Daniel S. Meatte
(taken from the Prehistory of the Western Snake River Basin (1990) pp.63-70)
THE ETHNOGRAPHIC MODEL
The random descriptions of Native American peoples in the study area recorded in historic journals of western fur traders and explorers (Fremont 1970; Ogden 1950, 1961, 1971; Stuart 1935; Talbot 1931; Wyeth 1851) and more comprehensive, systematic descriptions made by ethnographers after the turn of the century (Liljeblad 1957; Lowie 1909; Murphy and Murphy 1960; Stewart1941) have cumulatively formed a rich and varied ethnographic record. Archaeologists have attempted to use this ethnographic record, by way of analogy, as a comparative model for aboriginal lifeways that might strengthen a variety of archaeological inquiries (Chang 1967; Fowler and Jennings 1982:121).
The ethnographic record has also served archaeologists as an ethnographic datum from which cultural and material affiliations can be traced backward in time, using the Direct Historical Approach (Steward 1941; Strong 1935). Using Steward's model of Shoshone lifeways, Jesse Jennings conceived of a generalized adaptive model, the Desert Culture Concept, of prehistoric peoples occupying the arid lands of the Great Basin region (Jennings 1957). Similarly, most archaeologists working in the study area have implicitly operated with the Steward model and the Jennings Desert Culture Concept when interpreting archaeological assemblages (Butler 1978b; Plew 1976a; Swanson 1965; Tuohy 1963).
A more critical test of Steward's model was applied by Plew with data from the Owyhee Uplands region of southwestern Idaho (1980a:80). Plew found that some, but not all, aspects of the Steward model pertinent to the region could be archaeologically demonstrated (1980a:80).
Ames has recently suggested that any "confirmation of Steward is inevitable and perhaps trivial" (1982a:86). He bases this critique in the belief "that the patterns described ethnographically for the area are generally those of any mid-latitude foragers" (Ames 1982a:86). Ames suggests that if we are to assess the applicability of Steward's model to the archaeological record we must build and employ explicit tests and/or build settlement models "without reference to the ethnographic record" for comparison (Ames 1982a:86).
Clearly, with only one critical test available, any judgment about the relevance of Steward's model is premature at this time. A need for comparative testing of the Steward model in most of the principal ecological settings of the study area (i.e., riverine settings, high intermontane prairies, and lowland deserts) exists, and future researchers should pursue such endeavors (for examples see Bettinger 1977; C. Fowler 1982; Janetski 1983; Thomas 1973).