The Archaeology of the Caddo

7. Exploring Prehistoric Caddo Communities through Archaeogeophysics

177 7. Exploring Prehistoric Caddo Communities through Archaeogeophysics chester p. walker and duncan p. mckinnon Introduction Our understanding of how the prehistoric Caddo Indians organized their communities across natural and cultural landscapes has been heavily influenced by two primary source images. These are the 1691 map of the Upper Nasoni community drawn by an anonymous scribe from the Domingo Terán de los Ríos expedition to the Red River (in what is now northeast Texas, see Sabo, this volume), and a series of late nineteenthcentury historic photographs taken by William S. Soule of the Smithsonian Institution. The Terán map depicts a dispersed Caddo community of partitioned farmsteads spread out over several kilometers along the Red River. Each partitioned farmstead consists of one or more large circular thatchedcovered dwellings with open-air arbors, storage structures or granaries, and ramadas. Located on the western periphery of the community is a large temple mound. A vacant area surrounds the mound, with a templo or temple on the mound and an open-air structure at the base. The community is presumed to be the Hatchel-Mitchell-Moores archaeological site complex located in Bowie County, Texas (Wedel 1978:14; see Perttula 2005). The Soule photographs were taken between 1868 and 1872 of the farmstead of Chief Long Hat near Binger, Oklahoma, complete with beehive-shaped house clusters, outdoor arbors, and storage shelters. The combination of farmstead structures in the Soule photographs is quite similar to the combination of structures in the Terán map farmsteads , despite the two centuries that separate the two sources. 178 Walker and McKinnon These two sources are often conceptualized as ethnographic examples of Caddo community spatial organization across the landscape (cf. Lockhart 2007; Schambach 1982a; Trubowitz 1984). However, they have been overused in investigations and characterizations of community organization at Caddo archaeological sites, in part as a result of the scarcity of primary archaeological data of Caddo community organization, as they only represent a synchronic view of Caddo community spatial organization. Currently the most completely excavated Caddo community in the Caddo archaeological area may be the Oak Hill Village site (41RK214) in east Texas (Perttula and Rogers, chap. 8, this volume; Perttula and Rogers 2007). Prior to the remote sensing or archaeogeophysical “revolution ” in Caddo archaeology (Perttula et al. 2008), the excavations at the Oak Hill Village site provided the most complete archaeological view of the organization and evolution of a prehistoric Caddo community. Now, however, recent archaeogeophysical investigations at a number of Caddo sites have obtained primary data sets of geophysical features that allow a more detailed consideration of community organization at a landscape scale as well as a better appreciation of the character and intrasite distribution of Caddo architectural features. It is becoming increasingly economically feasible to use archaeogeophysics as a complementary investigative tool in the study of Caddo community spatial organization across the landscape. Kenneth Kvamme (2003) was the first North American archaeologist to urge a shift from using archaeogeophysics exclusively as a tool for locating cultural anomalies to an approach where it is holistically integrated as an investigative strategy with archaeological investigations and interpretations. This paradigm shift concerning how to approach the study of the overall spatial structure of prehistoric Native American communities is one that is unlikely to be achieved solely through modern excavation strategies, if for no other reason than cost. With expansive archaeogeophysical investigations of large coverage areas at sites, it is often possible to reveal the spatial arrangements of individual structures in addition to various attributes about their geometric shape, spatial orientation, and interrelationships. Furthermore, it is often possible to gather archaeogeophysical data suitable for interpreting the character of interior features within a structure, such as fire hearths, storage pits, or burials underneath house floors. The ability to 179 Exploring Prehistoric Communities acquire information on the settlement structure of prehistoric American Indian sites over large coverage areas, compared with slow-paced, expensive , and spatially limited excavations, has affected the way archaeology is being conducted today in North America and will have an even larger effect in the future. In this chapter, we discuss the archaeogeophysical findings from the George C. Davis (41CE19), Hill Farm (41BW169), and Battle Mound (3LA1) sites (see figs. 1-2 and 1-3) to highlight the various architectural attributes of Caddo communities that can be defined using landscape archaeogeophysical methodology. At the most basic and perhaps analytically most powerful resolution, the use...