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

5 Middle Ceramic Period Earthlodges as the Products of Craft Traditions Donald J. Blakeslee From the beginning of archaeological research on the houses of the Middle Ceramic period (a.d. 1000–1400) in the Central Plains, the earthlodges of the historic village peoples were used both explicitly and tacitly as interpretive models. Thus, when Robert F. Gilder (1907) initiated excavation of the houses of the Nebraska phase early in the twentieth century, he assumed that the structures were circular and never bothered to determine their actual shape. It took the ¤rst professional archaeology in the region, done by Frederick Sterns (1915), to demonstrate that the houses were square. Similarly, when WPA archaeologists began excavating similar sites in Iowa, they ¤rst found what appeared to be circular remains of houses, and only after A. T. Hill showed them a better excavation method were they able to report square structures (Hotopp 1978). Still later, W. Raymond Wood (1952) used information on historic Hidatsa earthlodges to interpret variations in the number of center posts in Nebraska phase lodges. This reliance on the historic lodges is fraught with dif¤culties. To begin with, the number of detailed descriptions of how the historic lodges were built is quite meager. Gilbert Wilson’s (1934) monograph on the Hidatsa earthlodge is the only relatively complete account, reported by Buffalobird Woman, who owned the sacred knowledge required to be a Hidatsa architect. M. R. Gilmore’s (1931) account of how to lay out an Arikara medicine lodge was provided to him by two priests who differed in regard to the proper way to ensure the correct proportions within the lodge. Accounts for other Plains societies (e.g., Fletcher and La Flesche 1972; Welt¤sh 1965) are far more general and could not serve as guides to actually constructing a lodge. This makes it impossible to assess adequately the differences in styles of earthlodges from society to society. In addition to lack of detailed descriptions for building lodges, there is the additional problem of moving from historic descriptions of standing structures to what one can observe in the archaeological record. What archaeology provides, with metronomic regularity, are ®oor plans of excavated lodges and depths of house pits when they are present. What it seldom provides is any detailed information about the superstructure. Hotopp’s (1978) reporting on the kinds of wood used in wall posts and center posts in the Glenwood locality of western Iowa stands alone in that he not only analyzed the charcoal for species identi¤cation but did so within a project that excavated a series of houses. Elsewhere, wood identi¤cation has been exceedingly sporadic and incomplete. Loy Neff’s (1988) analysis of the construction of a roof of a house from the St. Helena phase, based on analysis of the daub from a burned structure, remains unique in Plains archaeology. Although he was able to determine the structure of the roof and to show that it was quite different from that of the reported Hidatsa earthlodges (although it did have a sod covering), no one has repeated his analysis, and we do not know whether the roof he found was a common type or rare. Unfortunately, earlier archaeologists tended to view roof ¤ll as something to be gotten through so that the materials on the house ®oor could be revealed, and many opportunities to determine how roofs were constructed have been lost (e.g., Kivett and Metcalf 1997:35, 38). Wall construction is also seldom addressed. Wall post pits are usually mapped, but what went on the wall posts, and how, is seldom addressed. Neff (1988:23–24) was able to show that the daub from the house he analyzed did not differ from the walls to the center of the house, so either the walls were made in the same fashion as the roof or the walls of that structure were not daubed. Given that the roof was made like a sandwich, with a core of thatch bordered with layers of bark, with sod on top and roof beams below, it is not very likely that the same technique could have been applied to vertical or steeply sloping walls. Therefore, undaubed walls were probably present on that house. A test pit in the wall of another house in the same site, however, yielded a large piece of daub that appeared to have been spread thickly over typical wattle—a set of parallel small poles with a single large pole at right angles to...

Share