A BRIEF HISTORY OF ARCHEOLOGICAL SURVEY
Modern archeology in North America has roots in the dilletantism and antiquarianism of the 19th century. Some of the earliest "archeologists" were explorers, traveling journalists, soldiers, and natural scientists who described archeological sites that they sought out or stumbled upon during travels in the little known regions of the west and Latin America (cf. Stephens 1841). In this sense, archeological survey has a long tradition in American archeology.
Some of the earliest archeological publications were essentially survey volumes, describing the ruins or mounds that had been discovered in some particular area of the country, discussing collections that had been derived from them, and speculating on their origins and functions (cf. Squier and Davis 1848). These studies were a far cry from the systematic surveys conducted by archeologists today; they were simple explorations in which the fieldworker described those phenomena that came to his attention with no pretense of identifying all the vestiges of past human activity in the area. Such full descriptions were not necessary to the authors' purposes.
As archeology became a recognized discipline in the United States in the latter part of the 19th century, such general exploratory surveys were a normal part of its research repertoire. The purposes of survey were almost totally descriptive; sites might generally be compared and contrasted with one another on the basis of survey data, but in most cases the survey was regarded primarily as a prelude to excavation. One surveyed to locate sites to dig. Survey methods were not the subjects of much concern. The archeologist presumably knew what kinds of sites he wanted to dig, and the survey simply involved looking for them. If other sorts of sites were missed in the process this was of no importance, because the archeologist did not want to dig them anyway. The survey was not itself seen as a research tool, since relationships among sites were not generally considered important.
At this time, archeology was primarily oriented toward the study of change in artifact types, structural types, and other attributes of archeological sites through time. The aims of such studies were the characterization of ethnographic groups in time-depth, the search for the origins of particular cultures, and the reconstruction of culture history (cf. Willey and Sabloff 1974: 42-64). Early studies were directed toward demonstrating what were thought to be universal patterns of human cultural evolution, showing that given societies had advanced inevitably through stages equitable with "savagery" and "barbarism" to "civilization."
In the early 20th century, the concepts of unilinear cultural evolution began to go out of vogue, to be replaced with what became known as "historical particularism." Historical particularism denied the possibility of readily demonstrating large-scale evolutionary changes; particularists argued instead for the painstaking reconstruction of the histories of particular peoples and cultures. These small histories, it was thought, could eventually be synthesized to permit the development of an understanding of cultural evolution in general. Archeologists trained in the historical particularist tradition naturally tended to direct their research toward the reconstruction of the culture-histories of particular sites or small areas. Very careful study of local culture change sequences became the rule of the day (cf. Willey and Sabloff 1974: 88-98).
The physical focus of the study for the particularist archeologist, however, was not greatly different from that for the unilinear evolutionist. Both approaches to culture-historical study caused archeologists to value large, deep sites with many strata or other indicators of change through time, and with many artifacts that could be equated with things used by ethnographic peoples. In such sites, cultures could be speculatively described through analogy with living groups that used similar artifacts, and could be seen succeeding one another in more-or-less orderly progression through the strata. When surveying an area, most archeologists sought sites of this type, and virtually or entirely ignored the small, or shallow, or recent sites that did not promise to contribute directly to culture-historical reconstruction.
During the 1930s, archeology became deeply involved in the emergency employment programs initiated by the Roosevelt Administration. Large numbers of workers could be committed to archeological activities, to do socially useful work under relatively low-cost supervision. As a result, huge crews were thrown together and sent into the field under the leadership of archeologists--themselves often young graduate students or avocationalists. While some of these projects were utter disasters, others provided extremely important bodies of data, and the exercise had profound effects on the nature of archeological practice. One such effect was on archeological survey.
Large areas were surveyed, in advance of construction projects such as reservoirs or simply in order to deploy large numbers of people in activities that would do minimum damage to archeological sites. Because the workers were unskilled, and their supervisors often not highly trained or broadly experienced either, it was necessary to develop somewhat standardized methods of recording sites. Work was usually undertaken in areas where large populations of unemployed workers existed, not necessarily where an archeologist, left to his own devices, might have chosen to work. In consequence, the archeologists often found themselves dealing with areas they did not know well, where they were not sure just what kinds of sites to seek. Thus it became necessary to think about what constituted an archeological site, what gave them importance, and to record a wider range of sites than would have been recorded by an archeologist simply seeking sites to dig for pure research purposes.
The rationale for survey remained the discovery of sites for excavation, and reconstruction of culture-history was still the main reason for excavation. Changes were in the making, however, springing in part from other effects of the depression and its archeological activity. The make-work programs had often forced archeologists into work they would not have done otherwise, at sites that would normally not have tempted them. In some cases they found the results very interesting. The big work crews made possible the stripping of large site-areas, revealing the organization of entire prehistoric villages and showing that more could be studied about extinct human groups than the ways their artifacts changed through time. The Roosevelt Administration itself, with its somewhat socialistic policy overtones, may have set the stage for the rise of cultural materialism in anthropological theory, which characterized the 1960s. This in turn would contribute to a major change in the ways in which archeologists looked at their world and their research base.
Before these changes took place, however, archeology lapsed into general quiescence during World War II. After the war, with the initiation of huge water-control projects across the nation, archeology was faced with a major challenge, and the era of "salvage archeology" began in earnest. Initially, the Smithsonian Institution and the National Park Service undertook the river basin salvage program; after passage of the Reservoir Salvage Act of 1960 (Public Law 86-523), the responsibility became more and more concentrated within the National Park Service's Interagency Archeological Salvage Program (now Interagency Archeological Services) until, in 1969, the Smithsonian Institution divorced itself from the program entirely.
Surveys were obviously required as the first step in most reservoir salvage projects. These surveys were still aimed almost exclusively at locating sites to dig, and sites were still chosen for excavation primarily when it was thought that they would contribute to the construction of culture-historical sequences. The scant funds appropriated to the National Park Service for salvage were largely reserved for excavation, so surveys were done cheaply and fast. The results were sometimes appalling by modern standards. At the proposed New Melones Reservoir in California, for example, a river basin salvage program survey in 1948 resulted in the recording of only four sites, none of which was regarded as being of sufficient importance to justify expenditure of the program's limited salvage money (Frederickson 1969). Based on recent survey, the still uncompleted New Melones Reservoir is now recognized as a National Register District because of its 190 prehistoric and over 400 historic sites, structures, buildings, and objects (Moratto 1967).
When the Interstate Highway System went into construction during the 1950s, archeological surveyors faced a new set of challenges. Lacking the Congressional mandate for a salvage program like the one provided for reservoirs, archeologists developed such arrangements as they could with their state highway agencies for the conduct of surveys and salvage. Pipeline construction companies also began to make arrangements for salvage. One of the first and most successful highway and pipeline salvage programs was that developed by Fred Wendorf and his colleagues at the Museum of New Mexico. Wendorf's experiences resulted in the preparation of a manual on salvage archeology. The conditions under which many salvage surveys were done were described as follows: "The archeological teams follow as closely behind the surveyors and as far ahead of the right-of-way clearing machinery as possible. Even under ideal conditions the timing will still be close, and there may not be more than three to four weeks between the survey and dozer clearing the right-of way." (Wendorf 1962:54)
Working under such pressures, archeologists found themselves having to think in new ways about old and little considered questions: what sites were worth recording? What made a site worth digging? How could they locate, record, and excavate them most efficiently?
As the 1950s progressed, anthropologists in many of the Nation's universities had become dissatisfied with historical particularism as their basic approach to understanding human society (see Harris 1963, for an extended discussion). Generally, the construction of local culture-histories had not provided the basis for syntheses that revealed much about culture-change. Some began to believe that the cart had been placed before the horse, i.e., that precise questions should be developed about culture-change, and hypothetical answers proposed, before data could be collected in a fruitful manner. It was recognized that it was physically impossible to collect all the data that might exist about any living culture or any archeological site; data collection is always selective. Without having formulated questions to be answered before fieldwork began, it was unlikely that the anthropologist or archeologist in the field would select the data necessary to answer them. At the same time, with the decline of the mid-5O's extreme anti-communism, theories that purported to explain culture-change using models derived from Marx and other materialist thinkers became popular. Testing materialist propositions about the nature and causes of culture change required the study of relationships between human society and the material resource base--the natural environment. The rise of environmental anthropology (Steward 1955; White 1959) coincided nicely with the growing recognition among salvage archeologists that there were scientifically valuable data in sites that were not large or deeply stratified. In fact, it was clear that if one really wanted to understand the relationships between human groups and their environments, one needed to look at all kinds of sites, representing all kinds of interactions with the environment. Small sites representing a small range of activities carried out during a single season with reference to a single economic resource were at least as important to understanding human-environment relations as were big, deep sites where people lived repeatedly or year-round and engaged in a diversity of activities. On the upper Texas coast for example, very small, unstratified sites provided an unequaled source of clear information about short-term activities--involving periods ranging from probably one day up to two-or-three weeks. Such information could virtually not be obtained from complex, stratified sites (Aten 1977).
Not only have small, shallow sites begun to get new attention from environmentally oriented archeologists; the relationships among sites, and between constellations of sites and the environment in which they existed, have become fruitful objects for study. The focus of archeology during the 1960s shifted rapidly, from the individual site to the regional settlement pattern (cf. Chang 1968). Archeological survey itself began to be recognized as an important research tool, one which, even without associated excavation, could show how human populations and their activities had been distributed within a natural environment. Survey was gradually redefined, no longer being viewed simply as exploration to find sites to dig but as a systematic effort to "provide information on the number, the location, and the nature of the sites within a given region" (Heizer and Graham 1967:14).
The importance of archeological survey as a research activity has continued to grow during the last decade. Strangely, however, the literature concerning survey methods remains unusually limited. In 1966, Reynold Ruppe published a case study concerned with demonstrating how well-organized, problem-oriented archeological survey could "be made to produce information that is usually considered procurable only by excavation" (Ruppe 1966:331). Many studies based on survey data have been published since (cf. Thomas 1975; Schiffer and House 1975, 1977; Matson and Lipe 1975), all devoted at least in part to the discussion of the advantages and deficiencies of particular survey strategies. Some papers have also been devoted exclusively to the discussion and, in some cases, the comparison of survey strategies (cf. Mueller 1974, 1975; Lovis 1976). These studies and discussions have all concerned themselves only with predictive sample surveys. With the exception of Lovis, all have reported on work done in the arid to semiarid west, and most have focused primarily on the problems of selecting appropriate sample areas for inspection rather than on the problems of inspection itself. These are important problems, and predictive sample survey is of great value for management purposes. Discussion of such surveys follows in Chapter VII. First, however, we need to consider some more basic questions about archeological survey from a management perspective.