Patterns of Elite Faunal Utilization in Moundville, Alabama H

Home , Abo (historic place), Black drink

The University of Southern Mississippi The Aquila Digital Community

Faculty Publications

7-1-2003 Patterns of Elite Faunal Utilization in Moundville, Alabama H. Edwin Jackson University of Southern Mississippi, [email protected]

Susan L. Scott University of Southern Mississippi

Follow this and additional works at: https://aquila.usm.edu/fac_pubs Part of the Anthropology Commons

Recommended Citation Jackson, H. E., Scott, S. L. (2003). Patterns of Elite Faunal Utilization in Moundville, Alabama. American Antiquity, 68(3), 552-572. Available at: https://aquila.usm.edu/fac_pubs/8247

This Article is brought to you for free and open access by The Aquila Digital Community. It has been accepted for inclusion in Faculty Publications by an authorized administrator of The Aquila Digital Community. For more information, please contact [email protected]. PATTERNS OF ELITE FAUNAL UTILIZATION AT MOUNDVILLE, ALABAMA

H. Edwin Jackson and Susan L. Scott

In recent years, zooarchaeological research has begun to examine the roles of animals aspart of the suite of symbols employed in the ongoing social, ceremonial, and political dynamics ofprehistoric cultural systems. In the southeastern United States, studies of late prehistoric Mississippian chiefdoms have documented differences in species composition and meat cuts associated with particular social contexts of consumption-for instance, ceremonialfeasting vs. private meals-and also with gross distinctions in social rank-elite vs. commoner Differences in the latter reflect elite control of procurement as well as cultural rules that assign meanings to certain species, which in so doing regulates access to their consumption. Faunal samples collected by recent mound excavations at the Moundville site in west-central Alabama provide the basis for an examination of more subtle differences in the consumption patterns of elite residents. Zooarchaeological samples produced by two elite households, although generally similar and fitting expectations for elite consumption well, are distinguished by differences in the distribution of rare species, the role offish, and possibly by evidence of differences in food waste, distinctions that can be associated with interpretations of these households' relative status in Moundville society drawn from other classes of archaeological data.

En anios recientes, la investigacidn zooarqueologica ha comenzado a examinar las funciones de los animales dentro de la dindmica del contexto social, ceremonial y politico como parte de una serie de simbolos empleados continuamente en los sis- temas culturales prehistoricos. En el sureste de los Estados Unidos, estudios sobre los asentamientos del periodo prehistorico tardio de la cultura Mississippi, han documentado diferencias en la composicidn de especies y cortes de came asociadas con contextos sociales particulares de consumo alimenticio (por ejemplo, diferencias entrefestines ceremoniales y comidas domes- ticas) y tambien con grandes diferencias en el estrato social (por ejemplo, entre la elite y los plebeyos). Las diferencias entre este ultimo reflejan el control de la elite sobre las compras asi como en las normas culturales relacionadas con el significado de ciertas especies y que al hacerlo, regulan el acceso al consumo. En excavaciones recientes llevadas a cabo en el sitio arque- ol6gico de Moundville, localizado en la zona centro occidental del estado de Alabama, se han recogido muestras defauna que proveen la base para un andlisis de diferencias mds sutiles en los patrones del consumo alimenticio de la elite. A pesar de algunas semejanzas generales y que encajan bien dentro de las expectativas del consumo de la elite, las muestras zooarque- ologicas excavadas en dos viviendas pertenecientes a ese contexto se distinguenpor las diferencias en la distribucidn de especies raras, la funcion del pescado y la evidencia de desperdicios de comida o metodos alternos para la preparacidn de alimentos. Estas distinciones pueden estar asociadas con las interpretaciones del estrato social de los habitantes de estas viviendas en Moundville, que han sido formuladas en base a otros tipos de datos arqueologicos.

I n recent years anthropological research, both demonstration of some of the ways in which food ethnographic and archaeological, increasingly remains reflect social distinctions in prehistoric has turned its attention to the social, political, societies, particularly among those ranked middle- and symbolic underpinnings of food practices in range societies, chiefdoms. Studies focusing on small-scale societies (Clarke and Blake 1994; Mississippian chiefdoms of the southeastern United Dietler 1996; Dietler and Hayden 2001; Hayden States have demonstrated the important role of 1995, 1996; Knight 2001b; Potter 2000; Wiessner feasting in political and ritual events (Blitz 1993b; and Schiefenhovel 1996). Among the archaeolog- Kelly 2001; VanDerwarker 1999) and the nature of ical contributions to this discussion has been the economic relations between elite and commoner

H. Edwin Jackson and Susan L. Scott * Department of Anthropology and Sociology, University of Southern Mississippi, Hattiesburg, MS 39406-5074

American Antiquity, 68(3), 2003, pp. 552-572 Copyright? 2003 by the Society for American Archaeology

552

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 553

(Jackson and Scott 1995a; Scott 1983; Welch mound 1991; function. The excavation of mound summit Welch and Scarry 1995). They have also begunand slope to refuse deposits (Knight 1995, 2001a; delineate how differences in access to particular Markin 1997) has provided a host of data related foods and taxa provided an important symbolicto elite activity, including good evidence for elite medium for distinguishing the Mississippian faunal eliteuse. from the rest of society (Bogan 1980; Jackson and Scott 1995b; Maxham 2000; Rees 1997; Scott Expectations and of Faunal Use among Jackson 1998; Welch 1991; Welch and Scarry Mississippian Elite 1995). Our analysis of faunal samples Social from and political inequality and their expression Moundville elite contexts has been guided in by Mississippian pre- economy, religious authority, and vious work building on Scott's (1983) analysis the symbolic of system that sustained this system of Mississippian faunal remains from the relationships,Lubbub conditioned access to meat and cer- Creek Archaeological Locality. Lubbub Creek tain other is a animal products (Jackson and Scott single mound and village site that served as the1995b). civic This is not surprising since meat is often and ceremonial center of a simple chiefdom accorded occu- high social and symbolic value, particu- pying a stretch of the Tombigbee River in west-cen-larly among groups that depend on hunting to tral Alabama, about 65 km from Moundville obtain (Blitz it (e.g., Kent 1989), and animals are com- 1993a). monly used to portray power, dominion, and a host Previous studies demonstrated distinctions in of other characteristics that might be associated subsistence patterns at a fairly gross scale: urban with leadership positions (Hudson 1976:128-130). vs. rural, elite vs. commoner. In this paper, we Nutritional characteristics (e.g., fat content), cul- report on fauna recovered by excavations of mound tural perceptions of meat quality (e.g., stringiness, summit and slope midden contexts at Moundville taste), and culturally defined proscriptions might in Alabama, conducted by the University offind themselves expressed as socially determined Alabama between 1989 and 1998 under the direc- differences in access to certain cuts or taxa. We tion of Vernon J. Knight. These data provide an would expect differences in animal resource use to opportunity not only to test some previous predic- be a product of both the manner in which social tions regarding Mississippian elite animal use pat- difference is symbolized by foodways and the terns (Jackson and Scott 1995b; see below), but also elites' access to labor and its effect on the mix of to examine how the faunal record may indicate dif- subsistence commodities. For instance, part of this ferences in the relative rankings of social units com- variability may relate to the economic mechanisms prising the Moundville elite. by which the elite were provided with animal prod- Moundville was the paramount center of a Mis- ucts. The distinctive nature of elite refuse would sissippian polity on the Black Warrior River independ in part on the degree to which elite fami- west-central Alabama from roughly A.D. 1050 tolies relied on the efforts of their followers for food, 1450. During this interval the Moundville chiefdom through gifts, tribute, or systematic provisioning. was among the most centralized and complex chief- With greater or more regular levels of subsistence doms to have developed in the southeastern U.S. Itprovisioning, transport considerations might be has been the subject of investigations for more than expected to come into play (e.g., Welch and Scarry a century, resulting in an extraordinarily good 1995), shifting the focus of hunting efforts to larger understanding of the historical trajectory and orga- animals (Speth and Scott 1989), and increasing the nizational aspects of this important Mississippian likelihood and extent of field butchery. Under such chiefdom. We and other investigators (e.g., Knight conditions, elite households are predicted to have 1995; Peebles 1983) are confident in interpreting received field-dressed carcasses or solely the most mound-related refuse contexts as the product ofdesirable cuts. If the latter were the case, then elite domestic or ritual activity. Further, both Knight hunters retained less valued portions of the prey. (1998) and Peebles (1971, 1978, 1983) interpret dif- Thus, in contrast to the refuse that accumulated in ferences in location, mound size, and artifact asso- non-elite contexts, elite bone accumulations gen- ciations as reflecting differences in corporate group erally can be expected to include a higher propor- affiliation and status, as well as differences in tion of meat-bearing anatomical units and greater

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 554 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

representation of the highest-quality cuts (Jackson Arkansas, samples from within an elite early Cad- and Scott 1995b). Mississippian period sites withdoan structure produced more than 90 percent of evidence for distinctive elite patterns of faunal the use excavated passenger pigeon remains. In con- include Cahokia in Illinois (Kelly 2001), Crenshaw trast, turkey remains were nearly evenly split in southwest Arkansas (Scott and Jackson 1998), between elite structure samples and presumed non- Toqua in Tennessee (Bogan 1980; VanDerwarker elite contexts elsewhere on the site (Scott and Jack- 1999), and Lubbub Creek in west-central Alabama son 1998). The pattern suggests that passenger (Jackson and Scott 1995a; Scott 1983). pigeon was a delicacy reserved for the elite. Different patterns of food preparation may also To this point we have considered meat primar- distinguish elite households from those of their ilyfol- as a commodity. However, we know from eth- lowers. Greater tendency to roast rather than nohistoricstew accounts that certain animals were meat implies an abundance of available meat consideredand to represent or have qualities emanat- could even serve as a sign of conspicuous waste; ing from their place in the cosmological system of roasting results in the loss of drippings, and is moresoutheastern Indians (Hudson 1976:128-134). That likely to emphasize the meatiest cuts. The alterna- the symbolic importance of sometimes mythical tive, stewing, is a more effective way to make andcom- sometimes real animals had roots in the pre- plete use of prey, by stretching quantities with historic the past is indicated by their ample depictions addition of more water and including the meat in on Southeastern Ceremonial Complex (SECC) irregularly shaped elements, for instance vertebrae. iconography (e.g., Galloway 1989). Since SECC If we can assume that in non-elite households com- iconography served to display the sanctity and plete carcass use was an overriding goal, then power we of the Mississippian elite, the extension of would expect greater reliance on stewing for its animaleffi- symbolism to acts of consumption is not cient use of meat rather than on roasting and such its a far reach. In this way, meat consumption attendant wastefulness. More intensive bone offerspro- a potentially important mechanism for gain- cessing (smashing to gain access to marrow or ingboil- or increasing desirable qualities, and in cultural ing to render grease) may distinguish preparation contexts wherein asymmetric power relations dic- activities in non-elite contexts, although this tate dis- dietary choice, restrictions limiting that con- tinction could be mitigated by a high cultural valuesumption only to those in power may help to explain placed on marrow and grease. Marrow extraction, variation in species representation. requiring just a single break, may well be equiva- Faunal samples representing elite contexts often lent in households of varying status. However, include we a broader range of uncommonly recovered suggest that the greater processing intensity taxa. Moreover, ethnohistoric accounts indicate that required to render grease (extensive bone breakage specific characteristics of certain animals were plus boiling) is less likely to occur in high-status transferable by consumption of their meat (Jack- households with the same frequency found else-son and Scott 1995b). While in some cases rare taxa where, resulting in a smaller proportion of verysimply may be delicacies, very often it is the qual- fragmentary bone. It should be possible to gaugeities assigned to particular species that appear to the effective utilization of bone products or, alter-promote their representation. An example is the natively, wasteful behavior, by the relative degree greater representation of "dangerous" taxa, such as of bone fragmentation exhibited by bone samples bear, cougar, or bobcat, implying consumption to from different contexts. obtain their power. Carnivores in particular are Mississippian elite refuse appears to include interesting a inclusions in elite middens, since his- higher proportion of birds than are found in torically,non- at least, southeastern Indians considered elite contexts, although the importance of birdsflesh-eating animals taboo (Hudson 1976:318). varies significantly among Mississippian societies, Birds, especially raptors such as hawks, owls, and turkey in particular was staple fare in the falcons,diets and eagles, are prominent in southeastern of most Mississippians. Other taxa have more vari-Indian cosmology as well as in Mississippian able distributions that in at least some cases seem iconography, representing another category of to be a function of differences in social status. For "charged" taxa, with political as well as religious instance, at the Crenshaw site in southwest connotations. The distribution of these taxa is vari-

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 555

ably restrictive. Eagles, for instance, while notproportions com- of large mammal remains, a function mon, are found in non-elite contexts. In contrast,of maximizing the amount of meat available for the the remains of peregrine falcons, often depicted event, inrather than by diverse assemblages (e.g., SECC iconography as a falcon warrior (Galloway VanDerwarker 1999:26). However, a high diversity 1989), have an extremely limited distribution "pot luck" in approach to supplying feasts also has Mississippian sites and seem to be restricted been suggestedto (VanDerwarker 1999; Zeder 1996), mound centers, such as Cahokia (Chmumy in which1973; multiple participants provide food. For Kelly 2001) or Etowah (van der Schalie andinstance, Par- according to Charlevoix (Swanton malee 1960). A large number of swans, a taxon 1911:122): that "Each private person contributes some- is generally rare in American Bottoms assemblages, thing of his hunting, his fishing, and his other pro- was identified by Kelly (2001; also Pauketat visions, et al. which consist in maize, beans, and melons" 2002) in feasting and ritual refuse found to belowthe midsummer harvest festival held by the Mound 51 at Cahokia. No wing elements Natchez. were Very large quantities of food refuse may identified, suggesting that these had been madebe key into to recognizing feasting episodes (e.g., Kelly fans and ultimately disposed of elsewhere 2001). (Kelly 2001:349). The data suggest that in addition Finally,to the and this point takes on importance for meat provided by this large bird, associated understanding sym- the present Moundville case, Mis- bolic meaning or prescribed ritual use may sissippian under- chiefdoms were quite variable in their lie its extraordinary frequency. Smaller birds, scale andsuch degree of centralization. Mississippian as crows, jays, and other songbirds, though polities prob- ranged from apparent "big man" systems ably not important for their contribution (e.g., to Lorenzelite 1996) to complex or paramount chief- meals, nonetheless provide colored plumage doms, that such as Cahokia or Moundville, with con- can often be related to color symbolism such siderable as that variation in scale or centralized authority associated with the cardinal directions, or inwar between. and Limitations on distribution of certain peace. At Lubbub Creek, birds limited to taxa, mound as well as the degree to which foodstuffs were contexts include cardinal, mockingbird, mobilizedCarolina to support ceremony or social group, are parakeet, crow, bluejay, and a merlin (Scott partly 1983). a function of degree of political centraliza- The result is that in addition to greater-than- tion. This is itself a dynamic feature of Mississip- expected large mammal meat-bearing elements, pian polities (e.g., Anderson 1994). We can assume elite contexts are distinguished by more a diversewaxing of the political power of the Moundville assemblages that result from the preferential elite during or the period in which they mobilized the exclusive access to certain species. However, labor necessary a to undertake the massive scale of number of factors confound such a simple mound picture. and palisade construction in evidence at the First, we should not expect perfect uniformity site. This period might also represent a time of throughout the Mississippian world in howmore regimentedor organization in the provisioning which animals symbolized ideological constructs of Moundville elite, requiring more intensive or status differences. Second, meals may not exploitation be the of available faunal resources, than may only source of bone refuse; craft or paraphernalia have occurred either before or after, or than may manufacture that used animal parts, or have animal characterized smaller, less complex polities. remains resulting from ritual activities, may also contribute to the elite faunal record. Further, eliteBackground to the Present Study private refuse often may be mixed with the remains The Moundville Chiefdom of ceremonial activities such as feasting (Pauketat et al. 2002:273-275). Where the economic Between orga- approximately A.D. 1150 and 1500, nization provided daily fare to the elite, these Moundville same served as the political and ceremonial mechanisms likely also served to provision center cere- of a complex chiefdom in the Black Warrior monial feasts, so that it may be difficult to River differ- valley in Alabama (Figure 1). At its zenith entiate these sources of refuse when there the is spatialsite covered 75 ha, included at least 29 earthen overlap. If depositional events can be distinguished, mounds, and was surrounded by a bastioned pal- feast provisioning should be reflected by very isade highearly in its history (Knight and Steponaitis

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 556 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

\ o oT OH

<_M L^ ?K J? /?

Projected Palisade Line 0 150

METERS

Figure 2. The Moundville site.

Early in the fourteenth century, there was an apparently rapid evacuation of the general popula-

Figure 1. Location of Moundville and other sites tion, men- leaving only a small elite group and their tioned in text. retainers as residents of Moundville (Steponaitis 1998). The balance of the population inhabited scattered small farmsteads. Three not necessarily 1998) (Figure 2). Based on ceramic evidence, mutually four exclusive propositions have been sug- chronological phases have been distinguished, gested to account for population dispersal (Knight Moundville I through IV (Figure 3). In the and pre- Steponaitis 1998:19). Local resources may ceding Late Woodland West Jefferson phase, have prior been depleted by the large late Moundville toA.D. 1050, occupation of the Black Warrior I-early val- Moundville II site population, forcing ley consisted of a series of nucleated settlements. movement of the bulk of the population elsewhere. Moundville's initial occupation began early Alternatively,in the removal of the non-elite may have Moundville I phase, represented by a single been small aimed at increasing the sanctity of the mound and evidence of a small but growing Moundville resi- center. Finally, the consolidation of dent population. Late in the Moundville I regionalphase, political power by the Moundville elite betweenA.D. 1200 and 1250, population increased may have resulted in a relatively peaceful period, substantially, the plaza area was made level byreducing the the need to live within a palisaded settle- addition of fill in low areas, and work began ment. on all Regardless of cause, the effect of this phys- the major mounds, providing the site with the ical spa- separation would have been to accentuate the tial configuration depicted in Figure 2 (Knight symbolic and distancing of elites from the rest of Steponaitis 1998:15). A palisade that encircled Moundville the society. At the same time, burials at the mound group and associated residential area site was increased significantly in number, leading constructed at this time, undergoing several rebuild- researchers to interpret a change in site function ing episodes in the thirteenth century. The residentfrom town to necropolis (Knight and Steponaitis population peaked during this interval, estimated 1998:19). The most lavish burials appear to date to to have been approximately 1,000, all living thiswithin interval (e.g., Peebles and Kus 1977). the confines of the palisade. Several outlying Bysec- the end of the fourteenth century (late ondary mound centers were established during Moundville the II), there is evidence that a number of late Moundville I phase. the platform mounds had been abandoned, though

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 557

Moundville's political power. By the end of the AD 1600- Moundville IV sixteenth century, Moundville had been entirely abandoned. a. Knight (1998) and previously Peebles (1983)

AD 1500- have built the case that the spatial arrangement of Moundville III the plaza periphery mound group provided a diagrammatic representation of the hierarchical and corporate-group relationships among the newly AD 1400- * emerged Moundville elite. Surrounding the largest mound in the center of the plaza, the plaza periphery mounds are organized in what appear to be Moundville II pairs of large and small mounds, with the largest AD 1300- of the major mounds located on the north edge of

. the plaza, and as the distance from this northern tier increases, the sizes of the major mounds AD 1200- decrease. Mound burials more frequently occur in the minor mounds, such that the pairs were origi- Moundville I nally interpreted as representing domicile (major mound) and temple or charnel house (minor AD 1100- mounds) functions (Knight 1998). Each pair is sug-

a gested to represent a different elite corporate group organized in order of descending status from north

AD 1000- to south. Recent excavation indicates that this West Jefferson dichotomy is too simplistic, however; architecture on major mounds suggests more than one function and a domiciliary function for the minor mounds seems also to be the case (Knight 1995; 2001a). Figure 3. Moundville chronology. Previous Moundville Faunal Studies

others continued to be inhabited and underwent The archaeological research on which the forego- additional building episodes. Although the inten- ing framework is based has also produced infor- sity of prestige-goods exchange appears to have mation about dietary patterns of the Moundville diminished somewhat, elite burials continued to elite,be including the basic character of elite patterns lavish, and many of the iconographically rich SECC of meat consumption. Lauren Michals (1992) items are attributed to this interval (Knight andreported on faunal samples from several socially Steponaitis 1998:19). Additional secondary cen- differentiated contexts at Moundville, including ters were established at this time, suggesting theoff-mound middens north and west of Mound R destinations of at least some of the elites who had that are interpreted as elite residential areas occu- evacuated the site. pied during the late Moundville I phase. Michals By approximately A.D. 1450 (Moundville III), found that anatomical unit representation is indica- additional mounds were abandoned, with only three tive of deer provisioning and she identified a pos- on the northern side of the site still occupied itive correlation between social rank and increased (Knight and Steponaitis 1998:21). Only a small representation of upper forequarters and axial off-mound residential area was occupied. The num- remains. bers of burials declined, mounds ceased their mor- Welch's (1991, 1998) excavations at the White tuary function, and nucleated settlements site, a Moundville III single mound center located reappeared in the valley. Cemeteries at outlying 13 km from Moundville, produced a modest fau- settlements were established at this time, suggest- nal assemblage associated with the elite residents ing that Moundville's role as necropolis had waned. of that site. Welch's (1991) original analysis The evidence is taken to represent the decline of assumed that the site served as a local center in the

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 558 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

Moundville chiefdom's political economy. sive. The Samples of fauna recovered from three nearby revised history of the latter cast doubt that the farmsteads,two Oliver (Michals 1997), Mill Creek site sites were integrated in this way. Constructed at (Michals the 1987), and 1TU768 (Holm 1997), located time when Moundville mounds were being aban- upriver from Moundville, are small and dominated doned, the White site includes one of two ceme- by large mammal remains. Although the data are teries outside of Moundville and evidence for limited, it does appear that more primary butcher- population nucleation there during the ing late refuse is found at this class of sites (Michals Moundville III subphase, which may reflect 1997). the One exceptional rural site is the Grady Bobo progressive disintegration of the Moundville chief-site (1TU66), where excavation of a large shallow dom (Welch 1998:164-165). With respect to Moundvillethe I pit produced an exceptionally high fauna, overall taxonomic contributions to the percentagetotal of bird remains (Holm 1997; Jackson admittedly small sample are not significantly 2002).dif- It is unlikely that the pit represents every- ferent from those documented for either Lubbub day household refuse, however, and how rural rit- Creek or Michals's small sample from Moundville ual may have played a role in the early stages of (Welch 1991). Deer body-part representation sug- the development of the Moundville polity is an gests off-site butchering and possibly provisioning issue currently under study (Maxham 2000; Scarry by smaller communities. In contrast to the pattern and Scarry 1997). reported by Michals, hind limbs are considerably The University of Alabama better represented than forelimbs. Other than fox,l Mound Excavation Program no "exotic" carnivores are represented in the White site sample, and no birds other than turkey and aThe present interpretation of the history of mound teal-sized duck were identified. construction and occupation at Moundville is in no Pertinent to the role of meat and animal prod- small part the result of University of Alabama exca- ucts in the broader foodways patterns of the vations between 1989 and 1998, directed by Ver- Moundville chiefdom are several ethnobotany and non J. Knight. Knight's excavations in the flanks trace element studies. Research focused on status- and summits of five mounds have provided data for related variation in plant processing and con- a number of new and important studies (Knight sumption evidence in the ethnobotanical record 1992, 1995,2001a; Markin 1997; Ryba 1995; Taft (Welch and Scarry 1995) provided greater evidence 1996; Wilson 2001), clarifying chronology and for plant food (maize, nuts) processing at outlying expanding our understanding of mound summit settlements than at Moundville, but found no sig- activities. Knight (1989, 1992) and his student nificant difference in evidence for the amount con- Robyn Astin (1996) also reanalyzed materials from sumed. Similarly, stable isotope analysis of bone nine other mounds produced by previous investi- samples from Moundville burials by Schoeninger gations. The University of Alabama project pro- and Schurr (1998) failed to distinguish social dif- duced faunal samples from the five mound contexts. ferences in either amount of maize in the diet (mea- Our analysis of these (Jackson and Scott 2002), in sured by stable carbon isotope ratio) or in the particular, large samples from Mounds Q and G, contributions made by fish (stable nitrogen iso- are the focus of this discussion. tope). An earlier study (Peebles and Schoeninger Mounds Q and G were initiated late in the 1981) of strontium levels indicated that elites con- Moundville I phase and occupied during sumed more meat (from terrestrial animals) than Moundville II and Early Moundville III (ca. A.D. did commoners. These studies suggest that at least 1250-1450). Mound Q, located in the northwest for the Moundville chiefdom, the kinds of differ- corer of the plaza, is a modest construction just ences that may be discovered in faunal samples under 4 m tall and approximately 45 by 30 m at its from socially distinct contexts are largely symbolic base. Mound summit and flank excavations pro- of those social distinctions, rather than contribut- duced a wide array of material. Much of the bone ing in a profound way to differences in nutritional was recovered from the northern flank where a health. thick midden deposit was encountered. Summit At present, data on animal use by commoner excavation exposed architectural remains repre- households of the Moundville polity are not exten- senting multiroom domestic structures. Excava-

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 559

tions produced abundant pottery representing on Mound both Q were used for mixing paint. A strik- cooking and serving vessels, subsistence remains,ing difference in material culture is the paucity of and a wide array of artifacts and debris represent- evidence for craft manufacture. Ethnobotanical ing craft goods manufacture. Craft activities data, like that from Mound Q, suggest high con- included copper working, production of sumption tabular of corn and similar levels of processing. stone artifacts, and woodworking (Knight According2001a; to Knight (2001a), both mounds rep- Markin 1997; Wilson 2001). There is also resent abun- platforms on which Moundville elites dant evidence for pigment use, including minerals,resided, but the evidence suggests that we should paint palettes, and sherds from vessels notused assume for equivalent status within the broad mixing paints. In addition to the evidence superordinatethat the social category. Just as the mounds structure served a residential function and themselves a place suggest social distinctions in their rel- for artifact crafting, other remains, including ative highly sizes and position around the plaza, differ- fragmented bits of human bone, tobacco, ences andin the activities of the elites residing on yaupon (Ilex vomitoria, the leaves of which Mounds were G and Q appear to correspond to different used to make "black drink," an emetic concoction positions in social space. Knight suggests that the used in southeastern Indian purification residents rituals) of Mound G were more elite than their point to more esoteric activities (Knight counterparts 2001a). on Mound Q, which provided greater Subsistence-related botanical remains, analyzed access to by certain categories of artifacts and allowed Scarry (1996), are dominated by corn with them modest to remain aloof from the day-to-day activi- amounts of nutshell, squash, chenopod, knotweed, ties of Moundville's population. In contrast, those and maygrass. Scarry's analysis indicates aresiding greater on the more modest Mound Q actively amount of corn was processed in non-elite interacted resi- with elites and non-elites through their dence areas than is the case for elite contexts artisanry, at the ritual practices, and ritual bone manipu- northern end of the site. The latter, dating lation. to theThe social distinction drawn on the basis of late Moundville I phase when the site's residential artifacts and inferred activity differences can be population peaked and during the most vigorousfurther evaluated through a comparison of the fau- construction period, is interpreted to represent nal samples the from the two mounds. higher level of provisioning needed to support the Analysis site's residents at that time. After Moundville's population dispersed, local fields apparently could Mound meet Q produced 10,577 specimens from con- the needs of those still residing there (Knighttrol trenches recovered by a 6-mm-mesh screen 2001a). (number of identified specimens or NISP = 9,628). Mound G, located on the southeast margin of Flotation samples added 2,587 specimens to the the plaza, is a larger structure, 6.5 m high and sample. Controlled excavation of Mound G pro- roughly 60 by 60 m square (Knight 1995). Flank duced 3,299 specimens (NISP = 3,119) by screen- trenches revealed four building stages dating from ing and 60 additional identifiable specimens in early Moundville II through early Moundville III, analyzed flotation samples. In addition, bone from coeval with the excavated deposits of Mound Q. unscreened reference trench excavation was Summit excavations were not conducted but 25 test scanned to identify taxa not represented in the con- holes placed there by Clarence Moore in 1905 trolled samples and to collect additional informa- failed to produce burials, indirectly indicating its tion on deer element representation and breakage residential function (Knight 2001a). Midden patterns. These were kept separate analytically. At deposits on the mound's flank produced artifacts least 58 taxa are represented, 45 species in the and botanical and faunal remains. A diverse assem- Mound Q sample and 34 from Mound G (Table 1). blage is represented by the ceramic sample. Com- Examination of reference trench material yielded pared with that from Mound Q, the assemblage one additional taxon, a whooping crane (Grus cf. from G is distinguished by a higher representation americana) from Mound Q. The other much of bottle forms, reflecting the more privileged smaller samples from other mound excavations lifestyle of Mound G's residents (Knight 2001a). produced only one additional species, woodchuck There were also fewer hemispherical bowls, which (Marmota monax) (Jackson and Scott 2002).

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 560 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

Table 1. Taxa Identified in Mounds Q and Deer and large mammal (comprised almost G 6-mm Samples. entirely of fragments of deer bone too small to identify confidently) make up the bulk of the sam- Taxon Common Name Q G ples. Small mammals, birds, reptiles, and fish make Didelphis virginiana Opposum x x Sylvilagus floridana Eastern Cottontail x x near-identical contributions based on relative bone x x Sylvilagus aquaticus Swamp Rabbit weight. Following deer, turkey provides the next- Peromyscus sp. Mouse x Cricetidae Rat/Mouse x x greatest contribution to both samples. Several of the Sciurus carolinensis Eastern Gray Squirrel x x taxa, including mice, rats, and frog/toad, are Sciurus niger Eastern Fox Squirrel x x Castor canadensis Beaver x x assumed to be commensal. Presumably, the remain- Procyon lotor Raccoon x ing taxa, including at least 17 mammals, 15 birds, Mustela vison Mink x 7 reptiles, and 12 fish, were consumed or used in Mephitis mephitis Skunk x x Lynx rufus Bobcat x some other way (Jackson and Scott 2002). Felis concolor Cougar x Ursus americanus Black Bear x x Deer and Large Mammal Urocyon cinereoargenteus Gray Fox x Canis familiaris Domestic Dog x x By count, deer and large mammal comprise 70 per- Canidae Dog x x Carnivora Carnivore x x cent of the identifiable portion of the sample; by Odocoileus virginianus Whitetail Deer x x weight their contribution exceeds 90 percent (Fig- Bos/Bison sp. Cow/Bison (cf. Bison) x ure 4). Deer comprise 13-15 percent of the site Eudocimis alba White Ibis x Branta canadensis Canada Goose x x Minimum Number of Individuals (MNI). Deer ele- Aix sponsa Wood duck x ment representation corresponds well with patterns Redhead x Aythya americana identified in elite contexts by Michals (1992) for Aythya marilla Greater Scaup x Anatidae Medium Duck x x Moundville and from elite contexts at other Mis- x Grus canadensis Sandhill Crane sissippian sites. Element representation was first Buteo jamaicensis Redtail Hawk x x Buteo sp. Hawk x evaluated in the following manner. Ignoring sym- Falco peregrinus Perigrine Falcon x metry (left or right elements), the number of ele- Meleagris gallopavo Turkey x x Colinus virginianus Common Bobwhite x ment portions (e.g., proximal humerus, distal Corvus brachyrhynchos Crow x humerus) was tallied and expressed as a percent- x x Ectopistes migratorius Passenger Pigeon age of the minimal animal units (MAU, essentially Passerine Songbird Chelydra serpentina Snapping Turtle x the number of carcass portions necessary to account Chrysemys picta/ Painted/Cooter x for the most common element portion) (Jackson Pseudemys floridana Chrysemys/Graptemysl Painted/Map/Cooter x and Scott 2002). In Figure 5, element portions are Pseudemys sp. grouped into broad utility categories, following Box Turtle x Terrapene carolina Kelly (2001). Anatomical units represented in the Sternotherus sp. Musk Turtle x Kinosternidae Mud/Musk Turtle x north flank midden of Mound Q are primarily high- Apalone sp. Soft Shell Turtle x utility cuts, specifically upper fore- and hindquar- Coluber/Masticophus sp. Racer/Coachwhip x Viperidae Viper x ters. As Michals noted in her analysis of elite Rana/Bufo sp. Frog/Toad x samples from elsewhere at Moundville, forequar- Amia calva Bowfin x ters are somewhat better represented than hind Atractosteus spatula Alligator Gar x Lepisosteus platystomus Short Nosed Gar x limbs. Elements representing low and medium util- Gar x Lepisosteidae ity cuts or primary butchering debris are decidedly Ictiobus bubalus Small Mouth Buffalo x Moxostoma carinatum River Redhorse x more poorly represented, but are present. MAU Moxostoma poecilurum Blacktail Redhorse x derived from Mound G deer elements presents a Moxostoma sp. Redhorse sp. x Pylodictus olivaris Flathead Catfish x similar pattern, although hindquarters are better Ictalurus furcatus Blue Catfish x represented. Posterior axial material-lumbar ver- Channel Catfish x Ictalurus punctatus tebrae and sacra-is also better represented in the I. furcatus/punctatus Blue/Channel Catfish x Ictulurus melas Black Bullhead x Mound G sample, suggesting that either a wider Micropterus salmoides Largemouth Bass x range of cuts was consumed there or else there was Micropterus sp. Bass x Micropterus/Pomoxis Bass/Crappie x less destruction of this portion of the skeleton (the Aplodinotus grunniens Freshwater Drum x lumbar region contains the tenderloin, the source Carcharhinidae Shark x of filet mignon).

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 561

100

* Mound Q 80 E Mound G 70

0 Lge Mam Sm Mam Bird Rept/Amph Fish

Figure 4. Proportional contributions of major taxonomic groups based on weight.

100 I

I r 90 EQ IG vFk I Y r 1 r I 80 I 'kr

v vr Y I 1 70 r ^m1 W. e tk

I 41 %I rI 1 60 le I v11 I 41 I1 rv1 Ir 41r rl 50 I rl I a) C i le I 40 ^ d : PO r rl r I 30 rw

20 10 illl 111111 1 I cn 6 > , i03 U E a U) .U aE _ 0 nH I I _ 0 LL Mdu oU - 0 5t Low Utilit Medium Utility High Utility Low Utility Medium Utility High Utility

Figure 5. Relative proportions of low, medium and high utility carcass parts reflected by the percent of expected minimum anatomical units.

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 562 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

?-? Skull

Axial

Upper Fore Mound Q Limbs [ Mound G IIIIIIIIIIIIIIIIIIII Upper Hind mI White Limbs

Lower Limbs, -II Feet

IIW

-20 -15 -10 -5 0 5 10 15 20

Percent Less Percent Greater

Figure 6. Proportions of major anatomical units, measured by weight, represented by Mound Q, Mound G, and White site deer remains, in reference to proportions of units in a complete skeleton.

In addition to the graphical assessment oferal the anatomical region (skull, axial, longbone, and MAU pattern, we compared the percent MAU indeterminate, dis- the latter comprised mainly of can- tribution for each sample to those of an index cellous of bone fragments). If extensive fragmentation carcass portion utility; for comparability with is responsiblepre- for reduced identifiability, bone will vious work we used Binford's (1978) modified be gen- relegated to the large mammal category. Figure eral utility index (MGUI) calculated for caribou.2 6 compares proportional contributions by weight We also assessed the potential effects of cultural of bone or identified as deer and large mammal bone natural attrition on the samples by comparing and per- assigned to anatomical categories, expressed cent MAU with element portion bulk bone-density in terms of the expectable proportional contribu- values provided by Lyman (1984, 1991). In tions both of these anatomical units to the total skeleton cases, rank order correlation was used to weightassess of a modern deer. The similarity reinforces these relationships, using Spearman's rank the order interpretation that attrition is not significantly correlation statistic (r ). Significant positive responsible rank for the element patterning exhibited by order correlations exist between percent MAU deer. and Although not reported here, patterning of deer MGUI (for Mound Q, rS = .521; for Mound elementsG, r = recovered from reference trenches is iden- .66). In contrast, no significant correlation tical was (Jackson and Scott 2002). found between percent MAU and bulk-density Overall,val- element representation is similar to that ues, suggesting that the sample, from a preserva- observed in previous studies of deer remains from tion standpoint, is reliable. elite Mississippian contexts, with generally low Despite the lack of correlation between proportionsMAU of primary butchering debris and high and bulk density, there is still a possibility that proportions attri- of meat-bearing elements. In a previ- tion is responsible for the dominance of ousupper study (Jackson and Scott 1995a; Scott 1983), limbs. An additional control is provided by deer com- element distribution exhibited by remains bining the weights of deer elements with that recovered por- from Lubbub Creek was contrasted with tion of the assemblage classified as large mammal. that of a rural single-family farmstead, the Yarbor- This is possible because in the analysis large oughmam- site also located in the Tombigbee drainage. mal specimens were classified according to We gen- interpreted the complementary distributions of

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 563

80 . .

70 * Deer E Lg Mammal 60

50 c C, 40

1L 30 30

Skull Axial Upper Lower Limbs Limbs, Feet

Figure 7. Comparison of deer and large mammal anatomical unit representation. deer elements in the two assemblages to reflect as eliteat contexts in the Mississippian settlement at least periodic provisioning of the Lubbub Creek Lubbub Creek (Scott 1983), forelimbs are more elite by hunters residing at outlying sites. Nearer plentiful. Hind limbs and forelimbs are roughly to Moundville in the Black Warrior drainage, Welch equally represented in the Mounds Q and G sam- (1991) also suggested that the deer remains fromples. The high representation of forelimbs in these the single mound White site reflected provisioning. cases runs counter to assessments of differential The similarity in patterning among the mound sam-carcass part utility (e.g., Binford 1978, 1984; Bin- ples and that from the White site can be observed ford and Bertram 1977; Metcalfe and Jones 1988). by comparison of bone weight contributions to gen-Only at the White site are hind limbs better repre- eral anatomical categories, expressed with refer- sented. Also interesting and possibly related is the ence to these categories in the proportions found fact that hind limbs present in the Mound Q and G in a complete deer skeleton (Figure 7). While samples we tend to be from somewhat younger indi- don't disagree with Welch's characterization of viduals the than those represented by forequarters White site data, it is interesting to note that there (Jackson is and Scott 2002). This observation is based greater evidence for primary butchering (metapo- on patterns of epiphyseal fusion, an admittedly very dials, feet) there than is present in the Moundville coarse estimate of an individual's age (Purdue sample, suggesting local hunting. Moreover, 1983). as It is possible that the pattern represents a cul- Welch noted, forequarters are under-represented tural preference that defies more "rationally" deter- compared to hindquarters, in contrast to the minedmore predictions about utility. However, two other even representation in the Mound G and Q sam-(not necessarily mutually exclusive) explanations ples. Nonetheless, we are reasonably confident can that be proposed. One possibility is that transport the similarities in element representation in the considerations deer warranted smaller, more manage- remains from these elite contexts reflect similar able cuts. Shoulders can be easily removed from the economic relationships. carcass by cutting through the soft tissue behind the Accounting for the variable distribution of fore- scapula, rendering a small meat package of about limbs vs. hind limbs in Moundville elite contexts 5-6 kg. Hindquarters are considerably more diffi- may present some difficulties. In the north of Mound cult to disarticulate and are larger and more irregu- R elite contexts analyzed by Michals (1992), as larlywell shaped. This could explain why those

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 564 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

120

100 Mound Q Z Mound G a Lubbub

80 I I I

I 0 I a r Q) II

L. 60

I (L 7 s i ? ^ i ?= 40 h

4-^ ' r . w - C -' --. o o- :Eo ._''- a_ a_ a_ r-7

Figure 8. Comparison of the degree of deer element fragmentation, expressed as the proportion of element specimens 50 percent or more complete. Data from Mound Q, Mound G, and Mississippian period Lubbub Creek Archaeological Locality assemblages. hindquarters that were carried to Moundville they were are available from Lubbub Creek, which thus more often from younger and presumably, on provides aver- something of a baseline for assessing frag- age, smaller individuals. A second explanation mentation (commoner and elite contexts com- focuses on tension that may have existed between bined). Figure 8 compares the percentage of hunters and the provisioned elite. By supplying elements fore- represented by specimens greater than 50 quarters to the elite, hunters may have abided percent by the complete in the Moundville deer samples letter of the law in providing the requisite to contri- those recorded for the Lubbub Creek sample bution of venison, while maximizing the amount (Scott 1983:Tableof 6). Overall, the Moundville sam- remaining meat available for their own families. ples have In been subjected to less fragmentation, and the case that either or both of these scenarios are those from Mound G even less than Mound Q. Of correct and at the same time food utility indices do particular note is the greater percentage of more indeed reflect prehistoric meat valuation, it is inter- complete vertebrae in the Moundville samples esting to note the greater representation of hind (axis, cervical, and lumbar vertebrae). This suggests limbs in the Mound G sample. that the vertebral column was not subjected to the One way to assess the intensity of deer utiliza- degree of processing evident in the Lubbub sam- tion is by examining patterns of bone breakage. Ini- ple. If boiled in stews, vertebrae were simply dis- tial butchering, chopping to produce pieces carded once the meat fell away, rather than being appropriately sized for cooking vessels, further pro- further processed to render grease. Similarly, pha- cessing for marrow and grease, and finally natural langes are less fragmented, suggesting that these postdepositional factors all contribute to the extent were more often discarded whole, without being to which bones are found in fragmentary condition split open for marrow. In contrast, long bones (Scott 1983:286-298). To examine fragmentation exhibit similar fragmentation at both sites, with of identifiable deer bones, each specimen was only small percentages (less than 15 percent) recorded as a fraction of a whole element. While greater than 50 percent complete, indicating that comparable data are not presently available from these bones were regularly broken to extract mar- commoner contexts directly related to Moundville, row. Even greater processing is in evidence at well-

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 565

documented Mississippian farmsteads such presence as the of bison at the latter lends support to Yarborough site in northeast Mississippi Schambach's(Jackson case. and Scott 1995a; Scott 1982). Other Mammals Bison Excluding probable commensal taxa (mice and Three elements, a metatarsal, a lateral malleolus, rats), 14 mammalian species were identified in the and a first phalanx, all from Mound G, were iden- Mound samples (Table 1), though none in great tified as probable bison. The three were recovered numbers. Gray squirrels are most plentiful, fol- near one another and appear to have been the result lowed by fox squirrels, along with smaller num- of a single disposal episode. All of the elements arebers of small and medium-sized mammals that from a calf too young to be absolutely certain commonlyof occur in southeastern faunal assem- our provisional identification. However, clear abo- blages, such as cottontail, swamp rabbit, opossum, riginal (stone tool) skinning marks present on the raccoon, and dog. anterior face of the first phalanx running perpen- That gray squirrel remains outnumber those of dicular to the shaft lend credence to the identifica- fox squirrels is interesting in light of Moundville's tion; cattle were not introduced to the area until the developmental history. Fox squirrels are more late nineteenth century by which time iron tools likely to be found in open habitats, while gray squir- were in common use. Four additional specimens, rels inhabit woodland settings. Scott (1983) found a rib fragment and an indeterminate fragment from significant increases in the ratio of fox squirrel to Mound G, and two indeterminate fragments from gray squirrel from Late Woodland to Mississippian Mound Q, identified as very large mammal, may phases at Lubbub Creek, corresponding to an also represent bison. The only other possible can- increased representation of domesticated crops in didates for this identification are bear, which fre- the archaeobotanical record. The shift in taxa was quently can be recognized on the basis of surface interpreted as the result of land clearance for food texture, and elk, which are absent from late production. In the Moundville case, just the oppo- Holocene archaeological assemblages as far south site pattern is exhibited. The ratio of fox squirrel to as central Alabama. Based on size and morphol- gray squirrel (based on NISP) decreases from .85 ogy, bovid is the most likely candidate. (35:41) in Moundville II subsamples to .09 (5:51) Our present evidence for bison east of the Mis- in Moundville III subsamples. The shift suggests sissippi River in the mid-South dates to the proto- that after the dispersal of the general populace early historic period. Among the sites producing bison in Moundville II phase, local forests returned to are the protohistoric/historic Futorian site (Johnson abandoned fields, so that by Moundville III, gray et al. 1994) and at ImmoKakina'Fa', an early his- squirrel was the dominant species. Although over- toric Chickasaw site (Scott and Tuma 1998), both all cottontails, also more common in open habitats, in northeast Mississippi, and the Milner site, a mid- outnumber swamp rabbits in the Mound Q sample, seventeenth-century site on the Coosa River in the latter are found only in Moundville III sub- Alabama (Smith et al. 1993). Since bison seem to samples, providing some corroboration for the have been a very late intrusion east of the Missis- trend. Only swamp rabbit, the larger of the two sippi, we suspect that the Mound G specimens most species, was identified in the Mound G sample. likely represent exchange of bison products. We Three taxa (bobcat, cougar, and bear) fall into suggest these bones arrived as riders on bison hides the category of "dangerous prey" and are interest- used to transport dried meat or other Plains prod- ing in light of their possible roles in the symbol- ucts, left in the hide to serve as handles for the bun- ization of power. Bear occurs in both mound dles, a pattern documented at Plains village sites samples while cougar and bobcat are present only (e.g., Jackson and Scott 1992). As for the source of in Q. All three species are represented by either limb bison products, we note that Schambach (1993) or vertebral elements; none is burned or otherwise has presented a convincing argument that Spiro modified. Fox might also fall into this category, not served as a conduit that funneled Plains products because it is a particularly dangerous animal, but into the Mississippian world. Given other evidence because of its potential supernatural connotations. of connections between Spiro and Moundville, the It is present only in the Mound G sample. Altera-

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 566 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

tively, fox may reasonably be included with other tral feature of Mississippian iconography, most fur-bearing taxa, such as raccoon, mink, and skunk, often depicted in human bird form as a falcon war- which are present in both samples, but are slightly rior. As noted earlier, peregrine falcon remains are more common from Mound Q, perhaps reflecting exceedingly rare in Mississippian faunal assem- the craft activities there. blages. One interesting characteristic of both of the Turkey was likely second only to deer in the Moundville samples is the paucity of commensal amount of meat a single taxon contributed to elite rodents in the mound samples examined. We meals.have Body, wing, and leg elements are well rep- found in other elite samples an abundance of resented,rats although extremities (phalanges, pollex, and mice, which we have surmised were attracted tarsometatarsus) and skull elements are present as to elite residential areas because of their proxim- well, indicating that whole turkeys were prepared ity to large storage structures containing the plant on the mounds. Sex composition was estimated foods received as tribute. Only seven rodent bones, based on the size of each element for both con- representing both mice and rats, were identified trolled in samples and for turkey specimens from the 6-mm samples and an additional 26 in the flotation reference trenches. Smith (1975), based on samples from Mound Q. Three additional rodent Schorger's (1966) study of the wild turkey, suggests elements were identified in the Mound G 6-mm that we should expect aboriginal kill assemblages sample and none in the fine screen. By way of con- to have more females and pre-adults than males, trast, 227 rodent bones, nearly seven times as many, mirroring flock composition and reflecting the dif- were identified from an elite house structure and ficulty of capturing gobblers. Smith (1975:Table associated midden at Crenshaw in southwest 18) found that males comprised an average of only Arkansas (Scott and Jackson 1998), in a sample 23 percent of turkeys from seven Middle Missis- only 50 percent larger than the combined samples sippi sites, based on the presence or absence of from mounds Q and G. The paucity of rodent spurs on tarsometatari. In the Moundville samples remains suggests that storage facilities were maleslocated comprise 37 percent of the controlled sam- away from the mounds (although it is possible ple thatby NISP, and 40 percent of the larger sample surplus grain storage at Moundville was not impor-including bones from scanned proveniences, sig- tant). nificantly different from Smith's results (x2 = 6.063, p = .014). Why males are better represented in the Birds Moundville sample is of some interest. Large gob- Birds, dominated by turkey, comprise the second blers simply may have been preferred by the elite. most plentiful taxonomic category. Turkey com- Manipulation of local turkey populations (raising prises 85-87 percent of the bird NISP identified to wild poults?) raises another, though unevaluated, levels more specific than class. Turkey plus uniden- possibility. tifiable large bird constitutes 91-95 percent of bird Reptiles and Amphibians remains measured by NISP and nearly 95-97 percent measured by weight. Waterfowl include A variety of turtles, including snapping turtle, Canada goose, wood duck, redhead, greater scaup, aquatic emydids, box turtle, musk turtle, and soft- white ibis, sandhill crane, and whooping crane shell turtle, are represented by carapace and plas- (from a reference trench sample), each with an tron fragments. Box turtle is the most common, MNI of 1. The white ibis is an uncommon inclu- based on both NISP and weight, followed by soft- sion in southeastern faunal assemblages, particu- shell turtle. Only two snake taxa were identified, larly from inland sites. Passenger pigeon is theincluding coachwhip or racer, represented by two second most common taxon (NISP = 13, MNI =vertebrae, and a viper represented by four verte- 3). Six bones from raptors were identified in thebrae. Reptiles are less well represented in the Mound Q sample, but only one could be identified Mound G sample, although sample size may be the to species, a redtail hawk. Particularly significant cause. is the presence of peregrine falcon in Mound G. The Fish peregrine falcon is perhaps most telling of the status of the elite residents of the mound, being a cen- Fish appear to have made a minor contribution to

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 567

35 * Mound Q [ Mound G 30

0) 20 0

0. 15

10 .

Figure 9. Relative contributions made by fish taxa to Mound Q and Mound G assemblages. elite meals, compared to greatest large contribution mammal of any single or taxon large (29 per- bird. However, the samples cent), whileare suckers diverse collectively and constitute there the are some differences between the two. From largest portion of the sample (35 percent). Unlike Mound Q (6-mm fish NISP = 409), the single themost Mound Q sample, that from Mound G lacks abundant species is freshwater drum, comprising individuals larger than 60 cm. One unusual speci- 16 percent of fish MNI (Figure 9). The most abun- men in the Mound G fish sample is an unfossilized dant family, suckers, including blacktail redhorse, shark tooth. It is unmodified, and while it clearly river redhorse, and smallmouth buffalo, comprise indicates contact with coastal populations or a visit nearly 29 percent of the sample. Catfish (25 toper- the shore, it cannot be determined whether meat cent of the fish sample) include mainly blue or and simply the tooth was obtained. channel cats, and a single black bullhead. Gars, Discussion which comprise 12 percent of the sample, include both alligator gar and shortnose gar. Modal bodyThere is no reason to doubt that the residents of both length for most taxa falls within the 30-45-cm mound summits were elite members of Moundville range (standard length), although certain taxa, society, based on multiple lines of archaeological including gar, redhorse, and channel/blue catfish evidence. all This social position afforded access to had individuals in excess of 55 cm, and one considerablealli- variety in the meat portion of the diet, gator gar specimen was from an individual greater including choice cuts of venison, turkeys, and a than 100 cm in length. A general emphasis on varietyriver of other, often uncommonly recovered taxa. channel fishing is indicated by fish species com-Animals with hypothesized symbolic meaning are position, casting doubt on an oft-voiced suggestion in evidence as well: bobcat, cougar, fox, black bear, of the importance of fishing in borrow-pit ponds white ibis, redtail hawk, sandhill and whooping scattered across the site (e.g., Walthall 1980:216), cranes, peregrine falcon, and some unknown num- at least for provisioning the elite. ber of songbirds. Bison and shark reflect the far- Fish composition in the Mound G sample lacksflung ties that the elite maintained in the bowfin (Figure 9), although sampling error may Mississippian be world. While there is evidence for responsible given the small number of specimens provisioning or at least off-mound butchering, there (6.4-mm fish NISP = 88). Drum again makes is thealso evidence that culinary preparation occurred

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 568 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

within each household although care to completely sissippian pattern of faunal exploitation, which use animal products was apparently not important, shows a decrease in the range of taxa utilized com- judging from the relatively low level of bone pared frag- to Late Woodland assemblages (Muller 1997; mentation compared to other Mississippian sitesSmith in 1975). The narrower range is attributable to the region. scheduling conflicts presented by intensive agri- In the context of recent interest among archae- culture as well as a restructuring of the distribution ologists and ethnographers alike in the role of feast- of animal populations resulting from extensive field ing in social and political interactions, our initial clearing and higher, more sedentary, human popu- expectation was that the mound faunal assemblages lations (Muller 1997:227; Scott 1983:322). would fit hypothesized profiles of feasting refuse. Nonetheless, at least in the Moundville case, the However, the deer element profiles and general luxury of variety appears to have been enjoyed, or diversity of the assemblage are not consistent with perhaps required, by its elite residents. the expectation that bulk meat was the ultimate Conclusions goal. In contrast, Kelly's (2001:347) interpretation of the sub-Mound 51 pit at Cahokia as a deposit The present study suggests that in addition to dis- produced by feasting is based on evidence for ini- tinguishing broad contextual (feasting vs. private tial processing of deer elsewhere (a nearly complete consumption) and social (elite vs. commoner) absence of skull or feet) and considerable butch- dimensions of prehistoric middle-range societies, ery waste (articulated and presumably uncooked faunal remains may be added to the list of artifact vertebral columns, little bone breakage attributable categories with which useful, sometimes subtle, to marrow extraction, etc.). Moreover, entomolog- distinctions in social rank can be identified. In the ical evidence (abundant blowfly pupae) from this Moundville case, differences in rank between extraordinary pit feature indicates the disposal of mound-top households based on ceramic assem- raw meat presumably still attached to limbs (Kelly blages, craft debris, and other artifacts are mirrored 2001:348). Other large taxa, including large river in differences in foodways and other uses of ani- channel fish and a preponderance of swans, also mal products. The differences reflect broad princi- suggest that the goal was to gather a large quantity ples related to economic organization, differential of meat to supply the feasts, a faunal profile that access, and animal symbolism. As stated at the out- was consistent throughout the strata of the pit. This set, we don't intend to propose a formulaic method simply is not the case for the Moundville samples. for ascertaining rank differences in Mississippian While similarities between the Mound Q and societies. However, to the extent that foodways Mound G assemblages are greater than the differ- incorporate cultural meanings related to political ences, it is those subtle differences that express dis- power, social differences, ethnic identity, or ritual tinctions among the elite social group, which matters, faunal assemblages resulting from partic- Knight (2001) has established based on more ular activities or from different household contexts durable classes of archaeological remains. Most can be expected to vary as a consequence. For the striking is that the most unusual taxa-bison, shark, Mississippian case, some of these can be predicted and peregrine falcon-are all from Mound G. Fur- by "objective" measures such as utility or waste- bearing taxa, possibly related to craft production, fulness, while other measures are clearly rooted in are more common in Mound Q. Further, the sam- the symbolic imagery of prehistoric southeastern ple from Mound G is even less fragmentary than Indian culture. In relying on the latter class of mea- that from Q, suggesting less-frequent bone pro- sures, our work with Mississippian societies is cessing or less-frequent consumption of stews. Dif- clearly informed by the ethnographic record of ferences in networks, as well as the activities myth and beliefs of southeastern peoples. However, through which those networks were maintained, the cross-cultural value placed on meat and meat seem to distinguish the faunal records of the two eating (e.g., Kent 1989) leads us to suggest that elite residential groups that inhabited their respec- symbolic attributes of animals may have broader tive mounds. application. Just as Helms (1992) has demonstrated In general, the variety of meat in the diet in the that exotic materials or particular forms of work- Moundville elite runs counter to the general Mis- manship are useful clues for identifying symboli-

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 569

cally "charged" artifacts, the attributes of Clarke, certain John A., and Michael Blake animals in faunal assemblages (for instance, 1994 iden- The Power of Prestige: Competitive Generosity and the Emergence of Rank Societies in Lowland Mesoamer- tification of rare or dangerous taxa, etc.) recovered ica. In Factional Competition and Political Development from suspected high-status contexts may be usefulin the New World, edited by E. M. Brumfiel and J. W. Fox, in evaluating the ways in which rank differences pp. 17-30. Cambridge University Press, Cambridge. Chmurny, William W. affected foodways variation in other archaeologi- 1973 The Ecology of the Middle-Mississippian Occupation cal contexts. of the American Bottom. Ph. D. Dissertation, University of Illinois, Champaign-Urbana. University Microfilms, Ann Arbor. Acknowledgments. The National Science Foundation (Award Dietler, Michael 920658) and the University of Alabama supported the exca- 1996 Feasts and Commensal Politics in the Political Econ- vations on which this study is based. The University of omy: Food, Power and Status in Prehistoric Europe. In Southern Mississippi provided laboratory space as well as Food and the Status Quest: An Interdisciplinary Perspec- support for the research described here. We gratefully tive, edited by Polly Wiessner and Wulf Shiefenhovel, pp. acknowledge access to the Ornithology Skeletal Collection 87-126. of Berghahn Books, Providence. the University of Michigan's Museum of Zoology, permittingDietler, Michael, and Brian Hayden identification of some of the rarer bird taxa and collections 2001 Feasts:at Archaeological and Ethnographic Perspec- the American Museum of Natural History and tives the on Food, Politics, and Power. Smithsonian Institu- Smithsonian Institution, which greatly facilitated the identifi- tion Press, Washington, D.C. Galloway, Patricia cation of bison from Moundville. Thanks to Elizabeth Reitz, 1989 The Southeastern Ceremonial Complex:Artifacts and director of the Georgia Museum of Natural History, for Analysis. University of Nebraska Press, Lincoln. access to collections there. Special thanks are Gibson, due Jon the L. Principal Investigator, Vernon J. Knight, for his particular 2000 The Ancient Mounds of Poverty Point, Place of Rings. patience during the course of the analysis, and for his University com- Press of Florida, Gainesville. ments on an earlier version of this manuscript. Hayden,We've Brianalso benefited from many discussions with Paul Welch 1995 Pathwaysand to Power: Principles for Creating Socioe- Margaret Scarry about Moundville archaeology and conomic from Inequalities. In Foundations of Social Inequal- Paul's comments on an earlier version of the paper. Francisco ity, edited by T. Douglas Price and Gary Feinman, pp. 15-86. Plenum Press, New York. Melara kindly prepared the Spanish abstract. Finally, we 1996 Feasting in Prehistoric and Traditional Societies. In appreciate the insightful comments of Christopher S. Peebles, Food and the Status Quest: An Interdisciplinary Study, Nancy M. White, and two anonymous reviewers, all of whom edited by Polly Wiessner and Wulf Schiefenhovel, pp. helped to clarify the presentation of our work. 127-147. Berghahn Books, Providence. Helms, Mary References Cited 1992 Political Lords and Political Ideology in Southeast- Anderson, David G. ern Chiefdoms: Comments and Observations. In Lords of the Southeast: Social Inequality and the Native Elites of 1994 The Savannah River Chiefdoms: Political Change in the Late Prehistoric Southeast. University of Alabama Southeastern North America, edited by Alex. W. Barker Press, Tuscaloosa. and Timothy. R. Pauketat, pp. 185-194. American Anthro- Astin, Robyn pological Association Archaeology Papers 3, Washington, D.C. 1996 Mound M: Chronology and Function at Moundville. Holm, Mary A. Unpublished Master's thesis, Department of Anthropol- ogy, University of Alabama, Tuscaloosa. 1997 Appendix A: Zooarchaeological Remains from Blitz, John H. Moundville I Phase Features at 1TU66 and 1TU768. In West Jefferson Community Organization in the Black War- 1993a Ancient Chiefdoms of the Tombigbee. The Univer- sity of Alabama Press, Tuscaloosa. rior Valley, Alabama, by C. M. Scarry and J. F. Scarry. Report of Investigations prepared for the National Geo- 1993b Big Pots for Big Shots: Feasting and Storage in a graphic Society. Report on file, Research Laboratories in Mississippian Community. American Antiquity 58:80-96. Binford, Lewis R. Anthropology, University of North Carolina, Chapel Hill. Hudson, Charles 1978 Nunamiut Ethnoarchaeology. Academic Press, New York. 1976 The Southeastern Indians. University of Tennessee Press, Knoxville. 1984 Faunal Remains from Klasies River Mouth. Academic Jackson, H. Edwin Press, Orlando. Binford, Lewis R., and J. B. Bertram 2002 An Analysis of Faunal Remains from 1TU66, 1977 Bone Frequencies and Attritional Processes. In For Tuscaloosa County, Alabama. Report on File, Research Laboratories in Anthropology, University of North Car- Theory Building in Archaeology, edited by L. R. Binford, pp. 77-153. Academic Press, New York. olina, Chapel Hill. Jackson, H. Edwin, and Susan L. Scott Bogan, Arthur E. 1980 A Comparison of Late Prehistoric Dallas and Over- 1992 Vertebrate Remains from Big Village. In Archaeol- hill Cherokee Subsistence Strategies in the Little Ten- ogy and Ethnohistory of the Omaha Indians: The Big Vil- nessee River Valley. Unpublished Ph.D. Dissertation, lage Site, edited by John M. O'Shea and John Ludwickson, pp. 329-340. University of Nebraska Press, Omaha. Department of Anthropology, University of Tennessee, Knoxville. 1995a Mississippian Homestead and Village Subsistence Organization: Contrasts in Large-Mammal Remains from

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 570 AMERICAN ANTIQUITY [Vol. 68, No. 3, 2003]

Two Sites in the Tombigbee Valley. In Mississippian Lyman, Com- R. Lee munities and Households, edited by J. Daniel Rogers 1984 and Bone Density and Differential Survivorship of Fossil Bruce D. Smith, pp. 181-200. University of Alabama Press, Classes. Journal of Anthropological Archaeology Tuscaloosa. 3:259-299. 1995b The Faunal Record of the Southeastern Elite: The 1991 Taphonomic Problems withArchaeologicalAnalyses Implications of Economy, Social Relations, and Ideology. of Animal Carcass Utilization and Transport. In Beamers, Southeastern Archaeology 14:103-119. Bobwhites, and Blue-points, edited by James R. Purdue, 2002 Appendix E: Faunal Utilization by the Moundville William E. Klippel, and Bonnie W. Styles, pp. 125-138. Elite: Zooarchaeology of Mounds Q, G, E, F, and R. In Illinois State Museum Scientific Papers Vol. 23, and the Chronology and Use ofPublicArchitecture atMoundville: University of Tennessee, Department of Anthropology Excavations in Mound Q, by Vernon J. Knight. Report sub- Report of Investigations No. 52. Illinois State Museum, mitted to the National Science Foundation, Award No. Springfield. 9727709. Report on file, Department of Anthropology, Madrigal, T. Craig, and Julie Zimmerman Holt University of Alabama, Tuscaloosa. 2002 White-Tailed Deer Meat and Marrow Return Rates Johnson, Jay K., Susan L. Scott, James R. Atkinson, andAndrea and the Application to Eastern Woodlands Archaeology. Shea American Antiquity 67:745-759. 1994 Late Protohistoric Settlement and Subsistence on the Markin, Julie G. Black Prairie: Buffalo Hunting in Mississippi. NorthAmer- 1997 Elite Stoneworking and the Functions of Mounds at ican Archaeologist 15:167-179. Moundville. Mississippi Archaeology 32:117-135. Kelly, Lucretia S. Maxham, Mintcy D. 2001 A Case of Ritual Feasting at the Cahokia Site. In 2000 Rural Communities in the Black Warrior Valley, Feasts: Archaeological and Ethnographic Perspectives on Alabama: The Role of Commoners in the Creation of the Food, Politics, and Power, edited by M. Dietler and B. Hay- Moundville I Landscape. AmericanAntiquity 65:337-354. den, pp. 334-367. Smithsonian Institution Press, Wash- Metcalf, Duncan, and Kevin T. Jones ington, D.C. 1988 A Reconsideration of Animal Body Part Utility Kent, Susan Indices. American Antiquity 53:486-504. 1989 Cross-cultural Perception of Farmers as Hunters and Michals, Lauren the Value of Meat. In Farmers as Hunters: The Implica- 1987 Faunal Remains. In The Mill Creek Site, 1TU265, tions of Sedentism, edited by Susan Kent, pp. 1-17. Cam- Black WarriorRiver Valley, Alabama, edited by Tim S. Mis- bridge University Press, Cambridge. tovich, 171-182. Report of Investigations 54. University Knight, Vernon J. of Alabama Office of Archaeological Research, 1989 Certain Aboriginal Mounds at Moundville: 1937 Tuscaloosa. Excavations in Mounds H-L. Paper presented at the 46th 1992 The Nature of Faunal Exploitation in Mississippian Annual Meeting of the Southeastern Archaeological Con- Societies. Paper presented at the 57th Annual Meeting of ference, Tampa. the Society for American Archaeology, Pittsburgh. 1992 Preliminary Report on Excavations at Mound Q, 1997 The Oliver Site and Early Moundville I Phase Eco- Moundville. Paper presented at the 49th Annual Meeting nomic Organization. In Archaeology of the Moundville of the Southeastern Archaeological Conference, Little Chiefdom, edited by V. J. Knight and V. P. Steponaitis, pp. Rock. 167-182. Smithsonian Institution Press, Washington, D.C. 1995 Chronology and Use of Public Architecture at the Muller, Jon Moundville Site: Excavations in Mounds R, F, G, and E. 1997 Mississippian Political Economy. Plenum Press, New Second Interim Report to the National Science Founda- York. tion, Award No. 920658. Report on file, Department of Pauketat, Timothy R., Lucretia S. Kelly, Gayle J. Fritz, Neal H. Anthropology, University of Alabama, Tuscaloosa. Lopinot, Scott Elias, and Eve Hargrave 1998 Moundville as a Diagrammatic Ceremonial Center. 2002 The Residues of Feasting and Public Ritual at Early In Archaeology of the Moundville Chiefdom, edited by Cahokia. American Antiquity 67:257-279. Vernon J. Knight and Vincas Steponaitis, pp. 44-62. Smith- Peebles, Christopher S. sonian Institution Press, Washington, D.C. 1971 Moundville and Surrounding sites: Some Structural 2001a Characterizing Elite Midden Deposits at Moundville. Considerations of Mortuary Practices. In Approaches to Paper presented at the 66th Annual Meeting of the Soci- the Social Dimension of Mortuary Practices, edited by ety for American Archaeology, New Orleans. James A. Brown, pp. 68-91. Memoir 15, Society for Amer- 2001b Feasting and the Emergence of Platform Mound Cer- ican Archaeology, Washington, D.C. emonialism in Eastern North America. In Feasts: Archae- 1978 Determinants of Settlement Size and Location in the ological and Ethnographic Perspectives on Food, Politics, Moundville Phase. In Mississippian Settlement Patterns, and Power edited by Michael Dietler and Brian Hayden, edited by Bruce D. Smith, pp 369-416. Academic Press, pp. 311-333. Smithsonian Institution Press, Washington, New York. D.C. 1983 Moundville: Late Prehistoric Sociopolitical Organi- Knight, Vernon J., and Vincas P. Steponaitis zation in the Southeastern United States. In The Develop- 1998 A New History of Moundville. InArchaeology of the ment of Political Organization in Native North America, Moundville Chiefdom, edited by Vernon J. Knight and Vin- edited by Elizabeth. Tooker, pp. 183-201. American Eth- cas P. Steponaitis, pp. 1-25. Smithsonian Institution Press, nological Society, Washington, D.C. Washington, D.C. Peebles, Christopher S., and Susan M. Kus Lorenz, Karl 1977 Some Archaeological Correlates of Ranked Societies. 1996 Small-scale Mississippian Community Organization American Antiquity 42:421-448. in the Big Black River Valley of Mississippi. Southeast- Peebles, Christopher S., and Margaret J. Schoeninger ern Archaeology 15:145-171. 1981 Notes on the Relationship between Social Status and

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms REPORTS 571

Diet at Moundville. Southeastern Archaeological Confer-1975 Middle Mississippian Exploitation of Animal Popu- ence Bulletin 24:96-97. lations. Anthropological Papers No. 57, University of Potter, James M. Michigan Museum of Anthropology, Ann Arbor. 2000 Pots, Parties, and Politics: Communal Feasting in the Smith, Marvin T., Vernon J. Knight, J. B Smith, and K. R. Turner American Southwest. American Antiquity 65:471-492. 1993 The Milner Village, a Mid 17t Century Site near Purdue, James R. Gadsden, Alabama. In Archaeological Survey and Exca- 1983 Epiphyseal Closure in Whitetail Deer. Journal of vations in the Coosa River Valley, Alabama, edited by V. Wildlife Management 47:1207-1213. J. Knight, pp. 49-70. Alabama Museum of Natural His- Rees, Mark A. tory Bulletin 15, Tuscaloosa. 1997 Coercion, Tribute and Chiefly Authority: The Regional Speth, John D., and Susan L. Scott Development of Mississippian Political Authority. South- 1989 Horticulture and Large-Mammal Hunting: The Role eastern Archaeology 16:113-133. of Resource Depletion and the Constraints on Time and Ryba, Elizabeth A. Labor. In Farmers as Hunters: The Implications ofSeden- 1995 Summit Architecture on Mound E at Moundville. tism, edited by Susan Kent, pp. 71-79. Cambridge Uni- Unpublished Master's thesis, Department of Anthropol- versity Press, Cambridge. ogy, University of Alabama, Tuscaloosa. Steponaitis, Vincas P. Scarry, C. Margaret 1998 Population Trends at Moundville. In Archaeology of 1996 Archaeobotanical Analysis. In Chronology and Use the Moundville Chiefdom, edited by Vernon J. Knight and of PublicArchitecture at the Moundville Site: Excavations Vincas P. Steponaitis, pp. 26-43. Smithsonian Institution in Mounds R, E, F; and G, edited by V. J. Knight. Second Press, Washington, D.C. Interim Report to the National Science Foundation. Asward Swanton, John R. No. 920658. Report on file, Department of Anthropology, 1911 The Indians of the Lower Mississippi Valley. Bureau University of Alabama, Tuscaloosa. of American Ethnology Bulletin 43. U.S. Government Scarry, C. Margaret, and John. F Scarry Printing Office, Washington, D.C. 1997 West Jefferson Community Organization in the Taft, Black Kristi E. Warrior Valley, Alabama. Report of Investigations pre- 1996 Functionally Relevant Classes of Pottery at pared for the National Geographic Society. Research Lab- Moundville. Unpublished Master's thesis, Department of oratories in Anthropology, University of North Carolina, Anthropology, University of Alabama. Chapel Hill. van der Schalie, Henry, and Paul Parmalee Schambach, Frank F 1960 Animal Remains from the Etowah Site, Mound C, 1993 Some New Interpretations of Spiroan Culture History. Bartow County, Georgia. The Florida Anthropologist In Archaeology of North America: Papers in Honor of 13:37-54. Stephen Williams, edited by James B. Stoltman, pp. Van Derwarker, Amber M. 187-230. Archaeological Report No. 25, Mississippi 1999 Feasting and Status at the Toqua Site. Southeastern Department of Archives and History, Jackson. Archaeology 18:24-34. Schoeninger, Margaret J., and Mark R. Schurr Walthall, John S. 1998 Human Subsistence at Moundville: the Stable Isotope 1980 Prehistoric Indians of the Southeast: Archaeology of Data. In Archaeology of the Moundville Chiefdom, edited Alabama and the Middle South. University of Alabama by Vernon J. Knight and Vincas P. Steponaitis, pp. 120-132. Press, Tuscaloosa. Smithsonian Institution Press, Washington, D.C. Welch, Paul D. Schorger, A. William 1991 Moundville's Economy. University of Alabama Press, 1966 The Wild Turkey. University of Oklahoma Press, Nor- Tuscaloosa. man. 1998 Outlying Sites within the Moundville Chiefdom. In Scott, Susan L. Archaeology of the Moundville Chiefdom, edited by Ver- 1982 Yarborough Site Faunal Remains. In Archaeological non J. Knight and Vincas P. Steponaitis, pp. 133-166. Investigations at the Yarborough Site (22CL814), Smithsonian Clay Institution Press, Washington, D.C. County, Mississippi, edited by Carlos Solis Welch, and Richard Paul D., and C. Margaret Scarry Walling, pp. 140-152. Report of Investigation 1995 30, Status-Related Office Variation in Foodways in the of Archaeological Research, University of Alabama,Moundville Chiefdom. American Antiquity 60:397-419. Tuscaloosa. Wiessner Polly, and Wulf Schiefenhovel (editors) 1983 Analysis, Synthesis, and Interpretation of Faunal 1996 Foodand the Status Quest:An Interdisciplinary Study. Remains from the Lubbub Creek Archaeological Local- Berghahn Books, Providence. ity. In Prehistoric Agricultural Communities in West Wilson,Cen- Gregory D. tral Alabama, edited by Christopher S. Peebles, pp.2001 Small Celt, Big Celt, Old Celt, New Celt: The Early 272-290. Report submitted to the U.S. Army Corps of Mississippian Greenstone Tool Industry at Moundville. Engineers, Mobile District. Southeastern Archaeology 20:118-127. Scott, Susan L., and H. Edwin Jackson Zeder, Melinda A. 1998 Early Caddo Ritual and Patterns of Animal Use: 1996 Zooarchaeological Approaches to Complexity: A Analysis of Faunal Remains from the Crenshaw Site View from the Old World. Paper presented at the 53rd (3MI6), Southwestern Arkansas. The Arkansas Archeolo- Annual Meeting of the Southeastern Archaeological Con- gist 37:1-38. ference, Birmingham. Scott, Susan L., and Michael Tuma 1998 Analysis of Faunal Remains from ImmoKakina'Fa'. Notes Unpublished report on file, Cobb Institute of Archaeology, Mississippi State University, Starkville. 1. Fox is nonetheless interesting, appearing in southeast- Smith, Bruce D. ern representational art at least as early as the Late Archaic

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms 572 AMERICAr N A .NTIQUITY [Vol. 68, No. 3, 2003]

Poverty Point and Jones' period (1988) analysis reports in little changethe in the rankLower Mississippi Valley and playing a role orderingin ofChoctaw anatomical portions; thus, weconceptions are comfortable of supernatural beings (e.g., Gibson staying2000:191). with Binford's MGUI, at least for the time being. 2. Other food utility indices have been developed, building on Binford's analysis, for instance Metcalfe and Jones (1988). Recently, Madrigal and Holt (2002) reported meat and marrow values Received August 26,for 2002; Revised whitetail December 17, 2002; deer carcass portions, unfor- tunately after Accepted the December present 17, 2002. analysis was completed. Metcalfe

This content downloaded from 131.95.218.41 on Tue, 29 Aug 2017 19:49:05 UTC All use subject to http://about.jstor.org/terms