Feasting with Shellfish in the Southern Ohio Valley: Archaic Sacred Sites and Rituals

6. Ohio River Valley Shell-bearing Sites: Villages?

CHAPTER 6 OHIO RIVER VALLEY SHELL-BEARING SITES: VILLAGES? Prior to the late 1980s, the shell mounds that are the focus of this study were assumed to be villages, lived in at least half of each year. The quantity of debris and the size of some of these accumulations were just two of the lines of evidence. The types of artifacts and features found in them were even more significant in the village logic. Places such as Carlston Annis are comprised of fire-cracked rock, sandstone, ash, charcoal, shells, fractured bones, flakes, cobbles, and various bone and stone tools. DeWeese, a neighboring shell heap, is estimated to have 17 percent shell, 6 percent sandstone, 42 percent pore space, and 34 percent matrix (Crothers 1999:206).The features suggest small cook fires, earth ovens, and storage pits—all the stuff of domestic life. The presence of shells was another expectation of villages, as they were thought to be food debris only. There are very good reasons for interpreting these Archaic shell-heap sites as more than mere habitation areas, and the shells as representing something more than daily food collected and amassed in family-sized portions over thousands of years. The evidence I have cited in the past for these ideas were 1) the disconnect between the potential for this type of site to be found on every river in the eastern United States yet their appearance on only a few, 2) the accumulation of shells into piles instead of lineal spreads mimicking the bed, 3) accumulation of shells under difficult circumstances (such as carrying them up a bluff), 4) the often voluminous strata of clean,whole shells,5) the occurrence of paired valves, which indicated little disturbance of parts of these heaps, as well as wasted food, 6) the large number and high density of burials, and 7) the death/rebirth symbolism that is known to be accorded shells later in time. I have discussed each of these points of contention in previous publications (e.g., Claassen 1991, 1996), and some of them in Chapter 86 Ohio River Valley Shell-bearing Sites: Villages? 3 in this study. There are other aspects of these heaps that raise doubts about their assumed function and should be added to the list: 8) the volume and stratigraphy of the heaps, 9) the quantity of dog burials, 10) the numbers and proportions of artifacts, 11) the type of grave goods, 12) the number of violent deaths, and 13) the characteristics of the initial burials. Each of these new topics will be discussed in some detail at this time. Heap Construction and Volume The excavators and reporters of these sites referred to them as mounds, yet assumed that the mound shape was inconsequential. Bringing shellfish to a central spot, however, when this resource occurs linearly in beds (see Chapter 3) suggests otherwise. Both construction evidence— layers of clean, whole shells—and the monumentality of these sites have received little attention and deserve more. Clean Shell Layers Recent attention has been called to the intentional construction of Middle Archaic mounds using shells at several sites on the St. Johns River: Harris Creek (Aten 1999), Hontoon Island North (Randall and Sassaman 2005), and Hontoon Dead Creek (Sassaman 2005:98). Loose, clean, whole shells are also the principal constituent of Rollins, Fig Island, and many more shell rings and arcs (Ken Sassaman,personal communication, 1994). The mounds of marine shell at Horr’s Island are also said to be intentionally constructed (Russo 1994a, b). Given that shell was being used as construction material in Florida by 7,000 years ago, we must take a closer look at the midcontinent shell heaps for intentional construction clues. Layers consisting of only shell suggest little disturbance and building episodes. Such layers existed at Mulberry Creek, with four alternating pairs of shell and sand layers (Webb and DeJarnette 1942:264), as the basal layer of Long Branch (Webb and DeJarnette 1942:180), at Perry, where“layers of nearly pure shell. . . were separated by more compact layers containing silt” (Webb and DeJarnette 1942:60),and at the base of Bluff Creek (Morrison 1942) (Figure 6.1). The O’Neal site in Pickwick Basin had nine pairs of thick, clean, shell lenses alternating with ashy midden (Webb and DeJarnette 1942:Figure 37). Ohio River Valley Shell-bearing Sites: Villages? 87 Watson (2005) and Cridlebaugh (1986) are comfortable with shell as construction material. For...