Page: List of Figures: 1051–1120, Page Number: xxxi
Figure 1051. Structure 9, plan view (view to north).
Figure 1052. Top of Burial 2 before excavation (view to northwest).
Figure 1053. Top of Burial 3 before excavation (view to northwest).
Page: List of Tables, Page Number: xxxii
List of Tables
Table 1. Summary of pottery recovered from the Fredricks site.
Table 2. Distribution of pottery from features, burials, and structures.
Table 3. Formal attributes for the whole vessels and reconstructed vessel sections from the Fredricks site.
Table 4. Metric attributes and contexts for whole vessels and reconstructed vessel sections from the Fredricks site.
Table 5. Distribution of pottery by feature and burial clusters.
Table 6. Analyzed stone tools from the Fredricks site.
Table 7. Historic artifacts found at the Fredricks site.
Table 8. Summary of glass trade beads from the Fredricks site.
Table 9. Animal remains from the Wall site (1983 and 1984 excavations).
Table 10. Animal remains from the Fredricks site (1983 and 1984 excavations).
Table 11. Expected and actual representation of deer skeletal elements (1983 and 1984 excavations).
Table 12. Estimated meat yield in pounds (1983 and 1984 excavations).
Table 13. Animal remains from the Fredricks site (1985 excavations).
Table 14. Estimated meat yield in pounds (1985 excavations).
Table 15. Animal remains from the Fredricks site (1986 excavations).
Table 16. Age of deer from the Fredricks site.
Table 17. Estimated meat yield in pounds (1986 excavations).
Table 18. Summary of plant remains from the Wall, Fredricks, and Mitchum sites (weights in grams).
Table 19. Percent of plant food remains from the Wall, Fredricks, and Mitchum sites.
Table 20. Ubiquity of plant remains, as percent of flotation samples at the Wall, Fredricks, and Mitchum sites.
Table 21. Seed counts from the Fredricks site.
Table 22. Seed counts from the Wall and Mitchum sites.
Table 23. Percent of nutshell from the Wall, Mitchum, and Fredricks sites.
Table 24. Comparison of plant remains from burials, features, and structures at the Fredricks site.
Table 25. Distribution of seeds from the Wall, Mitchum, and Fredricks sites (number per gram of plant food remains).
Table 26. Plant remains from 1985 flotation samples (weights in grams).
Table 27. Plant food remains from 1985 flotation samples (weights in grams).
Table 28. Percentage of plant food remains from 1985 flotation samples.
Table 29. Seed/fruit counts for burials, features, and structure from 1985 flotation samples.
Table 30. Plant ubiquity at the Fredricks site as percentage of features (1985 samples).
Table 31. Percentages of nutshell from 1983, 1984, and 1985 flotation samples.
Table 32. Percentages of plant food remains from 1983, 1984, and 1985 flotation samples.
Table 33. Seed/fruit counts and proportions from 1985 flotation samples.
Table 34. Seed/fruit counts and proportions from 1983, 1984, and 1985 flotation samples.
Table 35. Plant remains from Features 35 and 37 (1985 flotation samples; weights in grams).
Table 36. Plant remains from Feature 36 (1985 flotation sample; weights in grams).
Table 37. Plant remains from Feature 30 (1985 flotation sample).
Table 38. Plant remains from 1986 flotation samples (weights in grams).
Table 39. Weights (in grams) of plant food remains from 1986 flotation samples.
Table 40. Percentage of plant food remains from 1986 flotation samples.
Table 41. Densities of plant remains in features (1986 flotation samples).
Table 42. Seed and fruit counts from 1986 flotation samples.
Table 43. Absolute and relative quantities of nutshell.
Table 44. Ubiquity of plant taxa from the Fredricks site as percentage of features.
Table 45. Percentage of plant food remains from the Fredricks site (1983–1986 flotation samples).
Table 46. Percentage of seeds and fruits from the Fredricks site, 1983–1986.
Table 47. Acculturation indices for Fredricks site burial groupings.
Table 48. Distribution by age category of utilitarian (U) and ornamental (O) artifacts associated with the Fredricks site burials.
Table 49. Wall post densities at the Wall site.
Table 50. Wall post densities at Upper Saratown.
Table 51. Inventory of European trade artifacts from feature and burial contexts at Upper Saratown and the Fredricks site.
Page: Archaeological Background: Historical Background, Page Number: 1
When European explorers first entered the Virginia and Carolina Piedmont, they found it occupied by several small Indian tribes who shared a common culture and a similar language. These Siouan tribes also shared a mixed subsistence of hunting, gathering, and agriculture, and a social system regulated by ties of kinship and reciprocity (see map).
As the colonial frontier was pushed into the Piedmont and as Indian and European interaction was intensified, the Occaneechi tribe became prominent among the Siouan groups. The Occaneechi controlled much of the deerskin trade, and their language became the lingua franca of the Piedmont. Their pivotal role in the fur trade came about partly because one of their villages, on an island in the Roanoke River, was astride the Great Trading Path from Virginia to Georgia.
The island village of the Occaneechi was visited by John Lederer in 1670 (Cumming 1958). After the Occaneechis "barbarously murthered" six Cherokees who were attempting to establish trade relations with the Virginia colonists, Lederer, fearing for his life, cut short his visit. James Needham and Gabriel Arthur, who traveled through the same territory in 1673, observed that the Occaneechis controlled the colonial trade, which endowed them with an importance that far exceeded their numbers (Alvord and Bidgood 1912). They seem to have maintained and reinforced their role in the trade network through warfare and intimidation. Thus, the Occaneechi tribe earned a fierce and pugnacious reputation, which eventually led to an eruption of armed hostilities with Nathaniel Bacon's militia in 1676.
After pursuing a group of Susquehannock Indians into Occaneechi territory, Bacon convinced some "Manakins" and "Annalectins," who had also joined the Occaneechi, to aid his forces in defeating the Susquehannocks. After that victory was accomplished, Bacon then attacked the Occaneechis (Billings 1975:267–269).
After the battle with Bacon, the Occaneechis were so reduced in numbers that they could no longer defend their island stronghold on the Roanoke (see map). The survivors abandoned their home territory, retreated southward, and reestablished a village on the Eno River, near present Hillsborough, North Carolina (see map). In 1701, English surveyor John Lawson visited the relocated Occaneechi Town where he observed that there were "no Indians having greater Plenty of Provisions than these" (Lefler 1967:61) (see Lawson's journal: title page, text).
After Lawson's visit, conditions worsened for the Occaneechi, as well as for the other Siouan tribes, and by 1722, disease, warfare, and rum had virtually destroyed Indian societies in the Piedmont. Remnants of once autonomous groups either huddled together around Fort Christanna in Virginia or moved to join their cousins, the Catawba, in South Carolina. By 1730, except for a few isolated Indian families, the North Carolina Piedmont lay mostly vacant, awaiting the arrival of hordes of colonists from Virginia, Maryland, and Pennsylvania.
Page: Archaeological Background: Siouan Archaeology, Page Number: 2
Archaeologists first became interested in studying the remains of the Piedmont Siouans in the 1930s, when village sites thought to be associated with the Keyauwee, Sara, Saponi, and Occaneechi were subjected to excavations of varying intensity (Coe 1937; Lewis 1951). Though broad in scope, these early efforts were not focused by a structured research design. At most sites, only small areas were tested, and collections were gathered primarily with an eye toward identifying pottery types of the different tribes. As part of this early research, extensive excavations were carried out between 1938 and 1941 at the Wall site on the Eno River near Hillsborough (see Hillsborough locality and archaeological district maps). This site was thought to represent the Occaneechi village visited by Lawson in 1701.
The next archaeological research in the Siouan area was undertaken in the 1940s on the Roanoke River, prior to the inundation of Kerr Reservoir in North Carolina and Virginia (Miller 1962). Under the auspices of the Smithsonian Institution's River Basin Salvage Program, extensive excavations were conducted in the reservoir area at the Clarksville site on the east bank of the river opposite "Occaneechi Island," and on the island itself at the Tollifero site. These two sites contained information on the prehistoric Siouan inhabitants of the area, but no evidence was found of the 1670 Occaneechi village visited by Lederer, Needham and Arthur, and Bacon.
In 1972, the Research Laboratories of Anthropology began excavations at the Upper Saratown site on the Dan River in Stokes County, North Carolina (see regional map). These investigations, which lasted for 10 consecutive field seasons, exposed a group of circular houses with associated storage pits and burials, as well as a sequence of village palisades (Ward 1980; J. Wilson 1983). Most of the burials were accompanied by nonutilitarian (ornamental) European trade items. Ethnohistoric records and the recovered trade artifacts suggested that this site was occupied during the late 1600s by the Sara, one of the Piedmont Siouan tribes and neighbors of the Occaneechi.
When combined, these initial efforts to investigate the "Siouan problem" seem substantial. Each project, however, was developed as an end in itself and was not guided by an overall set of research objectives. Consequently, archaeological coverage of the Siouan area was uneven. For example, the upper Dan River valley was extensively investigated, whereas the Haw and Eno drainages to the southeast received relatively little attention. Surveys were opportunistic rather than systematic, and a few larger sites were tested and excavated at the virtual exclusion of many small ones. Despite all of their shortcomings, these previous investigations provided a foundation for more systematic studies of Piedmont Siouan culture.
Although the need to approach Siouan archaeology with a set of specific goals, operationalized by an overall research strategy, was obvious, such a course of study was not formulated until 1983. At that time, staff of the Research Laboratories of Anthropology developed a research design which included a set of questions focused on Siouan culture change and the archaeological correlates of that change.
The archaeological investigations that followed became known as the Siouan Project and focused on the Dan, Eno, and Haw River drainages, heartland of the Piedmont Siouans during the Historic period. Extant ethnohistoric and archaeological information suggested that there was considerable cultural diversity among the groups in these three river systems, reflecting possible differences in ethnicity, microenvironmental adaptation, and intensity of interaction with the English. Although the Siouan tribes seem to have commonly shifted their villages and to have even changed their territories, by 1675 the locations of their settlements were more or less stabilized within the confines of these three drainages. The Sara, Tutelo, and Saponi occupied the territory drained by the Dan and its tributaries; the Eno basin was the homeland of the Eno, Shakori, and Occaneechi (after 1680); and the Haw River area was occupied by the Sissipahaw and possibly others (see regional map).
Since the Siouan Project was concerned with studying changes in aboriginal culture brought about by contact and interaction with English colonists, a primary goal was to locate and identify towns occupied by the various Indian tribes at specific time intervals from the Late Prehistoric through Contact periods. These intervals are: Late Prehistoric (A.D. 1300–1525), Protohistoric (A.D. 1526–1625), Early Contact (A.D. 1626–1675), Middle Contact (A.D. 1676–1710), Late Contact (A.D. 1711–1740), and Euroamerican (A.D. 1741–present). Once sites representing all (or most) intervals were located in each drainage area, it was possible to address more specific questions concerning how the different Piedmont groups adapted within local environments to increasing exposure to European materials, ideas, and institutions.
Some initial questions to be addressed were: What were the Siouan cultures like prior to European contact? After initial European contact, what aspects of culture changed first, and with what relative intensity? As contact became protracted, did the Indians move more toward the adoption of European ways, or more toward making adjustments in their existing cultural patterns to cope with the European presence? What were the short-term and long-term effects of European epidemic diseases? What effects did the deerskin trade have on the native economy, technology, and social organization? How did man-land interactions change through time?
Page: Archaeological Background: Fredricks Site Discovery, Page Number: 3
Fredricks Site Discovery
Although the ethnohistoric record contains little precise information on the locations of Siouan towns, a description in John Lawson's journal (Lefler 1967) and the survival of "Occaneechi" as a place name provide a strong case for locating the 1701 town of Occaneechi immediately southeast of present-day Hillsborough in Orange County, North Carolina (Rights 1957; Lefler 1967). This case is strengthened further by Edward Moseley's 1733 map of North Carolina which places "Acconeechy" just east of the Great Trading Path on the north side of the Eno River. A reconstruction of this path's route through piedmont North Carolina has shown that it crossed the Eno River at Hillsborough (Rights 1931). Finally, the Eno River floodplain just east of Hillsborough, formed by a bend in the river's course, comprises the only large (ca. 25 acres) expanse of land in the vicinity that would have been suitable for native agriculture (see map of Hillsborough archaeological district).
When the Wall site was first investigated archaeologically during the late 1930s and early 1940s, it was reasonably thought to be the remains of Lawson's Occaneechi Town because of its location and presumed late date of occupation (Coe 1952). This interpretation was generally accepted until the inception of the Siouan Project in 1983. At that time, a cursory re-examination of the 1938–1941 excavation data immediately called to question this interpretation. It was felt that, given the Occaneechis' active participation in the deerskin and fur trade with the Virginia colony and the comparatively late date for Occaneechi Town, the archaeological remains of that community should contain numerous European-made artifacts. However, most of the European and Euroamerican artifacts found at the Wall site either dated too late or had been found in disturbed contexts. Because of this, it was felt that additional field investigations were needed to clarify the chronological placement of the Wall site and to critically evaluate its identification as Occaneechi Town.
A Second Look at the Wall Site
In the summer of 1983, excavations were resumed at the Wall site. Initially, sections of the old 1938–1941 excavations were isolated and a site grid was re-established. A portion of the rich midden surrounding the village was excavated and subjected to fine-scale recovery techniques. In addition, three burials were excavated and portions of two circular house patterns were exposed. The few European and Euroamerican artifacts found were from plow-disturbed soil, and most of these dated to the latter half of the eighteenth century or early nineteenth century, well after the time when Occaneechi Town was occupied. Just as important, no European artifacts were found in undisturbed contexts despite careful waterscreening through 1/16-inch mesh. Radiocarbon dating provided additional evidence that the Wall site village predated the period of European contact. Three radiocarbon samples from undisturbed contexts yielded an average corrected date of A.D. 1545±80 years [not calibrated]. These data, in conjunction with a review of the earlier investigations, led to the irrefutable conclusion that the Wall site was too old to be historic Occaneechi Town.
Discovering the Fredricks Site
During initial surveys to relocate the Wall site in the spring of 1983, two other village sites were discovered nearby. At one of these, the Fredricks site, several European artifacts, along with Indian-made pottery fragments, were found on the surface. Once it appeared that the Wall site was not Occaneechi Town, UNC archaeologists dug several shovel-sized test pits on this newly discovered site in the hope of finding evidence for intact archaeological deposits. One of these small test pits exposed the top edge of a pit filled with refuse and dark soil. This discovery was followed by the excavation of eight contiguous 10-ft x 10-ft squares. These revealed five sharp-cornered rectangular pits (see excavated square) and a line of small postholes (see excavated square). Both the posthole pattern and the pits were neatly arranged in a northwest-southeast direction. (See video describing the discovery of Occaneechi Town.)
Four of the pits were excavated, and three contained human skeletal remains accompanied by grave goods of both European and Indian manufacture. The fourth pit, identical in shape to the other three, appeared to represent a burial but contained no bones or grave goods. Two of the burials were the remains of children between three and eight years old at death. Included with these burials were European trade items such as knives, scissors, and a variety of glass beads. Aboriginal artifacts included shell gorgets, shell beads, and a ceramic vessel. One adult male burial contained a wealth of European artifacts, including an intact rum bottle, scissors, knives, a pewter pipe and buttons, a pair of ember tongs, and an iron axe head.
Most of the trade artifacts from these burial pits dated to the late 1600s or very early 1700s, the appropriate period for Occaneechi Town. Archaeological finds from subsequent investigations at the site confirmed this initial chronological assessment. The site appeared to be well preserved, with no evidence of disturbance other than shallow plowing. From these observations, it became obvious that more intensive work was needed at the Fredricks site. Hence, plans were immediately begun for a project in 1984 that would combine a major excavation and testing program at the Fredricks site, along with continued work at the neighboring Wall site.
Page: Archaeological Background: Fredricks Site Excavation, Page Number: 4
Fredricks Site Excavation
Because the Fredricks site was discovered late in the 1983 field season, investigations that summer were relatively brief. A limited excavation of 800 sq ft revealed a portion of a cemetery lying just outside the village (see photo) and a segment of the village palisade (see photo). Three human burials within the cemetery were excavated. All three pits were rectangular with sharp corners (indicating that they probably were excavated with metal tools) and contained numerous artifacts of Euroamerican manufacture. A fourth pit excavated within the cemetery contained neither human remains nor grave associations (see photos of sifting plowed soil, burial excavation).
A second field season at the Fredricks site, conducted during the summer of 1984 and sponsored by the National Geographic Society, uncovered a much larger area of the cemetery and the adjacent village (Dickens et al. 1984, [ed.] 1987) (see aerial view of Wall and Fredricks site excavations; see aerial view of Fredricks site excavation). These investigations were designed to obtain additional data on mortuary behavior and to begin sampling domestic areas. In addition, systematic subsurface testing was undertaken on unexcavated portions of the site to delimit probable site boundaries and to make a preliminary assessment of internal site structure.
During 1984, 27 new 10-ft by 10-ft units (2,700 ft2) were excavated, and six of the eight units excavated in 1983 were re-exposed (see general view of excavation). These excavations uncovered six additional burials within the cemetery, a 90-ft palisade segment, and approximately 2,250 ft2 of the village area inside the palisade (see cemetery, burial excavation). Mapping of postholes revealed two complete domestic structures. In addition, an oval, wall-trench sweat lodge with an interior fire pit was exposed in the southwesternmost corner of the excavation (see sweat lodge). Subsurface testing of unexcavated areas consisted of auger sampling at 2.5-ft intervals to identify archaeological features. This procedure proved to be highly reliable and was successful both in delimiting the remainder of the cemetery and in identifying areas of intensive domestic activity within the village. It was somewhat less effective, however, in providing a precise definition of site boundaries (see Davis and Ward 1987) (see 1984 crew).
In 1985, a third season of fieldwork was made possible by additional funding from the National Geographic Society (Dickens et al. 1985; Dickens et al. [ed.] 1986). These excavations exposed 62 10-ft by 10-ft units, more than doubling the total area uncovered during the previous two field seasons (see general view of excavation). The large excavated area made it possible to estimate the overall size of the village as well as to predict its internal spatial configuration. Twenty-five features and three burials were excavated (see feature excavation). The burials were the last remaining in the cemetery, bringing the total to 12 with an additional probable burial. Six new structures also were defined as a result of the 1985 work, and approximately 100 ft of the palisade were exposed as it continued to encircle the habitation area (see cleaning house patterns). At the end of the 1985 field season, it was estimated that the village compound within the palisade was small, comprising only about .25 acres. A total of 11–12 houses were estimated to have sheltered approximately 50–75 individuals (see 1985 crew).
The 1986 excavation at the Fredricks site uncovered all the remaining village area enclosed within the palisade except for a small section in the southwest corner where large trees prevented soil removal (see general view of excavation). Although still small, the village shape was more irregular than predicted after the 1985 field season. The irregular outline resulted from the fact that the palisade "bowed-out" or expanded to the southwest, thus creating a D-shaped rather than oval configuration. Although much of the structural evidence continued to consist of vague posthole clusters, two additional wall-trench structures were defined and the 13 pit features associated with the structures were extremely rich. Their depositional character and contents added significant new data that aided in clarifying general as well as specific behavioral patterns within the village. In addition, two human burials, two possible burial pits, a possible hearth, and an irregular trench were excavated. Another probable pit (Feature 60) and a shallow basin (Feature 52) were mapped but not excavated (see 1986 crew).
In 1995, while excavating the nearby Jenrette site, Feature 60 and the three final 10-ft by 10-ft units at the Fredricks site were excavated.
For a map showing the area excavated during each field season, click here.
Archaeological field methods were consistent throughout all field seasons at the Fredricks site. Each season, preliminary site preparation consisted of bushhogging the work area and re-establishing the site grid and reference point for elevations. All plowzone (0.5–1.6 ft thick) was excavated in 10-ft by 10-ft units, with soil being dry screened through 1/2-inch wire mesh using hand sifters. A 20-liter soil sample from the plowzone of each unit was waterscreened through 1/16-inch mesh to assess small artifact content.
Following the removal of plowzone, the bottom of each excavation unit (top of subsoil) was carefully trowelled in order to identify and record pits and postholes. The trowelled surface was documented by black-and-white and color photographs and was mapped at a scale of 1 in=2 ft. The drawings of each excavation unit were subsequently combined to produce an overall plot of the excavation. Photographs were also made of all procedures and of the general progress of work. Horizontal and vertical control was maintained through reference to the site grid, using a transit and leveling rod to determine elevations.
Excavation of features and burials was accomplished using trowels, grapefruit knives, brushes, and other small tools. Sunscreens, constructed of wooden frames and bedsheets, were erected over features during excavation to minimize the damage to feature contents by the summer sun. Feature fill was removed in natural zones, when evident, and all fill was waterscreened through sluice boxes having a sequence of 1/2-inch, 1/4-inch, and 1/16-inch screens. This technique permitted the recovery of minute artifacts, including shell and glass beads, lead shot, small animal bones, and carbonized plant remains. Standard 10-liter soil samples from each zone of each feature were simultaneously processed by flotation to retrieve very small, extremely fragile carbonized seeds and plant parts that might otherwise be lost in the waterscreening. Elevations were taken following the removal of each soil zone of a feature in order to establish precise provenience for zone contents and to permit the calculation of soil volume.
After completion of excavation, all features and burials were extensively documented by black-and-white and color photography, and by drawings in profile and plan at a scale of 1 in=1 ft. Also, extensive notes were kept by all excavators in both field journals and on standardized feature and burial data forms.
Following completion of each field season, the excavation was immediately backfilled using a front-end loader.