Extracts from a 1989 NSF grant proposal that was funded





Tiwanaku Colonies of the Lower Osmore Valley:

The Expansion of an Early State

Bruce Owen


Project Summary

Tiwanaku was one of the early expansive states that should be considered in any comparative study of the origins of complex society. Current research in the Tiwanaku altiplano and the middle Osmore valley on its western edge is shedding light on the center of the Tiwanaku phenomenon. The proposed research, part of the Programa Contisuyu, seeks to understand the frontiers of Tiwanaku expansion, thought to comprise colonies producing low altitude crops for the altiplano core. The project will map, surface collect, clear, and excavate in large areal exposures the residential areas of two such colonies in the lower Osmore valley: El Algodonal and Loreto Viejo. The goals are to establish the specific resources that motivated Tiwanaku expansion into the region, the internal administrative, functional, and class structure of the colonies, the relationship between the colonies, and the relations of the colonists to any local population. Studies of mortuary remains will test the hypothesis that the colonists were literally immigrants from the altiplano. The data will be interpreted primarily through comparisons with Programa Contisuyu findings from local Chiribaya and middle valley Tiwanaku sites, and the Inka empire as documented in the literature and in recent research.

Tiwanaku Colonies of the Lower Osmore Valley:

The Expansion of an Early State


Social scientists have a fundamental curiosity about the origins and nature of state-level society, that scale of social organization which, only some 5000 years after its first appearance, affects the lives of virtually everyone on earth. To understand this theme of increasing social, political, and economic extension and complexity, we need to know what early states were like. By comparing them, we discover the possible forms of organization, how they were the same and different, and which features were essential and which free to vary.

Of the several regions of the world that independently developed state-level societies, the South American Andes cradled a number of true expansive states or empires. From perhaps 300 to 1200 AD, long before the better-known Inka, a state centered on the city of Tiwanaku dominated much of the southern Andes, with fingers of settlement ultimately reaching out from its high, cold altiplano core to the warm jungle lowlands on the east and the Pacific coast on the west, and with indirect trade contacts, at least, stretching as far as 700 kilometers south (fig. 1) (Mujica 1985, Kolata 1982, Ponce 1972). Tiwanaku was probably the largest Andean polity before the Inka, and was arguably "pristine", without precedent at least in the southern Andes. If success can be judged by stability and duration, then Tiwanaku was the most successful state ever to arise in the Andes.

Yet, with no written sources and with comparatively little archaeological data, our understanding of Tiwanaku expansion is still somewhat speculative. Tiwanaku has been considered a market trade center with a religious or ideological system disseminated by proselytizing merchants (Browman 1978, 1980), a religious and military movement (Serracino 1980), an "apolitical, ideo-economic network" (Wallace 1980), and, most commonly, an example of "Andean verticality", a form of socioeconomic organization purportedly unique to the Andes (Lumbreras 1973, Nuñez 1973, Mujica 1985, Murra 1975, etc.). Berenguer et.al. (1980) suggest convincingly that Tiwanaku expansion took different forms in different places. Ponce (1972) imagines Tiwanaku growing by establishing colonies that later served as bases for military conquest. Others imply that Tiwanaku settlements spread unhindered by indigenous opposition, coexisting more than interacting with smaller, less highly organized groups (Lumbreras 1973, Rivera 1980, Muñoz 1981, Kolata 1988). Most assert that peripheral Tiwanaku sites on the western and eastern slopes of the Andes were colonies established specifically to export low-altitude crops such as maize, cotton, ají peppers, coca, and fruits to the altiplano; the coastal sites would also have exported marine resources such as dried fish, shellfish, seaweed, and guano (Kolata 1988, Mujica 1985, Flores 1972, Rivera 1980).

These sketch models are based largely on our incomplete knowledge of Tiwanaku settlement patterns, the geographic distribution of decorated status or ritual artifacts, and grave goods from Tiwanaku cemeteries. While Ponce and others (Bennett 1937, Ponce 1972, Dauelsberg 1960) have laid the crucial descriptive and chronological groundwork, we are just now seeing the important fruits of Kolata's (1982, 1988) more explicitly problem-oriented, ongoing excavations in the Tiwanaku heartland, and of Goldstein's (1985, in press) illuminating work on Tiwanaku in the middle Osmore valley. Both reveal large populations organized to extract agricultural produce on a grand scale. The research proposed here takes the next step away from the altiplano heartland to the Tiwanaku colonies of the lower Osmore near the Pacific coast, with the intention of assessing the nature of Tiwanaku expansion towards the coast, what motivated it, how it functioned, and how it was similar to, and distinct from, other early state-level expansive developments in the Andes and elsewhere.


This project will focus on the two known Tiwanaku colonies in the lower Osmore drainage, El Algodonal and Loreto Viejo (fig. 2). It will be one of several relatively independent research efforts operating under the aegis of the Programa Contisuyu, a loose group of cooperating archaeologists working in the Osmore drainage.

The Osmore is a small drainage on the western slope of the Andes (fig. 2), comprising three branches that join near the modern town of Moquegua and flow towards the Pacific through a fertile but narrow valley cutting across the arid desert of Peru's far south coast. About 47 kilometers from the coast, the river disappears from the surface and flows underground beneath a barren, dry quebrada. Some 17 kilometers from the Pacific, the river emerges again for several weeks a year between July and October, watering crops in the constricted valley bottom and emptying into the sea near the port city of Ilo. We refer to the branches of the Osmore above Moquegua as the "upper valley", the stretch between Moquegua and the dry gorge as the "middle valley", and the seasonally-watered portion on the coast as the "lower valley".

The Osmore Tiwanaku sequence was reconstructed by Paul Goldstein (1985, in press). The first phase, Omo, features a hierarchy of indefensible Tiwanaku IV ("classic") agricultural settlements established in the upper and middle valley between perhaps 780 and 980 AD (see figs. 3 and 4 for dates). The following Chen Chen phase, from perhaps 960 to 1170 AD, corresponds roughly to the Tiwanaku V ("expansive") phase of the altiplano. Radiocarbon dates suggest that intrusive colonies with Tiwanaku ceramics and burials were founded in the lower reaches of several coastal valleys, including the Osmore, during this phase. These colonies apparently persisted to perhaps 1280 AD.

In the middle valley, the Tumilaca phase marks the declining influence of Tiwanaku. Finally, archaeologists have generally considered the Chiribaya and Maitas-Chiribaya ceramic styles to be local developments on Tiwanaku stylistic models that arose after Tiwanaku influence waned (Dauelsberg 1973, Goldstein 1985, but see Lumbreras 1973). However, radiocarbon dates (figs. 3 and 4) suggest that much of this local development occurred contemporaneously with the later Tiwanaku-related phases, from 1000 AD at the latest to perhaps 1400 AD. If this overlap is real, then understanding the relationship between the state "colonists" and the local population could be crucial for reconstructing the economic, political, and cultural nature of Tiwanaku expansion.

The proposed research will focus on the two known Tiwanaku sites in the lower Osmore valley, El Algodonal and Loreto Viejo. El Algodonal (fig. 5) is a well-preserved, 1.8 hectare terraced habitation site with a .3 hectare cemetery. Using estimated population densities of 100 to 600 individuals per hectare (Earle et.al. 1988:11), El Algodonal probably housed between 180 and 1080 people. Although Tiwanaku-style round cist burials predominate, there are also rectangular Chiribaya-style burials, and both Tiwanaku and Chiribaya ceramics are found on the surface. Organic preservation is excellent, with plentiful wood, textile, basketry, animal, and plant remains. Profiles of looter's pits and eroding terraces suggest that several centimeters of midden can be expected in residential contexts, allowing large areas to be opened in a reasonable time. Programa Contisuyu experience suggests that brushing the surface should reveal the stubs of cane-walled structures (Goldstein 1985, Don Rice, pers. comm.). Fieldwork will concentrate on El Algodonal as the best-preserved Tiwanaku colony in the lower valley.

Loreto Viejo (fig. 6), located on the next alluvial fan upstream from El Algodonal, is the type site for the loosely-defined coastal Tiwanaku ceramic style of the same name. No description of this badly looted and alluviated site has been published. With 5.6 hectares of habitation area and a 1.3 hectare cemetery, it is substantially larger than El Algodonal, housing between 560 and 3360 individuals. Smaller-scale fieldwork at Loreto Viejo will look for differences and similarities in organization, activity, and occupants between the sites.

Recent research in this region has been dominated by Murra's "vertical archipelago" model of socio-economic organization (Murra 1975, 1985). Murra summarizes his model with five points: 1) ethnic groups each control a variety of econiches at differing elevations; 2) most people live in a core area, with the rest in "colonies" in continuous social and trade contact with the core; 3) families in the colonies retain rights to land and products of the core; 4) diverse groups may place colonies in the same vicinity; and 5) as the group grows large, colonies are founded at greater distance, colonists lose their rights to the core, and colonies are founded for reasons that are not determined by the econiche, such as military control or craft centralization.

The proposed research is not primarily intended to test the applicability of Murra's model to this case. Nevertheless, Murra's points 1, 2, 4, and aspects of 5 should all be confirmed or rejected easily enough for the lower Osmore case in the course of treating other questions. Point 3 would be difficult to address archaeologically, but in point 5 Murra suggests properly that it would not apply to a large state such as Tiwanaku. The results of this research should go beyond these general aspects of Murra's model to address in cross-culturally applicable terms the specifics of Tiwanaku expansion towards the Pacific coast.

Specific research questions

Each of the following questions will be answered for the Tiwanaku colonies of the lower Osmore valley through archaeological survey and excavation. The results will be interpreted by making comparisons with the Inka, Wari (where possible), and other, non-Andean cases in the literature to determine what about Tiwanaku colonialism was similar to other cases, and what was distinct and requires special explanations. Approaches to answering these questions are outlined under "analytical methods".

A. Motivation: What specifically motivated Tiwanaku expansion, at least towards the Pacific coast? Was it a need for specific crops, marine resources, guano, minerals, specialized craft production centers, or military footholds? Was the motivation more for products in the subsistence sphere or in the wealth/status/political sphere?

B. Constitution: Were the colonies occupied by altiplano people, local people, or both, and was occupation permanent, occasional, or seasonal?

C. Social and political structure: What sort of class/status/control structure was there within the colonies, and how did it relate to local versus altiplano ethnicity? Were colonists primarily producers with little stratification, a mix of producers and administrators, or administrators controlling production by, or products of, non-colonists?

D. Relations with Tiwanaku: Did the colonies receive altiplano subsistence goods, status/wealth goods, or logistical support such as camelids? Did they depend on these altiplano goods? Were the various colonies organized as a single hierarchical or parallel system, or were they independent, perhaps outposts of different altiplano groups in the sense of Murra's "multi-ethnic" colonization?

Field methods

The field crew will consist of Bruce Owen, one or two other North American graduate students yet to be determined, including one specializing in human osteological remains, one or two Peruvian archaeology students also not yet determined, and several local laborers hired as needed. Because the season is long, the membership of the crew may vary. Bruce Owen will supervise work for the entire season. The crew will be divided into three teams that will survey and excavate separate areas concurrently. An application for a permit to carry out this research has been approved by the Peruvian Instituto Nacional de Cultura and lacks only the presidential signature, so the permit should be issued well before fieldwork is scheduled to begin.

A brief surface survey of the lower valley from the coast to 17 kilometers inland, where the dry section of the valley begins, will help to sketch in the settlement pattern, population size, and scale of organization prior to and during the period of Tiwanaku intrusion. The major sites are known, but a systematic survey is needed to check for ephemeral pre-Tiwanaku sites and Tiwanaku-contemporary local settlement. Sites will be dated by surface ceramics, and by radiocarbon dates from eroding occupation deposits where necessary. Prehistoric canals and field systems will be recorded in order to estimate the potential production capacity of the lower valley. This pedestrian survey, recorded on air photographs in the field, will cover both margins of the narrow valley; the valley bottom has been completely disturbed by farming. Spot checks of higher slopes and ridgetops will indicate whether the survey should be extended to these areas. Sites discovered during the survey will be mapped sufficiently to estimate their area, population, and major architectural features.

Attention will then shift to El Algodonal. The site will be mapped in order to distinguish residential, public, and other areas, and to define natural collection and excavation units. An electronic distance meter (EDM) may be available through the Programa Contisuyu, which would allow rapid and accurate mapping. Otherwise, a plane table and alidade will be used.

Next, the site will be surface-collected in natural and arbitrary units, depending on the superficial definition of terraces. The collections will be analyzed immediately in order to identify variability that could indicate ethnically distinct sectors or areas of differing status or function. These data will be useful in themselves for understanding the sites, and will guide subsequent surface clearing. Areas where Tiwanaku and local ceramics are mixed could indicate locations of ethnic interaction or places where sequential occupations might be stratigraphically distinguishable, especially on the uphill margins of the sites where terraces are most likely to have been covered by slumping and reoccupied. Since the contemporaneity of Tiwanaku and Chiribaya styles is not generally accepted, stratigraphic superposition of Tiwanaku over Chiribaya would be important evidence of contemporaneity.

Natural units of up to 10 X 10 meters will be selected from sampling strata defined by ethnic, functional, or status differences indicated by the map and surface collections. These units will be brushed to expose the expected cane wall stubs and stonework needed for a more detailed analysis of residential patterns, status differences, and structures pertaining to administration, storage, or control of access. Again, the data will be useful both directly in analyzing the site and as a guide to locating excavation units.

Guided by the surface collections and brushed architecture, purposefully selected samples of the brushed surfaces will be opened as shallow excavation units of up to 5 X 10 meters, to recover refuse suggesting variable activities, ethnicity, status, and function in architectural and feature context, and to seek chronologically significant stratigraphy. Time is allocated for six large units and two smaller ones. All soil will be screened. Soil samples will be taken to the field lab in order to recover small botanical and faunal remains by fine screening or seed-blowing, which are preferable to floatation in these dry, loose, and salty deposits. Proveniencing, labeling, and excavation recording formats will be consistent with those of other Programa Contisuyu projects in order to ensure comparability of data.

In the lab, ceramics will be dry-brushed clean for identification as much as possible, since the water supply will be limited. Because the Osmore soils are salty, ceramics that must be washed will be soaked for a week in a series of water baths to extract salt and to prevent crystal growth and exfoliation. Organic materials will only be mechanically cleaned.

Finally, at least 25 burials will be excavated, to recover both associated artifacts and human remains. Burial cists will be located by clearing several randomly selected broad areas in different cemetery sectors, to minimize spatial bias. If a good sample of remains from Tiwanaku and, ideally, non-Tiwanaku graves can be assembled, numerous important questions concerning the origins of the colonists and their relationship with the local population can be addressed as described under "analytical methods".

When work at El Algodonal is completed, the same sequence of steps will be applied to Loreto Viejo, but on a shorter schedule and on a smaller scale. Although the site is larger than El Algodonal, mapping, surface collection, and analysis of surface collections are not expected to take any longer because alluviation has covered many surface details and the sherd density is low.

Analytical methods

Count, weight, type, and provenience of all artifacts will be entered into computer files in the field lab by a Peruvian typist to reduce the cost and delay of data entry in the United States and to allow basic in-field analysis. Area, volume, and interpreted context (occupation deposit, midden, fill, mixed) will be entered for each provenience for standardization of artifact densities.

Each of the major research questions will be addressed as follows:

A. Motivation: Evidence of production, storage, and control of stored products will suggest the function of the colonies and the motivation for establishing them. This evidence will be compared to a baseline of local production, storage, and control patterns at nearby Chiribaya sites currently being excavated by David Jessup of the Programa Contisuyu, to detailed production and consumption data gathered by the Upper Mantaro Archaeological Research Project, and to other sites in the literature. These comparisons will suggest, for example, whether the total volume of storage features at the colonies is unusually high, whether production and storage activities are abnormally concentrated in areas subject to administrative control, and whether there is an unusual emphasis on certain production activities or crops. Evidence considered will include the quantity, nature, contents, and architectural context of storage features, and the variable densities of tools associated with production tasks, such as hoes, stone drills, and grinding equipment.

To test whether the colonies produced surplus low-altitude crops for export,

food evidence from domestic consumption contexts such as in-house storage, middens, and crusts inside cooking vessels will be compared to evidence of food in public or controlled-access storage contexts and evidence of food production such as agricultural hoes, stone blades with sickle gloss, and grinding equipment. A marked difference in the crop inventories in consumption and production/storage contexts would suggest production for export.

The total potential production of various crops will be estimated by using catchment analysis combined with the types and densities of agricultural tools found. An analysis of population size and consumption patterns will suggest total potential consumption of various crops. Comparing the two analyses will suggest which crops could have been produced in significant surplus, and which might have offered the best returns.

Once the crops, marine resources, minerals, and/or craft goods that were exported are identified, they will be assessed on a scale of function ranging from subsistence goods such as guano to status/prestige goods such as copper or fine cotton textiles. Placing specific crops on this scale may be difficult, but will be aided by comparisons of the relative statuses of their contexts in coastal Tiwanaku and Chiribaya sites and by ethnographic analogy. The goal is to assess to what extent the colonies served the political as opposed to the domestic economy of the altiplano. This interpretation will be aided by comparing the distances and quantities of similar goods moved by the Inka.

B. Constitution: The hypothesis to be tested is that some or all of the occupants of the colonies were literally immigrants from the altiplano or their descendants. Recognizing altiplano, lower valley, or both types of artifacts in residential and mortuary contexts will suggest the ethnicity of the colonists. However, a more conclusive approach will be a comparison of the human remains from Tiwanaku-style round cist tombs and those from Chiribaya-style rectangular tombs. Grave goods will support this initial grouping. The two groups will be compared on a variety of cranial traits thought to be under genetic control to establish whether they represent one or two distinct breeding populations. Additional comparisons using the same traits will be made to over forty Chiribaya burials recently excavated near Ilo, and to mid-valley local and Tiwanaku burials stored by the Programa Contisuyu in Moquegua. This sub-project will be undertaken by a human osteology specialist.

Skin, muscle, and hair samples will be taken from the burials for a study based on DNA amplification, restriction fragment length polymorphisms, and DNA hybridization. This sub-project is still in negotiation, but if successful, it would offer additional tests of relatedness that should confirm or reject the hypothesis of an intrusive population. The DNA work will be done by a molecular biologist in collaboration with the principal investigators.

The possibility of seasonal occupation will be addressed by tabulating the evidence of consumption of seasonally-available resources. For example, sea urchins indicate gathering from April through June and September through November, sea lions suggest hunting in December and January, lomas plants signify collecting between June and September, and advanced fetal or neonate llamas in non-ritual contexts imply occupation between November and April (Parsons 1970, Flores 1979). The age and sex distributions of the burial populations will be examined to see if they are consistent with a self-maintaining population, or if they represent, for example, a visiting population of working-age males. To control for the possibility that the burial population is a biased sample of the occupants, comparisons will be made to the age and sex distributions of the burial groups referred to above, which should represent permanent occupations with similar biases in burial practices.

C. Social and political structure: Variability in the size, quality, and layout of architecture and the features and artifact assemblages contained in it will suggest differences in social status and function in different sectors of the site. For example, domestic, production, and administration areas should be distinguishable, as should higher and lower status residences. The arrangements of these structures should aid in reconstructing the sociopolitical organization of the colonies. The production and consumption activities suggested by their artifact assemblages should suggest the extent to which the colonists were producers, administrators, or both.

The hypothesis that the Tiwanaku colonies were contemporary with a local Chiribaya occupation will be tested stratigraphically as described earlier, and through radiocarbon dating. Fifteen samples from excavated and surveyed sites will be dated and compared with dates from current Programa Contisuyu excavations at the nearby Chiribaya sites of El Algorrobal and Chiribaya Alta.

To the extent possible, status and functional differences will be correlated to the presence of artifacts related to altiplano or lower valley ethnicity, primarily Tiwanaku and local ceramics, in order to determine what differing social, economic, and political roles may have been played by altiplano-related and local people. Care will be taken not to use the same markers for ethnicity and status, to avoid circular arguments. Additional tests of ethnic role differences will focus on the burial remains, seeking similarities or differences in general health, stature, mortality, and stress-related pathologies between altiplano and local people.

D. Relations with Tiwanaku: Botanical, bone, lithic, and ceramic artifacts will be sourced to determine what the colonies received from the altiplano. Identifiable imports include high-altitude crops such as potatoes or quinoa, chuño (freeze-dried potatoes), and vicuña, alpaca, or guanaco, which will be identified by analysis of the lower incisors. Llamas cannot be considered necessarily altiplano products (Shimada and Shimada 1985). Lithic materials such as obsidian may have been imported. A ceramic thin-section study will determine whether the Tiwanaku style ceramics in the colonies were made from the same paste as local non-Tiwanaku ceramics, or if they were imported from the altiplano, as Goldstein (1985) suggests. Inclusions will be identified and compared to the local geology to suggest or rule out local manufacture.

If the colonies were in extensive trade relations with the altiplano, there may be evidence of llama keeping, stockpiling of goods, harness gear, and pack bags. The density of this evidence will be compared to Chiribaya sites to assess whether it is unusually high or within the normal range for local economic units.

The populations and architectural features of all Tiwanaku sites in the lower valley will be compared, based on the survey, detailed maps, and better density estimates from surface clearing and excavation, in order to assess the likelihood of a hierarchical versus a parallel organization. Artifacts and botanical assemblages from each site will be compared to detect differences that might indicate different ethnic or ecological microniche origins of the occupants. For example, site-level differences in ceramic style, technology, or paste between contemporary sites could suggest independent colonization by different altiplano groups in accordance with Murra's "multi-ethnic" model; the same conclusion could be drawn from site-level differences in assemblages of plants or other imports from the altiplano.

In conjunction with the production and consumption estimates discussed earlier, the total surplus production of the colonies will be estimated for various goods to assess whether the colonies could have produced a surplus large enough to merit state investment. If not, it might suggest that they were outposts of communities rather than of the state itself. The density of altiplano goods in the colonies will be weighed against estimates of colonial production and consumption to assess whether the colonies were independently viable, or dependent upon the altiplano for a significant fraction of their diet or artifact assemblage.

Finally, once any imported goods have been identified, they will be placed on a scale of subsistence to status/prestige goods, just as the possible exports were, in order to assess to what extent colonial connections with the altiplano related to the economic and subsistence sphere as opposed to the political and status sphere. Combined with information concerning the exports, internal structure, and intersite organization of the colonies, this data should give us a good idea of specifically why and how the Tiwanaku state expanded beyond its altiplano heartland.






May 1, 1989

2 weeks

Survey valley margins; check slopes and ridgetops

May 15

1 week

Map El Algodonal

May 22

1 week

Surface-collect El Algodonal

May 29

1 week

Analyze surface collections

June 5

2 weeks

Brush terrace surfaces to expose walls

June 19

12 weeks

Excavate six large areas, two weeks each

Sept 11

2 weeks

Excavate two smaller areas, one week each

Sept 25

3 weeks

Excavate tombs in cemetery at El Algodonal

Oct 16

1 week

Map Loreto Viejo

Oct 23

1 week

Surface-collect Loreto Viejo

Oct 30

1 week

Analyze surface collections

Nov 6

1 week

Brush surfaces to expose walls where feasible

Nov 13

4 weeks

Excavate two large areas, two weeks each

Dec 11

3 weeks

Excavate three smaller areas, one week each


1 week

Christmas and New Year's holidays interspersed above

Jan 8, 1990

2 weeks

Excavate tombs in cemetery at Loreto Viejo

Jan 22

16 weeks

Analyze artifacts in field lab

May 14

2 weeks

Prepare initial field report

Total excavations at El Algodonal: 17 weeks

Total excavations at Loreto Viejo: 9 weeks

Total field season: 56 weeks (about 13 months)


Bennett, Wendell C.

1934 Excavations at Tiahuanaco. American Museum of Natural History Anthropological Papers 34(3).

Berenguer, Jose, Victoria Castro, and Osvaldo Silva

1980 Reflexiones acerca de la presencia de Tiwanaku en el norte de Chile. Estudios Arqueológicos 5:81-93. Antofagasta.

Browman, David L.

1978 Toward the Development of the Tiwanaku State. In Advances in Andean Archaeology. Pp. 327-349.

1980 Tiwanaku Expansion and Altiplano Economic Patterns. Estudios Arqueológicos 5:107-120.

Dauelsberg, Percy

1960 Algunos problemas sobre la cerámica de Arica. Boletín del Museo Regional de Arica 5:7-17.

1973 La Cerámica de Arica y su Situación Cronológica. Chungara 1:15-26.

Earle, Timothy, Terence D'Altroy, Christine Hastorf, Catherine Scott, Cathy Costin, Glenn Russell, and Elsie Sandufer

1988 Archaeological Field Research in the Upper Mantaro, Peru, 1982-1983: Investigations of Inka Expansion and Exchange. Monograph XXVIII: Institute of Archaeology, University of California, Los Angeles.

Flores Ochoa, Jorge

1972 El reino Lupaca y el actual control vertical de la ecología. Historia y Cultura 6:195-201. Lima.

1979 Pastoralists of the Andes: The Alpaca herders of Paratía. Ralph Bolton, trans. Philadelphia: Institute for the Study of Human Issues.

Goldstein, Paul Samuel

1985 Tiwanaku Ceramics of the Moquegua Valley, Peru. Unpublished masters thesis, Department of Anthropology, University of Chicago.

i.p. The Tiwanaku Occupation of Moquegua. In Ecology, Settlement, and History in the Osmore Drainage. Donald Rice and Charles Stanish, eds. British Archaeological Reports, International Series.

Kolata, Alan

1982 Tiwanaku: Portrait of an Andean Civilization. Field Museum of Natural History Bulletin 53(8):13-28.

1988 The urban-rural axis and the organization of the Tiwanaku state.

Lumbreras, Luis G.

1973 Sobre la problemática arqueológica de Arica. Chungara 1/2:25-27.

Mujica, Elias

1985 Altiplano-Coast Relationships in the South-Central Andes: From Direct to Indirect Complementarity. In Andean Ecology and Civilization. Pp. 103-140. Tokyo: University of Tokyo Press. Muñoz Ovalle, Ivan

1981 Dinámica de las estructuras habitacionales del extremo norte de Chile (valle-costa). Chungara 8:3-32.

Murra, John V.

1975 El control vertical de un maximo de pisos ecologicos en la economia de las sociedades andinas. Formaciones Economicas y Politicas del Mundo Andino. Pp. 59-115. Lima: Instituto de Estudios Peruanos.

1985 The limits and limitations of the "Vertical Archipelago" in the Andes. In Andean Ecology and Civilization, An Interdisciplinary Perspective on Andean Ecological Complementarity. Pp. 15-20. Tokyo: University of Tokyo Press.

Nuñez Atencio, Lautaro

1973 Carta respuesta a Luis G. Lumbreras sobre la problemática arqueológica de Arica. Chungara 1:30-37.

Parsons, Mary Hrones

1970 Preceramic subsistence on the Peruvian coast. American Antiquity 35(3):292-304.

Ponce Sangines, Carlos

1972 Tiwanaku: Espacio, Tiempo y Cultura (Continuación). Pumapunku 4:7-24. La Paz: Instituto de Cultura Aymara de la H. Municipalidad de La Paz.

Rivera, Mario A.

1980 Algunos fenómenos de complementaridad económica a través de los datos arqueológicos en el Area Centro Sur Andina--la fase Alto Ramirez reformulada. In Temas antropológicos del norte de Chile. Estudios arqueológicos, numero especial. Pp. 71-103. Antofagasta: Universidad de Chile.

Serracino, George

1980 Tiwanaku desde San Pedro de Atacama. Estudios Arqueologicos 5:95-106. Antofagasta.

Shimada, Melody, and Izumi Shimada

1985 Prehistoric Llama Breeding and Herding on the North Coast of Peru. American Antiquity 50(1):3-26.

Stuiver, M., and G.W. Pearson

1986 High-Precision Calibration of the Radiocarbon Time Scale, AD 1950-500 BC. Radiocarbon 28:805-838.

Stuiver, M., and P.J. Reimer

1986 [1987] University of Washington Quaternary Isotope Lab Radiocarbon Calibration Program, Rev. 2.0, bidecadal calibration data, intercept method A. Radiocarbon 28:1022-1030.

Wallace, Dwight T.

1980 Tiwanaku as a symbolic empire. Estudios Arqueologicos 5:133-144.








Figure 1. Tiwanaku settlement/influence pattern. The large triangle represents Tiwanaku; the smaller ones indicate sites attributed to any Tiwanaku period for any reason, from large habitation sites with primarily Tiwanaku ceramics to cemeteries with a few isolated Tiwanaku grave goods. The shaded area is above 4000 meters in elevation; the contour line is at 2000 meters. Sources: Ponce 1972; Mujica 1985; Goldstein in press.

Figure 2. Middle and lower Osmore valley.

Figure 3. Radiocarbon date chart

Figure 4. Radiocarbon date table

Figure 5. El Algodonal, sketch map.

Figure 6. Loreto Viejo, sketch map.


Budget Explanation

Fieldwork costs

Wages, 3 workers: $3/day X 180 days (36 weeks)

$ 1,620

Vehicle gasoline and maintenance: $50/month X 9 months




Cloth bags: 3,000 @ .25


Plastic bags: 6,000 @ .10


Recording forms, air photos


Miscellaneous string, pens, stakes, etc.


Film and processing


20 rolls Plus-X, 36 exposures @ $11


20 rolls Kodachrome 64, 36 exposures @ $12


INC supervision fee: $20/day X 130 days (26 weeks)




Laboratory analysis costs


Wages, 2 workers: $3/day X 80 days (16 weeks)


Wages, 1 data entry worker: $3/day X 40 days (8 weeks)




Plastic bags: 5,000 @ .10


Computer coding sheets and diskettes


Miscellaneous labeling supplies, packing material, etc.


Photocopying for INC and archiving in US


Radiocarbon dates: 10 @ $200




Total requested funds:

$ 10,000



Vehicle and computer to be supplied through Programa Contisuyu.

INC supervision fee is charged by the Peruvian Instituto Nacional de Culturafor the period of excavation of all archaeological projects in the country.

Airfare for Bruce Owen and some personnel maintenance costs to be covered by Department of Education Fulbright grant (see Current and Pending Support statement).

Additional radiocarbon dates, labor, and supplies will be covered by funds to be sought from other sources (see Current and Pending Support statement).