Bison Rockshelter is one of the two most significant archaeology sites found within the Birch Creek area, the other being Veratic Rockshelter (10CL3) which lies just to the North. These shelters were discovered during the Birch Creek Archaeology Project which revealed 136 archaeology sites, including 54 open camp sites, 77 rockshelters, 1 “tipi ring,” and 4 circular rock structures (Swanson 1964). The geological formation of the two previously mentioned rockshelters was begun by the erosive action of Birch Creek as it flowed against the rock face during the late Pleistocene. The shelters were enlarged to their present size by the process of insolation, frost, weathering, and structural failure. They were eventually partially filled by alluvial fan deposits (Swanson 1972). The Birch Creek Project was conceived by Earl H. Swanson Jr. to test his theories regarding the prehistory of the Shoshonean speaking peoples of the Northern Rocky Mountains. Swanson believed that the spread of Shoshonean dialects could be explained as traveling South and Westward out of the Rocky Mountains as well as diverging from the dialects spreading in a North and Eastward movement from the Great Basin (Butler 1981). There is archaeological evidence of continual human occupation, alternating between Bison and Veratic Rockshelters, dating as far back as 11,000 years ago (Butler 1981). Initial work in this rich valley region was begun with an archaeological reconnaissance survey in 1958, which led to the rockshelter excavations during the field seasons of 1960 and 1961 (Swanson 1972).
Bison Rockshelter lies within the middle course of the Birch Creek Valley. This valley and the Lemhi River Valley to the north constitute the largest contiguous valley of north-south orientation in the Pacific Northwest. Birch Creek is a high mountain valley whose headwaters begin from natural springs in the north, just below the Gilmore Divide, and flow south to the sinks at the northern edge of the Snake River Plain (Swanson 1964). There is a wide variety of larger and small game indigenous to the area. Plants include bunch grass, sage brush, birch and willow trees, lodgepole pine, Engelmann spruce, mountain hemlock, alpine fir, aspen, and perennial as well an annual herbs and flowers (Swanson 1972, 1964).
The published accounts of Bison Rockshelter offer limited ranges of artifact types, dating from 11,000 to 150-years ago (Butler 1981). These artifact types include modified stone and bone tools, large and small faunal remains, historic trade items, as well as samples of charcoal, ash, and soil. The shelter walls were noted to have several examples of pictographs drawn on them. Original estimates place the number of artifacts from this site at 2,625. There are, in fact, 4,494 artifacts, with an additional 261 artifacts currently missing or unaccounted for. Discrepancies in artifact counts may be due in part to a lack of consistent controls in the original field and lab work as well as the decades long delay in both analysis activity and the processing of the collection and related documentation for long term curation. Given the amount of time that has lapsed since the original excavation occurred, there are a few problems with the collection that are likely to never be resolved, but these concerns are minor enough that they do not affect the overall integrity of the collection and its importance to the scientific community.
A great deal of work went into processing the artifacts from Bison Rockshelter, partially due to the time laps between when the artifacts were collected and finally processed. The artifacts were removed from their storage containers and individually checked against the original field inventory account logs and the catalog cards. This process exposed several problems with the documentation and the artifacts. Many of the artifacts were dirty and some still in field lot bags with no catalog numbers or reference to provenience. Most of the faunal remains and the stone tools were stored in large bags not meant for long-term curation, and artifacts from several other sites were mixed into the bags.
Some of the notes in the logbooks made obscure reference to objects that were never located during the course of this project. These objects may have been discarded as “insignificant” by the original field or laboratory analysis, this is not clearly stated, although, in some instances it is alluded to. It is possible that there are objects that are on loan to researches involved with the original project that never submitted the artifacts they were working on to the museum.
An analyst examined the faunal remains and some were reclassified. The same was done for the stone tools, the majority of which were reclassified. The artifacts with duplicate and missing numbers were assigned new numbers, which were referenced to artifacts that were found in the same storage bags. A new catalog was generated as a result of these changes. The entire cleaned and reconciled collection was then digitally photographed with the images placed online along with a site overview and the catalog for the purpose of research and public access. All the artifacts were then individually bagged and placed into proper storage containers appropriate for long-term curation.
The importance of Bison Rockshelter in its prehistoric context must be interpreted in conjunction with its nearby partner, Veratic Rockshelter. Even though both sites are examples of secondary refuse (artifacts discarded away from their primary use location, an activity in which much of the cultural significance is lost), together the shelters provide evidence of a continuous occupation spanning 11,000 years for the hunting and gathering culture commonly associated with the Northern Shoshoni in the Northern Rocky Mountain setting (Schiffer 1985). This evidence lends strong credence to Dr. Swanson’s theories about the diffusion of the Shoshonean ancestors.
The Bison Rockshelter assemblage is being reexamined for the dual purpose of preparing the collection for long term curation at the Idaho Museum of Natural History and placing the resulting information into an on-line catalog to improve public access to this important cultural material. Processing activities include examining the artifacts and faunal remains for proper identification and conservation needs, and relabeling and repackaging where needed. Each object will be then digitally photographed and information about it entered into a database that will include catalog numbers and descriptions necessary for scientific study and research. Funds for this undertaking are being made available through a 2001 Cost Share Challenge Grant between the Bureau of Land Management – Idaho Falls District Field Office, and the Idaho Museum of Natural History, Idaho State University, where both parties share in the cost of bringing the collection up to the condition needed to better preserve it for future use while making the information about it available to the general public, the ultimate owners of this collection.
Butler, B. Robert
1981. When did the Shoshoni Begin to Occupy Southern Idaho?: Essays on Late Prehistoric Cultural Remains from the Upper Snake and Salmon River country. Occasional Papers of the Idaho Museum of Natural History, No. 32
Swanson, Earl H. Jr. and Alan Lyle Bryan
1964. Birch Creek Papers No.1 An Archaeological Reconnaissance in the Birch Creek Valley of Eastern Idaho. Occasional Papers of the Idaho State University Museum, No. 13