dietler highlighting the ritual aspect of feasting and the shared consumption often accompanied by performances and social transactions. Hayden takes the more liberal view of any atypical shared meal being a feast. Generally, it appears Hayden’s de¤nition has gained the greater currency (e.g., Mills 2004). Hayden (2001) provides a most thorough typology of feasts, dividing all feasts into three major divisions: (1) alliance and cooperation feasts, (2) economic feasts, and (3) diacritical feasts. While this segmentation would seem to indicate a wide variety in the kinds of feasts encountered archaeologically, it is obvious that the majority of feasting literature emphasizes competition (often through showiness), politics, asymmetry, and division (e.g., Blitz 1993; Costin and Earle 1989; Dietler 2001; Hayden 2001; Jackson and Scott 1995; Joffe 1998; Kelly 2001; Knight 2001; Kolb 1994; Pauketat et al. 2002; Perodie 2001; Phillips and Sebastian 2004; Smith and Williams 1994; VanDerwarker 1999; Welch and Scarry 1995), though it would be a gross misstatement to characterize all feasting literature in this manner (e.g., Potter and Ortman 2004; Wilson and Rathje 2001).

130

v irgil roy be asley iii

This emphasis on feasting as political discourse provides a fertile ¤eld from which to reap insights into social interaction and practice. However, this focus has a tendency to create a situation in which competition is expected as a component of feasting behavior. Perhaps there always is competition—observe a modern family reunion, church picnic, or barbeque and you will certainly ¤nd sport in food presentation and quality—but archaeologically, the expectation is that the competition has meaning in politics and cultural evolution. The sharing of food and drink, as a component of togetherness, rather than distinction, is lost.

Archaeology, Identity, and Feasting Identity as a conceptual framework holds tremendous potential in archaeology. De¤ned here as the acknowledgment/expression of social and/or cultural sameness, identity shares some of the characteristics of what is traditionally thought of as archaeological systematics (e.g., cultures and phases). However, identity differs in that while it is often expressed materially, the scale of expression can range widely from a family group to a nation-state, or even the entire planet (“I am a citizen of Planet Earth”), and it is based, at least theoretically, on emic notions instead of assigned classi¤cations. While we cannot make simplistic assumptions that shared material culture equals shared identity (Mills 2004; Shennan 1994), the often strong correspondence between the two, especially when multiple classes of artifacts, context, and timing are considered, provides an ideal opportunity for exploring this basic human phenomenon. Identity as a construct allows for comparison of a behavior that produces a recognizable material signature, for example, feasting, cross-culturally and intra-culturally. Given that identity is an integral part of life, be it personal, political, religious, ideological, or even avocational, it is no surprise that food and feasting are strongly expressive of belonging. Food, as a necessary condition of life, is the economic product around which much of the material universe revolves. Food is the central fruit of our labors. Consumables and related material paraphernalia thread through many social interactions. Food is “sign, symbol, and product” (Weismantel 1988:7), having perhaps the greatest malleability in practice. Certain acts of sharing food are recognized as feasting and the material remains of feasting behavior are recoverable archaeologically (Dietler and Hayden 2001). Nearly all levels of social complexity likely carry out the practice of feasting, though Hayden (2001:54) has doubts about generalized hunter-gatherers. Combining the ubiquity of feasting behavior with the con-

a nna site exc avations in the natchez bluffs

131

Figure 7.2. Site map of Anna Mounds group (after Jennings and Wagner 1940).

struct of identity provides opportunity for robust and meaningful comparisons across and within cultures.

The Anna Site The Anna site (22AD500) is one of the largest of all the Natchez Bluffs Plaquemine sites. Material and geographical continuities indicate that Anna is likely one of several sites that were occupied by predecessors to the famed historic Natchez (Brain 1978; Brown 1998c:57). Within the area presently de¤ned as the site, a minimum of eight visible earthen mounds occur, ranging from approximately 2 m high to about 16.5 m (Figure 7.2). This study adds another small mound, not visible from the surface, to the inventory. Due in large part to the nature of the terrain, the site has never been accurately surveyed and mapped in total. Jesse Jennings and C. A. Wagner produced the most complete and reliable map in 1940, a sketch map with elevations established for mound heights. The site has changed somewhat since their survey, the most drastic cultural alteration being a re-

132

v irgil roy be asley iii

routing of Pine Ridge Road, which now splits the area between Mounds 3 and 4 (Figure 7.2). In the Natchez Bluffs region, the most striking physiographic features are the extremely dissected loess bluffs. These bluffs, which are composed of a ¤nely grained, velvety soil, date to the Pleistocene and are believed to be aeolian in origin (Saucier 1994:186). It is the nature of the loess that has most visibly impacted the Anna site. When tree cover is removed, the leaching out of calcium causes the soil to lose cohesion, and erosion can be severe. Loess that has not been weathered can maintain a vertical slope if protected from surface runoff, but weathering creates an acute situation of erosion susceptibility (Saucier 1994:185). Gaping ravines encircle and cut through the site and await the diligent soul who ventures into the surrounding forest. These ravines have destroyed some portion of the original site and have surely resulted in the destruction of valuable archaeological contexts. Although this erosion was greatly accelerated by the removal of tree cover in the mid-nineteenth century and by plow agriculture since then (Neitzel 1983:19), the bluffs were probably of higher relief in prehistoric times. Modern agricultural and forestry practices have only accelerated a preexisting condition. This pattern of bluff tops and ridges combined with large ravines may have affected settlement, as the lack of large expanses would have predicated the need for a dispersed farmstead pattern. The Mississippi River ®oodplains do provide large, ®at expanses. However, archaeological surveys have not systematically covered these areas or the talus slopes. It is possible given local erosional tendencies that many sites are covered in colluvium. Elsewhere in the Plaquemine culture area, evidence points toward limited occupation of the principal sites, with the majority of population scattered throughout the countryside, in contrast to Mississippian sites to the north (Brown 1985b:253). Unfortunately, in the Natchez Bluffs area, no one mound site and its surroundings have been so thoroughly sampled as to make a supportable conjecture as to the settlement pattern. The non-mound site of Lookout was intensively investigated and appears to be a small village with an associated cemetery (Brown 1985a:56). Future sampling of sites such as Anna may provide better answers to the question of whether the large mound sites in the Natchez Bluffs region represent inhabited villages or abandoned ceremonial centers, where the majority of the population lived away from the site (Williams and Brain 1983:340). Work in the Lower Big Black River basin supports the dispersed settlement model, with small sites, possibly hamlets, scattered along the river ( Johnson 1997). Various archaeologists working under the auspices of the Lower Mississippi

a nna site exc avations in the natchez bluffs

133

Survey have extensively surveyed the region, including the Yazoo Basin, the Tensas Basin, and the Natchez Bluffs, areas that differ physiographically but are culturally similar (Brain 1978:332–334; Brain et al. 1994; Brown 1985a; Brown and Brain 1983; Phillips et al. 1951). Within the Natchez Bluffs region, there are over 125 recorded sites with at least some representation of Native American populations, but the archaeology on the preponderance of these is extremely limited (Brain et al. 1994). The majority of professionally excavated sites are mound sites (e.g., Fatherland, Emerald, Feltus), but there are exceptions (Brown 1985a). While this scarcity of information makes large-scale interpretations and explanations very dif¤cult, it emphasizes the need for future research in the region.

Previous Investigations at the Anna Site The Anna site was well known to scholars who made their homes in the Natchez region or visited it in the nineteenth and early twentieth centuries (Brown 1926:40–42; Brown 1997:19–20; Culin 1900:124–127; Moorehead 1932:162–163). Several different names have been applied to the site, including the Voucherie group (B. L. C. Wailes in Brown 1996), the Robson group (C. Brown 1992), the Lewis group, the Stowers group (Brown 1997), and the Anna group (Brown 1926:40–42; Brown 1996). Despite this widespread recognition of the importance and immensity of the site, very little archaeology beyond artifact acquisition took place in these early years. B. L. C. Wailes, a noted scholar and recorder of Native American remains in the Natchez region, made a cursory visit to the site in April 1853, giving one of the best descriptions of the era (Brown 1998a). Warren K. Moorehead, in the company of Calvin S. Brown, conducted excavations at the site in 1924, opening up a minimum of ¤ve units in various locations throughout the site (Moorehead 1924, 1932). At least one of the units excavated was in Mound 4, the area of which this chapter is primarily concerned. Calvin Brown gives a description of some of the Moorehead excavations, contributing measurements of the mounds and relating a brief description of the materials found in Mound 4, deferring the more intensive report to Moorehead (Brown 1926:40–42). Unfortunately, Moorehead, in typical fashion, provides only a ®eeting description of his activities at Anna, mentioning his excavations into Mounds 4 and 6 and giving a brief account of the cultural materials found in Mound 4. Moorehead refers the reader to Calvin Brown’s work for further details (Moorehead 1932:162–163). James A. Ford made a surface collection at the site sometime prior to 1936, probably stemming from his earlier work in the area with Henry B. Collins.

134

v irgil roy be asley iii

This collection was illustrated by Ford (1936:111) and placed within his Tunica complex. Ford mentions that ravines surrounded the site but notes no further damage or alterations. Jesse Jennings, working under the auspices of the National Park Service’s Natchez Trace Parkway project, visited the site in 1939 or 1940. In his unpublished report to the National Park Service, Jennings notes that the owner had opened Mound 5 for the purposes of a silage trench. Silage is a type of feed given to cattle that is made from fodder that is fermented. Jennings recommended that the site undergo further archaeological work, be acquired by the National Park Service, and be converted into a public park with a museum ( Jennings 1940:5–11). These recommendations were never achieved, and the site received no professional attention for almost a decade. Philip Phillips visited the site in 1947 and noticed that two of the mounds, 1 and 5, had been opened up for silage trenches (Brain et al. 1994). Phillips suggested that Mound 5 be professionally excavated and John Cotter, Jennings’s successor with the National Park Service, followed up on this suggestion in 1948. At that time, Cotter opened up a 5-by-10-foot unit in the south wall of the silage trench in Mound 5 (Cotter ca. 1950:3). Cotter’s work led to the only detailed publication of archaeological excavations conducted at Anna, an American Antiquity article dealing with stratigraphic tests at the Anna and Emerald sites (Cotter 1951a). The landowner had discovered 15 whole or partial vessels during his trenching of Mound 5. Cotter’s excavations revealed at least four stages of mound construction, all except the last capped by a distinct occupational surface including evidence of structures (Brain et al. 1994; Cotter 1951a). Ceramic classi¤cation and seriation were the main purposes of Cotter’s work, and the excavations at Anna helped to provide the basis for the Natchez Bluffs chronology. The material recovered by Cotter was reanalyzed by Vincas P. Steponaitis using the type-variety system prevalent in the area today (Phillips 1970; Steponaitis 1981, 1995) and incorporated into Ian W. Brown’s section on the Anna site in the Lower Mississippi Survey volume (Brain et al. 1994). The results obtained by the Lower Mississippi Survey have supported and re¤ned Cotter’s original conjectures.

Importance of the Anna Site in Prehistory As one of the largest sites in the Lower Mississippi Valley (LMV), Anna has been accorded great signi¤cance in the prehistory of the region. On the basis of Cotter’s Mound 5 excavations (Cotter 1951a), it appears that the ¤rst substantial occupation of the site occurred during the Anna phase (a.d. 1200– 1350; Figure 7.3). The 1997 excavations reveal at least some Balmoral phase occupation (Beasley 1998). The Anna phase is the initial expression of Plaque-

a nna site exc avations in the natchez bluffs

135

Figure 7.3. Neo-Indian chronology in the Lower Yazoo Basin and Natchez Bluffs region of the Lower Mississippi Valley (from Steponaitis 1995:Figure 1).

mine culture, known as a variant of the Mississippian cultural tradition (Brown 1985b). Plaquemine culture, which has been termed “Mississippianized Coles Creek” by Brain (1978:345), is surprisingly tough to de¤ne. For some, while material culture and practice differ, there is no qualitative difference between Plaquemine and Mississippian (Steponaitis 1998:15). Plaquemine is characterized by a substantial increase in mound-building activities and perhaps a greater emphasis on beans, maize, and squash agriculture. For much of the Plaquemine area, shell tempering of pottery is always in a minority, and burial treatments rarely approach the sumptuousness of Cahokia, Moundville, or Etowah. Williams and Brain (1983) distinguish a general decrease in Mississippian ceramic traits as one progresses south. Increasing population and an expansion of major centers such as Anna occurred throughout the region, perhaps re®ective of the new subsistence base (Brown 1985b:254). Because Anna is the largest site in the Natchez Bluffs during the Anna phase, it has been hypothesized that the site was the apex of a multitier hierarchy characteristic of a complex chiefdom (Brain 1971b, 1978). Unfortunately, despite the large amount of survey and archaeology carried out within the region (Brain et al. 1994), there is not enough information to either support

136

v irgil roy be asley iii

or refute this hypothesis. Recently, scholars have again questioned the idea that the historic Natchez Indians were organized into a complex chiefdom (Lorenz 1997; Muller 1997:63–68), an assumption long taken for granted by archaeologists (Brain 1978; Knight 1990a; Steponaitis 1978). While Lorenz and Muller fail to take into account several factors in their analysis, most signi¤cantly the in®uence of two different European powers attempting to manipulate native alliances, the caution is warranted. Only with a much more careful examination of the settlement data at hand and improved recovery at the presumed lower-order centers can such hypotheses be supported.

1997 Block 1 Excavations Mound 4 and the accompanying “®ats” are topographically a bluff remnant formed by road construction and erosion that has bisected the original site. This remnant was chosen for excavation on the basis of a number of factors. The ¤rst of these considerations was that while the road cut and erosion have seriously de®ated the area surrounding Mound 4 and the ®ats, what remains appears to be fairly intact. Earlier exploratory investigations by Ian Brown and Richard Fuller had revealed the presence of large amounts of daub and pottery fragments in the eastern cut, known as the Old Spanish Road. The main intention of the 1997 project was to recover architectural information that would expand our knowledge of Plaquemine culture house types (Brown 1997:4, 1985b). With these factors in mind, it was determined that this area would be an excellent candidate for excavation. A 34-by-14-m grid was laid out on a magnetic north–south axis and soil probe tests were conducted every 4 m. The tests revealed a concentration of daub and artifacts in the area between the N24W00–N32W00 stakes and the N24W08–N32W08 stakes. Soil probe testing was followed by shovel testing to the edge of the areas where artifact concentrations were found. The shovel tests produced an insight into the stratigraphy and provided us with an estimate of the depth to subsoil. The combination of surface collections, soil probe testing, and shovel testing enabled us to place an 8-by-8-m block where it had the best chance of yielding architectural remains. The excavation block (Block 1) was divided into 2-by-2-m squares and each square was given a numerical designation. Two units were initially selected for excavation, G211 and G213. Each of these units was excavated using arbitrary 10-cm levels. G213 consisted entirely of a dark brown-black midden soil, with abundant bone, charcoal, daub fragments, and pottery. The base of G211 revealed the presence of four small post features in a linear arrangement, contained within an arc of yellowish-brown soil. This was thought to indicate a

a nna site exc avations in the natchez bluffs

137

structure, and the ensuing excavation strategy was aimed at attempting to follow out this pattern. As more units were opened, it became apparent that there was a detectable pattern to the yellow-brown soil. Eventually it was found that this distinct soil formed a general oval pattern (Figure 7.4). This was thought initially to indicate either a structure or perhaps a soil cap placed over an earlier structure. The excavation strategy continued to be horizontal in nature, attempting to expose as much of the yellow-brown soil as possible. Each subsequent unit excavated was brought down to the top of this soil level. In unit G214, there were multiple partial vessels found smashed on an occupational surface in situ. The type-variety classi¤cation of these vessels, one Chicot Red, var. Fairchild bowl (Figure 7.5) and one Addis Plain, var. Addis bowl (Figure 7.6), indicates that this yellow-brown surface was at one time an activity area dating to the Foster phase (a.d. 1350–1500). Several large mammal bones were also found on this surface (Table 7.1). Several concentrations of daub were found throughout the units, some fragments being quite substantial. However, as no other structural indications were found, the initial assumption that this yellow-brown surface was related to a structure was questioned. Unit G216 was brought down further and a layer of daub was found, but without further structural information. Pottery sherds were found at various angles throughout the yellow-brown matrix, indicating that the area had been ¤lled with soil over a relatively brief temporal span and was not the result of

138

v irgil roy be asley iii

gradual accumulation. The area around the yellow-brown soil continued to be a dark brown-black soil, thought to be indicative of a midden. While it was apparent that the yellow-brown soil indicated a distinct cultural activity, the nature of this activity was still unclear. After discussion, it was thought that the pitch of the ground surface and the nature of soils in the region would be conducive to sheet erosion, wherein the combination of rainfall and a small angle will cause large horizontal areas of soil to erode. This is apparently what has happened. Part of the original surface of the Mound 4 ®ats has eroded into the Old Spanish Road. Within the road cut, large amounts of artifact-rich colluvium can be found that originated on the ®ats. This erosion, in addition to limited plowing of the surface in historical times, altered the uppermost layers, washing away evidence of the latest activities. Also, it was speculated that the yellow-brown layer was evidence of a dumping episode, not a structure. Eventually it was decided to bring unit G216 down further in order to gain a better idea of the vertical stratigraphy. This decision proved to be vital to explaining the nature of the block. By leaving a balk on the western edge of the unit, we were able to get a section for the entire block. This section clearly showed that the yellow-brown layer was underlain by several ¤ll layers, all overlying a slanting dark ¤ll indicative of mound construction. Just above the subsoil, an Anna phase (a.d. 1200–1350) wall trench was found that cut into an earlier Coles Creek component of the Balmoral phase. The latter component is underlain by the sterile subsoil and is probably indicative of the earliest occupation in the Mound 4 ®ats, although the presence of a few sherds of Tchefuncte Plain, var. unspeci¤ed indicates an earlier occupation in the vicinity. It seems reasonable in light of this information to consider the Block 1 locale a small mound, possibly a house mound, of which the uppermost layers have been washed away. The yellow-brown layer is a cap of this mound onto which further ¤ll was added that has eroded away over the years. Within unit G211 were found a series of small post features (Figure 7.4). Other than the presence of daub, these features are the only structural indicators on this surface. It is possible that these features represent the remains of a small scaffold or screen. The Natchez used such ephemeral structures during feasts (Swanton 1911:114–115, 118). Du Pratz, speaking of the Natchez harvest feast, describes their construction: “The feast day being ¤xed, the necessary arrangements for this ceremony are made some days before. The cabin of the great chief is built opposite the granary and that of the great war chief at the side of this granary. That of the sovereign is on an elevation of earth about 2 feet high, which has been brought hither. It is made by the warriors

a nna site exc avations in the natchez bluffs

139

Figure 7.4. Plan view, Block 1 excavations.

of grass and leaves” (Swanton 1911:114). Dumont, in describing the same feast, also narrates the construction of temporary structures: “Eight days before it began [the harvest feast] the savages cut all the grass on the trail over which their great chief, who was then the Tattooed-serpent, would have to pass; that is to say, for the space of about a league and a half, the distance between the great village and this tun [granary]. At the same time, they prepared many cabins of branches around a beautiful open space which they had prepared

Figure 7.6. Addis Plain, var. Addis vessel.

Figure 7.5. Chicot Red, var. Fairchild vessel.

a nna site exc avations in the natchez bluffs

141

beside the tun, and at the end of this open space they raised another cabin more ornate than all the others, destined to serve as the palace of their chief ” (Swanton 1911:118). Thus, substantial structural remains would not be expected at the scene of a feast. The Mound 4, Block 2 locale, also excavated during the 1997 season and between the Block 1 locale and Mound 4, revealed the presence of several post features. Interestingly, almost all of the smaller, nonstructural post features were found to contain ¤sh vertebrae and/or gar¤sh scales (Boudreaux 1997). Artifacts found within these features date to the Foster phase or later. Brown (1997:6) proposes that post features in the Block 2 locale may represent an open warm-weather structure. Alternatively, these features could be representative of a temporary structure of the type employed by the Natchez during feasts (Swanton 1911:114–115, 118).

Stratigraphic Interpretations Though not the focus of this study, it is also useful to brie®y consider the overall excavation block at this point in order to place the deposits in chronological context. The reader is referred to Beasley (1998) for a fuller discussion. The earliest stratigraphic unit at the Mound 4, Block 1 locale occurs in levels G216E1–F1. On the basis of the presence of sherds of Beldeau Incised, var. Beldeau and Coles Creek Incised, var. Mott, this unit dates to the Balmoral phase of the Coles Creek period (Brown 1998b). However, this unit is intruded by a wall trench believed to date to the Anna phase and the Balmoral phase elements may be of secondary context. The inclusion of three sherds of Anna Incised, var. Anna argues for some type of intrusion. Gordon phase diagnostics were absent, arguing against this being a transitory deposit. On the basis of the available information, the best interpretation is one in which an Anna phase structure was placed over and through an earlier Balmoral phase component, which represents the premound occupation of this area. Units G216C and G216D both contain sherds that are strongly indicative of an Anna phase occupation, including Anna Incised, var. Anna, L’Eau Noire Incised, var. L’Eau Noire, and Plaquemine Brushed, var. Plaquemine (Steponaitis 1981, 1995). This dating lends credence to the above interpretations and provides further insight into the construction sequence of this area. The west and south pro¤les indicate a substantial accretion of detritus that dates to the Anna phase occupation, consisting of a structure that appears to extend toward the southeast. The material from this structure, a black clayey silt with midden properties, slopes toward the periphery of the excavation block, lending support to the interpretation of this being a mound. A post feature not

142

v irgil roy be asley iii

detected during the horizontal excavation but apparent in the pro¤le was associated with this wall trench and midden-like deposit. This feature may be the remnants of a corner post associated with the Anna phase structure. Rectangular wall-trench patterns are known for the Anna phase (Brown 1985b: Figure 10, Table 2, 277), and this type of construction continued to be employed throughout the Natchez occupation of the area (Brown 1985b:277; Neitzel 1965:Figure 12, 1983:Figures 7, 14). The initial purpose of these excavations was to recover architectural information relating to Plaquemine culture houses. As such, most of the excavations were horizontal in nature, resulting in a large areal exposure but a minimal vertical exposure. The majority of the units, excepting G216, were taken down only to the yellow-brown soil change or slightly below. These units, which are immediately below the plow zone, date to the Foster phase. Excavation below the initial yellow-brown soil change revealed that the ¤ll below also dated to the Foster phase. The few units analyzed that included the plow zone (e.g., G216A) tended to represent mixed deposits, although there is a substantial representation of Emerald phase materials, which may indicate the ¤nal occupation of the area. Those units that fell out of the area where the yellow-brown soil existed, such as G213B, also tended to contain mixed deposits. Plowing of the surface in the early part of this century ( J. Hall 1997) and continued sheet erosion have created a situation wherein the area around the mound has been overlain with a thick zone of eroded soil. Pro¤les support this conclusion, as the dark soil of the plow/erosion zone thickens toward the east, which is the lower side of the area and toward the Old Spanish Road. The combination of partial vessels and faunal materials discussed earlier was found overlying the yellow-brown soil. The artifacts found on this surface, including the Chicot Red, var. Fairchild and Addis Plain, var. Addis bowls mentioned above, date the deposition of the materials to the Foster phase or later.

Discussion The 1997 excavations north of Mound 4 at the Anna site revealed the presence of a small mound. On one of the latest summits of the mound, faunal and ceramic materials were recovered that are thought to represent the material remains of a small feast. Ethnohistoric observations of feasting among the Natchez are congruent with the recovered archaeological signature. It is the location and scale of this feast that makes arguments of competition and displays of power dif¤cult to support. Alternatively, if we approach this assemblage as the result of a small-scale feast intended to express and

a nna site exc avations in the natchez bluffs

143

promote identity within one segment of Anna society, such interpretation is unnecessary. The obvious question is, what group is expressing identity by participating in this mound summit feast? Without further research, any answers proposed here can only be tentative. The feasts observed by du Pratz and Dumont clearly involved a component of identity, with wide-®ung members of Natchezan society traveling to a central location to participate in an event that not only demonstrated the position and authority of elites but also served to unite people as a means of materializing community. The feast described in this chapter involved only a small number of participants and was not likely of the scale of the harvest feasts described above. Rather, this deposit seems to indicate a few people coming together for a single communal meal. The location of the deposit on the summit of a small mound indicates that the feast may be associated with the adding of a mound mantle, but this again begs the question of why the mound was built. A major life event (e.g., death, birth, marriage), the creation of alliances, celebrations, or numerous other events could have served as impetus for the mound construction/feast; no single answer is certain without further research.

Plaquemine Culture and Identity While again pushing the margins of this chapter, a brief discussion of Plaquemine culture as an analytical concept seems warranted. After years of uncritical usage, the notion that Plaquemine is “Mississippianized Coles Creek” (Brain 1978; Williams and Brain 1983), a hybridization of indigenous Coles Creek people and Mississippian ideas, is at issue (Brown, this volume; Kidder 1998a; Rees and Livingood, this volume). Unfortunately, such terminology obfuscates indigenous contributions. Coles Creek appears as a static entity, passively accepting the rising waters of the Mississippian ®ood. If the LMV is merely a receptacle for northern hegemonic practices, local originality, including the existence of one of the earliest traditions of mound building (Saunders et al. 1997), is lost. Only the porous defense of distance and backwater swamp provided any protection from the Cahokian juggernaut (Williams and Brain 1983:411–414). Approaching Plaquemine culture as the materialization of identity allows us to view constellations of material culture, dispersed settlement pattern, and distinctive ideological and ritual practices as analytical units comparable to regional Mississippian traditions, such as Moundville, Etowah, and Cahokia, rather than a diminished version of a pan-Mississippian phenomenon. Plaquemine people seem to have been expressively different from contemporary

144

v irgil roy be asley iii

groups, whatever the mechanics of “in®uence.” The historic Natchez are often appealed to for models of southeastern chie®y organization, but the archaeological record does not meet the expectations of chiefdom organization (I. Brown 1990; Lorenz 1997; Mason 1964; Rees 1997). It is the individuality of Plaquemine that promises the greatest and most immediate research potential.

Conclusions The Anna mound group remains one of the most important sites in North America, and the tiny amount of excavation accomplished so far only provides a tantalizing glimpse of the archaeological potential of the site. The 1997 excavations detailed here indicate that at least part of the site had some Coles Creek occupation, with a continuous usage throughout much of the Plaquemine time frame. Results con¤rm and reinforce earlier chronological observations (Cotter 1951a; Steponaitis 1981, 1995), with the ¤nal occupation occurring during the Emerald phase (a.d. 1500–1650) and no historic Native American utilization clearly demonstrated. Identity as a theoretical construct holds promise for readdressing the material culture of Plaquemine societies at the local and regional levels. By emphasizing the materialization of behaviors, small areal excavations can expose episodes of deposition and build upon the long practice of ceramic typology in the LMV. Creating a synthetic description of Plaquemine culture and theorizing the diversity of late protohistoric materializations in the Southeast will provide important insights into historically particular routes to complexity.

8 Plaquemine Culture in the Natchez Bluffs Region of Mississippi Ian W. Brown

In this study I explore the archaeological expression of Plaquemine culture along the loess bluff hills of southwest Mississippi (Figure 8.1). It was during late prehistoric times, circa a.d. 1200–1550, that this region experienced its greatest sociopolitical complexity, but the culture itself continued into historic times as re®ected in the Natchez polity. Plaquemine is known to have developed in the lower Mississippi River alluvial valley out of the preceding Coles Creek culture (Kidder 1998a, 1998b:129–130), a vibrant cultural expression whose people were some of the ¤rst to make use of platform mounds for the support of their buildings. Ever since James Ford’s publication of Analysis of Indian Village Site Collections from Louisiana and Mississippi, it has been well known that the Natchez Bluffs region has a strong representation of Coles Creek sites (Ford 1936:172– 216). Even though the Greenhouse site (Belmont 1967; Ford 1951) in the Lower Red River region is the site that contributed most to Ford’s de¤nition of this culture, the Natchez Bluffs region was recognized early on as one of its major expressions. Sites like Feltus (often referred to as Truly or Ferguson; see Brain et al. 1994; Brown 1926:45–46, 265; Culin 1900:122–123; Ford 1936:198–199; Moorehead 1932:163–164; Phillips 1970:920, 948), Smith Creek (Boggess and Ensor 1993; Brain et al. 1994; Ford 1936:193–198; Phillips 1970:920), and Mazique (Baca 1989:38; Brain et al. 1994; Brown 1926:34–35; Brown 1998a:172; Cotter 1948; Culin 1900:127–128; Frank 1975; Ingraham 1835:2:180; Moorehead 1932:165; Phelps and Jennings ca. 1940s; Phillips 1970:920; Rowland 1907:1:911, 1925:1:45, 266–269, 289) are important mound sites in the Natchez Bluffs that have produced large numbers of typical Coles Creek pottery (Brown 1998b; Ford 1935b, 1936:172–218). But what often goes unrecognized is that these sites also have strong Plaquemine components (Phillips 1970:920). As with the

Figure 8.1. Selected Plaquemine sites in the Natchez Bluffs region of Mississippi.

pl aquemine culture in the natchez bluffs

147

Lake George site in the Lower Yazoo Basin (Phillips 1970:278–304; Williams and Brain 1983), the thought has tended to be that these Coles Creek towns were already in existence prior to Plaquemine development and that there was a relatively smooth transition between the two. Presumably Coles Creek people received inspiration in some manner from Mississippian societies to the north, from Cahokia in particular, and then transformed into something that we have come to know as Plaquemine culture (Brain 1978, 1989; Brown 1985b, 1998e; Brown and Brain 1983; Grif¤n 1946, 1967; Hally 1972; Jennings 1952:267; Kidder 1998b:130–132; Phillips 1970:967–968; Quimby 1951, 1957; Steponaitis 1981; Weinstein 1987a; Williams 1967:9; Williams and Brain 1983). Over the years Brain (1971a:76–77, 1978:344–350, 1989:122–125; Williams and Brain 1983:412–414) has consistently argued that the Plaquemine variant of Mississippian, which characterizes much of the Lower Mississippi Valley in late prehistory, basically represents a hybridization of Coles Creek and Mississippian cultures. If indeed one ¤nds a strong local mound-building population whose sites continue with unbroken occupation and that also exhibits material continuities, then it is reasonable to presume that one developed into the other. That certainly seems to be the case in the Lower Yazoo Basin (Phillips 1970:12–13, 560; Williams and Brain 1983), and it may also be true for the Natchez Bluffs, but it is important to recognize that the transition between Coles Creek and Plaquemine in the Natchez Bluffs is really not as clear as might be expected.

The Historic Expression of Plaquemine Culture Let us start by examining the historic manifestation of Plaquemine culture in the Natchez Bluffs. The name for the region highlights the fact that the Natchez Indians were the last hurrah of Plaquemine culture (Barnett 1998; Brain 1978; Brown 1998c, 1998e; Oswalt and Neely 1999; Quimby 1942; Stern 1977). Spanish and French explorers were fascinated by the complex social and political structure of the protohistoric and historic Natchez (Albrecht 1946, 1948; Brain 1971b; Brown 1982a, 1989b, 1992; Clayton et al. 1993; Fischer 1964; Hart 1943; Knight 1990b; Loren 1995; Lorenz 1997; Quimby 1946; Steponaitis 1978:421–423; Tooker 1963; White et al. 1971). French adventurers were wont to draw comparisons between the Great Sun of the Natchez and Louis XIV, France’s own Sun King, but just how great was the Great Sun? As a result of work by James A. Ford, Robert S. Neitzel, Moreau B. Chambers, Andrew C. Albrecht, George I. Quimby, and others, we know that the Fatherland site was most certainly the Grand Village of the historic Natchez (Albrecht 1944; Barnett 1998:4–8; Brown 1998a:175–176, 1998d; J. Brown 1990;

148

i a n w. brow n

Ford 1935b, 1936:59–64; Green 1936; Lorenz 1997:106; Neitzel 1964, 1965, 1978, 1981, 1983; Phillips 1970:948–949; Quimby 1942, 1953; Steponaitis et al. 1983; Williams 1962, 1966, 2003:xiii, xvi–xvii). In the early eighteenth century it was the principal town of the Natchez nation and, as such, it has received great attention in the writings of Le Page du Pratz (1758, 1972 [1774]), Dumont de Montigny (1753), and other French writers (Delanglez 1935; French 1869; Giraud 1974; McWilliams 1988; Rowland and Sanders 1927, 1929; Rowland et al. 1984a, 1984b; Surrey 1968; Swanton 1911). Despite the fact that Fatherland has received an inordinate amount of attention historically and archaeologically, it really is a rather small and unassuming site. Three mounds were built on the banks of what was once a shallow, gently ®owing St. Catherine Creek. In historic times only two mounds were in use, with Mound B supporting the Great Sun’s house and Mound C the temple. A small plaza stretched between these two edi¤ces. Mound A seems to have been retired by historic times. According to Lewis et al. (1998:5), Fatherland ¤ts the de¤nition of a “Mississippian town,” which consists of a number of mounds arranged around a public arena and substantial evidence of habitation. A “Mississippian mound center,” in contrast, is a planned site that exhibits earthworks but yields little to no evidence of occupation. I do not mean to imply that Fatherland was ever a large town, though, as Neitzel (1983) certainly was hard pressed to discover houses between the mounds. However, he did detect some evidence of habitation—enough to show that there was indeed a resident population. The French were impressed with the Great Sun and his perceived power, but they really could not have been overly excited by the town of this noble and of his brother the Tattooed Serpent. One wonders just how powerful the Natchez suns really were, especially considering the frequency of in¤ghting. The related nobles at the White Apple, Jenzenaque, and Grigra villages always seemed to have been ®exing their muscles over something or other (Albrecht 1946; Brain 1976; I. Brown 1990:235; Lorenz 1997; Swanton 1911:186–257). Most students of the region have tended to lean toward the observed factionalism as having been primarily a product of the turbulent seventeenth and eighteenth centuries, as French and English agents wrestled for Indian allegiance (Brain 1976; Brain et al. 1994; Brown 1982a, 1985a, 1989a, 1989b, 1992; Crane 1929; Neitzel 1965:55–57; Quimby 1946). European in®uences are thought to have literally torn the social and political order apart in the early eighteenth century, but that alone cannot explain the presence of factions. The prominence of the Fatherland site and its noble family should really be considered the last blip of what was probably a normal Plaquemine politi-

pl aquemine culture in the natchez bluffs

149

cal process (Lorenz 1997; Muller 1997:63–68, 1998:185). Too often we tend to forget that the historic mounds at Fatherland are but the end product of a long history of construction and usage ( J. Brown 1990; Neitzel 1965). It is important to remember that Fatherland was continuously occupied for over ¤ve centuries before its abandonment in 1730. Although there was signi¤cant habitation at Fatherland during the Anna phase, between a.d. 1200 and 1350, mound construction does not appear to have begun until the following Foster phase, between a.d. 1350 and 1500. It is extremely doubtful, however, that Fatherland was a prime center in the region until late in its history.

Emerald, Anna, and the “Bubbling Porridge” Model Far more important than Fatherland in the Natchez Bluffs region between a.d. 1200 and 1500 were sites like Emerald and Anna. Emerald is located along the Natchez Trace at the headwaters of Fairchild’s Creek, while Anna is situated about 10 miles north of the city of Natchez on a bluff top overlooking the Mississippi River’s alluvial valley. Over the years there has been considerable archaeological literature relating to these two sites, but we have still just barely come to terms with their overall signi¤cance. Emerald is by far the more imposing of the two sites (Brain 1978:Figure 12.6; Brain et al. 1994; Brown 1926:36–40, 256–264; Brown 1985b:266; Cotter 1951a; Cotter, ed. 1951; Moorehead 1932:161–162; Phillips and Brown 1978: 204–205; Steponaitis 1974). Its base mound is just shy of being the largest mound structure in the United States, second only to Monks Mound at Cahokia. Part of the reason Emerald is so spectacular perhaps has to do with an overzealous National Park Service restoration program in the late 1940s, but its grandness is also related to the fact that a natural hill was incorporated in the construction of the enormous base mound. The large secondary mound that was built upon the western end of the base mound is comparable in height to the principal mound at Anna. Emerald’s construction began circa a.d. 1200 in Anna phase times, with additional stages having been added during the Foster and Emerald phases. Anna’s occupational history overlaps Emerald’s as it, too, is known to have had mound construction that began during the Anna phase and continued through both the Foster and Emerald phases. There is also some evidence for a very slight historic Natchez Indian occupation at Anna (Beasley 1998; Brain 1978; Brain et al. 1994; Brown 1926:40–42; Brown 1985b:265–266, 1989b:17–19, 1997, 1998a:173–174; Cotter 1951a; Downs 2004; Ford 1936:111, 114, Figure 22k–s; Frank 1980a; Moorehead 1932: 162–163; Warhop 2005). So which of these two towns supported the prehistoric and protohistoric

150

i a n w. brow n

Great Suns? I suspect that they both did, but at different times as power bounced back and forth between the two. Even while Anna was the residence of one such Great Sun, it is probable that Emerald and other Plaquemine towns in the region had nobility who were continually vying for power, supreme or otherwise. In An Archaeology of the Soul, Robert Hall has likened Mississippian towns to “city-states governed by all-powerful hereditary rulers who would have been known as princes had the seats of their authority been located in France or Italy” (R. Hall 1997:146). I believe this is certainly applicable to Plaquemine lifeways in the Natchez Bluffs region. Towns like Anna, Emerald, Glass (Brain et al. 1994; Brown 1984; Ford 1936:69–71; Moore 1998 [1911]:41–48), and Windsor (Brain et al. 1994; Brown 1998a:172–173) are impressive features on the landscape because of the size of their mounds, but that does not mean that they were always enormous. As Blitz and Livingood (2004) reveal, the building of mounds at Mississippi period towns is an extremely complex issue. The structural sequence of mounds at major towns, for example, seems to have had a different grammar than that witnessed at smaller towns. Moreover, although a large portion of the size of mounds can be attributed to duration of site use, perhaps as much as 40 percent, those who prefer “powerist” explanations can highlight the fact that as much as 60 percent of the volume of mounds at the big sites cannot be attributed to duration of occupation; other factors were involved. For the Natchez region, most of the larger and smaller sites were occupied throughout the greater portion of the Mississippi period, but mound construction and usage at these sites seems to have had its ups and downs, or rather its ups and not so ups. In essence, the evidence is not supportive of the model of a paramount chie®y family that was located at one continually dominant town. Rather, the Plaquemine political order in the Natchez Bluffs is analogous to a “bubbling porridge” wherein underlying currents continually gave rise to air blasts that popped away at the surface. Every once in a while one of the bubbles became considerably larger than the others in what appears to be a random fashion. And when that happened, the nobility of that particular bubble ruled for a time of varying duration. If this is a valid model, towns like Fatherland, Foster, Gordon, Feltus, and Bayou Pierre, all of which have mounds that encircle plazas, may have supported the most powerful potentate at any given time, losing out to other nobles at other sites in future generations. No site in the Natchez region seems to have been forever paramount. Nor does it appear that smaller towns constantly catered to speci¤c sites in a consistent and predictable fashion. I do not mean to imply from all of this that I question the overall chiefdom

pl aquemine culture in the natchez bluffs

151

concept, as I do not. Clearly, there were both great chiefs and lesser chiefs in the Natchez region. We already know this to be the case historically, so it is also probably true prehistorically. The nobles who ruled this society, presumably related matrilineally, were distributed across the region in a rather uniform fashion. When one chief was greater than the others, for whatever reason, he or she probably received tribute from the others for the time this status was enjoyed. The main point to emphasize is that the nobility were all located at similarly structured towns. We are not talking about a Moundville here (Beck 2003:651–654; Knight and Steponaitis, ed. 1998; Peebles 1978; Steponaitis 1983; Welch 1991), in which there was one supreme mega-site and all the rest paled in comparison. A site like Anna certainly is big, but its perceived size is in part related to its physical setting. The narrow con¤nes of the bluffedge landscape would seem to have been a major problem for site growth at Anna, but for some reason the mounds did indeed grow large, enormously so. This was not an inevitable situation, but the bubbling porridge model does provide a ¤tting explanation for the phenomenon. Regional consolidation seems to be a critical component in the formation of chiefdoms (Anderson 1996a; Beck 2003:642–643; Blitz 1999; Carneiro 1981; Earle 1987, 1991, 1997; Hally 1996; Milner and Schroeder 1999), and this is also true for Plaquemine culture in the Natchez Bluffs. At the same time, a typology of simple chiefdom versus complex chiefdom really does not work very well for the region (Anderson 1994b:8, 1996b; Blitz 1999; Lorenz 1997; Smith 1978:494–498; Steponaitis 1978, 1991). Robin Beck’s (2003) apical-constituent model seems broadly applicable to the Natchez region. This model, which is not a chiefdom typology per se, looks at how authority is vested in Mississippian chiefdoms. On one end of the scale is the ultimate apical system in which a regional chief delegates authority downward to local-level chiefs. On the other end of the scale is the constituent system wherein community leaders cede authority upward to the regional leader. The Plaquemine variant of the Mississippian tradition, as represented in the Natchez region, ¤ts somewhere along the scale, but its actual position no doubt shifted back and forth through time. Again, much depended on the leadership capabilities and charisma of the various leaders in the respective local towns. I basically agree with John Blitz’s (1999) ¤ssion-fusion process study, which argues that the organizing force of most Mississippi period polities really had little to do with ef¤ciently administering regional tribute. Instead, polity formation and settlement patterns were more likely shaped by compromises made between the various local communities as each sought to maintain maximum autonomy and, at the same time, have mutual security (Blitz

152

i a n w. brow n

1999:589). How to enjoy both freedom and protection simultaneously is a continuing struggle for societies around the world. Sometimes smaller, less powerful communities learn rather quickly that they have to give up a measure of one to obtain the other. Under the ¤ssion-fusion process, the various Plaquemine towns in the Natchez Bluffs could have bubbled forth at any one time and established allegiance linkages if the underlying political currents happened to favor them. By virtue of the ¤nal size of their mounds, I suspect that sites like Anna, Emerald, Glass, and Windsor were favored more often than the others. These towns grew as various nobles ®exed their muscles and as the surrounding leaders of other towns ceded authority to them, not necessarily because the latter were happy to do so but because they knew it was in their best interests in terms of security.

Settlement Hierarchy: Where Are the Plaquemine Villages? It is important to note that towns are not the only form of Plaquemine settlement that occurs in the Natchez Bluffs region. Although each of the towns had a residential population that lived on or around the mounds, most of the people resided either in isolated households or in hamlets scattered along the many bluff tops and ridges (Brain 1978; Brain et al. 1994; Brown 1973, 1985a; Lorenz 1997:106–108; Steponaitis 1978:421–423; see Kidder 1998b for an analogous situation in the Tensas Basin). Almost every little hill in the Natchez region sports a Plaquemine sherd or two if one looks hard enough. These small residential occupations probably consisted of the houses and outbuildings of one or several families. The inhabitants may have been related, but without skeletal data we have no way of knowing for sure. The Lookout site is perhaps the best example of a Plaquemine hamlet in the Natchez Bluffs that has been excavated (Brown 1983, 1985a:11–56, 1985b:264–265). The Antioch, Trinity, and Green¤eld sites are other good examples, but unfortunately they have not yielded much in the way of structural evidence (Brown 1985a:98–147, 167–187). What we are missing in the Plaquemine settlement system are villages, zones of habitation that have substantial populations of related and nonrelated families but that lack mounds or plazas. To my knowledge, Plaquemine villages did not exist in the Natchez Bluffs region. Instead, the settlement hierarchy starts with the isolated households or hamlets. They, in turn, are believed to have catered to nearby towns that supported the local nobility. At any given time, a number of these towns (not necessarily always the same ones) probably also offered their allegiance to yet another town that housed

pl aquemine culture in the natchez bluffs

153

the then current Great Sun, wherever that might be located. But eventually, if they were lucky, some of these subservient towns advanced to “principal town” status, and whenever that happened to a particular town it is likely that major stages were added to its own mounds. The ritual burial of mounds and their associated temple/chie®y structures must have occurred each time that power transference occurred at a town. The archaeology suggests that this was a rather frequent occurrence in the Natchez Bluffs region. Of course a big question is whether it was possible to have stages applied to mounds when a speci¤c town did not serve as the principal seat of government. The Fatherland site alone would seem to argue that this did occur.

Mortuary Patterns Physical anthropology would obviously help a lot in examining sociopolitical connections such as shifts in noble lines, but it is doubtful that much data will be forthcoming on this matter. First, the political situation of the twenty-¤rst century is not really conducive to extensive grave excavation and analysis. It is unlikely that we will see much in the way of research programs directed at reconstructing regional burial patterns. And second, past research has already demonstrated that the mortuary behavior of the Plaquemine people is uneven with regard to archaeological visibility (Brain et al. 1994). For example, it has long been recognized that whenever local informants tell us about looters ¤nding new burial sites, we do not have to even look at the artifacts to know the dates of the burials. Almost all non-mound Plaquemine cemeteries seem to occur late in the sequence, primarily during the late Foster, Emerald, or Natchez phase. And more often than not, these cemeteries are on hilltops or isolated ridges, examples being the Mangum (Bohannon 1963; Brain et al. 1994; Brain and Phillips 1996:363–364; Cotter 1951b, 1952b; Dailey 1974; Phillips 1970:948; Phillips and Brown 1978:205–206; Spaulding 1941), Burthe (Brain et al. 1994; Ford 1936:71; Hally 1972:377–382; S. Williams 1979:21), Ring (Brain et al. 1994; Ford 1936:69), O’Quinn (Brain et al. 1994; Brown 1985a:57–73), and Rice (Brain et al. 1994; Brown 1985a:80–97; Frank 1980b) sites. Admittedly, some isolated historic Natchez interments have been found, such as at the North site (Brain et al. 1994), but these are rare overall. As revealed at Fatherland, the historic Natchez Indians did bury some of their dead in mounds, but these interments are believed to have been either the elite or their retainers (Barnett 1998:24; Ford 1936:57, 61, 64; Neitzel 1965:9, 52–54, 93–95; Quimby 1942:262). Burials in mounds occurred in prehistoric Plaquemine times too, as represented at Glass, Oak Bend Landing (Brain et al. 1994; Moore 1998 [1911]:38–41; Phillips 1970:945–946), Feltus,

154

i a n w. brow n

Quitman (Brain et al. 1994; Brown 1926:34, 44; Culin 1900:118–122; Moorehead 1932:165), Emerald (Brain et al. 1994), and Smith Creek. Almost all of these sites have one or more of their mounds devoted to the deposition of bodies, in either primary or secondary condition. Interestingly enough, we have no idea whatsoever of what Coles Creek culture mortuary patterns were like in the Natchez Bluffs region.

Mound Construction It may seem curious to the reader that Feltus and Smith Creek are included in the distribution of Plaquemine mound sites, seeing that these two sites, along with Mazique, were critical to Ford’s (1936:172–218) de¤nition of Coles Creek culture. However, work by Harvard’s Lower Mississippi Survey (LMS) has revealed that mound construction for at least two of these sites, Feltus and Mazique, probably occurred during both Coles Creek and Plaquemine times (Brain et al. 1994). Although my colleagues Brain and Steponaitis do not necessarily agree with me, I tend to believe that Plaquemine mound building was more dominant at these two sites. There are strong Coles Creek period components at all three sites, to be sure, but the diagnostic pottery in the mounds could just as easily relate to Coles Creek village debris that ended up as ¤ll. My use of the term village simply means that a relatively large number of people lived together on the same site in close proximity. As mentioned above, a village is different from a “town” in that there are no ®at-topped mounds or plazas in the former. It is possible that many if not most of the Coles Creek components in the Natchez Bluffs ¤t this de¤nition, as there really are few isolated Coles Creek households or hamlets in the region (see also Kidder 1998b:130; Nass and Yerkes 1995). Let us look more closely at the dating of mound construction at Feltus, Mazique, and Smith Creek. Feltus is known to have had four pyramidal mounds de¤nitely, but it could possibly have had as many as seven. Mound D, a small mound leveled by Warren K. Moorehead in 1924, yielded eight skeletons, as well as large sherds of Fatherland Incised, var. Fatherland pottery. Clearly a portion of the mound was used during or after Emerald phase times, but that does not necessarily mean it was constructed at that time (Brown 1995). We do know from LMS excavations in Mound B that that mound was built during or after terminal Coles Creek times, but it is important to note that our data come from only two 2-m squares positioned on the ®anks of the mound (Brain et al. 1994). It is quite possible that there may be an earlier Coles Creek construction deep within Mound B. The situation as it presently stands is that although a large amount of Coles

pl aquemine culture in the natchez bluffs

155

Creek pottery is found at Feltus, there is no clear evidence that Coles Creek people actually constructed any of the mounds. Having said that, it is also not clear just how much of the mounds were built by Plaquemine residents. It should be noted that the plaza between the mounds is absolutely saturated with Coles Creek pottery. Considering that plazas in Plaquemine towns are usually void of occupational material, the obvious suggestion is that Coles Creek occupation at the site occurred before the plaza came into use. I should point out, however, that Moorehead might have been responsible for spreading Coles Creek pottery in a portion of the plaza when he leveled Mound D in 1924. Let us now turn our attention to the Smith Creek site, located near Fort Adams in Wilkinson County. Smith Creek consists of three mounds arranged around a plaza, with its principal mound situated on a low bluff overlooking the Mississippi ®oodplain. Smith Creek also has its largest mound on the west side of the plaza, a structural arrangement identical to both Anna and Feltus and that Brain (1978:345) has referred to as a distinctly non–Coles Creek trait. J. Ashley Sibley, Jr. (1967) dug four large pits and a 3-m-wide trench in Mound B in the 1960s, thereby uncovering a number of skeletons, but the actual dating of this mound is unclear from his study.1 The LMS investigations at Smith Creek (Brain et al. 1994) consisted solely of surface collecting on and around the mounds. Although the LMS did indeed recover a large amount of Coles Creek pottery, as was typical of earlier surveys (Ford 1936:Figure 1), a Plaquemine occupation is revealed by the presence of Addis Plain, Anna Incised, and Plaquemine Brushed types. Test excavations made in the plaza in 1993, beneath what is now a ¤re department building, produced early Coles Creek period pottery almost exclusively (Boggess and Ensor 1993). In the absence of additional mound excavation, the verdict is still out as to whether the visible expression of Smith Creek, that is, the mound-and-plaza complex, is Coles Creek, Plaquemine, or both (Phillips 1970:920). For the Mazique site the data are more robust. This site is located along the west bank of Second Creek, north of the Homochitto River. At least three pyramidal mounds once existed at Mazique (Culin 1900:128) but only two survived when Calvin Brown surveyed it in 1916 (Brown 1926:34–35). Even at that time the larger of the two mounds was severely damaged by the ®ow of the creek. Additional destruction of this mound occurred later as a result of a museum having been established on the so-called White Apple Village (Brown 1926:34). The only professional excavations conducted at Mazique were by John Cotter and W. P. Lancaster in 1948 in their capacity as National

156

i a n w. brow n

Park Service archaeologists (Cotter 1948). In a day trip to the site they happened to observe a thick deposit of village earth lying beneath the principal mound, as seen from the creek side. Fortunately, they also excavated some pottery from this midden, which was later analyzed by Steponaitis (in Brain et al. 1994). Although the sample is small, it contains Addis Plain and Fatherland Incised sherds, certainly not the kind of material that would be expected beneath a supposed Coles Creek mound. Despite the quantity of Coles Creek pottery that occurs on the surface of the Mazique site (Ford 1936:Figure 1; Phillips 1970:948), current evidence, limited as it is, supports only Plaquemine mound construction activity. Even the Gordon site, which is the type site for the terminal Coles Creek phase in the region, is now known to have had its two mounds constructed during Plaquemine times, probably during the Emerald phase (Brain et al. 1994; Cotter 1952a; Johnson et al. 1983a, 1983b; McCullough 1981; Martinez 1981; Phillips 1970:919–920, 947–948; Walker 1981). So where does that leave us? The four prime “Coles Creek period towns,” Feltus, Smith Creek, Mazique, and Gordon, appear to have been Coles Creek villages. We cannot say with any con¤dence that these communities also had well-developed mounds and plazas. All of these sites have substantial Coles Creek occupation, to be sure, but there is no concrete evidence as yet that mound construction occurred to any signi¤cant extent in the Natchez Bluffs region during the Coles Creek period. It is of course possible that Coles Creek mound construction is hidden deep within the cores of the principal mounds, so we must keep our minds open to that possibility. There are four Plaquemine towns in the region that either lacked Coles Creek occupation altogether or had only minimal usage during this period. These are the Glass, Yokena (Brain et al. 1994; Ford 1936:234–235), Bayou Pierre (Brain et al. 1994; Brookes and Inmon 1973:19–21), and Anna sites. Seven other clearly recognized Plaquemine towns do have evidence for some Coles Creek occupation, so it is worthwhile looking at each of them to see whether they might yield evidence for Coles Creek mound construction. We have already considered the Emerald site to some extent. In terms of chronology, it is now known that Emerald began as a village during the Gordon phase, but the earliest constructional stage of the platform mound dates to the Anna phase. The later stages were added during both the Foster and Emerald phases (Brain et al. 1994). The Windsor site, located in Claiborne County, also exhibits some Coles Creek occupation. This site consists of four widely spaced pyramidal mounds situated near the edge of the bluffs overlooking both Bayou Pierre and the

pl aquemine culture in the natchez bluffs

157

Mississippi ®oodplain. LMS investigations at Windsor in 1971 were limited to two 2-m square test units, one in the summit of Mound A and one on its ®ank near the base (Brain et al. 1994). These investigations revealed strong evidence for construction and usage of the mound during both the Anna and Foster phases. There is a possibility that construction may have begun in the late Gordon phase, but that is not certain. We also have no idea of the core construction date of the mound, because our tests were unable to penetrate deep enough. Continuing to the south, the Pumpkin Lake site consists of one oval rounded mound, which once probably was pyramidal (Brain et al. 1994). It is situated to the south of Feltus and may originally have been part of that town. The LMS excavated a test pit at the base of the Pumpkin Lake mound in 1971. Although the site itself is multicomponent, as revealed by abundant pottery in the ¤ll (Toth 1988:181), mound construction is believed to have occurred during the Emerald phase. The Ratcliffe site is located at the headwaters of St. Catherine Creek. One pyramidal mound exists at this site, heavily disturbed as the result of Warren K. Moorehead’s explorations in 1924. Excavations by the LMS in 1971 revealed the mound to have been constructed during the Emerald phase (Brain et al. 1994). An Emerald phase construction is also true for at least one of the mounds at the Foster site (Brain et al. 1994; Brown 1926:34; Ford 1936:65; Ingraham 1835:2:219; Phillips 1970:948). Foster is located along the middle reaches of St. Catherine Creek in Adams County. The site consists of two pyramidal mounds separated by a plaza. Mound B, the smaller mound, located on the south end of the plaza, dates to the Emerald phase. Mound A may date earlier than Mound B, but we do not know that for sure. An LMS crew examined the nearby Henderson site in 1971. As with Foster, Henderson once had two pyramidal mounds separated by a plaza. The exact date of the one surviving mound is unknown, but from pottery in the ¤ll it is believed to have been constructed during or after the Gordon phase (Brain et al. 1994). Finally, we come back to Fatherland, which was the starting point of this discussion. Intensive excavations by Neitzel (1965, 1983) revealed Anna phase structures beneath Mound B, with construction stages dating to the Foster, Emerald, and Natchez phases. Because Mound C, the historic temple mound, lacked buildings on its early mantles, dating has been dif¤cult, but it is believed to have been constructed primarily during the Foster, Emerald, and Natchez phases. Despite evidence for some Coles Creek occupation having

158

i a n w. brow n

occurred at Fatherland, there is no indication that these people constructed any of the mounds (Brain et al. 1994; Brown 1985b:267–272, 1998d). In summary, there are four prime Plaquemine towns in the Natchez Bluffs region that exhibit major mound construction but that lack signi¤cant Coles Creek occupation. These are Glass, Yokena, Bayou Pierre, and Anna. An additional eight Plaquemine towns have some Coles Creek occupation but have yielded no evidence for Coles Creek people having been involved in any form of mound construction. These are Windsor, Gordon, Pumpkin Lake, Emerald, Ratcliffe, Foster, Henderson, and Fatherland. Finally, three Plaquemine towns seem to have been built upon what had earlier been Coles Creek villages— Feltus, Mazique, and Smith Creek—but whether the people of these last three villages were the direct ancestors of the Plaquemine people in this region has not been demonstrated. Mound construction activities would argue otherwise, but as the cores of most of these mounds have not been penetrated, we cannot say for sure. It is worth pointing out, however, that for both Feltus and Smith Creek the principal mounds were erected on the west side of the plaza, which is a decidedly Plaquemine trait (Brain 1978:345).2 With regard to building patterns, I have argued elsewhere (Brown 1985b) that there is no clear continuity between circular Coles Creek buildings and rectangular Plaquemine structures in the Natchez Bluffs. The circular singleset post structures of Gordon phase components probably developed out of the earlier Coles Creek circular wall-trench structures that have been observed in other regions, but they do not seem to continue after a.d. 1200 in the Natchez Bluffs. Plaquemine structures tend to be rectangular wall trenches almost exclusively until late in the sequence, at which time rectangular singleset post buildings make an appearance. In short, there seems to be a break in building types in the transition from Coles Creek to Plaquemine in the Natchez Bluffs region.

Conclusions For years it has been assumed that Coles Creek culture blended rather comfortably into Plaquemine culture, what with some helpful stimuli from Mississippians to the north. If this was the case for the Natchez region, all that can be said at this point is that it was not a comfortable transition from an archaeological perspective. I myself believe that we may be dealing with different people, but I should stress that this is not a view necessarily shared by my colleagues Brain and Steponaitis. One thing that we do know is that Mississippian peoples really did move about the Southeast, a lot it would seem, and in their wake they impacted many indigenous populations (Williams

pl aquemine culture in the natchez bluffs

159

1994). That much we cannot deny, but when they or their in®uences moved into a new region, the local population did not necessarily “up and die.” It should be stressed that the Coles Creek culture, as represented in the Natchez Bluffs, was not a mediocre phenomenon. These people lived in substantial villages and made a signi¤cant and lasting impression on the landscape. Although the peak expressions of Coles Creek culture played out to the north, west, and south, its representation in the Natchez Bluffs is quite impressive and these are not a people who would have left easily, if they even left at all. Did the Coles Creek people of the region give rise to the Plaquemine people, both biologically and culturally, or are we dealing with distinct populations? That is a big question but, unfortunately, it cannot yet be resolved with the data at hand.

Acknowledgments It is to be emphasized that the above thoughts are not necessarily those shared by my colleagues Jeff Brain and Vin Steponaitis. I thank them profusely for their wisdom, as I have pro¤ted immensely from suggestions and critique in our many conversations over the years. We have had many friendly arguments concerning the relationship of Coles Creek to Plaquemine culture and each of us, quite curiously, has occasionally found himself defending opposing viewpoints. In the past quarter century I myself have done a 180-degree turn in support of Plaquemine mound construction having had dominance over Coles Creek in the Natchez region. In sum, the archaeological information that is presented here is the product of a team effort, but how that information has been handled interpretively in this speci¤c study is my responsibility alone.

Notes 1. We know that Sibley was the perpetrator of the massive un¤lled excavations at Smith Creek solely from an interview made in 1971 during the LMS survey (Brain et al. 1994). In his book Louisiana’s Ancients of Man Sibley only reports having “conducted numerous training sessions and weekly programs the last ten years” as cosponsor and co-organizer of the Junior Archaeological Society (Sibley 1967:72). There is no mention of Smith Creek in that volume, beyond its being plotted on Figure 46 as a Coles Creek site. 2. Philip Phillips also wrestled with the issue of Coles Creek mound building. He offered a similar warning to that which is expressed in this essay, so it is worth revisiting his thoughts on the matter: “Re-examination of the situation in period D-C (Coles Creek) is more surprising. Of 24 small ceremonial centers listed in table 12,

160

i a n w. brow n

nineteen are mixed sites with Mississippi period components. Three more are single component sites of Coles Creek date, but our information on mound shapes is not suf¤ciently explicit. This leaves only two sites, Green River (13-O-2) and Lessiedell (20-M-4)—I hesitate to add that Lessiedell is not above reproach (cf. p. 498)—to carry the burden of the argument. Only two ‘large ceremonial centers’ were assigned to this period (ibid., table 13). Both have, according to my analysis, Mississippi period occupations, and must be left out of account. These distributions, be it noted, do not involve the southern half of the Lower Mississippi area where ‘Coles Creek’ and ‘temple mounds’ have become so ¤rmly associated in archaeological minds. They do suggest, however, the exercise of caution in accepting assignments of unexcavated rectangular platform mounds to the Coles Creek period, even in this part of the area where the proposition is not in serious doubt” (Phillips 1970:964–965).

9 The Outer Limits of Plaquemine Culture A View from the Northerly Borderlands Marvin D. Jeter

Lower Mississippi Valley (LMV) Plaquemine is one of several late prehistoric, multiregional cultural variants that archaeologists have generally regarded as distinct from, but interacting with, contemporary Mississippians. Coeval non-Mississippian archaeological cultures include Caddoan in the Trans-Mississippi South (TMS), plus Fort Ancient, Oneota, and Plains Village in other areas. These cultures are generally rather vaguely de¤ned and are classi¤ed mainly on the basis of ceramics. Identi¤cations are often problematic in regions close to Mississippian or other cultural variants. This is certainly true for the situations summarized here.

De¤nitions, Concepts, and Problems Interestingly, the impetus for this volume came from denizens of the Louisiana bottomlands, as did the original Plaquemine concept. As Rees and Livingood note in Chapter 1, Plaquemine was ¤rst published (as “Placquemine”!) in two charts derived from James A. Ford’s Louisiana work (Ford and Willey 1941:Figures 2 and 5), but it was not really described until the 1950s (Ford 1951:85–89, 100ff, 125, 127–129; Quimby 1951, 1957; cf. Gregory 1969; Neitzel 1965). By contrast, major publications of the Harvard-based Lower Mississippi Survey (LMS) downplayed the Plaquemine concept. Ford and his cohorts operated in a sort of parallel universe, distinct from their LMS colleagues and successors, who distanced themselves from Ford on various issues. Instead, Phillips (1970) wrote only of Coles Creek and Mississippi as culture periods, overextending both spatially. He referred variously to “Plaquemine culture” but avoided the P-word in his minimal discussion of his southeast Arkansas Bellaire phase (Phillips 1970:944; see below). In a similarly brief Fitzhugh phase section (Phillips 1970:945), he referred to relationships with “the Plaquemine phase,” also mentioned elsewhere (Phillips 1970:50, 560), but

162

m arv in d. jeter

in the end did not actually name such a phase. Phillips (1970:946–947) also suggested that Plaquemine was an outgrowth of Ford’s earlier (1936) Caddoan complex, but actually it was Ford’s homegrown interpolation between his Coles Creek and Natchez periods, displacing the “Caddoan” concept that he had erroneously intruded into those regions ( Jeter et al. 1989:205). In the Lake George report, the drama was recast in terms of interactions between Coles Creek and Mississippian cultures, which again got headings and period names, while Plaquemine did not (Williams and Brain 1983:391, 405, 408). In the big-picture discussion (Williams and Brain 1983:408ff ), Plaquemine was only mentioned in passing in a section entitled “Ecumene (a.d. 1200–1400).” A discussion of the Winterville, Routh, and Anna phases did not mention their Plaquemine cultural af¤liation (Williams and Brain 1983:413–414). Instead, following Brain’s (1969) lead, it concentrated on “Cahokian and continuing (but unspeci¤ed) [sic] Mississippian in®uence” on Coles Creek culture. Plaquemine was mentioned only in the next few paragraphs, once with “culture” in quotes, after which it was “de¤ned as Mississippianized Coles Creek” (Williams and Brain 1983:414). This report also whittled away at the time span and artifacts allotted to Plaquemine, by subtracting a key phase. Whereas Brain’s (1969:Table 20) Winterville-based dissertation and Phillips’s (1970:558–560, 945, Figure 450) tome had included Crippen Point as an early phase of Plaquemine culture, dating from a.d. 1000 to 1200, Williams and Brain (1983:373) suggested an extension of the Coles Creek period to include Crippen Point, with its basic Addis 1 set of 10 ceramic varieties as a terminal expression of Coles Creek culture, dating about a.d. 1100–1200. The ¤rst phase of Plaquemine culture became the Winterville phase, starting around a.d. 1200, featuring the Greenville set of grog/shell-tempered types, plus the Yazoo 2 set of shell-tempered types (Williams and Brain 1983:Figures 9.4, 10.16, 11.4, and 12.1). So at least in the Lower Yazoo Basin, the Fordian concept of Plaquemine culture was gutted, even deprived of its basic Addis ware (Quimby 1951:107– 109), which had already been subsumed under the northerly type Baytown Plain as var. Addis by Phillips (1970:50–51). Instead, Plaquemine was characterized as having primarily northerly in®uenced wares with shell tempering! However, Addis ware still appears primarily associated with the Plaquemine tradition farther south, on into historic Natchezan culture (Brown 1985a:Table 2). De¤nitions and criteria for Addis ware have varied signi¤cantly ( Jeter et al. 1989:206; Ryan 2004:91–93). Recently, they have emphasized heterogeneous tempering agents, mainly crushed-sherd grog, sometimes with shell, more

outer limits of pl aquemine culture

163

rarely bone, and often dark apparently organic material, in a relatively homogeneous paste. Aspects of form (carinated bowls, open-bowl shapes, deep jars, bottles) are also semidiagnostic, as are several decorated types. However, a new and more rigorous analysis, using quantitative standards derived from geology (Ryan 2004:89–100, Figures 7-1, 7-3, 7-4), suggests that in the southern Tensas Basin at least, the main sorting criteria for var. Addis should be grog size (“medium-¤ne”) and especially grog quantity (30 percent or more of the paste). Thus de¤ned, Addis did not become a major presence until early Plaquemine times at the Hedgeland site, whereas similar varieties with signi¤cantly less temper were more characteristic of late to terminal Coles Creek phases (Ryan 2004:91, 93, 95–96). In Ryan’s (2004:97) analysis, the Greenville variety is subsumed under the Bell Plain type and also associated with the early Plaquemine component, though it is quite rare in that southerly location. In two LMS publications presaging the Lake George report (Belmont 1982b:Figure 3; Belmont and Williams 1981:Table 1), the phrase “Transitional Coles Creek” was introduced without discussion as a subperiod between late Coles Creek and early Plaquemine/Mississippian or Mississippi I, dating from circa a.d. 1100 to 1200. It included phases such as Crippen Point, Preston in the Tensas, Gordon in the Natchez Bluffs, and Spring Bayou in the Lower Red region. Another LMS report (Brown 1985b:Figure 2) showed these phases (plus St. Gabriel in the Baton Rouge region) in the Coles Creek period but indicated a Plaquemine af¤liation in their latter portions. In an overview, I emphasized a “deeply rooted continuum” in material cultures and expanded the phrase to “Transitional Coles Creek–Plaquemine” to underscore this ( Jeter and Williams 1989:165ff, 172, Figure 17, 205ff ). LMS publications have focused on Plaquemine along the Mississippi River, especially east of it. Hally’s (1972) lengthy Harvard dissertation, which focused on northeast Louisiana, was an exception, but it has never been published and has been downplayed, like Ford’s work, in the LMS literature. Meanwhile, in southeast Arkansas, which Phillips (1970:861) erroneously thought was basically “Caddoan” territory, other Plaquemine research was going on, also mostly overlooked in the LMS literature (but cf. Kidder 1991).

Introduction to Southeast Arkansas This chapter is mainly about northern to northwestern Plaquemine variants, as seen from southeast Arkansas (Figure 9.1). The eastern boundary of my research territory is the Mississippi River. Along it, here and in northeast Louisiana, a few important sites with Plaquemine af¤liations are known

164

m arv in d. jeter

Figure 9.1. Southeast Arkansas, showing major streams, county boundaries, and locations of some key sites mentioned in the chapter.

but have not been assimilated into a phase. They are summarized in the next section. At a casual glance, the rest of our “delta” lowlands (generally around 100 to 125 feet, or 30 to 40 m asl) west of the Mississippi appear to be its ®oodplains, but in fact they are mostly ancient, abandoned Arkansas River meander belts. Plaquemine culture is fairly well represented, especially on the most recent belt, along Bayou Bartholomew.

outer limits of pl aquemine culture

165

My northern boundary is the recent to modern Lower Arkansas meander belt. Plaquemine culture apparently never expanded north enough to encroach on these lowlands, but some Plaquemine in®uence is evident in the Menard locality on the north side of the river ( John House, personal communication, 2003). The western boundary is the greater Ouachita Valley, occupied by the Ouachita River and tributaries such as the Saline River. The Felsenthal National Wildlife Refuge bottomlands contain the lowest land and water in Arkansas (around and even below 65 feet, or 20 m asl). Again, there was a signi¤cant Plaquemine variant, with Caddoans beyond to the west and northwest. My southern boundary is the Louisiana state line. I will leave the Plaquemine variants below there largely to my Louisiana colleagues but will poach a bit in that direction. The so-called uplands (usually 200 to 400 feet, or 60 to 120 m asl) in the midst of my territory are formed on West Gulf Coastal Plain and old high terrace deposits. They are mainly covered by timber and pastures and are poorly known but probably were occupied by minor offshoots of Plaquemine groups based to the east and west. Two Willey-and-Phillips-style archaeological regions, Bartholomew-Macon and Felsenthal, are of relevance here. The former region, named after two major bayous in former Arkansas meander belts, includes the delta lands west of the Mississippi and the eastern part of the uplands ( Jeter 1982:80, Figure 6-2). To the south, the Upper Tensas and Boeuf regions are largely in Louisiana. The Felsenthal region includes the western part of my territory and extends into poorly known localities in Louisiana (Schambach 1981:105). Farther south, the Ouachita Valley has not been systematically explored, although part of it was included in the Boeuf Basin region by Belmont (1985). No of¤cial region name has been designated along the Ouachita Valley proper by Louisiana archaeologists, but I have called it the Lower Ouachita region ( Jeter et al. 1989:29, 211–212). Kidder (1990a:52ff ) has also used this term. Four relevant Willey-and-Phillips-style phases have been de¤ned, or at least named, for our southeast Arkansas regions. They are the Bellaire and Bartholomew phases in the Bartholomew-Macon region and the Gran Marais and Cypress Swamp phases in the Felsenthal region. Gran Marais is the best documented and Bartholomew is also fairly well known. Cypress Swamp is a promising newcomer but the Bellaire phase is a dubious concept, questionable at best. A relatively well-de¤ned sequence exists for the Tensas Basin (Hally 1972),

166

m arv in d. jeter

but only preliminary data have emerged for the Boeuf Basin and Lower Ouachita regions (Belmont 1985; Jeter et al. 1989:211–212, Table 20; Kidder 1990a:52ff, Figures 2–7).

Mississippi River Sites The lands along the Mississippi in southeast Arkansas are very poorly known, since the batture between the river and levee is heavily vegetated. However, one site (3DE24) in this vicinity was potted by a collector in the 1960s, allegedly yielding a remarkable pipe that has been called a “Corn Goddess” or “Corn Woman” ef¤gy (Reilly 2002, 2004:124–125). It is questionable whether the plants depicted actually represent corn stalks (Fritz and Kelly 2004). The pipe may well have been made of Missouri ®int clay at or near Cahokia (Reilly 2004:131–132; cf. Emerson and Hughes 2000). Stylistically similar items were made of this material during the Stirling phase (ca. a.d. 1100– 1200) in the Cahokia vicinity and may have been exported to distant sites (Emerson and Girard 2004; Emerson et al. 2003), though some may have been interred with burials somewhat later. Alternatively, they may have been retained at Cahokia until after it declined and subsequently disbursed to Caddoan and southeastern recipients in the mid to late 1200s (Emerson et al. 2003:306, 308). The alleged 3DE24 ¤nd thus could be another link between Cahokia and this segment of the LMV during the transitional subperiod or slightly later during early Plaquemine times. A few other ¤nds from this vicinity, including a stone “kneeling prisoner” pipe, also may point in this direction (notes and photos on ¤le at the Arkansas Archeological Survey’s University of Arkansas at Monticello Research Station). Near the southeast corner of Arkansas is the Lakeport mound site, in a bend of the Mississippi. It has 11 mounds, ¤ve arranged around an apparent plaza. One is about 10 m high and roughly rectangular. Only brief tests have been done here. No major midden was found, suggesting a possible vacant ceremonial center. The sparse artifact sample suggests late or transitional Coles Creek, Plaquemine, and late Mississippian occupations, with the latter perhaps representing the major component (Rolingson 1971b). Some 50 km to the south, on a Mississippi oxbow in northeast Louisiana, recent work at the Lake Providence Mounds suggests occupation during the Coles Creek to Plaquemine transition, around a.d. 1150–1250. Exotic items suggest northern contacts, including the American Bottom and Cahokia (Wells and Weinstein, this volume). North of Lake Providence at Panola, Louisiana, a low mound was bulldozed in 1965, revealing burials. A collector uncovered at least 24 skeletons,

outer limits of pl aquemine culture

167

mainly extended-supine but also including four bundle burials. The collector also recovered more than 60 vessels of which 44 were retained. These ¤nds were summarized by a leading amateur ( Jones 1984), who was advised by John Belmont of the LMS. They appear to represent terminal Coles Creek to Plaquemine use or uses of the mound, with a few hints of a very late Mississippian presence ( Jones 1984:142ff, Figures 1–5). These sites, along with Winterville, suggest that along the Mississippi itself the Plaquemine situation was more elaborate, with more connections to northerly Mississippian cultures than along the old Arkansas River meander belts in the southeast Arkansas interior. However, more intensive surveys and excavations would be required to document the settlement-subsistence system along the Mississippi.

Deconstructing the “Bellaire Phase” Pipe Dream Bellaire, the ¤rst designated but least documented phase under scrutiny here, was named as a “particularly tentative” concept by Phillips (1970:944–945), largely to create a cultural home for a remarkable serpent-cat stone pipe (Figure 9.2). This pipe was found in 1886, either by a farmhand plowing or during a dig into a mound on Crooked Bayou in northernmost Chicot County. In 1931 it was purchased by Judge Harry J. Lemley of Hope, Arkansas. His collections were later acquired by the Gilcrease Museum of Tulsa, Oklahoma. Phillips regarded Bellaire as related to coeval phases of Plaquemine culture, but there are a number of problems with this. Lemley also sponsored a 1934 expedition to southeast Arkansas. At Bellaire, a “badly damaged” mound, “approximately 80′ in diameter and 12′ high,” in the speci¤ed location was brie®y tested, producing only a few sherds of “Tunica ware” (late shell-tempered) pottery (Lemley and Dickinson 1937:31). The mound remnant was revisited in 1971 and found to be only about a foot high and 6 to 10 feet across, reportedly having been plowed down gradually. Nothing was found on or near the mound, and the site has not been revisited by professionals. The Bellaire pipe very closely resembles one reportedly found at Moundville, Alabama, around 1860 (Moore 1905:131, Figures 1–3), now in the Peabody Museum at Harvard (a replica is lent to the LMS’s C. B. Moore Award winner each year). A very similar specimen was excavated at Moundville by Moore himself (see below). Another was reportedly found at Lake George around 1900 (Williams and Brain 1983:255, Figure 7.32a–c). Several similar pipes depicting mythical animals or humans are known, including those from the Selsertown or Emerald Mounds site just north of Natchez and one report-

168

m arv in d. jeter

Figure 9.2. The Bellaire “serpent-cat” pipe (Arkansas Archeological Survey Negative No. 693878).

edly found at or near Hot Springs, Arkansas (Brown 1926:256–264, Figures 218–226). The Bellaire pipe was originally described as made of felsite, an igneous rock (Lemley and Dickinson 1937:31). However, Vincas Steponaitis, who has studied similar pipes from Moundville and other sites, states they were all made of Glendon limestone, best known from outcrops in the vicinity of Vicksburg (Steponaitis, personal communication, 2004). He states that Moundville burials in general date between a.d. 1200 and 1500 and those with more elaborate or exotic artifacts tend to date between a.d. 1300 and 1450. Moore’s (1905:237–240, Figures 165–171) own ¤nd was in a burial with a remarkable stone ef¤gy bowl and several unillustrated items. Steponaitis suggests that these materials date in the a.d. 1400–1450 range. A similar specimen from Emerald probably dates to the Foster-Emerald phase transition, circa a.d. 1450 to the early 1500s, still within the Plaquemine tradition in that region.

outer limits of pl aquemine culture

169

These comparisons cast doubt on the association of the Bellaire pipe with Plaquemine culture in southeast Arkansas. Here, a transition to Mississippian culture seems to have occurred by about a.d. 1400 ( Jackson 1992; Jeter et al. 1989:171ff; Rolingson 1976, 1993). It may have been made by Plaquemine people near Vicksburg and traded to Mississippians at Bellaire, where the 1930s test produced only shell-tempered pottery. Phillips (1970:944–945) did not discuss other artifacts of the Bellaire phase, but merely compared it to his Crippen Point, Mayersville, and Fitzhugh phases, that is, Plaquemine culture as he saw it or transitional Coles Creek in the revised LMS scheme. Major ceramic types of those phases were listed by Phillips (1970:558, 945), but so far there is very little evidence for any of them from the Bellaire locality. Alba Stemmed arrow point varieties, rare Bayou Goula “¤shtail” points, and bone and antler points have been reported for the Crippen Point and Mayersville phases (Brain 1989:105; Williams and Brain 1983:336, 338). Of these, only Alba-like points are known from southeast Arkansas. The Bellaire locality is a blank with regard to well-provenienced points. Phillips (1970:944) named four other southeast Arkansas sites as having Bellaire components. Almost no further ¤eldwork or analysis has been done, but in the light of present knowledge only one of them, Alma Brown, appears to approach even minimal evidence of a Plaquemine component. It once had at least six mounds, of uncertain dates (Lemley and Dickinson 1937:37ff ). It was revisited by the LMS (Phillips et al. 1951:Table 1) and was discussed by Phillips (1970:892–893, 916), who remarked that it “has everything from the Marksville period on,” which might imply a Plaquemine component. The LMS Archives (Steponaitis et al. 2002, site ¤les: 17-K-7) indicate that over 3,800 sherds were collected in 1940. A few are of types that might have come from a Plaquemine component but are not diagnostic. In summary, the entire Bellaire phase of Plaquemine culture might well be taxonomically sunk unless de¤nitive evidence of Plaquemine occupation can be found in this locality. Relevant components at some of its putative sites might instead be included in a Bellaire phase of late Mississippian culture or lumped into the late prehistoric to protohistoric Mississippian Hog Lake phase.

The Bartholomew Phase To the west in the same region, Rolingson (1976) de¤ned the Bartholomew phase of Plaquemine culture on the basis of extensive surveys and tests along and near Bayou Bartholomew. Well ahead of the LMS revision, she suggested an overall chronology of about a.d. 1200–1400 (Rolingson 1976:117–119).

170

m arv in d. jeter

However, the artifact assemblage included some items that probably belong to the transitional subperiod of about a.d. 1100–1200.

Ceramics The ceramics included mainly Addis-like Baytown Plain with some shelltempered Mississippi Plain and rare bone-tempered ware (Morris Plain), which hints at Caddoan in®uence. Relatively scarce decorated sherds included varieties of Mazique Incised and, more rarely, Harrison Bayou Incised as rim decorations. Evansville Punctated is apparently almost diagnostic of this period in these regions. Also present were a Hardy-like variety of Coles Creek Incised, Plaquemine Brushed, Avoyelles Punctated, Hollyknowe Ridge-Pinched, L’Eau Noire Incised, and early varieties of shell-tempered Winterville Incised (Rolingson 1976:115–116, 1993). A ¤ner-grained sequence of these types, or varieties of them, might be developed to subdivide this phase (see the Gran Marais-Cypress Swamp section).

Ashley versus Homan Points Rolingson (1971a; cf. Rolingson 1976, 1993) de¤ned the Ashley arrow point type, distinguished mainly by its rounded, “bulbous” stem, as common on Bartholomew sites. Subsequently, she also recognized this type in the Felsenthal region (Rolingson 1981:73ff ). However, her illustrations included points that converge with the somewhat similar Homan type. Although individual specimens no doubt intergrade, Ashley points may tend to be later and more closely associated with northern Plaquemine culture. The Homan type was de¤ned by Wood (1963), who noted that corner notching produced “a ®ared, fan-shaped stem,” which often has distinct corners where its edges meet the “evenly convex” base. He suggested a Coles Creek cultural af¤liation but gave an estimated date range of a.d. 1000–1300, which might extend them too late. According to Brown (1976:79, 92), a more widely published Homan description (Perino 1968:34, Plate 17) confused the situation by including illustrations of Ashley-like points with more bulbous, rounded stems. Similar overlaps are present in Rolingson’s (1971a, 1981) and Webb’s (2000:16) illustrations of Ashley and Homan points. In one report, Rolingson (1981:73–75) included as Ashley points some specimens in which “the stem is more fan-shaped than is typical of the type.” Others have not been consistent with regard to criteria for these types (cf. illustrations in Weinstein and Kelley 1984). Weber (1973:34–36) sorted them mainly by presence or absence of shoulder barbs. Her other criteria were shared among types and she did not emphasize stem shape. Waddell and King

outer limits of pl aquemine culture

171

(1990:64), citing “vague or overlapping sorting criteria” and using an unwieldy computerized database, simply lumped the Ashley, Alba, and Homan types as an “ASALHO cluster.” In contrast, I prefer a strict, literal emphasis on the rounded, bulbous stem of the Ashley type versus the fan-shaped basal-cornered stem of the Homan type, with an “in-between” or “Homan-Ashley” category for specimens with ambiguous stems. Homan points discussed and illustrated by Hemmings (1982:229, Figure 58n, o, Table 20) appear to meet these criteria. Hemmings suggested that Ashley points may have been derived from Homan points, perhaps overlapping from circa a.d. 1000 to 1200. There is evidence for earlier occurrences of Homan points than Ashley points, strictly speaking. At Toltec, Hoffman (1998:68–70) de¤ned a broad “Rockwall cluster” of arrow points, including two specimens that “generally ¤t within the Homan type” (Hoffman 1998:69, Figure 72o, p) and meet my preferred criteria. This suggests that they were in use as early as a.d. 950–1050, if not earlier (Rolingson 1998:100, 107). Webb (2000:16) estimated the time range of Homan points at a.d. 800–900 but also said that they were found with late Coles Creek ceramics, implying a somewhat later date range. Although her typology may be questionable, Weber (1973:61–65, 73) found Homan-like points in strata that she considered early to late Coles Creek and Plaquemine. She found Ashley-like points only in the late Coles Creek and Plaquemine strata. She also stated that most arrow points of all types from Plaquemine deposits were found in their lower levels, hinting at their later replacement by bone or antler points, as in the Plaquemine heartland (Hally 1972:335). But there is at least some evidence that Ashley points may still have been in use after a.d. 1400 (Hemmings 1982:191–193).

Sites and Settlement Data Rolingson’s (1976) Bartholomew phase data were from 51 sites divided into four types: 6 small ceremonial centers, 14 hamlets, 29 house sites, and only 2 camp sites. The centers were spaced about 3 to 11 km (ca. 2 to 7 miles) apart along Bayou Bartholomew and nearby streams. They were relatively small, despite the ®urry of major mound building that was going on to the south and east just before, and especially after, a.d. 1200. They had only one or two mounds, with no evidence of plazas. The largest mounds were probably ®at topped but only medium sized at best. Midden deposits, sometimes extending more than 100 m along low natural-levee rises, were generally nearby. Lacking extensive excavations, it is unknown whether these represent contemporary houses or palimpsests of separate occupations involving only one or a few

172

m arv in d. jeter

houses at any time, and the same uncertainty applies to the hamlets. Nor is it known whether occupations were seasonal or year-round. The hamlets are generally small, with no evidence for nucleated (let alone forti¤ed) villages with numerous structures as in coeval Mississippian sites to the north (cf. Morse and Morse 1983:237ff ). Five sites with Bartholomew phase components were tested (Rolingson 1993). The Henderson site, the southernmost center, had one mound about 4 m high with a base ca. 40 by 40 m and occupation zones extending about 60 m east and 90 m west of it (Rolingson 1993:45ff ). Reportedly there was once also a low mound about 60 m away. The McArthur site, a small hamlet or house site, produced post molds from two overlapping circular structures of individual-post construction (Rolingson 1993:61ff, Figure 29). Ceramics were dominated by the usual types, with no diagnostics of other periods. Lithics included eight Ashley-like or Homanlike points (cf. Rolingson 1993:Figure 34a). Seven burial pits contained at least eight individuals. Importantly, they were analyzed and found to represent three partially disabled adults plus ¤ve infants, suggesting that perhaps “only peripheral members [of this society] were accorded habitation site burial” (Rose and Powell 1993:71). The adult burials were extended, two with heads to the south. One was buried with a Baytown Plain jar and a turtle-carapace cup. One infant had a Mississippi Plain carinated bowl and another a Mazique Incised rim sherd. The Currie site (Rolingson 1993:85ff ), a low midden mound about 12 m across on a low rise in a swampy zone, was occupied during several periods. Mixed components made analysis dif¤cult. The Ellis Pugh site (Rolingson 1993:115ff ) was similarly multicomponent and disturbed by deep plowing. The Wilson Brake site (Rolingson 1993:27ff ) had only a minor Bartholomew component in the plow zone.

Subsequent Bartholomew Phase Research Salvage of Mound A at Boydell, Rolingson’s northernmost Bartholomew phase site, revealed a middle Coles Creek period Stage I, a middle to late Coles Creek Stage II, a late Coles Creek to Plaquemine Stage III, and hints of later stages disturbed by earth-moving equipment (House and Jeter 1994:73–77). Eight disturbed burials in the upper mound contained the remains of at least 15 individuals with basically Plaquemine ceramics, including Addis Plain varieties, Coleman Incised, L’Eau Noire Incised, and Mississippi Plain. A smashed Beldeau Incised vessel may have come from a slightly earlier burial (House and Jeter 1994:22f, 26ff ). One large, nicely made plain open bowl from a Boydell burial is made of

outer limits of pl aquemine culture

173

Figure 9.3. Addis Plain bowl with deeply notched rim from the Boydell site in southeast Arkansas. Diameter of rim exterior, about 40 cm.

grog-tempered Addis ware with possible bone or shell inclusions (Figure 9.3). It has a deeply notched rim reminiscent of the edges of circular sandstone palettes used for grinding pigments at Moundville (Moore 1905:Figures 4, 5, 19, 65–66; Whitney et al. 2002). Fragments of similar palettes have also been found at Lake George (Williams and Brain 1983:263–266, Figure 7.41). Steponaitis (personal communication, 2004) states that similar bowls of the Anna Interior Engraved (or Incised) type from the Natchez vicinity may also have held pigments. An intensive catchment survey on relatively high ground within a 2-km radius of Boydell found seven small hamlet or household sites with probable Bartholomew components, as well as possibly four others ( Jeter et al. 2007). Artifacts included most of the usually expected ceramics. Ashley points were fairly common, especially at two moderate-sized household sites. Because of access problems, the total intensively surveyed area was only about 8.5 km2. Rolingson’s (1993:Figure 5) map shows about 350 km2 of relatively high land in her project area, by my estimate, so there might easily be 250 to 300 Bartholomew phase sites on the higher land in her area. Additional sites of this phase, like Currie, undoubtedly exist on the lower ®ood-prone lands. Still more surely exist beyond her area’s con¤nes.

Bartholomew-like Sites to the North In a preliminary paper, Rolingson (1970) noted the lack of mounds for about 50 km (ca. 30 miles) above Boydell, after which other mounds were associated with “quite different” cultural materials. We now refer to this as the Tillar complex or phase of late to protohistoric Mississippian culture ( Jackson 1992;

174

m arv in d. jeter

Jeter 1986; McKelway 1990), but we also know now that there are Plaqueminelike components in that locality. About halfway between Boydell and Tillar, my surface surveys found several small sites with varying proportions of grog- and shell-tempered ceramics, possibly representing various episodes during a Plaquemine-Mississippian transition, rather than overlapping separate occupations. One site produced a Mill Creek (southern Illinois) chert hoe chip with high polish on one surface. Just north on the bayou, intensive salvage at the Ables Creek site recovered predominantly Tillar burials and ceramics but also a few sherds from an ephemeral Plaquemine component ( Jackson 1992:86). At nearby Harrell Bend on the bayou, a small-scale salvage effort produced mostly shell-tempered sherds but also several late Coles Creek to Plaquemine types, plus an Ashleylike point (Wesolowsky 1974:20–31). At the Taylor site, on the bayou in northeastern Drew County, there were once four mounds around a plaza. Testing of two of them revealed early Coles Creek construction starts in both ( Jeter, ed. 2007; Jeter et al. 1991). Only one, Mound 3, had a substantial above-ground remnant. Its ®at-topped Stage III surface was occupied during late or transitional Coles Creek times, with an archaeomagnetic date of “late 1100s” (Dan Wolfman, personal communication, 1993) associated with Vicksburg-like ware, Caddoan Holly Fine Engraved sherds, ®akes of probable Crescent Quarry (near St. Louis) chert ( John Kelly and Neal Lopinot, personal communications, 2001), and one ®ake of Mill Creek chert, plus an arc of post molds from a circular structure and a smudge pit containing a number of charred corncobs. The ®at Stage IV burned surface produced an early 1200s Plaquemine period date (Wolfman, personal communication, 1993). Stage V had lost its upper surface but yielded some later materials. Another Taylor mound was allegedly the source of a pipe fragment that has been identi¤ed as being of Missouri ®int clay (Bruce Smith, personal communication, 1982). My limited surveys farther up Bayou Bartholomew found additional Plaquemine-like sites into central Lincoln County, as did a contract project (Giardino 1979:126). A few generally small mounds are known up to this vicinity, but they have not been excavated under control. Surface collections from the Yorktown Mound, on the bayou north of Star City, consist mainly of Coles Creek ceramics but include a transitional variety of Coles Creek Incised and an Evansville Punctated rim.

The “Louisiana Bartholomew” Locality Probably related are sites along Bayou Bartholomew in northeast Louisiana, including the Matheny mound group (Kidder 1990a:67, 1994). It was mapped

outer limits of pl aquemine culture

175

in 1880 as including three mounds in a nearly linear con¤guration, but only two were visible by the early 1900s (Kidder 1994:138–139, Figures 2 and 3). Matheny Mound A is about 7 m high and over 40 by 40 m at the base, with a ®at top ca. 15 by 15 m. Construction started, as at Taylor, early in the Coles Creek period over a Baytown midden. Second and possibly third stages may have been added during a Plaquemine occupation, and the fourth and ¤nal stage represented a Mississippian component (Kidder 1994:144–145). Mound B was lower and severely damaged. It apparently had a similar origin and “abundant Plaquemine remains” on its plowed surface, along with a few Mississippian sherds (Kidder 1994:146). Overall, Kidder (1994:149) stated, “the Plaquemine culture is better represented than any other component” despite the lack of undisturbed deposits, and he added that C. B. Moore had excavated a Plaquemine cemetery west of the mounds. Moore (1909:166–169) reported 13 adult burials, in extendedsupine position with varied orientations, plus two layers of massed bones, a bundle burial, and several isolated skulls, with a total of 14 ceramic vessels. Only one vessel was obviously shell tempered.

Subsistence, Agriculture, and Maize Subsistence data are scarce for this phase, because of lack of intensive excavations and ®otation. At the McArthur site, seven irregular pits or midden remnants produced the remains of deer and other mammals, turkey, turtles, and ¤sh, plus some shell¤sh and nut hulls (Rolingson 1993:80–82, Table 26). Rolingson’s (1993:42, 106–112, 130, Tables 20, 29, 31) other Bartholomew sites produced some similar faunal remains and a few nut hulls, but interpretations were limited by disturbance and poor preservation. Although Williams and Brain (1983:408) suggested that Coles Creek culture was based on maize agriculture, subsequent research has not borne this out. Instead, it now appears that wild plant and animal foods, plus the Native Eastern domesticates, were the basis. Maize was present by the eighth century a.d. but apparently not important until the 1200s or later in these latitudes (cf. Lopinot 2003). Our late-twelfth century a.d. maize at Taylor is thus not surprising and is consistent with recent ¤ndings elsewhere. Maize has been found at Toltec in small amounts in a probable feasting context as early as the late-eighth century a.d. and less abundantly in a feature assemblage dominated by Native Eastern cultigens, along with some wild plant remains dating from circa a.d. 925 to 1025. Maize was present at related sites by a.d. 700–900 northwest of Little Rock and between a.d. 960 and 1005 closer to Toltec (Rolingson 1998:105–106, 2003:23, 29; C. Smith 1996). Overall, the context was one of “considerable reliance on, and experimenta-

176

m arv in d. jeter

tion with, cultivated crops in the 750–950 period” (Rolingson 2003:23). Maize of the Eastern complex, which apparently evolved in the Midwest rather than in Mesoamerica, has been identi¤ed at the Caddoan George C. Davis site (Ford 1997:107), possibly as early as the late-ninth century a.d. (Story 1997:96). Maize has also been found in a pre-Caddoan context, more de¤nitely in the 800s, at the Ray site in the Red River valley (Bruseth 1998:53). Over 3,000 fragments of maize were found at Lake Providence in northeast Louisiana in contexts dating circa a.d. 1150–1250. It was present in about 75 percent of some 140 sampled features and 80 midden samples (Roberts 2005; Weinstein 2005; Wells and Weinstein, this volume). Farther south in the Tensas, maize became more common between a.d. 1000 and 1200, a trend that accelerated after a.d. 1200 but may not have reached the intensi¤cation stage until the 1300s (Fritz and Kidder 1993:8, 9–11). In the southern Tensas Basin, work at the Hedgeland site has indicated that “both wild and domesticated native seed crops . . . were heavily utilized in the early Coles Creek period. . . . Maize occurs in minor amounts beginning in the late Coles Creek period, but does not play a major role in subsistence until the Plaquemine occupation” (Ryan 2004:iii; see also Roberts 2004:217–219, 224–226). This calls into question Kidder and Fritz’s (1993) hypothesis of a nonagricultural food base for early Coles Creek heartland populations. Instead, at Hedgeland, “a substantial degree of native seed cropping was underway” as early as the late-eighth century a.d. (Roberts 2004:225). These ¤ndings reinforce part of Rolingson’s (1976:119) suggestion that Bartholomew subsistence “was based on corn-beans horticulture supplemented by wild resources including deer, various nuts, and other bottom-land ®ora.” However, no domesticated beans have been identi¤ed in these regions. Indeed, they are absent from Cahokia’s peak period and apparently were not signi¤cant in the eastern United States until after a.d. 1250 or 1300 (Hart et al. 2002; Hart and Scarry 1999). In their absence, acorns may have provided an adequate nutritional supplement to maize (Ford 1997:104–105).

The Gran Marais and Cypress Swamp Phases Schambach (1979:30, Figure 3.1) de¤ned the Gran Marais phase as representing Plaquemine culture in the Felsenthal region. However, he soon changed its cultural af¤liation to unknown due to proximity to Caddoan cultures and estimated its date at circa a.d. 1250–1350 (Schambach and Rolingson 1981:189– 190, Table 18). Later, with more evidence in hand, he cautiously called it “a northern regional variant of the Plaquemine culture, although that concept is still so nebulous that close comparison is impossible” (Schambach 1990:122,

outer limits of pl aquemine culture

177

Figure 78). From my perspective, I have leaned toward the Plaquemine designation for this phase ( Jeter et al. 1989:210; Jeter and Early 1999:52–55). It might be seen as essentially continuous with the Bartholomew phase, if not for the poorly known intervening uplands. In contrast to Rolingson’s sequence, which placed Bartholomew continuously between Coles Creek and Mississippian phases, the ¤ner-tuned Felsenthal sequence includes some gaps. At Bangs Slough, one component was made the basis of a new Cypress Swamp phase of the “Late Coles Creek–Early Mississippi period,” estimated at about a.d. 950–1050 on the basis of ceramics and 11 radiocarbon dates (Schambach 1990:116–118, Table 32). On one chart it was directly aligned with a Plaquemine cultural af¤liation but also with a date of a.d. 1000 (Schambach 1990:Figure 78). The text stated it was “clearly ancestral” to the Gran Marais phase and compared it to “Coles Creek and Plaquemine adaptations to the east and south” (Schambach 1990:120–121). Gran Marais itself was again restricted to circa a.d. 1250–1350 and the protohistoric Mississippian Caney Bayou phase was placed around a.d. 1550–1700, leaving two-century gaps before and after the Gran Marais phase (Schambach 1990: 121–122). However, this chronology itself is actually about 20 years old and may need revision. It used the Phillips (1970) period terminology and chronology, probably because the Bangs Slough contract report had been prepared before the Lake George revisions were available (Schambach 1985), and the published report closely followed the contract report’s format. Kidder (1991:81–83) remarked that Bangs Slough provided “the strongest, most consistent set of [radiocarbon] dates in the central and southern Lower Mississippi Valley.” He added that “the Felsenthal sequence is in many ways more secure than that originally de¤ned” in the LMV to the east, but he also noted that the calendrical dates needed calibration. Recent recalibrations for this time range have suggested that calendrical equivalents should be moved up to 50–100 years later (Knight 2003:135), so the Cypress Swamp phase might actually date circa a.d. 1000–1200.

Artifacts This section will be dominated by discussions of ceramics, but I will note that Ashley points are fairly common. Some Homan and Alba-like points also may be associated, as discussed above (Hemmings 1982:229, Figure 58; Rolingson 1981:73–75, 97, Figure 15; Waddell and King 1990:64, Figure 52). The Gran Marais phase is characterized mainly by, once again, Baytown/ Addis grog-tempered plainwares. The decorated ceramics, in®uenced by Cad-

178

m arv in d. jeter

doan practices, have such complex combinations of rim and body decorations, at times involving different decorative techniques on the two ¤elds, that the rim-oriented Phillips type-variety system proved inadequate. Schambach (1981) consequently devised his innovative “descriptive” classi¤cation system. Some level or levels in its hierarchy of categories can be applied to virtually all decorated sherds, whereas their classi¤cation is often ambiguous for previously established types of the LMV or TMS. Nevertheless, the Felsenthal region is considered part of the LMV (Schambach 1981:103–104) and useful if tentative comparisons can be made to some familiar LMV types and varieties. These include Mazique Incised and a regional variant, Lapile Incised (Schambach 1981:166); Harrison Bayou Incised (a panregional type, fairly common in the LMV, but, as noted by Ford 1936:96, Figures 1 and 2, Harrison Bayou itself is actually in northeast Texas!); Coleman-like Pargoud Incised (Schambach 1981:167); punctated or pinched variants (as in Avoyelles Punctated, Evansville Punctated, and Hollyknowe Ridge-Pinched or Caddoan types); and, more rarely, Plaquemine Brushed (or Caddoan brushed types). Also present are variants of Kiam Incised, a panregional type ¤rst described in the Caddoan literature (Suhm and Jelks 1962:89), with some resemblance to Coles Creek Incised, var. Hardy, often with diagonal overincising. Introducing his descriptive system, Schambach (1981:106) speci¤ed that it was designed for the “clay-tempered [i.e., grog-tempered] Mississippi period decorated pottery” of the Felsenthal region (emphasis added). Subsequently, its coverage has been expanded spatially, temporally, and culturally, taking in later shell-(and grog-) tempered Caddoan pottery (Schambach and Miller 1984:112ff ) and earlier grog-tempered ceramics (Schambach and Waddell 1990:54–58, Table 3; Schambach 1990:113ff ). The Cypress Swamp and Gran Marais assemblages shared Baytown Plain, Harrison Bayou Incised, and Kiam Incised, on the type level (Schambach 1990:116, 121). Brushed pottery was also present in both, but Plaquemine Brushed was not included in the lists because it was not sortable from Caddoan brushed types (Schambach 1990:57, 61). Notably, only Cypress Swamp had Coles Creek Incised varieties, Evansville Punctated, and Hollyknowe Ridge-Pinched. Conversely, only Gran Marais had Avoyelles Punctated, Lapile Incised, and Pargoud Incised. Actually, this overall Gran Marais type list represents the Shallow Lake site assemblage. At Bangs Slough, where the Gran Marais component was estimated as perhaps “50 years later than the Shallow Lake assemblage,” Harrison Bayou Incised was absent, apparently “replaced by Mazique Incised.” That would imply a late, Manchac-like variety of the

outer limits of pl aquemine culture

179

Mazique type, which was represented by a Mazique-like variety in the Cypress Swamp assemblage (Schambach 1990:55, 62).

Sites and Settlement Data in the Felsenthal Heartland An intensive bankline survey by Hemmings (1982) along the Ouachita and lower Saline rivers, in the heart of the Felsenthal region, found a number of Gran Marais components, representing temporary extractive camps in this ®ood-prone environment. Beyond the banklines, Hemmings called attention to seven Gran Marais mound centers, which apparently also served as multiseasonal, if not permanent, base settlements, on or near the overlooking terrace edges or on “pine island” ridges within the ®ood zone (Hemmings 1982: 151–153, Figure 34). One center, Watts Field in Union County, has at least nine mounds, including two rectangular ®at-topped mounds at the ends of a long plaza, and several adjacent midden zones (House 1982:69–71, Figure 14). Another Union County center, Shallow Lake, had at least ¤ve small mounds and one plaza, plus several dispersed midden zones (Rolingson and Schambach 1981:7–10, Figure 3). One mound produced Gran Marais materials associated with a circular structure, about 12 m in diameter, with widely spaced post molds (Rolingson and Schambach 1981:37–40, Figure 38) and a number of pit features. Radiocarbon and archaeomagnetic analyses suggested a date range of a.d. 1250–1300 (Rolingson and Schambach 1981:52–53). The Eagle Lake center, on a terrace edge in southern Bradley County above the Saline-Ouachita juncture, has four mounds, one of them fairly large, around an elongated plaza, in addition to two late prehistoric habitation zones. Surface collections, plus tests atop a medium-sized mound and well into (and below) the smallest one, suggest late Coles Creek to Plaquemine components, with no evidence yet of Mississippian occupation ( Jeter 2007a).

The Saline Valley and Other Tributaries In the bankline zone near the mouth of the Saline River, Hemmings (1982:168ff, 255–263, Figures 68–69, Tables 23–25 and A-3) surface collected and brie®y tested the False Indigo site, ¤nding a small but dense concentration and scatter of over 1,000 sherds. Also present were two Ashley points and an Alba point, along with midden or pit remnants, all consistent with a brie®y occupied, early Gran Marais phase extractive camp interpretation. A short distance upstream, at Jug Point Cutoff, Hemmings (1982:185–187) tested a small, compact midden lens associated with a later Gran Marais ceramic assemblage and an Ashley point (Hemmings 1982:Tables 25 and A-3). Again, a brief seasonal extractive function was likely. Nearby at the Jug Point

180

m arv in d. jeter

2 site, he found a probable early Caney Bayou extractive camp, with nearly 75 percent shell-tempered pottery and probably dating after a.d. 1400 (Hemmings 1982:191–193; Tables 25 and A-3). It is mentioned here because its surface produced two Ashley points, hinting at a late survival of this type. In northwest Ashley County, the Gordon site (not the Gordon site near Natchez) was intensively excavated by amateurs (White 1987). It yielded evidence of Coles Creek occupations, plus a fairly substantial Gran Marais ceramic assemblage (White 1987:Table 1) and a protohistoric Mississippian cemetery. Particularly noteworthy were several rim-body sherds referred to as “Gordon Brushed-Incised” (White 1987:9, 18, Figure 11). The “brushing” actually is sloppily executed incising, using the criteria of Schambach (1981:144). In the less common variant, overincised diagonal lines are often present, and the body is often ¤ngernail-punctated or sometimes plain. In the more common variant, overincising seems to be absent, and the body is always plain. I regard both as varieties of Kiam Incised. Similar materials were found in more limited amateur investigations at the Gary site (3AS247) on the Saline about 5 km above Gordon (Scott 2007), at Gee’s Landing on the river in southwestern Drew County (White 1970:14, Figure 15), and at the Fraser site on a low terrace just east of Warren (White 1987:18, 32). A newly discovered small mound site, a few miles to the north, also appears to have had one or more Coles Creek–Plaquemine occupations. Brief salvage excavations at the Saline-Fifteen site, on the upland edge overlooking the Saline Valley in northeast Bradley County, produced evidence of transitional, late Gran Marais to Caney Bayou occupation. Four separate test units found arbitrarily de¤ned lower midden deposits with about 50-percent shell-tempered pottery below an upper midden with about 70-percent shelltempered ware ( Jeter 2007b). Still farther north, occasional ¤nds in the largely wooded or pastured landscapes suggest that Plaquemine occupations of small sites continued into Cleveland and western Lincoln counties ( Jeter and Early 1999:52–55). A small site in northeast Cleveland County is our northernmost Plaquemine candidate. Between the Saline and Ouachita rivers, the situation along smaller streams such as L’Aigle and Moro creeks is not at all clear, because of lack of ¤eldwork. Only minimal information is available for the Felsenthal region between Moro Creek and the Ouachita Valley. An extensive survey in the central Calhoun County uplands located 49 small sites, 12 of which produced a total of only 121 sherds, more than half of them from one site. Five sites yielded evidence of one or more Mississippi period occupations, but some may represent the Caney Bayou phase rather than Plaquemine culture (House 1986).

outer limits of pl aquemine culture

181

Upstream in the Ouachita Valley Weinstein and Kelley (1984) extensively surveyed bankline levees and natural rises in the Ouachita Valley from Calion to Camden, with testing of selected sites and a regional reconnaissance. They suggested that the maximum northwestward extent of any LMV culture had occurred during Plaquemine times, almost reaching the Camden vicinity, with Caddoan culture beyond in the Middle Ouachita region of the TMS. Going up valley, their ¤rst tested site with a signi¤cant Gran Marais component was Bangs Slough (Weinstein and Kelley 1984:337–375). It was later tested more extensively by the Survey, resulting in a major report (Schambach 1985, 1990). At Little Mud Lake, a short distance up the valley, testing produced evidence of Archaic, transitional Coles Creek, Gran Marais, and Caney Bayou occupations (Weinstein and Kelley 1984:382–414). A slightly disturbed burial included the remains of a large adult male, placed in extended-supine position with the head to the east, fragmentary bones of an infant, and a few intermixed bones of a newborn (Weinstein and Kelley 1984:403–412, Figures 7-44 through 7-46). Near the adult was a ceramic elbow pipe, tempered with grog, ¤nely crushed shell, and bone, with a barred oval design (a “Southern Cult” motif ) on its front and L-shaped variants on its sides (Weinstein and Kelley 1984:403, Figure 7-43A). The ¤ll contained early Gran Marais ceramics, along with an Alba-like point. A sample from the adult’s ribs produced a radiocarbon date centered on the early a.d. 1200s (Weinstein and Kelley 1984:412, Table 7-47). An associated pit contained another grog-tempered pipe, with some attributes of both Caddoan and LMV pipes (Weinstein and Kelley 1984:406, Figure 7-43B). The skeletal remains were thoroughly analyzed. The adult had pronounced occipital ®attening, plus evidence of multiple episodes of metabolic upset between the ages of 1.5 and 6.5 years (Mires and Owsley 1984). A short distance upstream, Boone’s Mounds was tested by C. B. Moore (1909:90–91) “without success.” Even then, the main mound had been partially eroded by the river (Moore 1909:Figure 91). Subsequent relic-collectors’ efforts were more productive. The Survey’s Southern Arkansas University Research Station has photographed a number of their artifacts, sketch-mapped the site, and made additional collections. Weinstein and Kelley (1984:419– 446) partially mapped, collected, and tested the site. It presently includes eight mounds, in various stages of preservation. Seven are arranged in a squarish trapezoidal pattern around a plaza. Apparently they were all truncated pyramidal mounds, from 2 to 6.5 m above the adjacent surface. The other

182

m arv in d. jeter

mound, off to the east, is now only “an amorphous, low rise” that “has succumbed to prodigious pothunting activity” and may have been a domed burial mound (Weinstein and Kelley 1984:424–430, Figures 7-51 and 7-61). The deep, extensive middens and lower mound stages appear to date to the Coles Creek period and Gran Marais phase, whereas only part of the site and some mound tops were used by Caney Bayou peoples. Whole or nearly complete Gran Marais vessels from Boone’s Mounds were illustrated by Weinstein and Kelley (1984:Figures 7-53 through 7-55) and include an Addis Plain open bowl with a scalloped rim (cf. the Boydell bowl), an Anna Incised open bowl, an Evansville Punctated jar, and a Lapile Incised jar with similarities to the Sanson Incised type (cf. Phillips 1970:159). Also present are several specimens of Caney Bayou, Caddoan, or “multicultural” types. An archaeomagnetic date range of a.d. 1270–1355 from one mound was attributed to the Gran Marais phase (Weinstein and Kelley 1984:421). In summary, this was the “paramount site” of this phase in this portion of the region and “construction of most, if not all, of the mounds was probably completed during this time” (Weinstein and Kelley 1984:505). The site has been acquired by The Archaeological Conservancy. Several smaller Gran Marais sites are known in this vicinity (Weinstein and Kelley 1984:505, Figure 9-7). Beyond this cluster, only two Gran Marais components were found in the middle reaches of the survey and none in its upper reaches. The Keller Place, a short distance back from the river, described by Moore (1909:91–96), was more recently visited by Schambach, then by Weinstein and Kelley (1984:124– 130), although it was out of their project area. It has at least six mounds, plus midden areas. Moore (1909:91ff ) gave a height estimate of 13 feet for Mound A, which he described as quadrangular, and 9.5, 3, and 2 feet respectively for three “circular” mounds. He found “human remains throughout” the second mound, representing at least 52 individuals. He also recovered 52 vessels without shell tempering, mostly badly crushed, but including some complete specimens (Moore 1909:Figures 94, 96, 97, Plate VI). One was a miniature with a decoration combining two Southern Cult motifs, the ogee and the barred oval. Also found were a pipe with a notched and engraved design and a copper-covered wooden rattle shaped like a large carnivore canine tooth (Moore 1909:Figures 92, 93, 95). Keller ceramic vessels in private collections include two Coleman Incised jars and one Kiam Incised jar with a punctated body (Weinstein and Kelley 1984:127, Figure 6-23). The site appears to have minor Coles Creek and Caney Bayou components, before and after a major Gran Marais component.

outer limits of pl aquemine culture

183

The Paw Paw site, a short distance upstream from Keller, has a succession of middens on a relict natural levee. It has been tested extensively and found to have been occupied variously between the Middle Archaic and Mississippi periods. It has major Coles Creek components and two Gran Marais components, including a deep midden (Kelley 1992:229–230; Weber 1973:8). Writing before the Gran Marais phase and the transitional Coles Creek concept had been de¤ned, Weber (1973:68ff ) made three subdivisions of “Plaquemine” strata, based on decorated ceramics. In the earliest, late varieties of Coles Creek Incised predominated. In the middle, sherds comparable to Mazique Incised, var. Manchac were dominant. In the latest levels, Plaquemine Brushed was most prevalent.

The “Louisiana Felsenthal” and Lower Ouachita Regions Schambach (1981:105) tentatively placed the southern boundary of the Felsenthal region near the juncture of Bayou Bartholomew with the Ouachita, a short distance above Monroe, Louisiana. Little modern work has been done in that locality, but some similarities to Arkansas have been found (Price and Heart¤eld 1977). The region’s southwestern quadrant was explored by Moore (1913) with little success. Recent work has produced some data from the Scott Place on Big Corney Bayou, just above its con®uence with Bayou D’Arbonne (Harty 2005; Saunders 1998). Two of the ¤ve mounds are ®at-topped, about 3.6 m and 1.5 m high. In one, Moore encountered a burnt feature, which was relocated in 1998 and produced a radiocarbon date centering on a.d. 1250. No diagnostic sherds were associated, but those found elsewhere included Coles Creek Incised, var. Hardy. Another mound site, about 25 km up the same stream, seems contemporary but ceramically closer to Caddoan culture (Harty 1997:136–137). The major Plaquemine center in the Lower Ouachita region is Pargoud Landing, at Monroe (Gibson 1996:69). It may have had six mounds, the largest about 8.5 m high; at least one contained burials (Gibson 1996:70; cf. Kidder 1990a:64). A “Pargoud phase” had been named (see Jeter et al. 1989:211ff, Table 17) but was not really discussed until Kidder (1990a:63–65, Figure 4) summarized and mapped it. He remarked that it probably belonged to “the same, as-of-yet unde¤ned, culture as the Gran Marais phase . . . heavily in®uenced by Lower Mississippi Valley groups, probably by way of the Bartholomew phase of Plaquemine culture.” Meanwhile, Schambach (1990) moved Gran Marais back to Plaquemine cultural status, a subject not addressed by Kidder (1990a, 1991).

184

m arv in d. jeter

Subsistence, Agriculture, and Maize Again Again, ®otation has been inadequately applied. Not surprisingly, macroremains from ®ood-prone bankline sites include deer and small animal bones, mussel shells, some ¤sh bones, and possible nutshell or seed fragments (Hemmings 1982:171, 185, Table B-1; Kelley 1990, 1992). Schambach (1979:30, 1981:105–106) ¤rst hypothesized that some form of slash-and-burn agriculture had been practiced in the Felsenthal uplands. But, in the Bangs Slough report, he suggested that agriculture or even “horticulture” might have been “absent or unimportant in the Felsenthal region in all periods” (Schambach 1990:120–122; emphasis added). Kidder (1991), however, remarked that sampling and recovery had been inadequate, and I ( Jeter and Early 1999:53–54) have suggested that too much emphasis has been placed on the ®ood-prone lowlands, which even now have never been claimed for agriculture. Other sites have produced indirect evidence with regard to late prehistoric agriculture. Faunal data from the lowland Paw Paw site suggested, among other possibilities, that late Coles Creek “clearing of gardens for horticulture was responsible for changes in small mammal frequencies” and that scheduling con®icts might have decreased summer and fall deer hunting and mussel collecting (Kelley 1990, 1992:237–238). Even if agriculture was well established by Plaquemine times, it may not have involved signi¤cant production of maize. The early Gran Marais adult male burial at Little Mud Lake “showed no evidence of dental caries, suggesting that his diet did not include large quantities of domesticated carbohydrates such as maize” (Weinstein and Kelley 1984:412) but instead had a “moderate degree of dental wear” suggestive of “abrasive hunting and gathering diets” (Mires and Owsley 1984:565). Analyses by scanning electron microscopy of a sample of teeth from protohistoric burials at the Gordon site found molar surfaces “fairly smooth” but with some large and small striations, indicating “a relatively soft diet, few abrasive particles, and the consumption of hickory nuts . . . substantially different from all other Lower Mississippi Valley molars . . . a distinctive diet” (Harmon and Rose 1989:331). These teeth also had a fairly high caries rate, possibly suggesting maize or other agricultural carbohydrates in the diet (Harmon and Rose 1989:324, 331–332, Table 38). Fritz and Kidder (1993:10) questioned the correlation between increased caries rates and maize, or even domesticated starchy seeds, suggesting instead that changes in wild plant processing technology might be implicated. More recent work suggests that starchy-seed domesticates were signi¤cant by early

outer limits of pl aquemine culture

185

Coles Creek times (Roberts 2004; Ryan 2004) and maize was at least present in the late Coles Creek subperiod. At Saline-Fifteen, the region’s ¤rst direct evidence of maize emerged. Lopinot (2007) found that it was virtually ubiquitous in feature and midden ®otation samples. He suggested that its cultivation probably went beyond mere gardening and possibly involved production of surpluses. The maize itself has produced one date. Some if not most of it might belong to the Caney Bayou component, but the features, from which most maize remains were recovered, had relatively low percentages of shell-tempered pottery. Although the samples are small, this suggests that the features may represent the Gran Marais component, as does the AMS 1-sigma range, cal. a.d. 1280–1300 (Beta-218342). Other plant remains included abundant hickory nuts, and acorns were common (Lopinot 2007). Persimmon and sumac were also present, but domesticates of the old Eastern starchy/oily seed complex were absent. In this regard, this site resembles the protohistoric (ca. a.d. 1600) Goldsmith Oliver 2 site at Little Rock (Lopinot 1990). Faunal remains (Scott 2007) suggested at least multiseasonal, if not year-round, occupation and a strong reliance on deer, plus a few smaller mammals, along with ¤sh, turtles, mussels, and even some calcined craw¤sh claws (cf. Hemmings 1982:185). Schambach’s original intuition about upland maize farming by Gran Marais peoples may well have been correct, at least for the later portion of this phase. Hemmings (1982:271) also suggested that the higher terrace margins, near the mound centers overlooking the Felsenthal lowlands, had provided land for farming. Such local ecotonal positions were probably emphasized (cf. also Saline-Fifteen).

The Uplands The arbitrary boundary between the Bartholomew-Macon and Felsenthal regions runs along the upland drainage divide between Bayou Bartholomew and the Saline-Ouachita rivers ( Jeter 1982:78–80, Figure 6-2). However, the upland zone, about 40 km (ca. 25 miles) wide, is generally heavily wooded, punctuated by pastures and some small ¤elds. Although known sites are relatively scarce and small, the database is quite inadequate, so it is not clear whether there is a clinal gradation or de¤nite break between these regions’ Plaquemine phases or whether the uplands might have been occupied by a third group with a discernibly different assemblage. The few known upland sites are typically located along small streams and suggest cultural similarities with functional differentiation, that is, smaller and less permanent sites closer

186

m arv in d. jeter

to the drainage divide (e.g., Mooney et al. 2004:141ff; University of Arkansas at Monticello Research Station ¤les). There are hints of differential use of the uplands, most notably at the Clark site in eastern Ashley County, on the upland edge. There, in 1970, amateurs salvaged at least 50 vessels from a site that was being destroyed by sandpitting. Preservation was very poor, but at least 10 to 12 burials were apparently of adults in the extended-supine position, with heads mainly to the north, some possibly in rows. The vessels are mainly crude, grog-tempered, relatively deep jars, some plain, some with combinations of sloppy Gran Marais rim incising and body punctating or pinching. Some vessels resemble Caddoan specimens, but there are more similarities to vessels from Pargoud, though at Clark, punctated body decorations were more common ( Jeter 1982:105; Jeter et al. 1989:216). No evidence of habitation was noted, so this may have been a specialized mortuary site. A similar situation, on a smaller scale, had been found at the Norrell site in northwestern Ashley County, again very close to the upland edge. In 1968, amateurs excavated 14 poorly preserved burials, all adults in extended-supine positions with heads generally to the southeast. Several were probably in rows. Nine vessels were recovered. They generally resemble those from the Clark site. Again, no evidence of habitation was observed. Sites like these may represent a major mortuary practice of the peoples of these phases, possibly comparable to the Mangum “Plaquemine necropolis” on an upland knoll in southwestern Mississippi (Bohannon 1963).

Antecedents of Plaquemine Culture Were the changes from Coles Creek to Plaquemine culture largely gradual and continuous or mainly abrupt and discontinuous, and were their causes more likely endogenous or exogenous? I suspect complex combinations of such processes and causal factors, on varying spatial, temporal, and cultural scales (cf. Jeter 2003a). On the basis of work at major centers like Winterville and Lake George, Brain (1969, 1989) suggested that the change was rather sudden and at least partly exogenous, engendered by Mississippian out of Coles Creek, as it were: “an intrusion that had a vitally stimulating effect, which resulted in the climactic Plaquemine culture” (Williams and Brain 1983:376). This was seen as being evidenced by “Cahokia horizon” artifacts around the late 1100s, with the Winterville phase of Plaquemine culture as a hybrid of Mississippian and Coles Creek traits (Brain 1989:108–110, 117–122; Williams and Brain 1983:374– 375). Grif¤n (1990:68–72) argued that Cahokian in®uence had been over-

outer limits of pl aquemine culture

187

emphasized and that any such contacts had occurred signi¤cantly earlier, around a.d. 1050 in the then-extant chronology (Grif¤n 1990:71). Recent recalibrations would move this into the 1100s (cf. Emerson and Girard 2004). Rolingson (1976:119) suggested that the Bartholomew phase might represent instead a northward “intrusion of Plaquemine peoples . . . out of either the Tensas Basin or the Ouachita . . . Valley.” Belmont (1985:279) found a Boeuf Basin settlement shift from south to north between Coles Creek and Plaquemine times (cf. Kidder 1990a:67), indirectly supporting Rolingson’s hypothesis. However, there is also evidence for in-place continuity from Coles Creek antecedents, which are dif¤cult to recognize because of the dominance of plainware. For Boydell Mound A, House saw “a strong case . . . for continuity across this span of a.d. 900 to 1200 and beyond” (House and Jeter 1994:77). Intensive surveys around Boydell ( Jeter et al. 2007) produced artifacts suggestive of at least two and possibly eight small Coles Creek sites. At Taylor Mound 3, the late 1100s Stage III surface yielded terminal Coles Creek artifacts, topped by Stage IV with an early 1200s date ( Jeter, ed. 2007). Schambach (1979:30) remarked that at least some mounds in each of 11 major Felsenthal groups were built or begun during the Coles Creek period, and all of these groups appeared to have been maintained, added to, and used as cemetery areas by Plaquemine people. Later, he saw the Cypress Swamp phase as transitional and “clearly ancestral” to Gran Marais (Schambach 1990:120–121). Weinstein and Kelley (1984:505) found in Gran Marais “a continuation of site occupation,” with Boone’s Mounds continuing as the paramount site, as during Coles Creek times. Data from Paw Paw also support continuity (Weber 1973). Down the Ouachita Valley, Gibson (1996:69) saw Plaquemine culture as “little more than an evolved pottery complex, not a radically new way of life.” He added that although realignments of settlements took place, they could have been merely changes of locations or “adaptational, as peoples took to the swamps paralleling the rivers.” There seems to have been a strong Coles Creek– Plaquemine continuum in the Tensas (Rolingson 1976:118; cf. Hally 1972:308ff ).

Sociopolitical Organization, Change, and Relations Social organization has not been intensively explored for Plaquemine phases in these regions. But it can at least be outlined on several scales: within phases, between phases (or site clusters), between regions, and by comparisons with the LMV and greater Southeast. About all we have for the dubious Bellaire phase are hints of associations with a few sites and that elite-looking serpent-cat pipe. For the Bar-

188

m arv in d. jeter

tholomew phase, Rolingson (1976) noted the presence of small ceremonial centers, plus a simple, size-based hierarchy of non-mound sites and an “outstanding lack . . . of status goods” in burials (Rolingson 1976:116). She did add, though, that “some status distinctions” might be indicated by ¤nds of better ceramics and a ¤gurine at a ceremonial center. Along the bayou, I have noted a tendency for nonlocal quartz crystals to be associated with mound centers and perhaps also burial sites. Another indication of distinctions is the possibility that marginal individuals were buried at small hamlets like McArthur, while higher-status people may have been buried in mounds as at Boydell and/or at specialized upland cemeteries, such as Clark. Use of charnel houses, as evidenced by bundle burials, suggests more elaborate treatment for some of the populace. The Cypress Swamp phase has been identi¤ed only at one extractive site. The Gran Marais phase is associated with some large mounds and multimound sites, and the apparent hierarchy of such sites may re®ect a “complex chiefdom” though site functions and seasonality may be factors (Weinstein and Kelley 1984:511–512). Despite a few Southern Cult ¤nds, de¤nite highstatus burials are unknown. However, it is of interest that some of the largest Gran Marais centers are close to the apparent border with Caddoan cultural territory and that Southern Cult items have been found in that vicinity. Similarly, the multimound Taylor site is near the northern limit of the Bartholomew phase and has produced some likely Cahokian materials. Meanwhile, east of the Mississippi very large mound sites like Winterville and smaller mound centers were ®ourishing, though again high-status burials are unknown. To our south, especially in the Tensas Basin, similar things were going on (Hally 1972). By comparison with those heartland ceremonial centers, the Bartholomew and even Gran Marais centers look relatively simple. Another large-scale question deals with interactions between Arkansas Plaquemine peoples and other Plaquemine groups, Mississippians, and Caddoans. The most distinctive ceramic and arrow point types have different distributions, which might be related to differing male and female roles in interregional communication and/or marriage and residence rules. Ashley points may suggest a relatively restricted male communication zone, including southeast Arkansas and regions to the west. It did not extend east across the Mississippi or much south of the present Louisiana state line, and was smaller than the preceding Homan distribution zone. Such an interaction zone may relate to communications about how to shape points rather than actual exchange, as the points were made predominantly on local pebble raw materials. Assuming that the pottery was basically produced and decorated by women, they apparently had a different zone of communication and interac-

outer limits of pl aquemine culture

189

tion. It extended about as far north and perhaps as far west (the “boundary” with Caddoan ceramics was certainly permeable, at least for some decorative techniques and styles), but it also reached across the Mississippi and signi¤cantly farther to the south, as indicated by the basic ceramic paste recipes, decorative techniques, and designs. This possibility of a larger interaction zone for Plaquemine women would be in accord with some recent Old World studies, which indicate that “surprisingly, in general females seem to have stirred the genetic melting pot by dispersing their DNA more widely than their brothers” (Pennisi 2001:1733), implying that “patrilocal societies . . . have dominated during the past several millennia, most likely emerging with the growing dependence on agriculture,” although these ¤ndings are still tentative and contested (Pennisi 2001:1734). There is a tremendous potential for DNA studies in the LMV, but because of political considerations it may remain unful¤lled ( Jeter 2002a:195–196, 2003b:186–188). On a still larger scale, an interesting contour map suggests that chiefdom organization spread very rapidly down the Mississippi Valley (Anderson 1999:Figure 15.5), from northeast Arkansas around a.d. 950 to the Red River mouth by about 1000. Whether or not this is precisely correct (it postdates major developments at Toltec), it certainly suggests that chiefdoms preceded Plaquemine culture and were fueled largely by the Native Eastern domesticated plants and wild resources, before the southward spread of signi¤cant dependence on maize. Other recent studies have found some evidence for incipient strati¤cation by late Coles Creek times in northeast Louisiana (Ryan 2004:247–248; Wells 1998). Gibson (1996:60) professed a “general impression . . . of population reduction and growing sectionalism.” Whereas there had been 17 known Coles Creek mound centers along the Ouachita in Louisiana, there were only 13 with Plaquemine components, “usually minor, and . . . dominant only at six sites” (Gibson 1996:69), perhaps indicating unsettled times. Some of this instability, along with settlement shifts, could be symptomatic of sociopolitical factors. Perhaps there was a disintegration of earlier Coles Creek–Plaquemine spatialcultural continuity in the Tensas–Lower Ouachita regions, followed by some northward migration and, especially in the Felsenthal region, closer af¤liations with Caddoan peoples. Maybe some people voted with their feet to escape incipient chie®y elitism along the Mississippi River.

Environmental Factors Perhaps there were also exogenous factors, such as climate change. Rolingson (1976:104ff ) not only began studying the southeast Arkansas environment via nineteenth-century General Land Of¤ce notes and historic ®ood data but also

190

m arv in d. jeter

looked for connections with global climatic change. She suggested that a warm, moist climatic optimum between about a.d. 1000 and 1200, followed by a relatively warm, dry climate, might have made the alluvial valley environment more attractive. A Little Ice Age, dating around a.d. 1430–1850, might then have been a factor in a post-Bartholomew population drop (Rolingson 1976:109–110). The dates for these global changes have recently been pushed back and their nature has been clari¤ed, with implications of cultural impacts. According to LeBlanc (1999, 2001:45; LeBlanc and Rice 2001:12), a “Medieval [European] Warm Period,” with virtually worldwide, relatively warm and wet conditions, occurred about a.d. 900–1150. During that span, maize-based populations grew substantially in the agriculturally marginal Southwest. This period also saw the rise of Cahokia, where some see a “Big Bang” of consolidation around a.d. 1050, followed by a century or more of relative prosperity and cultural climax or hegemony (Emerson 2002:135–136; Pauketat 2002). It appears that this was not initially fueled by maize, which was present but up to a.d. 1100 or later was only one of a number of cultigens, along with Native Eastern plants, in a “multicropping” situation (Lopinot 2003:144–145). In the LMV, Coles Creek cultures ®ourished and began their transitions to Plaquemine, apparently with even less dependence on maize (Fritz and Kidder 1993; Kidder and Fritz 1993:293–295; Roberts 2004, 2005). By the late 1100s, populations in the marginal Southwest “may have caught up with the increased [carrying] capacity, and by the middle 1200s the climate began to deteriorate into . . . the Little Ice Age” (LeBlanc and Rice 2001:12). There were signi¤cant “drier and colder” conditions (LeBlanc 2001:45). In the Southwest, “con®ict . . . was the consequence” (LeBlanc and Rice 2001:12), with large-scale warfare beginning around a.d. 1250 (cf. Lekson 2002). In the less marginal Midwest, between a.d. 1100 and 1250 maize became dominant (Lopinot 2003:146). Cahokia prospered on a reduced scale for some decades but suffered a “precipitous and abrupt decline” around a.d. 1300. This is attributed by some to “political and social collapse” (Emerson 2002:137–139) but the picture of subsistence during the decades of decline is “less clear” (Emerson 2002:138), and climate change may have been an underlying factor. Between Cahokia and Plaquemine country, maize also took over between a.d. 1100 and 1250 (Lopinot 2003:144, 146). A “shift to greater control by [chiefdoms]” accompanied by the spread of early Southern Cult martial iconography and aggregation of formerly independent chiefdoms “may have begun in earnest after about . . . 1250” (Morse and Morse 1983:247–250, 271). Whether a change to colder, drier conditions was a major factor remains to be

outer limits of pl aquemine culture

191

explored on a region-by-region basis, but Mississippian elements expanded southward into northern Plaquemine territory during the 1300s and beyond the present Arkansas-Louisiana line by around 1500 ( Jeter et al. 1989:171ff, Figures 17–21), as adverse climate peaked (LeBlanc 2001:45). Meanwhile, within northeast Louisiana Plaquemine culture, maize agriculture may not have really intensi¤ed until shortly before 1400 (Fritz and Kidder 1993:11; but cf. Roberts 2005; Wells and Weinstein, this volume). Other aspects of such a process might have included settlement-subsistence shifts. One alternative to emphasizing maize in this sunny “Sportsman’s Paradise” would have been to turn back to more reliance on hunting, gathering wild plants, and wetland-riverine resources. Much of the western Plaquemine territory is taken up by lowlands, including very low, seasonally ®ooded wetlands. A detailed comparative analysis of the Felsenthal situation with that in the Catahoula Basin of eastern Louisiana would be of great interest (Gregory 1969; Gregory et al. 1987; Jeter et al. 1989:213).

Mississippian(?) Successors Shell tempering gradually became more common in Arkansas Plaquemine assemblages until it became dominant and they became Mississippian for archaeologists. Inconsistently, the dominance of shell tempering has not conferred full Mississippian status on other societies such as Caddoan, Fort Ancient, and Oneota. The dominance of shell appears to have marched down the Mississippi Valley at an average rate of about 60 to 75 km per century ( Jeter 2003b:184; Jeter et al. 1989:Figures 16–21). By the mid-1300s, it had crossed the Arkansas River. By the 1500s it had moved into northeast Louisiana. I continue to interpret this process as mainly one of cultural diffusion of shell-tempering technology rather than migration with population replacement ( Jeter 1986:49, 2002a:210), but there are other possibilities. Noting Neolithic expansion northwestward into Europe at about 100 km per century (Cavalli-Sforza 1996), I wonder whether our situation was also largely or partly a case of population replacement via migration (“demic diffusion” in Cavalli-Sforza’s terms) or some complex combination of these processes, along with some biological interbreeding ( Jeter 2003b:183–188). The large-scale DNA analyses necessary to test this hypothesis seem unlikely to be conducted. Meanwhile, several lines of evidence suggest a good degree of cultural continuity. In the Yazoo Basin, “Mississippianization” was “a gradual affair” (Williams and Brain 1983:392). In southeast Arkansas, like the Plaquemine peoples before them, the regional Mississippians had dispersed settlement pat-

192

m arv in d. jeter

terns of small hamlets, farmsteads, and temporary foraging sites, focused on a local ceremonial (mound and/or charnel house/cemetery) center, rather than the nucleated agricultural villages and larger polities of the Mississippians to the north. In a preliminary paper, Rolingson (1972) referred to the Bartholomew phase as a rural variant of Plaquemine culture. Independently, I used the phrase “rural Mississippian” for these same regions ( Jeter et al. 1979), borrowing the phrase from Goldstein’s (1976) work in west-central Illinois. Furthermore, there is some artifactual evidence for continuity. Major decorative ideas apparently spread from south to north (e.g., Mazique to Barton rectilinear incising, Coleman to Winterville and other curvilinear incising, Evansville to Parkin ¤ngernail punctating, Plaquemine to Grace brushing), while the transition from grog to shell tempering spread in the other direction (Hally 1972:624–625). Although settlement concentration shifted from the Bartholomew locality to the Tillar vicinity, our surveys in between have turned up small habitation sites with varying percentages of shell- versus grogtempered pottery. I see them as something like Ford’s ideal seriation subjects, catching brief intervals of time in the process of change from grog to shell tempering. In his Felsenthal bottomland surveys, Hemmings (1982:155ff, 263–266) found evidence of varying percentages of grog- and shell-tempered ceramics in the Gran Marais–Caney Bayou transition. Similarly, at SalineFifteen ( Jeter 2007b), four test units all produced about 50-percent shelltempered sherds from the lower portion of a midden and over 70 percent from its upper portion.

More Distant Successors? If the rural Mississippians of southeast Arkansas were indeed basically the descendants of rural Plaquemine peoples in these same regions, who were their further protohistoric and early historic descendants, in the ethnic/tribal sense and linguistically? One possibility, assuming still further continuity, is that Tunican speakers, perhaps including the Koroa, who were ethnohistorically documented in these regions in the late 1600s, were their descendants (cf. Jeter 1986; Kidder 1988; Rolingson 1976:118; Schambach 1981:106). Brain (1988:273) suggested that “the” Tunica homeland at the time of Spanish contact in 1541 was in northwest Mississippi but that “Tunica forebears” had arrived there from southeast Arkansas a short time earlier, “leaving behind those who went on to produce the Tillar complex.” A radically different view of the Tunican homeland, as being in the Oklahoma-Arkansas borderlands along the Arkansas Valley, was later proposed by Schambach (1999). I disagree with

outer limits of pl aquemine culture

193

his placement (and Brain’s) but agree that Tunicans had not yet been based in southeast Arkansas by Spanish-contact times. I recently suggested another alternative, a “Tunicans and Northern Natchezans” scenario, implying major protohistoric discontinuities ( Jeter 2002a:206– 213, Figure 2). I still believe that this provides a better explanation of what Soto and his Merry Men found along their winding 1540s transect of Arkansas. Using the Hudson (1985, 1993) revision of the route for the locations of various Indian provinces and settlements, but assuming that Swanton (1939) was basically right about which Indian language families their names represent, the Tunican names are clustered along the southeastern to southern Ozark margins and adjacent lowlands, from extreme northeast Arkansas (Pacaha) to well up the Arkansas Valley (Tanico). This Tunican homeland placement is between the extremes suggested on the east by Brain and to the west by Schambach. More relevant here, this also places speakers of Natchezan languages all over the rest of eastern Arkansas, including all of southeastern to south-central Arkansas, and well up the Arkansas Valley, in the 1540s. I suggest that this implies a basic human biological and linguistic “Northern Natchezan” continuity in and near southeast Arkansas from late prehistoric times, perhaps back to a.d. 1000 or earlier, until contact in the 1540s. Meanwhile, the cultural (archaeological) assemblage had been changing from Coles Creek to Plaquemine in the 1100s and, largely via technological diffusion, to Mississippian in the 1300s to 1400s. Similarly, a bit later, the Caddoan biological, ethnic, and linguistic traditions were probably continuous in and near southwest Arkansas, but their ceramics were also becoming mostly shell tempered. Rolingson (1976:118) also mentioned possible connections between the Bartholomew and subsequent Wilmot phases and the historic Taensa, instead of Tunican speakers. The Taensa were almost certainly Natchezan speakers, ¤rst documented in northeast Louisiana in the late 1600s, but I have suggested that they may have migrated there from eastern Arkansas and/or northwestern Mississippi in the late 1500s to mid-1600s ( Jeter 2002a:210–211). After the Spanish entrada, and possibly because of it (Ramenofsky and Galloway 1997) but not necessarily so (Brain 1988:293), the Tunican and “Northern Natchezan” peoples of Arkansas may have been decimated by European diseases during the later 1500s and earlier 1600s, the 130-year “Protohistoric Dark Ages” with no direct European contact at all but some indirect contact via trade goods ( Jeter 2007c; M. Smith 1987, 1990). Alternatively, they may have been decimated by severe droughts between the 1540s and 1590s

194

m arv in d. jeter

(Burnett and Murray 1993:235–236; Stahle et al. 1985; Stahle et al. 2000; cf. Mainfort 2004:6–7). In any event, I suggest ( Jeter 2002a:210ff ) that they began moving southward during this period, as per the general trend in the rest of the interior Southeast (cf. Smith 1987, 2000, 2002). By the time of the ¤rst French contacts in the 1670s and 1680s, the “Northern Natchezans” were already out of Arkansas and northwest Mississippi and may be represented by peoples such as the Taensa, whose Mississippian archaeological assemblage in their supposed northeast Louisiana homeland surprised Phillips (1970:945). Some remnant Tunicans were then in the Lower Arkansas Valley and southeast Arkansas but were out of Arkansas themselves by a.d. 1700 ( Jeter 1986). In their place, or near it, in the vicinity of the Arkansas-Mississippi river juncture, were the Quapaw. I have suggested that they had come from well up the Ohio Valley around 1660, as very late arrivals in what came to be known as their historic homeland by this ®uke of discovery ( Jeter 2002a:213ff )—but that’s another story. To summarize, I see a basic biological and linguistic continuity of southeast Arkansas populations of late prehistoric (including Plaquemine) to middle protohistoric (Mississippian) “Northern Natchezans,” on the margins of more elaborate cultural entities. Finally, they were caught up in the storm of “Columbian consequences” ( Jeter 2002b) and forced to migrate at some time between the 1540s and 1670s southward toward their “Southern Natchezan” kin.

Acknowledgments and Dedication This is an expanded version of my paper in the Plaquemine Problems symposium at the 2003 Southeastern Archaeological Conference meeting in Charlotte, North Carolina, approximately 25 years after my arrival in southeast Arkansas and my ¤rst encounters with the Plaquemine scene. Many thanks are due to Mark Rees and Patrick Livingood for organizing the symposium and riding herd on us contributors through the expansion-for-publication process, to the staff at the University of Alabama Press for their editorial and production services, and to their reviewers for helpful comments. I am also much obliged to Vincas Steponaitis, Director of the Research Laboratories of Archaeology, University of North Carolina at Chapel Hill, and a former denizen of the Natchez Bluffs region (referred to in my SEAC remarks as “the only Plaquemine relic I’ve seen in North Carolina”). Vin offered helpful comments at the symposium and afterwards by phone and provided internet access to the Lower Mississippi Survey Archives, now maintained by UNC-RLA. Stephen Williams, Emeritus Director of the LMS, has also provided much in the way of information and opinions over the past

outer limits of pl aquemine culture

195

25 years, as have other LMS alumni including John Belmont, Jeffrey Brain, Ian Brown, David Hally, and T. R. Kidder, and our Louisiana and Mississippi colleagues, especially Doug Wells and Joanne Ryan, and also including Jon Gibson, Pete Gregory, Rich Weinstein, David Kelley, Joe Saunders, Chip McGimsey, Sam Brookes, John Connaway, and Jay Johnson. Various members of the Arkansas Archeological Society have helped over the long haul, most notably Ed and Patsy White, who have been formally involved in Arkansas archaeology since the 1960s. Especially memorable are the Gordon site project, our survey work around the Boydell site, and our work at Eagle Lake, performed along with Bob Cooper, the late Armin Dressel, and a number of other Society members and University of Arkansas at Monticello students. My former assistant Hank McKelway helped us with the Boydell survey and a number of other efforts, including some in the Felsenthal bottomlands, where it was our pleasure to assist Tom Hemmings and his crew a bit and to learn a lot. My present assistant, Bob Scott, has worked very productively with our collections. As always, I am grateful to my beloved blushing bride of a decade now, Charlotte Copeland, who accompanied me on the arduous highway journey to Charlotte, North Carolina, and back, and provided constant support and encouragement while I was expanding, then contracting, the manuscript for publication. This essay is dedicated to my distinguished Survey colleagues Martha Rolingson, Frank Schambach, and John House, with great respect and appreciation for a quarter century of collegial research, providing much of the information that I have used (if not misused) here. By the time I got here, Martha had already done the groundwork and ¤rst major publications on the Bartholomew phase and Frank was well on the way to de¤ning the Gran Marais phase and setting up the innovative descriptive ceramic classi¤cation system. Martha has gone on to great things at Toltec, and I wish her a long and productive retirement. Frank has contributed very extensive and welcome emailed comments and discussions. More recently, he has announced his own impending retirement; may it bring still more signi¤cant publications! And when I ¤rst arrived, John was starting the ¤nal salvage at the doomed Boydell Mound A. When I joined him there, I asked, “Just what is this Plaquemine culture, anyhow?” and he replied, “It’s . . . just a bunch of sherds.” Through the efforts of “all of the above,” I think it is something more than that now. However, I am responsible for any errors or misinterpretations that may be found in this essay.

10 Contemplating Plaquemine Culture Tristram R. Kidder

The essays in this volume demonstrate that Plaquemine research is alive and well. Moreover, prospects for further research are excellent; new ideas are being explored and old ones put to the test. Central problems in Plaquemine research emphasized in these contributions include, Where does this society (or is it “these societies”?) come from? How do we identify, de¤ne, and conceptualize Plaquemine in relation to its ancestors, its contemporaries, and its descendants? What are the spatial, economic, ritual, and political variations among Plaquemine societies and how does this variability re®ect on concepts of Plaquemine as a cultural construct? What external in®uences affect the historical development of Plaquemine peoples? What is the place of Plaquemine in the larger social and economic landscape of the Southeast? These new ideas are developing at a time when research on Mississippian societies throughout the Southeast is at an exciting new stage: increasing evidence of regional variation is complemented by emerging ideas of economic, political, and ritual domination by a limited number of centers, notably Cahokia, Etowah, and Moundville (Anderson 1997; King 2001, 2003; Knight 1997; Knight and Steponaitis 1998; Pauketat 2002, 2004). The essays in this volume place Plaquemine in a local and regional context and serve not only as a regional synthesis of a poorly understood culture and temporal period in the Lower Mississippi Valley (LMV) but also as a springboard for developing new research and for probing the larger issues of how late pre-Hispanic societies in the Southeast developed and were maintained. In his landmark summary of the archaeology of the LMV, Phillips (1970:5– 8, Figure 2) made an important but often ignored distinction in terminology. He de¤ned the Mississippi period as a temporal unit but distinguished Mississippian from Plaquemine culture. These are subtle nomenclatural distinctions,

contempl ating pl aquemine culture

197

but they bear serious consideration in light of our emerging understanding of the southeastern cultural landscape after circa a.d. 1000. Phillips needed these terms to set Plaquemine apart as a unique cultural entity, but he also recognized the remarkable cultural, material, political, social, settlement, subsistence, and ritual variability among the populations now lumped into Mississippian “culture” during what is usually referred to as the Mississippian “period.” Not all people living in or at the margins of the Southeast after a.d. 1000 were culturally Mississippians and perhaps it is time our terminology caught up with this apparent reality. In the following discussion I use the term Mississippi period as a temporal unit for the time frame from circa a.d. 1000 to European contact. Plaquemine and Mississippian are reserved to denote different cultural groups or, probably more accurately, groups practicing different cultural behaviors. Archaeologists working with the late pre-Hispanic populations in the Southeast should give thought to the value of distinguishing temporal units from cultural content. Like many culture historical constructs in the Southeast, Plaquemine is built on a foundation made of ¤red clay. This use of ceramics as an indicator of cultural af¤liation is most notable in the distinction between Mississippian and Plaquemine. Mississippian culture in the Mississippi Valley has often been associated with the use of shell as a tempering agent in pottery (Phillips 1970). Shell temper in pottery has been used for some time to denote the distinction between Mississippian and Plaquemine and is the basis for assuming that people who used one or the other of these tempering technologies were culturally distinct populations. Research presented in this volume (e.g., Livingood, Wells and Weinstein) reinforces arguments that shell tempering has been overemphasized as a marker of cultural distinction. While there is no doubt different groups in the LMV had diverse behaviors and distinct cultural af¤liations, to suggest that the use of a single technological attribute is suf¤cient to distinguish large-scale cultural differences is probably simplistic and does not do justice to the remarkable complexity witnessed in late prehistory. In the LMV, shell-tempered pottery is neither an absolute hallmark of Mississippi period occupations nor is it a certain marker of cultural af¤liation. Shell tempering is most common in the northern portions of the valley and the technology appears to have been slowly spreading southward after a.d. 1200 (Hally 1972; Williams and Brain 1983). In the Delta region of the coast shell tempering appears to have been introduced or diffused from farther east, notably the Mobile Bay area of Alabama (Brown 2004; Kidder 2004b). Pottery manufactured in the area between Vicksburg, Mississippi, and Baton Rouge was most commonly tempered with a mixture of well-crushed grog supple-

198

tristr a m r. k idder

mented with bone, shell, and even vegetable matter alone or in combination (identi¤ed as Addis ware) (Phillips 1970; Ryan 2004; Williams and Brain 1983). Mississippi period ceramic technology is quite sophisticated, with vessels often being very well ¤red with thin walls and ®at or round bases. Vessel shapes are very diverse and probably functionally specialized. In addition to standard bowls, jars, and beakers, Mississippi period vessel assemblages frequently include bottles, jars of various forms, bowls with complex vessel shapes, and unique or exotic compound vessels. Mississippi period ceramic decoration in the LMV embodies both holdovers from earlier times as well as new styles and expressions. Rectilinear patterns so common in Coles Creek times diminish in frequency but are never fully replaced. Simple rectilinear incising parallel to or set at an angle to the rim of simple vessels is a familiar characteristic in most assemblages. Brushing becomes common, and various curvilinear incised design styles come into vogue by the thirteenth century. Across the region so-called Mississippian styles are frequently found on local (non–shell tempered) wares while ancestral (mostly rectilinear) styles are in many instances executed on shell-tempered vessels. Painting, most often monochrome reds or red and white polychrome designs, is a minority style in the northern and central parts of the valley but never catches on in the Delta region or along the coast. Many of the designs or design combinations are unique to speci¤c sites or regions, but some pansoutheastern styles can be detected. A common shared trait is the sun circle, which is manifest in a variety of forms (including running scroll, cross-in-circle, guilloche, and swastika). Although pottery with Southeastern Ceremonial Complex designs and motifs is found in surrounding regions, especially to the north and east, it is rare in the Mississippi Valley south of the mouth of the Arkansas River. In the Mississippi Delta vessel fragments with Southeastern Ceremonial Complex design characteristics have been recovered and indicate an extensive trade relationship with Pensacola (Moundville-related or in®uenced) groups along the eastern Gulf coast. A better grasp of Plaquemine ceramic technology, style, and decoration is important for our understanding of temporal, economic, and social variation; however, as noted by Livingood (this volume), the returns on traditional type-variety research at present are limited. New analytic methods and approaches are required to breathe new life into our rather moribund approach to ceramics that historically emphasizes typology over any other information. In reading these essays there is a sense of déjà vu. Many of the issues raised here echo those surrounding the debates on Mississippian Emergence in the 1980s (Fortier and McElrath 2002; J. Kelly 1987, 1990; B. Smith 1984, 1990).

contempl ating pl aquemine culture

199

Then, as now, there were debates about cultural identity and cultural, social, economic, and political variability; I will focus on these issues for the rest of this discussion with a few excursions into related matters. One of the issues raised by these essays is that of analytical scale (notably spatial scale and temporal resolution). Issues of scale are embedded in archaeological research and are certainly not new to the archaeology of eastern North America (Blakeslee 2002; Neitzel 2000; Neitzel and Anderson 1999). Concepts of Plaquemine cultural identity and variability play out at different scales of analysis and thus these debates are critical because they frame theoretical and methodological investigations of Plaquemine archaeology. At what spatial level do we analyze Plaquemine? Scholars working in the LMV have oscillated between polar extremes in identifying Plaquemine as a cultural entity. For example, James Grif¤n (1967) in his summary of eastern North America classi¤ed Plaquemine as a “variant” of Mississippian culture, and Jeffrey Brain (1978, 1989, 1991) has argued that Plaquemine arose from local roots as a consequence of stimulation by northern “Mississippian” neighbors. More recently, I have championed a parochial de¤nition emphasizing the local roots of Plaquemine (Kidder 1998b, 2002, 2004b). Regardless of their merits, these discussions emphasize that scale matters in understanding Plaquemine. The discussions in this volume and in earlier papers direct our focus toward the processes of “Mississippianization” in the LMV. This emphasis is very relevant, as it places the archaeology of the LMV within the larger context of discussions of how Mississippi period societies evolved and how they were connected and integrated. As a theoretical dimension these issues challenge us to explore how “Mississippian” is internalized and expressed—or if it is at all. Such a challenge removes us considerably from debates about shell versus non-shell tempering. While these essays show that we can argue about the how and why and speci¤c details of what makes Plaquemine a distinctive cultural entity, we should not lose sight of the broadest context because, no matter how we parse it, Plaquemine is a “Mississippian” culture when scaled to a continental perspective. Acknowledging the relevance of the spatial scale also begets a comment about temporal resolution. The origins of Plaquemine seem to be reasonably ¤xed at circa a.d. 1150–1250, with a.d. 1200 as a reasonable but arbitrary beginning date. However, we should recognize that how we view temporal resolution is important to our interpretation of Plaquemine history. Clearly, ¤negrained temporal resolution is vital for our understanding of the historical development of societies through time. This challenge becomes critical when we explore the nature and form of apparently contemporary material culture

200

tristr a m r. k idder

variability (and, by extension, social and cultural variability). The essays in this volume demonstrate that contemporary Plaquemine groups manifest considerable variability across space. De¤ning the temporal boundaries of this variability and exploring time-space relationships within the LMV will be critical, as will using these data to explore how this region ¤ts within the broader “Mississippian” world. There appears to be a growing sense that Mississippian origins are the result of rapid transformations, possibly at a few key centers (Pauketat 1992, 2004). This punctuated equilibrium model has important rami¤cations for understanding Plaquemine but it requires temporal sensitivity and a continued emphasis on solid, high-resolution chronologies. Part of the issue of identity is recognizing boundaries, and many of the essays in this volume implicitly touch on this issue. What is the core of Plaquemine? How do we de¤ne this core? These issues are especially evident along a north–south geographic axis. The valley walls largely (but clearly not entirely) bound Plaquemine, but is the geographic and cultural core centered on Larose, Baton Rouge, Natchez, or Greenville? If mound building de¤nes the core then the center is between Greenville, Mississippi, and Jonesville, Louisiana; but should we be in®uenced by mass? Perhaps we should be asking whether there is even a core of Plaquemine culture. This issue is not entirely an esoteric argument among specialists. If the study of Plaquemine is to have relevance beyond the LMV then understanding where Plaquemine exists is as important as understanding what it is. Models of cultural origin (e.g., Mississippianization vs. indigenous origins) turn on how cultural processes took place and where (and when) they occurred. In this regard, Jeter’s documentation (this volume; see also Jeter and Williams 1989) of “rural” Plaquemine in the northwest “frontier” region suggests several important issues. One, of course, is that there is such a concept as a frontier or non-core population. And by his use of the term and his explicit discussion, Jeter clearly feels that there is a core for Plaquemine. Jeter also invokes the concept of a cosmopolitan population living in the major river valley in contrast to “rural” groups in small tributary valleys. Research in the Ouachita Valley provides evidence that there is a considerable Plaquemine-like occupation extending north to the Felsenthal region (Kelley 1984; Kidder 1990a; Schambach 1990). Are these groups, sometimes identi¤ed as “Pladdoan” (a supposed fusion of Plaquemine and Caddoan cultural traits), really peripheral, or do they deserve their own status as equal partners in some sort of larger Plaquemine political and cultural universe? Similarly, work by Livingood (this volume) in the Pearl River valley demonstrates that characteristics of Plaquemine culture moved outside of the valley wall con¤nes of the

contempl ating pl aquemine culture

201

Mississippi and invites the question of where the eastern boundaries of this culture lie. Perhaps more important, these populations living “on the edge” provoke the question of how they developed the characteristics we recognize as Plaquemine. Are these groups derived from populations that were once living in the core area (presumably the LMV proper) and that budded off to inhabit new territories? Are they local indigenous populations that have adopted Plaquemine cultural traits? If we recognize peripheral populations then we encourage questions about how these populations interacted and how (and if ) they were related. In this regard I would note that the distribution of Plaquemine populations is largely isomorphic with that of Coles Creek. Scholars have in many instances observed that the geographic, ethnographic, and historical “center” of Plaquemine culture can be de¤ned by the ethnographically known Natchez, who were the ¤nal manifestation of Plaquemine in historic times (Brain 1978, 1989; Brown 1985a, 1985b; Brown and Brain 1983; Neitzel 1965, 1983). But the Natchez were not the only inheritors of the Plaquemine legacy. Rees (this volume) points out that the Natchez were not the biological end of Plaquemine; there are living Plaquemine descendants today in the Chitimacha. Further, the archaeological boundaries of Plaquemine culture encompassed at the time of European contact signi¤cant ethnic and linguistic diversity (Foster 1996; Haas 1940, 1956, 1971, 1979), with not only Natchezan speakers (e.g., the Natchez, Taensa) but also those who spoke Tunican (the Tunica, Koroa), Chitimacha (the Chitimacha, Ouacha/Chawasha), and southern Muskhogean (the Bayougoula). Thus, Plaquemine “culture” crosses ethnic and linguistic boundaries and reminds us that the archaeological de¤nition of a culture is based on only one segment of what really de¤nes culture in a contemporary context. But the crux of the identity issue is thus (to reframe Rees and Livingood’s [this volume] simpli¤cation): is Plaquemine Mississippianized Coles Creek, is it a local historical outgrowth from Coles Creek, or is it the result of a new population that had minimal ties to populations living in the Coles Creek “heartland.” These issues permeate all of the chapters in this volume. On a larger scale these are the same issues that have bedeviled archaeologists studying Mississippian cultures since the late nineteenth century (Phillips 1970; Phillips et al. 1951; Smith 1984) and that allow us to recognize that a better understanding of Plaquemine contributes to a fuller comprehension of how later pre-Hispanic societies in the Southeast developed after a.d. 1000. The Mississippianized Coles Creek argument was the outgrowth of Jeff Brain’s work at Winterville and Lake George (Brain 1978, 1989, 1991; Williams

202

tristr a m r. k idder

and Brain 1983). As presented by Brain, Plaquemine was the climactic outcome of Coles Creek peoples receiving the fruits of Mississippian knowledge and wisdom, evidently transported in a relatively small number of pots. Elsewhere, I have argued that Plaquemine is the outgrowth of indigenous Coles Creek populations (especially, one presumes, the elite) coalescing in response to external challenges and events (Kidder 1998b, 2002, 2004b). One key factor I have cited to argue for a largely indigenous origin for Plaquemine is the absence of “Mississippian” material culture traits outside of a limited number of sites in the Yazoo Basin. Some of the best data on exogamous contacts between Coles Creek people and their northern neighbors come out of the new and exciting work at Lake Providence (Wells and Weinstein, this volume). Here we have important contexts rarely witnessed and a well-dated and precisely de¤ned stratigraphic sequence showing the development of this site (Weinstein 2005). The presence of abundant Mississippian material culture elements, many sourced directly to Cahokia, is hard to ignore. Wells and Weinstein, however, carefully (and, to my thinking, quite reasonably) avoid assigning causality to the Cahokia connection. Clearly, however, it is impossible to ignore this connection and these data provide an opportunity to rethink notions of external causality and internal historical processes. I suspect that we will ¤nd that the evidence increasingly points to a combination of causal factors leading to the transition from Coles Creek into Plaquemine. The Lake Providence data are all the more intriguing because that site is essentially fully contemporary with the later occupations at Raffman, located roughly 38 (straight-line) km to the west and south. Raffman, which is physically larger and has more and bigger mounds than Lake Providence (Kidder 2004a), has an entirely different ceramic assemblage in the period a.d. 1000– 1200 (Roe, this volume). Despite considerable work at Raffman there is almost no evidence of long-distance trade or exchange of any sort and certainly nothing like the abundance of data from Lake Providence. Subsistence data also indicate that Raffman and Lake Providence were very different. At Lake Providence there is considerable evidence that the site’s inhabitants were engaged in a maize-based subsistence economy while the contemporary population at Raffman was only occasionally using corn (Roberts 2005; Trachtenberg 1999). Roe suggests that at sites such as Raffman complexity of the sort expected for Plaquemine is already present in advance of maize agriculture, long-distance contacts, or the expansion/expression of Plaquemine cultural identity (see also Kidder 2004a). These two sites emphasize as we have never seen before the existence of extensive cultural variability (at least as measured

contempl ating pl aquemine culture

203

by ceramics and other material culture traits) at the time of the “transition” to Plaquemine. How we understand these issues has yet to be resolved, but the work presented in this volume is a ¤rst step to appreciating such variability. We also should be cognizant of the evidence for both western and eastern long-distance contacts circa a.d. 1000–1200. One can make an argument that the east–west axis from Moundville to the Great Bend of the Red River is just as important for the movement of trade items (especially prestige goods) as a north–south axis descending from Cahokia (Brain and Phillips 1996; Kidder 1998c; Steponaitis et al. 1996). Taken in concert, these data emphasize again that events in the LMV are not occurring in a vacuum; understanding Plaquemine requires a pan-southeastern perspective. Ian Brown (this volume) advocates a slightly different view of the transition to Plaquemine, suggesting that the evidence for the long-assumed continuity (cultural and perhaps biological) from Coles Creek to Plaquemine may not be as strong as it is generally thought. His model argues that the Plaquemine folk of the Natchez Bluffs region may be derived from a migrant population presumably from more northern (but not that far north) regions. To my view the data are not clear on the issue of population origins and I am more inclined to see continuity. However, this work provides an important perspective that cannot be ignored. We know that historic populations moved and it is not unreasonable to see precontact groups moving about as well. Brown’s hypothesis further emphasizes the remarkable variability that exists within the Plaquemine “world” (or perhaps it really emphasizes differences among archaeologists working on Plaquemine issues). Brown’s data derive from the Natchez Bluffs region, which is directly across the Mississippi River from a series of major sites (e.g., Osceola, Routh, Fitzhugh, Raffman, Lake Providence) that have been argued to demonstrate convincing evidence of continuity from Coles Creek into Plaquemine (Hally 1972; Kidder 1998b, 2002). At this point I would not be especially surprised to ¤nd these two adjacent areas had distinctly different political and perhaps even population histories. Brown’s chapter also introduces us to the complexity and subtlety of the emerging (existing?) political landscape. His hypothesis of shifting political capitals is one that merits attention and should be tested. It certainly ¤ts existing data from the LMV (Brown and Brain 1983) and elsewhere (Anderson 1996b; Blitz 1999; Hally 1996; Scarry 1996). These concepts remind us that these people had distinct histories and these histories count when we examine the social, political, and cultural trajectories of Plaquemine “culture.” Additionally, Brown’s discussion serves to focus our attention on the tran-

204

tristr a m r. k idder

sition from Coles Creek. Here the question resolves itself in the following form: How simple was Coles Creek? This argument is critical for modeling Plaquemine as an exceptional departure from what had come before. As noted by Roe (this volume) and Jeter (this volume), recent research suggests that Coles Creek leaders had the capacity to mobilize labor and to direct it in speci¤c and seemingly complex ways before about a.d. 1000. In fact, if we emphasize earth moving and mound construction as a key attribute of complexity and a signi¤cant manifestation of the emergence of permanent, centralized leadership, one might locate the source of Mississippian emergence in Coles Creek times rather than at a.d. 1100–1200 (Kidder 2002, 2004a, 2004b). Some characteristics that traditionally de¤ne Mississippi period societies (e.g., large mound sites constructed around an open plaza, hierarchical settlement patterns) are present at sites such as Pritchard Landing, Raffman, and Osceola in advance of Plaquemine hallmarks such as shell-tempered or Addis pottery. Assuming for a moment that Coles Creek is perhaps more complex than we sometimes think and that Brown’s porridge is, as he suggests, simmering by a.d. 1000, then perhaps it is time we consider situating the presence of long-distance trade goods in the context of how different communities were jockeying or contending for power. One of the major and incontrovertible shifts in the Tensas Basin of northeast Louisiana in the period circa a.d. 1150– 1250 is a movement from more interior locations toward those situated along the Mississippi River. Raffman, Mott, and Pritchard Landing are all in decline (for lack of a better term) by a.d. 1200 while Transylvania, Fitzhugh, and Routh are in ascendancy (Kidder 2002). Lake Providence at circa a.d. 1150– 1200 was a minor center, all things considered. Could the elites of this minor center have been trying to leverage Lake Providence’s location on the river in an attempt to increase its political and social prestige—especially at the expense of its more interior and evidently less cosmopolitan neighbors? In contrast, could Lake George, located in the interior, be manipulating longdistance trade goods to capture power from Winterville and Mayersville, which were located on the river? While such scenarios are wholly hypothetical they do suggest that we place long-distance trade in a context of local political and social behavior and that we assume the local populations were actively engaged in de¤ning themselves. These issues also suggest that we should keep in mind the signi¤cance of the shifting physical geography of the Mississippi Valley and its (potential) impact on the cultural and political landscape. In sum, cultural changes in Plaquemine times (e.g., bigger mound groups arranged in a hierarchical settlement pattern, increasing emphasis on maize, greater material culture diversity) were occurring throughout the LMV and

contempl ating pl aquemine culture

205

can be recognized as historical trends dating prior to a.d. 1000. To me, the major architectural and material culture changes in the Mississippi period have clearly local Coles Creek antecedents. There is no doubt that people living in the LMV, especially those in the northern range of the region, were feeling the economic, political, and possibly religious in®uence of Mississippian cultures from the north and east. I suspect that local elites were mobilizing contacts with populations and polities beyond the LMV as part of a strategy to enhance or solidify their political and social position. Unspeci¤ed Mississippian in®uences, however, cannot account for the signi¤cant changes taking place in the Mississippi Valley after a.d. 1000. Perhaps because of the rich resources of the region, and in contrast to other parts of the Mississippi Valley, overt warfare and hostility were not elements constraining social and settlement organization. We know from ethnohistoric documents, however, that armed con®ict and interpersonal competition were important elements of the social and political landscape. Plaquemine groups in all parts of the valley and along its fringes exhibited a high degree of variability and differentiation. Our research in the future should not ignore this pattern. After a.d. 1200 local (and recently arrived?) populations continued to re¤ne their adaptation to the rich resources of the region while selectively adding new technologies and concepts to existing cultural systems. Political centralization in the Mississippi period was an outgrowth of competition among and between individuals and communities in Coles Creek times. These political shifts were likely enhanced or encouraged by external contacts and opportunities. What changes after a.d. 1200 is the scale of events, not their underlying structure. Can the origins and perpetuation of Plaquemine be both endogenous and exogenous? The essays in this volume suggest the answer is yes. These studies demonstrate that the culture we know as Plaquemine is far more complex, varied, and nuanced than we have traditionally believed. The challenge now is to take these ideas and to test them against the archaeological data.

Acknowledgments I am grateful to the editors for providing this opportunity to put my thoughts on paper and for patience with my slow progress. I am also indebted to Vin Steponaitis for his insights about Plaquemine and contemporary groups.

References Cited

Albrecht, Andrew C. 1944 The Location of the Historic Natchez Villages. Journal of Mississippi History 6(2):67–88. 1946 Indian-French Relations at Natchez. American Anthropologist 48(3):321–354. 1948 Ethical Precepts among the Natchez Indians. Louisiana Historical Quarterly 31(3):569–597. Anderson, David G. 1994a Factional Competition and the Political Evolution of Mississippian Chiefdoms in the Southeastern United States. In Factional Competition and Political Development in the New World, edited by E. M. Brum¤el and J. W. Fox, pp. 61–76. Cambridge University Press, Cambridge. 1994b The Savannah River Chiefdoms: Political Change in the Late Prehistoric Southeast. University of Alabama Press, Tuscaloosa. 1996a Chie®y Cycling Behavior and Large-Scale Abandonments as Viewed from the Savannah River Basin. In Political Structure and Change in the Prehistoric Southeastern United States, edited by J. F. Scarry, pp. 150–191. University Press of Florida, Gainesville. 1996b Fluctuations between Simple and Complex Chiefdoms: Cycling in the Late Prehistoric Southeast. In Political Structure and Change in the Prehistoric Southeastern United States, edited by J. F. Scarry, pp. 231–252. University Press of Florida, Gainesville. 1997 The Role of Cahokia in the Evolution of Southeastern Mississippian Society. In Cahokia: Domination and Ideology in the Mississippian World, edited by T. R. Pauketat and T. E. Emerson, pp. 248–268. University of Nebraska Press, Lincoln.

208

1999

references cited

Examining Chiefdoms in the Southeast: An Application of Multiscalar Analysis. In Great Towns and Regional Polities in the Prehistoric American Southwest and Southeast, edited by J. H. Neitzel, pp. 215–242. Amerind Foundation New World Studies Series 3. University of New Mexico Press, Albuquerque. Arnold, Dean E. 1985 Ceramic Theory and Cultural Process. Cambridge University Press, Cambridge. Autin, W. J., S. F. Burns, B. J. Miller, R. T. Saucier, and J. I. Snead 1991 Quaternary Geology of the Lower Mississippi Valley. In The Geology of North America, Vol. K-2: Quaternary Nonglacial Geology: Conterminous U.S. Geological Society of America, Boulder, Colorado. Baca, Keith A. 1989 Archaeological Collections of the Mississippi Department of Archives and History. Mississippi Archaeology 24(2):28–44. Bareis, C. J., and J. W. Porter 1984 Research Design. In American Bottom Archaeology, edited by C. J. Bareis and J. W. Porter, pp. 1–14. University of Illinois Press, Urbana. Barker, Alex W. 1992 An Archaeological Survey Along the Western Margin of the Upper Tensas Basin, Louisiana. Tensas Regional Archaeological Project, Monograph No. 1. Report on ¤le, Louisiana Division of Archaeology, Baton Rouge. 1999 Chiefdoms and the Economics of Perversity. Unpublished Ph.D. dissertation, Department of Anthropology, University of Michigan, Ann Arbor. Barker, Alex W., and Timothy R. Pauketat 1992 Introduction: Social Inequality and the Native Elites of Southeastern North America. In Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America, edited by A. W. Barker and T. R. Pauketat, pp. 1–10. Archeological Papers of the American Anthropological Association, No. 3. American Anthropological Association, Arlington, Virginia. Barker, Alex W., and Timothy R. Pauketat (editors) 1992 Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America. Archeological Papers of the American Anthropological Association, No. 3. American Anthropological Association, Arlington, Virginia. Barnett, Jim 1998 The Natchez Indians. Division of Historic Properties, Mississippi Department of Archives and History, Natchez. Beasley, Virgil R. 1998 Feasting and Mound Construction at the Mound 4, Block 1 Locale, the Anna

references cited

209

Site (22Ad500), Adams County, Mississippi. M.A. thesis, Department of Anthropology, University of Alabama, Tuscaloosa. Beck, Robin A., Jr. 2003 Consolidation and Hierarchy: Chiefdom Variability in the Mississippian Southeast. American Antiquity 68(4):641–661. Belmont, John S. 1967 The Culture Sequence at the Greenhouse Site, Louisiana. Proceedings of the 23rd Southeastern Archaeological Conference. Southeastern Archaeological Conference Bulletin 6:27–35. 1982a Toltec and Coles Creek: A View from the Southern Lower Mississippi Valley. In Emerging Patterns of Plum Bayou Culture: Preliminary Investigations of the Toltec Mounds Research Project, edited by Martha Ann Rolingson, pp. 64– 70. Research Series No. 18. Arkansas Archeological Survey, Fayetteville. 1982b The Troyville Concept and the Gold Mine Site. Louisiana Archaeology 9:65–98. 1983 Untitled Series of Notes Concerning Lower Mississippi Valley Ceramic Typology, Compiled between 1964 and 1983. Notes on ¤le, Coastal Environments, Inc., Baton Rouge. 1985 A Reconnaissance of the Boeuf Basin, Louisiana. Louisiana Archaeology 10:271–285. Belmont, John S., and Stephen Williams 1981 Painted Pottery Horizons in the Southern Mississippi Valley. In Traces of Prehistory: Papers in Honor of William G. Haag, edited by F. H. West and R. Neuman, pp. 19–42. Geoscience and Man No. 22. School of Geoscience, Louisiana State University, Baton Rouge. Binford, Lewis R. 1968 Some Comments on Historical versus Processual Archaeology. Southwestern Journal of Anthropology 24(3)267–275. Binford, Lewis R., and Sally R. Binford 1966 The Predatory Revolution: A Consideration of the Evidence for a New Subsistence Level. American Anthropologist 68(1):508–512. Bishop, C. A. 1987 Coast-Interior Exchange: The Origins of Strati¤cation in Northwestern North America. Arctic Anthropology 24(1):72–83. Blakeslee, Donald J. 2002 Fractal Archaeology: Intra-Generational Cycles and the Matter of Scale, an Example from the Central Plains. In The Archaeology of Tribal Societies, edited by W. A. Parkinson, pp. 173–199. Archaeological Series 15. International Monographs in Prehistory, Ann Arbor, Michigan.

210

references cited

Blandino, Betty 1997 Coiled Pottery: Traditional and Contemporary Ways, rev. ed. A and C Black, London. Blitz, John H. 1993 Big Pots for Big Shots: Feasting and Storage in a Mississippian Community. American Antiquity 58(1):80–96. 1999 Mississippian Chiefdoms and the Fission-Fusion Process. American Antiquity 64(4):577–592. Blitz, John H., and Patrick Livingood 2004 Sociopolitical Implications of Mississippian Mound Volume. American Antiquity 69(2):291–301. Blitz, John H., and C. Baxter Mann 2000 Fisherfolk, Farmers, and Frenchmen: Archaeological Explorations on the Mississippi Gulf Coast. Archaeological Report No. 30. Mississippi Department of Archives and History, Jackson. Boggess, Elizabeth, and Bradley E. Ensor 1993 Archaeological Investigations at the Fort Adams Volunteer Fire Department, Inc. Firehouse Site: Smith Creek Site 22Wk526. Report prepared for Fort Adams Volunteer Fire Department, Inc., Woodville, Mississippi. Archaeologists Unlimited, Natchez, Mississippi. Bohannon, Charles F. 1963 The Mangum Site: A Plaquemine Necropolis in Claiborne County, Mississippi. Resource study proposal NATR-A-1, Natchez Trace Parkway. National Park Service, Washington D.C. Boudreaux, Edmond A. 1997 Mound 4 Flats, Block 2. In Excavations at the Anna Site (22Ad500), Adams County Mississippi: A Preliminary Report, edited by Ian W. Brown, pp. 44– 45. Unpublished manuscript, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama Museums, Tuscaloosa. Brain, Jeffrey P. 1969 Winterville: A Case Study of Prehistoric Culture Contact in the Lower Mississippi Valley. Ph.D. dissertation, Department of Anthropology, Yale University. University Micro¤lms, Ann Arbor. 1971a The Lower Mississippi Valley in North American Prehistory. Report submitted to the National Park Service, Southeast Region, and the Arkansas Archeological Survey, Fayetteville. 1971b The Natchez “Paradox.” Ethnology 10(2):215–222. 1976 French Indian Interaction: Political Neutralization and the Destruction of the Natchez. Paper presented at the 5th French Regime Symposium, Fort De Chartres Historic Site, Illinois.

references cited

1978

211

Late Prehistoric Settlement Patterning in the Yazoo Basin and Natchez Bluffs Regions of the Lower Mississippi Valley. In Mississippian Settlement Patterns, edited by B. D. Smith, pp. 331–368. Academic Press, New York. 1988 Tunica Archaeology. Papers of the Peabody Museum, No. 78. Harvard University, Cambridge. 1989 Winterville: Late Prehistoric Culture Contact in the Lower Mississippi Valley. Archaeological Report No. 23. Mississippi Department of Archives and History, Jackson. 1991 Cahokia from the Southern Periphery. In New Perspectives on Cahokia: Views from the Periphery, edited by J. B. Stoltman, pp. 93–100. Monographs in World Archaeology 2. Prehistory Press, Madison, Wisconsin. Brain, Jeffrey P., Ian W. Brown, and Vincas P. Steponaitis 1994 Archaeology of the Natchez Bluffs. Manuscript on ¤le, Research Laboratories of Anthropology. University of North Carolina, Chapel Hill. Brain, Jeffrey P., and Philip Phillips 1996 Shell Gorgets: Styles of the Late Prehistoric and Protohistoric Southeast. Peabody Museum Press, Harvard University, Cambridge. Brightman, Robert A. 2004 Chitimacha. In Southeast, edited by Raymond D. Fogelson, pp. 642–652. Handbook of North American Indians, vol. 14, W. Sturtevant, general editor. Smithsonian Institution Press, Washington, D.C. Brookes, Samuel O., and Byron Inmon 1973 Archaeological Survey of Claiborne County, Mississippi. Mississippi Archaeological Survey Report No. 3. Mississippi Department of Archives and History, Jackson. Brown, Calvin S. 1926 Archeology of Mississippi. Mississippi Geological Survey and University of Mississippi, Oxford. Reprinted by University Press of Mississippi, Jackson, 1992. Brown, Ian W. 1973 Settlement Patterns in the Bluff Area of the Lower Mississippi Valley. Honors thesis, Department of Anthropology, Harvard College, Cambridge. 1978 Archaeology on Avery Island, 1978. Louisiana Archaeological Society Newsletter 5(4):9–11. 1979 Archaeological Investigations at Salt Mine Valley (35-I-5). Petite Anse Project Research Notes No. 8, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. 1980 Archaeological Investigations on Avery Island, Louisiana, 1978. Southeastern Archaeological Conference Bulletin 22:110–118. 1981 The Morgan Site: An Important Coles Creek Mound Complex on the

212

1982a

1982b 1983 1984

1985a

1985b

1989a 1989b 1990

1992

1995

1996

references cited

Chenier Plain of Southwest Louisiana. North American Archaeologist 2(3):207–237. An Archaeological Study of Culture Contact and Change in the Natchez Bluffs Region. In La Salle and His Legacy: Frenchmen and Indians in the Lower Mississippi Valley, edited by Patricia K. Galloway, pp. 176–193. University Press of Mississippi, Jackson. The Southeastern Check Stamped Pottery Tradition: A View from Louisiana. Midcontinental Journal of Archaeology Special Paper 4. Plaquemine Culture Houses in the Natchez Bluffs Region, Mississippi: Excavations at the Lookout Site. Mississippi Archaeology 18(1):14–26. Late Prehistory in Coastal Louisiana: The Coles Creek Period. In Perspectives on Gulf Coast Prehistory, edited by David D. Davis, pp. 94–124. Ripley P. Bullen Monograph in Anthropology and History No. 5. University of Florida Press, Gainesville. Natchez Indian Archaeology: Culture Change and Stability in the Lower Mississippi Valley. Archaeological Report No. 15. Mississippi Department of Archives and History, Jackson. Plaquemine Architectural Patterns in the Natchez Bluffs and Surrounding Regions of the Lower Mississippi Valley. Midcontinental Journal of Archaeology 10:251–305. The Calumet Ceremony in the Southeast and Its Archaeological Manifestations. American Antiquity 54(2):311–331. Natchez Indians and the Remains of a Proud Past. In Natchez before 1830, edited by Noel Polk, pp. 8–28. University Press of Mississippi, Jackson. Historic Indians of the Lower Mississippi Valley: An Archaeologist’s View. In Towns and Temples along the Mississippi, edited by David H. Dye and Cheryl A. Cox, pp. 227–238. University of Alabama Press, Tuscaloosa. Certain Aspects of French-Indian Interaction in Lower Louisiane. In Calumet and Fleur-de-Lys: Archaeology of Indian and French Contact in the MidContinent, edited by J. A. Walthall and T. E. Emerson, pp. 17–34. Smithsonian Institution Press, Washington, D.C. Summary of Warren K. Moorehead’s Work at the Ferguson Mounds (Feltus Site) as Reconstructed from His Unpublished Field Notes and Drawings, 1924. Unpublished manuscript, Gulf Coast Survey Files, Alabama Museum of Natural History, University of Alabama, Tuscaloosa. B. L. C. Wailes and Mississippi Archaeology, Journals of B. L. C. Wailes, Vols. 1–13, 1852–1854, transcribed by Neil Wailes Brandon. Unpublished manuscript, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama, Tuscaloosa.

references cited

213

Excavations at the Anna Site (22Ad500), Adams County Mississippi: A Preliminary Report. Unpublished manuscript, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama Museums, Tuscaloosa. 1998a Benjamin L. C. Wailes and the Archaeology of Mississippi. Mississippi Archaeology 33(2):157–191. 1998b Decorated Pottery of the Lower Mississippi Valley: A Sorting Manual. Mississippi Archaeological Association, Mississippi Department of Archives and History, Jackson. 1998c The Eighteenth-Century Natchez Chiefdom. In The Natchez District in the Old, Old South, edited by Vincas P. Steponaitis, pp. 49–65. Southern Research Report No. 11. Center for the Study of the American South, University of North Carolina, Chapel Hill. 1998d Fatherland Site. In Archaeology of Prehistoric Native America: An Encyclopedia, edited by Guy Gibbon, pp. 269–270. Garland Publishing, New York. 1998e Plaquemine Culture. In Archaeology of Prehistoric Native America: An Encyclopedia, edited by Guy Gibbon, pp. 657–659. Garland Publishing, New York. 1999 Salt Manufacture and Trade from the Perspective of Avery Island, Louisiana. Midcontinental Journal of Archaeology 24(2):113–151. 2003 Concluding Thoughts on Bottle Creek and Its Position in the Mississippian World. In Bottle Creek: A Pensacola Culture Site in South Alabama, edited by Ian W. Brown, pp. 205–226. University of Alabama Press, Tuscaloosa. 2004 Prehistory of the Gulf Coastal Plain after 500 B.C. In Southeast, edited by R. D. Fogelson, pp. 574–585. Handbook of North American Indians, vol. 14, W. Sturtevant, general editor. Smithsonian Institution Press, Washington, D.C. Brown, Ian W., and Jeffrey P. Brain 1983 Archaeology of the Natchez Bluffs Region, Mississippi: Hypothesized Cultural and Environmental Factors In®uencing Local Population Movements. Southeastern Archaeological Conference Bulletin 20:38–49. Brown, Ian W., and Nancy Lambert-Brown 1978 Archaeological Investigations at the Banana Bayou Mound (33-I-6). Petite Anse Project Research Notes No. 5, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. 1979 The Mississippian Ceramics from Salt Mine Valley (33-I-5). Petite Anse Project Research Notes No. 10, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. Brown, Ian W., Richard S. Fuller, and Nancy Lambert-Brown 1979 Site Survey in the Petite Anse Region, Southwest Coast, Louisiana. Petite Anse 1997

214

references cited

Project Research Notes No. 11, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. Brown, James A. 1976 Spiro Studies, Vol. 4: The Artifacts. Stovall Museum of Science and History, University of Oklahoma, Norman. 1990 Archaeology Confronts History at the Natchez Temple. Southeastern Archaeology 9:1–10. Bruseth, James E. 1998 The Development of Caddoan Polities along the Middle Red River Valley of Eastern Texas and Oklahoma. In The Native History of the Caddo: Their Place in Southeastern Archaeology and Ethnohistory, edited by T. Pertulla and J. Bruseth, pp. 47–68. Texas Archeological Research Laboratory Studies in Archeology, vol. 30. Texas Archeological Research Laboratory, University of Texas at Austin. Burnett, Barbara A., and Katherine A. Murray 1993 Death, Drought, and de Soto: The Bioarchaeology of Depopulation. In The Expedition of Hernando de Soto West of the Mississippi, 1541–1543, edited by Gloria A. Young and Michael P. Hoffman, pp. 227–236. University of Arkansas Press, Fayetteville. Carneiro, Robert L. 1981 The Chiefdom: Precursor of the State. In The Transition to Statehood in the New World, edited by G. D. Jones and R. R. Kautz, pp. 37–79. Cambridge University Press, Cambridge. Cavalli-Sforza, Luigi Luca 1996 The Spread of Agriculture and Nomadic Pastoralism: Insights from Genetics, Linguistics, and Archaeology. In The Origin and Spread of Agriculture and Pastoralism in Eurasia, edited by David R. Harris, pp. 51–69. Smithsonian Institution Press, Washington, D.C. Clark, John E., and Michael Blake 1994 The Power of Prestige: Competitive Generosity and the Emergence of Rank Societies in Lowland Mesoamerica. In Factional Competition and Political Development in the New World, edited by Elizabeth M. Brum¤el and John W. Fox, pp. 17–30. Cambridge University Press, Cambridge. Clayton, Lawrence A., Vernon J. Knight, Jr., and Edward C. Moore (editors) 1993 The De Soto Chronicles: The Expedition of Hernando de Soto to North America in 1539–1543. 2 vols. University of Alabama Press, Tuscaloosa. Cobb, Charles R. 1996 Specialization, Exchange, and Power in Small Scale Societies and Chiefdoms. Research in Economic Anthropology 17:251–294.

references cited

215

Mississippian Chiefdoms: How Complex? Annual Review of Anthropology 26:63–84. Collins, Henry B., Jr. 1927 Archaeological Work in Louisiana and Mississippi. In Explorations and Fieldwork of the Smithsonian Institution in 1926. Smithsonian Miscellaneous Collections 78:200–207. Smithsonian Institution, Washington, D.C. Comaroff, Jean, and John Comaroff 1991 Of Revelation and Revolution: Christianity, Colonialism, and Consciousness in South Africa. University of Chicago Press, Chicago. Cordell, Ann S. 2004 Petrographic Analysis of Four Addis Plain Thin Sections. Manuscript on ¤le, Florida Museum of Natural History, Gainesville. Cordell, Ann S., and Patrick C. Livingood 2004 Point/Counter Point: The Accuracy and Feasibility of Digital Image Techniques in the Analysis of Ceramic Thin-Sections. Paper presented at the 61st Annual Meeting of the Southeastern Archaeological Conference, St. Louis. Costin, Cathy Lynne, and Timothy Earle 1989 Status Distinction and Legitimation of Power as Re®ected in Changing Patterns of Consumption in Late Prehispanic Peru. American Antiquity 54(4):691–714. Cotter, John L. 1948 NTP Archeological Survey, Inspection Visit, November 11, 1948, White Apple Mound and Village Site, Adams Co., Mississippi. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. ca. 1950 Results of a Stratigraphic Test of Mound 5 in the Anna Mounds Group, Adams County Miss. Unpublished manuscript, National Park Service, Southeast Regional Of¤ce, Tallahassee. 1951a Stratigraphic and Area Tests at the Emerald and Anna Mound Sites. American Antiquity 17(1):18–32. 1951b Mangum Site Test, April 3, 4, 5, 6, 1951. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. 1952a The Gordon Site in Southern Mississippi. American Antiquity 18(2):110–126. 1952b The Mangum Plate. American Antiquity 17(1):65–68. Cotter, John L. (editor) 1951 Dr. Monette’s Observations on Emerald Mound. Journal of Mississippi History 13(1):31–40. Coxe, Carey L., and David B. Kelley 2004 Faunal Remains. In Data-Recovery Excavations at the Hedgeland Site (16CT19), 2003

216

references cited

Catahoula Parish, Louisiana, edited by Joanne Ryan, pp. 227–239. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Crane, Verner W. 1929 The Southern Frontier 1670–1732. University of Michigan Press, Ann Arbor. Crumley, Carole L. 1994 Historical Ecology: A Multidimensional Ecological Orientation. In Historical Ecology: Cultural Knowledge and Changing Landscapes, edited by C. L. Crumley, pp. 1–16. School of American Research Press, Santa Fe. Culin, Stewart 1900 The Dickeson Collection of American Antiquities. Bulletin of the Free Museum of Science and Art of the University of Pennsylvania 2(3):113–168. University of Pennsylvania, Philadelphia. Dailey, R. C. 1974 Osteological Analysis of the Human Skeletal Remains of the Mangum Site, Natchez Trace Parkway, Mississippi. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. Dalton, George 1977 Aboriginal Economies in Stateless Societies. In Exchange Systems in Prehistory, edited by Timothy K. Earle and Jonathon E. Ericson, pp. 191–212. Academic Press, New York. Davis, Dave D. 1981 Ceramic Classi¤cation and Temporal Discrimination: A Consideration of Later Prehistoric Stylistic Change in the Mississippi River Delta. Midcontinental Journal of Archaeology 6(1):55–89. 1984 Protohistoric Cultural Interaction along the Northern Gulf Shore. In Perspectives on Gulf Coast Prehistory, edited by David D. Davis, pp. 216–231. Ripley P. Bullen Monograph in Anthropology and History No. 5. University of Florida Press, Gainesville. Davis, Dave D., and Marco J. Giardino 1980 Some Notes Concerning Mississippian Period Ceramics in the Mississippi River Delta. Louisiana Archaeology 7:53–66. DeBoer, W. R., and D. Lathrap 1979 The Making and Breaking of Shipibo-Conibo Ceramics. In Ethnoarchaeology: Implications of Ethnography for Archaeology, edited by C. Kramer, pp. 102–138. Columbia University Press, New York. DeJarnette, D. L., and S. B. Wimberly 1941 The Bessemer Site: Excavation of Three Mounds and Surrounding Village Ar-

references cited

217

eas near Bessemer, Alabama. Geological Survey of Alabama, Museum Paper 17. University of Alabama, Tuscaloosa. Delanglez, Jean 1935 The French Jesuits in Lower Louisiana (1700–1763). Loyola University, New Orleans. Di Caprio, N. C., and S. Vaughn 1993 An Experimental Study in Distinguishing Grog (Chamotte) from Argillaceous Inclusions in Ceramic Thin Sections. Archeomaterials 7:21–40. Dietler, Michael 2001 Theorizing the Feast: Rituals of Consumption, Commensal Politics, and Power in African Contexts. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 65–114. Smithsonian Institution Press, Washington, D.C. Dietler, Michael, and Brian Hayden 2001 Digesting the Feast—Good to Eat, Good to Drink, Good to Think: An Introduction. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 1–22. Smithsonian Institution Press, Washington, D.C. 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. 3–17. Routledge, New York. Downs, Lauren E. 2004 Plaquemine Culture Structures in the Natchez Bluffs: Architectural Grammar at the Mound 3 Summit Locale, the Anna Site, Adams County, Mississippi. Unpublished Master’s thesis, Department of Anthropology, University of Alabama, Tuscaloosa. Duhe, Brian J. 1981 A Study of Prehistoric Coles Creek–Plaquemine Cultural and Technological Adaptations in the Upper Barataria Basin. Southeastern Archaeological Conference Bulletin 24:34–37. Dumont de Montigny 1753 Mémoires Historiques sur la Louisiane. 2 vols. Edited by Le Mascrier. Bauche, Paris. Earle, Timothy K. 1987 Chiefdoms in Archaeological and Ethnohistorical Perspective. Annual Review of Anthropology 16:279–308. 1991 The Evolution of Chiefdoms. In Chiefdoms: Power, Economy and Ideology, edited by T. K. Earle, pp. 1–15. Cambridge University Press, Cambridge.

218

references cited

How Chiefs Come to Power: The Political Economy in Prehistory. Stanford University Press, Stanford. Emerson, Thomas E. 2002 An Introduction to Cahokia 2002: Diversity, Complexity, and History. Midcontinental Journal of Archaeology 27:127–148. Emerson, Thomas E., and Jeffrey S. Girard 2004 Dating Gahagan and Its Implications for Understanding Cahokia-Caddo Interactions. Southeastern Archaeology 23:57–64. Emerson, Thomas E., and Randall E. Hughes 2000 Figurines, Flint Clay Sourcing, the Ozark Highlands, and Cahokian Acquisition. American Antiquity 65:79–101. Emerson, Thomas E., Randall E. Hughes, Mary R. Hynes, and Sarah D. Wisseman 2003 The Sourcing and Interpretation of Cahokia-Style Figurines in the TransMississippi South and Southeast. American Antiquity 68:287–313. Everett, Nicole D., Jon L. Gibson, Danny M. Guillory, and Thomas J. Johnson 2002 Dating Bayou Portage Guidry Mounds (16SM38), St. Martin Parish, South Central Louisiana. Louisiana Archaeological Society Newsletter 29(1):7–11. Fischer, John L. 1964 Solutions for the Natchez Paradox. Ethnology 3(1):53–65. Flannery, Kent V. 1967 Culture History v. Cultural Process: A Debate in American Archaeology. Scienti¤c American 217(2):119–122. Ford, James A. 1935a Ceramic Decoration Sequence at an Old Indian Village Site near Sicily Island, Louisiana. Anthropological Study No. 1. Department of Conservation, Louisiana State Geological Survey, Baton Rouge. 1935b Outline of Louisiana and Mississippi Pottery Horizons. Louisiana Conservation Review 4(6):33–38. 1936 Analysis of Indian Village Site Collections from Louisiana and Mississippi. Anthropological Study No. 2. Department of Conservation, Louisiana State Geological Survey, Baton Rouge. 1938 A Chronological Method Applicable to the Southeast. American Antiquity 3(3):260–264. 1951 Greenhouse: A Troyville-Coles Creek Period Site in Avoyelles Parish, Louisiana. Anthropological Papers of the American Museum of Natural History, No. 44, Part 1. American Museum of Natural History, New York. 1952 Measurements of Some Prehistoric Design Developments in the Southeastern States. Anthropological Papers of the American Museum of Natural History, No. 44, Part 3. American Museum of Natural History, New York. 1997

references cited

219

Ford, James A., and George I. Quimby 1945 The Tchefuncte Culture: An Early Occupation of the Lower Mississippi Valley. Memoirs of the Society for American Archaeology Memoir, No. 2. Menasha, Wisconsin. Ford, James A., and Clarence H. Webb 1956 Poverty Point: A Late Archaic Site in Louisiana. Anthropological Papers of the American Museum of Natural History, No. 46, Part 1. American Museum of Natural History, New York. Ford, James A., and Gordon R. Willey 1941 An Interpretation of the Prehistory of the Eastern United States. American Anthropologist 43(3):325–363. Ford, Richard I. 1997 Preliminary Report on the Plant Remains from the George C. Davis Site, Cherokee County, Texas: 1968–1970 Excavations. Bulletin of the Texas Archeological Society 68:104–107. Fortier, Andrew C., and Dale L. McElrath 2002 Deconstructing the Emergent Mississippian Concept: The Case for the Terminal Late Woodland in the American Bottom. Midcontinental Journal of Archaeology 27(2):171–215. Foster, Michael K. 1996 Language and the Culture History of North America. In Languages, edited by I. Goddard, pp. 64–110. Handbook of North American Indians, vol. 17, W. Sturtevant, general editor. Smithsonian Institution Press, Washington, D.C. Frank, Joseph V. III 1975 In Defense of Hutchin’s Natchez Indian. Mississippi Archaeological Association Newsletter 10(NL-4):7–11. 1980a A European Trade Bead from the Anna Mounds Site, Adams County, Mississippi. Mississippi Archaeological Association Newsletter 15(NL-3):5. 1980b The Rice Site: A Natchez Indian Cemetery. Mississippi Archaeology 15(2):32–41. French, Benjamin F. 1869 Historical Collections of Louisiana and Florida. J. Sabin and Sons, New York. Fritz, Gayle J., and John E. Kelly 2004 Maize Symbolism in Context: Plants on Mississippian Figurines and Pipes. Paper presented at the 69th Annual Meeting of the Society for American Archaeology, Montreal, Canada. Fritz, Gayle J., and Tristram R. Kidder 1993 Recent Investigations into Prehistoric Agriculture in the Lower Mississippi Valley. Southeastern Archaeology 12:1–14.

220

references cited

Fuller, Richard S., Sylvia Timmons Duay, and Susan L. Scott 1995 Artifact Analysis. In The Rock Levee Site: Late Marksville through Late Mississippi Period Settlement, Bolivar County, Mississippi, by Richard A. Weinstein, Richard S. Fuller, Susan L. Scott, C. Margaret Scarry, and Sylvia Timmons Duay, pp. 77–155. Volume 3 of The Beulah Levee Project: Archaeology and History, edited by Richard A. Weinstein and Charles E. Pearson. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Fuller, Richard S., and Diane S. Fuller 1987 Excavations at Morgan: A Coles Creek Period Mound Complex in Coastal Louisiana. Bulletin 11, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. Fuller, Richard S., and D. B. Kelley 1993 Archaeological Survey and Testing within Items 4-A and 4-B of the Sicily Island Levee Project, Catahoula Parish, Louisiana. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Fuller, Richard S., and Stephen Williams 1985 Management Summary: Archaeological Survey in the Southern Boeuf Basin, Louisiana, 1984. Lower Mississippi Survey, Peabody Museum, Harvard University, Cambridge. Submitted to Louisiana Division of Archaeology, Baton Rouge. Gagliano, Sherwood M. 1967 Occupation Sequence at Avery Island. Louisiana State University Press, Baton Rouge. 1984 Geoarchaeology of the Northern Gulf Shore. In Perspectives on Gulf Coast Prehistory, edited by Dave D. Davis, pp. 1–40. Ripley P. Bullen Monograph in Anthropology and History No. 5. University of Florida Press, Gainesville. Gagliano, Sherwood M., Richard A. Weinstein, and Eileen K. Burden 1975 Archeological Investigations along the Gulf Intracoastal Waterway, Coastal Louisiana Area. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, New Orleans District. Galaty, Michael 2006 Ceramic Petrography and the Classi¤cation of Mississippi’s Archaeological Pottery by Fabric: A GIS Approach. In Interstate 69 in Mississippi: A Regional Archaeological Perspective, edited by Janet Rafferty and Evan Peacock. Unpublished manuscript, under press review. Galloway, Patricia 1992 The Unexamined Habitus: Direct Historic Analogy and the Archaeology of the Text. In Representations in Archaeology, edited by J. Gardin and C. Peebles, pp. 178–195. Indiana University Press, Bloomington.

references cited

221

Ethnohistory. In The Development of Southeastern Archaeology, edited by J. K. Johnson, pp. 78–108. University of Alabama Press, Tuscaloosa. 1995 Choctaw Genesis: 1500–1700. University of Nebraska Press, Omaha. 1998 Commentary and Re®ection on Long-Term Continuities and Discontinuities. In The Natchez District in the Old, Old South, edited by Vincas P. Steponaitis, pp. 67–78. Southern Research Report No. 11. Center for the Study of the American South, University of North Carolina, Chapel Hill. 2002 Colonial Period Transformations in the Mississippi Valley: Disintegration, Alliance, Confederation, Playoff. In The Transformation of the Southeastern Indians, 1540–1760, edited by Robbie Ethridge and Charles Hudson, pp. 225– 248. University Press of Mississippi, Jackson. Gatschet, Albert S. 1883 The Shetimasha Indians of St. Mary’s Parish, Southern Louisiana. Transcript of the Anthropological Society, Washington, D.C. 1907 Chitimacha. Bureau of American Ethnology Bulletin 30(1). Washington, D.C. Giardino, Marco J. 1977 An Osteological Analysis of the Human Population from the Mount Nebo Site, Madison Parish, Louisiana. Unpublished Master’s thesis, Department of Anthropology, Tulane University, New Orleans. 1979 Cultural Resources Survey and Evaluation of the Cane Creek RC & D Measure, Lincoln County, Arkansas. New World Research, Inc., New Orleans. Report submitted to Interagency Archaeological Services, Atlanta. 1984 Documentary Evidence for the Location of Historic Indian Villages in the Mississippi Delta. In Perspectives on Gulf Coast Prehistory, edited by Dave D. Davis, pp. 232–249. Ripley P. Bullen Monograph in Anthropology and History No. 5. University Press of Florida, Gainesville. 1994 Ceramic Analysis for 16SC27. In Cultural Resources Survey and Testing for the Davis Pond Freshwater Diversion, St. Charles Parish, Louisiana, vol. 2, by Kenneth R. Jones, Herschel A. Franks, and Tristram R. Kidder, pp. 323– 395. Earth Search, Inc., New Orleans. Prepared for the U.S. Army Corps of Engineers, New Orleans District, and on ¤le with the Louisiana Division of Archaeology, Baton Rouge. Gibson, Jon L. 1975 The Prehistory of Acadiana. In The Culture of Acadiana: Tradition and Change in South Louisiana, edited by Stephen L. Del Sesto and Jon L. Gibson, pp. 16–40. University of Southwestern Louisiana, Lafayette. 1976 Archaeological Survey of Bayou Teche, Vermilion River, and Freshwater Bayou, South Central Louisiana. Center for Archaeological Studies Report No. 2. University of Southwestern Louisiana, Lafayette.

1993

222

references cited

1978a Archaeological Survey of the Lower Atchafalaya Region, South Central Louisiana. Center for Archaeological Studies Report No. 5. University of Southwestern Louisiana, Lafayette. 1978b The Land of the Chitimacha: Historical and Archaeological Data Relative to the Geographic Territory Occupied and Utilized by the Chitimacha Indians. Archaeology, Inc., Lafayette, Louisiana. Submitted to John C. Hose, New Orleans. 1979 Perceptions of Atchafalaya Drainage Basin Archaeology. Prepared for U.S. Army Corps of Engineers, New Orleans District. 1980 Documentary Evidence Bearing on Chitimacha Land Claims in the Bayou Plaquemine Tract, Iberville Parish, Louisiana. Archaeology, Inc., Lafayette, Louisiana. Submitted to John C. Hose, New Orleans. 1982 Archaeology and Ethnology on the Edges of the Atchafalaya Basin: A Cultural Resources Survey of the Atchafalaya Protection Levees. Center for Archaeological Investigations, University of Southwestern Louisiana, Lafayette. Submitted to U.S. Army Corps of Engineers, New Orleans District. 1983 Ouachita Prehistory. Louisiana Archaeology 10:319–335. 1990 Archaeological Survey of the Mid-Teche Ridge, South Louisiana: From Bayou Gerimond to Bayou Portage Guidry. Center for Archaeological Studies, University of Southwestern Louisiana, Lafayette. 1996 Ancient Earthworks of the Ouachita Valley in Louisiana. Technical Reports No. 5. Southeast Archeological Center, Tallahassee. Giraud, Marcel 1974 A History of French Louisiana, Vol. 1: The Reign of Louis XIV, 1698–1715, translated by Joseph C. Lambert. Louisiana State University Press, Baton Rouge. Goldstein, Lynne G. 1976 Spatial Structure and Social Organization: Regional Manifestations of Mississippian Society. Unpublished Ph.D. dissertation, Department of Anthropology, Northwestern University, Evanston, Illinois. Goodwin, R. Christopher, Stephen Hinks, William P. Athens, Ralph Draughon, Jr., Jennifer A. Cohen, William A. Morgan, Allen R. Saltus, Jr., and Paul V. Heinrich 1991a Historical and Archaeological Investigations of Fort Bisland and Lower Bayou Teche, St. Mary Parish, Louisiana. R. Christopher Goodwin and Associates, Inc., New Orleans. Submitted to the U.S. Army Corps of Engineers, New Orleans District. Goodwin, R. Christopher, Stephen Hinks, William P. Athens, Jennifer A. Cohen, Ralph Draughon, Jr., and Paul V. Heinrich 1991b Supplemental Archaeological Investigations of Lower Bayou Teche, St. Mary Par-

references cited

223

ish, Louisiana. R. Christopher Goodwin and Associates, Inc., New Orleans. Submitted to the U.S. Army Corps of Engineers, New Orleans District. Goodwin, R. Christopher, Jill-Karen Yakubik, Galloway W. Shelby, Kenneth R. Jones, Debra Stayner, and Janice Cooper 1985 An Archaeological and Historic Sites Inventory of Bayou Teche between Franklin and Jeanerette, Louisiana, vol. 1. R. Christopher Goodwin and Associates, Inc., New Orleans. Submitted to the Division of Archaeology, Department of Culture, Recreation and Tourism, Baton Rouge. Green, John A. 1936 Governor Périer’s Expedition against the Natchez Indians. Louisiana Historical Quarterly 19(3):547–577. Greenwell, Dale 1984 The Mississippi Gulf Coast. In Perspectives on Gulf Coast Prehistory, edited by Dave D. Davis, pp. 125–155. Ripley P. Bullen Monograph in Anthropology and History No. 5. University Press of Florida, Gainesville. Gregory, Hiram F. 1969 Plaquemine Period Sites in the Catahoula Basin, A Cultural Microcosm in East-Central Louisiana. Louisiana Studies 8:111–134. Gregory, Hiram F., Clint Pine, William S. Baker, Jr., Aubra Lee, Reinaldo Barnes, and George A. Stokes 1987 A Survey of Catahoula Basin, 1987. Department of History and Social Science, Northwestern State University, Natchitoches, Louisiana. Submitted to Louisiana Division of Archaeology, Baton Rouge. Grif¤n, James B. 1943 The Fort Ancient Aspect: Its Cultural and Chronological Position in Mississippi Valley Archaeology. University of Michigan Press, Ann Arbor. 1946 Culture Change and Continuity in Eastern United States. In Man in Northeastern North America, edited by F. Johnson, pp. 37–95. Papers of the Robert S. Peabody Foundation for Archaeology 3. Peabody Foundation, Andover, Massachusetts. 1952 Culture Periods in Eastern United States Archeology. In Archeology of Eastern United States, edited by J. B. Grif¤n, pp. 352–364. University of Chicago Press, Chicago. 1967 Eastern North American Archaeology: A Summary. Science 156(3772):175–191. 1985 Changing Concepts of the Prehistoric Mississippian Cultures of the Eastern United States. In Alabama and the Borderlands: From Prehistory to Statehood, edited by R. R. Badger and L. A. Clayton, pp. 40–63. University of Alabama Press, Tuscaloosa. 1990 Review of Winterville: Late Prehistoric Culture Contact in the Lower Mississippi Valley, by Jeffrey P. Brain. Mississippi Archaeology 25(1):64–73.

224

references cited

Haag, William G. 1965 Louisiana in North American Prehistory. Southern Studies 4(4):279–323. 1978 A Prehistory of the Lower Mississippi River Valley. In Man and Environment in the Lower Mississippi Valley, edited by Sam B. Hilliard, pp. 1–8. Geoscience and Man No. 19. School of Geoscience, Louisiana State University, Baton Rouge. 1988 Plaquemine Culture. In Historical Dictionary of North American Archaeology, edited by Edward B. Jelks and Juliet C. Jelks, pp. 378. Greenwood Press, New York. Haas, Mary R. 1940 Tunica. Extract from Handbook of American Indian Languages, vol. 4, edited by Franz Boas, pp. 1–143. J. J. Augustin, New York. 1956 Natchez and the Muskogean Languages. Language 32:61–72. 1971 Southeastern Indian Linguistics. In Red, White, and Black, edited by Charles Hudson, pp. 44–54. Southern Anthropological Society Proceedings No. 5. University of Georgia Press, Athens. 1979 Southeastern Languages. In The Languages of Native America: Historical and Comparative Assessment, edited by L. Campbell and M. Mithun, pp. 299–326. University of Texas Press, Austin. Hall, John C. 1997 Report on the Trees of the Mound 3 Summit and the Mound 4 Flats. In Excavations at the Anna Site (22Ad500), Adams County Mississippi: A Preliminary Report, edited by Ian W. Brown, pp. 46–49. Unpublished manuscript, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama Museums, Tuscaloosa. Hall, Robert L. 1997 An Archaeology of the Soul: North American Indian Belief and Ritual. University of Illinois Press, Urbana. Haller, Mikael 1998 Zooarchaeological Analysis of Materials from the Mound 4, Block 1 Locale, the Anna Site, 22Ad500. Unpublished manuscript, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama Museums, Tuscaloosa. Hally, David J. 1967 Post-Coles Creek Cultural Development in the Upper Tensas Basin of Louisiana. Southeastern Archaeological Conference Bulletin 6:36–40. 1972 The Plaquemine and Mississippian Occupations of the Upper Tensas Basin, Louisiana. Unpublished Ph.D. dissertation. Department of Anthropology, Harvard University, Cambridge.

references cited

225

The Territorial Size of Mississippian Chiefdoms. In Archaeology of Eastern North America: Papers in Honor of Stephen Williams, edited by J. B. Stoltman, pp. 143–168. Archaeological Report No. 25. Mississippi Department of Archives and History, Jackson. 1996 Platform-Mound Construction and the Instability of Mississippian Chiefdoms. In Political Structure and Change in the Prehistoric Southeastern United States, edited by J. F. Scarry, pp. 92–127. University Press of Florida, Gainesville. 1999 The Settlement Pattern of Mississippian Chiefdoms in Northern Georgia. In Settlement Pattern Studies in the Americas: Fifty Years since Virú, edited by B. R. Billman and G. M. Feinman, pp. 96–115. Smithsonian Institution Press, Washington, D.C. Harmon, Anna M., and Jerome C. Rose 1989 Bioarchaeology of the Louisiana and Arkansas Study Area. In Archaeology and Bioarchaeology of the Lower Mississippi Valley and Trans-Mississippi South in Arkansas and Louisiana, edited by M. D. Jeter, J. C. Rose, I. Williams Jr., and A. M. Harmon, pp. 323–354. Arkansas Archaeological Survey, Fayetteville. Final report to the U.S. Army Corps of Engineers, Southwestern Division Study Unit 6 of the Southwestern Division Archaeological Overview. Hart, Charles W. M. 1943 A Reconsideration of the Natchez Social Structure. American Anthropologist 45(3, pt. 1), pp. 379–386. Hart, John P., David L. Asch, C. Margaret Scarry, and Gary W. Crawford 2002 The Age of the Common Bean (Phaseolus vulgaris L.) in the Northern Eastern Woodlands of North America. Antiquity 76(292):377–385. Hart, John P., and C. Margaret Scarry 1999 The Use of Common Beans (Phaseolus vulgaris) in the Northeastern United States. American Antiquity 64:653–658. Harty, James 1997 The Lindsey Site: A Mound Complex on Big Corney Bayou in Union Parish, Louisiana. Louisiana Archaeology 22:125–142. 2005 Archaeological Investigations at the Scott Place Mounds (16UN4), Union Parish, Louisiana. Louisiana Archaeology 26:55–61. Hayden, Brian 1998 Fabulous Feasts: A Prolegomenon to Archaeological and Theoretical Importance of Feasting. Paper presented in the Symposium “The Archaeological Importance of Feasting Rituals.” Meetings of the Society for American Archaeology, Seattle. 2001 Fabulous Feasts: A Prolegomenon to the Importance of Feasting. In Feasts: 1993

226

references cited

Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 23–64, Smithsonian Institution Press, Washington, D.C. Helms, Mary W. 1993 Craft and the Kingly Ideal. University of Texas Press, Austin. Hemmings, E. Thomas 1982 Human Adaptation in the Grand Marais Lowland: Intensive Archeological Survey and Testing in the Felsenthal Navigation Pool, Ouachita and Saline Rivers, Southern Arkansas. Arkansas Archeological Survey Research Series, No. 17. Arkansas Archeological Survey, Fayetteville. Hoffman, Teresa L. 1998 The Lithic Assemblage. In Toltec Mounds and Plum Bayou Culture: Mound D Excavations, by Martha Ann Rolingson, pp. 54–79. Arkansas Archeological Survey Research Series, No. 54. Arkansas Archeological Survey, Fayetteville. Holley, George R., and Gary B. DeMarcay 1977 Preliminary Report on the Prehistory of Barataria. Paper presented at the Third Annual Meeting of the Louisiana Archaeological Society, New Orleans. On ¤le with the Louisiana Division of Archaeology, Baton Rouge. Holmes, William H. 1886 Ancient Pottery of the Mississippi Valley. Fourth Annual Report of the Bureau of American Ethnology, 1882–1883, pp. 361–436. Government Printing Of¤ce, Washington, D.C. 1903 Aboriginal Pottery of the Eastern United States. Twentieth Annual Report of the Bureau of American Ethnology, 1898–1899. Government Printing Of¤ce, Washington, D.C. House, John H. 1982 Prehistory of the Felsenthal Region. In Human Adaptation in the Grand Marais Lowland: Intensive Archeological Survey and Testing in the Felsenthal Navigation Pool, Ouachita and Saline Rivers, Southern Arkansas, by E. Thomas Hemmings, pp. 51–77. Arkansas Archeological Survey Research Series, No. 17. Arkansas Archeological Survey, Fayetteville. 1986 Aboriginal Ceramics in the Sparta Mine Area: 1980 and 1982 Field Seasons. In Model Validation in Sparta, edited by Robert H. Lafferty III and John H. House, pp. 112–127. Arkansas Archeological Survey Research Reports, No. 25. Arkansas Archeological Survey, Fayetteville. House, John H., and Marvin D. Jeter 1994 Excavations at Boydell Mound A (3AS58), Southeast Arkansas. The Arkansas Archeologist 33:1–82.

references cited

227

Hudson, Charles 1985 De Soto in Arkansas: A Brief Synopsis. Arkansas Archeological Society Field Notes 205:3–12. 1993 Reconstructing the De Soto Expedition Route West of the Mississippi River: Summary and Contents. In The Expedition of Hernando de Soto West of the Mississippi, 1541–1543, edited by Gloria A. Young and Michael P. Hoffman, pp. 143–154. University of Arkansas Press, Fayetteville. Ingraham, Joseph H. 1835 The South-West. By a Yankee. 2 vols. Harper and Brothers, New York. Jackson, H. Edwin 1992 The Ables Creek Site: A Protohistoric Cemetery in Southeast Arkansas. Arkansas Archeological Survey Research Series, No. 42. Arkansas Archeological Survey, Fayetteville. Jackson, H. Edwin, and Susan L. Scott 1995 The Faunal Record of the Southeastern Elite: The Implications of Economy, Social Relations, and Ideology. Southeastern Archaeology 14(2):103–119. 2003 Patterns of Elite Faunal Utilization at Moundville, Alabama. American Antiquity 68(3):552–572. Jennings, Jesse D. 1940 Archeological Report on Anna Mound Group, Anna, Mississippi. Unpublished manuscript, National Park Service, Southeast Regional Of¤ce, Tallahassee. 1952 Prehistory of the Lower Mississippi Valley. In Archaeology of Eastern United States, edited by J. B. Grif¤n, pp. 256–271. University of Chicago Press, Chicago. Jennings, Jesse D., and C. A. Wagner 1940 Sketch Field Map of the Anna Mound Group, Anna, Mississippi. On ¤le with the Natchez Trace Parkway, National Park Service, Tupelo, Mississippi. Jeter, Marvin D. 1982 The Archeology of Southeast Arkansas: An Overview for the 1980s. In Arkansas Archeology in Review, edited by Neal L. Trubowitz and Marvin D. Jeter, pp. 76–131. Arkansas Archeological Survey Research Series, No. 15. Arkansas Archeological Survey, Fayetteville. 1986 Tunicans West of the Mississippi: A Summary of Early Historic and Archaeological Evidence. In The Protohistoric Period in the Mid-South: 1500– 1700, edited by David H. Dye and Ronald C. Brister, pp. 38–63. Archaeological Reports No. 18. Mississippi Department of Archives and History, Jackson. 1989 Protohistoric and Historic Native Americans. In Archeology and Bioarcheology

228

references cited

of the Lower Mississippi Valley and Trans-Mississippi South in Arkansas and Louisiana, edited by Marvin D. Jeter, Jerome C. Rose, G. Ishmael Williams, Jr., and Anna M. Harmon, pp. 221–248. Arkansas Archeological Survey Research Series, No. 37. Arkansas Archeological Survey, Fayetteville. 2002a From Prehistory through Protohistory to Ethnohistory in and near the Northern Lower Mississippi Valley. In The Transformation of the Southeastern Indians, 1540–1760, edited by Robbie Ethridge and Charles Hudson, pp. 177–223. University Press of Mississippi, Jackson. 2002b Shatter Zone Shock Waves along the Lower Mississippi. Paper presented at the 59th Southeastern Archaeological Conference, Biloxi, Mississippi. 2003a Review of Great Towns and Regional Polities in the Prehistoric American Southwest and Southeast, edited by Jill E. Neitzel. Southeastern Archaeology 22:112–114. 2003b Discussant Remarks. In Woodland-Mississippian Transition in the MidSouth: Proceedings of the Twenty-Second Mid-South Archaeological Conference, June 2–3, 2001, edited by Charles H. McNutt, Stephen Williams, and Marvin D. Jeter, pp. 73–76, 182–189. University of Memphis Anthropological Research Center, Occasional Papers, No. 25, Memphis. 2007a The Eagle Lake Mounds (3BR4): A Late Prehistoric Ceremonial Center in the Felsenthal Region, Southeast Arkansas. Report in preparation, for publication in the Arkansas Archeological Survey Research Series. 2007b The Saline-Fifteen Site (3BR119) and Late Prehistoric to Protohistoric Cultural Ecology in the Felsenthal Region Uplands, Southeast Arkansas. Report in preparation, for publication in the Arkansas Archeological Survey Research Series. 2007c Oliver, as Seen from Goldsmith Oliver—and Beyond. Chapter in preparation, to be included in a volume on the Oliver site in Mississippi, edited by John M. Connaway, for publication by the Mississippi Department of Archives and History, Jackson. Jeter, Marvin D. (editor) 2007 The Taylor Site (3DR2): Prehistoric Cultural Ecology and Ceremonialism along Bayou Bartholomew in Southeast Arkansas. Report in preparation, for publication in the Arkansas Archeological Survey Research Series. Jeter, Marvin D., John S. Belmont, Ann M. Early, John H. House, and Jeffrey M. Mitchem 1991 The Taylor Mounds Site (3DR2), Southeast Arkansas: A Preliminary Summary. Paper presented at the 48th Annual Meeting of the Southeastern Archaeological Conference, Jackson, Mississippi.

references cited

229

Jeter, Marvin D., and Ann M. Early 1999 Prehistory of the Saline River Drainage Basin, Central to Southeast Arkansas. In Arkansas Archaeology: Essays in Honor of Dan and Phyllis Morse, edited by Robert C. Mainfort, Jr., and Marvin D. Jeter, pp. 31–63. University of Arkansas Press, Fayetteville. Jeter, Marvin D., David B. Kelley, and George P. Kelley 1979 The Kelley-Grimes Site: A Mississippi Period Burial Mound, Southeast Arkansas, Excavated in 1936. The Arkansas Archeologist 20:1–51. Jeter, Marvin D., Henry S. McKelway, and Patsy White 2007 The Boydell “Catchment Survey,” Southeast Arkansas. Manuscript in preparation, on ¤le at UAM Station, Arkansas Archeological Survey. Jeter, Marvin D., Jerome C. Rose, G. Ishmael Williams, Jr., and Anna M. Harmon 1989 Archeology and Bioarcheology of the Lower Mississippi Valley and TransMississippi South in Arkansas and Louisiana. Arkansas Archeological Survey Research Series, No. 37. Arkansas Archeological Survey, Fayetteville. Jeter, Marvin D., and G. Ishmael Williams, Jr. 1989 Late Prehistoric Cultures, AD 1000–1500. In Archeology and Bioarcheology of the Lower Mississippi Valley and Trans-Mississippi South in Arkansas and Louisiana, edited by Marvin D. Jeter, Jerome C. Rose, G. Ishmael Williams, Jr., and Anna M. Harmon, pp. 171–220. Arkansas Archeological Survey Research Series, No. 37. Arkansas Archeological Survey, Fayetteville. Joffe, Alexander H. 1998 Alcohol and Social Complexity in Ancient Western Asia. Current Anthropology 39(3):297–322. Johnson, Hunter B. 1997 An Archaeological Survey of the Lower Big Black River Valley near Vicksburg, Mississippi. Unpublished Master’s thesis, Department of Anthropology, University of Alabama, Tuscaloosa. Johnson, Jay K., Abigayle Robbins, and John T. Sparks 1983a Excavations at the Mud Island Creek Archaeological Complex and the Gordon Mounds Site, Jefferson County, Mississippi. Archaeological Papers No. 4. Center for Archaeological Research at the University of Mississippi, Oxford. 1983b Prehistoric Subsistence in the Natchez Bluffs, Jefferson County, Mississippi. Mississippi Archaeology 18(2):4–12. Jones, Dennis, and Malcolm K. Shuman 1987 Archaeological Atlas and Report of Prehistoric Indian Mounds in Louisiana, Vol. 1: Ascension, Iberville, Point Coupee, St. James, and West Baton Rouge Parishes. Museum of Geoscience, Louisiana State University. Submitted to the

230

references cited

Louisiana Division of Archaeology, Department of Culture, Recreation and Tourism, Of¤ce of Cultural Development, Baton Rouge. 1991 Archaeological Atlas and Report of Prehistoric Indian Mounds in Louisiana, Vol. 6: Acadia, Lafayette & St. Landry Parishes. Museum of Geoscience, Louisiana State University. Submitted to the Louisiana Division of Archaeology, Department of Culture, Recreation and Tourism, Of¤ce of Cultural Development, Baton Rouge. Jones, Reca Bamberg 1984 The Panola Plantation Site: A Plaquemine Burial Mound in Northeast Louisiana. Louisiana Archaeology 11:139–164. Keesing, Roger M. 1983 ’Elota’s Story: The Life and Times of a Solomon Islands Big Man. Holt, Rinehart, and Wilson, Fort Worth. Kelley, David B. 1984 Aboriginal Occupation of the Upper Felsenthal Region. In Archaeology and Paleogeography of the Upper Felsenthal Region: Cultural Resources Investigations in the Calion Navigation Pool, South-Central Arkansas, by Richard A. Weinstein and David B. Kelley, pp. 25–49. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. 1990 Coles Creek Period Faunal Exploitation in the Ouachita River Valley of Southern Arkansas: The Evidence from the Paw Paw Site. Unpublished Ph.D. dissertation, Department of Anthropology, Tulane University, New Orleans. 1992 Coles Creek Period Faunal Exploitation in the Ouachita River Valley of Southern Arkansas. Mid-Continental Journal of Archaeology 17:227–264. Kelley, David B., Douglas C. Wells, Dana B. Lee, Richard A. Weinstein, and Josette LeBoeuf 2000 Cultural Resources Evaluation of the Lower Atchafalaya Backwater Area, South Louisiana. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, New Orleans District. Kelly, John E. 1987 Emergent Mississippian and the Transition from Late Woodland to Mississippian: The American Bottom Case for a New Concept. In The Emergent Mississippian: Proceedings of the Sixth Mid-South Conference, June 6–9, 1985, edited by R. A. Marshall, pp. 212–226. Occasional Papers 87-01. Cobb Institute of Archaeology, Mississippi State University, Starkville. 1990 The Emergence of Mississippian Culture in the American Bottom Region.

references cited

231

In The Mississippian Emergence, edited by B. D. Smith, pp. 113–152. Smithsonian Institution Press, Washington, D.C. Kelly, Lucretia S. 2001 A Case of Ritual Feasting at the Cahokia Site. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 334–367. Smithsonian Institution Press, Washington, D.C. Kidder, Tristram R. 1988 The Koroa Indians of the Lower Mississippi Valley. Mississippi Archaeology 23(2):1–42. 1990a Ceramic Chronology and Culture History of the Southern Ouachita Basin: Coles Creek to the Early Historic Period. Midcontinental Journal of Archaeology 15(1):51–99. 1990b Final Report on the 1989 Archaeological Investigations at the Osceola (16TE2) and Reno Brake (16TE93) Sites, Tensas Parish, Louisiana. Center for Archaeology, Archaeological Report No. 1. Tulane University, New Orleans. 1991 Review of Coles Creek and Mississippi Period Foragers in the Felsenthal Region of the Lower Mississippi Valley, edited by Frank F. Schambach. Mississippi Archaeology 26(2):78–85. 1992a Coles Creek Period Social Organization and Evolution in Northeast Louisiana. In Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America, edited by A. W. Barker and T. R. Pauketat, pp. 145– 162. Archeological Papers of the American Anthropological Association, No. 3. American Anthropological Association, Arlington, Virginia. 1992b Timing and Consequences of the Introduction of Maize Agriculture in the Lower Mississippi Valley. North American Archaeology 13(1):15–41. 1993a 1992 Archaeological Test Excavations in Tensas Parish, Louisiana. Center for Archaeology, Archaeological Report No. 2. Tulane University, New Orleans. 1993b The Glendora Phase: Protohistoric–Early Historic Culture Dynamics on the Lower Ouachita River. In Archaeology of Eastern North America: Papers in Honor of Stephen Williams, edited by J. B. Stoltman, pp. 231–260. Archaeological Report No. 25. Mississippi Department of Archives and History, Jackson. 1994 Matheny: A Multicomponent Site on Bayou Bartholomew, Northeast Louisiana. Mid-Continental Journal of Archaeology 19:137–169. 1995 Archaeological Data Recovery at 16JE218, Jefferson Parish, Louisiana. Report submitted to the U.S. Army Corps of Engineers, New Orleans District, and on ¤le with the Louisiana Division of Archaeology, Baton Rouge.

232

references cited

1998a Coles Creek Culture. In Archaeology of Prehistoric Native America: An Encyclopedia, edited by Guy Gibbon, pp. 171–173. Garland Publishing, New York. 1998b Mississippi Period Mound Groups and Communities in the Lower Mississippi Valley. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R. B. Lewis and C. Stout, pp. 123–150. University of Alabama Press, Tuscaloosa. 1998c Rethinking Caddoan–Lower Valley Interaction. In The Native History of the Caddo: Their Place in Southeastern Archaeology and Ethnohistory, edited by T. Pertulla and J. Bruseth, pp. 129–143. Studies in Archeology, vol. 30. Texas Archeological Research Laboratory, University of Texas at Austin. 1998d The Rat That Ate Louisiana: Aspects of Historical Ecology in the Mississippi River Delta. In Advances in Historical Ecology, edited by William Balee, pp. 141–168. Columbia University Press, New York. 2002 Woodland Period Archaeology of the Lower Mississippi Valley. In The Woodland Southeast, edited by David G. Anderson and Robert C. Mainfort, Jr., pp. 66–90. University of Alabama Press, Tuscaloosa. 2004a Plazas as Architecture: An Example from the Raffman Site, Northeast Louisiana. American Antiquity 69:514–532. 2004b Prehistory of the Lower Mississippi Valley after 800 B.C. In Southeast, edited by Raymond D. Fogelson, pp. 545–559. Handbook of North American Indians, vol. 14, W. Sturtevant, general editor. Smithsonian Institution Press, Washington, D.C. Kidder, Tristram R., and Gayle J. Fritz 1993 Investigating Subsistence and Social Change in the Lower Mississippi Valley: The 1989 and 1990 Excavations at the Reno Brake and Osceola Sites, Tensas Parish, Louisiana. Journal of Field Archaeology 20(3):281–297. Kidder, Tristram R., Gayle J. Fritz, and Christopher Smith 1993 Emerson (16TE104). In 1992 Archaeological Test Excavations in Tensas Parish, Louisiana, by T. R. Kidder, pp. 100–137. Center for Archaeology, Archaeological Report No. 2. Tulane University, New Orleans. Kidder, Tristram R., and Lori M. Roe 1999 Archaeology at the Raffman Site, Madison Parish, Louisiana. Paper presented at the 56th Annual Meeting of the Southeastern Archaeological Conference, Pensacola. Kidder, T. R., and D. C. Wells 1992 Baytown Period Settlement Organization in the Lower Mississippi Valley. Paper presented at the 49th Annual Meeting of the Southeastern Archaeological Conference, Little Rock.

references cited

233

King, Adam 2001 Long-Term Histories of Mississippian Centers: The Developmental Sequence of Etowah and Its Comparison to Moundville and Cahokia. Southeastern Archaeology 20:1–17. 2003 Etowah: The Political History of a Chiefdom Capital. University of Alabama Press, Tuscaloosa. Kniffen, Fred B. 1936 Preliminary Report on the Indian Mounds and Middens of Plaquemines and St. Bernard Parishes. In Lower Mississippi River Delta: Reports on the Geology of Plaquemines and St. Bernard Parishes, edited by Richard J. Russell, H. V. Howe, J. H. McGuirt, C. F. Dohm, W. Hadley, Jr., F. B. Kniffen, and C. A. Brown, pp. 407–422. Geological Bulletin No. 8. Geological Survey, Louisiana Department of Conservation, New Orleans. 1938 Indian Mounds of Iberville Parish. In Reports on the Geology of Iberville and Ascension Parishes, pp. 189–207. Geological Bulletin No. 13. Geological Survey, Louisiana Department of Conservation, New Orleans. Kniffen, Fred B., Hiram F. Gregory, and George A. Stokes 1987 The Historic Indian Tribes of Louisiana: From 1542 to the Present. Louisiana State University Press, Baton Rouge. Knight, Vernon J., Jr. 1986 The Institutional Organization of Mississippian Religion. American Antiquity 51:675–687. 1990a Excavations of the Truncated Mound at the Walling Site. Alabama State Museum of Natural History, Division of Archaeology, Report of Investigations 56. University of Alabama, Tuscaloosa. 1990b Social Organization and the Evolution of Hierarchy in Southeastern Chiefdoms. Journal of Anthropological Research 46(1):1–23. 1997 Some Developmental Parallels between Cahokia and Moundville. In Cahokia: Domination and Ideology in the Mississippian World, edited by T. R. Pauketat and T. E. Emerson, pp. 229–247. University of Nebraska Press, Lincoln. 1998 Moundville as a Diagrammatic Ceremonial Center. In Archaeology of the Moundville Chiefdom, edited by V. J. Knight, Jr., and V. P. Steponaitis, pp. 44–62. Smithsonian Institution Press, Washington, D.C. 2001 Feasting and the Emergence of Platform Mound Ceremonialism in Eastern North America. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 311–333. Smithsonian Institution Press, Washington, D.C. 2003 Discussant Comments. In Woodland-Mississippian Transition in the Mid-

234

references cited

South: Proceedings of the Twenty-Second Mid-South Archaeological Conference, June 2–3, 2001, edited by Charles H. McNutt, Stephen Williams, and Marvin D. Jeter, pp. 133–135. University of Memphis Anthropological Research Center, Occasional Papers, No. 25, Memphis. Knight, Vernon J., Jr., and Vincas P. Steponaitis 1998 A New History of Moundville. In Archaeology of the Moundville Chiefdom, edited by V. J. Knight, Jr., and V. P. Steponaitis, pp. 1–25. Smithsonian Institution Press, Washington, D.C. Knight, Vernon J., Jr., and Vincas P. Steponaitis (editors) 1998 Archaeology of the Moundville Chiefdom. Smithsonian Institution Press, Washington D.C. Kolb, Michael J. 1994 Ritual Activity and Chie®y Economy at an Upland Religious Site on Maui, Hawai’i. Journal of Field Archaeology 21(4):417–436. Kottak, C. P. 1999 The New Ecological Anthropology. American Anthropologist 101(1):23–35. Krause, Richard A. 1985 The Clay Sleeps. University of Alabama Press, Tuscaloosa. Krieger, Alex D. 1946 Culture Complexes and Chronology in Northern Texas, With Extension of Pueblo Datings to the Mississippi Valley. University of Texas, Austin. Leach, Edmond R. 1964 Political Systems of Highland Burma. London School of Economics Monographs on Social Anthropology No. 44. Athlone Press, London. LeBlanc, Steven A. 1999 Prehistoric Warfare in the American Southwest. University of Utah Press, Salt Lake City. 2001 Warfare and Aggregation in the El Morro Valley, New Mexico. In Deadly Landscapes: Case Studies in Prehistoric Southwestern Warfare, edited by Glen E. Rice and Steven A. LeBlanc, pp. 19–49. University of Utah Press, Salt Lake City. LeBlanc, Steven A., and Glen E. Rice 2001 Southwestern Warfare: The Value of Case Studies. In Deadly Landscapes: Case Studies in Prehistoric Southwestern Warfare, edited by Glen E. Rice and Steven A. LeBlanc, pp. 1–18. University of Utah Press, Salt Lake City. Lekson, Stephen H. 2002 War in the Southwest, War in the World. American Antiquity 67:607–624. Lemley, Harry J., and Samuel D. Dickinson 1937 Archeological Investigations on Bayou Macon in Arkansas. Bulletin of the Texas Archeological and Paleontological Society 9:11–47.

references cited

235

Le Page du Pratz, Antoine S. 1758 Histoire de la Louisiane. 3 vols. Paris. 1972 [1774] The History of Louisiana, or of the Western Parts of Virginia and Carolina. Claitor’s Publishing Division, Baton Rouge. Lewis, R. Barry, Charles Stout, and Cameron B. Wesson 1998 The Design of Mississippian Towns. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R. B. Lewis and C. Stout, pp. 1–21. University of Alabama Press, Tuscaloosa. Lightfoot, Kent G. 1995 Culture Contact Studies: Rede¤ning the Relationship between Prehistoric and Historical Archaeology. American Antiquity 60(2)199–217. 2001 Traditions as Cultural Production: Implications for Contemporary Archaeological Research. In The Archaeology of Traditions: Agency and History before and after Columbus, edited by T. R. Pauketat, pp. 237–252. University Press of Florida, Gainesville. Livingood, Patrick 1999 Investigation of Mississippian Mounds on the Middle Pearl River, Mississippi. Paper presented at the 56th Annual Meeting of the Southeastern Archaeological Conference, Pensacola. 2002 The Study of Ceramic Temper Using Digital Image Analysis. Paper presented at the 59th Annual Meeting of the Southeastern Archaeological Conference, Biloxi, Mississippi. 2004 Digital Image Analysis of Ceramic Thin-Sections: Present and Future. Paper presented at the 61st Annual Meeting of the Southeastern Archaeological Conference, St. Louis. Lopinot, Neal H. 1990 Floral Remains from the Goldsmith Oliver 2 Site. In Goldsmith Oliver 2 (3PU306): A Protohistoric Archeological Site near Little Rock, Arkansas, by Marvin D. Jeter, Kathleen H. Cande, and John J. Mintz, pp. 424–482. Arkansas Archeological Survey, Fayetteville. Submitted to Federal Aviation Administration, Fort Worth. 2003 Lifeways: An Archaeobotanical Perspective on the Transition. In WoodlandMississippian Transition in the Mid-South: Proceedings of the Twenty-Second Mid-South Archaeological Conference, June 2–3, 2001, edited by Charles H. McNutt, Stephen Williams, and Marvin D. Jeter, pp. 142–149. University of Memphis Anthropological Research Center, Occasional Papers, No. 25, Memphis. 2007 Archeobotany of the Saline-Fifteen-A Site. In The Saline-Fifteen Site (3BR119) and Late Prehistoric to Protohistoric Cultural Ecology in the

236

references cited

Felsenthal Region Uplands, Southeast Arkansas. Report in preparation, for publication in the Arkansas Archeological Survey Research Series. Loren, Diana A. 1995 Forgotten Actors: French Interest Groups, Natchez Indians and African Slaves in Louisiana: 1699–1731. Unpublished Master’s thesis, Department of Anthropology, State University of New York, Binghamton. Lorenz, K. G. 1997 A Re-examination of Natchez Sociopolitical Complexity: A View from the Grand Village and Beyond. Southeastern Archaeology 16(2):97–112. McCullough, David L. 1981 Natchez Trace Parkway: An Archeological Testing Project at the Gordon Mounds Site, Jefferson County, Mississippi. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. McGimsey, Charles R. 1995 Annual Report for Management Unit 3, 1994/1995 Grant Year. Regional Archaeology Program. Submitted to the Louisiana Division of Archaeology, Baton Rouge. 1998 Chenier Plains and Civil Wars: Archaeology in Southwest Louisiana. Regional Archaeology Program, Management Unit III. 1997/1998 Annual Report. Submitted to the Louisiana Division of Archaeology, Baton Rouge. 2001 The Rings of Marksville, and Other Stories of Southwest Louisiana Archaeology. Regional Archaeology Program, Management Unit III. 2000/2001 Annual Report. Submitted to the Louisiana Division of Archaeology, Baton Rouge. McGimsey, Charles R., and Daniel Cring 2003 Report of Investigations, Burial 1, Patterson Mounds Site (16SMY10), St. Mary Parish, Louisiana. Submitted to the Louisiana Division of Archaeology, Baton Rouge. McGimsey, Charles R., and Josette van der Koogh 2001 Louisiana’s Archaeological Radiometric Database. Special Publication No. 3. Louisiana Archaeological Society, Baton Rouge. McIntire, William G. 1958 Prehistoric Indian Settlements of the Changing Mississippi River Delta. Louisiana State University Press, Baton Rouge. McKelway, Henry S. 1990 The McClendon Site: A Protohistoric Cemetery in Southeast Arkansas. The Arkansas Archeologist 29:37–99. McNutt, C. H. 1996 The Central Mississippi Valley: A Summary. In Prehistory of the Central Mississippi Valley, edited by C. H. McNutt, pp. 187–257. University of Alabama Press, Tuscaloosa.

references cited

237

McWilliams, Richebourg G. (editor and translator) 1988 Fleur de Lys and Calumet, Being the Pénicaut Narrative of French Adventure in Louisiana. University of Alabama Press, Tuscaloosa. Originally published 1953, Louisiana State University Press, Baton Rouge. Mainfort, Robert C., Jr. 2004 Arkansas History and Prehistory in Review: The Nodena Phase. Arkansas Archeological Society Field Notes 318:3–10. Malinowski, Bronislaw 1922 Argonauts of the Western Paci¤c. E. P. Dutton, New York. Mann, C. B. 1988 Archaeological Classi¤cation of Ceramics from the Pearl River Mounds (22Lw510), Lawrence County, Mississippi. Unpublished Master’s thesis, Department of Anthropology and Sociology, University of Southern Mississippi, Hattiesburg. Mann, Rob 2001 2001 Annual Report for Management Units IV and V. Regional Archaeology Program, Museum of Natural Science, Louisiana State University. Division of Archaeology, Department of Culture, Recreation and Tourism, Baton Rouge. 2002 2002 Annual Report for Management Units IV and V. Regional Archaeology Program, Museum of Natural Science, Louisiana State University. Division of Archaeology, Department of Culture, Recreation and Tourism, Baton Rouge. 2005 Intruding on the Past: The Reuse of Ancient Earthen Mounds by Native Americans. Southeastern Archaeology 24(1):1–10. Manning, K., P. C. Armstrong, E. C. Poplin, and R. C. Goodwin 1987 Cultural Resources Survey of East Atchafalaya Basin Protection Levee Item E-44, Iberville Parish, Louisiana. R. Christopher Goodwin and Associates, Inc., New Orleans. Submitted to U.S. Army Corps of Engineers, New Orleans District. Martinez, C. A. 1981 Archeological Data Recovery Plan, Gordon Mounds Site, Jefferson County, Mississippi. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. Mason, Carol I. 1964 Natchez Class Structure. Ethnohistory 11(2):120–133. Mauss, Marcel 1990 The Gift: The Form and Reason for Exchange in Archaic Societies, translated by W. D. Halls. W. M. Norton, New York. Originally published 1950, Presses Universitaires de France.

238

references cited

Melancon, Mark A. 1999 Seriation of Certain Tchefuncte Ceramic Decoration Styles in Avoyelles, Lafayette, St. Landry, and St. Martin Parishes, Louisiana. Louisiana Archaeology 23:37–47. Miller, Cinder Grif¤n, R. Christopher Goodwin, Glenn Walter, T. R. Kidder, Marie Danforth, Roger T. Saucier, John Jones, William Hayden, Allen Green, Angele Montana, Nathanael Heller, Charlene Keck, Michele Williams, and Dave D. Davis 2000 Phase III Data Recovery Excavations at Site 16LF66, the Discovery Site, a Site Identi¤ed within the Discovery Producer Services LLC Larose Gas Processing Plant in Lafourche Parish, Louisiana. Discovery Gas Transmission LLC, St. Rose, Louisiana. R. Christopher Goodwin and Associates, Inc., New Orleans. Mills, Barbara J. 2004 Identity, Feasting, and the Archaeology of the Greater Southwest. In Identity, Feasting, and the Archaeology of the Greater Southwest, edited by Barbara J. Mills, pp. 1–26. University Press of Colorado, Boulder. Milner, George R. 1984 The Julien Site. American Bottom Archaeology, FAI-270 Site Reports, vol. 7. University of Illinois Press, Urbana. Milner, George R., and Sissel Schroeder 1999 Mississippian Sociopolitical Systems. In Great Towns and Regional Polities in the Prehistoric American Southwest and Southeast, edited by J. E. Neitzel, pp. 95–108. Amerind Foundation New World Studies Series 3. University of New Mexico Press, Albuquerque. Mires, Ann Marie, and Douglas W. Owsley 1984 Analysis of Human Skeletal Remains from the Little Mud Lake Site (3CA265). In Archaeology and Paleogeography of the Upper Felsenthal Region: Cultural Resources Investigations in the Calion Navigation Pool, South-Central Arkansas, by Richard A. Weinstein and David B. Kelley, Appendix A, pp. 555–573. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Misner, Elizabeth J., and Elizabeth J. Reitz 1994 Vertebrate Fauna from 16SC27. In Cultural Resources Survey and Testing for the Davis Pond Freshwater Diversion, St. Charles Parish, Louisiana, vol. 2, by Kenneth R. Jones, Herschel A. Franks, and Tristram R. Kidder, pp. 439– 502. Earth Search, Inc. Submitted to U.S. Army Corps of Engineers, New Orleans District. Mooney, James T., Susan Wilkerson, Troy Mead, and James P. Wilson 2004 Remaining Phase I Archaeological Survey and Phase II Testing of the Southeast

references cited

239

Arkansas I-69 Corridor. Michael Baker Jr., Inc., Field Of¤ce, White Hall, Arkansas. Submitted to Arkansas Highway and Transportation Department, Little Rock. Moore, Clarence B. 1905 Certain Aboriginal Remains of the Black Warrior River. Journal of the Academy of Natural Sciences of Philadelphia 13:125–244. Reprinted in 1996 in The Moundville Expeditions of Clarence Bloom¤eld Moore, edited by Vernon James Knight, Jr., pp. 21–142. University of Alabama Press, Tuscaloosa. 1909 Antiquities of the Ouachita Valley. Journal of the Academy of Natural Sciences of Philadelphia 14:1–170. Reprinted in 2003 in The Louisiana and Arkansas Expeditions of Clarence Bloom¤eld Moore, edited by Richard A. Weinstein, David B. Kelley, and Joe W. Saunders, pp. 215–394. University of Alabama Press, Tuscaloosa. 1913 Some Aboriginal Sites in Louisiana and in Arkansas: Atchafalaya River, Lake Larto, Tensas River, Bayou Maçon, and Bayou D’Arbonne in Louisiana; Saline River in Arkansas. Journal of the Academy of Natural Sciences of Philadelphia 16:5–93. Reprinted in 2003 in The Louisiana and Arkansas Expeditions of Clarence Bloom¤eld Moore, edited by Richard A. Weinstein, David B. Kelley, and Joe W. Saunders, pp. 567–659. University of Alabama Press, Tuscaloosa. 1911 Some Aboriginal Sites on Mississippi River. Journal of the Academy of Natural Sciences of Philadelphia 14:365–480. Reprinted in 1998 in The Lower Mississippi Valley Expeditions of Clarence Bloom¤eld Moore, edited by Dan F. Morse and Phyllis A. Morse, pp. 25–144. University of Alabama Press, Tuscaloosa. Moorehead, Warren K. 1924 Field Notes, Natchez, Mississippi, Mounds and Village Sites. Unpublished manuscript on ¤le at the Robert S. Peabody Foundation, Phillips Academy, Andover, Massachusetts. 1932 Explorations near Natchez, Mississippi. In The Etowah Papers, Exploration of the Etowah Site in Georgia, pp. 158–165. Yale University Press, New Haven, and Oxford University Press, London. Morgan, William N. 1999 Precolumbian Architecture in Eastern North America. The Ripley P. Bullen Series, Florida Museum of Natural History. University Press of Florida, Gainesville. Morse, Dan F., and Phyllis A. Morse 1983 Archaeology of the Central Mississippi Valley. Academic Press, New York.

240

references cited

Muller, Jon 1997 Mississippian Political Economy. Plenum Press, New York. 1998 Towns along the Lower Ohio. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R. B. Lewis and C. Stout, pp. 179–199. University of Alabama Press, Tuscaloosa. Nass, John P., Jr., and Richard W. Yerkes 1995 Social Differentiation in Mississippian and Fort Ancient Societies. In Mississippian Communities and Households, edited by J. D. Rogers and B. D. Smith, pp. 58–80. University of Alabama Press, Tuscaloosa. Nassaney, Michael S. 1991 Spatial-Temporal Dimensions of Social Integration during the Coles Creek Period in Central Arkansas. In Stability, Transformation, and Variation: The Late Woodland Southeast, edited by M. S. Nassaney and C. R. Cobb, pp. 177–221. Plenum Press, New York. 1992 Communal Societies and the Emergence of Elites in the Prehistoric American Southeast. In Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America, edited by A. W. Barker and T. R. Pauketat, pp. 111–143. Archeological Papers of the American Anthropological Association, No. 3. American Anthropological Association, Arlington, Virginia. 2001 The Historical-Processual Development of Late Woodland Societies. In The Archaeology of Traditions: Agency and History before and after Columbus, edited by T. R. Pauketat, pp. 157–173. University Press of Florida, Gainesville. Neitzel, Jill E. 2000 What Is a Regional System? Issues of Scale and Interaction in the Prehistoric Southwest. In The Archaeology of Regional Interaction: Religion, Warfare and Exchange across the American Southwest and Beyond, edited by M. Hegmon, pp. 25–40. University Press of Colorado, Boulder. Neitzel, Jill E., and David G. Anderson 1999 Multiscalar Analyses of Middle Range Societies: Comparing the Late Prehistoric Southwest and Southeast. In Great Towns and Regional Polities in the Prehistoric American Southwest and Southeast, edited by J. E. Neitzel, pp. 243–254. Amerind Foundation New World Studies Series 3. University of New Mexico Press, Albuquerque. Neitzel, Robert S. 1964 The Natchez Grand Village. Florida Anthropologist 17(2):63–66. 1965 Archeology of the Fatherland Site: The Grand Village of the Natchez. Anthropological Papers of the American Museum of Natural History, No. 51, Part 1. American Museum of Natural History, New York.

references cited

241

A Doublebarreled Detective Story. In Codex Wauchope: A Tribute Roll, edited by M. Giardino, B. Edmonson, and W. Creamer, pp. 19–34. Human Mosaic, Tulane University, New Orleans. 1981 A Suggested Technique for Measuring Landscape Changes through Archaeology. In Traces of Prehistory: Papers in Honor of William G. Haag, edited by F. H. West and R. Neuman, pp. 57–70. Geoscience and Man No. 22. School of Geoscience, Louisiana State University, Baton Rouge. 1983 The Grand Village of the Natchez Revisited: Excavations at the Fatherland Site, Adams County, Mississippi, 1972. Archaeological Report No. 12. Mississippi Department of Archives and History, Jackson. Neuman, Robert W. 1972 Archaeological Investigations at the Morton Shell Mound, Weeks Island, Iberia Parish, Louisiana. Report submitted to the National Science Foundation, Washington, D.C. 1977 An Archaeological Assessment of Coastal Louisiana. Melanges 11. Museum of Geoscience, Louisiana State University, Baton Rouge. 1984 An Introduction to Louisiana Archaeology. Louisiana State University Press, Baton Rouge. 2002 Louisiana Archaeology: A Selective History. In Histories of Southeastern Archaeology, edited by S. Tushingham, J. Hill, and C. H. McNutt, pp. 88–98. University of Alabama Press, Tuscaloosa. Neuman, Robert W., and A. Frank Servello 1976 Atchafalaya Basin Archaeological Survey. Department of Geography and Anthropology, School of Geoscience, Louisiana State University. Submitted to U.S. Army Corps of Engineers, New Orleans District. Nicklin, K. 1981 Pottery Production and Distribution in Southeast Nigeria. In Production and Distribution: A Ceramic Viewpoint, edited by H. Howard and E. Morris, pp. 169–186. BAR International Series 120. British Archaeological Reports, Oxford. Orlove, B. S. 1980 Ecological Anthropology. Annual Review of Anthropology 9:235–273. Oswalt, Wendell H., and Sharlotte Neely 1999 The Natchez: Sophisticated Farmers of the Deep South. In This Land Was Theirs: A Study of Native Americans, 6th ed., by W. H. Oswalt and S. Neely, pp. 477–500. May¤eld Publishing, Mountain View, California. Patterson, Thomas C. 1994 Toward a Properly Historical Ecology. In Historical Ecology: Cultural Knowledge and Changing Landscapes, edited by C. L. Crumley, pp. 223–238. School of American Research Press, Santa Fe. 1978

242

references cited

Pauketat, Timothy R. 1992 The Reign and Ruin of the Lords of Cahokia: A Dialectic of Dominance. In Lords of the Southeast: Social Inequality and the Native Elites of Southeastern North America, edited by A. W. Barker and T. R. Pauketat, pp. 31–51. Archeological Papers of the American Anthropological Association, No. 3. American Anthropological Association, Arlington, Virginia. 1995 Cahokian Political History as Punctuated Disequilibrium. Paper presented at the 60th Annual Meeting of the Society for American Archaeology, Minneapolis. 1997a Cahokian Political Economy. In Cahokia: Domination and Ideology in the Mississippian World, edited by T. R. Pauketat and T. E. Emerson, pp. 30–51. University of Nebraska Press, Lincoln. 1997b Specialization, Political Symbols, and the Crafty Elite of Cahokia. Southeastern Archaeology 16(1)1–15. 1998a Re¤guring the Archaeology of Greater Cahokia. Journal of Archaeological Research 6(1):45–89. 1998b The Archaeology of Downtown Cahokia: The Tract 15A and Dunham Tract Excavations. Studies in Archaeology, No. 1. Illinois Transportation Archaeological Research Program, University of Illinois, Urbana. 2000 The Tragedy of the Commoners. In Agency in Archaeology, edited by MarciaAnne Dobres and John E. Robb, pp. 113–129. Routledge, New York. 2001a A New Tradition in Archaeology. In The Archaeology of Traditions: Agency and History before and after Columbus, edited by T. R. Pauketat, pp. 1–16. University Press of Florida, Gainesville. 2001b Practice and History in Archaeology: An Emerging Paradigm. Anthropological Theory 1(1):73–98. 2002 A Fourth-Generation Synthesis of Cahokia and Mississippianization. Midcontinental Journal of Archaeology 27(2):149–170. 2004 Ancient Cahokia and the Mississippians. Cambridge University Press, New York. Pauketat, Timothy R., and Thomas E. Emerson (editors) 1997 Cahokia: Domination and Ideology in the Mississippian World. University of Nebraska Press, Lincoln. Pauketat, Timothy R., L. S. Kelly, G. J. Fritz, N. H. Lopinot, S. Elias, and E. Hargrave 2002 The Residues of Feasting and Public Ritual at Early Cahokia. American Antiquity 67(2):257–279. Peebles, Christopher S. 1978 Determinants of Settlement Size and Location in the Moundville Phase. In Mississippian Settlement Patterns, edited by B. D. Smith, pp. 369–416. Academic Press, New York.

references cited

243

Peebles, Christopher S., and Susan M. Kus 1977 Some Archaeological Correlates of Ranked Societies. American Antiquity 42(3):421–448. Pennisi, Elizabeth 2001 Tracking the Sexes by Their Genes. Science 291:1733–1734. Perino, Gregory 1968 Guide to the Identi¤cation of Certain American Indian Projectile Points. Special Bulletin No. 3, Oklahoma Anthropological Society, Cheyenne. Perodie, James R. 2001 Feasting for Prosperity: A Study of Southern Northwest Coast Feasting. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 185–214. Smithsonian Institution Press, Washington, D.C. Phelps, Dawson A., and Jesse D. Jennings 1940s A Preliminary Study of the Natchez Trace, Proposed Locations of the Natchez Trace Parkway, and Historical and Archeological Sites in and near Natchez, Mississippi. Unpublished Report, Southeast Archeological Center, National Park Service, Tallahassee. Phillips, David A., and Lynne Sebastian 2004 Large-Scale Feasting and Politics: An Essay on Power in Precontact Southwestern Societies. In Identity, Feasting, and the Archaeology of the Greater Southwest, edited by Barbara J. Mills, pp. 233–260, University Press of Colorado, Boulder. Phillips, Philip 1970 Archaeological Survey in the Lower Yazoo Basin, Mississippi, 1949–1955. Papers of the Peabody Museum of American Archaeology and Ethnology, No. 60, Parts 1 and 2. Harvard University, Cambridge. Phillips, Philip, and James A. Brown 1978 Pre-Columbian Shell Engravings from the Craig Mound at Spiro, Oklahoma, pt. 1. Peabody Museum Press, Harvard University, Cambridge. Phillips, Philip, James A. Ford, and James B. Grif¤n 1951 Archaeological Survey in the Lower Mississippi Alluvial Valley, 1940–1947. Papers of the Peabody Museum of American Archaeology and Ethnology, No. 25. Harvard University, Cambridge. Phillips, Philip, and Gordon R. Willey 1953 Method and Theory in American Archaeology: An Operational Basis for Culture-Historical Integration. American Anthropologist 55:615–633. Potter, James M., and Scott G. Ortman 2004 Community and Cuisine in the Prehispanic Southwest. In Identity, Feast-

244

references cited

ing, and the Archaeology of the Greater Southwest, edited by Barbara J. Mills, pp. 173–191. University Press of Colorado, Boulder. Prentice, Guy 2000 Ancient Indian Architecture of the Lower Mississippi Delta Region. Southeast Archaeological Center, National Park Service, Tallahassee. Price, G. R. Dennis, and Lorraine Heart¤eld 1977 Archeological Test Excavation of Site 16OU15, Located on the Monroe to Sandy Bayou Levee, Ouachita River Levees, Ouachita Parish, Louisiana. The Research Institute, College of Pure and Applied Science, Northeast Louisiana University, Monroe. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Prichard, Walter, Fred B. Kniffen, and Clair A. Brown (editors) 1945 Southern Louisiana and Southern Alabama in 1819: The Journal of James Leander Cathcart. Louisiana Historical Quarterly 23(3):735–921. Quimby, George I. 1942 The Natchezan Culture Type. American Antiquity 7(3):255–275. 1946 Natchez Social Structure as an Instrument of Assimilation. American Anthropologist 48(1):134–137. 1951 The Medora Site, West Baton Rouge Parish, Louisiana. Field Museum of Natural History, Anthropological Series 24(2). Chicago. 1953 Natchez Archaeology: A Tribute to the Natchez for Their Seeming Consistency in the Production of the Fictile Fabric. Southeastern Archaeological Conference Newsletter 3(3):22–24. 1957 The Bayou Goula Site, Iberville Parish, Louisiana. Fieldiana: Anthropology 47(2). Chicago Natural History Museum, Chicago. Ramenofsky, Ann F., and Patricia K. Galloway 1997 Disease and the Soto Entrada. In The Hernando de Soto Expedition: History, Historiography, and “Discovery” in the Southeast, edited by Patricia K. Galloway, pp. 259–279. University of Nebraska Press, Lincoln. Rappaport, Roy 1968 Pigs for the Ancestors. Yale University Press, New Haven. Redmond, E. M. 1996 Alternative Paths to Centralized Leadership among South American Tribesmen. Paper presented at the 61st Annual Society for American Archaeology Meetings, New Orleans. Rees, Mark A. 1997 Coercion, Tribute and Chie®y Authority: The Regional Development of Mississippian Political Culture. Southeastern Archaeology 16(2):113–133. 2002 Bayou Portage Guidry: A Plaquemine Mound Site in St. Martin Parish,

references cited

245

Louisiana. Paper presented at the 59th Annual Meeting of the Southeastern Archaeological Conference, Biloxi, Mississippi. 2003 Bayou Portage Guidry and the Plaquemine Mounds Archaeological Project. Louisiana Archaeological Society Newsletter 30(1):16–21. Reilly, F. Kent III 2002 A New Mississippian Female Image from Southern Arkansas. Paper presented at the Mid-South Archaeological Conference, Memphis. 2004 People of the Earth, People of the Sky: Visualizing the Sacred in Native American Art of the Mississippian Period. In Hero, Hawk, and Open Hand: American Indian Art of the Ancient Midwest and South, edited by Richard F. Townsend, pp. 124–137. Yale University Press, New Haven. Reuss, Martin 2004 Designing the Bayous: The Control of Water in the Atchafalaya Basin: 1800– 1995. Gulf Coast Studies No. 4. Texas A & M University Press, College Station. Rice, Prudence M. 1987 Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago. Riffel, Judy (editor and compiler) 1985 Iberville Parish History. Le Comité des Archives de la Louisiane, Baton Rouge. Roberts, Katherine M. 2004 Plant Remains. In Data Recovery Excavations at the Hedgeland Site (16CT19), Catahoula Parish, Louisiana, edited by Joanne Ryan, pp. 207–226. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. 2005 Plant Remains. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 431– 490. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Rodríguez-Alegría, E., R. J. Speakman, and M. D. Glascock 2005 Instrumental Neutron Activation Analysis of Late Coles Creek/Early Mississippian Ceramics from the Tensas Basin, Louisiana. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 565–578. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Rolingson, Martha Ann 1970 Settlement Pattern of the Plaquemine Culture along Bayou Bartholomew. Paper presented at the 29th Southeastern Archaeological Conference, Morgantown, West Virginia.

246

references cited

1971a The Ashley Point. The Arkansas Archeologist 12:50–52. 1971b Lakeport: Initial Exploration of a Late Prehistoric Ceremonial Center in Southeastern Arkansas. The Arkansas Archeologist 12:61–80. 1972 The Bartholomew Phase: Evidence for the Eastern Boundary of the Caddoan Area. Paper presented at the 14th Annual Caddo Conference, Austin, Texas. 1976 The Bartholomew Phase: A Plaquemine Adaptation in the Mississippi Valley. In Cultural Change and Continuity: Essays in Honor of James Bennett Grif¤n, edited by C. E. Cleland, pp. 99–119. Academic Press, New York. 1981 Artifacts: Lithics. In The Shallow Lake Site (3UN9/52) and Its Place in Regional Prehistory, by Martha Ann Rolingson and Frank F. Schambach, pp. 73–99. Arkansas Archeological Survey Research Series, No. 12. Arkansas Archeological Survey, Fayetteville. 1993 Archeology along Bayou Bartholomew, Southeast Arkansas. The Arkansas Archeologist 32:1–138. 1998 Toltec Mounds and Plum Bayou Culture: Mound D Excavations. Arkansas Archeological Survey Research Series, No. 54. Arkansas Archeological Survey, Fayetteville. 2003 Discussant Comments. In Woodland-Mississippian Transition in the MidSouth: Proceedings of the Twenty-Second Mid-South Archaeological Conference, June 2–3, 2001, edited by Charles H. McNutt, Stephen Williams, and Marvin D. Jeter, pp. 22–24, 29, 69–73, 110. University of Memphis Anthropological Research Center Occasional Papers, No. 25, Memphis. Rolingson, Martha A., and Frank F. Schambach 1981 The Shallow Lake Site (3UN9/52) and Its Place in Regional Prehistory. Arkansas Archeological Survey, Research Series No. 12. Arkansas Archeological Survey, Fayetteville. Rose, Jerome C., Barbara A. Burnett, and Mark W. Blaeuer 1984 Paleopathology and the Origins of Maize Agriculture in the Lower Mississippi Valley and Caddoan Culture Areas. In Paleopathology at the Origins of Agriculture, edited by M. N. Cohen and G. J. Armelagos, pp. 393–424. Academic Press, Orlando. Rose, Jerome C., Murray K. Marks, and Larry L. Tieszen 1991 Bioarchaeology and Subsistence in the Central and Lower Portions of the Mississippi Valley. In What Mean These Bones? Studies in Southeastern Bioarchaeology, edited by Mary Lucas Powell, Patricia S. Bridges, and Ann Marie Wagner Mires, pp. 7–21. University of Alabama Press, Tuscaloosa. Rose, Jerome C., and Mary Lucas Powell 1993 Analysis of Eight Skeletons from the McArthur Site. In Archeology along

references cited

247

Bayou Bartholomew, Southeast Arkansas, by Martha Ann Rolingson, pp. 67– 71. The Arkansas Archeologist 32. Rowland, Dunbar 1925 History of Mississippi: The Heart of the South. 4 vols. S. J. Clarke, Chicago. Rowland, Dunbar (editor) 1907 Mississippi, Comprising Sketches of Counties, Towns, Events, Institutions, and Persons, Arranged in Cyclopedic Form. 3 vols. Southern Historical Publishing Association, Atlanta. Rowland, Dunbar, and Albert G. Sanders (editors and translators) 1927 Mississippi Provincial Archives, 1729–1740, French Dominion, vol. 1. Mississippi Department of Archives and History, Jackson. 1929 Mississippi Provincial Archives, 1701–1729, French Dominion, vol. 2. Mississippi Department of Archives and History, Jackson. Rowland, Dunbar, Albert G. Sanders, and Patricia K. Galloway (editors and translators) 1984a Mississippi Provincial Archives, French Dominion, 1729–1748, vol. 4. Louisiana State University Press, Baton Rouge. 1984b Mississippi Provincial Archives, French Dominion, 1749–1763, vol. 5. Louisiana State University Press, Baton Rouge. Russ, John C. 1999 The Image Processing Handbook, 3rd ed. CRC Press. Russo, Mike 1992 Annual Report for Management Unit 3, Regional Archaeology Program, Department of Sociology and Anthropology, University of Southwestern Louisiana. Division of Archaeology, Department of Culture, Recreation and Tourism, Baton Rouge. 1993 Annual Report for Management Unit 3, Regional Archaeology Program, Department of Sociology and Anthropology, University of Southwestern Louisiana. Division of Archaeology, Department of Culture, Recreation and Tourism, Baton Rouge. Russo, Mike, and James Fogleman 1996 Stelly Mounds (16SL1): An Archaic Mound Complex. Louisiana Archaeology 21:127–158. Ryan, Joanne (editor) 2004 Data-Recovery Excavations at the Hedgeland Site (16CT19), Catahoula Parish, Louisiana. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Santeford, L. G., J. Yakubik, S. Dawdy, and B. Maygarden 1995 Cultural Resources Survey and Testing of Sanitary Sewer System Improvements

248

references cited

and Rehabilitation, St. Martin Parish, Louisiana. Earth Search, Inc., New Orleans. Submitted to Professional Engineering Consultants Corporation, Baton Rouge. Saucier, Roger T. 1981 Current Thinking on Riverine Processes and Geologic History as Related to Human Settlement in the Southeast. In Traces of Prehistory: Papers in Honor of William G. Haag, edited by F. H. West and R. Neuman, pp. 7–18. Geoscience and Man No. 22. School of Geoscience, Louisiana State University, Baton Rouge. 1994 Geomorphology and Quaternary Geologic History of the Lower Mississippi Valley. U.S. Army Engineer Waterways Experiment Station. Vicksburg, Mississippi. Saunders, Joe W. 1998 A Preliminary Report on Scott Place Mounds. Louisiana Archaeological Society Newsletter 25(3):6–8. Saunders, Joe W., Rolfe D. Mandel, Roger T. Saucier, E. Thurman Allen, C. T. Hallmark, Jay K. Johnson, Edwin H. Jackson, Charles M. Allen, Gary L. Stringer, Douglas S. Frink, James K. Feathers, Stephen Williams, Kristen J. Gremillion, Malcolm F. Vidrine, and Reca Jones 1997 A Mound Complex in Louisiana at 5400–5000 Years before the Present. Science 277:1796–1799. Scarry, John F. 1996 The Nature of Mississippian Societies. In Political Structure and Change in the Prehistoric Southeastern United States, edited by J. F. Scarry, pp. 12–24. University Press of Florida, Gainesville. Scarry, John F. (editor) 1996 Political Structure and Change in the Prehistoric Southeastern United States. University of Florida Press, Gainesville. Schambach, Frank F. 1979 Summary of Local and Regional Archeology. In Hampton: An Archeological and Historical Overview of a Proposed Strip Mine in South-Central Arkansas, assembled by Timothy C. Klinger, pp. 21–30. Arkansas Archeological Survey Research Reports, No. 19. Arkansas Archeological Survey, Fayetteville. 1981 A Description and Analysis of the Ceramics. In The Shallow Lake Site (3UN9/52) and Its Place in Regional Prehistory, by Martha Ann Rolingson and Frank F. Schambach, pp. 101–176. Arkansas Archeological Survey Research Series, No. 12. Arkansas Archeological Survey, Fayetteville. 1999 Spiro and the Tunica: A New Interpretation of the Role of the Tunica in the Culture History of the Southeast and the Southern Plains, A.D. 1100–

references cited

249

1750. In Arkansas Archaeology: Essays in Honor of Dan and Phyllis Morse, edited by Robert C. Mainfort, Jr., and Marvin D. Jeter, pp. 169–224. University of Arkansas Press, Fayetteville. Schambach, Frank F. (editor) 1985 The Bangs Slough Site (3CA3): A Multidisciplinary Investigation of a Multicomponent Prehistoric Site in the Ouachita River Valley of South-Central Arkansas. Arkansas Archeological Survey, Fayetteville. Submitted to U.S. Army Corps of Engineers, Vicksburg District. 1990 Coles Creek and Mississippi Period Foragers in the Felsenthal Region of the Lower Mississippi Valley: Evidence from the Bangs Slough Site, Southeast Arkansas. Arkansas Archeological Survey Research Series, No. 39. Arkansas Archeological Survey, Fayetteville. Schambach, Frank F., and John E. Miller III 1984 A Description and Analysis of the Ceramics. In Cedar Grove: An Interdisciplinary Investigation of a Late Caddo Farmstead in the Red River Valley, edited by Neal L. Trubowitz, pp. 109–170. Arkansas Archeological Survey Research Series, No. 23. Arkansas Archeological Survey, Fayetteville. Schambach, Frank F., and Martha Ann Rolingson 1981 Conclusions. In The Shallow Lake Site (3UN9/52) and Its Place in Regional Prehistory, by Martha Ann Rolingson and Frank F. Schambach, pp. 177– 203. Arkansas Archeological Survey Research Series, No. 12. Arkansas Archeological Survey, Fayetteville. Schambach, Frank F., and David B. Waddell 1990 The Pottery from the Bangs Slough Site. In Coles Creek and Mississippi Period Foragers in the Felsenthal Region of the Lower Mississippi Valley: Evidence from the Bangs Slough Site, Southeast Arkansas, edited by Frank F. Schambach, pp. 19–64. Arkansas Archeological Survey Research Series, No. 39. Arkansas Archeological Survey, Fayetteville. Schwab, C. A. 1996 A Study of Structure Patterns at the Phillip Nick Farm Site, Avoyelles Parish, Louisiana. Unpublished Master’s thesis, Department of Geography and Anthropology, Louisiana State University, Baton Rouge. Scott, Susan L. 2005 Vertebrate Fauna. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 411– 430. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. 2007 Faunal Remains. In The Saline-Fifteen Site (3BR119) and Late Prehistoric to Protohistoric Cultural Ecology in the Felsenthal Region Uplands, Southeast

250

references cited

Arkansas. Report in preparation, for publication in the Arkansas Archeological Survey Research Series. Shennan, Stephen 1994 Introduction: Archaeological Approaches to Cultural Identity. In Archaeological Approaches to Cultural Identity, edited by Stephen Shennan, pp. 1–32. Routledge, London. Shepard, Anna O. 1976 Ceramics for the Archeologist. Carnegie Institution of Washington, Washington, D.C. Shuman, Malcolm K. 1985 Cultural Resources Survey of an Area of Proposed Water Works Improvements along Bayou Grosse Tete, Iberville Parish, Louisiana. Surveys Unlimited, Baton Rouge, Louisiana. Submitted to Water Works District No. 4, Grosse Tete, Louisiana. 2002 Cultural Resources Survey of a Portion of the Northwest Shore of Lake Salvador, St. Charles Parish, Louisiana. Unpublished report for St. Charles Parish Planning and Zoning Department and on ¤le with the Louisiana Division of Archaeology, Baton Rouge. Shuman, Malcolm K., and Mary M. Shuman 2003 Phase II National Register Testing of Three Archaeological Sites on the Northwest Shore of Lake Salvador, St. Charles Parish, Louisiana. Report submitted to the St. Charles Parish Planning and Zoning Department and on ¤le with the Louisiana Division of Archaeology, Baton Rouge. Sibley, J. Ashley, Jr. 1967 Louisiana’s Ancients of Man: A Study of Changing Characteristics of Louisiana Indian Cultures. Claitor’s Publishing Division, Baton Rouge. Sinopoli, Carla M. 1991 Approaches to Archaeological Ceramics. Plenum Press, New York. Smith, Bruce D. 1978 Variation in Mississippian Settlement Patterns. In Mississippian Settlement Patterns, edited by B. D. Smith, pp. 479–503. Academic Press, New York. 1984 Mississippian Expansion: Tracing the Historical Development of an Explanatory Model. Southeastern Archaeology 3(1):13–32. Smith, Bruce D. (editor) 1990 The Mississippian Emergence. Smithsonian Institution Press, Washington, D.C. Smith, Christopher J. 1996 Analysis of Plant Remains from Mound S at the Toltec Mounds Site. The Arkansas Archeologist 35:51–76. Smith, Lawson M., Joseph B. Dunbar, and Louis D. Britsch 1986 Geomorphological Investigation of the Atchafalaya Basin, Area West, Atcha-

references cited

251

falaya Delta, and Terrebonne Marsh, vol. 1. Department of the Army Waterways Experiment Station, Corps of Engineers, Vicksburg, Mississippi. Prepared for the U.S. Army Engineer District, New Orleans. Smith, Marvin T. 1987 Archaeology of Aboriginal Culture Change in the Interior Southeast: Depopulation During the Early Historic Period. University Press of Florida, Gainesville. 1990 Glass Beads from the Goldsmith Oliver 2 Site. In Goldsmith Oliver 2 (3PU306): A Protohistoric Archeological Site near Little Rock, Arkansas, by Marvin D. Jeter, Kathleen H. Cande, and John J. Mintz, pp. 217–223. Arkansas Archeological Survey, Fayetteville. Submitted to Federal Aviation Administration, Fort Worth. 2000 Coosa: The Rise and Fall of a Southeastern Mississippian Chiefdom. University Press of Florida, Gainesville. 2002 Aboriginal Population Movements in the Postcontact Southeast. In The Transformation of the Southeastern Indians, 1540–1760, edited by Robbie Ethridge and Charles Hudson, pp. 3–20. University Press of Mississippi, Jackson. Smith, Marvin T., and Mark Williams 1994 Mississippian Mound Refuse Disposal Patterns and Implications for Archaeological Research. Southeastern Archaeology 13(1):27–35. Smith, Rhonda Lee 1996 Vertebrate Subsistence in Southeastern Louisiana between A.D. 700 and 1500. Unpublished Master’s thesis, Department of Anthropology, University of Georgia. Spaulding, Albert C. 1941 Trip Report [Mangum Site]. Memorandum for Acting Superintendent Gardner, May 5, 1941. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. Springer, James 1973 The Prehistory and Cultural Geography of Coastal Louisiana. Ph.D. dissertation, Yale University. University Micro¤lms, Ann Arbor, Michigan. 1977 The Bruly St. Martin Site and Its Implications for Coastal Settlement in Louisiana. Louisiana Archaeology 1:75–80. 1980 An Analysis of Prehistoric Food Remains from the Bruly St. Martin Site, Louisiana, With a Comparative Discussion of Mississippi Valley Faunal Studies. Midcontinental Journal of Archaeology 5(2):193–223. Stahle, David W., Malcolm K. Cleaveland, and John M. Hehr 1985 A 450-Year Drought Reconstruction for Arkansas, United States. Nature 316:530–532.

252

references cited

Stahle, David W., Edward R. Cook, Malcolm K. Cleaveland, Matthew D. Therrell, David M. Meko, Henri D. Grissomo-Mayer, Emma Watson, and Brian H. Luckman 2000 Epic Sixteenth Century Drought over North America. EOS 81:121–125. Steponaitis, Vincas P. 1974 The Late Prehistory of the Natchez Region: Excavations at the Emerald and Foster Sites, Adams County, Mississippi. Honors thesis, Department of Anthropology, Harvard College, Cambridge. 1978 Location Theory and Complex Chiefdoms: A Mississippian Example. In Mississippian Settlement Patterns, edited by B. D. Smith, pp. 417–453. Academic Press, New York. 1981 Plaquemine Ceramic Chronology in the Natchez Region. Mississippi Archaeology 16(2):6–19. 1983 Ceramics, Chronology, and Community Patterns: An Archaeological Study at Moundville. Academic Press, New York. 1986 Prehistoric Archaeology in the Southeastern United States, 1970–1985. Annual Review of Anthropology 15:363–404. 1991 Contrasting Patterns of Mississippian Development. In Chiefdoms: Power, Economy, and Ideology, edited by T. Earle, pp. 193–228. Cambridge University Press, Cambridge. 1995 Cultural Chronology, Natchez Bluffs Region. Manuscript on ¤le, Gulf Coast Survey, Alabama Museum of Natural History, University of Alabama, Tuscaloosa. 1998 Native American Cultures in the Precolonial South. In The Natchez District in the Old, Old South, edited by Vincas P. Steponaitis, pp. 1–22. Southern Research Report No. 11. Center for the Study of the American South, University of North Carolina, Chapel Hill. Steponaitis, Vincas P., M. James Blackman, and Hector Neff 1996 Large-Scale Patterns in the Chemical Composition of Mississippian Pottery. American Antiquity 61:555–572. Steponaitis, Vincas P., Jeffrey P. Brain, and Ian W. Brown 1983 Glossary of Ceramic Types and Varieties. In The Grand Village of the Natchez Revisited: Excavations at the Fatherland Site, Adams County, Mississippi, 1972, by R. S. Neitzel, pp. 139–145. Archaeological Report No. 12. Mississippi Department of Archives and History, Jackson. Steponaitis, Vincas P., Stephen Williams, R. P. Stephen Davis, Jr., Ian W. Brown, Tristram R. Kidder, and Melissa Salvanish (editors) 2002 LMS Archives Online. http://rla.unc.edu/archives/lms1/ June 1, 2004. Stern, Theodore C. 1977 The Natchez. In The Native Americans, 2nd ed., edited by R. F. Spencer and J. D. Jennings, pp. 414–424. Harper and Row, New York.

references cited

253

Stevenson, Joe 1992 Zooarchaeology of the Kleinpeter Site. Unpublished Master’s thesis, Louisiana State University, Baton Rouge. Stoltman, James B. 1989 A Quantitative Approach to the Petrographic Analysis of Ceramic Thin Sections. American Antiquity 54:147–160. 1991 Ceramic Petrography as a Technique for Documenting Cultural Interaction: An Example from the Upper Mississippi Valley. American Antiquity 56:103–121. 2000 The Role of Petrography in the Study of Archaeological Ceramics. In Earth Sciences and Archaeology, edited by Paul Goldberg, Vance T. Holliday, and C. Reid Ferring, pp. 297–326. Kluwer Academic/Plenum, London. 2005 Report on Petrographic Observations of Ceramic Thin Sections from Lake Providence, Louisiana. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 545–564. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Story, Dee Ann 1997 1968–1970 Archeological Investigations at the George C. Davis Site, Cherokee County, Texas. Bulletin of the Texas Archeological Society 68:1–103. Stuiver, M., and P. J. Reimer 1986 A Computer Program for Radiocarbon Age Calibration. Radiocarbon 28:1022–1030. Stuiver, M., P. J. Reimer, E. Bard, J. W. Beck, G. S. Burr, K. A. Hughen, B. Kromer, G. McCormac, J. van der Plicht, and M. Spurk 1998 INTCAL98 Radiocarbon Age Calibration, 24,000–0 cal BP. Radiocarbon 40(3):1041–1083. Suhm, Dee Ann, and Edward B. Jelks 1962 Handbook of Texas Archeology: Type Descriptions. Bulletin of the Texas Archeological Society Special Publication No. 1. Surrey, N. M. Miller 1968 The Commerce of Louisiana during the French Regime, 1699–1763. AMS Press, New York. Originally published 1916, Columbia University Press, New York. Suttles, Wayne 1968 Coping with Abundance. In Man the Hunter, edited by Richard B. Lee and Irven DeVore, pp. 56–68. Aldine, Chicago. Swanton, John R. 1911 Indian Tribes of the Lower Mississippi Valley and Adjacent Coast of the Gulf of Mexico. Bureau of American Ethnology Bulletin 43. Smithsonian Institution, Washington, D.C.

254

references cited

Source Material for the Social and Ceremonial Life of the Choctaw Indians. Bureau of American Ethnology Bulletin No. 103. U.S. Government Printing Of¤ce, Washington, D.C. 1939 Final Report of the United States De Soto Expedition Commission. 76th Congress, 2nd Session, House Document 71. Reprinted in 1985 by Smithsonian Institution Press, Washington, D.C. Talma, A. Siep, and John C. Vogel 1993 A Simpli¤ed Approach to Calibrating C14 Dates. Radiocarbon 35(2):317–322. Tooker, Elizabeth 1963 Natchez Social Organization: Fact or Anthropological Folklore? Ethnohistory 10(4):358–372. Toth, Edwin A. 1988 Early Marksville Phases in the Lower Mississippi Valley: A Study of Culture Contact Dynamics. Archaeological Reports No. 21. Mississippi Department of Archives and History, Jackson. Trachtenberg, Samuel 1999 Macrobotanical Remains from the Raffman Site (16MA20). Unpublished Master’s thesis, Department of Anthropology, Tulane University, New Orleans. Traverica, Forrest 1977 Miscellaneous unpublished documents relative to 16SC14. On ¤le with the Museum of Natural Science, Louisiana State University, Baton Rouge. Trigger, Bruce 1980 Archaeology and the Image of the American Indian. American Antiquity 45(4):662–676. van der Leeuw, Sander E. 1977 Toward a Study of the Economics of Pottery Making. In Ex Horreo, edited by B. L. Van Beek, R. W. Brandt, and Willy Goeman-van Waateringe, pp. 68–76. University of Amsterdam, Amsterdam. VanDerwarker, Amber M. 1999 Feasting and Status at the Toqua Site. Southeastern Archaeology 18(1):24–34. Van Nest, J., D. K. Charles, J. E. Buikstra, and D. L. Asch 2001 Sod Blocks in Illinois Hopewell Mounds. American Antiquity 66(4):633–650. Velde, Bruce, and Isabelle C. Druc 1998 Archaeological Ceramic Materials. Springer-Verlag, Berlin. Vigander, H., and B. Maygarden 1994 Phase I Cultural Resources Inventory of Public Access Lands in the Atchafalaya Basin, Vicinity of the Sherburne Wildlife Management Area, Point Coupee, St. Martin and Iberville Parishes, Louisiana. Earth Search, Inc., New Orleans. Submitted to U.S. Army Corps of Engineers, New Orleans District. 1931

references cited

255

Waddell, David B., and Kathryn A. King 1990 The Other Artifacts from the Bangs Slough Site. In Coles Creek and Mississippi Period Foragers in the Felsenthal Region of the Lower Mississippi Valley: Evidence from the Bangs Slough Site, Southeast Arkansas, edited by Frank F. Schambach, pp. 63–84. Arkansas Archeological Survey Research Series, No. 39. Arkansas Archeological Survey, Fayetteville. Walens, Stanley 1981 Feasting with Cannibals: An Essay on Kwakiutl Cosmology. Princeton University Press, Princeton. Walker, John W. 1981 Summary of Previous Investigations. In Natchez Trace Parkway: An Archeological Testing Project at the Gordon Mounds Site, Jefferson County, Mississippi, by D. L. McCullough, pp. 2–6. Unpublished report, Southeast Archeological Center, National Park Service, Tallahassee. Warhop, Jennifer R. 2005 Investigations of a Unique Structure at the Anna Site (22Ad500), Mississippi. Unpublished Master’s thesis, Department of Anthropology, University of Alabama, Tuscaloosa. Webb, Clarence 1959 The Belcher Mound: A Strati¤ed Caddoan Site in Caddo Parish, Louisiana. Memoirs of the Society for American Archaeology, No. 16. Menasha, Wisconsin. 1961 Relationships between the Caddoan and Central Louisiana Sequences. In Symposium: Relationships between the Caddoan Area and Neighboring Groups, edited by E. Mott Davis, pp. 11–25. Bulletin of the Texas Archeological Society 31. 2000 Stone Points and Tools of Northwestern Louisiana, 2nd rev. ed. Special Publication No. 1. Louisiana Archaeological Society, Baton Rouge. Weber, J. Cynthia 1973 The Paw Paw Site (3OU22): A Report on Excavations in Area Four. Manuscript on ¤le, Arkansas Archeological Survey, Fayetteville. Weiner, Annette B. 1988 The Trobrianders of Papua New Guinea. Harcourt, Brace, Jovanovich, Fort Worth. Weinstein, Richard A. 1986 Tchefuncte Occupation in the Lower Mississippi Delta and Adjacent Coastal Zone. In The Tchula Period in the Mid-South and Lower Mississippi Valley, edited by David H. Dye and Ronald C. Brister, pp. 107–127. Archaeological Reports No. 17. Mississippi Department of Archives and History, Jackson. 1987a Development and Regional Variation of Plaquemine Culture in South Loui-

256

references cited

siana. In Emergent Mississippian: Proceedings of the Sixth Mid-South Archaeological Conference, edited by Richard A. Marshall, pp. 85–106. Occasional Paper No. 87-01. Cobb Institute of Archaeology, Mississippi State University, Mississippi State. 1987b The Rosedale and Shellhill Discs: “Southern Cult” Evidence from Southeastern Louisiana. Louisiana Archaeology 11:65–88. Weinstein, Richard A. (editor) 2005 Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana. 2 vols. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Weinstein, Richard A., and David B. Kelley 1984 Archaeology and Paleogeography of the Upper Felsenthal Region: Cultural Resources Investigations in the Calion Navigation Pool, South-Central Arkansas. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. 1992 Cultural Resources Investigations in the Terrebonne Marsh, South-Central Louisiana. Cultural Resources Series Report No. COELMN/PD-89/06. Coastal Environments, Inc., Baton Rouge. Prepared for the U.S. Army Corps of Engineers, New Orleans District. Weinstein, Richard A., David B. Kelley, and Joe W. Saunders (editors) 2003 The Louisiana and Arkansas Expeditions of Clarence Bloom¤eld Moore. Classics in Southeastern Archaeology Series. University of Alabama Press, Tuscaloosa. Weismantel, Mary J. 1988 Food, Gender, and Poverty in the Ecuadorian Andes. Waveland Press, Prospect Heights, Illinois. Welch, Paul D. 1991 Moundville’s Economy. University of Alabama Press, Tuscaloosa. Welch, Paul D., and C. Margaret Scarry 1995 Status Related Variation in Foodways in the Moundville Chiefdom. American Antiquity 60(3):397–419. Wells, Douglas C. 1997 Political Competition and Site Placement: Late Prehistoric Settlement in the Tensas Basin of Northeast Louisiana. Louisiana Archaeology 22:71–91. 1998 The Early Coles Creek Period and the Evolution of Social Inequality in the Lower Mississippi Valley. Unpublished Ph.D. dissertation, Department of Anthropology, Tulane University, New Orleans. 2001 Cultural Resources Evaluation of the Upper Atchafalaya Backwater Area, Iberville and Pointe Coupee Parishes, South Louisiana. Coastal Environments,

references cited

257

Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, New Orleans District. 2005 Aboriginal Ceramics. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 311–390. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Wells, Douglas C., and Richard A. Weinstein 2005 The Preston Phase in Regional Context. In Lake Providence: A Terminal Coles Creek Culture Mound Center, East Carroll Parish, Louisiana, edited by R. A. Weinstein, pp. 491–511. Coastal Environments, Inc., Baton Rouge. Submitted to U.S. Army Corps of Engineers, Vicksburg District. Wesolowsky, A. B. 1974 Investigations at Two Sites in Southeastern Arkansas. The Arkansas Archeologist 15:20–26. Wesson, Cameron B. 1998 Mississippian Sacred Landscapes: The View from Alabama. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R. B. Lewis and C. Stout, pp. 93–122. University of Alabama Press, Tuscaloosa. Wesson, Cameron B., and Mark A. Rees 2002 Protohistory and Archaeology: An Overview. In Between Contacts and Colonies: Archaeological Perspectives on the Protohistoric Southeast, edited by C. B. Wesson and M. A. Rees, pp. 1–11. University of Alabama Press, Tuscaloosa. White, Douglas R., George P. Murdock, and Richard Scaglion 1971 Natchez Class and Rank Reconsidered. Ethnology 10(4):369–388. White, Patsy K. 1970 Investigation of the Cemetery at the Gee’s Landing Site. The Arkansas Archeologist 11:1–20. 1987 The Gordon Site: A Middle Coles Creek to Late Mississippi Period Occupation in Ashley County, Southeast Arkansas. The Arkansas Archeologist 2526:1–51. Whitney, Cynthia, Vincas P. Steponaitis, and John J. W. Rogers 2002 A Petrographic Study of Moundville Palettes. Southeastern Archaeology 21:227–234. Willey, Gordon R. 1966 An Introduction to American Archaeology: 2. North and Middle America. Prentice-Hall, Englewood Cliffs, New Jersey. Willey, Gordon R., and Philip Phillips 1958 Method and Theory in American Archaeology. University of Chicago Press, Chicago.

258

references cited

Williams, Mark 1994 The Origins of the Macon Plateau Site. In Ocmulgee Archaeology 1936–1986, edited by D. J. Hally, pp. 130–137. University of Georgia Press, Athens. Williams, Stephen 1956 Settlement Patterns in the Lower Mississippi Valley. In Prehistoric Settlement Patterns in the New World, edited by Gordon R. Willey, pp. 52–62. Viking Fund Publications in Anthropology, No. 23. Wenner-Gren Foundation for Anthropological Research, New York. 1962 Historic Archaeology in the Lower Mississippi Valley. Southeastern Archaeological Conference Newsletter 9(1):53–63. 1966 Historic Archaeology, Past and Present. School of American Research Annual Report for 1966, pp. 23–26. Santa Fe, New Mexico. 1967 On the Location of the Historic Taensa Villages. Conference on Historic Sites Archaeology Papers 1965–1966 1:3–13. 1979 Research Resources on Lower Mississippi Valley Archaeology. Bulletin No. 5, Lower Mississippi Survey. Peabody Museum, Harvard University, Cambridge. 2003 Introduction to 2003 Edition: Phillips, Ford, and Grif¤n’s Lower Valley Survey. In Archaeological Survey in the Lower Mississippi Alluvial Valley, 1940– 1947, by P. Phillips, J. A. Ford, and J. B. Grif¤n, edited by S. Williams, pp. xi–xxxii. University of Alabama Press, Tuscaloosa. Williams, Stephen, and Jeffrey P. Brain 1983 Excavations at the Lake George Site, Yazoo County, Mississippi, 1958–1960. Papers of the Peabody Museum of Archaeology and Ethnology, No. 74. Harvard University, Cambridge. Wilson, Douglas C., and William L. Rathje 2001 Garbage and the Modern American Feast. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 204–422, Smithsonian Institution Press, Washington, D.C. Wolf, Eric R. 1959 Sons of the Shaking Earth. University of Chicago Press, Chicago. Wood, W. Raymond 1963 Two New Projectile Points: Homan and Agee Points. The Arkansas Archeologist 4(2):1–6. Young, Biloine Whiting, and Melvin L. Fowler 2000 Cahokia: The Great Native American Metropolis. University of Illinois Press, Urbana.

Contributors

Virgil Roy Beasley III received a B.A. from the University of Central Florida and an M.A. from the University of Alabama. He is currently conducting doctoral research at Northwestern University. Mr. Beasley’s research interests include the archaeology of ¤shing populations, the use of shell as a building material, and landscape archaeology. Ian W. Brown is Professor of Anthropology at the University of Alabama and Curator of Gulf Coast Archaeology at the Alabama Museum of Natural History. He specializes in the archaeology of the southeastern United States. Most of his research has been in the Lower Mississippi Valley, along the southwest coast of Louisiana, and in the Mobile-Tensaw delta of Alabama. He is the editor of Bottle Creek: A Pensacola Culture Site in South Alabama, which is also published by the University of Alabama Press (2003), and author of “The Calumet Ceremony in the Southeast as Observed Archaeologically” in the revised edition of Powhatan’s Mantle: Indians in the Colonial Southeast (U. of Nebraska Press, 2006). Marvin D. Jeter (Ph.D., Arizona State University, 1977) has been the UAM Station Archeologist for southeast Arkansas with the Arkansas Archeological Survey since 1978. He has also worked in Alabama, Arizona, Tennessee, Louisiana, and Illinois. His publications include a 1989 overview of Arkansas and Louisiana archaeology, the prize-winning book Edward Palmer’s Arkansaw Mounds (U. of Arkansas Press, 1990), and Arkansas Archaeology (U. of Arkansas Press, 1999), plus a number of reports, articles, book chapters, and reviews. His main research interests are the late prehistoric and protohistoric-contact periods in the Lower Mississippi Valley and Southeast and the history of archaeology.

260

contributors

Tristram R. Kidder received a B.A. from Tulane University in 1982 and an M.A. (1987) and Ph.D. (1988) from Harvard University. He is currently Professor of Anthropology at Washington University in St. Louis. Patrick C. Livingood completed his Ph.D. at the University of Michigan in 2006 with a dissertation on Mississippi period Pearl River mound sites. He received a B.S./B.A. from the University of North Carolina at Chapel Hill in computer science and anthropology and an M.A. in anthropology from the University of Michigan. He is currently Assistant Professor of Anthropology at the University of Oklahoma. Mark A. Rees is Assistant Professor of Anthropology in the Department of Sociology and Anthropology at the University of Louisiana at Lafayette. He received a Ph.D. in anthropology from the University of Oklahoma in 2001 and an M.A. in historical archaeology from the University of Massachusetts at Boston in 1991. His research interests include Native American political culture and historical anthropology. He is co-editor with Cameron Wesson of Between Contacts and Colonies (U. of Alabama Press, 2002) and a contributor to The Archaeology of Traditions, edited by Timothy R. Pauketat (U. Press of Florida, 2001). Lori Roe is currently pursuing a doctoral degree in North American archaeology at Tulane University. Her dissertation research focuses on Coles Creek mound use at the Raffman site and its implications for Coles Creek social organization. Malcolm K. Shuman is the owner of SURA, Inc., a contract archaeology ¤rm in Baton Rouge. He was educated at Louisiana State University, the University of New Mexico, and Tulane University. He has done archaeology in the Southeast, in the Southwest, and in France. Richard A. Weinstein is a Principal Investigator for Coastal Environments, Inc., in Baton Rouge. He has investigated sites of all descriptions in the Lower Mississippi Valley, Florida, and coastal Texas. Douglas C. Wells is an Archaeologist/Project Manager for Coastal Environments, Inc., in Baton Rouge. He has won the Nobel Piece Prize for his work with underprivileged potsherds and reserves the right to author his own biographical pieces.

Index

Ables Creek site, 174 Addis Plain, 4, 5, 108–9, 110, 112, 117–8, 123, 124, 137, 142, 155, 156, 162, 172–3, 182, 198. See also Baytown Plain Agriculture, 13, 96, 132, 135, 175–6, 184–5, 189, 191, 202. See also botanical; domesticates; maize Alba point, 75, 78, 169, 171, 177, 179, 181 American Bottom, 15, 35, 52, 54, 56, 58, 60, 62, 64, 166 Anna: phase, 17, 111, 134, 135, 138, 141–2, 149, 156, 157, 162; site, 5, 17–18, 127– 44, 149, 150, 151, 152, 155, 156, 158 Anna Incised, 46, 48, 85, 104, 112, 123, 124, 141, 155, 182 Antioch site, 152 Architecture, 1, 8, 15, 18, 21, 24, 34, 36, 37, 91 Ashley point, 170–74, 177, 179, 180, 188 Atchafalaya: basin, 12, 17, 66–69, 70, 72, 78, 87–92; river, 67, 68, 71, 72, 75, 76, 84 Avery Island, 69 Avoyelles Punctated, 46, 49, 75, 78, 100, 104, 170, 178 Balmoral phase, 33–34, 44, 46–47, 50, 58, 60, 134, 138, 141 Bangs Slough site, 177, 178, 181, 184 Barataria: basin, 17, 101, 106; phase, 10, 17, 100–01, 104

Bartholomew: basin, 11, 19, 165; bayou, 164, 169, 171, 174, 183, 185; phase, 18, 19, 165, 169–76, 177, 183, 187, 188, 190, 192, 193, 195; region, 165, 185, 192 Bartholomew-Macon region, 18–19, 165, 185–86 Barton Incised, 47, 103 Bayou Cutler phase, 76, 81, 83, 85 Bayou Goula: culture, 3; point, 75, 169; report, 3; site, 5 Bayougoula Indians, 201. See also Bayou Goula Bayou Petre phase, 9, 10, 100–01, 104 Bayou Pierre site, 150, 156, 158 Baytown period, 23, 25, 28, 31, 61, 99, 175 Baytown Plain, 4, 5, 30–33, 46, 49, 60, 76, 77, 83, 87, 102–06, 109, 162, 170, 172, 177, 178 Beldeau Incised, 33, 83, 141, 172 Bellaire phase, 18–19, 161, 165, 167–69, 187 Bell Plain, 102, 103, 109, 110, 117, 123, 124, 163 Belmont, John, 27, 94, 165, 167, 187, 195 Berwick site, 71 Boeuf basin, 11, 165–66, 187; region, 165. See also Bartholomew, basin Bois Chactas site, 97, 99, 101–03, 104–06 Boone’s Mounds site, 181–82, 187 Botanical (archaeobotanical), 42, 129. See also agriculture; domesticates; maize Bottle Creek site, 112

262

index

Boydell site, 172–73, 174, 182, 187, 188, 195 Brain, Jeffrey, 8–9, 20, 52–56, 61, 96, 97, 106, 108–09, 135, 147, 154, 155, 158, 159, 162, 175, 186, 192, 193, 195, 199, 201–02 Buras Incised, 100 Burial, 24, 25, 30, 61, 87, 96, 135, 153, 166–67, 168, 172, 174, 175, 181, 183, 184, 186, 188. See also mortuary; mound, burial Burthe site, 153 Caddo (Caddoan), 4–5, 10, 11, 12, 60, 161– 63, 165, 166, 170, 174, 176, 178, 181–83, 186, 188, 189, 191, 193, 200 Cahokia: contact and in®uence, 61–62, 143, 147, 162, 166, 186, 188, 202, 203; horizon, 52–58, 60, 63, 186; site, 36, 52, 58, 96, 135, 143, 149, 166, 176, 190, 196 Caney Bayou phase, 177, 180–82, 185, 192 Carter Engraved, 33, 46, 49, 51, 60, 81, 83, 87, 100, 104, 112, 123 Catahoula Basin, 11, 12, 191 Central Mississippi Valley, 4, 35, 96 Ceramic petrography. See petrographic analysis Ceramics. See type names, complexes; grog temper; shell, temper Chenopod, 24 Chicot Red, 137, 140, 142 Chiefdom, 18, 23, 25, 34, 62, 67, 135–36, 144, 150–51, 188–90 Chitimacha, 17, 68, 71–72, 78, 80, 82–84, 86–93, 201 Choctaw, 97 Clark site, 186, 188 Coker set, 52–54 Coleman Incised, 52, 54, 172, 182 Coles Creek: culture, 1, 4, 5, 6–15, 17–26, 28, 33–37, 46, 58, 61, 64, 66, 69, 76, 83, 90,–92, 100, 138, 141, 143–45, 147, 154–56, 157–62, 170, 175, 180, 184, 186, 187, 189, 190, 201–02, 203–05; period, 3, 9, 14, 15, 21, 26, 30–32, 34, 35, 44– 46, 52, 58, 60, 61, 63, 71, 75, 76, 78, 81, 83, 84, 86–88, 104–06, 154, 156,

162, 163, 172, 174–77, 179, 182, 185, 189, 198. See also transitional Coles Creek Coles Creek Incised, 4, 5, 30, 32, 33, 44, 46–48, 75, 78, 81, 83, 85, 100–04, 141, 170, 174, 178, 183 Collins, Henry B., 68, 71, 83, 133 Complexity, 4, 13, 21, 24, 35, 129, 130, 140, 145, 197, 202–04. See also chiefdom Cotter, John L., 5, 134, 155 Cracker Road Incised, 102 Crescent Quarry chert, 174 Crippen Point phase, 7–9, 54–55, 162, 163, 169 Culture history, culture historical approach, 4, 12, 17, 68, 72, 87, 88, 94, 197 Cypress Swamp phase, 165, 176–79, 187, 188 Delta Natchezan phase, 85, 101. See also Natchez, phase Discovery site, 99, 101, 104 D’Olive Incised, 112, 121–24 Domesticates, 175–76, 184–85, 189. See also agriculture; botanical; maize Eagle Lake site, 179, 195 Elite, 14, 15, 24–26, 36, 52, 58–64, 91, 143, 153, 187, 189, 202, 204, 205; residence, 24, 34. See also chiefdom; complexity Emerald: phase, 142, 144, 149, 153, 154, 156, 157, 168; site, 5, 18, 133, 134, 149– 50, 152, 154, 156, 158, 167, 168 Evansville Punctated, 46, 49, 75, 78, 83, 103, 170, 174, 178, 182, 192 False Indigo site, 179 Fatherland site, 18, 133, 147–49, 150, 153, 157–58 Fatherland Incised, 85, 154, 156 Fauna, faunal, 18, 32, 42, 102, 103, 105, 129, 142, 175, 184, 185 Feast, feasting, 14, 18, 127–31, 138–41, 142– 43, 175 Felsenthal region, 11, 12, 165, 170, 176–85, 187, 189, 191, 192, 195, 200

index

Feltus site, 133, 145, 150, 153–58 Fitzhugh: phase, 7, 28, 161, 169; site, 34– 36, 203, 204 Ford, James A., 3–4, 6–7, 11, 12, 66, 94, 95, 108, 123, 133–34, 145, 147, 154, 161– 62, 163, 192 Foster: phase, 111, 137, 141, 142, 149, 153, 156, 157, 168; site, 18, 150, 157–58 Fraser site, 180 French Fork Incised, 30, 33, 46, 48, 75, 82, 83, 85 Gary site, 180 Gee’s Landing site, 180 Glass site, 150, 152, 153, 156, 158 Gordon: phase, 9, 141, 157, 158, 163; site in Arkansas, 180, 184, 195; site in Mississippi, 5, 150, 156, 158 Grace Brushed, 112, 121, 124, 192 Gran Marais phase, 18, 19, 165, 176–88, 192, 195 Greenhouse site, 5, 7, 24, 25, 77, 91, 145 Grif¤n, James, 6, 7, 10, 186, 199 Grog temper (grog-tempered), 17, 23, 33, 46, 62, 63, 101–06, 108–10, 113–14, 116– 17, 119–24, 126, 162, 163, 172–73, 174, 177, 178, 181, 186, 192, 197–98 Harrison Bayou Incised, 47, 75, 78, 81, 85, 104, 170, 178 Hedgeland site, 46, 60, 63, 109, 121, 163, 176 Henderson site, 157, 158, 172 Holly Fine Engraved, 174 Hollyknowe Pinched (Hollyknowe RidgePinched), 46, 49, 81, 170, 178 Holmes, William Henry, 3–4, 96 Homan point, 170–71, 172, 177, 188 Identity, 12, 127, 128, 130–31, 142–44, 199, 200–02 Interaction, 19, 20, 21, 35–37, 64, 88, 90, 97, 100, 106, 127, 130, 162, 188–89, 201, 203 Jug Point 2 site, 179–80

263

Keller Place site, 182 Kiam Incised, 178, 180, 182 Koroa, 192, 201 Lake George: phase, 61, 111; site, 8–9, 15, 20, 24–25, 35, 52, 54–55, 64, 147, 162– 63, 173, 177, 186, 201, 204 Lake Providence site, 15, 20, 36, 38–65, 166, 176, 202–4 Lake Salvador, 17, 94–106, Lapile Incised, 178, 182 L’Eau Noire Incised, 5, 78, 81, 83, 100, 112, 123, 141, 170, 172 Leland Incised, 112 Little Mud Lake site, 181, 184 Lookout site, 132, 152 Lower Mississippi Survey (LMS), 56, 69, 134, 154–55, 157, 159, 161, 163, 167, 169, 194–95 Lower Mississippi Valley (LMV), 1–4, 6, 8, 10, 12–21, 23, 34–37, 52, 58, 61–67, 71–2, 88–90, 94, 108, 110–11, 124, 134– 35, 143–44, 147, 161, 166, 177–78, 181, 183–84, 187, 189–90, 196–205, 208 Lower Yazoo Basin. See Yazoo Basin Maddox Engraved, 100–01 Maize, 13, 20, 24, 37, 42, 60, 135, 166, 170, 174–76, 184–85, 190–91, 202, 204. See also agriculture; botanical; domesticates Mangum site, 153, 186 Marksville: culture, 4, 31, 42; period, 28, 31, 169 Matheny site, 174–75 Mayersville phase, 109, 169, 204 Maygrass, 24 Mazique Incised, 4, 5, 33, 46–48, 78, 81, 83, 85, 100, 102, 123, 170, 172, 178–79, 183, 192 Mazique site, 145, 154–56, 158 Medora: phase, 7, 66, 76, 83, 85, 109; site, 5–6, 17, 66, 68, 70, 95 Mill Creek chert, 174 Mississippian culture, 1, 4, 6, 8, 19, 36,

264

index

46, 96–97, 100, 147, 162, 167, 169, 173, 197, 199, 201, 205 Mississippianization, 8, 12, 15, 17, 61, 191, 199–200 Mississippi period, 3–4, 6, 23, 28, 58, 61, 63, 68, 75, 83, 88, 104–05, 150– 51, 160, 177–78, 180, 183, 197–99, 204–05 Mississippi Plain, 4, 52, 56, 60, 102–03, 110, 112, 117, 121, 124, 170, 172 Moore, C. B., 68, 82–85, 167, 175, 181– 82, 183 Moorehead, Warren K., 133, 154–55, 157 Moorehead phase, 58 Morgan site, 24, 69 Mortuary, 15, 24, 25, 61, 63, 91, 153–54, 186. See also burial Mott site, 34–36, 204 Mound: burial, 8, 24–25, 153, 182; conical, 26, 73, 76, 90; platform, 8, 21, 24–26, 32, 61, 72, 75, 77, 83, 86, 90, 92, 95, 145, 156, 160; shell, 70, 99; temple, 3, 148, 153, 157, 160 Mound Place Incised, 101–03, 105, 112, 123 Moundville Incised, 101–03 Moundville site, 96, 101, 106, 112, 135, 143, 151, 167, 168, 173, 196, 198, 203 Mount Nebo site, 25, 61 Mulberry Creek Cord Marked, 32 Natchez: bluffs, 1, 3–4, 10, 11, 13, 14, 17, 18, 23, 36, 64, 66, 91, 109, 127, 131–135, 145–159, 163, 203; culture, 4, 5, 11, 66, 162, 193–94; Great Sun, 147–48; Indians, 14, 18, 89, 131, 136, 138, 141–43, 145, 147–49, 153, 201; phase, 153, 157, 162 Native American. See Bayougoula Indians; Caddo; Chitimacha; Choctaw; Coles Creek, culture; Koroa; Mississippian culture; Natchez; Plaquemine, culture; Quapaw; Taensa; Tunica Neitzel, Robert S., 27, 147, 157 Neutron Activation analysis (INA A), 56, 62, 64, 113

New Archaeology, 94. See also processual archaeology Norrell site, 186 North site, 153 Oak Bend Landing site, 153 Old Town Red, 52–54, 63 O’Quinn site, 153 Osceola site, 63, 203, 204 Ouachita: river, 23, 165, 179, 180, 185, 189; valley, 1, 10, 11, 165–66, 181, 183, 187, 189 Pargoud Incised, 178 Pargoud Landing site, 183, 186 Pargoud phase, 183 Parkin Punctated, 112 Paw Paw site, 183, 184, 187 Pearl River, 11, 17, 111–12, 117, 120, 124– 26, 200 Pensacola Incised, 101 Petite Anse region, 69 Petrographic analysis, 56, 108, 113–17 Pevey: phase, 113; site, 11, 17, 111–13 Pharr site, 71 Phillips, Philip, 5–10, 27, 66, 76, 83, 94, 97, 106–09, 134, 159, 161–62, 167, 169, 177, 194, 196 Pinola phase, 100, 105 Pinola Plain, 102–06 Plaquemine: Coles Creek origins, 4, 6–9, 15, 20–37, 88, 186, 199, 202–03; culture, 1–19, 20–21, 25–26, 35–37, 46, 63– 64, 66–67, 83, 127–28, 145–49, 159–59, 161–63, 186, 196–205; culture traits, 5, 7, 20; elite, 18, 61; Mississippian in®uence, 97, 108, 191–92; Mississippianized Coles Creek, 6–8, 12, 15, 17, 61, 66, 97, 143, 186, 191, 199–203; peoples, 13, 15, 66–67, 88, 90, 158, 196; period, 3–19, 20–21, 60, 94–95, 100, 108–09, 161–63, 196–205; pottery complex, 5, 17, 46, 64, 100, 106, 109–111, 197–98; town of, 1, 66. See also transitional Coles Creek

index

Plaquemine Brushed, 1, 4–5, 31, 46–48, 75–78, 81, 83, 85, 100, 104, 112, 123–24, 141, 155, 170, 178, 183 Plaza, 5, 21, 24–26, 28, 31, 34, 38, 63, 72, 76, 77, 80, 86, 91, 95, 112, 150, 152, 154– 58, 166, 171, 174, 179, 181, 204 Pontchartrain Check Stamped, 75, 76, 78, 81–83, 85 Pottery. See type names, complexes; grog temper, shell temper Powell Plain, 52–53, 56, 60, 62 Powell set, 52, 54 Preston: ¤neware complex, 46, 51, 60, 64; phase, 9, 15, 33, 36, 43–49, 58–64, 163 Processual archaeology (processualism), 8, 17, 88, 94 Pumpkin Lake site, 157 Quapaw, 194 Quimby, George, 3–6, 95, 108, 147 Quitman site, 154 Raffman site, 15, 20–37, 63, 92, 202–04 Ramey Incised, 52–54, 60, 62 Ratcliffe site, 157, 158 Red River, 1, 12, 67, 95, 109, 145, 176, 189, 203 Rhinehart Punctated, 46, 48 Rice site, 153 Ring site, 153 Routh: phase, 46–47, 50, 61, 162; site, 34– 36, 203–04 Saline-Fifteen site, 180, 185, 192 Saline river, 165, 179, 185; valley, 179–80 Sanson Incised, 182 Scallorn point, 75, 78 Scott Place site, 183 Settlement pattern, 8, 14, 21, 24, 35, 37, 72, 90, 95, 132, 143, 151, 204 Shallow Lake site, 178, 179 Shell: artifacts, 81–82, 84, 86, 97, 101–03, 184, 198; midden, 70, 80, 82–86, 90, 99; temper, 4, 7–8, 10–12, 17, 34, 37, 49, 51–52, 56, 60–64, 82, 91, 95–96,

265

100–06, 108–126, 135, 162, 167, 169– 70, 173–75, 178, 180–82, 184–85, 191– 93, 197–99, 204. See also mound, shell. Shell¤sh, 14 Smith Creek site, 145, 154–56, 158–59 Southeastern Ceremonial Complex, 198 Southern Cult, 96, 100, 104, 181–82, 188, 190. See also Southeastern Ceremonial Complex Stirling phase, 58, 166 Tabatiere Perdu site, 97, 99, 103 Taensa, 193–94, 201 Taxonomy, 106 Taylor site, 174, 175, 187, 188 Tchefuncte: culture, 43; period, 71 Tchefuncte Plain, 138 Tchula period, 28, 31 Teche: bayou, 68, 72, 76, 84–85; ridge, 68–69, 85, 90 Temper. See grog temper and shell, temper Temple Mound. See mound Tensas: basin, 3, 9–11, 13–15, 20, 23, 33–36, 38, 46, 55–56, 58–61, 63–64, 133, 152, 163, 165, 176, 187–189, 204; bayou, 26 Tillar: complex, phase, 173, 192; site, 174, 192 Toltec site, 171, 175, 189 Transitional Coles Creek, 6, 9, 13, 15, 17, 18, 20, 46, 58–64, 78, 86, 88, 94–97, 100–01, 104–06, 158, 163, 166, 167, 169, 181, 183, 193, 203–04 Transylvania: phase, 61; site, 11, 61, 204 Tribute, 151 Trinity site, 152 Troyville: complex, 4; culture, 61, 83, 95 Tunica: complex, pottery, 134, 167; Indians, 192, 201 War, 5, 62, 129, 138, 190, 205 Washa tribe, 97 Watts Field site, 179 Willey, Gordon, 3–4, 94–95, 165 Williams, Stephen, 8–9, 20, 27, 52, 54–55, 61, 64–65, 108–09, 135, 162, 175, 194

266

Windsor site, 150, 152, 156–58 Winterville: phase, 61, 111, 162; site, 8, 20, 35, 54, 64, 162, 167, 186, 188, 201, 204 Winterville Incised, 101, 103, 170, 192

index

Yazoo Basin, 1, 7–9, 10–11, 13–15, 18, 20, 23, 25, 35–37, 52, 54–56, 61, 63, 66, 91, 94–95, 108–09, 112, 133, 135, 147, 162, 191, 202 Yokena site, 156, 158