BACKGROUND TO THE INVESTIGATION


The presence of a prehistoric Indian site on the south bank of Lick Creek has been known for over one hundred years. The archaeological site in "Cox Woods" was first described by the State Geologist, E. T. Cox, in his report of Orange County for the year 1875 (Elrod and McIntire 1876: 238-239). Cox's brief account described the presence of a double-walled earthen enclosure that measured about twelve hundred feet in circumference. Associated with this enclosure were numerous, small "kitchen" mounds that were believed to represent the foundations for the "wigwams" of the inhabitants of the site.

A more detailed description of the site was provided by Goodspeed in an 1884 history of Lawrence, Orange, and Washington counties (Goodspeed 1884). This account described the original walls as measuring three feet in height and extending 1,300 feet in a "U-shaped" arc that opened on Lick Creek (Goodspeed 1884:374-375). Excavations of the inner wall uncovered abundant pottery, charcoal, and ash which were attributed to the aboriginal inhabitants of the site. The embankment itself was reported to contain a layer of "... large flat stones contiguously arranged (Goodspeed 1184:374)." No sign of the "kitchen" mounds reported by Cox were detected during Goodspeed's investigation.

A review of archival documentation and artifact collections curated at the Glenn A. Black Laboratory of Archaeology indicated that amateur survey and excavation activities were carried out periodically at the Cox's Woods site during the 1950's. Most of this work was done by Jesuit priests from West Baden College which occupied the former West Baden Springs Hotel in the 1950's and early 1960's. Recent correspondence with one of the participants, James Mohler, S. J., confirmed that non-systematic excavations were carried out in the vicinity of the preserved embankments on the east side of the site. Relatively large assemblages of Oliver Phase pottery, chipped stone tool debris, and animal bone were recovered from midden deposits located between the double walls of the enclosure; however, the exact locations of these excavations are unknown.

Local informants recall periodic digging of the embankment walls by residents of the county at least as far back as the 1930's. Very few details of these investigations are available, but the finding of "only arrowheads" and some pottery was a common occurrence. The resulting collections have been dispersed and most of the original collectors are deceased. One elderly gentleman did report that, during the early years of cultivation, a linear embankment was visible along the bank of Lick Creek but was rapidly plowed away through the deliberate efforts of the farmer.

The first professional investigation of the site on record was a brief reconnaissance survey in 1975 by David Sonner, then of the Glenn A. Black Laboratory of Archaeology. Sonner reported the existence of remnant sections of the enclosure walls, but no cultural material was observed. In the project report (Sonner 1975), recommendations were made for intensive survey of the preserved earthworks and systematic testing of the site area to locate intact cultural deposits.

The Cox's Woods site was discussed in greater detail in the 1989 Cultural Resource Management Overview for the Hoosier National Forest (Sieber et al. 1989). In this document, an unspecified number of "investigations" by unnamed personnel of the Glenn A. Black Laboratory were noted to have taken place in the 1980's; however, no formal archaeological activities were carried out. The authors of the overview did report the existence of earthen walls and signs of illegal excavations in the form of numerous pot-holes. The authors of the "Overview" concluded that the Cox's Woods site was, potentially, one of the most significant sites in the Hoosier National Forest and recommended that a formal evaluation involving archaeological investigations be carried out.

Under the direction of the senior author of this report, preliminary archaeological excavations at the Cox's Woods site took place, intermittently, between March 11 and April 4, 1993 (Redmond and McCullough 1993). The main objectives of this excavation were (1) to make a preliminary assessment of the archaeological deposits reported in the accounts of Cox (Elrod and McIntire 1876) and Goodspeed (1884), and (2) to evaluate past impacts from illegal digging or other human disturbances. An important focus of this project was the delineation of the village limits and the recovery of culturally diagnostic artifacts from intact deposits in the earthen embankments and elsewhere on the site.

The delineation of the village area was initiated with the excavation of forty 50.0 cm by 50.0 cm test units that were located along two north-south and three east-west transects (Figure 2). (An additional ten units of this kind were excavated in the summer of 1993.) These transects were located to the west of the earthen walls in an attempt to identify the southern, western, and northern limits of the settlement. Each test unit was excavated either to culturally-sterile subsoil or to a maximum depth of 50.0 cm below the surface. Information from Forest Service archives, local informants, and a 1938 aerial photograph demonstrated that this area of the site (i.e., west of the earthworks) had been cleared and plowed in the past (Redmond and McCullough 1993). The use of small test units was therefore seen as a way of systematically recovering artifactual material from the plowzone deposits of the site. All soils from each excavation unit were passed through one- quarter inch mesh screen, and the materials that were recovered were sorted into four artifact classes: chert, limestone, fire-cracked rock, and pottery. These materials were counted and recorded in the field, and the resulting tabulations were used to construct maps showing the spatial distributions of subsurface cultural material across the site (Redmond and McCullough 1993: Figures 41-46) (Figure 3).

The distribution of each artifact class was examined and compared in order to identify the limits of the site and to reveal traces of the former western and southern sides of the embankments. Wall profile maps showing the depth and nature of plowzone soils were drawn during the excavation of each test unit and examined in order to evaluate the spatial extent and degree of cultivation. A total of 12.5 square meters of surface area was excavated using this method.

In addition to the use of test units, the spring 1993 project excavated a 1.0 meter (m) by 9.0 m test trench (designated as "Trench 1") across one relatively undisturbed (although slightly eroded) section of the earthen walls. This trench was excavated in 1.0 m by 1.0 m subunits, each of which was taken down in 10.0 centimeter (cm) thick levels. All soil was passed through one-quarter inch mesh and samples of soil (for flotation processing) and charcoal were collected. All possible features were mapped in plan, profiled, and photographed, and representative wall profiles were documented in a similar fashion.

In summary, this investigation verified the prehistoric nature of the earthen walls through the discovery of diagnostic Oliver Phase materials and other cultural debris within the embankments. Furthermore, wood charcoal recovered from sealed contexts produced two (uncalibrated) dates of AD 1380 +/-70, and AD 1300 +/-110 (Redmond and McCullough 1993:103). The discovery of post molds within the inner embankment determined that this earthwork functioned to support a defensive stockade. Finally, the excavation of small test units across the site identified the southern and western edges of the settlement as an annular pattern of cultural debris (i.e., midden deposits and stockade fill) (Figure 3).
The center ofthe site contained little cultural material and was identified as a central plaza or public area.

[return to part 1 menu] [return to table of contents] [continueto research design]