|
THE ARCHAEOLOGICAL SITES MONITORING PROGRAM AT THE BIG SOUTH FORK NATIONAL RIVER AND RECREATION AREA, 1986 TO 1989
Tom Des Jean
Introduction
The Big South Fork of the Cumberland River flows northward along the Upper Cumberland Plateau. This formation is a part of the Appalachian Plateau Physiographic Province and represents a true peneplain or "a broad area of slight relief shaped by erosional processes". The swiftly moving waters of the river have scoured deep gorges into the usually resistant 250 million year old Pennsylvanian sandstones of the Plateau formation. This cutting action has produced literally thousands of rockshelters and "bench" or terrace areas. These locations have been occupied by prehistoric and historic peoples living in this area for the past 10,000 to 12,000 years. The Big South Fork National River and Recreation Area (BISO) was created by Congress (Public Law 93-251) in 1974 "...for the purposes of conserving and interpreting an area containing unique cultural, historic, geologic, fish and wildlife, archaeologic, scenic, and recreational values...for the benefit and enjoyment of present and future generations". At present there are 848 known archaeological sites recorded within the BISO boundary (Figure 1) many of which have been looted for the artifacts they contain. Only ten percent of this National Area has been surveyed but if the remaining lands possess sites at the same density as the areas already surveyed then the National Park Service (NPS) is congressionally mandated to protect a huge archaeological resource that may well exceed 8,000 sites. An Archaeological Sites monitoring program was one step toward developing site preservation at the Big South Fork National River and Recreational Area (Faust 1986). The program was funded by the U.S. Army Corps of Engineers (COE) which presented a very generalized scope of work to assess and manage the effects that trail development and trail use in the National Area had and continues to have on the cultural resources at BISO. The program was then developed and conducted by the Southeast Archaeological Center (SEAC) of the NPS from December 1986 through March 1989. This two year program was divided into three developmental phases Phase I (six months), Phase II (six months), and Phase III (one year). The phases are discussed further in the research design section. A research design was developed for the monitoring program which outlined archaeological site impact data collection. The research design also proposed monitoring various trails to define and isolate different types of impacts. Baseline data collection for this program included monitoring control areas in order to gauge variations in impacts from area to area. The development of the monitoring program anticipated some logistical adjustments and technical problems that could only be overcome through field trials. Procedural changes were made throughout the first year of monitoring activities in order to streamline and "fine-tune" baseline data collection. Following these modifications the final phase of this study was completed. This article presents a monitoring program "Template" which can be modified for use in other areas.
Research Design
The goal of the Archaeological Sites monitoring program was to understand and define various types of site impacts (natural and human) and to develop a model for cultural resources monitoring in the BISO area of the Upper Cumberland Plateau. To accomplish this the research design followed a chronological set of operational steps and objectives described below.
The three phases of the archaeological sites monitoring program contained specific and overlapping objectives. The three primary objectives were:
1. Develop a Park Archaeological Sites monitoring program by:
2. Evaluate the monitoring program with respect to:
3. Report results and recommendations concerning:
Two secondary objectives were also defined as part of the monitoring program:
1. Increase the knowledge concerning the archaeological resources at BISO through the monitoring program by:
2. Increase the public awareness concerning the park's non-renewable archaeological resources through:
Phase I of the monitoring program addressed primarily, items 1a, 1b, 2a, and 3a. Phases II and III addressed the remaining objectives. The secondary objectives were collateral duties which were done by the monitoring program Archaeologist or BISO Interpretation staff whenever opportunity or occasion permitted.
Data Collection Baseline data collection, as indicated in the above discussion, required a lot of trial and error in the field. The basic procedure was to locate rockshelters in the three zones (hiking trails, horse/jeep trails, and remote areas) described in the research design and to define every rockshelter by length, height, depth, size category, color, aspect, presence/absence of artifacts, presence/absence of digging, photographs, locational description and field sketch, and a locational map. A permanent metal tag was then attached to every rockshelter that was recorded. All of this data was collated onto 5" X 8" field index cards and then entered into a "dbase file". Once an area had been recorded seasonal visits were made to all of the tagged sites in that area and any new disturbance was noted on the monitoring cards. The locational cards and photographs worked well but there were a few sites in steep, jagged and densely overgrown terrain that were not able to be relocated for a year. Other sites located in monitoring areas of rugged terrain were overlooked and only identified long after the program had been in place. Discerning evidence of new looting from old looting activity was easier as monitoring became more frequent and as the person monitoring became more familiar with each site in the database. It also became easier to recognize various clues about site disturbances. It would be a very helpful technique to be able to read computer "bar code dates" which are now found on most product labels discarded at sites during illegal digging. Often the trash left by looters did contain readable product expiration dates which provided a Terminus Ante Quem (time before which) for disturbances. Additionally, sun bleaching of labels and plastic, the growth of mold/lichen on backdirt piles, rounding of the sharp edges of backdirt piles, rainfall and erosion, blown in leaf litter, and the disappearance time for footprints, gave some indication of the time since disturbance. The rate of change for many of these signs differ for every disturbed rockshelter site. An example of this is that shallow footprints or handprints will disappear within a matter of a few hours to a day in rockshelter soils containing high to moderate amounts of Nitrates. Erosional processes also vary considerably from rockshelter to rockshelter and this tends to obscure episodes of "new digging". Many of the disturbed rockshelter sites were so impacted that discovering clues to indicate recent looting activity was very difficult. One idea proposed to address this situation was to "sweep" rockshelter sites so that it would be easier to identify new looting activity. An assistant was hired for three months to accompany the monitoring program researcher and "sweep" these heavily disturbed sites leaving a smooth surface which would reveal any new impacts. This technique worked well and helped to determine 32 new impacts and speed up the monitoring process. Determining types of impacts involved identifying human versus animal disturbances. This was not always easy to do but the presence of animal denning materials, animal feces and other animal signs were adequate to differentiate those sites recently disturbed by animals. Most of the time the presence of human garbage was enough to indicate that the observed disturbance was not of natural origin. Also the size of disturbances varied from minimal amounts in the case of animal activity, to unusually large and deep craters or numerous small impacts that resulted from human activities. Rockshelter locations were initially investigated using a three zone research design. This design proposed monitoring archeological sites within stratified zones. The zones were defined as, Zone I-hiking trails, Zone II-horse/jeep trails, and Zone III-Remote areas. It became clear early in Phase II that no prehistorically occupied rockshelter was immune to impacts. Even the "Remote Zone" sites have been and continue to be looted (Des Jean 1987a:8-9). While evidence was collected indicating that disturbed rockshelter sites along the more heavily used trails suffered a slightly lower rate of looting compared to other zones, no zone or area in BISO has remained free of intensive site looting or site destruction (Figure 2). This fact is not unique to the Big South Fork area. Lack of easy access is not a consideration to dedicated relic collectors and this fact has been noted by other researchers in numerous areas of the country (Lightfoot and Francis 1978:89; Ison et al. 1981:29; Lyneis et al. 1980:152; Williams 1978: Table 33). Verification of this factor at BISO convinced the monitoring program archeologist to modify the program's research design and eliminate the Stratified Zones originally defined (Des Jean 1987b). Rockshelters were then logged-in as either on or adjacent to jeep roads or trails (100m either side), or not on or adjacent to jeep roads or trails. This modification still provided the data which answered questions that the funding agency (COE) wanted addressed (impacts to rockshelters on trails) and it provided the control information (non-trail rockshelter impacts) with which to interpret that data. Data continued to be collected on all of the sites logged-in to the database from the beginning. One other modification to the research design was to eliminate use of BISO staff participation in monitoring activities. This was an optimistic goal which, because of the shortage in staffing, was never really able to be implemented. There were, however, several individuals from all of the BISO divisions who reported new impacts observed. One exception to staff participation in this program was a biology researcher who was able to monitor a prehistorically occupied rockshelter site associated with his year long monthly sampling regimen. After more than two years of data collection the monitoring program has identified a number of patterns in looting behavior not previously recognized.
Results The monitoring program archeologist logged-in 226 rockshelter locations and made an average of 5 visits per site during the two year program. The number of visits ranged from only one visit at some rockshelters to a maximum of thirteen visits to rockshelter #RS22. This latter site, Muleshoe rockshelter, was the one closest to BISO Headquarters and the one used in an interpretive tour to illustrate to the public the effects of looting and vandalism. Out of the 226 rockshelters that have been recorded in the monitoring program database 139 (62%) were identified to be prehistoric archaeological sites. Looting activity is evident at 128 (92%) of these sites. This is similar to the 95% looting impact rate given for the Daniel Boone National Forest (DBNF) which has a common boundary with BISO (Ison 1989). The 11 archeological sites in the monitoring program database which have not been looted are either intact due to roof-fall, have very difficult access and/or low visibility, of small size, have low artifact densities, or all of the foregoing. Since the beginning of the monitoring program 208 incidents of looting have been recorded. Most of these were past episodes but 86 incidents of looting were relatively new and 32 of these were repeat looting incidents occurring between monitoring visits. Eight other rockshelters suffered digging impacts made by animals. One hundred nine rockshelters were monitored that had no artifacts evident. Only 18 of these had been looted and many of these were in the small (<20 linear meters) category. It appears then, that the one critical element that looters use to determine site location is the presence of chert debitage. Usually this material is visible in the dripline of a rockshelter but the 18 "tested" sites indicate that some looters do understand that not all prehistoric rockshelter sites will be immediately evident. The 91 rockshelters that have had no disturbances will be dropped from future monitoring activities. Also archaeological sites which have had no new impacts for a year or more can be monitored on a yearly basis following the season of greatest activity, which in BISO is during the Fall. It was initially assumed that more archaeological site looting would be evident along the federally constructed trails where heavy public use occurs. It was also assumed that remote areas where few roads or trails are located would have the least amount of looting. Figure 3 demonstrates that these ideas are not correct. While the hiking trail areas, Zone I in the original monitoring program research design, have suffered a tremendous amount of site looting more impacted sites were found along Zone II, the horse trails and jeep roads. This fact was also found to be true for jeep trails by Lightfoot and Francis (1978:89) on US Forest Service lands in Colorado. A direct relationship has been reported to exist at BISO and elsewhere between rockshelter site looting and linear distance to jeep or secondary roads (Ferguson et al. 1986:232; Wylie 1989) but monitoring program data do not appear to reflect this very strongly. This may be due to the fact that there are a number of unrecorded roads in the area that serve to provide looter access. The same thing is evident in the National Forest (DBNF) according to Ison et al. (1981:29), who found that, "...accessibility is no problem because of numerous unrecorded roads and undeveloped trails". Another reason that this relationship appears to be weak is because of the intense site looting that occurred during TVA trail construction at BISO where unsupervised workers, some of them avid relic collectors, worked for five years from 1981-1985 (Personal Informant 1987). Remote areas of BISO, Zone III, evidenced more looting impacts numerically but in the Zone II areas no archaeological rockshelter sites were found that were not looted. Both Zones I & III, at least, possessed some non-looted prehistoric rockshelter sites. New disturbances continue to occur along hiking trails but to a lesser degree than in other areas of BISO. The explanation for slightly lower site looting along frequently used hiking trails may be found in a looters reluctance to dig in areas where discovery would be more likely. Regardless of accessibility or location rockshelter sites suffer from repeated looting. The one evident fact is that "Proximity to roads will have a powerful influence on the degree of disturbance which can be expected" (Ferguson et al. 1986:239; bold added). This has been observed to be the case during the monitoring program: remote location offers little protection from repeated site looting. Another statistic that was thought to have some significance in documenting looting behavior is that of rockshelter size. It was thought at the outset of this study that the larger rockshelters would indicate selectively greater evidence of impacts. This, however, was not the case; a fact also documented by the University of Tennessee Survey (Ferguson, 1986:234). Rockshelters were divided by opening length into small (1-20m), medium (>20m-50m), and large (>50m) sizes. The medium sized rockshelter sites exhibit about twice as much impact as both large and small rockshelter sites (Figure 4). This may be due to a selective bias of looters for rockshelters larger than 20m in length and because there are fewer prehistorically occupied small rockshelters. It may also be a result of fewer prehistorically occupied large (>50m) rockshelters that are able to be dug into (Figure 5). Many of the large category rockshelter sites have a great deal of roof-fall which prevents them from being looted. Ison, Norville, and Pollack (1981:28, 30) noted this fact too, in a study done in the Red River Gorge area of the DBNF, where it was reported that while roof-fall does protect some sites looters will attempt to break up roof-fall boulders to dig under them for artifacts. One collector interviewed by the monitoring program archeologist stated that he takes a block and tackle back to particularly good sites in order to remove roof-fall. Earlier findings of this study indicated that the hunting season (Fall) accounted for most of the looting of archaeological resources in the National Area (Des Jean 1987b:10). This situation has not changed but new evidence indicates that looting activity is now occurring in the summer at higher levels than anticipated (Figure 6). This summertime looting activity may be the result of a continually rising rate of visitation (Figure 7). The Red River Gorge study at DBNF (ISON et al. 1981:24) found that as visitation increased, so too did looting impacts, at the rate of 10% for both developed and remote sites. The increase observed in summertime looting may also be due in part to the fact that the monitoring program database was not adequately set up during the previous year (1987) to predict seasonal
trends by 1988. Impacts that occur during hunting season, based on current data, still account for the majority (53%) of known sites looted in BISO (Des Jean and Wilson 1989:5).
Assessing the Monitoring Program The three primary objectives of the monitoring program were accomplished or modified as the program developed. Most of these procedures worked well but a few problems that were encountered and a few successes that were noteworthy will be discussed. Objectives (1c, 2a-2b) concerning staff participation in monitoring activities were unworkable because of limited personnel and Big South Forks's large and dispersed cultural resource base. While no real systematic monitoring by BISO staff was possible they were enthusiastic and did report 13 instances of recent looting activity. Although not totally effective at BISO, staff monitoring may work well in other areas where staffing is larger and the cultural resources are better defined and concentrated. The concept of horizontally stratified zones of archaeological site impacts (2c) used in guiding the logistics and planning of the monitoring program was not applicable here and was discarded. Objective 2d, determining the frequency of monitoring needed with respect to site types, impacts, and zones, was generally not difficult to do but in a few cases did prove impossible. Various reasons account for this: some sites are assumed by looters to be "played out" while other sites continue to be dug. In a few cases, sites are repeatedly and predictably looted/vandalized allowing enforcement strategies to come into play with a greater prospect of success. The one useful variable for archaeological site protection strategies was the determination that site looting was a seasonal activity geared to greater visibility as a result of leaf fall and vegetation dieback in the Fall. The Fall season also coincides with snake hibernation and the hunting season. The latter providing a time to be in the woods and the opportunity to search in many areas without worrying about snakes, which are numerous on the Upper Cumberland Plateau. Secondary objectives included increasing the knowledge of cultural resources in the Big South Fork area and increasing public awareness. Secondary objective 1 called for surveying new areas of the park and surface collection of sites to help define a cultural chronology for this area of the Upper Cumberland Plateau. As a result the monitoring program identified approximately 150 new archeological sites and collections made at these sites has helped to establish a more reliable cultural chronology. The other Secondary objective was for the monitoring program archaeologist to provide education programs at local schools, to help develop interpretation programs at BISO, and to help law enforcement personnel in archaeological site protection efforts. A number of programs were developed to educate the public using school "chalk & talk" programs, a cultural hiking trail, an area "Dig Box" activity, and posters and activity hand outs for children. Primary objective 3d and secondary objective 2c concern recommendations for eliminating effects of adverse impacts. Suggestions were made to reduce looting impacts through law enforcement measures using up-to-date data supplied by the on-going monitoring program. This was done with the result that four suspects were apprehended in the act of looting an archaeological site. These suspects later pleaded guilty to a misdemeanor under the Archaeological Resources Protection Act. This conviction was one of the first made in the Eastern United States for the looting of a prehistoric site.
Conclusion The Applicability of various protection strategies has been discussed extensively at BISO and has resulted in an Archeological Resources Protection Plan (Des Jean 1988b). The plan outlined a number of strategies which rely on information developed from an ongoing archeological sites monitoring program. These suggestions include: law enforcement methods involving remote sensing, increased backcountry patrolling, increasing warning signs, and increased seasonal protection; management strategies, including focusing protection efforts in critical zones, selective protection of critically significant sites, and closing public access to critical areas; maintenance strategies, involving rerouting roads and trails and erecting barricades to prevent looting in critical rockshelter sites; and, interpretation efforts to increase public education programs. Information gathered during the monitoring program indicates that there are three types of looters: (1) opportunists, (2) those that collect for personal acquisition, and (3) those who collect for profit (Des Jean 1988a:6). Ison, Norville, and Pollack (1981:30-32) also identified 3 classes of looters/vandals; un-intentional (opportunists), relic collectors (acquisition), and commercial looters (profiteers). Law enforcement, Interpretation, and Education strategies may have to be tailored to the characteristics of each class of looter in order to effectively combat the impacts of this activity. In addition to the deterrence provided by increasing law enforcement, public education is the other necessary approach towards protection of the archaeological resources on our public lands. The Red River Gorge study (Ison et al. 1981), the Little Colorado Planning Unit Study (Plog 1978), and this monitoring program study have all come to the same conclusions: that Law enforcement and public education are the most important deterrents to archaeological site destruction (Ison et al. 1981:34); and that, "Education is the most important means of protecting archaeological resources" (Lightfoot and Francis 1978:34). Educating the general public on the elementary, middle school, high school, and adult level will instill a degree of appreciation and understanding for the value of cultural resources. Hopefully this will prevent unchanneled interest in prehistoric peoples from becoming a frenzy of site destruction and relic collecting. The dividends from such public education programs, though, will not be observed for years. Of course for some looters education will have little effect. For those the only deterrent is the criminal justice system.
References Cited
Des Jean, T. (1987a) Monitoring Archaeological Site Impacts. Paper Presented at the 44th Annual Southeastern Archaeological Conference, Charleston, South Carolina. Des Jean, T. (1987b) Monitoring Program Interim Report: Following the Completion of Phase I of the Archaeological sites Monitoring Program at Big South Fork National River and Recreation Area. National Park Service, Southeast Archaeological Center, Tallahassee
Des Jean, T. (1988a) Big South Fork National River and Recreation Area Archaeological Resources Protection Plan for FY88 through FY92. National Park Service, Southeast Archaeological Center, Tallahassee
Des Jean, T. (1988b) Looting Activity: A Folk Tradition of the Upper Cumberland Plateau. Lamar Briefs No. 11:6-7, Watkinsville, Georgia Des Jean, T. and Wilson, R. C. (1989) Vandalism Behavior in the National Parks: Diagnoses and Treatment. Paper presented at the 53rd Annual Meeting of the Soc. for Am. Archaeology, Atlanta Faust, R. (1986) Archaeological Resources Status Report, Big South Fork National River and Recreation Area. National Park Service, Southeast Archaeological Center, Tallahassee
Ferguson, T.A., Pace, R.A., Gardner, J.W., and Hoffman, R.J. (1986) Final Report of the Big South Fork Archaeological Project: Survey, Testing and Recommendations. U.S. Army Corps of Engineers, Nashville District
Ison, C. (1989) Oldest Indian Site in Kentucky, 7-12-89. Lexington Herald-Leader, Lexington
Ison, C., Norville, C., and Pollack, D. (1981) Vandalism of Rockshelter Sites in Red River Gorge, Kentucky: An Assessment. Dept. of Anthropology, Univ. of Kentucky, Lexington
Lightfoot, K.G., and Francis, J.E. (1978) Management Information and Recommendations. In An Analytical Approach to Cultural Resource Management: The Little Colorado Planning Unit., F.C. Plog, ed. Arizona State University, Anthropological Research Papers No. 13., Tempe, Arizona
Lyneis, M.M., Weide, D.L., and Warren, E. (1980) Impacts: Damage to Cultural Resources of the California Desert. Dept. of Anthropology, University of Nevada, Las Vegas
Plog, F. (1978) An Analytical Approach to Cultural Resource Management: The Little Colorado Planning Unit. Arizona State University, Anthropological Research Papers No.13. Tempe, Arizona
Wylie, J. (1989) Archaeological Protection Efforts. Federal Archaeology Report, August 1989 Vol.2, No.3:pp5. Washington, D.C.
Williams, L. R. (1978) Vandalism of the Resources of the Rocky Mountain West. Cultural Resources Report No. 21 USDA, United States Forest Service, Southwestern Region, Albuquerque.
Return to the Table of Contents Open the Microsoft Word Version of this Article
|
