Woodland Potters and Archaeological Ceramics of the North Carolina Coast
352
Woodland Potters and Archaeological Ceramics of the North Carolina Coast
352eBook
Related collections and offers
Overview
Pottery types, composed of specific sets of attributes, have long been defined for various periods and areas of the Atlantic coast, but their relationships and meanings have not been explicitly examined. In exploring these relationships for the North Carolina coast, this work examines the manner in which pottery traits cross-cut taxonomic types, tests the proposition that communities of practice existed at several scales, and questions the fundamental notion of ceramic types as ethnic markers.
Ethnoarchaeological case studies provide a means of assessing the mechanics of how social structure and gender roles may have affected the transmission of pottery-making techniques and how socio-cultural boundaries are reflected in the distribution of ceramic traditions. Another very valuable source of information about past practices is replication experimentation, which provides a means of understanding the practical techniques that lie behind the observable traits, thereby improving our understanding of how certain techniques may have influenced the transmission of traits from one potter to another. Both methods are employed in this study to interpret the meaning of pottery as an indicator of social activity on the North Carolina coast.
Product Details
| ISBN-13: | 9780817381196 |
|---|---|
| Publisher: | University of Alabama Press |
| Publication date: | 12/11/2009 |
| Sold by: | Barnes & Noble |
| Format: | eBook |
| Pages: | 352 |
| File size: | 15 MB |
| Note: | This product may take a few minutes to download. |
About the Author
Joseph M. Herbert is an archaeologist with the Directorate of Public Works, Cultural Resources Program, Fort Bragg Garrison Command.
Read an Excerpt
Woodland Potters and Archaeological Ceramics of the North Carolina Coast
By Joseph M. Herbert
The University of Alabama Press
Copyright © 2010 Joseph M. HerbertAll right reserved.
ISBN: 978-0-8173-5517-3
Chapter One
Sketch of the Region's Woodland History
The inception of pottery making that marks the beginning of the Woodland era lies far back in time, perhaps two millennia before the Christian era and more than three millennia before the first Europeans were noting linguistic differences and recording the whereabouts of the coastal Indian tribes. As we move back in time we quickly lose any surety that the cultural geography, as represented by regional similarities in language, remains faithful to those first ethnohistoric sketches. Nevertheless, it is assumed that the geography of precontact Indian cultures is reflected in the archaeological patterns of ceramic technological styles. Tracing the shifting boundaries of these ceramic techno-stylistic regions is one of the principal goals of this study.
For convenience, the Woodland era is divided into Early (2200-400 B.C.), Middle (400 B.C.-A.D. 800), and Late (A.D. 800-1600) periods. In theory, such boundaries are designed to coincide with fundamental shifts in technology and major differences in social and economic traditions. In practice, however, the data with which to measure shifting socioeconomic conditions are often uneven or totally lacking. Insuch cases, the temporal boundaries of subperiods are often based on differences in a limited suite of material culture traditions including pottery and are, more often than not, imposed as a convenient means of establishing a chronological framework. In many parts of eastern North Carolina, where even the most basic subsistence data are absent, economic model building must rely on inference.
Early Woodland Period (2200-400 B.C.)
The Early Woodland period in eastern North Carolina begins about 2200 B.C. and is distinguished from the preceding Late Archaic principally by the emergence of ceramic technology. Studies on the lower Savannah River and Coastal Plain of South Carolina have shown that the earliest pottery made in the Atlantic coastal region (fiber-tempered Stallings) has been found in contexts dated at least as early as 2500 B.C. (Sassaman 1993:102-110, Figure 11, Appendix). As the earliest Stallings pottery is contemporary with contexts bearing Late Archaic period Savannah River phase materials, many researchers prefer to assign the early portion of the Stallings phase to the Late Archaic period. Finding no advantage in assigning the North Carolina Stallings potters to the Archaic era, I have chosen to stick to the more traditional formula: as pottery emerges, so begins the Woodland era. Regardless of culture-era affiliation, ceramic technology first emerged on the Atlantic coast around 2200 B.C. and evolved into an essential technology over the next 500 years. In the Savannah River valley, perforated soapstone disks or slabs, presumably used in basket or bladder cooking, appeared about 3000 B.C. (Sassaman 1993:185). This was followed in about 300 years by the innovation of ceramic vessel technology, with which it coexisted for many generations. At present it appears that the practice of pottery making spread from the Savannah Valley into the North Carolina Coastal Plain during this same period, by both overland (inland) and maritime (coastal) transportation and trade networks.
Sassaman (1993) contends that the earliest fiber-tempered vessels in the Savannah region were not placed directly on cooking fires but were used as containers for boiling-stone cookery. This may also be true of early pottery from North Carolina, such as the Croaker Landing series (Byrd 1999; Egloff et al. 1988; Pullins et al. 1996). Although the Stallings and Croaker Landing pottery regions are found several hundred miles apart, they appear to reflect very similar technological styles. Vessels were typically thick-walled, slab-built, flat-bottomed containers that were inherently porous, rather soft as a result of low firing temperature, and probably cumbersome-in many ways, a poor substitute to soapstone bowls but eminently more efficient as one did not need to travel to the Piedmont to obtain the raw materials for making clay pots. In just a few centuries, this technological practice led the way to broad-scale experimentation in pottery making. On the southern and central coasts, quartz in a variety of sizes and densities became the standard ingredient in paste. The slab-built, flat-bottomed method of construction was superseded by coil building with the paddle-and-anvil system that produced thin-walled, conical-based pots that were fired at higher temperatures and, thus, represented extremely effective cooking vessels for use directly on the fire.
This second generation of Early Woodland vessels along the North Carolina coast were tempered with a variety of substances including primarily quartz sand (from fine to granule sized) and crushed rock (including quartz, sandstone, granite, and soapstone). Crushed metamorphic stone such as saprolitic granite was sometimes used to temper pottery found on sites in the Sandhills. Ingredients that reflect Piedmont sources suggest the perpetuation of the Late Archaic procurement networks by which soapstone slabs and bowls, along with rhyolite for projectile points and tools, were acquired (Culpepper et al. 1999; Sassaman 1993; Waselkov 1982).
Middle Woodland Period (400 B.C.-A.D. 800)
The greater archaeological visibility of Middle Woodland sites may be due to generally increased population size and dispersion, but it is also the result of more routine use, breakage, and discard of ceramic cooking vessels fired at higher temperatures, whose sherds preserve more readily than Early Woodland pottery. There were several significant changes in ceramic technology that occurred at this time. Fiber tempering was superseded by sand tempering on the South Carolina coast. Limestone tempering appears in the Hamp's Landing series on the southern coast of North Carolina and in the Wando series of northern South Carolina. For the Middle Woodland period in the northern coastal region, Phelps (1983) suggests that the size grade of sand temper used in the New River series (Deep Creek) shifts from coarse to medium, while the proportion of granule- and pebble-sized temper of the Mount Pleasant series increases (cf. results from the Bandon midden in Chapter 4).
Middle Woodland vessel shapes include larger jars with conical bases and straight walls, a transformation that emphasizes the intensified use of ceramic containers as cooking vessels. The carved-paddle stamping technique that emerged in the Refuge and Deptford phases in South Carolina and in the Hamp's Landing and New River series in North Carolina expanded to include wrapped-paddle cord marking and fabric and net impressing in North Carolina. Very few Middle Woodland sand-tempered specimens from the North Carolina Coastal Plain exhibit simple stamping. This stands in distinct contrast to the Early Woodland New River and Hamp's Landing series. Simple stamping occurs in both the Cape Fear and Hanover series in coastal South Carolina (Anderson et al. 1982) but not in North Carolina, where it is found to reappear only in the Late Woodland Townsend (Colington) and Cashie series (Phelps 1981a, 1983).
Late Woodland Period (A.D. 800-1600)
During the first centuries of the Late Woodland period, economic, organizational, and ideological structures in eastern Carolina are marked by increasing sedentism, tribalization, territoriality, and ceremonialism. Sedentary villages evidenced by site size, architectural and storage facilities, and the remains of cultivars such as squash, beans, and corn are particularly evident on the coast and in the tidewater region where abundant marine resources helped to sustain village life. Current evidence suggests significant differences in residential settlement strategies in the various areas of eastern Carolina. In the Sea Island and Sandhills sections of the Coastal Plain sites are small and broadly scattered, suggesting relatively high mobility and dispersed settlement (Culpepper et al. 1999). In the Embayed Section of the coast, larger village sites located along the sounds, estuaries, major rivers, and their tributaries suggest more sedentism and regionalization (Phelps 1983:39).
Three cultural regions have been proposed for coastal North Carolina that conform to ethnohistoric records of linguistic differences: Algonquian speakers resided in the northern and central Embayed Coastal region, Iroquoian speakers occupied the northern interior Coastal Plain, and Siouan speakers inhabited the southern Coastal Plain.
Archaeologically, these culture areas are recognized by regional differences in ceramics (Phelps 1983) and notable differences in burial customs and architectural forms (Loftfield 1990; Loftfield and Jones 1995). Phelps (1983:39) suggests that the southern extent of the Algonquian culture and the corresponding distribution of Colington ceramics were just south of the Neuse River (cf. Loftfield 1976). Loftfield and Jones (1995) have suggested that the distribution of Late Woodland Colington phase cultures may have extended as far south as Onslow County, North Carolina. These authors cite evidence including not only shell-tempered pottery but also mortuary and architectural features, for example, ossuary burials and long houses typical of Algonquian culture found to the north.
This regionalism, expressed in pottery technology, is perhaps best represented by the shell-tempered Townsend series Colington and White Oak types. Clearly bounded in geographic distribution, the makers of the shell-tempered Townsend series, found on the northern coast of North Carolina, constructed vessels with simple-stamped and fabric-impressed surfaces, often fired in a reduced atmosphere. Their southern neighbors made very similar shell-tempered ware with some interesting differences. The White Oak type, restricted to the central coast of North Carolina, does not appear to include a simple-stamped variant but does have a burnished type only rarely seen in Colington.
Archaeological data provide less evidence of Late Woodland adaptations on the southern coast and in the Sandhills. Whereas in the central and northern coastal regions evidence of small villages with several associated house structures has been found, both palisaded and open, nothing on this scale has been found on the southern coast or in the Sandhills. Although communication and social cohesion seem to have existed in these areas, extended- family dwellings appear to have been scattered across the landscape.
Chapter Two
Explaining Ceramic Patterns
Any exposition of the behavioral and social causes lying behind the spatial and temporal patterning of pottery in a given region requires some explanation, and in this chapter I consider some of the issues involved.
Methodological Issues
Constructing Types and Varieties
The taxonomic classes of pottery used in this study are considered to be definitional referents, that is, products of observation that focus on the modal expression of a set of traits exhibited by specimens in an assemblage, rather than on the variability continuously expressed in the sample. The implication of using defined (static) sets of attributes to model continuous (dynamic) variation is what Dunnell (1995a) has referred to as the materialist paradox. Th is conceptual problem draws our attention to the critical importance of maintaining the distinction between empirical and theoretical units. Artifact classes do not evolve; classes are counting units that partition variation so that when arrayed chronologically, distinct classes are differentially represented over time (Dunnell 1995a; Jones et al. 1995; Teltser 1995). The usefulness of types for assessing or resolving questions about variation in space and time depends upon the sets of traits of which they are composed.
Traditionally, the process of defining pottery types has arisen from the desire to identify criteria (attribute sets) sufficient to segregate groups of potsherds thought to have culture-historical significance. Logically, the process has moved inductively from the observation of variation to the formation of classes. Reasoning has then worked from hypotheses (often including assumptions about the linkage between regularly patterned data and cultural behavior) to inferences about the empirical data; that is, if culture groups are differentiated in space and time, then so should pottery types be. If the observations of empirical phenomena conform to the implications, then the hypotheses appear to be supported.
Thus, if pottery types are found to cluster in space and time, they are interpreted as evidence of discriminate culture groups. The problem, as Jones et al. (1995:15) point out, is that "although the observations may be sufficient to support the hypothesis, they may not be a necessary consequent of only that causal process specified in the hypothesis" (emphasis added). In short, several explanations may account for the same observation. The problem of equifinality is similarly encountered in evolutionary narratives. A solution to this problem may be sought by employing multiple working hypotheses or differential scales of measure (Dunnell 1995b) or through actualistic studies (Binford 1989, 1992), both of which seek to evaluate the accuracy of alternative explanations.
The potential deficiency of relying exclusively on a typological approach for exploring ceramic boundaries is that important variation may be overlooked. The definition of a pottery type comprises a temporally variable cluster of attributes including temper, surface treatment, decoration, vessel form, and any attributes reflecting construction and firing methods. Each sort of attribute reflects a different step in the chain of operations performed in making pottery, and each may vary independently with respect to space and time. The imperative for all samples to be sorted into mutually exclusive classes composed of attribute clusters may draw attention away from variation existing within classes that may be important for identifying boundaries or relationships at different scales and for explaining or accounting for the mechanisms that may have caused the observed variation. A partial solution, therefore, is to report ceramic data both at the class and at the attribute level.
If a recurring attribute pattern is found to be chronologically or spatially circumscribed, there is little reason to doubt that a typological classification describing this pattern represents past technological traditions that have a cultural basis. The identification of the temporal and spatial dimensions of such traditions is, in fact, a primary objective of this study. The utility of characterizing assemblages or collections in terms of such types is obvious-it simplifies the denotation of attribute combinations so that their temporal and spatial variation may be more easily observed and it provides a means of exploring the way material culture was patterned by those who used it. This study, therefore, employs both attributes and types, the first designed to improve our understanding of the variation in specific technological practices independent of types, and the second designed to reduce the complexity of pottery-making practices into more easily observed units that are hypothesized to correspond to identifiable patterns of cultural behavior and social structure. Ultimately, the goal is to explain these behaviors and structures.
(Continues...)
Excerpted from Woodland Potters and Archaeological Ceramics of the North Carolina Coast by Joseph M. Herbert Copyright © 2010 by Joseph M. Herbert. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.