E.B. Banning's Archaeological Survey is a volume I would recommend to any geoscientist who wants, and perhaps even needs, to better understand what an archaeologist means by the word "site." The book surveys the methods, technologies, and theoretical approaches that archaeologists use to find, record, and analyze archaeological sites. A central theme of the book is that sites, as recorded manifestations of past cultures, do not exist independently of the surveys that find and document them. The kinds of inferences that can be drawn from sites depend on the methods, techniques, and theoretical frameworks that led to their discovery by archaeological survey.
Chapter I organizes the literature on archaeological survey into nine "models" that have historically been used as guiding principles for archaeological surveys. With the exception of the Earthwork and Monument models, which merely focus attention on particularly obtrusive or spectacular classes of archaeological features, the models are, in essence, models of site-formation processes. They model how and why human activity, in concert with natural processes, produce artifacts and features that are distributed in nonrandom patterns across the land. The Distributional ("Nonsite"), Place, and Paleolandscape models will perhaps be of greatest interest to geoarchaeologists, since they involve searching for sites within the context of the landscape. The other models, however, are no less important. Once acquainted with the "Uniform Distribution," "Bulls-Eye or Fried-Egg," "Palimpsest," and "Offsite" models, the geoarchaeologist will have a better appreciation for the complex and variable construct that comprises an archaeological site.
Chapter II, "The Goals of Archaeological Survey," demonstrates how archaeological surveys can be designed for different objectives, with the choice of survey methods and sampling techniques tailored to the survey's purpose. How one goes about searching for sites determines the kinds of questions that can be answered from the data once the survey is complete. A survey designed to estimate the density of sites, for example, will not necessarily yield results useful for estimating the proportions of different types of sites.
Chapter III discusses methods for detecting archaeological sites and the factors that influence detectability. Methods include surface survey by visual inspection, subsurface surveys involving small excavations or augering, geophysical and geochemical surveys, and underwater surveys. Detectability is, in part, a function of intrinsic factors, such as the site's size, geometry, and physical properties, which affect its visibility or obtrusiveness. Other factors that influence detectability are extrinsic to sites, and can, therefore, be directly controlled by the archaeologist, including the intensity of survey effort, the resolution of survey coverage (i.e., transect or shovel test intervals), and the survey crew training, experience, and motivation.
Among the intrinsic properties of sites dealt with in Chapter III are postdepositional factors that affect the spatial patterning and preservation of archaeological remains. Geoarchaeologists will find the two pages devoted to this topic disappointingly brief, with few useful references to an extensive literature on the subject. On the other hand, most geoarchaeologists would benefit from the chapter's discussion of sampling grids. The spacing and orientation of core holes on grids, for example, can be explicitly designed, using statistical techniques, to optimize the probability of detecting buried phenomena of a given size. Although in Banning's examples the phenomena are sites or features within sites, in geoarchaeological applications, they might be geologic features, such as paleochannels on a floodplain or crevasses and splays on a levee.