| OCR Text |
Show THE VARIANTS OF THE FREMONT: A METHODOLOGICAL EVALUATION PATRICK HOGAN and LYNNE SEBASTIAN Washington State University ABSTRACT Much work in the Fremont region has centered about the question of "variants, "yel the implications of the concepts and assumptions that have structured the investigations related to this question have yet to be stated and considered. The basis of Fremont culture classification is examined, and the concept of variants is evaluated from this methodological perspective. Three points are emphasized. First, the system of cultural classification into which the Fremont culture and its variants have been incorporated implies phylogenetic relationships which are not determinable given the descriptive level at which the definitions have been formulated. Second, this system of cultural classification, despite connotations of varying settlement patterns and means of subsistence, is based on selected artifact assemblages with discrete spatio-temporal distributions. And finally, the variants established using this classification system have been used in regional comparison as if they were solely geographic, with possible internal temporal developments being ignored. INTRODUCTION Since the publication of Marwitt's definition of five Fremont variants (Marwitt 1970), work in the region has been directed toward an " . . . attempt to buttress or abandon the concept of several variants of Fremont culture thought to be recognized in Utah" (Jennings 1977:19-20). Several "attempts" have proposed changes in variant boundaries (Schroedl and Hogan 1975), questioned the distinctiveness of the Fremont from the Anasazi (Berry and Berry 1976), and proposed the variants as separate cultures (Madsen and Lindsay 1977). Yet the implications of the concepts and assumptions implicit in Marwitt's classification - the framework of our investigations - have not been stipulated. This paper, then, seeks to examine the basis of Fremont cultural classification and to evaluate the concept of variants from this methodological perspective. Three points are emphasized. First, the system of cultural classification into which the Fremont culture and its variants have been incorporated implies phylogenetic relationships which are not determinable given the descriptive level at which the definitions have been formulated. Second, this system of cultural classification, despite connotations of varying settlement patterns and means of subsistence, is based on selected artifact assemblages with discrete spatio-temporal distributions. And finally, the variants established using this classification system have been used in regional comparison as if they were solely geographic, with possible internal temporal developments being ignored. While the current designations of the Fremont and its variants could be retained to facilitate discussion, the limitations of the definitions and the implications of the classification system should be recognized. The current system of classification has been imposed on the data rather than having grown out of them. Thus, this system has structured our research instead of being a product of it. We propose that future research be oriented toward studies of cultural adaptation at a community level; definitions of regional variants, if warranted, could then be based on varying settlement and subsistence systems - definitions more appropriate to the processual and culture ecological questions now being emphasized. 13 HISTORICAL BACKGROUND Let us consider briefly the Southwestern system of culture classification, its application to the Fremont culture, and its inherent conceptual difficulties. Between 1914 and 1960 the primary concerns in Southwestern archeology were chronology and the areal distribution ol artifacts (Willey and Sabloff 1974:89). Artifacts became index fossils and typologies were deemed useless unless they aided in effecting spatial and temporal separation of materials (Ford 1954; Rouse 1955). Because of this emphasis, pottery classification assumed considerable prominence in Southwestern archeology. Ceramics in the Southwest varied greatly in time and space, and potsherds were numerous and almost indestructible - well suited to counting and manipulation. The ceramic classifications pioneered by Winifred and H.S. Gladwin (1930) and Colton and Hardgrave (1937) were based on biological taxonomies, both in the use of binomial designators for pottery types and in the phylogenetic system through which types were related. Gladwin's culture classification system was a direct outgrowth of ihis biological approach to ceramic classification. Using the analogy of a tree, Gladwin designated the most fundamental grouping of cultures as "roots." These split into "stems," and stems, in turn, into "branches." The working units from which these tree structures were constructed were "phases" (Gladwin and Gladwin 1934). As described by Colton (1939), a phase was determined by noting similarities in ceramics; architecture; stone, wood, and bone implements; and textiles from contemporaneous sites in a restricted geographic area. Similar phases in adjacent areas were combined into branches. Branches were assigned to stems, again on similarity of traits, especially ceramics. The process was one of gradual grouping based on artifact similarity, geographic location, and temporal provenience. For the Fremont, however, the process just described was bypassed. PHYLOGENY AND THE FREMONT The Fremont culture is described by Marwitt as ". an areal tradition taxonomically equivalent to Anasazi" (Marwitt (1970:137). As Madsen and Lindsay have observed (1977:89), this definition has arisen more through contrast of the materials with the Anasazi than through the grouping process described above. Once it was established that there were basic differences between the prehistoric horticulturalists of the "Northern Periphery" and the Pueblo culture of the San Juan drainage, the Fremont culture was accorded equal taxonomic status with the Anasazi - a distinct "stem" in Gladwin-Colton terms. Subsequently, Marwitt defined the Fremont variants (branches) with the same degree of regional specialization and cultural distinctiveness as the Mesa Verde or Kayenta Anasazi. Classification of regional variation within the Fremont was thus a dividing of the whole rather than a grouping of parts. Despite the anomalous and rather arbitrary manner in which the Gladwin- Colton system was applied to the Fremont, that classification has continued to structure our thinking and research, even as it has been abandoned in the rest of the Southwest. In the Gladwin-Colton classification, definition is based on variation in artifact style and assemblage composition, which leads to an emphasis on the trait list. A growing interest among archeologists in cultural ecology introduced the concepts of site location, settlement pattern, and subsistence pattern into definitions of cultural subdivisions, but locational and subsistence data do not permit the fine distinctions that are possible using artifact taxonomies, so these taxonomies were often retained And, as our working definition of culture has changed from "a distinct artifact assemblage" to "a dynamic system of human adaptation," the retention of culture classification based primarily on artifacts has frustrated attempts to deal with questions of origin, internal development, and cultural demise. Similarly, the phylogenetic scheme inherent in the Southwestern classification system has frustrated attempts io understand the scale and development of cultural variation. The problem here is a simple one: the classification scheme incorporates in its hierarchy assumptions about the relationship between a culture and its subdivisions that remain to be demonstrated for the Fremont culture and its variants. As long as the classification is applied a priori, research will be self-fulfilling and circular. The anomalous manner in which the Fremont was incorporated into the Gladwin-Colton classification system has a particular implication for any consideration of the problem of cultural origin and the basis of cultrual variation. Currently the most widely accepted hypothesis of Fremont origin is that Fremont was derived from an indigenous Archaic culture that adopted ceramics, horticulture, and pithouse architecture through the same early stimulus of Southwestern influence that triggered Anasazi development (Jennings J 978a: 155-156) This hypothesis is based on the presence in the Fremont material culture of typical Archaic artifacts, especially the Fremont moccasin and one-rod-and-bundle-basketry (both distinctive features of the Desert Archaic in the eastern Great Basin) and on the stratigraphic evidence lrorn Hogup Cave showing apparent continuity between Archaic and Fremont levels (Aikens 1970b:204). 14 Others have hypothesized that Fremont had a northern Plains origin (Aikens 1966b; Sharrock 1966) or was a northward extension of the Anasazi (Gunnerson 1969; Berry 1975). Our purpose here is not to evaluate the relative merits of these three hypotheses, but to point out how the incorporation of the Fremont into the Southwestern classification has crippled our attempts to resolve this question. As discussed, the phylogenetic nature of the Gladwin scheme makes Fremont culture a stem and root; if we accept this, then there can be but one origin of the Fremont, and the variants must represent subsequent diversifications. The possibility suggested by Madsen and Lindsay (1977) of different origins for the variants and subsequent convergence cannot be effectively analyzed as long as the present framework is in use. By the same token, their definition of separate cultures ol equivalcni taxonomic status would, if accepted, thwart attempts to analyze the Fremont as a single cohesive entity with regional variability. Manipulation of the present classification system cannot better explain the variation in the prehistoric agricultural cultures of Utah. Because the classification scheme contains genetic implications, the relations between the variants are predetermined by the way they are classified. There can be little doubt that the classification of Fremont culture has brought order where there was once only chaos. But given the ambiguities and pitfalls of that system, it is reasonable to ask whether such broad classifications may not be premature. Is the gain in regional clarity at this stage of research worth the price? TRAIT LISTS AND THE VALIDITY OF THE VARIANTS The reliance on trait lists to define a culture inherent in the Gladwin-Colton system bears directly on the definition, utility, and cultural reality of Marwitt's variants. Because of a concern for chronology and culture classification in the Southwest, artifact typologies became formalized, emphasizing those attributes that provided the greatest degree of temporal and spatial separation. Artifacts came to be treated as fossils with characteristically restricted distributions. Although the gains in the understanding of Southwestern prehistory that resulted from this strategy should not be underestimated, the focus on artifacts in time and space led to an increasing removal of the artifacts from any cultural context. With extension of this strategy to culture classification, the problem was magnified. During the early twentieth century it was accepted that the restrictions of archeological data would permit archeologists to contribute little to the field of anthropology. At best they could only hope to achieve an ordering of prehistoric cultures in time and space. And these cultures were literally collections of traits - assemblages of artifacts and features that co-varied in time and space, artifact traditions whose relationships were defined on morphological similarity. As culture historians attempted to flesh out their regional frameworks and concern about relating archeological material to cultures began to emerge, data on subsistence, settlement patterns, and other aspects of lifeway were added to the basic artifact definitions. This was done largely through analogy with ethnographic records of historic peoples. Other than providing an appearance of more comprehensive culture classification, this veneer of anthropological relevance has had little impact on the study of Fremont variants. In the area of subsistence, for example, little systematic work has been done. That there was great variation is evident; the use of marshland resources and bison by the Great Salt Lake Fremont (Aikens 1966b; Shields and Dalley 1978) differs fundamentally from the reliance on cultivated crops apparent in the San Rafael area. Madsen and Lindsay (1977:90) have suggested these differences are sufficiently great to warrant separate cultural designations. Our view is that any basing of classification on subsistence or settlement pattern at this time is premature. The necessary quantity of information is simply unavailable. In his definition of Fremont variants Marwitt states: " . . . a defensible ecological definition of each variant is not possible at present - primarily because environmental information is not available in sufficient quantity, but also because patterns of aboriginal land use have not been studied in detail and ecological districts are by no means coextensive with the geographical limits proposed here for each cultural variant (1970- 138)." Our knowledge of Fremont settlement and subsistence has increased little since then, and despite Marwitt's intentions, the variants of the Fremont remain simply **. . . geographic divisions reflecting slightly different assemblages of material culture . . ."(Marwitt 1970:138). The reason lor this, we believe, is that despite the tentative nature of the variants as proposed, those of us involved in Fremont research have operated as if these variants were established patterns of varying cultural adaptation as well as of artifactual material. Site artifacts are analyzed; the site is assigned to a variant based on location and artifact content; and interpretation of economic and demographic factors are taken from the assumptions of the classification scheme rather than from detailed study of the site material. Rather than providing a focus for more detailed study, as intended, the variants have merely given an illusion of sophistication to our pigeonholing of artifact complexes. Until careful studies of regional cultural adaptation provide some concrete data by which to assess the variants, this classification based on geographic distributions of artifacts will remain 15 an inappropriate Iramework for our inquiries into the "cultural reality" of the Fremont and its regional subdivisions. We might ask whether the variants, stripped on any ecological implications, could not be useful as a descriptive means of ordering systematic covariation ol artifact types in a regional sequence. In reality such a systematic-covariation does not occur; artifact distributions overlap, but not regularly. In order to define variants on the basis of artifact assemblages, those artifacts producing the greatest separation are used to establish the borders; the less separable artifacts that fall within that area arc then tacked onto the trait list lor that variant. As now defined, the Fremont variants are most defensible as a map of the distribution ol a ceramic tradition. II emphasis is given to a different cultural trait, marked changes in variant borders occur. This does not imply that distinctive artifact clusters cannot be formulated, but these aggregates should include a large number of artifact types possessed by most of the sites, with no single type necessary or sufficient for membership. TAXONOMY AND TEMPORAL VARIATION In discussing the conceptual problems of Fremom cultural classification we have so far emphasized two problems inherent in the typological system itself. A third problem relates to a shift in the theoretical orientation of Fremont specialists from culture history to culture ecology. This new emphasis has altered our conception of what a variant is, causing us to inadvertently ignore temporal variation in our taxonomic scheme. Marwitt conceived of the Fremont variants as regional sequences with the implicit assumption that chronological phases would be worked out as more data became available. Due largely to a general shift in interest from chronology to cultural ecology this remains to be done. The variants have rather become designators for different adaptive patterns, despite Marwitt's statement of the tenuous nature of such an interpretation. And as research has proceded, it is the variant itsell that has become the basic unit of comparison. The variation due to several hundred years of style change and cultural development within an area has thereby been effectively compressed into a single spatial grouping, and the resulting loss of temporal variability has made the question of regional variation more ambiguous. Do figurines occur only in a few variants throughout the Fremont occupation, or are they a later embellishment on a more basic Fremont cultural pattern? Are cultigens part of the cultural legacy that triggered development of a Fremont culture which deemphasized them in later periods in favor of other resources, or were they never accepted over the entire Fremont area? Such questions would require that finer chronologies be formulated for each variant separately and that inter-variant comparisons take chronological phases into account. SUMMARY In this paper, we have argued that the scheme of culture classification that marks the present level of Fremont regional synthesis has structured our inquiry into Fremont culture history. Inherent in the classification of Fremont and its variants in terms of the Gladwin-Colton Southwestern classification scheme are two problems. First, the genetic relationships assumed in the system frustrate efforts to determine the "cultural reality" of the variants, their relationship to one another, and indeed, the origin of the Fremont culture itself and its relationship to other Southwest traditions. Second, the classification of Fremont variants is based on the distribution of only a few diagnostic artifacts, so that their present use as ecological variants is premature and misleading. An additional difficulty with our current use of the variants is that in using them as designators of varying adaptation, we have come to ignore temporal variation within each variant when comparing artifact traits between regions. Given these criticisms, we conclude that the variants as defined have only minimal validity - a validity that is more than offset by the problems the framework causes in misdirecting our research goals. It seems unlikely that the present designators will be abandoned - the framework does have some utility in defining our interest area and in facilitating communication among areal specialists. It is imperative, therefore, that the implications of the way Fremont specialists have applied this classification be recognized and that the pitfalls we have outlined here be consciously avoided. In restructuring work in the Fremont region, it seems most profitable to drop back to concentrated studies of communities, exploring in detail local systems of adaptation. As studies in adjacent areas are compared, new variant definitions can be formulated, based less on traits than on technology, economics, and settlement patterns. Sequences of development can then be built up for the local area and gradually expanded, creating a classification system based on the empirical realities of Fremont culture or cultures rather than on a procrustean application of the Colton-Gladwin system. 16 |
