Tiahuanaco

By Paul Goldstein and Sarah Berry

Tiahuanaco, Tiahuanaku

ca. 1600 B.P. - 900 B.P.

Follows the Andean Regional Development tradition and precedes the Aymara Kingdoms and Andean Regional States traditions. In the terminology of the Tiwanaku region, the tradition includes the terminal "Early Intermediate Period" and "Middle Horizon" and precedes the "Late Intermediate Period."

The South Central Andes. The Tiahuanaco type site is at an elevation of 3800 m., in the southern Lake Titicaca basin of Bolivia. The Tiwanaku Core Region extends through most of the Peruvian and Bolivian Altiplano. The Tiwanaku peripheries include lowland regions of Southern Peru, Northern Chile, and Eastern Bolivia.

Stone sculptural traditions of anthropomorphic stelae and low relief carvings. Cut-stone public architecture including sunken rectangular courts and pyramidal or stepped platform mounds. Settlement pattern hierarchy with large urban site, secondary centers, and farmsteads with raised field agricultural systems in highlands. Serving and ceremonial pottery including "KERO" drinking goblets, anthropomorphic and zoomorphic vessels, polychromes typically red- slipped with motifs in black, white, orange and blue-gray. Characteristic utilitarian plainware pottery. Camelid wool and cotton textiles including polychrome striped warp-faced plain weaves and figural tunics of interlocked tapestry technique, both with loop stitch embroidery. Bronze, gold and silver metallurgy, diverse lithic and lapidary industries.

Tiwanaku Core Region, Cochabamba Tiwanaku, Moquegua Tiwanaku, Azapa Tiwanaku (Loreto Viejo/Cabuza), San Pedro de Atacama Tiwanaku.

Tiahuanaco, Lukurmata, Pajchiri, Iwawe, Omo, Azapa AZ-83, Piñami, San Pedro de Atacama.

Climate in the Altiplano Tiwanaku Core Region exemplifies extreme high altitude at a tropical latitude, with an annual mean temperature of 9° C and diurnal variation of up to 15° C due to extreme solar radiation. Night frosts are a significant limiting factor. Altiplano rainy season is from November through March, with a mean annual precipitation approaching 700 mm. Tiwanaku's hyper-arid western peripheries in Peru and Chile enjoy mild temperatures year- round, but are climatically constrained by the near-total rain shadow of the Pacific coast. Periodic El Niño Southern Oscillation (ENSO) events disrupt these normal patterns with highland droughts and coastal floods. Long term trends suggested by glacial ice cores correlate the Tiwanaku tradition with two relatively wet periods from 1340 - 1300 B.P. and 1190 - 910 B.P. Decreasing rainfall inferred after 910 B.P., culminating in severe drought and low lake levels from 705 to 640 B.P., have been implicated by Kolata et al. in the Tiwanaku collapse. However, Erickson argues that most Tiwanaku site abandonment took place 100 to 200 years before the onset of this drought. Major flood events in Moquegua, dated by Magilligan et al. to 1290 B.P. and 620 B.P. indicate that extreme ENSO cycles occurred during the Tiwanaku era.

The Tiwanaku Core Region is in the Altiplano, a flat depression that runs 800 km north to south between the Eastern and Western Cordilleras of the southern Andes. The Tiahuanaco site is located in an 11 km wide valley formed between the east - west running Kimsachatta and Achuta ranges. Elevations above 3800 m. for Tiahuanaco and nearby sites by the shores of Lake Titicaca make the Tiwanaku Core Region the world's highest ancient civilization center. Tiwanaku settlements are also found at lower elevations in the Pacific-draining river valleys of Tambo, Moquegua, Locumba, Sama, Caplina, Lluta and Azapa in the coastal desert of Peru and Chile. No sites have been found on the Pacific littoral, with colonial settlement concentrating inland in the steeply sloping oasis valleys. Variants of the Tiwanaku tradition also appear in the wetter Amazon-draining eastern slope valleys of Cochabamba, Mizque and Capinota in Bolivia.

The south central Andes are a tectonically active region, where uplift and volcanic activity are caused by the collision and subduction of the eastward- moving Nazca Plate with the South American continent. The geology of the south central Andes is highly variable, but might be typified as uplifted Pre- Cambrian conglomerate formations mixed with metamorphic and igneous rocks created by intrusive volcanics. There are a number of historically active volcanoes in the region, and significant mineral deposits of tin, silver, copper, gypsum and salt are found in the southern Bolivian Altiplano, northern Chile and southern Peru. The Altiplano of the Tiwanaku Core Region was formed by conglomerate and sandstone sediments deposited in the depression formed by the uplift of the Eastern Cordillera. Subsequent Early and Late Pleistocene lakes, known as Ballivian and Tauca, respectively, covered the altiplano to a level 50 m above the current level of Titicaca, leaving stranded terraces now visible in the Taraco Peninsula. Evaporation and sedimentation through the Holocene have left salt flats covering much of the southern Altiplano, and continue to decrease the levels of Lake Titicaca and the increasingly saline Lake Poopo.

The Tiwanaku Core Region comprises as many as seven micro-environmental zones, including marshes and periodically inundated lands in the Titicaca basin, extensive flat and rolling grasslands, foothills, alluvial terraces and colluvium. Predominant flora consists of Stipa ichu and a variety of other grass species, the lake reed TOTORA, and deciduous bushes such as THOLA and KHOA. Indigenous fauna include the domesticated and wild camelid species (llama, alpaca, guanaco and vicuña), rodents such as guinea pigs and viscacha, a wide variety of birds including hawks, condors and owls, rhea, flamingoes, ducks and other water fowl, and freshwater fish.

The first order site of Tiahuanaco, at least 420 hectares in area with a population approaching 40,000, dominated the Tiwanaku Core Region settlement pattern. Interpretations differ on the systemic integration of second order sites like Lukurmata, Pajchiri, Khonko Wankane, Chiripa, Iwawe and Omo, few of which exceeded 10,000 in population. Ponce Sangines and Kolata describe them as secondary centers in a highly hierarchical four-level Tiwanaku settlement system corresponding to state agricultural investment and administrative functions. McAndrews et al's. recent rank-size and cluster analyses of Tiwanaku Core Region settlement pattern data suggest a less-integrated arrangement in which the second order sites were the centers of autonomous sub-systems. A similar range of interpretations exists for settlement systems of the Tiwanaku peripheries. In Moquegua, Goldstein notes the sudden appearance of four large Tiwanaku towns, agricultural works, and specialized facilities for regional provincial administration. In contrast, Higueras notes virtually no change from pre-existing local settlement pattern in the Mizque and Capinota valleys of Cochabamba with the advent of Tiwanaku material culture.

The Tiahuanaco site's ceremonial, administrative and residential palace precincts were laid out on a cardinal orientation and surrounded by a moat to separate them from the rest of the city. Elite residential complexes have been identified on the summit of the Akapana pyramid and in the Putuni Palace, and it is likely that the monumental center was reserved for royal or high status residents. The four square kilometers of surrounding non-elite residential sectors at Tiahuanaco maintained the same cardinal orientation as the monumental center. Residential groups were bounded by perimeter walls into socially or functionally differentiated barrios, which in turn were subdivided into patio groups. These are believed to represent segmentary social units corresponding to AYLLUS or kin groups within the city. Similar community segmentation appears to be the case in other Core Region settlements like Lukurmata.

In Moquegua, desert preservation permits the exposure of entire Tiwanaku town plans, including extensive residential districts and ceremonial/administrative structures. Omo Phase occupations at the Omo site, dated to 1350 - 1200 B.P. consisted of freestanding multi-room buildings, arrayed around three large plazas. These plaza-centered residential groups appear to correspond to intentionally segregated AYLLU communities.

Tiwanaku Core Region domestic structures were constructed of adobe over stone foundations, with packed earth floors and informal hearths in each habitation room. Although circular structures are reported in later levels at Lukurmata, rectangular structures predominated. Dispersed single-structure household units were typical of earlier occupations. Walled compounds appear within Tiahuanaco's Akapana East residential center after 1350 B.P., and similar compact patio groups, consisting of two domestic structures, a storage building and sub-floor burials, appear at Lukurmata between 1200 and 1100 B.P. In both cases, it is believed that the reorganization from a minimal household plan into larger compounds corresponds to a coalesced family structure that maximized super-household production in the period of maximum state tribute. Lowland Tiwanaku settlers used cane wall and timber post construction rather than adobe in most domestic architecture. Early household units at Omo did not have independent patios or storage facilities. As in the Core Region, autonomous patio groups with contiguous roofed rooms, open patios, and storage units (mud-plastered stone cists or rectangular cribs) appeared after 1200 B.P., and may indicate changes in household organization, size, and production in later settlements. In the later phases, steeper domestic sites were terraced for occupation with stone retaining walls.

A comprehensive study of non-metric traits by Blom found no significant biological distance among skeletal populations of Tiahuanaco, Lukurmata and rural sites of the Tiwanaku Valley and the Katari Basin. This indicates relatively open migration and mating among Core Region majority populations. Small groups from the Akapana East and Mollo Kontu sectors at Tiahuanaco, who exhibit different mortuary practices and possibly biological distinction, could represent enclaved foreign populations within the city. The Tiwanaku population of the Chen Chen site of the Moquegua Valley was far closer genetically to that of the Tiwanaku Core Region than to non-Tiwanaku peoples of Moquegua, confirming that peripheral settlements were peopled by colonists and their descendents, and not acculturated local populations.

Most Tiwanaku populations practiced intentional cranial deformation, using annular binding or devices that produced variations of fronto-occipital flattening and parietal widening. Tiwanaku cranial modification styles correlated with regional, ethnic or clan affiliations rather than status, and stylistic variation has been noted among regions and among individual cemeteries. As might be expected for a cosmopolitan city, the Tiahuanaco site displays the greatest heterogeneity of styles.

Tiwanaku health was generally good, and no major epidemics or violent episodes have been identified. Child and infant mortality was relatively high, in keeping with other pre-Columbian societies. Periostosis and porotic hyperostosis suggest that Tiwanaku peoples suffered from moderate rates of anemia and systemic infections, perhaps more severely in lowland areas. Dietary differences may account for higher rates of caries in the lowlands and more extreme tooth wear in the Core Region.

Tiwanaku's Core Region depended primarily on the large scale herding of the domesticated camelids llama and alpaca and the cultivation of specialized frost-resistant crops in labor-intensive raised field systems. Raised field systems covering over 120,000 hectares in the Altiplano were the principal staple subsistence base for the Tiwanaku Core. Raised fields mitigated nightly frosts by storing solar energy in the water-filled swales between fields, managed drainage, and permitted fertilization of crops with organic materials dredged up in canal maintenance. Experiments by Erickson and Kolata's teams suggest that these advantages can dramatically decrease failure risk and increase yield over flatland agriculture in the Altiplano.

Lacustrine and some terrestrial wild foods, hunted with small reed water craft, snares, bow and arrow, darts, slings, and bolas, played a significant supplementary role in Tiwanaku diet. Husbandry of the lacustrine flora and fauna may have been a by-product of the extensive artificial wetlands created in the swales of the raised field systems.

Tiwanaku agriculture in the temperate lowlands of the Pacific slope concentrated on irrigation near the floodplains of the riverine oases. In Moquegua, Tiwanaku farmers modified the landscape to reclaim extensive areas of desert through the construction of laterally extended canal systems. A single preserved canal and field system investigated by Williams at the Chen Chen site alone irrigated over 90 hectares, and comparable systems may be assumed at other sites in the valley. Unlike Wari, Late Intermediate, and Inca agriculturists in the same region, the Tiwanaku seldom resorted to terracing of steep terrain, preferring to extend canals to relatively flat areas. Marine resources, such as mollusks, seaweed, bird guano fertilizer, seabirds, fish and sea mammals, were utilized in lowland Tiwanaku settlements. These were probably obtained through occasional foraging trips or exchange with maritime specialists, as no Tiwanaku colonies have been identified on the littoral, and there is no evidence of any Tiwanaku maritime technology.

Water birds composed a significant dietary contribution in the Tiwanaku Core Region, and freshwater fish and amphibians are also well represented in the faunal record. Reed plants were harvested for industrial purposes. Deer, the wild camelid species of vicuña and guanaco, and small rodents and amphibians were also consumed. Choromytilus, other sea mollusks, river crustaceans, and possibly fish, seabirds and sea mammals were consumed in the Moquegua and Azapa peripheries.

Agriculture in the Tiwanaku Core Region relied on a complex of frost-resistant tubers including potatoes, OCA, OLLUCO, and MASHUA and chenopod grains such as QUINOA. A freeze-drying process utilizing sunlight and frosts permitted the preparation and long term storage of tuber crops in a dehydrated form known as CHUÑO. Because of high altitude frost, maize could not be a significant staple cultigen in the altiplano, although it may have grown in protected areas near Lake Titicaca.

Maize was nonetheless extremely significant as the source of CHICHA, or maize beer, which ceramic evidence suggests was consumed in quantity in throughout the Tiwanaku tradition. Maize, along with hot peppers, coca leaf, peanuts, and beans was cultivated at low elevation Tiwanaku settlements in the eastern and western Andean slopes. Sandness' isotope analysis of human remains from the Omo site suggests that lowland Tiwanaku colonists consumed significant quantities of maize.

Domesticated food animals included llamas, (also used as pack animals), alpacas (valued for wool), guinea pigs, and probably ducks and other waterfowl.

Pottery production, despite considerable regional variation, can be considered a single technological tradition. Both Tiwanaku fine serving wares and utilitarian plainware pottery were dense, moderately hard, low-fired terra cottas, their principal differences being varieties, amounts, and sizes of sand temper, vessel thickness, and the quality of burnishing and surface treatment. Most fine serving vessels were red-slipped, often with polychrome decoration with principal motifs in black and subsidiary decoration in somewhat translucent white, orange and blue-gray mineral pigments. Plain and decorated wares at Chiji Jawira, a potter's barrio at Tiwanaku, were fired in open pits, using camelid dung and grasses for fuel. Elsewhere, however, a burnished black serving ware was produced through smudging or reduction firing that may have required different firing conditions or fuels.

Ceramic serving vessels included KEROS or drinking goblets, TAZONES or flaring sided bowls, basins, small pitchers, and modeled anthropomorphic and zoomorphic drinking vessels, each of which also appears in wood and metal. Ceramic ceremonial burners were used to burn llama fat or other offerings. Utilitarian plainware pottery, which accounts for roughly 90% of sherds in domestic and midden contexts, included several sizes of globular OLLAS , cooking vessels with horizontal rim strap handles, and cylindrical neck TINAJAS, storage and brewing vessels with horizontal body handles. Full sized flat handle wooden spoons, often decorated with a llama silhouette or geometric motif, were commonly used with TAZONES for eating.

Spinning was done on drop spindles with wooden shafts and whorls made of wood, drilled sherds, or specially made pottery whorls used for finer thread. Weaving tools such as picks or shuttles were typically made of camelid limb bones. Most Tiwanaku textiles were probably made on staked looms, although backstrap looms may have been used as well. As in most Andean cultures, textiles were used as woven or as two joined panels, and never cut or tailored.

Because of poor highland preservation, Tiwanaku textiles are best known from desert sites in Chile and Peru. Plainweave all-cotton textiles, found in domestic middens at the Omo site bore no decoration, were never included in tombs, and appear to have been of utilitarian function. Tiwanaku-contemporary cotton textiles have also been reported by Oakland in the Cochabamba region. Most decorated textiles were of camelid wool or occasionally cotton-wool composite, either in natural colors or dyed red, blue, purple, green, yellow, and white. Warp-faced plainweave tunics and blankets were most commonly decorated with warp stripes. Supplemental warp decoration does appear, although it appears to be more popular in post-Tiwanaku cultures. A characteristic loop stitch embroidery of geometric or figural motifs was used to finish selvages and edges. The most labor-intensive textiles were interlocked tapestry tunics with figures depicting mythical themes. These are rare, and usually associated with elite contexts.

Bone and stone tool industries provided most utensils in daily use. Bermann and Janusek have documented specialized workshops at Lukurmata that produced a characteristic hafted utensil made from the snapped mandible of a camelid and pan pipes. Musical instruments also included ceramic whistles and cane and bone flutes and pan pipes.

Tiwanaku ground stone utensils included bowls and palettes, mortars and pestles, manos and metates, and some very large rocker grinders in Moquegua probably used for maize. Chipped stone flake tools, and hoes were casually made from chert, basalt or volcanic stone in very large numbers.

Tiwanaku hunters and warriors used wooden bows and wood or cane arrows with narrow-stemmed points and double fletching with feathers. Atlatls with darts and pecked slingstones, woven slings, and grooved bola stones were also used. Characteristic to Tiwanaku are TROMPOS, top-like conical or cylindro-conical pecked stone or wooden objects of unknown function.

Bundles or kits related to hallucinogenic drug use have been found in elite Tiwanaku burial contexts, and occasionally in middens. These include decorated textile bags, wood, stone or bone snuff tablets, and small bone spoons, brushes, tubes and containers. Larger rectangular bags, carrying cloths and cordage were used to wrap coca leaves for practical purposes and for offering bundles.

Tiwanaku stonemasons produced some of the finest monuments in the Andes. Cut prismatic blocks were precisely fitted to face the sunken court temples, enclosures, and terraces of large pyramid structures at Tiahuanaco and Core Region sites. Figural stone carving included low relief decoration and three- dimensional sculpture that developed out of previous Altiplano sculptural traditions of Pukara, Chiripa and other sites. Massive rectangular abstract style humanoid stelae, including the seven meter Bennett Monolith, the largest stone sculpture in the New World, were carved from local sandstones with stone and bronze tools. Like lintels, jambs and other architectural elements, the classic Tiwanaku stelae were decorated in low-relief with figures of deities and geometrics that may represent tapestry tunics, tattoos or face paint.

Tiwanaku metallurgy was advanced, including gold, silver and copper, and bronze alloys of copper-arsenic-nickel, copper-tin and copper-arsenic, but was largely limited to ornamental purposes. Metal joining techniques did not attain the complexity achieved in the Northern Andes, and most small tools and ornaments such as pins, needles, axes and small figures were formed and hammered in one piece. The most common technique for decoration of metalwork was repoussé used on thin beaten sheet ornaments, headdresses and vessels. Bronze architectural cramps were cast and hammered into carved depressions in cut stones to join architectural elements of monumental buildings, and some gold decorative elements may have been added to stonework in a similar fashion.

Clothing, headgear, jewelry and body decoration marked Tiwanaku social and status identities. Both males and females wore sleeveless rectangular tunics and blankets, which women fastened with metal or cactus spine pins. Undecorated cotton plain weaves may have served as everyday clothing or underwear. Tapestry tunics, the most labor-intensive textile, were rare, and their representation in low relief, dressing Tiahuanaco's major stelae, supports that they were reserved for elites. Ceramic portrait head vessels depict often-mustachioed males with complex facial painting, wearing large cylindrical earspools, lip plugs, four-pointed hats or turbans. Wooden and ceramic pigment kits containing red and yellow ochres have been found in the Azapa and Moquegua valleys. Specialized workshops at Tiahuanaco and in Moquegua worked in rare and imported lapidary materials, producing metal headdress ornament, pendants, TUPUS, or cloak or turban pins, and drilled beads or inlay pieces of lapis lazuli, malachite, sharks' teeth, and seashell, including Oliva Peruvianus , Choromytilus Choro and occasionally Spondylus. Flamingo and other feathers found in domestic and midden contexts were probably used ornamentally as well.

It is generally accepted that Tiwanaku did not have a developed entrepreneurial market system, and that most trade involved either reciprocal relations among kin and affines or state-sponsored exchange. Llama caravans and human transport linked the Tiwanaku Core Region with the eastern and western peripheries. Except for Lake Titicaca, where water transport is possible, Andean terrain and the caloric requirements of llamas probably made bulk long distance transport of staples impractical. Exceptions would have been lowland crops such as maize, coca and hot peppers, which were highly valued in the Core Region for ceremonial use or as condiments. Long distance exchange either through traders or intermediaries obtained exotic raw materials such as Spondylus shell from the Ecuadorian coast, obsidian from sources in the Colca valley near Arequipa, Peru, lapis lazuli and copper from northern Chile and perhaps medicinal plants, hallucinogens, bird feathers and exotic wildlife from the Amazonian lowlands. Tiwanaku sumptuary objects such as snuff kits and tapestry tunics, often bearing complex Tiwanaku iconography, were traded to non-Tiwanaku elites in San Pedro de Atacama in Chile and perhaps as far as Northwest Argentina. Exotic pottery found at Tiahuanaco suggests trade or resident foreigners from several eastern Bolivian regions.

Tiwanaku crafts were produced both by individual households and by specialists. Simple bone and ground stone tools, flake implements, and stone hoes were probably made casually by their users, and most plain textile spinning and weaving also took place in the household. It has not been possible to identify gender-specific production activities except from iconography and grave goods, which suggest male rulers, priests and curers, and female weavers.

Production of more elaborate crafts was by corporate specialist groups residing in urban barrios or rural settlements. Pottery production was specialized to workshop groups like the Chiji Jawira barrio at Tiahuanaco, who oversaw all phases of production and maintained distinctive sub-styles. Janusek suggests that such co-residential crafts specialists may have corresponded to descent or ethnic groups, like AYLLUS, embedded in Tiwanaku's segmentary social sub-structure. Nonetheless, elite-attached specialists are also probable for highly skilled, labor intensive or sumptuary crafts like tapestry weaving, fine lithic production, bronze, gold and silver metallurgy, and lapidary industries. Workshop evidence in Moquegua suggests that stemmed chert projectile points were produced as preforms near the point of quarrying, and transported to other locations for finishing, and lapidary work was specialized as well.

It may be assumed that pasture and farmlands were held by households or corporate groups like AYLLUS, and it is likely that differential access to prime lands and herds among these groups was the basis for social stratification within Tiwanaku society. Ready access to labor for the construction and upkeep of public works and the collection of surplus to support the urban elite and their retainers would have been critical for maintaining political power. Tiahuanaco's ruling elites lived in palaces within the monumental center and enjoyed preferred access to labor-intensive sumptuary goods and those of exotic raw materials. Few intact elite tombs have been recorded by archaeologists in the Core Region, however, museum and private collections reflect looted elite burials with offerings of numerous metal objects. Provincial Tiwanaku sites have not revealed elite residences, although residential sectors enjoyed preferred access to elaborate pottery. Elite burials in Cochabamba and Moquegua also suggest differential access to metal objects, beads, and tapestry tunics.

The Tiwanaku-period Titicaca Basin encompassed diverse ethnic communities, and these were probably also subdivided structurally at several levels. Kolata argues that a social hierarchy grew out of a coalition of proto-Aymara herders who formed Tiwanaku's elite with lower status Pukina agriculturists and Uru aquatic specialists. Spatial variations of ceramic style and household ritual between barrios within Tiahuanaco are seen by Janusek to represent diverse social segments subsumed within the hierarchy of the Tiwanaku polity. Similar spatial division into socially-constituted sectors has also been noted by Goldstein in lowland Tiwanaku habitation sites. Most scholars believe that the fundamental social unit was the AYLLU, a descent-based segmentary group that pooled labor and resources above the level of the household.

Recent fieldwork in the Tiwanaku Core Region has engendered debate on the strength and centricity of the Tiahuanaco site as the capital of a bureaucratic and hierarchical state. Analysts focusing on the Tiahuanaco type tend to emphasize the centripetal nature of the capital city and a strongly centralized Tiwanaku state. Researchers focusing on smaller sites of Tiwanaku's Core Region have emphasized more segmentary interpretations of Tiwanaku as a loose ethnic confederacy of politically autonomous communities who shared only cultural and ceremonial ties to the capital.

State or elite sponsorship of festivals, performance of ritual obligations, effective redistribution of surplus and elaborate gift giving may have been critical for maintaining political power. In exchange, the Tiwanaku populace contributed labor for public works and surplus to support elites and specialists. There is relatively little evidence for coercive social control, and much of this relationship may have been consensual. Most evidence suggests social controls dedicated to maintaining distinctions among Tiwanaku's component ethnic or social groups, which were separated by walls or moats within Tiahuanaco, and identified with specific styles of dress, material culture, and cranial deformation.

Relatively little direct evidence of external warfare has been demonstrated for Tiwanaku. Evidence of possible human sacrifice on the Akapana pyramid, reworked cranea curated in Tiahuanaco domestic contexts and CHACHAPUMA stone sculptures that depict decapitator feline figures may point to sanctioned violence in Tiwanaku. However, iconographic evidence for organized combat, sacrifice or the taking of trophy heads is far rarer than in comparable Peruvian cultures like Moche and Nasca. Core Region Tiwanaku sites were not walled or defensibly located, and no correlation of Tiwanaku culture change with incidence of skeletal trauma, site burning or increased presence of weapons has been observed. Deliberate site destruction does accompany the abandonment of residential and ceremonial sites in Moquegua ca. 1000 B.P. and may indicate a generalized rebellion or attack by outsiders. The peripheral Tiwanaku collapse was followed by a transfer of settlement to walled and defensible hilltop sites.

Tiwanaku textile, stone and ceramic iconography suggests a set of state deities associated with the Staff or Gateway God, a frontal figure depicted wearing an elaborate headdress and holding a puma-headed staff in either hand. The Staff God, and a variable group of winged figures that accompany him on Tiahuanaco's archetypal "Gateway of the Sun" and numerous other carved architectural elements have been interpreted as a pantheon of sky, weather, mountain or cosmic deities, possibly ancestral to the ethnohistorically known deities Viracocha and Illapa. Portable depictions of the staff god on pottery and in textiles accompanied Tiwanaku long distance exchange and the colonization of new territories, but disappeared from the tradition at the time of societal collapse, supporting the god's association with the power of the Tiwanaku state.

Beyond state ideologies, the Tiwanaku tradition also encompassed enduring religious practices on the household and community level. These included the sacrifice of pottery, jewelry, and animals to dedicate homes, temples and fields. These offering traditions strongly resemble ethnographically recorded Aymara rituals dedicated to local mountain, earth and landscape deities.

It is believed that Tiahuanaco became an important center for pilgrimage, as well as a capital, and the city's monumental core is famous for terraced pyramids, platforms, enclosures, and sunken courts where religious ceremonies took place. Both at Tiahuanaco and at other Tiwanaku tradition sites, Conklin proposes that the architectural plan of temple structures suggests processional paths from public spaces through narrow doorways and staircases to progressively more restrictive sanctuaries. The inner temple complexes, typically walled sunken courts, were the location of the most exclusive ceremonies and housed the most elaborate stone sculptures or other sacred objects. Hallucinogenic drug use, probably restricted to ritual specialists in private contexts, played a role in prognostication, propitiation or curing ceremonies. Both religious and mundane buildings were consecrated with buried offerings of young or fetal camelids, ritually killed pottery, metal pins and other artifacts, amulets including architectural models, seashells, and bundled coca leaves, and by ritual burning of llama fat in ceramic censors.

One mass human interment excavated by Manzanilla on the Akapana pyramid may represent a human sacrifice to dedicate the pyramid's construction or to commemorate an event.

Ceramic evidence throughout the Tiwanaku tradition and analogy to Inca practices suggest that popular celebrations involved politically sponsored feasting and mass consumption of maize beer in civic, household, or rural contexts. Beer-drinking paraphernalia such as KERO goblets are found in domestic middens throughout the Tiwanaku tradition, but also occur in concentration in Moquegua in particular households that may have been ceremonial CHICHERIAS, near public plazas, and outside of settlements on promontories with views of temples or important landscape features. It is possible that restrictive and the public ceremonies may have been celebrated simultaneously by placing sunken courts out of sight at the top of artificial pyramids or platforms that were visible from below.

Complex figural iconography, such as portrayals of the Staff or Gateway God, appears on only a small proportion of Tiwanaku works in architectural stone, textile, metal and ceramic media. Although temples and palaces were stuccoed and painted in solid red, green and yellow pigments, no Tiwanaku mural painting has yet been found. Cut stone architecture was commonly ornamented with recessed jambs and lintels and deep carvings of crosses, step motifs or other motifs. When figures were carved, usually in low relief on stone architraves, there is no use of perspective and it is difficult to decipher any narrative or thematic organization of scenes. The same would apply to figures depicted on tapestry textiles and pottery.

In painted pottery and wood vessels, geometric motifs predominate, the most common being opposed step-stair designs that may represent terraced pyramids or mountains with cosmological embellishments. Spiral, volute and cross motifs are also common. Painted and wood-carved representations of llamas, felines , flamingos and humans are less common, and tend to be highly abstracted or rectangularized to fit panels.

In-the-round stone sculptures were highly stylized, and most analysis has focused on the representation of the humanoid stelae figures' dress, adornment, and utensils such as KEROS lime dippers and hallucinogenic snuff tablets, and the knives and decapitated heads held by feline figures known as CHACHAPUMAS. In contrast, many modeled ceramic vessels depicting human heads, full-body llamas, felines, and ducks and the heads of felines and eagles, often displayed great sensitivity and realism. It appears that at least some human representations on drinking vessels are true portraits of individuals. Relatively crude plainware figurines of females with long braids and deformed cranea are also found.

There is good evidence that Tiahuanaco curated stone sculpture of ancient and foreign styles, perhaps by practicing HUACA capture, the imprisonment of revered idols of subjugated peoples. Kolata argues that sculpture collections of varied styles depicting diverse costumes, such as the group of stelae and tenon heads collected at Tiahuanaco's Semisubterranean Temple, may have been assembled as microcosmic social maps of the Tiwanaku world.

Most Tiwanaku burials were individual primary interments, buried with offerings in simple pits or stone-lined cylindrical cists. Individuals were wrapped in one or more tunics or blankets, tightly bound with cordage, and interred in a seated, flexed position with offering objects, that typically included one or more KEROS, TAZONES or other serving vessel, small plain OLLAS, and wooden spoons. Less commonly, jewelry items, weaving implements, pigment boxes, musical instruments or drug paraphernalia were included. An east-facing orientation was customary in Tiwanaku interments in Moquegua, Lukurmata and the Coyo Oriental and Quitor cemeteries of San Pedro de Atacama. In the Tiwanaku Core Region, burials were placed either within household units or in separate cemeteries, while in the peripheries of Moquegua, Azapa, and San Pedro, burials were segregated to nearby cemetery areas and tombs were marked with wooden posts. The proximity of the interred to domestic space and the marking and probable maintenance of many tombs suggests a concept of the afterlife in which ancestors played important roles in the affairs of the living.

Documents referred to in this section are included in the eHRAF collection and are referenced by author, date of publication, and eHRAF document number.

The Tiahuanaco collection consists of 19 documents, all in English. The documents discuss the site and culture of Tiahuanaco (also spelled Tiwanaku) near Lake Titicaca in Bolivia and Peru. The collection covers the time period from 1900 B.P. to 700 B.P. (100 A.D. to 1300 A.D.) which is outside the absolute time period of 1900 B.P. to 800 B.P. The source that is outside the absolute time range is pointed out below.

Kolata (1993, no. 7) presents an in-depth overview of Tiwanaku culture and its rise and fall. This source covers the time period of 1900 B.P. to 700 B.P. to more fully explore the Tiwanaku collapse. Bennett (1934, no. 2) discusses his excavations at Tiahuanaco in-depth. Bermann (1994, no. 3) discusses the results of his fieldwork at Lukurmata, a second-order site in the Tiwanaku hierarchy. Goldstein (1989, no. 4) writes of his fieldwork in the Moquegua Valley, a Tiwanaku colony. Janusek (1999, no. 5) examines craft specialization, comparing the sites of Lukurmata and Tiwanaku.

The rest of the sources all explore raised field agriculture; many authors considered this the economic foundation of the Tiwanaku state. Albarracin- Jordan (1996, no. 1) surveyed the lower Tiwanaku Valley and he discusses both raised field agriculture and settlement patterns. Kolata (1986, no. 6) presents the results of his fieldwork in an area of raised fields. The rest of the works are from a book (1996) on Tiwanaku raised fields and the paleoecology of the southern Lake Titicaca basin. Kolata (no. 9) describes the research project and (no. 10) the environment. Analysis of the geology, soils, and water are discussed by Argollo (no. 11), Binford (no. 12), and Heath (no. 15). Ortloff (no. 14) examines the canals and other devises that were used in raised field agriculture. Kolata and Ortloff (no. 13) explore the distribution of the raised fields in the Catari and Tiwanaku river valleys. The collapse of the Tiwanaku state is discussed in Kolata (no. 16). More modern, ethnographic research and experimental archaeology was conducted by Kehoe (no. 19) and Kolata (no. 17) as the raised fields are again being used by the Aymara in the area. Finally, Kolata (no. 20) presents some social models as to how the Tiwanaku state may have run.

Kolata (1996, no. 8) is the bibliography for documents nos. 9-20. For further information on individual works in this collection, see the abstract in the citations preceding each document.

The major tradition summary is from the article, "Tiwanaku," by Paul Goldstein, in the Encyclopedia of Prehistory, Peter N. Peregrine and Melvin Ember, eds. New York: Plenum Publishing Corporation, 2001-2002. We thank Paul Goldstein and Peter N. Peregrine for bibliographic suggestions. Sarah Berry wrote the synopsis in 2002.

Albarracin-Jordan, J. Tiwanaku Settlement Systems: The Integration of Nested Hierarchies in the Lower Tiwanaku Valley. In: Latin American Antiquity. 3(3), 1996: 183-210.

Bennett, W. C. Excavations at Tiahuanaco. In: Anthropological Papers of the American Museum of Natural History, 34(3), 1934: 361-493.

Bennett, W. C. Excavations in Bolivia. In: Anthropological Papers of the American Museum of Natural History 35(4), 1936:329 - 508.

Berenguer, J. R., and Percy Dauelsberg H. El Norte Grande en la órbita de Tiwanaku (400 a 1,200 d.C). In Culturas de Chile, Prehistoria Desde sus Orígenes Hasta los Albores de la Conquista, edited by J. Hidalgo L., V. Schiappacasse F., H. Niemeyer F., C. Aldunate, I. Solimano R., pp. 129-180. Editorial Andrés Bello, Santiago de Chile. 1989.

Bermann, M. P. Lukurmata: Household Archaeology in Prehispanic Bolivia. Princeton:Princeton University Press, 1994.

Blom, D. Tiwanaku and the Moquegua Settlements: A Bioarchaeological Approach. Ph.D. Dissertation. PU: Chicago: University of Chicago, 1999.

Browman, D. L. Cultural Primacy of Tiwanaku in the Development of Later Peruvian States. In: Diálogo Andino 4, (La problematica Tiwanaku Huari en el contexto panandino del desarollo cultural, edited by M. Rivera), 1985: 59-72.

Browman, D. L. Political Institutional Factors Contributing to the Integration of the Tiwanaku state. In Emergence and Change in Early Urban Societies, edited by L. Manzanilla, pp. 229-243. New York: Plenum, 1997.

Chavez, S. J. The Arapa thunderbolt stelae: a case of stylistic identity with implications for Pucara influences in the area of Tiahuanaco. In: Nawpa Pacha, 13, 1976: 3-25.

Conklin, W. Tiwanaku and Huari: Architectural Comparisons and Interpretations. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by W. Isbell, Gordon McEwan, pp. 281-292. Washington, D.C.: Dumbarton Oaks, 1991.

Eisleb, D., and Renate Strelow. Altperuanische Kulturen III: Tiahuanaco. Berlin: Museum fur Volkerkunde, 1980.

Erickson, C. L. Neo-environmental determinism and agrarian 'collapse' in Andean Prehistory. In: Antiquity 73, 1999: 634-42.

Goldstein, P. S. Tiwanaku Temples and State Expansion: A Tiwanaku Sunken Court Temple in Moquegua, Peru. In: Latin American Antiquity 4(3), 1993: 22-47.

Goldstein, P. S. Tiwanaku Settlement Patterns of the Azapa Valley, Chile - New Data, and the Legacy of Percy Dauelsberg. In: Dialogo Andino 14/15, (Special issue Prehistoria del Norte de Chile y del Desierto de Atacama. Simposio Homenaje a Percy Dauelsberg Hahmann), 1996: 57-73.

Janusek, J. W. Craft and Local Power: Embedded Specialization in Tiwanaku Cities. In: Latin American Antiquity 10(2), 1999: 107-131.

Kolata, A. L. The Agricultural Foundations of the Tiwanaku State: A View from the Heartland. In: American Antiquity 51, 1986: 748-762.

Kolata, A. L. Economy, Ideology and Imperialism in the South-Central Andes. In Ideology and Pre-Columbian Civilizations, edited by A. A. Demarest, Geoffrey W. Conrad, pp. 65-87. Santa Fe: School of American Research Press, 1992.

Kolata, A. L. The Tiwanaku: Portrait of an Andean Civilization. Cambridge, Massachusetts: Blackwell Publishers, 1993.

Kolata, A. L., editor. Tiwanaku and its Hinterland: Archaeology and Paleoecology of an Andean Civilization. Washington, D.C.: Smithsonian Institution Press, 1996.

Lechtman, H. El Bronce Arsenical y el Horizonte Medio. In Arqueologia, Antropología e História en los Andes: Homenaje a Maria Rostworowski, edited by R. Varón G., Javier Flores E., pp. 153-186. Lima: Instituto de Estudios Peruanos, Banco Central de Reserva del Perú, 1997.

Llagostera, A. San Pedro de Atacama: Nodo de complementariedad reticular. In La Integracion Surandina Cinco Siglos Despues, edited by X. Albó, M. Arratia, J. Hidalgo, L. Nuñez, A. Llagostera, M. Remy, and B. Revesz, pp. 17-42. Cusco and Antofagasta: Centro de Estudios Regionales Andinos Bartolomé de Las Casas, and Universidad Católica del Norte, 1996.

Manzanilla, L. Akapana: Una Pirámide en el Centro del Mundo. Mexico City: Universidad Nacional Autónoma de México, Instituto de Investigaciones Antropológicas, 1992.

Moseley, M. E., Robert A. Feldman, Paul S. Goldstein, and Luis Watanabe M. Colonies and Conquest: Tiahuanaco and Huari in Moquegua. In Huari Administrative Structure: Prehistoric Monumental Architecture and State Government, edited by W. G. M. Isbell, pp. 91-103. Washington, D.C.: Dumbarton Oaks, 1991.

Moseley, M. E. The Incas and their Ancestors: The Archaeology of Peru. New York: Thames and Hudson, 1992.

Mujica B., E. La Integración Surandina durante el periódo Tiwanaku. In: La Integracion Surandina Cinco Siglos Despues, edited by X. Albó, M. Arratia, J. Hidalgo, L. Nuñez, A. Llagostera, M. Remy, and B. Revesz, pp. 81-116. Cusco and Antofagasta: Centro de Estudios Regionales Andinos Bartolomé de Las Casas, and Universidad Católica del Norte, 1996.

Muñoz Ovalle, I. Integración y Complementaridad en las Sociedades Prehispanicas en el Extremo Norte de Chile: Hipótesis de Trabajo. In La Integracion Surandina cinco Siglos Despues, edited by X. Albó, M. Arratia, J. Hidalgo, L. Nuñez, A. Llagostera, M. Remy, and B. Revesz, pp. 117-134. Cusco and Antofagasta: Centro de Estudios Regionales Andinos Bartolomé de Las Casas, and Universidad Católica del Norte, 1996.

Oakland Rodman, A. Textiles and Ethnicity: Tiwanaku in San Pedro de Atacama, North Chile. In: Latin American Antiquity 3(4), 1992: 316-340.

Ponce Sangines, C. Tiwanaku: Descripcion Sumaria del Templete Semisubterraneo. In: Academia Nacional de Ciencias de Bolivia 20, 1969.

Ponce Sangines, C. Tiwanaku: Espacio Tiempo y Cultura. In: Academia Nacional de Ciencias de Bolivia, La Paz, 1972.

Ponce Sangines, C. Arqueología de Lukurmata vol. 1. La Paz: INAR, 1989.

Reinhard, J. Chavín and Tiahuanaco: A New Look at Two Andean Ceremonial Centers. In: National Geographic Research 1, 1985: 395-422.

Rivera Diaz, M. A. The Prehistory of Northern Chile: A Synthesis. In: Journal of World Prehistory 5(1), 1991: 1-48.

Rivera Sundt, O. Resultados de la excavación en el centro ceremonial de Lukurmata. In: Arqueología de Lukurmata, edited by A. L. Kolata, pp. 59-89. vol. 2. La Paz: INAR, 1989.

Stanish, C. Ancient Andean Political Economy. Austin: University of Texas Press, 1992