A Late Prehistoric Platform Mound
in Western Arkansas

By Gregory Vogel

Mounds in the Arkansas River Valley
History of Investigations
Size and Shape of the Mound
The Artifacts
The Tunnels
The Historic Cemetery
The Gift Shop
Aerial Photographs
Mounds in the Arkansas River Valley
Conclusions and Further Questions
References Cited
Gallery of All Figures

Numerous Caddoan era mound sites are found along the Arkansas River and its tributaries. Spiro (34LF37), Norman (34WG2) and Harlan (34CK6) are the three largest and best known of these. Three primary mound types are recognized for this area, consisting of burial, structure, and platform mounds. Structure mounds are generally low, conical mounds erected on top of structures that appear to have been used as charnel houses or temporary burial locations. Burial mounds are conical, elongated, or multi-lobed features that are usually the result of numerous episodes of mound building. Grave goods are commonly associated with burials in these mounds.

Platform mounds are the largest earthen structures found in the Arkansas River valley. These are generally flat-topped or "truncated pyramids", built in multiple episodes. Platform mounds generally contain no artifacts or burials. Their internal structure can be quite complex, however, revealing numerous stages of mound construction. A typical sequence of platform mound construction begins with a "fill" stage of a few decimeters of soil, sometimes constructed of basket loads of highly contrasting matrix. Each fill layer was "active" for some time, with structures built on top and sometimes pits excavated into the fill and then backfilled. Fill layers were typically capped with a thin layer of burnt or compacted sediment, which was in turn covered by another fill layer, and so on. The actual internal structures of platform mounds are quite complex, however, so even this "typical" sequence is suggestive only. The similarities of area platform mounds that have been excavated lie in their complexity and expression of cyclical building more than in any particular sequence or progression of construction. This cyclical building pattern is clearly expressed at several platform mounds throughout the region (see Brown 1996:172-179 for a discussion of this patterning and platform mounds in general).

Sites that contain several mounds are termed civic ceremonial centers, and appear to have served as the location of important community ceremonies tied to regional political power. Brown et al. (1978) constructed a hierarchical typology of civic ceremonial centers which they termed first, second, and third echelon. Sites are considered first echelon if they contain at least one burial mound and an associated structure mound. Second echelon centers contain several structure mounds, at least one burial mound, and an additional platform mound. Third echelon centers are essentially large second echelon centers with the addition of a fourth mound type or other large architectural element. The additional mound type or architectural element is generally unique to the site, suggesting an "organizational discontinuity with the lower-order centers" (Brown et al. 1978:189). In the case of the Spiro site, the unique architectural element is the Craig mound with four conical burial mounds joined by earthen saddles.

Cavanaugh appears to be a typical platform mound for the region, except for its isolation from other mounds or from an associated residential area. Because it is a platform mound alone on the landscape, isolated from closely associated mounds, it does not fit into Brown et al.'s (1978) echelon system.

Figure 23 shows a shaded relief representation of Cavanaugh Mound compared to some other large Caddoan platform mounds in the region. The Craig Mound at Spiro is included as an aid in visualizing the scale of these earthworks; Craig has been reconstructed and undoubtedly numerous readers have visited the site and have a strong and personal impression of how large that structure is. For purposes of this comparison, Craig Mound was based on a topographic map of Spiro (Peterson 1989:2, Figure 1); Brown Mound was based on the description by Orr (1946:230); Skidgel was based on the "primary" flat-topped mound (before a final layer converted the mound to a conical shape), from Brown (1996:177, Figure 1-56); Harlan Unit 7 was based on a detailed topographic map (Bell 1972:186, Figure 21); and the reconstruction of Cavanaugh was based on the 2004 mapping efforts described in this paper.

Figure 23. Shaded relief representations of several
large mounds in the Arkansas River valley.

These representations are not meant to be exact; the edges and heights of the mounds are difficult to reconstruct even for those that have been fully excavated. From the accumulated effects of centuries of erosion and natural soil forming processes combined with decades of plowing, looting, and other historic disturbances, the mounds have likely been altered in ways we cannot reconstruct, and any determination of a mound's exact shape and size is only an approximation. Still, the general outlines and sizes may be compared within reasonable limits of uncertainty. By almost any measure, Cavanaugh is a large mound, and certainly ranks as one of the most impressive in the Arkansas River valley and adjacent regions.