Origins and Spread of Shell-Tempered Ceramics in the Eastern Woodlands: Conceptual and Methodological Frameworks for Analysis
Feathers, James K., Peacock, Evan, Southeastern Archaeology
We briefly review problems in understanding the temporal and spatial distribution of shell-tempered pottery in prehistoric eastern North American, drawing on evidence provided by other papers in this issue of Southeastern Archaeology. Many problems are chronological: When did shell-tempered pottery first appear, when did it rise to high frequencies, and how fast did it replace earlier pottery tempered with other materials? Direct dating of ceramics is required to address these questions. Difficulties with chronology aside, the largest challenge is in understanding why shell-tempered pottery achieved the popularity it did. Both historical and functional explanations are considered: Can migration or diffusion account for its appearance in different areas? Can the selective value of shell-tempered pottery be determined? How has shell's unique firing requirements limited its distribution?
The impetus for the series of papers that appear in this issue of Southeastern Archaeology is an article about shell-tempered pottery recently published by the senior author (Feathers 2006). In doing the research for that article, it became apparent that little information about the spatial and temporal distribution of shell-tempered pottery in Eastern North America was readily available in the published literature; much of what exists is confined to the "gray" literature of cultural mitigation reports or to state newsletters and journals of limited circulation. A major deficiency therefore in the paper's attempt to understand why shell-tempered pottery appears when and where it does was a lack of knowledge about precisely what was to be explained. To begin rectifying this situation, we organized a poster session at the 2007 Society for American Archaeology meetings in Austin, inviting archaeologists who work in the Eastern Woodlands to discuss the origin and spread of shell-tempered pottery in the particular region in which they work. The resulting appreciation for the complexity of the issue should be readily apparent to all who read these papers.
Scholarly interest in shell-tempered pottery began with the first European observers of surviving native potters (Swanton 1946). These early witnesses marveled at the fine pottery that was produced despite what appeared to be relatively low firing temperature (some questioned whether it was fired at all, e.g., Potter 1880). Explanations for the use of shell as temper were published as early as Hohnes (1903), but the realization of the temporal significance of shell-tempered pottery did not materialize until the 1930s when various seriation and stratigraphie works across the Midwest and Southeast placed shell-tempered pottery chronologically late. It was concentrated in the highest stratigraphie levels and often was associated with historic artifacts. Application of the Midwestern Taxonomic System and similar classifications (Deuel 1937; McKern 1939) elevated shell-tempered pottery to the status of diagnostic of the late prehistoric period (roughly after 900 A.D.). It has thus been used as a chronological marker when no other dating has been available. Its apparent association with sedentary maize agriculturalists led to its inclusion as a characteristic trait of what became known as "Mississippian." This early categorization of shell-tempered pottery, however, proved premature. More recent work suggests that shell tempering dates far earlier than the late prehistoric period, and that not all maize agriculturalists used shell tempered pottery nor was all shelltempered pottery made by agriculturalists.
In this essay, we briefly review some of the issues concerning shell-tempered pottery in the Eastern Woodlands, drawing from the evidence provided by the papers in this volume.
Three papers document examples of early shelltempered pottery, appearing in the Ozarks by A.D. 500 (Sabo and Hilliard, this issue), in the upper Mississippi River Valley between A.D. 250 and A.D. 500 (Boszhardt, this issue), and along the mid-Atlantic Coast before 400 B.C. (Herbert, this issue). In the latter two cases, a temporal hiatus is apparent between this initial use of shell-tempered pottery and a later, more substantial use. In both the upper Mississippi and mid-Atlantic, the early shell-tempered pottery coexists with pottery tempered with different materials. Although the technology of this early shell-tempered pottery is not well known, the loss in either case may represent localized technological extinctions, and apparently neither had any influence on subsequent use of shell temper. The Ozarks are a different matter (see also Hilliard and Mainfort 2007). Here the pottery has continuity into the late prehistoric period and is thought by many (e.g., Kelly 2000; Lafferty, this issue; Price and Price 1984) to be ancestral to shell-tempered pottery in the Mississippi Valley. Unfortunately, few if any technical studies have been done on this pottery and the transition from earlier grog-tempered or limestone-tempered pottery is poorly known.
Early shell-tempered pottery may also be present in other places in the Southeast. The senior author, using luminescence, has documented shell-tempered pottery dating to A.D. 200-600 in northern Alabama (Feathers 2008) and perhaps as early as A.D. 400 in northeastern Arkansas. Rafferty and Peacock (this issue) also mention the presence of a shell-tempered specimen in a Middle Woodland context in northeastern Mississippi. Such early evidence has occasionally been reported for other places as well (e.g., Faulkner 1972) and hints at a widespread, although very low frequency use, of shell temper across the Eastern Woodlands for several centuries prior to A.D. 800. A distinction should therefore be made between the first appearance of shell tempering and its later rise to high frequencies. Low frequencies may have persisted until better control of firing was obtained (Feathers 2006). Overfiring of shell-tempered pottery can lead to spalling and destruction of the vessel (Herbert, this issue), which, if occurring often enough, may possibly have caused the aforementioned extinctions and kept shelltempered pottery elsewhere as a minor variant. The real significance of the Ozark material may be that this is where shell-tempered pottery first rose to high frequencies (Hilliard and Mainfort 2007; Sabo and Hilliard, this issue).
But if shell-tempered pottery is present early in low frequencies for several centuries prior to A.D. 800, how has this eluded chronological detection for the most part? We think the problem might be our dating methodology. In all parts of the East, dating methods, particularly the use of radiocarbon, have targeted occupations, rather than manufacture (Feathers 2008). An occupation is an event when a group of artifacts came together in a single place; it is defined at the scale of assemblage with continuity in space, time and form (Rafferty 2001), and the duration can vary enormously. Manufacture, by contrast, is an event, usually of short duration, when certain materials were modified or brought together to form a single object. In tracing the origins of shell-tempered pottery, we really are interested in manufacturing events. Do small amounts of shell-tempered pottery occurring with other kinds of pottery mean that the shell-tempered ware was made at the same time as the other pottery or only that these different kinds of pottery were deposited together as part of a long-duration event?
Radiocarbon dating of organic materials in association with these ceramics cannot answer this question. Rather, methods that directly date pottery manufacture are required. Luminescence, which dates the last time a ceramic was heated (to at least 450°C), is the most straightforward of these methods (Feathers 2003a), but the precision provided by luminescence (5-10 percent) may not always provide the necessary resolution to track fine-scale changes. Radiocarbon dating of organic materials mixed with the clay when the pottery was made may offer a high precision alternative, for example, AMS dating of shell temper (Peacock and Feathers 2009), although an important consideration is whether the event dated by radiocarbon is the same as the manufacturing event (e.g., problem of old shell). In the case of shell, freshwater reservoir effects also must be taken into account (Peacock and Feathers 2009). Radiocarbon dating of residues (e.g., Hilliard and Mainfort 2007; Sabo and Hilliard, this issue) is an alternative, although here the dating event is use of the ceramic, and whether this use can be assumed to be close in time to manufacture needs attention. Separating use residues from postdepositiorial alterations also is required.
This is not to say that radiocarbon dating of associated organics has no value. The large number of available radiocarbon dates, when carefully evaluated, can be useful for detailing general trends, as Lafferty (this issue) ably demonstrates. But the detail suffers because of association assumptions, as Lafferty's frequent suspicion of "secondary deposition" makes clear.
Seriation, because it is based on cultural transmission at the time of manufacture, does address manufacturing events. However, it dates assemblages, not individual artifacts, and thus averaging effects may preclude the dating of low frequency artifacts, such as, for example, early shell-tempered specimens (Feathers 2009). Nevertheless, on a broader scale it can be very useful for identifying trends in manufacture. Rafferty and Peacock (this issue) apply seriation to plot changes in ceramic technology (temper and surface treatment) and use that information to make several inferences about settlement changes, population movements and abandonments.
As Rafferty and Peacock (this issue) point out, seriation, based on paradigmatic classification with mutually exclusive combinations of attributes, is a more elegant way of tracking the structure of change across space and time than the more commonly used comparison of phases, based on differing proportions of ceramic types. Phase terminology is employed throughout many papers in this issue, and for understandable reasons. It has structured Southeastern chronology for decades (see Dunnell 2008 for a critical review) and gives archaeologists a way to communicate with each other. But it has the unfortunate drawback of packaging the archaeological record into boxes that often get reified into sociological units of questionable validity. As differences between phases get emphasized, variability within phases is masked. Phase construction also puts a premium on very carefully defined types with minimal spatial change, leading to persistent problems in identifying types (Weinstein and Dumas, this issue) and to interanalyst bias (Pollack et al., this issue).
If early occurrences of shell-tempered pottery are differentiated from later high frequency use, a critical chronological problem becomes the rate at which shelltempered pottery replaced earlier pottery. In some regions, the change to high frequencies appears to be very rapid, almost instantaneous in terms of our dating resolution (e.g., Feathers 2003b), but in other regions the rate is much slower. Lafferty (this issue) uses abundant radiocarbon data from sites in the northern part of the lower Mississippi Valley to show that shelltempered pottery replaced sand-tempered pottery fairly rapidly after A.D. 900 in the western part of the region. But it replaced grog-tempered pottery at a much slower rate in the eastern part, where grogtempered pottery persisted as late as A.D. 1300, if not later. Lafferty (this issue) also points out that shell tempering in this region was first introduced as a supplemental temper (i.e., mixed with other tempering materials in the same ceramic), something which occurred in other places as well (Weinstein and Dumas, this issue; Pollack et al., this issue). Careful documentation also allows Pollack et al. (this issue) to contrast rates from two adjacent regions. The rate of change in frequency of shell-tempered pottery in northern Kentucky was faster than that in central Kentucky, a difference they attribute in part to differential access to shell. They note, however, as do Rafferty and Peacock (this issue), that even where access is limited the eventual near-100 percent use of shell-tempered pottery signals a high selective value. Such fixation is facilitated by high interaction levels among groups, as both Pollack et al. (this issue) and Cook and Fargher (this issue) argue. Different degrees of isolation earlier may account in part for the different rates of acceptance in Kentucky.
Such high selective value, which will be discussed further in the next section, is all the more interesting when one considers where shell-tempered pottery was not selected: eastern Alabama, much of Georgia, peninsular Florida, portions of the lower Mississippi Valley between Vicksburg and Baton Rouge, and southwestern Louisiana. Weinstein and Dumas (this issue) discuss the particularly puzzling spatial gradation from shell-tempered pottery to sand-tempered pottery in the Florida panhandle, despite similarity in settlement patterns, subsistence and other features across the same region. A similar kind of grathent appears to be present in the Mississippi Valley below the Arkansas River (Kidder 1998).
While obtaining an accurate documentation of the temporal and spatial distribution of shell-tempered pottery has proven difficult, at least in the details, the larger challenge is trying to explain that distribution. Explanations for shell tempering typically emphasize either historical origins or functional adaptation (Feathers 2006), although most archaeologists recognize the need for both kinds of explanations. Historical explanations can never be enough. Pottery is not analogous to mitochondrial DNA, that is, only a matter of transmission, and therefore cannot serve unambiguously as a proxy for groups of interacting people. Pots are first and foremost containers, and thus are subject to any number of selective forces that govern their properties to that end. On the other hand, selection only works on available variability, so it is equally important to consider historical factors. At least some aspects of ceramics - decorative features, for example - are heavily influenced by transmission. The mixture of decorative types in southeastern Louisiana, for example, probably mainly reflects transmission conduits to both the east and west (Weinstein and Dumas, this issue).
Scale is also important. What may serve as an explanation at one level may have little explanatory power at another level. For example, shell-tempered pottery may have been adopted in one place because of a certain adaptive property, such as having greater mechanical strength than previous pottery. But if the people who make this shell-tempered pottery move to some other location, bringing their pottery tradition with them, shell-tempered pottery at this second location cannot be explained by greater mechanical strength. It may even have lower strength than whatever pottery existed at this location previously. Rather, it replaced the earlier pottery simply because different people moved there. In other words, the change occurred at a higher scale, and functional explanations will need to be directed not at the kind of pottery made, which might suffice at the original location, but at what allowed the people to make such a move.
Such migration, in fact, has been a traditional explanation for the spread of shell-tempered pottery in the Eastern Woodlands (Caldwell 1958; Smith 1984), and it is still employed today (e.g., Blitz and Lorenz 2002) although more in a case-specific context than as a pan-Eastern phenomenon. Weinstein and Dumas (this issue) review evidence that the appearance of shelltempered pottery around Mobile Bay is the result of migration of people from central Alabama (Moundville) keen on establishing an outpost to tap salt and marine resources. Boszhardt (this issue) also argues that people migrated north along the Mississippi to establish villages in the upper Mississippi, bringing with them shell-tempered pottery. But indigenous Oneota groups also manufactured shell-tempered pottery. Chronological comparisons suggest the Oneota learned shell tempering from the immigrants, but because of the early experience with shell tempering in the region, its indigenous roots cannot totally be ruled out. Herbert (this issue) reports that shelltempered pottery along the mid-Atlantic traditionally has been considered part of an expansion of northern Algonkians into the area, but he also shows that shelltempered pottery has a much deeper tradition in the mid-Atlantic. Finally, Jenkins (2003) has argued that shell-tempered pottery in central Alabama (including that at Moundville) can be attributed to migration south from the Tennessee River. He points out that Woodland age groups evolved in place for a thousand years but, in his view, never developed the more complex subsistence and settlement patterns that characterized the sudden appearance of "Mississippian" groups, who brought shell-tempered pottery with them. He provides evidence that house shape, burial patterns, and lithic technology also changed along with the pottery and notices that shell tempering did not extend to eastern Alabama despite similar Late Woodland adaptations. The low frequencies of shelltempered pottery found in the Late Woodland West Jefferson sites he attributes to small numbers of immigrants. His position is controversial because others have interpreted the same evidence differently and have stressed continuities between the shelltempered and earlier pottery (e.g., Mistovich 1988; Rafferty and Peacock, this issue; Welch 1994, 1998). Jenkins (2007, personal communication) has cited climatic change, facilitating the spread of maize agriculture, as the reason for the migration. Rafferty and Peacock (this issue) also attribute agricultural expansion in the Black Prairie of Mississippi to favorable climate.
Whatever the explanation for central Alabama, there is less dispute about the case in the Ohio Valley, where neither Cook and Fargher (this issue) nor Pollack et al. (this issue) argue for a migration, at least at any large scale. Here, the early shell-tempered pottery resembles the earlier pottery in every way but temper, and there are continuities in other artifacts. Both papers attribute shell tempering there to diffusion, although with a different emphasis. Pollack et al. (this issue) see shell tempering being adopted at about the same time in both the Ohio Valley and the central Mississippi River valley, seeing it as an idea that spread along similar routes as the bow and arrow or maize agriculture. The rate of acceptance, as mentioned earlier, varied regionally. Cook and Fargher (this issue) are more interested in the later period, when shell tempering rose to high frequencies. This rise they attribute to contact with "Mississippian" culture, showing, for example, at one site, shell temper associated only with a wall trench house, which also reflects Mississippian influence. Rather than trying to understand some higher level cause, their research is aimed at understanding the mechanism of diffusion that led to high frequencies. They trace the movement of pottery among sites in southern Ohio through petrographic analysis, again attributing increasing use of shelltempered pottery to higher rates of interaction among villages.
The question of "Mississippian" influence, which is commonly referred to in discussions of shell temper origins, raises issues about what "Mississippian" means and what is implied by such influence. Smith (1984) addressed such questions more than 20 years ago, but they remain largely unresolved, although arguments for migration are much more sophisticated now than when Smith was writing (e.g., Blitz and Lorenz 2002). As a most common denominator, "Mississippian" is probably nothing more than an illdefined constellation of traits involving subsistence, settlement and technology. Any implication of higher scale organization is still a matter of conjecture (although see Lipo et al. 1997). What is probably more important for this discussion is that each of these traits, such as shell tempering, has its own history independent of the others. Few of them may actually be confined to the late prehistoric period, and they did not coalesce at the same time or in the same place (Smith 1984). Likewise "Mississippian influence" probably cannot mean much more, at this point, than an indication of contact and sharing of some of these traits.
Cook and Fargher's (this issue) study also demonstrates the kind of information that can be obtained through sourcing. Not all pottery is locally made. Imported pottery can be identified by sourcing studies, using either pétrographie information (Galaty 2008; Stoltman et al. 2008) or chemical analyses (Neff 2008). Of particular interest where shell-tempered pottery is concerned is a new method that employs Laser Ablation-Inductively Coupled Plasma-Mass Spectrometry to chemically analyze the temper particles, based on the fact that mussels in different waterways incorporate different chemical loads into their shells (Peacock et al. 2007). In one example, petrographic work by Stoltman (2005) and chemical analysis by Rodríguez-Alegría (2005) showed that shell-tempered pottery from the Lake Providence site in northeastern Louisiana was all imported between A.D. 1150 and A.D. 1250. Petrographic work has also suggested early shell-tempered pottery was imported into the American Bottom (Kelly 1991) and from the American Bottom into the upper Mississippi River Valley area (Boszhardt, this issue; Stoltman et al. 2008).
Diffusion, of course, is a historical explanation. A more complete explanation also requires understanding why shell tempering was adopted in the receiving society. This is not a trivial matter, given the different rates of acceptance that have been documented and the fact that people in some places did not adopt shell tempering at all despite exposure to it.
Because of continuity in other ceramic attributes, Pollack et al. (this issue) argue that it was the use of shell temper alone that diffused, not other aspects of the pottery. This is an interesting observation. Temper is an archaeological term used to classify pottery. For the potter, it is an integral part of the production sequence and what temper is used depends on a host of variables including the nature of the clay, accessibility, forming and firing strategies and the uses to which the pottery is put. Temper is not something that can be interchanged at will, even though it is employed in classifications and seriations as if transmission were the only factor influencing its presence. Rather, temper has selective value and replacing one temper with another will affect all of the aforementioned variables. It is not surprising that when temper changes, other variables change as well. In parts of the Mississippi Valley, for example, the increase in frequency of shell-tempered pottery is also accompanied by an increase in slipping (Kelly 2000; Boszhardt, this issue; Feathers 2006). The senior author (Feathers 2006) has hypothesized that this may have to do with the increased porosity that using shell caused (at least until potters figured out how to avoid this), requiring some kind of surface finish as a counteraction. So it is remarkable that shell-temper replaced earlier tempers in the Ohio Valley without any other changes. How could this have happened?
It is well known that presence of calcium carbonate in pottery requires a careful firing strategy (Herbert, this issue). Carbonate decomposes into lime and CO2 at about 600-800°C, the exact temperature depending on the rate and duration of heating. Lime (CaO) readily hydrates when cooled and exposed to moisture, causing an expansion that can result in spalling or destruction of the vessel. If a potter is using sand temper and regularly firing vessels to 800°C in an oxidizing atmosphere, a switch to shell temper cannot be undertaken without altering the firing strategy. A number of solutions to this problem have been proposed: lower firing temperatures, addition of salt to the fabric (Boszhardt, this issue; see also Avery 1983, 1987; Stimmell 1978; Stimmell et al. 1982), wetting with water while still hot, or reducing atmospheres where presence of CO2 in the atmosphere delays decomposition. (Herbert, this issue, has questioned the efficacy of some of these methods.) The senior author has proposed that in southeastern Missouri an increased use of reduction in firing, perhaps brought about by a change in fuel, allowed shell temper to be used without frequent failures (Feathers 2006). Where such solutions were not realized, shell-tempered pottery may never have gotten a foothold, perhaps explaining why shell tempering reached a dead-end in Georgia and Florida.
So how could shell simply replace other tempering material in the Ohio Valley? One possibility is that the potters there already had experience with another carbonate temper, limestone, which perhaps could be easily exchanged for shell. A similar argument could be made for the southern part of the American Bottom (Kelly 2000; see also Hoard et al. 1995) and in eastern Tennessee, where the transition from limestone to shell did not appear to change the properties of the ceramics (Claassen 2007, personal communication). (Limestone pottery in peninsular Florida, in contrast, was never replaced by shell tempering). Another consideration, however, is the differences in size and shape between shell and limestone particles. The platy shell, with large surface-to-volume ratio, may have produced advantages in strength and workability (Feathers 2006) but may have also increased the risk of spalling. An important clue may be the increase in reduction-firing documented in eastern Kentucky just as shell-tempered pottery was reaching 100 percent (Pollack et al., this issue). Perhaps the gradual acceptance of shell-tempered pottery in this region was not just due to accessibility but also due to lack of sufficient control of firing. When that control was achieved, shell tempering became fixed no matter what the accessibility. The early mixture of shell with other tempering materials, noted in many papers for several areas, may also represent an attempt to lessen the impact of shell on successful firing.
Much of the previous paragraph, of course, is pure speculation, because we do not have the detailed technological studies that are necessary to make these kinds of arguments. Boszhardt (this issue) reports some initial work in the upper Mississippi Valley. Herbert's (this issue) experimental work, while just beginning, also is important in this regard. An interesting line of evidence to pursue would be geographical variation of forest cover (as reconstructed), and hence fuel wood, that would have been readily available to potters. In environmentally constricted areas where prime fuel wood could have been seriously depleted by sedentary populations, the adoption of shell temper may be explainable by the necessity of using poorer fuels and consequent selection for a kind of pottery that could be produced under such conditions. This is a hypothesis that should be testable with sufficient chronological control, survey data, and archaeobotanical analysis at both long-term and short-term occupations. It would be particularly interesting to test in areas such as the Ozarks where high levels of shell tempering became fixed at a relatively early date.
Where this hypothesis does not hold, selection of shell-tempered pottery should not have occurred, even if control of firing was obtained, unless the properties of the pottery provided advantages that the earlier pottery did not have - advantages that were sufficient to overcome risk aversion to altering established recipes. (Weinstein and Dumas [this issue] discuss how conservative potting traditions may have impeded the spread of shell tempering along parts of the Gulf Coast.) Unfortunately, this is the one area where Southeastern archaeologists have the least information. Few comparisons of ceramic properties of shelltempered pottery versus other kinds of pottery are available. Even where properties have been measured, it is difficult to link such measurements with how pottery was actually used, to gauge whether changed properties make a significant difference (Weinstein and Dumas, this issue). Correlations often have been drawn between shell-tempered pottery and maize agriculture or even domesticated seed cultivation (Sabo and Hilliard, this issue), but the correlation is far from perfect and the exact role that shell-tempering has with maize agriculture is unclear. The above-mentioned hypothesis concerning fuel depletion could explain the approximate temporal correlation between shell tempering and maize agriculture that occurs in many areas, given the pressures on fuel wood that can accompany agricultural settlement (e.g., Lopinot and Woods 1993). On another tack, no one has ever followed up on the suggestion (put forth by Morse and Morse 1983, but elaborated by Osborn 1988) that shell temper provided a source of lime for increasing the nutritional value of maize, although it is unlikely that shell-tempered pottery was ever fired high enough to assume this role.
A number of papers have mentioned how variation in the shapes and sizes of pottery vessels increased after the advent of shell-tempered pottery. Another interesting pattern is the gradual change from loop to strap handles, something that occurs as far north as the upper Mississippi River Valley (Boszhardt, this issue) to as far south as at least the central Tombigbee and Black Warrior River valleys (Peacock 1995; Steponaitis 1983) and east up the Ohio River (Pollack et al., this issue) and into eastern Tennessee (Lewis and Kneberg 1946). But these changes may be consequences of the adoption of shell temper, not the cause (Feathers 2006). The workability of shell-tempered clay, discussed in several papers in terms of ideal amounts, sizes and shapes of temper particles, may have facilitated the expansion in morphology. The resulting variability has interesting implications for evolutionary studies of Mississippian and Protohistoric period ceramics.
When vessel uses are unknown in any detail, and how properties of different ceramics may relate to them are not clear, it is tempting to fall back upon social interpretations, such as prestige or adoption of a new belief system. While such scenarios are plausible, they are difficult to test. More attention needs to be paid to how such social variables as prestige can be identified in the archaeological record. Recent research into costly signaling, for example, may provide clues as to how such explanations could be structured (McGuire and Hildebrandt 2005). It is also easy to confuse cause with effect. Prestige may be a consequence of shell-tempered pottery, not necessarily the reason for its adoption. Both Cook and Fargher (this issue) and Lafferty (this issue) argue that prominent individuals provided the conduit for the acceptance of shell-tempered pottery, but as Cook and Fargher acknowledge, this is only a proximal explanation for the acceptance, not the ultimate one. Many social explanations involve intention, which at best has only proximal explanatory value as well. There is a difference between what potters did and what they thought they were doing. Archaeologists have the benefit of the long hindsight of time. They are able to put changes into a broader context that the potters, confined to brief intervals of space and time, could not appreciate. Perhaps we are fortunate that we only have a record of what they did, unencumbered by what they thought. Because even understanding what they did, as the papers in this volume testify, is likely to occupy our research efforts for some time.
Acknowledgments. We wish to thank Gayle Fritz and T. R. Kidder for encouragement and flexibility as we worked with several authors to compile this special issue of Southeastern Archaeology and for the comments they made on this paper. Bob Mainfort provided information relating to Ozark ceramics. We also wish to thank all the authors who participated in the Society for American Archaeology symposium that led to this compilation, including those whose papers are not in this issue but hopefully will appear in future issues of the journal.
Avery, George E.
1982 Salt, Pots, and Diet: Replication Studies of Late Prehistoric Shell-Tempered Ceramics. Unpublished master's thesis, Department of Anthropology, Southern Illinois University, Carbondale.
1987 Salt and Shell-Tempered Pottery: An Examination through Replicative Experiments and X-Ray Diffraction Analysis. In The Emergent Mississippian: Proceedings of the Sixth Mid-South Archaeological Conference, June 6-9, 1985, edited by Richard A. Marshall, pp. 22-31. Occasional Papers No. 87-01. Cobb Institute of Archaeology, Mississippi State University, Mississippi State.
Blitz, John H., and Karl G. Lorenz
2002 The Early Mississippian Frontier in the Lower Chattahoochee-Apalachicola River Valley. Southeastern Archaeology 21:117-135.
Caldwell, Joseph R.
1958 Trend and Tradition in the Prehistory of the Eastern United States. Memoirs of the American Anthropological Association 88. Menasha, Wisconsin.
1937 The Application of a Classificatory Method to Mississippi Valley Archaeology. In Rediscovering Illinois: Archaeological Explorations in and around Fulton County, edited by F.-C. Cole and T. Deuel, pp. 207-219. University of Chicago Press, Chicago.
Dunnell, Robert C.
2008 Archaeological Things: Languages of Observation. In Time's River: Archaeological Syntheses from the Lower Mississippi River Valley, edited by Janet Rafferty and Evan Peacock, pp. 45-68. University of Alabama Press, Tuscaloosa.
Faulkner, Charles H.
1972 The Mississippian-Woodland Transition in the Middle South. Bulletin of the Southeastern Archaeological Conference 15:38-45.
Feathers, James K.
2003a Use of Luminescence Dating in Archaeology. Measurement Science and Technology 14:1493-1509.
2003b Comments I: Accounting for Ceramic Change. Archaeometry 45:163-169.
2006 Explaining Shell-Tempered Pottery in Prehistoric Eastern North America. Journal of Archaeological Method and Theory 13:89-133.
2009 Problems of Ceramic Chronology in the Southeast: Does Shell-Tempered Pottery Appear Earlier than We Think. American Antiquity, in press.
Galaty, Michael L.
2008 Ceramic Petrography and the Classification of Mississippi's Archaeological Pottery by Fabric: A GIS Approach. In Time's River: Archaeological Syntheses from the Lower Mississippi River Valley, edited by Janet Rafferty and Evan Peacock, pp. 243-273. University of Alabama Press, Tuscaloosa.
Hilliard, Jerry E., and Robert C. Mainfort, Jr.
2007 The Ira Spradley Field Site: A Late Woodland Cemetery in the Arkansas Ozarks. Southeastern Archaeology 26:269-291.
Hoard, Robert J., Michael J. O'Brien, M. G. Khorasgany, and V. S. Gopalaratmam
1995 A Materials Science Approach to Understanding Limestone-Tempered Pottery from the Midwest. Journal of Archaeological Science 22:823-832.
Holmes, William H.
1903 Aboriginal Pottery of the Eastern United States. Twentieth Annual Report of the Bureau of American Ethnology, 1898-1899, pp. 1-201. Smithsonian Institution, Washington, D.C.
Jenkins, Ned J.
2003 The Terminal Woodland/Mississippian Transition in West and Central Alabama. Journal of Alabama Archaeology 49:1-62.
Kelly, John E.
1991 The Evidence for Prehistoric Exchange and Its Implications for the Development of Cahokia. In New Perspectives on Cahokia: Views from the Periphery, edited by James B. Stoltman, pp. 65-92. Monographs in World Archaeology No. 2. Prehistory Press, Madison, WI.
2000 The Nature and Context of Emergent Mississippian Cultural Dynamics in the Greater American Bottom. In Late Woodland Societies: Tradition and Transformation over the Midcontinent, edited by Thomas E. Emerson, Dale L. McElrath, and Andrew C. Fortier, pp. 163-175. University of Nebraska Press, Lincoln.
Kidder, Tristram R.
1998 Mississippian Period Mound Groups and Communities in the Lower Mississippi Valley. In Mississippian Towns and Sacred Spaces: Searching for an Archaeological Grammar, edited by R. Barry Lewis and Charles Stout, pp. 123-149. University of Alabama Press, Tuscaloosa.
Lewis, Thomas M. N., and Madeleine Kneberg
1946 Hiwassee Island: An Archaeological Account of Four Tennessee Indian Peoples. University of Tennessee Press, Knoxville.
Lipo, Carl P., Mark E. Madsen, Tim Hunt, and Robert C. Dunnell
1997 Population Structure, Cultural Transmission, and Frequency Seriation. Journal of Anthropological Archaeology 16:301-333.
Lopinot, Neal H., and William I. Woods
1993 Wood Overexploitation and the Collapse of Cahokia. In Foraging and Farming in the Eastern Woodlands, edited by C. Margaret Scarry, pp. 206-231. University of Florida Press, Gainesville.
McGuire, Kelly R., and William R. Hildebrandt
2005 Rethinking Great Basin Foragers: Prestige Hunting and Costly Signally during the Middle Archaic Period. American Antiquity 70:693-710.
McKern, William C.
1939 The Midwestern Taxonomic Method as an Aid to Archaeological Study. American Antiquity 4:301-313.
Mistovich, Timothy S.
1988 Early Mississippian in the Black Warrior Valley: The Pace of Transition. Southeastern Archaeology 7:21-38.
Morse, Dan F., and Phyllis A. Morse
1983 Archaeology of the Central Mississippi Valley. Academic Press, New York.
2008 Review of Ceramic Compositional Studies from In and Around the Mississippi Valley. In Time's River: Archaeological Syntheses from the Lower Mississippi River Valley, edited by Janet Rafferty and Evan Peacock, pp. 223-242. University of Alabama Press, Tuscaloosa.
1988 Limitations of the Diffusionist Approach: Evolutionary Ecology and Shell-Tempered Ceramics. In The Transfer and Transformation of Ideas and Material Culture, edited by P. J. Hugill and D. B. Dickson, pp. 23-44. Texas A&M University Press, College Station.
1995 Test Excavations at an Upland Mississippian Site in Oktibbeha County, Mississippi. Mississippi Archaeology 30(2):1-20.
Peacock, Evan, and James K. Feathers
2009 AMS Radiocarbon Dating of Temper in Shell-Tempered Ceramics: Test Cases from Mississippi, Southeastern U.S.A. American Antiquity, in press.
Peacock, Evan, Hector Neff, Janet Rafferty, and Thomas Meaker
2007 Using Laser Ablation-Inductively Coupled PlasmaMass Spectrometry (LA-ICP-MS) to Source Shell in Shelltempered Pottery: A Pilot Study from North Mississippi. Southeastern Archaeology 26:319-329.
Potter, W. B.
1880 Archaeological Remains in Southeast Missouri. Contributions to the Archaeology of Missouri by the Archaeological Section of the St. Louis Academy of Sciences, 5-19. New York.
Price, James E., and Cynthia R. Price
1984 Phase II Testing of the Shell Lake Site, 23WE-627, Near Wappapello Dam, Wayne County, Missouri, 1984. St. Louis District, Cultural Resources Management Report No. 11. U. S. Army Corps of Engineers.
2001 Determining Duration at Prehistoric Sites: Short-Term Sedentary Settlement at Josey Farm, NE Mississippi. Journal of Field Archaeology 28:347-366.
Rodríguez-Alegría, Enrique, Robert J. Speakman, and Michael 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 Richard A. Weinstein, pp. 565-578. Coastal Environments, Inc., Baton Rouge. Submitted to Vicksburg District, U.S. Army Corps of Engineers.
Smith, Bruce D.
1984 Mississippian Expansion: Tracing the Historical Development of an Explanatory Model. Southeastern Archaeology 3:13-32.
Steponaitis, Vincas P.
1983 Ceramics, Chronology, and Community Patterns: An Archaeological Study of Moundville. Academic Press, New York.
1978 A Preliminary Report on the Use of Salt in Shell Tempered Pottery of the Upper Mississippi Valley. Wisconsin Archeologist 59(2):266-274.
Stimmell, Carole, Robert B. Hermann, and R. G. Hancock
1982 Indian Pottery from the Mississippi Valley: Coping with Bat Materials. In Archaeological Ceramics, edited by Jacqueline S. Olin and Alan D. Franklin, pp. 219-228. Smithsonian Institution Press, Washington, D.C.
Stoltman, James B.
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 Richard A. Weinstein, pp. 545-563. Coastal Environments, Inc., Baton Rouge. Submitted to Vicksburg District, U.S. Army Corps of Engineers.
Stoltman, James B., Danielle M. Benden, and Robert F. Boszhardt
2008 New Evidence in the Upper Mississippi Valley for PreMississippian Cultural Interaction with the American Bottom. American Antiquity 73:317-336.
Swanton, John R.
1946 The Indians of the Southeastern United States. Smithsonian Institution Bureau of American Ethnology Bulletin 137: 549-555.
Welch, Paul D.
1994 The Occupational History of the Bessemer Site. Southeastern Archaeology 13:1-26.
1998 Middle Woodland and Mississippian Occupations of the Savannah Site in Tennessee. Southeastern Archaeology 17:79-92.
James K. Feathers, University of Washington, Department of Anthropology, Box 353100, Seattle, WA 98195 (email@example.com)
Evan Peacock, P.O. Box AR, Mississippi State University, Mississippi State, MS 39762 (firstname.lastname@example.org)…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Origins and Spread of Shell-Tempered Ceramics in the Eastern Woodlands: Conceptual and Methodological Frameworks for Analysis. Contributors: Feathers, James K. - Author, Peacock, Evan - Author. Journal title: Southeastern Archaeology. Volume: 27. Issue: 2 Publication date: Winter 2008. Page number: 286+. © Southeastern Archaeological Conference Summer 2008. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.