Introduction

Distinct culinary and ritual uses of animals between members of different social strata are powerful and visual means to establish and maintain social disparity. How did the elites of an early Andean state society use animals to distinguish themselves from the general populace and bolster social and political inequality? The site of Cerro Baúl, located in far southern Peru, is a provincial capital of the Andean Wari imperium (Andean Middle Horizon) whose faunal assemblage demonstrates that the “luxury of variety” in animal use was an elite prerogative that fostered social inequality.

The ability to acquire a variety of animals reflects elite control of trade networks, the means to transport food items from distant lands, and the ability to order specialists to acquire local wild animals through hunting or other capture methods (also see López Luján et al., chapter 2, this volume; Sugiyama et al., chapter 1, this volume). A variety of fauna may also signify gifts or offerings that are brought by supplicants or individuals invited from the hinterland to the regional capital. A diverse range of foods circumscribed in their spatial distribution may indicate that elites created a class of luxury foods and restricted the intake of these foods to enforce their social standing (Van der Veen 2003). As has been demonstrated for Mississippian societies in the southeastern United States, the “luxury of variety” was something that the elites could “afford” (Jackson and Scott 2003).

Faunal variety is also achieved by the inclusion of nonfood animals used in ritual. Interpretation of ritual animals is complicated by the fact that many animals can serve in both dietary and ritual capacities. In addition, the remains of animals with low dietary value that are exotic (i.e., that do not naturally occur in the vicinity of the site) may represent the refuse from ritual activity (López Luján et al., chapter 2, this volume). Ethnohistorical and ethnographic studies of traditional Andean societies indicate that people create strong associations between the indigenous fauna and an Andean landscape that is considered animate (e.g., Allen 1988; Gade 1999; Guaman Poma de Ayala 1980; Urton 1985). Andean rituals related to death, fertility, and seasonal renewal often involved either the sacrifice of animals or the use of animal products (e.g., fat, blood) in ritual performance (Allen 1988; Guaman Poma de Ayala 1980). The curation and display of portions of exotic animals (e.g., feathers, pelts, teeth, worked-bone specimens) may also have served as symbols of power or prestige.

The summit of Cerro Baúl is characterized by spatial variability in architecture as well as in the distribution of animal remains across the site. The remnants of stone masonry architecture can be divided into residential compounds, ceremonial/ritual structures, and industrial complexes, particularly for lapidary and beer production. Following a discussion of the site, I describe the faunal remains from ten summit contexts to demonstrate how Wari elites created social inequality through the luxury of variety. The ability to acquire, consume, and display diverse animals bolstered the political and social standing of the upper echelon of Wari society.

The Archaeology of Cerro Baúl

Originating from the Ayacucho region of central Peru, the Wari colonized the Osmore River drainage sometime early in the seventh century AD and held sway over the territory for roughly 400 years (Figure 3.1) (Williams 2001). Settling in the upper part of the river drainage (2000–2500 masl), the Wari elite established their colonial capitol on the summit of Cerro Baúl, a steep-sided trunk-shaped mountain (Moseley et al. 2005; Williams 2001). Subsidiary settlements and ethnic barrios were built on the slopes of Cerro Baúl and on neighboring hilltops. The Wari transformed the sierra habitat into productive agricultural land through the construction of a high-elevation canal (Williams 2001). In addition to a reliance on domesticated camelids, the elite who resided at the administrative center of Cerro Baúl were able to acquire a variety of both local and exotic food animals and several nonfood animals, presumably for use in ritual, symbolism, and display. The dietary and ritual uses of animals at Cerro Baúl differ from both neighboring sites in the Wari settlement hierarchy and Wari sites elsewhere in Peru.

Figure 3.1. Location of Cerro Baúl in southern Peru.

The Wari capital at Cerro Baúl is enigmatic. The site is an illogical location for the capital of an imperial colony. There are no natural resources to sustain the population on the summit; with the exception of some ground birds (e.g., doves) and small mammals (e.g., foxes, vizcachas [Andean hares]), all food and water had to be transported to the hilltop. It is also not easily defensible, despite its appearance as a natural bastion. However, what the site lacked in infrastructure was compensated for by its place in the cosmic landscape. Today, Cerro Baúl is considered an apu, or sacred mountain, with modern devotees trekking 600 meters from the valley floor to the summit to make dedicatory offerings. Once on the summit, some of the highest and most cosmologically important peaks in the southern Andes are visible, including Arundane and Picchu Picchu (Williams and Nash 2006). Clearly, the physical prominence and spiritual might of Cerro Baúl overshadowed its logistical difficulties. The political authority of the residents of Cerro Baúl was legitimated though by their ability to commandeer and control a sacred apu (Williams and Nash 2006).

The scale of stone masonry construction and labor investment at Cerro Baúl is evidence of a sophisticated political economic apparatus (Moseley et al. 2005; Williams 2001). Covering three hectares, the Wari settlement consisted of high-walled agglutinated buildings, some more than two stories high. The inward orientation of most compounds guaranteed social differentiation and restricted access. Stone quarries are present on the summit, but thatch for roofing material, stone paving for patios, and other construction materials had to be brought to the summit. The labor for construction and maintenance of the facility had to be secured, organized, and fed.

After thriving for several centuries, the settlement at Cerro Baúl came to a spectacular end (Moseley et al. 2005). Although some buildings fell out of use or were ritually “closed” prior to the demise of the site, the full-scale abandonment was a dramatic event that involved the destruction of much of the material culture, particularly ceramics, in combination with the torching of buildings. Material discarded on floor surfaces was covered and sealed by the collapsed and burned buildings. Because the summit was not reoccupied again until probably the second half of the fifteenth century AD, many of the contexts lay undisturbed for centuries. Despite burning, taphonomic destruction of bone was not a major factor affecting the faunal assemblage.

Excavations at Cerro Baúl and neighboring Wari sites began in 1989 and continue today as part of a multiyear program of research on Wari colonization in southern Peru (Feldman 1989; Moseley et al. 2005; Williams 2001). The primary excavation method consisted of identifying structure complexes (designated as Units) from surface architecture, and then excavating rooms, patios, and associated architectural features as distinct spatial entities using a gridded layout of one-meter squares. Thus far, excavations have been completed in twenty summit contexts, various barrios located on the slopes, and at the neighboring site of Cerro Mejía.

During the 1989 field season faunal remains were collected with one-quarter-inch (6.35 mm) mesh. In subsequent field seasons excavators used site stratigraphy to make decisions regarding mesh size. Excavation units contain a layer of ash from the AD 1600 eruption of the Huaynaputina volcano. Excavators screened all strata below the ash lens with one-sixteenth-inch (1.8 mm) mesh. All faunal material was picked from screens in the field. All excavations within structures or plazas were terminated once intact floors were encountered; no subfloor excavations were completed with the exceptions of feature fill or looters’ pits, including burials that extended below floor level. Faunal remains present in the stratum from the ground surface to the ash deposit and those within the ash layer are excluded from this analysis. Summit excavations produced an abundance of faunal remains from diverse domestic, industrial, and ceremonial structures. This chapter reports on ten of the most diverse and best-preserved summit contexts (Table 3.1, Figure 3.2).

Excavation Contexts

Unit 1 is a large brewery compound (259 square meters) for the production of maize beer or chicha (Figure 3.2). The trapezoidal compound consists of four rooms where the brewing, fermentation, and consumption of beer took place. Remnants of fermentation vats indicate that the brewery could have produced approximately 1800 liters of chicha per batch, making this one of the largest pre-Inca breweries in the Americas (Moseley et al. 2005:17267). Serving pitchers and consumption jars (keros) indicate that ritual libation was a very important component of Wari political economy. As with several other buildings, the brewery was torched as part of site abandonment. The discarded remains of ornate keros attest to the conspicuous ritual disposal of material wealth.

Figure 3.2. Ten summit contexts, Cerro Baúl.

A large body of labor was needed for all aspects of beer production and serving. For the later Inca state, the social production of beer on a massive state-sponsored scale is interpreted as having mirrored household systems of beer production and consumption in which women’s labor was instrumental in creating the beverage that sustained the household and, symbolically, society as a whole (see Bray 2003). The recovery of several shawl pins (tupus) in the brewery assemblage indicates that women had a significant role in the production of Wari beer, as was the case in the Inca state. Brewers in the service of the state would have been fed for their labor. Abundant animal remains throughout the structure indicate that food refuse as well as remains of animals of symbolic value were discarded in the building.

Unit 2 is a four-room compound at the northernmost end of the settlement (Feldman 1989; Williams 2001). The architecture is less formal than in some other structures and consists of open work areas and one roofed structure. One of the northern rooms that was originally an open patio served as a trash dump where a considerable amount of faunal material was discarded. The other rooms also had a significant quantity of well-preserved faunal remains.

Unit 7 located just south of Unit 2 is a domestic complex of nine small rooms and corridors. Built on the terraces that slope to the east and at a lower elevation than the monumental residential and ceremonial architecture on the highest part of the summit, the rooms and work areas of Unit 7 are on different elevations, the lowest of which contained refuse from food and chicha de molle (a type of maize beer made with molle seeds) preparation. In addition to everyday faunal staples, the animal remains include some probable nonfood taxa.

Units 5 and 10 are D-shaped structures similar to those previously interpreted as Wari ceremonial temples (Figure 3.2) (Cook 2000). The Baúl D-shaped buildings had tall stone walls and both opened to the northeast (Moseley et al. 2005). The interior of Unit 10, the larger of the two temples, was excavated in its entirety while that of Unit 5 was partially excavated. An agglutinated room associated with Unit 5 was also excavated. The presence of these two ceremonial structures probably relates to Andean dual ayllu (kin and nonkin political group) organization. Phosphate analysis of soils indicates the presence of food and beverages (Coleman 2004). However, the faunal remains are not abundant in either structure, suggesting that the floors were kept relatively clean. There is no evidence for the ceremonial destruction of these temples, as was evident with other buildings, although associated buildings exhibit evidence of burning.

Unit 26 is an elaborate complex in the same walled complex as Unit 10, the larger of the two D-shaped temples. Within a large walled-compound, the temple annex consists of a large central patio surrounded by smaller rooms along the north and south walls. The painted and plastered room in the northwest corner was the most ornate, possibly owing to its use as a funerary locale for the burial of an infant and a prepubescent youth (Moseley et al. 2005). Also found in this room were the remains of an unusual painted ceramic drum in Late Nazca style. South of the funerary room and along the west wall the remains of a subadult camelid offering were found; this represents the most intact offering thus far excavated from Cerro Baúl. Other animal remains include a second, less complete camelid offering, marine fish, and nonlocal animals.

Unit 9, located in the residential area east of the brewery, is the most elaborate palace complex excavated on the summit. The multiroom elite palace featured a large (9 square meters) interior patio that was accessible only through a series of corridors and passageways. The patio floor of cut-and-polished rhyolite slabs was surrounded by benches that provided seating for guests and dignitaries along all four walls. A looted burial of an adult is among the features excavated in the central patio (Williams and Ruales 2002). The five contiguous roofed buildings surrounding the patio also had paved rhyolite floors, indicating a significant labor investment in architectural detail. These rooms yielded a variety of domestic artifacts associated with food preparation, weaving, and other activities. The ritual closing of this structure was accompanied by a spectacular “pot smash” in which over sixty varied ceramic vessels were destroyed. Moreover, the patio floor and surrounding rooms are scattered with the remains of an impressive final feast. The animal remains include the most diverse fauna of any structure thus far analyzed in terms of both species composition and provenance as determined by stable isotopic analysis (deFrance and Nash 2007; Thornton et al. 2011). A radiocarbon date indicates that the structure was “ritually closed” at about AD 800, although the remainder of the site continued to be used for another 200 years.

The current working hypothesis regarding the final feast and building closure is that this event represents an elaborate example of diacritical feasting (sensu Dietler 2001). Presumably, a party of select guests celebrated the interment of the individual in the central patio and then the building was destroyed in an act of reverence and remembrance (deFrance and Nash 2007; Nash 2010).

Southwest of the main palace complex (Unit 9) partial excavations were completed in two associated structures (Units 25 and 40). Employing similar architectural principles to Unit 9, access to an interior patio (Unit 25) was through a narrow corridor. Once inside the open courtyard, visitors could be seated along interior benches, presumably according to rank and status. Along the western wall was a small U-shaped roofed room. Using analogy with Inca ritual space, the U-shaped room was probably a focal point of state ritual (Moseley et al. 2005:17268). A small doorway on the opposite east wall opens into a simple open courtyard that served as a ceramic workshop (Unit 40). The animal remains from these two contexts are not as diverse as Unit 9 but nonetheless include some unique taxa.

Unit 24, located south of Units 40 and 25, appears to be a residential area associated with a household of non-elite status. The architecture is more modest than the buildings to the north. A large rectangular and at least partially roofed structure to the south had a central hearth and a relatively simple burial of an infant. Features, pottery remains, and some stone refuse and tools from lapidary work suggest multiple activities occurred in this compound. Midden material accumulated in some rooms, and well-preserved faunal remains are present.

Zooarchaeological Methods

Identification of the faunal remains from Cerro Baúl was completed using the vertebrate comparative collection housed at the Contisuyo Museum in Moquegua, Peru. The Peruvian National Institute of Culture (INC) granted permission for the export of a small number of faunal remains for which there were no modern comparative skeletal specimens at the Contisuyo museum. These remains were identified using vertebrate comparative specimens housed at the Florida Museum of Natural History, Environmental Archaeology Program, the Division of Mammalogy, or the Division of Ornithology.

All of the remains were identified to the lowest taxonomic level. When possible, camelid specimens were classified as representing either small or large varieties. This distinction was based on measurements and size criteria of modern comparative specimens. The small camelids presumably include the alpaca and the vicuña, whereas the large camelids include the llama and the guanaco. Although there is overlap in the size ranges of these four species of camelids, the categorization of specimens into small and large varieties aids in identifying possible economic and functional variation in how camelids were used.

Methods of quantification include the Number of Identified Specimens (NISP) and estimates of the Minimum Number of Individuals (MNI). The faunal material from different rooms within structures was combined analytically for calculating all measures of relative abundance.

The Cerro Baúl Faunal Assemblage

The Cerro Baúl faunal assemblage from the ten contexts presented here consists of 17,194 identified specimens representing a minimum of 483 individuals. Table 3.2 presents the scientific and common names of all the taxa represented, and summaries of the total NISP and MNI by structure. The assemblage contains the remains of at least thirteen mammalian taxa, seven taxa of birds, one reptile, one amphibian, one cartilaginous fish, and nine taxa of bony fishes (Figure 3.3).

Figure 3.3. Key to animals found at Cerro Baúl.

The remains of camelids and guinea pigs are present throughout the site, indicating that these animals were the subsistence base of the population. In addition to the dietary staples, some other animals also represent food refuse. The remains of several rare and exotic animals are found in limited contexts and are often represented by singular or few elements.

The mammalian assemblage consists of one probable bat, at least two species of rodents, of which the leaf-eared mouse is by far the most common, and the vizcacha, or Andean hare, as well as guinea pigs, puna fox, domestic dog, mountain lion, pampas cat, taruca (north Andean deer), white-tailed deer, and at least two species of camelids. The bat and rodents are interpreted as commensal, nonfood animals. The vizcacha, guinea pigs, two species of deer, and the camelids are all interpreted as food remains. The pampas cat, mountain lion, and canines probably do not represent food animals, but rather, were probably of ritual or symbolic significance.

The camelids include relatively small and very large individuals. However, the majority of the individuals are relatively uniform in size and cluster in the larger size range, suggesting that they are llamas. Using dimensional measurements and size comparisons, I was conservative in identifying specimens as representing small-sized camelids. Specimens in the smaller size range may represent the alpaca or the vicuña. Some very large individuals may be either large male domestic llamas or hunted guanaco. In the absence of hunting paraphernalia it is most probable that the larger specimens are remains of domestic llama. The measurement data do not indicate that size variation in the use of camelids was spatially correlated. As described below, camelids served dual roles as both offerings (i.e., ritual animals) and food animals.

The avian assemblage includes remains of the Andean condor, tinamous, spot-winged pigeon, eared doves, pygmy owl, and the small-sized tyrant flycatcher. Remains of at least two other unidentified medium-sized birds are present. The tinamous, doves, and pigeons are interpreted as food remains, whereas the condor, owl, and songbird probably served nondietary purposes. All of the bird remains are from species found in the highlands and the western slopes; none are from the eastern rainforest habitat.

Reptiles and amphibians are represented by remains of small-sized lizards and a toad. Neither is thought to represent food refuse. I discuss the possible role of toads in Wari ritual below.

Fishes are represented by the remains of one cartilaginous fish (a short-fin mako shark) and by at least nine species of bony fish, including anchovies, shad, herring, sardines, silversides, jacks, flying fish, two species of blennies, lorna (drum fish), jurel, and at least one tuna. The shark is represented by only one large tooth and therefore is probably not food refuse. All of the bony fishes are marine species and are considered to have been food remains.

Intrasite Variability in Faunal Remains

The Cerro Baúl faunal assemblage exhibits great variety in the range of animals present, suggesting that access to diverse food and nonfood animals was a privilege of the elite inhabitants. Dietary animals include nineteen taxa. Abundant remains of camelids and guinea pigs are present in all contexts and vizcachas occur in several contexts. In addition, rare or exotic animals are also present. The zooarchaeological assemblage includes at least ten nonfood taxa that are interpreted as animals that fulfilled ritual and/or symbolic roles.

Spatial variability in the distribution of food and nonfood taxa is present in the ten summit contexts (Figure 3.4). Units 1, 2, 7, 9, and 40 contain the greatest variety of fauna, ranging from seven to twenty taxa, with Unit 9 containing the greatest diversity; the two D-shaped temples (Units 5 and 10) contain the least (n = 3). The Unit 26 assemblage is not very diverse (n = 5), but two camelid offerings occur in this structure. Eight of the exotic and probable nonfood taxa (pampas cat, small feline, puna fox, mountain lion, short-fin mako shark, condor, tyrant flycatcher, pygmy owl) that occur in either Units 1, 2, 9, 24, or 40 are represented by single elements. Units 1, 2, and 9 contain the greatest number and variety of probable symbolic and ritual animals (see Figure 3.4).

Figure 3.4. Distribution of animal remains across the site.

Nonfood, exotic, or symbolic fauna from Unit 1, the large brewery, include the taruca, puna fox, pampas cat, tinamou, short-fin mako shark, jurel, and tuna. A largely intact camelid cranium found along the wall of the brewing room is a probable offering. The cranium was lodged between the stones that supported the cooking vessels used for the production of chicha; however, it was unburned, suggesting that it was deposited after the brewery was no longer in use. The context suggests that the cranium was purposefully placed along the wall, rather than simply representing abandoned butchering waste.

Unit 2 contains remains of domestic dog, mountain lion, Andean condor, toad, and a small unidentified feline. The Andean condor specimen is a cut and polished distal portion of an ulna. Domestic dog is also present in Units 7 and 24, and a specimen from an unidentified canid is present in Unit 25. A large portion of a male taruca cranium is also present in Unit 2. Although deer meat is highly comestible, the presence of only an antlered cranium may indicate that the taruca served some nonfood purpose, such as for display.

The remains of one large toad recovered in Unit 2 are interpreted as nonfood refuse. There was no suitable natural habitat for toads on the summit; therefore, these animals were transported to the site. The recovery of toad remains in only a high-status summit context supports the interpretation that these are remains of animals used for ritual. Like other bufonid toads, the skin covering the backs of Andean Bufo contains parotid glands that secrete toxic substances. These substances include bufotenin (bufotoxin), an alkaloid neurotransmitter (Lyttle 1993). Although Bufo remains in other geographic areas are interpreted as evidence of their use to induce ritual hallucinations (Weil and Davis 1994), others question the ability of the toxin to affect human neurological function (Cooke 1989; Lyttle et al. 1996). It is thus speculation as to whether anuran toxins were ingested by the Wari culture (deFrance 2004). Alternatively, these animals may have played a symbolic role in the life of Wari elites or been associated with water rituals, as has been documented among modern highland populations in the Lake Titicaca basin (Binford and Kolata 1996:45). Although we have a poor understanding of how and why toads were used, iconographic representations of toads with clearly visible parotid glands occur in zoomorphic ceramics from Cerro Baúl as well as at Conchopata, the large Wari capital near Ayacucho (Anita Cook, personal communication), indicating that they were important to the Wari.

Unit 9 has a very large and diverse assemblage of fauna. Uncommon animals include puna fox, white-tailed deer and taruca, Andean condor, pygmy owl, and tyrant flycatcher. The small tyrant flycatcher may represent the remains of a songbird that was used for its colorful feathers. Tyrant flycatchers inhabit a variety of habitats and elevations (Ridgely and Tudor 1994); therefore, this specimen may have been captured locally. Both the pygmy owl and the Andean condor inhabit the western reaches of the Andes, with the pygmy owl occurring primarily at lower elevations.

Unit 9 also has the most diverse assemblage of marine fishes of any summit context. Of the marine fishes, jurel (NISP = 157) are particularly abundant and include many fragments of the neurocranium, suggesting that whole fish were brought to the site. The quantity and variety of marine fish (NISP = 1358, n = 8 taxa) in Unit 9 alone is greater than in any other summit context (deFrance 2004). All marine fish originated from the coastal region roughly 100 kilometers away.

In addition to the abundance of marine fish, Unit 9 also contained the remains of three camelid individuals with bulk carbon and nitrogen isotopic signatures indicating a probable coastal diet (Thornton et al. 2011). These camelids were recovered from the central patio of Unit 9. Along with other abundant food refuse, they may be associated with a diacritical mortuary feast (deFrance and Nash 2007) connected with the ceremonial closure of this structure around AD 800 (Moseley et al. 2005), 200 years earlier than the final site abandonment. These data suggest that either the Wari extracted goods from the coast or that supplicants made the trip themselves, bringing a variety of fish, possibly via llama caravan. Due to restricted isotopic sampling it is not known if other structures excavated more recently (e.g., Units 24, 25, 26, 40) contain camelids that also exhibit coastal dietary signatures.

The faunal remains represented in Unit 26 consist of the remains of an almost complete large-sized subadult Lama cf. glama individual that is interpreted as an offering. The remains of the individual show extensive butchering, with evidence of knife cuts to disarticulate the carcass and probably to remove the flesh. However, the majority of the skeletal elements were left intact (i.e., only a small number of specimens such as ribs were hacked into smaller portions for either preparation or consumption and only some long bones were fractured). Once the specimen was butchered and defleshed, it was rearticulated in approximate anatomical position and interred within the room (Figure 3.5). Interestingly, the age of this camelid—just under three years, based on dental eruption and epiphyseal fusion (Wheeler 1982, n.d.)—corresponds to the age at which modern camelids in indigenous communities are first bred, shorn, and trained to work as pack animals (Wing 1988:169), indicating that the sacrificed animal had reached maturity as an animal of utility. Dimensional measurements of the proximal breadth of the first phalanges and the distal breadth of the scapula are within the size range of measurements for domestic llama reported by Miller and Burger (1995:431–432, 437) (n = 6; average first phalanx Bp is 20.6 mm).

Figure 3.5. Plan view of Unit 26, with detail of llama offering.

A second partially complete camelid individual is present in a second room in Unit 26. The humeri and femora along with portions of thoracic vertebrae, one rib, and a metapodial fragment from a large-sized individual were interred as a “bundle” within one of the smaller rooms. These remains exhibit butchering marks that include hacks (two specimens) and multiple small cuts (two specimens). One humerus and one femur were hacked into smaller portions of approximately one-fourth to one-half of the specimen. The proximal and distal humeri and femora are fused, whereas the centrum epiphysis of a thoracic vertebra is partially fused, indicating that the individual is greater than 3.5 years of age (Wheeler n.d.).

Although no other artifacts are in direct association with the camelid offerings, other rooms within this structure contained the remains of imported artifacts, including a Nazca-style pottery drum, nonlocal pottery, and two child burials (Moseley et al. 2005). The association of the camelid remains with human burials and imported pottery indicates these animals were afforded special status in death or were part of specific set of ritual practices.

Units 24, 25, and 26 have lesser quantities of nonlocal and nonfood animals. All three of these units have abundant camelid refuse as well as the remains of the imported marine fish, jurel. Both of the D-shaped temples have little variety of faunal remains, although Unit 10 also contains jurel. In addition, Units 7 and 24 have dog remains.

Discussion and Conclusions

The distribution of animal remains from ten high-status contexts across the summit of Cerro Baúl indicates that faunal use by the elite inhabitants was characterized by the luxury of variety. Domesticated camelids and guinea pigs were the dietary mainstays, but significant culinary diversity was achieved through the addition of locally hunted mammals (e.g., vizcachas, deer) and birds (e.g., doves and tinamous). The elite residents also imported food animals from distant habitats, most notably several species of marine fish and camelids that probably had been reared on the coast. Although the coastal trading partners of the Wari are not known at this time, some animals were obtained from over 100 kilometers away.

In addition to dietary variety, the remains of several nonfood animals suggest that animal products were used for either display purposes or other symbolic roles. The worked condor ulna from Unit 2 and the butchered condor wing digit in Unit 9 may have been remnants from feathered costumes. The plumage of the colorful flycatcher (Unit 9) might also have been used for clothing ornamentation. The display of feline pelts is suggested by the remains of pampas cat, a small unidentified feline, and the mountain lion from Units 1 and 2. Fox pelts might also have been used for display (Unit 1). Although deer meat might have been eaten, deer crania with antlers could have been powerful symbols or used for display, and the lone tooth of a short-fin mako shark is strongly suggestive of ornamentation. Finally, toads may have been ingested or related to rainfall rituals. Thus, the remains of thirty-four taxa at Cerro Baúl are significant in demonstrating the ability of elites to acquire exotic and wild animals for diverse purposes.

Comparisons of faunal diversity with both local Wari sites and sites elsewhere in the Andes indicate that the pattern of animal use at Cerro Baúl is unique. In contrast to the summit, the taxonomic variety of faunal material from primarily domestic contexts on the slopes of Cerro Baúl is limited, consisting only of camelids, guinea pig, and unidentified bony fish (see deFrance 2004; Moseley et al. 2005). In general, faunal remains from slope contexts are less well-preserved than those on the summit due to greater erosion from wind and precipitation. Also, the structures on the slope were smaller in size with more shallow deposits, which also contributed to greater erosion. Although some possible taphonomic factors contribute to poor bone preservation, the fauna is local and mundane.

Adjacent to, and at a lower elevation than, the elite center of Cerro Baúl, the Wari inhabitants of Cerro Mejía are interpreted as having occupied a lower social niche than that of the elite administrators of Baúl (Moseley et al. 2005; Nash 2002). The site of Cerro Mejía represents a second-tier provincial center in the Wari empire. In contrast to Cerro Baúl, the diet of people who resided on Cerro Mejía was local and mundane, consisting almost exclusively of camelids and three other taxa with no hunted, imported, exotic, or ritual fauna (deFrance 2004; Moseley et al. 2005).

Faunal remains from Wari sites in other areas of the Andes are also far less diverse than that of Cerro Baúl. Ongoing zooarchaeological research by Silvana Rosenfeld at the sites of Conchopata located in the Wari heartland of Ayacucho and at provincial sites near Cuzco indicates a dominance of camelids with little or no use of exotic or imported fauna (Rosenfeld 2011). At Conchopata, camelids are present as are abundant guinea pig remains, some canids, a possible ferret, small and medium-sized unidentified birds, and some amphibians. Rural and administrative sites in the Cuzco region include camelids and some deer, but few other animals (i.e., few guinea pigs, no canids, no birds).

Burial contexts from the site of Beringa in the Majes valley of southern Peru, another provincial center, contain a variety of animal offerings that include some imported or exotic fauna, although the majority of the animals probably were obtained locally (Gladwell 2001). The well-preserved offerings are dominated by camelids but also include a possible fox, a small dog, antlered deer crania, a song bird, two colorful species of macaws (probably from the eastern Andes), worked bird bone from large-sized birds (e.g., vulture, flamingo), and unidentified riverine or marine fishes. The Beringa assemblage is most similar to Cerro Baúl in terms of diversity and evidence for long-distance trade in fauna, indicating that elites in some other areas of the Wari realm were also able to obtain a variety of animals.

Wari political economy and symbolic life incorporated local and exotic animals that fostered social inequality. The diverse pattern of animal use exhibited at Cerro Baúl is significant among the regional Wari sites and those elsewhere in the Central Andes. Elites controlled local pastoral production, the rearing of guinea pigs, and some aspects of hunting. They also engaged in trade with distant populations or accepted animals or animal products from guests as gifts or payments, particularly from coastal regions. Life for the elite inhabitants of Cerro Baúl consisted of a table of plenty along with personal and public ornamentation that used feathers, fur, and teeth. In the South Central Andes, one means through which the Middle Horizon Wari created and institutionalized inequality was through the luxury of variety in animal use.