The Nanchoc Tradition: The Beginnings of Andean Civilization

Tom D Dillehay, Jack Rossen, Patricia J Netherly. American Scientist. Volume 85, Issue 1. Jan/Feb 1997.

One of the cherished goals-some would call it the fatal Cleopatraof anthropology is to explain the emergence of civilization in early human communities. Anthropologists interested in this question take as their laboratory five areas of the world where civilizations are thought to have developed autonomously rather than by diffusion: Mesopotamia, China, the Indus Valley, Mesoamerica (Mexico) and the central Andes. Of these “pristine” civilizations, the Andean case seems the most exceptional; the Inca, for example, never developed writing, considered a defining characteristic of Mesopotamian civilization. Yet it is the least studied and least understood.

One of the intriguing problems of Peruvian archaeology is the apparent absence of antecedents for the earliest conspicuous civilizations. Between 3000 and 2000 B.C., during what is called the Late Preceramic period, there was an explosion of pyramid building on the northern and central coasts of Peru and in the central highland basins. These salient material markers of civilization were associated with complex societies characterized by sedentism, the production of food, pastoralism and large villages. The monumental buildings appear rather abruptly on the coast, however, and earlier forms also have not been found in the highlands.

The archaeological record for the period preceding monument building, called the Middle Preceramic, is scant and fragmentary. Moreover, a bias toward the investigation of large, elaborate ceremonial centers has made it difficult to consider antecedent, small-scale public spaces or architectural forms. These are both harder to detect archaeologically and, given the tendency to assume that the accumulation of wealth and the attempt to gain social power are the bases for socioeconomic complexity, harder to accept paradigmatically.

The record is also distorted by a geographical bias. Most research on the origins of Andean civilization has been done in the major coastal valleys and the highland basins of Peru, where the remains of the later Moche, Chimu, Inca and Tiahuanaco civilizations were found. On the coast, anthropologists have studied shell middens and exploitation of the rich marine resources off the coast of Peru. In the highlands, they have excavated caves and studied the initial domestication of tuberous plants, such as potatoes and oca, and of the camelids, such as the llama and alpaca. Compared to the coast and the highlands, the Amazonian rain forest on the eastern slopes of the Andes has been neglected, as have the western mountain slopes, which were thought of as arid.

In 1978 the authors discovered a rich Preceramic complex in the Nanchoc Valley on the western slopes of the Andes in northern Peru. One location, the Cementerio de Nanchoc (named after a modern cemetery), turned out to be a specialized nonresidential site associated with the production of lime from travertine and calcite deposited in springs at the headwaters of local creeks. The processed lime may have served as a mineral supplement to the diet or, more likely, as an extractive agent taken with coca. Excavations at residential sites across the valley from the Cementerio de Nanchoc suggest that the lime was processed by a small community of specialized hunter-gatherers who were heavily dependent on plant resources and just beginning to practice horticulture.

The Nanchoc Valley excavation revitalizes the debate over the origins of Andean civilization. The earliest cultural developments in this area do not seem to fit any of the standard explanations for the appearance of social complexity There is no evidence, for example, that population pressure and growth or climatic change played a role in the forging of a communal identity. Instead the Nanchoc Valley seems to fit a different model, where public activity and the development of a ritually sanctioned extraction technology was an instrument for consolidating social and cultural identity.

Surprises in the Peruvian Record Most Andeanists agree that the first great culture in the Central Andes was Chavin, named after Chavin de Huantar, an archaeological site in the northern highlands of Peru. The Chavin civilization, which developed about 1000 B.C., was characterized by a dominant art style and architecture that influenced other, later cultures. This led Peruvian archaeologist Julio C. Tello to propose in the 1930s that Chavin is the oldest of Peru’s civilizations and the foundation on which all the later ones are built.

The chronological framework for the archaeology of Peru, which was suggested by Max Uhle of the University of California, Berkeley, and refined by John H. Rowe of Berkeley in 1962, uses the diffusion of Chavin style, called the Early Horizon, as one chronological marker. The period before the Early Horizon, called the Initial Period, begins with the introduction of pottery into Peru. (Horizons refer to times where a wave of stylistic influence swept across regional boundaries and periods refer to times characterized by regional styles.) The interval before the Initial Period, called the Preceramic, is divided into Late, Middle and Early periods. The Late Preceramic begins with the appearance of cotton in the archaeological record of the Peruvian coast at about 3000 B.C.

It is typical of the surprises the Peruvian record holds that monumental architecture predates not just the Chavin civilization but even the appearance of pottery and primary dependence on agriculture for food. In 1941 Gordon Willey and John Corbett, archaeologists who were excavating a large platform mound at the Aspero site on the coast of Peru, were disconcerted to find no ceramics. In the following decades, additional coastal sites were discovered that lacked pottery but featured major mound constructions. In 1960, a team from the University of Tokyo found large public buildings at Kotosh, a highland site, in stratigraphic layers below the remains of the earliest ceramic cultures, and other examples of highland Preceramic public architecture were soon found.

Monumental architecture in Peru, then, dates to the Late Preceramic and Early Initial periods, not to the Early Horizon as Tello had thought. But as Michael Moseley of the University of Florida and Peter Kaulicke of the Universidad Catolica de Peru have recently noted, the preconditions for monument building must be sought within the earlier Middle Preceramic period, between 6000 and 3000 B.C. It is within the hunter-gatherer societies of this period that the initial moves toward the domestication of plant and animal species, sedentism, technological innovation, occupational specialization and the construction of public architecture took place. The archaeological problem, of course, is that the first tentative steps toward a complex society left much subtler traces than the later monument-building civilizations.

At first glance Peru is an unlikely spot for a pristine civilization. It is composed of three contrasting biogeographic zones-the arid Pacific coast, the high, steep and rugged Andes Mountains and the lush Amazonian lowlands-all of which are habitable but none of which is notably hospitable. Indeed, acre for acre, Peru has one of the lowest carrying capacities found in the Western Hemisphere.

The coast of Peru is a strip of desert intersected by small rivers and intermittent streams that arise to the east in the Andes mountains. The coastal climate derives from the Humboldt Current, which makes a broad sweep from Antarctica up the coast of South America. Because the prevailing winds that blow from the southwest during the Southern Hemisphere winter are cooled by an upwelling of cold waters at the coastal edge of the current, almost no rain falls on the coastal plain. The early coastal communities were sustained by the resources of the ocean, which are acknowledged to be among the richest in the world, and, later, by irrigated farming.

Traditionally the western slopes of the Andean cordillera below an elevation of 2,400 meters above sea level have also been considered to be arid, an assertion based on observations made in the central coast. Unlike the valleys in central Peru, however, those along the north coast contain a variety of microclimates. When we first visited our study site, the Zana Valley, in 1972 and 1975 we were struck by its ecological richness.

The Zana River rises in two branches-the Rio Zana River and the Rio Nanchoc. The Upper Zana Valley (the area north of the Zana River) is covered by remnants of a tropical montane forest complete with parrots, deer, black bears and boa constrictors. The forest is sustained by montane rainfall characteristic of the highlands during the summer, but it also generates a self-perpetuating microclimate with dense fog from condensation and light rainfall at night. Downslope from the forest, which lies between the elevations of 1,300 and 2,400 meters, there is an ecotone, or area of overlapping ecological zones, in which the tropical montane forest grades into a subtropical forest, a drier thorn forest and a thorn steppe.

Beryl Simpson of the University of Texas at Austin has established that an extensive belt of montane forest covered the western slopes of the Andes during terminal Pleistocene times. The forest survives today as an archipelago in widely separated relict patches. The patch at the head of the Zana Valley is one of the largest of these.

We were first led to the Zana Valley by research we had done on late preInca and Inca societies on the western slopes of the Andes and at an ethnohistorically described Inca center in the tropical forest of the upper Zana Valley. We carried out a preliminary survey of the Upper Zana region in 1978 that located some 100 archaeological sites. In 1981 test excavations were undertaken to determine the nature and intensity of human occupation in the different ecological zones within the study area. We found significant occupation during the later Preceramic period and generally intense occupations during all the subsequent ceramic periods, ending with the Inca occupation.

Cementerio de Nanchoc

The Preceramic occupations were all located off the floors of the main valleys on alluvial fans at the mouths of quebradas (dry side canyons) or else some distance up the quebradas themselves. The most important of these sites is Cementerio de Nanchoc, which was located in 1978, and defined and excavated in 1981 and 1985. This nonresidential site lies on a small alluvial fan at the mouth of a narrow lateral quebrada on the north side of the Nanchoc Valley. Fiftyone residential sites were also discovered along streams and adjacent to springs in three quebradas that lie directly across the valley from Cementerio de Nanchoc. The Cementerio de Nanchoc site, which covers about 2 hectares (5 acres), lies on both banks of a small arroyo that cuts through the alluvial fan. The first of two areas we excavated (zone A), on the east side of the stream, contains two low, three-tiered, roughly triangular mounds made of a meter of earth placed on a slight rise in the quebrada alluvium. The second excavated area (zone B) is an open-air workshop on the west side of the stream that was in use at the same time as the mounds.

When we first mapped the site in 1978, we identified only the western mound. At the time, its outer perimeter and the tiers were defined by aligned stones. Most of the stones were unworked basalt boulders or limestone slabs, but a few had been deliberately split and thinned.

Two entrances on the south face were marked by larger boulders that formed stone steps onto the lowest tier. When we returned to the site in 1981 we realized that there were two mounds and an associated work area. The second, eastern mound falls within the cemetery for the modern village of Nanchoc and has been nearly destroyed. Between 1978 and 1981 many of the stones delineating the mounds were removed and incorporated in a wall that was built around the cemetery and mounds.

Excavation of the western mound uncovered two distinct floors, each of which was lightly strewn with flakes of local volcanic stones and a few conicalshaped chunks of worked lime, called cal. An electrical resistivity survey of the mound and test pits revealed two parallel lines of postholes and a litter of small stones and flakes at the level of the lower floor, suggesting that debris had accumulated along the side walls of an architectural structure.

A radiocarbon assay of charcoal from the lower floor yielded a date of 5770 B.C. 100 years. At this time, the site was probably a ritual area marked only by a low hummock or aligned stones. A piece of wood from the upper floor was dated to 3700 + 60 B.C. The layered habitational floors and the radiocarbon dates suggest that the mound was used intermittently and lightly but over a long period. Use of the west mound ended sometime between 4000 and 3000 B.C., when it was capped with stone slabs.

Excavation of zone B disclosed an occupational floor, a light scatter of stone artifacts and many hearths, which contained ash, unworked travertine and calcite, burned rock, heat-shattered metates, or milling stones, and manos (the handstone used as the upper millstone), and small fragments of burned and partially burned lime. No domestic refuse was found.

Radiocarbon dates for the hearths and floor range from 5300 to 4800 B.C., that is, over roughly the same period as the mounds. Identical stone artifacts from the lowest hearth and the lower level of the western mound, together with analysis of the natural and cultural stratigraphy, confirm that the work area and mounds are associated.

Given the scant domestic refuse, such as floor stains and midden debris, the narrow range and low density of the artifacts and the presence of the mounds, a reasonable supposition is that Cementerio de Nanchoc is a site where specialized nondomestic production took place, probably the manufacture of lime. Ethnographic analogy suggests that this material might have been used either as a mineral supplement with either tuber or grain meals or as an extractive agent with coca. Chewed with a pinch of lime, or cal, the leaf of the coca bush (Erythroxylum coca) releases a mild dose of cocaine alkaloid, numbing sensory nerves and dulling hunger. It has been central to religious rituals and the daily life of Peruvians for centuries.

There is little direct evidence of coca use at the Nanchoc Valley sites. Three leaves were found underneath fallen architectural stone buried in the floor of a dwelling at one of the excavated residential sites, but their radiocarbon dates are inconsistent with those of human bone remains found with them, which date around 4000-3500 B.C. (The dates given here are accelerator mass spectrometry, or AMS, radiocarbon dates, which can be thrown off if an artifact is contaminated by organic residues from the soil matrix.) No leaves were found at Cementerio de Nanchoc. But as anyone who has observed coca chewing knows, little remains, and that little might not be identifiable archaeologically. The use of coca is suggested both by the heavy investment in the production of lime and, perhaps, by the presence of exotic goods that could have been received in exchange for coca. Moreover, the montane forest of the Nanchoc Valley was probably a prime coca-growing region in the past, just as it is today.

Among the Aymara of modern Bolivia and the Quechua of modern Peru, cal is prepared by burning calciumbearing rocks and grinding the residue into powder. The powder is mixed with water and salt and condensed into small, cake-like concretions. The presence of burned, unburned, flaked and unflaked travertine and calcite, prepared lime concretions, and stone scrapers and grinding stones whose edges still bore particles of burned and unburned lime suggest that production and use of lime of Cementerio de Nanchoc might have been similar.

We interpret the mounds, then, as small-scale public structures that probably served as a focus for communal activities, including the production of lime but possibly also feasting and decision making. The technological activity of extracting cal was probably sanctioned by communal rituals. No direct evidence of ritual activity survives, but it is suggested by the tiered mound architecture, by the isolation of the mounds and work area from the domestic sites across the valley, and by the archaeology of the domestic sites.

Early Occupation: Las Pircas What people used the Cementerio de Nanchoc site, and how did they live?

In the three quedabras across the valley we found evidence of two phases of occupation, which we named the Las Pircas phase (65004500 B.C.) and the Tierra Blanca phase (4500-3000 B.C.), after the quedabras where the primary settlements were located.

Excavation of these sites suggested that household autonomy may have declined as communal activity at the lime-processing site assumed greater importance. The Las Pircas phase can be viewed as a florescence, characterized by the production of a variety of high-quality stone tools, household rituals (garden magic may have been practiced), regular access to exotic goods and elaborate treatment of the dead. In contrast, although there is evidence of greater horticultural skill (though still relying primarily on wild plants) during the Tierra Blanca phase, fewer exotic items were found, stone tools became cruder and treatment of the dead more haphazard.

The Las Pircas and Tierra Blanca occupations consist of a network of households scattered high on the alluvial fans in the quedabras. The sites are typically found on slightly elevated ground next to tracts of naturally terraced soil in the middle and lower courses of streams.

There are no signs of contemporaneous occupation on the floodplain below. This pattern is so characteristic of the Preceramic sites that it suggests the habitat of the high fans in lateral canyons must have been particularly favorable. Perhaps stream flooding, which might have been regular but not deep or prolonged, was advantageous for peoples just beginning to practice horticulture along with hunting and the gathering of wild plants.

The Las Pircas quedabra has the largest alluvial fan in the Nanchoc area. The fan is dotted with small (30 to 100 meters across) Preceramic residential and activity sites, several of which have been excavated. Radiocarbon dates range from about 6500 to 5000 B.c., with the earliest sites higher on the fan. Features at the sites appear to be interrelated and are never superimposed, suggesting a single, continuous occupation.

Although domestic refuse was found at all of the excavation sites, the foundations of houses and food refuse were found primarily at one site (CA-09-27), possible garden plots at this and one other site (CA-09-52) and human bone mainly at a third site (CA-09-28). Excavation of the first site revealed the foundations of a small (2.3 meters by 2 meters) elliptical hut. The foundations of the hut were adobe and stone and the sidewalls were made of quincha cane and daub (numerous daub fragments with cane impressions were found at Las Pircas). Small bedrock outcrops incorporated into the foundation of the house defined a northeast-facing entrance.

At the second and third sites, undulating areas without features suggest ancient garden furrows. These sites are situated on alluvial terraces in the upper quebradas near springs, where ditch-irrigated gardening would have been facilitated by the grade and shallow cut of the stream and associated rich alluvial soils. The garden plots each covered about an eighth of a hectare.

A series of secondary burials and one primary burial were excavated at the fthird site. The human remains were most often cut longbones of adult males that had been carefully placed in piles or shallow pits. Stone anvils apparently used in treating the dead lay nearby. Analysis of some of these bones by John Verano of Tulane University recently revealed evidence of possible cannibalism. The primary burial was flexed and covered with a rock pavement.

The Las Pircas artifact collections are dominated by an exclusively unifacial lithic industry characterized by formal consistency and relatively fine workmanship. We termed this industry, which can be distinguished by its workmanship from other northern Peruvian unifacial industries identified by James Richardson of the University of Pittsburgh, the Nanchoc Lithic Tradition.

All of the quedabras have occasional deposits of easily worked volcanic stones, principally basalts and andesites. The tools were generally made by striking a flake from a blocky stone core and then shaping the tool by chipping the flake with a harder stone. The edge might then be sharpened by retouching, much as one sharpens a knife. The stone tools made in this way include flake and slab choppers, crude blades, side-scrapers, gravers and a few endscrapers.

The Nanchoc Lithic Tradition appears to have been heavily oriented toward plant processing and woodworking. Of the approximately 50,000 lithics collected from all cultural periods in the upper Zana Valley, only two are bifacial and they are obviously imports. The absence of locally produced bifadally flaked stone tools, such as spear points, suggests that hunting may not have been of prime importance, although a small faunal assemblage indicative of a tropical forest environment was found. (In addition to abundant land snails, it included the remains of jaguarundi, a boa-like snake, and tinamou, a bird.) Bright plant polish and fibers found on the edges of some tools also suggest reliance on a vegetarian diet, as does heavy wear on human teeth and the presence of hundreds of metates and manos.

A small collection of plant remains was recovered through water flotation from the floors inside the structures. The collection of wild and semi-cultivated plants included squash, peanuts, Chenopodium (the plant that produces quinoa, a grain still eaten in Peru), fleshy solanaceous species with fruits resembling tomatillos, and cactus fruits. The collection is problematic because radiocarbon analysis has yielded inconsistent dates. Yet the plants have ancient forms, the stratigraphic context in which they were found is impeccable, and their presence is consistent with other evidence, such as the types of tools that were found and the wear patterns on the tools.

Exotic items recovered in the excavation of the Las Pircas alluvial fan included modest amounts of burned and broken shells, which must have come from the coast 80 kilometers to the west. The shellfish were probably a food item, but other, rarer exotic items may have had a less mundane purpose. These items, which include quartz crystals, stingray spines, colorful marine shells, fossils, beads and amulets made of green malachite, and a broken projectile point made of bright-red jaspar in the Paijan style, come from widely scattered sites in the modern Cajamarca department immediately to the east of the Zana Valley and the Lambayeque department immediately to the north.

Novel objects such as these may have been the earliest nonfood resources to be procured and “managed” by the people in this area. Michael Harner of Columbia University has shown that the modern Jivaro in eastern Peru plant exotic stones and other items in their gardens to protect their crops from evil forces. Ritualization, perhaps connected to exotic items, would be expected of a group that was experimenting with horticulture and perhaps also with related changes in settlement patterns and social structure. Some of these items, especially the quartz crystals, may have been used in garden magic, or at least in a general intensification of household ritual associated with exploratory manipulation of plants.

The Las Pircas phase, in summary, appears to have been characterized by a permanent settlement of the sort that Bennett Bronson of the Field Museum in Chicago has termed “pseudo-dense.” The clustering of homes, in other words, appears to reflect the presence of key natural resources such as the springs, alluvial soils and deposits of easily worked stone, rather than population density or pressure.

Together the archaeological evidence suggests that major economic and social changes were under way in the orbit of the Nanchoc area and that the local people were in the early stages of forging a corporate identity. No coca leaves were found at Las Pircas, but a few fragments of burned lime were found at the garden sites. The fragments were small and amorphous, suggesting that the extraction and processing of lime-bearing rocks during this phase was in its initial or early stages.

The Tierra Blanca Occupants

Along a stretch of the Tierra Blanca alluvial fan, we located a number of small domestic sites, five of which have been partially excavated. The dates obtained by radiocarbon analysis of human bone and charcoal recovered from hearths inside buildings range from 4000 to 3000 B.C.; this was a later occupation than the one found at Las Pircas. The cultural levels within the stratigraphy of these sites are not thick, and there are clear lenses of habitational debris, suggesting that, unlike Las Pircas, this site was periodically abandoned.

The Tierra Blanca occupation supertidaily resembled the Las Pircas one, but it turned out to be different in many subtle ways. At one site (CA-09-77) we found a rectangular house with rounded corners, rock dividing walls and two, small basinshaped hearths. Within the house were a variety of stone tools, shattered and sometimes burned animal and human bone, plant remains, burned and unburned malachite and conical pieces of lime. The tools found at this and at other Tierra Blanca sites included end-scrapers, polishing stones, and grinding stones and slabs. The plant remains included fragments of cotton, squash, Chenopodium and three leaves of wild coca. Donald Ugent of Southern Illinois University identified the species of the coca leaves.

The range of stone tools found at Tierra Blanca was smaller than that at Las Pircas, and the tools were comparatively crude and poorly worked. We associate this decline in the lithic industry with the scarcity of novel artifacts and the less systematic treatment of the dead. A few exotic lithics and marine shells were found, but these were generally of poorer quality that those recovered from the Las Pircas sites. (A greater variety and quantity of marine shells were found, however, suggesting that imported shellfish had become a more important component of the diet.) Instead of being carefully cut and placed, human bone was roughly broken, shattered and trampled on the floors of some sites.

Excavation of another site (CA-09-50), a low residential mound located not on an alluvial fan but lower down, at the edge of the Nanchoc River’s floodplain suggests, however, that the Tierra Blanca population may have been more accomplished horticulturists than their predecessors. The mound was in use from the Middle to the Late Preceramic period, and the more recent layers were associated with fragments of peanuts and Chenopodium. Garden furrows and small irrigation canals that appear to be Preceramic were found near the site and at the conjunction of the Terra Blanca fan and the main valley floor. A reasonable supposition is that this site reflects increased exploitation of forest-fringe and floodplain resources.

How can these differences between the two Nanchoc Valley occupations be understood? The later Tierra Blanca site suggests that a degradation, or devolution, of technologies and practices, including household ritual practices, accompanied the full-scale development of a lime- and food-producing economy. Whereas the Las Pircas sites contained only small, irregularly shaped chunks of lime, the Tierra Blanca sites yielded regular cones, suggesting a more standardized and formal system of lime production. The differences between the sites suggest that household autonomy decreased as ritually sponsored communal activity was established.

Ritual, Technology and Civilization To grasp the significance of the Cementerio de Nanchoc site, one must distinguish the technology of lime extraction from the complex communal support for this technology. The technology involved procurement of the local raw materials and knowledge of the process by which lime could be extracted from them. The multiple hearths and processing areas in the open-air workshop at Cementerio de Nanchoc indicate that lime processing was done communally. Communal support for lime processing must have entailed some kind of economic arrangement, but the evidence also suggests it involved social regulation by means of public ceremony.

Use of the Cementerio de Nanchoc mounds may actually have preceded the establishment of permanent domestic settlements. The mound deposits are thicker and include distinct floors, only a few unifacial lithic types are present, and debitage, the refuse of stone flaking, is absent. The deposits at the residential sites, on the other hand, are usually thin and contain many more artifacts. This difference suggests that the mounds saw long-term, but comparatively light, specialized use.

The Las Pircas people were just beginning to develop the skills needed to organize persistent production of lime for local consumption or exchange and to meet the social obligations of communal ritual. They were territorial-valued resources were not freely available to everyone-and they had developed rituals involving death and fertility, perhaps in response to the influx of new resources, such as coca, lime, possibly other plants, and exotic items.

We would argue that communal ritual developed together with a specialized lime-extractive technology during the later, Tierra Blanca phase. The refashioning of the mounds during the Tierra Blanca phase may have been directed by a specialized corporate body that was responsible both for conducting local affairs and overseeing exchanges with distant groups. The pooling of resources and the organization of labor were probably regulated by local ceremonies that may have attracted people from distant settlements.

The Preceramic population in the upper Zana Valley experienced some of the most fundamental transitions in human history. These include the transition from a nomadic to a sedentary life-style, identification with a group larger than the family, the capacity to develop rather than simply claim resources, and the establishment of public ceremonial places. Although these transitions may seem primitive, they are the basis of urban society. In a sense, the developments that followed were merely a scaling up of what had already occurred.

Earlier this century the archaeologist V Gordon Childe argued that a great urban revolution had taken place in the Near East about 3000 B.c. Childe defined the Urban Revolution as the development of densely populated settlements whose farmers supported a small army of craftspeople, priests and traders by producing massive food supplies. This model was thought to be general and was widely quoted in world histories and other popular works about the past, and although it has been discarded, its ghost still lingers in the popular imagination.

The lesson of Peruvian archaeology in general and of the Nanchoc Valley site in particular is that it may not be possible to define a universal suite of cultural mechanisms and processes that set the stage for civilization. Any hypothesis with pretensions to universality must be examined in local contexts and must be flexible enough to take into account local ecology, political economy, religious beliefs and historical conditions.

But the Nanchoc Valley record supports one general notion: the importance of communal social relations and ritually sponsored activities in the rise of civilization. At Nanchoc a specialized, ritually sanctioned extractive technology served as a means of promoting group identity and cohesiveness during the early phases of civilization. In Real Alto, an early Ceramic site in southern Ecuador, similar functions seem to have been served by a fiesta house and charnal house. Neither finding is a surprise to anthropologists. Cultural activity and a sense of shared purpose have always bound people together and given them power as a society; in each part of the world where preindustrial civilizations appeared, ceremonial places were the first seats of power exchange, authority and identity. Participation in rituals enhanced the sense of group identity, coordinated the actions of individual members of the group and prepared the group for cooperative action.

Beyond these observations, the excavated record is still too incomplete to answer other questions these sites raise, such as whether the later public architecture was a direct outgrowth of the earlier appropriations of public space. The final message of the Nanchoc Valley is a simple one: The foundations of Andean civilization are both older, more widespread and more complex than had been suspected.