Occupations, and Field Methods
Settlement pattern analysis is dependent on well-thought-out delineation of occupations, following the understanding that occupations should be basic units, along with individual artifacts and features, for studying both variability and change in landscape use. In turn, occupation delineation is dependent on employing field methods that accurately record the three-dimensional location of artifact samples at all scales, from assemblages to mounds to microartifacts. In order to support use of methods that have the greatest potential to provide appropriate data, it is important to examine recent field practices in the Lower Mississippi Valley (LMV) and suggest ways they can be improved.
Much of our understanding of settlement patterning in the LMV, and especially in the Yazoo Basin, is based on data recovered by the Lower Mississippi Survey (LMS) in the 1940s and 1950s (Phillips 1970; Phillips et al. 2003 ). The Survey recovered artifacts from 383 sites in 1940–1947 (Phillips et al. 2003 :219) and made a number of other collections, some from the same sites and some from newly recorded ones, in 1949–1955 (Phillips 1970:241– 242), for a total of 899 sites (Phillips 1970:961). The emphasis in collecting and analysis was on pottery. Aceramic assemblages, aside from those containing Poverty Point objects, were not prominent in the LMS collections. Projectile points were relatively rarely found (Phillips 1970:247–248; Phillips et al. 2003 :45) and neither Ford, in the 1951 volume, nor Phillips (1970) made an attempt to use them to order materials through time. Carr (this volume) comments on this issue.
Ford's settlement pattern analysis consisted largely of a study of different kinds of mound sites using seriation-based periods and five regions (Phillips et al. 2003 :309–344). Few villages without mounds, and no hamlet-sized or smaller habitation sites, were included in the study. This meant that the main distinctions Ford could make were between places with conical or flat-topped