In lieu of an abstract, here is a brief excerpt of the content:

the initial site plan consisted of a single plaza and that this unique site form emerges only in the middle part of the Mississippi period. Although these major sites clearly dominate the landscape, there are a number of large mound groups in the Yazoo that, although not of the size of Lake George or Winterville, are still not inconsiderable. Sites such as Arcola , Jaketown, Mayersville, Grace, Leland, Magee, and Haynes Bluff were all built or added onto during the Mississippi period (Phillips 1970). These communities support one or two large mounds arranged around a plaza. Most commonly the plan includes three prominent mounds forming a triangular plaza. Frequently smaller mounds bound the plaza or lie outside of the axis of the plaza. Across the river in southeastern Arkansas and northeastern Louisiana are contemporary large sites which, although quite large, still rank below the largest mound centers. The Transylvania and Lakeport mounds, for example, supported important late Mississippi period occupations dominated by shell-tempered ceramics (Hally 1972; Rolingson 1971). Farther south in the Natchez Bluffs region, there is a pattern of dominance by one or two major centers through time (Brain 1978). The Anna site appears to represent the largest community during the early part of the Mississippi period , while the locus of occupation shifted to the interior during the later part of the period and was centered on the Emerald mound site (Brown and Brain 1983). In virtually every regard, the Mississippian mound site plan is highly exclusionary. Historically, Lower Mississippi Valley mound groups had supported one or more major mounds. The trend through time was to add subsidiary mounds along the boundaries of the plaza, effectively increasing the degree of separation between the plaza and the outside world. Mound construction appears to emphasize access only through the plaza, with ramps facing this space and the slope of mounds becoming more pronounced on the outside of the plaza group. Few sites in the Lower Mississippi Valley appear to have supported deliberate structures for the exclusion of peoples from the mound group. Only at the Lake George site do we see the erection of a moat and palisade to bound the site. At Toltec, possibly Winterville, and even perhaps Routh and Fitzhugh, there may have been ditches and/or low walls surrounding parts of the site. While these features may have served to keep people out, they could also have functioned to help drain the sites as well (Kidder and Saucier 1991). Thus it is not possible to assign a single 146 Kidder function to the many features that could have served to inhibit access into mound groups. Although our survey data are patchy, it seems that Mississippi period house sites and hamlets were clustered into what could be broadly termed dispersed communities—that is, small settlements separated from other contemporary groups by as yet unknown amounts of unoccupied space. It is possible that these communities occupied space bounded by certain geographic features. For example, the historic Taensa “community” appears to have been largely con¤ned to the banks and immediate areas around a single oxbow lake (Jones and Kidder 1994; Swanton 1911). Presumably, these “communities” were centered around a small, usually single-mound, ceremonial center, which was in turn integrated into the larger polity by some relation to the occupants of the major mound sites. The relationship between the very largest communities and those of lesser size is unknown, but we can infer that the sociopolitical system was strongly ranked. Such a model is in keeping with ethnographically documented Mississippian chiefdoms elsewhere in the Southeast, but it is still not adequately proven in the Lower Mississippi Valley. It is worth noting that the nature of Mississippi period communities seems to change as one moves south in the Lower Mississippi Valley. In the northern Yazoo Basin, for example, Mississippi period communities are frequently large and densely occupied and manifest numerous square to rectangular houses. A good example of this pattern is seen at the Wilsford site in Cohoma County, Mississippi (Connaway 1984). Similar examples have been found, especially in the Clarksdale area (Connaway 1984; Phillips, Ford, and Grif¤n 1951; Starr 1984, 1991). Farther south, however, especially south of the Greenville-Greenwood line, Mississippi period sites appear to be structurally different. Nucleation of communities is less evident, the number of houses appears to decrease (on the basis of surface densities of daub), and the architecture appears to differ in a number of subtle regards (Brown 1985b). At some point in the...

Share