Introduction

In the great civilizations of the Old and New Worlds, particularly where domesticated livestock provided the primary source of meat, animals were commodities, leveraged in economic and political transactions by those in power just as other resources were (deFrance 2009). Autocratic rulers controlled the goals and scale of production in centralized economies, and could divert livestock resources as needed to support the goals of state, as well as control access to animal products in a manner that ensured that the social and political order was reflected by patterns of consumption. In the late prehistoric Mississippian societies that populated the southeastern United States from about AD 1000 to 1550, meat was hunted, trapped, fished, and gathered, precluding animals from political strategies that required sustainable surplus production, herd management, or livestock movement.

The Mississippian elite’s impact on production and consumption of animal resources is best regarded as a part of the broader framework of political authority that relied variably on local efforts to forge social solidarity and on a complex system of cosmologically related symbols, ritual responsibilities, and access to a regional prestige-goods economy (e.g., King 2003). Faunal remains from Mississippian sites attest to variable levels of elite provisioning, episodes of intense labor mobilization, an apparent preferential access to certain cuts of meat and certain rare or symbolically charged species, and the use of certain animals in ritual activities (e.g., Jackson and Scott 2003; Pauketat et al. 2002; Scott 2005).

DeFrance (2009) identified three broad themes in discussions of the impact of status distinctions on the zooarchaeological record: (1) evidence for the role of animal resources in the political economic relationships forged between elite and non-elite; (2) how differences in access to certain animals reflect the social differences of consumers; and (3) the role of animals in the expression of ideology and the performance of ritual as these relate to the differential distribution of social position and political power. Mississippian zooarchaeology has produced evidence to illustrate each of these themes, although most research has dealt with the first two. Therefore in this chapter, and in the hope of encouraging further research into this dimension of Mississippian animal use, I briefly examine evidence for political-economic relationships and how social standing may have determined access to certain faunal resources, and then consider how ritual manipulation of animals may affect elite faunal samples.

Meat in Mississippian diets was provided by deer, turkeys, and where abundant, waterfowl and fish, supplemented by a wide range of medium and small mammals, birds, reptiles, and amphibians (e.g., Compton 2009; Kelly 1997; Smith 1975). Deer was the only sustainably exploited large game animal, with much smaller contributions made by other large-mammal taxa, such as bear or elk. The relative contributions of taxa varied from region to region and household to household.

Faunal Resources and Mississippian Political Economy

Bogan’s (1980, 1983) analysis of material from the Toqua site, a Mississippian center in eastern Tennessee and Scott’s (1983) analysis of fauna from the Lubbub Creek site, a single-mound polity center on the Tombigbee River in west central Alabama (Figure 5.1) provided the earliest consideration of political-economic ramifications for Mississippian faunal patterning, using the frequency distribution of deer elements to argue that elite were provided venison hunted and field-butchered by non-elite—conclusions based on an underrepresentation of low-value anatomical units. Scott demonstrated that the pattern at Lubbub Creek was due to provisioning rather than simply field butchering by comparing it to the assemblage from a small farmstead, the Yarborough site, which displayed an excess of low-value anatomical units (skulls and lower limbs) (Jackson and Scott 1995a; Scott 1982).

Figure 5.1. Sites discussed in the text: (1) Cahokia, IL; (2) Toqua, TN; (3) Etowah, GA; (4) Moundville, AL; (5) White Site, AL; (6) 1TU66, AL; (7) Lubbub Creek, AL; (8) Yarborough, MS; (9) Winterville, MS; (10) Lake Providence, LA; (11) Crenshaw, AR; and (12) Tom Jones, AR.

Using deer-element representation, similar evidence for political-economic relationships between rural producers and elite has been documented for mound and other elite contexts at Moundville in Alabama (Jackson and Scott 2003, 2010; Michels 1992), and the Crenshaw site in southwest Arkansas (Scott and Jackson 1998). Crenshaw differs from both Lubbub Creek and Moundville in having a preponderance of hindquarters, in contrast to greater representation of shoulders from the Alabama sites. At Cahokia, Kelly’s (1997) analysis of samples from elite and non-elite contexts demonstrated that regardless of social position, low-utility elements of deer were poorly represented, suggesting provisioning of the general populace. Samples from elite contexts were dominated by high-utility cuts (mainly hindquarters) while those from non-elite contexts were mainly comprised of moderate-utility cuts (essentially shoulders) (Kelly 1997:80).

A second component of Mississippian political economy is the capacity of the elite to mobilize labor, including intensive procurement for particular events. Labor mobilization is manifest in mound and other construction projects conducted at the centers. It is also manifest in meat procurement to supply feasts that were part of these events. Whether feasts were aimed at solidarity-building efforts or restricted to elite participants (e.g., Blanton et al. 1996; King 2003), they required the capacity of hosts to draw upon the labor of hunters to supply the necessary meat.

Provisioning feasts on a grand scale is evident at Cahokia, the most dramatic example coming from refuse that was generated by feasts during mound construction on the main plaza and that was then deposited in a large borrow pit now beneath Mound 51 (Pauketat et al. 2002:258). Deer remains dominate the more than 10,000-specimen sample (Kelly 2001; Pauketat et al. 2002). Axial and upper limbs are well represented, with low-utility elements extremely poorly represented, a distribution that supports the interpretation of the fauna in the feature fill’s as the remains from feasting. Minimal bone breakage indicates processing for meat. Extrapolating from excavation volume to total pit size, Pauketat (2010:109) suggests that the sub-Mound 51 feature might contain several thousand butchered deer carcasses. Overall, taxonomic diversity is low, but those species present are abundant, suggesting intensive procurement. The assemblage includes large numbers of birds, especially swans and prairie chickens, and relatively few fish. Swan-wing elements are absent, suggesting these were removed to produce curated fans (Kelly 2001). Prairie chicken is typically only found in elite contexts (Kelly 1997), pointing to the unique nature of the pit contents.

Two features containing feasting refuse have been documented at Winterville, in northwest Mississippi (Kowalski et al. 2009). One large probable borrow pit dating to the commencement of mound construction is filled with several refuse-rich strata, each separated by lenses of clean fill. Deer and fish dominate the faunal assemblage, and the volume suggests intensive procurement efforts. Deer remains are limited to meat-bearing anatomical units. Fish are large-sized and include taxa representing the main Mississippi River channel. A much later feature, dating near the end of Winterville’s occupation, has a much different character, but based on the apparent intentional discard of large fineware serving vessels, the presence of nonlocal ceramics, and botanical materials that include a tobacco seed and more than 1500 clasping coneflower seeds (Rudbeckia amplexicaulis) (Flosenzier 2010), it is interpreted as evidence for small-scale ritual feasting, possibly hosted by the resident of a nearby mound. Deer, swamp rabbits (Sylvilagus aquaticus), and fish, mainly bowfin and catfish (chiefly bullheads, Amieurus sp.), dominate the faunal assemblage. The rabbits and the fish suggest local procurement efforts (VanDyck 2010). Deer elements include meat-bearing elements but also primary butchering debris, suggesting local procurement. Low levels of bone breakage suggest meat consumption rather than the more intensive processing. Whether the particular meats chosen for the feast have particular ritual meaning is not clear and the particular combination of swamp rabbits, bowfin, and catfish may simply reflect localized procurement.

Faunal Resources and Social Distinctions in the Mississippian World

The second theme is how social differences in access to resources engendered socially distinctive faunal assemblages. In an early consideration of this issue, Jackson and Scott (1995b:107) suggested five dimensions that distinguish elite Mississippian faunal refuse: high sample diversity; an abundance of prime cuts of meat, particularly when considering venison; greater representation of rare taxa; low frequency of butchering debris; and finally, a higher proportions of birds. Sample diversity may be a product of simple access to a more varied diet, but one must consider also the potential implications of the wide range of symbolic meanings assigned to birds, as well as rare or dangerous taxa that might engender restrictions in their consumption or use. Further, it is sometimes difficult to distinguish between animals used for meat and those employed in non-culinary activities, such as production of craft goods or use in ritual.

Mound-related samples from Moundville demonstrate that elite were accorded access to a greater variety of animal resources, including a range of taxa that do not ordinarily occur in non-elite assemblages (Jackson and Scott 2010). Sample sizes from the two mounds are quite different (Mound Q NISP = 9,628; Mound G NISP = 3119 (Jackson and Scott 2010:Table 8.2) but both show considerable richness. Mound Q produced forty-six taxa (fifteen mammal, ten bird, eight reptile, one amphibian, and twelve fish), while the much smaller Mound G sample included thirty-two taxa (thirteen mammal, nine bird, three reptile, and seven fish). In addition to ample venison, residents of both mounds enjoyed large amounts of turkey as well as passenger pigeon, ducks, geese, other waterfowl and other birds, and a variety of fish. Birds comprise a particularly large proportion of both samples: 17 percent of NISP from Mound Q and 16 percent from Mound G. Samples include a variety of small and medium mammals, numerically dominated by squirrels, and bear is present in both. Deer long-bones were cracked open to extract marrow, but otherwise breakage was minimal, indicating the absence of intensive processing for grease.

Distinctions between the two assemblages may indicate differences in status or in how animals were incorporated into mound-summit activities. On Mound Q, where artifact and other evidence suggests intensive crafting and ritual (exotic raw materials, sandstone saws, paint-mixing bowls, paint palettes, and pieces of cut human bone) (Knight 2010), there is a significant variety of furbearing animals. Mound G, lacking evidence for craft production, but distinctive in its higher ratio of fine serving vessels and bottles (Knight 2010), produced a number of animals unique for Moundville, including gray fox, sandhill crane, peregrine falcon, shark, and bison. Peregrine falcon is a possible referent to the falcon-warrior motif in Southeastern Ceremonial Complex (SECC) art (King 2007), whereas bison, shark, and sandhill crane were not locally available and point to far-flung connections.

Kelly’s (1997:82) American Bottoms research documents a number of distinctions between elite and non-elite faunal use during the Stirling phase (AD 1100–1200), including a significantly higher representation of birds, in particular turkey and prairie chicken, in elite faunal samples. Twelve bird taxa are represented in elite samples from Tract 15A, compared with only four taxa produced from non-elite contexts in the ICT-II tract (Kelly 1997:Table 4.4). Similarly more small and medium mammals are present in elite samples (six taxa from Stirling phase contexts at Tract 15A versus two from ICT-II).

Ritual Use of Animals by the Mississippian Elite

It is not always clear that greater species diversity can or ought to be attributed to variety in the diet, as distinct from other kinds of animal uses, (e.g., Moundville’s Mound Q). However, there are instances where species representation points strongly to the probability that particular animals, because of associated symbolic meanings, became part of the archaeological record as a consequence of their use in ritual. Although ritual knowledge was likely to have been parceled out along several social dimensions, it is clear the elite maintained and perpetuated their access to an important portion of this body of knowledge and to a range of material symbols. They did so through exchange and production that underscored the supernatural roles of the elite segment of society. The special relationship of the elite to cosmological forces provided legitimacy to political authority and social inequality (e.g., Barker 1992). Ritual performance was both a responsibility to one’s followers and a strategy for harnessing cosmological forces to increase one’s political power.

Ethnohistoric and ethnographic accounts indicate the importance of animals as representatives of different parts of the earthly and cosmic realms of Indian conceptualizations of the universe. These associations were often inferred from particular characteristics of certain taxa; simultaneously, as determined by the realms they represented, the associations conferred supernatural qualities on the elite social stratum (Hudson 1976; Jackson and Scott 1995b). There were likely to have been multiple dimensions on which these qualities were measured: pure–polluted, order–disorder, weak–strong, and harmless–dangerous, for example. To be able to project this system of beliefs about the animal world back in time is at least partially borne out by the choices of animals depicted in Mississippian art and iconography (e.g., King 2007; Reilly and Garber 2007; Steponaitis and Knight 2004; Sullivan 2007). Patterning in the zooarchaeological record might partially reflect this system of beliefs and the extent to which it propagated rules of proscription that affected access to certain animal taxa as well as the uses of these animals in ritual performance. Mediating against clear-cut patterning, ritual behavior was not restricted to elite arenas, and there is evidence of animal use in ritual activities from non-elite contexts (Jackson, Scarry, and Scott 2009; Maxham and Scarry 2009).

With that said, distinguishing the quotidian from the symbolic can be difficult to operationalize. Animal completeness, specific portions, presence of other artifacts or materials interpreted to have symbolic meaning, and animal associations with burials provide some criteria for identifying ritual use of animals (deFrance 2009:135). Treatment of material (burning or distinctive butchery, for instance) and context also provide potential clues. Mortuary inclusions or caches would seem to be the most recognizable instances of ceremonial animal use. However, it is possible that refuse from the use of animals in rituals might be commingled with everyday trash, as could be the case for Mounds G and Q at Moundville, if some or all of the rarer taxa were used for ritual purposes rather than simply sustenance.

Mound 34 at Cahokia provides an example of a faunal assemblage that is the product of ritual activities emphasizing birds, though incorporated into middens that also included feast-related refuse (Kelly n.d.). Mound 34, located 400 meters east of Monks Mound, was a focus of excavations in the 1950s (Kelly et al. 2007). Along with spatulate celts, real and chert-effigy sharks’ teeth, wooden-bowl fragments, negative-painted pottery, copper, and engraved marine-shell cup fragments, is a faunal assemblage that includes fifty-eight bird taxa. Waterfowl comprises three-fourths of the bird NISP, of which 20 percent is swan. Unusual and rarely found birds include a variety of hawks and eagles (including peregrine falcon, golden eagle, and kestrel), Carolina parakeet, ivory billed and other woodpeckers, and four different owls. Ethnohistoric references to hawks and eagles, owls, and woodpeckers, as well as their depictions in Mississippian art, lead Kelly to conclude that the birds from Mound 34 were involved in rituals that resulted in the eventual incorporation of war symbolism into Mississippian religious iconography, and ultimately into SECC art that included bird (or bird-warrior) symbolism as a major theme.

Kelly and Kelly (2007) tracked the distribution of swans in American Bottoms. Despite the fact swans could produce a relatively large amount of meat, they do not regularly occur in subsistence refuse in the Mississippian period. Instead they are found mainly at Cahokia and at three subsidiary centers. At Cahokia, they are found in the sub-Mound 51 pit, in the vicinity of Mound 34 and in the ICT-II tract. Although the first two have already been interpreted here as ritual-related, the ICT-II tract was a non-elite residential area. There, the vast majority of swan remains came from a single Lohmann phase (AD 1050–1100) feature (NISP = 217) and 92 percent of the elements were from wings, suggesting wing feathers were plucked and the wings discarded. By the Stirling phase, the height of the Cahokian reign (AD 1100–1200), the only examples of swans outside of Cahokia are wing elements and a high proportion of these were ringed and snapped or otherwise cut. Kelly and Kelly surmise that the distribution of swans was highly regulated by the elite.

Mound excavations at Lubbub Creek produced several bird taxa likely related to ritual use, including cardinal, mockingbird, Carolina parakeet, crow, blue jay, and merlin, all only found in mound context (Scott 1983). Plumage colors of the cardinal, blue jay, and crow were associated by southeastern Indians with the cardinal directions, and the merlin is a close relative of the peregrine falcon. Bear and bobcat were found only in the mound as well (Scott 1983). Finally, half the rodent bones from the entire site were collected from mound contexts, though the mound sample amounted to only 15 percent of the entire site assemblage. At the time Scott (1983) interpreted the rodent remains as commensal taxa, drawn to the mound by (inferred) large corn granaries.

Feast provisioning may simultaneously provide evidence of the political economy and symbols employed in rituals. The Lake Providence site, a terminal Coles Creek mound center in northeast Louisiana (ca. AD 1200) (Weinstein 2005), produced feasting refuse from middens associated with one of the site’s four mounds. The deposit produced a large percentage of local fineware ceramics, local ceramics that mimic motifs found on American Bottoms Ramey Incised, and some 350 sherds of vessels from the American Bottoms (Weinstein 2005). Nearly 20 percent (N = 4800) of the associated fauna (NISP = 24,000) are the bones of squirrels (Scott 2005), a proportion to which no other Lower Mississippi Valley site comes even close. Use and then discard of local ceramic fineware, along with Cahokian imports and a meal that included a main course of squirrels, seems to qualify as a ritual feast, occurring during an interval in the Lower Mississippi Valley described as the Cahokia Horizon (Williams and Brain 1983). Different interpretations are possible, but the intersection of the largest sample of Cahokian ceramics outside the American Bottoms and a feast with a main course of squirrels is unlikely to be coincidence. Aside from the obvious demonstration that the Lake Providence elite could command an astonishing display of procurement effort for the feast, there may be a more specific connotation (Scott 2005). Among the historic Chickasaw there was an appointed office given the name Fane Mingo, which translates as Squirrel King. The Fane Mingo served a diplomatic function in interactions within and between tribes. Perhaps a symbolic connection between squirrels and diplomacy had its origins several centuries earlier, reflected in the consumption of squirrels in diplomatic feasting between foreign nations. The Lake Providence case suggests that zooarchaeologists must be sensitive to the possibility of deployment of specific foods due to their symbolic meanings for specific political interactions.

Two Caddoan ceremonial centers in southwest Arkansas add additional insights into ritual manipulation of animal remains. The assemblage from Crenshaw has been reported previously (Jackson and Scott 1995b; Scott and Jackson 1998), but its relevance to the discussion warrants review. Excavators at Crenshaw uncovered a ritual deposit of more than two thousand deer frontals with attached antlers (Schambach 1996). Other taxa included a great horned owl, unmistakably an avian analogy for deer. Adjacent to this “deer temple” is the residence of a ritual specialist whose faunal diet belied an elite status. Excavations produced 12,000 specimens from the structure deposit and another 4,000 from adjacent deposits. Eighteen mammal taxa are present, including bear and cougar. At least fifteen bird species are represented, including a disproportionately large amount of passenger pigeon, compared to the rest of the site. Other unusual birds include woodpeckers, cuckoo, blue jay, mockingbird/thrasher, blackbirds, and two unidentified small passerines, a suite suggesting ritual use rather than simply consumption.

In contrast to other Caddo sites on the Red River where fish amounts to between 20 and 40 percent of NISP, fish contributes less than 3 percent (Scott and Jackson 1998:24), a fact attributable to the dominance of deer in the assemblage. Structure-floor deposits produced evidence of ritual behavior, including pipe fragments, native copper beads and other ornaments, freshwater-pearl beads, marine-shell beads, finely carved bone pins, and seventy-three human teeth representing at least ten individuals, sometimes attached to bits of mandible (Schambach 1996; Scott and Jackson 1998:4). One last but potentially important aspect of the structure’s faunal remains are 379 bones of moles, shrews, pocket gophers, rats, and mice. The original analysis considered their presence to be commensal, based on the significantly lower proportion of burned specimens compared with squirrels and rabbits (Scott and Jackson 1998:11). Like the Lubbub Creek mound sample, this would seem to be a reasonable interpretation.

Recent excavations at the Tom Jones site, also in southwestern Arkansas, provide further evidence for deer-related ritual in the Caddo area (Lockhart 2010; Schambach 2003), and intriguing indications of ritual use of other taxa. The Tom Jones site includes a large temple mound and five house mounds. Unlike other Caddo centers in the region that are situated along major rivers or their tributaries, Tom Jones is located in a remnant prairie that is perched on the watershed divide between the Red River and Ouachita River watersheds (Lockhart 2010).

Excavations exposed four purposefully burned mound structures that were then covered with a new mound mantle. Three of these structures are elite mound-summit residences, whereas the fourth is a structure that was built at the foot of the large temple mound and is interpreted as a ceremonial cookhouse, based on the ceramic inventory of twenty-four vessels destroyed by the fire. Meals prepared there were presumably consumed on the mound summit. In addition to purposeful burning and immediate burying, the structures shared the ritual placement of deer scapulae on each floor. For each of the samples, scapula was the most frequent deer element. In the case of the cookhouse, five scapulae were included in this final offering, one of which was placed under an overturned vessel. Excavation produced an NISP of approximately 6,295 specimens from the cookhouse built into the side of Mound A and another 958 from the domestic structure on the summit of Mound B (Jackson 2011).

Despite their modest size, the samples from both the cookhouse and residential floor display considerable diversity, including not only the expectable suite of deer, squirrel, raccoon, and turkey, but also duck, quail, passenger pigeon, pileated or ivory-billed woodpecker, flicker, and mourning dove, plus several unidentifiable passerines. Given the site’s upland setting, a surprising variety of fish is present as well, including main-channel species such as buffalo fish, freshwater drum, pickerel, and white bass. Faunal assemblage characteristics of the two samples clearly exhibit the profile of elite consumption.

Of particular interest to the discussion of ritual animal use are the small and very small mammal remains in the samples from A and B. In the Mound B floor sample, squirrel is considerably more plentiful (NISP = 67) than rabbit (NISP = 9) despite the expectable abundance of rabbits in a prairie setting. Because the bones come from a floor deposit, it is possible that sweeping would be more likely to collect larger bones, while missing those of squirrels and smaller animals. Indeed there is an even larger number of microfaunal remains (small rodents, voles, etc.; NISP = 71, seven percent of total NISP).

The larger sample from the Mound A cookhouse also includes a significant number of squirrel bones: nearly 850 specimens were identified as fox squirrel, gray squirrel, or tree squirrel (Sciurus sp.), roughly 13 percent of the total NISP. Unlike the Mound B sample, only thirty-four microfaunal specimens were recovered from the cookhouse floor, although the sample was seven times larger. Because it is hard to imagine that a cookhouse would have been less attractive to rodents and other small mammals than a domicile, two things seem apparent. First, it is at least possible that microfauna from Mound B are there because of human action. Second, the preponderance of squirrels, particularly in the floor refuse of a cookhouse that served activities on the temple mound summit, suggests that Tom Jones offers a second example of ritual squirrel-consumption like that documented at the Lake Providence site.

Pohl (1983) has documented the common use of small animals, in particular reptiles and amphibians, in Mayan ritual contexts. Building on Pohl’s observations, Claassen (2005, 2007) suggests that the small size of animals in ritual is symbolic of the beginning or distant time, transformed in spatial terms to distant places where one’s view of things appears small. Claassen’s own research focuses on Archaic-era ritual use of cave and rockshelters in the Southeast, making the case that the presence of small taxa in cave deposits represents ritual offerings. Ritual on the Mound B summit is already implied by the presence of a large woodpecker, mourning dove, flicker, and three unidentified songbirds, and the ritual placement of deer scapulae during the decommissioning of the structure. If the abundance of microfauna is a function of ritual, the same may be true of the large microfaunal representation in the Crenshaw religious specialist’s domicile, where other lines of evidence point to ritual performance. A third possible example to which this alternative interpretation may apply is the microfaunal assemblage from mound deposits at Lubbub Creek. In contrast to the high concentrations of microfaunal remains in these ritual contexts, the samples from Moundville’s mounds G and Q had notably scant microfaunal specimens (N = 7 for Q, and N = 3 for G) (Jackson and Scott 2010:347). Although maize-storage differences or general housekeeping may explain the contrast, further consideration of the role of microfauna in Mississippian ritual contexts deserves exploration.

With regard to the numerical dominance of squirrels in the cookhouse sample, the historic diplomatic symbolism of this creature noted earlier may be relevant. The Ouachita River and Red River valleys were home to clusters of Caddoan settlements and mound centers each probably representing independent polities. The Tom Jones site is uniquely situated on the watershed divide between these territorial units, and as such it is at least possible that its function was in part in the arena of intersocietal relations, serving as an intermediary between these distinct riverine-oriented social entities.

Conclusions

Animal use by the Mississippian elite was conditioned by a complex matrix that included political authority and negotiation, economic relations that bound elite to commoner and elite to elite, kinship, proscriptive rules, and ritual performance. The product of this matrix is elite faunal assemblages that differ sufficiently from those of commoners to reinforce archaeological interpretations of status differences based on other categories of artifacts, architecture, and use of community space (also see Sharpe et al., chapter 4, this volume). There is, however, no single suite of animal resources that can be isolated and used to demarcate a single “Mississippian elite diet.” There is regional variation that can only be understood by careful comparison of assemblages from elite and non-elite contexts. Moreover, there is subtle variation in faunal refuse from different elite contexts, offering opportunities to evaluate differences among elite households related to economic support and distinctive animal needs as defined by, for instance, group memberships, particular craft or other production efforts, the size and scale of hosted meals, and the rituals that underpin their political or social office.

The last of these has been explored by zooarchaeologists the least. As the emerging research on SECC iconographic art is showing, a significant underpinning of political power in Mississippian chiefdoms is provided by artistry that refers to concepts of cosmology and myths of supernatural beings in human and animal form. If production, exchange, and possession of SECC paraphernalia served as a display of the elite’s connections with the cosmic forces that legitimized political authority, maintaining or influencing those connections must have required the performance of rituals. With the present evidence provided by SECC iconography, dramatic assemblages such as from Cahokia’s Mound 34 (Kelly et al. 1997), and historic references (Hudson 1976), we should expect certain animals—such as swans, woodpeckers, falcons, snakes, or cougars—to be found in ritual contexts. But there is no a priori reason to believe that the list of possible taxa incorporated into ritual performance was limited to those depicted in art or described in ethnohistorical accounts. Indeed, assumptions, such as small animals being commensal, may mask important evidence for elite manipulation of the animal world in their efforts to retain their special and possibly quite individualized relationships to the supernatural.

Through careful consideration of context, species composition, anatomical representation, bone modification, and associated artifacts, zooarchaeologists are poised to advance our understanding about how, for particular cases, the political, social, and ideological dimensions of Mississippian elite behavior contributed to the distinctive composition of their faunal refuse.