Chapter Two Table of Contents
·
II. MODEL
BUILDING: A BACKGROUND
II. MODEL
BUILDING: A BACKGROUND
Although the study goals are not anthropological entirely, our ideas about how
best to produce a “forecast” of where one was most likely to find
archaeological sites derived from the anthropology of the Great Basin.
Anthropology – the study of man – has always been closely tied to archaeology
in the Great Basin. Our approach to model-building drew from the seminal work
of Julian Steward, as did the work of many, many, other archaeologists in the
region.
In this chapter, we present brief discussions of these major models. Relatively little of what we are doing here is new, novel, or untried. So, these short summaries inform as to why we took particular decisions in this research.
Environmental
Models
Julian Steward laid the groundwork for much of the research done in the Great Basin subsequent to the publication of his seminal work Basin-Plateau Aboriginal Sociopolitical Groups (1938). Researchers used the results of Steward’s ethnographic reports as springboards to study the environment as a limiting factor in the level of cultural complexity attained by people in semi-arid landscapes. Limited resources in these areas would force the inhabitants to spend the majority of their time and effort procuring food and producing the technology required to aid in these tasks. Subsistence and settlement patterns could then be explained and explored in these terms. Steward’s information provided many research topics and continues to be a valuable source of information for archaeologists.
Archaeologists quickly picked up from Steward the importance of pinyon nuts as a staple resource. Several studies examined whether the ethnographic reliance on pinyon had antiquity (Thomas 1971, 1973; Thomas and Bettinger 1976). Central Nevada studies showed that pinyon was, indeed, one of the long-term staples. This led to a focus on pinyon as the determinant of prehistoric settlement pattern.
More generally, Steward’s work created a family of models that Wilde (1994) describes as paleoecological models, including several relating to the Great Basin. These include:
· Steward’s ethnographic model discussed above which argued that the Great Basin had a “socially fragmenting effect upon its prehistoric inhabitants”;
· Jenning’s “Desert Culture” model based explicitly on Steward’s work, originally set out to account for the record at Danger Cave in northwestern Utah and proposed a cultural ecological model in which a stable settlement and subsistence pattern was evidenced for the past 10,000 years;
· The Warner Valley model as delineated by Weide (1968) which is a lake and marsh-oriented pattern, but with increased reliance on upland faunal resources;
· O’Connell’s Surprise Valley model with early and later variants. The early period (6500-4500 B.P.) was based primarily on marsh and grassland resources supplemented by upland animals during the winter; and,
· The Steens Mountain Model which shows an inverse relationship between site frequency and site size, which suggests high resource productivity, allowed population aggregations (Wilde 1994:97-102).
For a period of about 15 to 20 years, the pinyon-centric model of aboriginal settlement was truly the dominant paleoecological model. Because pine nuts are easily harvested and give a rich return, their absence can be predicted up to two years in advance, and their presence at least predictable in location, they were given primacy in many of the models of Great Basin prehistory. However, further archaeological work showed that the pinyon-centric model of aboriginal subsistence and settlement was too narrow. Other natural settings in the Great Basin, especially wetland and lacustrine environments, have long and rich archaeological records too. Resources in these settings are not so easily understood as pinyon nuts. For instance, why would one gather cattail pollen instead of harvesting pinyon nuts? Was this an alternative subsistence strategy or equal to “king pinyon”? Exploration of these questions brought anthropologists and archaeologists in to a consideration of caloric maximization in patchy environments: optimal foraging theory. Optimizing theory attempts to understand, and thus predict, the choices that a rational forager will make.
Optimizing models have been very successful as a deductive form of environmental model. Many studies in the relatively stark Great Basin have used models of what foragers should have done as rational behavior. These studies then examine whether the archaeological record matches the predicted behavior. Generally, such studies have been successful over areas of about half a million acres, such as a typical basin and range valley (Bonstead 2000; Connolly 1999; Gehr 1980; Jones et al. 2002; Mehringer 1986; Nials 1999, 2000; Pendleton 1979; Pettigrew 1984; Pinson 1999; Thomas 1971).
In the central Great Basin, the primary GIS optimal foraging models for the Great Basin have been proposed by Zeanah et al. (1995), Zeanah (in press), and Raven and Elston (1989). Beck and Jones (2000) provide a thoughtful overview of how these efforts fit within contemporary regional research directions. Optimal foraging approaches are not without problems. One of the main criticisms of the use of these models is that they are not easily replicated. Though they go far in description, they offer little in explanation outside of resource return rates in the form of calories expended and/or gathered per hour. Optimal models provide detailed formulas for energy return rates, but do not account for resources used in other contexts such as medicine, ritual, fuel, or shelter.
Overall, then, the history of inquiry in Great Basin archaeology has gone from informal paleoenvironmental models to ever more detailed and quantitative approaches. The latter methodologies, especially optimal foraging, provide numeric baselines from which to understand prehistoric settlement patterns.
Geomorphic
Approaches
A geomorphic site preservation approach has been applied to archaeological studies in the Great Basin. This model developed and refined by Nials (1999, 2000) uses geomorphic principles to identify areas likely to retain in situ cultural materials. Suitable locations include those lying on and adjacent to:
· Late shorelines of pluvial lakes, including dunes contemporary with late pluvial lake shorelines;
· Distributary drainages entering open basins;
· Upland valley bottoms where stream gradient locally flattens out and the valley widens;
· Near springs active at the appropriate times and;
· Rockshelter and caves (Nials 1999).
This approach contains some tautological assumptions pointed out by Nials (personal communication; 2002) that make it problematic for elucidating patterns in the archaeological record. The geomorphic model promotes the survey of landforms that are favorable for, and have a high probability of, containing intact sites. In other words, well-preserved sites are looked for in the exact environments in which they should be found. Whitley (2000) notes that cause and effect in the record become difficult to discern:
For instance, correlating 97% of
sites with floodplains is meaningless if 97% of the survey areas from which the
data is derived occur on floodplains. . . it is assumed that geomorphological
setting was a constraint on site locations for instance, yet it is rarely clear
how important certain geologic structures were in comparison with the
relationship to a permanent source of water.
Secondly, it is unclear whether it is the geomorphological setting or
the distance to water is important, if there is already a spatial correlation
between the two. . .Whitley (2000:27).
Approaches to
Model Formation
The analysis done for the GBRI project does not attempt to falsify other models but points to the fact that they may not be the best approaches for GBRI. Selectionist models and others that attempt to explain human behavior in terms of natural selection are based on biological principles of animal behavior. In many cases, applying that theoretical approach to model human behavior does not provide adequate explanations for social components of the system, though they do provide general descriptive frameworks useful in explaining optimal utilization within broad environments. Perfect information about people’s environment is rarely available, “which means that they never really forage optimally, but base decisions on their best guesses” (Kelly 1995:100). Optimal models work well in small, delimited areas, but would not be practical to derive for the 20 million acres under consideration for the GBRI project area, which covers parts of three states in different topographic and environmental settings.
Because the units of analyses are hydrographic units (HUC) explored in a Geographic Information System (GIS) environment, the next section provides a very basic discussion of the premises of GIS. GIS will be used in the plural when referred to in a general sense, as there are many GIS programs, and in the singular when used in reference to the results of this particular project, as the final models were built using one particular product, ArcViewâ version 3.3.
Return to Top
|
|
|
|
|
|
|
