Current Research

Andean South America

[Submitted by Dan Sandweiss]

Greetings fellow Andeanists! It is a pleasure to introduce the first installment of SAAWeb's Current Research for Andean South America under my editorship. Thanks to those who submitted reports for this issue and apologies for the tremendous delay in getting this installment posted. We hope that postings will be more prompt in the future, so now is a good time for everyone to send me recent news of your research dan_sandweiss@umit.maine.edu. Reports can be submitted in English or Spanish. Postings will be twice a year,and past postings will remain available (see submission information elsewhere on the web page). Further, by agreement with the SAA, all Andean postings will be published approximately every two years in Andean Past http://kramer.ume.maine.edu/~anthrop/AndeanP.html. Given the generally long delays in publication of research results, SAAWeb's Current Research is a timely way to let the community know what you've been doing. I look forward to hearing about your work and helping share it through the web.
Dan Sandweiss

Archaeological test excavations were initiated in 1997 at the site of Shanshipampa in the Pimampiro district of northern highland Ecuador under the direction of Tamara L. Bray (Wayne State University, email tbray@cms.cc.wayne.edu). The Pimampiro district, described in sixteenth-century sources as a gateway to the eastern lowlands and an important multiethnic trade center, has been the focus of a long-term regional study by Bray since 1991. With the help of the Wayne State University archaeological field school and personnel from the Instituto Nacional de Patrimonio Cultural, the site of Shanshipampa was mapped , and a range of features was tested during the 1997 field season.

On the basis of ethnohistoric information, previous archaeological work, oral history, and local informant input, Shanshipampa has been tentatively identified as the original site of Chapí. Late sixteenth-century sources indicate that Chapí had once functioned as an important commercial center for highland and tropical lowland merchants, and housed a multiethnic population that included affiliates of eastern montaña groups. The site was abandoned at some point in the seventeenth century, and memory of its location lost to local residents thereafter.

Test excavations in the 1997 season focused on the upper portion of the site, which, according to residents, had only been cleared of forest in the past 80 years. The landowner indicated that the area was used as pasture and had never been plowed. The focus of investigations was a series of parallel linear features, associated terraces, and a small hemispherical mound. The linear features run parallel to the slope, their uphill end abutting unfaced terrace risers. These features range in width from 3 to 6 m, are approximately 1B1.8 m tall, and vary from 25 to 60 m in length. Within the area of study, there is an upper and a lower tier of these linear mounds, which are arranged parallel to one another and spaced at 12B16-m intervals. Test excavations indicate that these mounds are comprised of piled colluvium with considerable amounts of ceramic materials mixed into the upper levels. Testing between the linear mounds yielded substantially lower densities of cultural materials. Based on the evidence, Bray is presently positing an agricultural field-dividing function for these features.

A nearby oval-shaped mound, ca. 12 x 23 x 2 m, was also tested. Several possible hearths were found in the middle and lower levels of this mound. The possible hearths, in conjunction with the ceramic evidence and the number of broken manos and metates recovered, suggest a domestic function for this feature. Seven meters to the east of this mound, a dome-shaped feature of white clay with a vented, box-like structure in its center (56 x 52 cm x 83 cm deep) was uncovered. The walls of this rectangular central portion of the feature were nearly vitrified, suggesting exposure to extremely high firing temperatures. Preliminary review of the literature suggests the possibility of a smelting furnace associated with metal-working activity. While gold ornaments are relatively common in this region, we have no reported finds of metallurgical production sites anywhere in the area. This makes the preliminary identification of this feature very tentative but also potentially very important. Initial sorting of the ceramic materials from the 1997 excavations suggests the presence of Caranqui, Capulí, and Panzaleo pottery, as well as some previously unrecorded blackware types.

A series of petroglyphs was also documented around the site of Shanshipampa, offering possible evidence of trans-sierran connections. Two carved stone monuments, both of which carry imagery suggestive of tropical lowland fauna (monkeys and snakes), were documented during a visit to the area in 1996. In 1997, an additional four petroglyphs evidencing spiral, circular, and linear motifs were recorded and mapped. Several of the motifs depicted, specifically a bicephalous, snake-like creature and a monkey, are identical to those observed on the flagstone pavement at the nearby site of La Mesa salvaged by Padre Porras in the early 1970s. Similar motifs have also been observed on rock art from the Nariño district of southern Colombia and the Quijos region to the east. Preliminary analysis of the iconography suggests that the residents of Shanshipampa participated in a widespread visual system potentially indicative of a shared interregional ideographic system, interaction sphere, or panregional information network. The distribution of the petroglyphs, in conjunction with ethnographic information, suggests that they may have performed an ethnic boundary maintenance function as well.

The project also had a paleobotanical component. The Pimampiro district was a reknowned center of coca production during the late prehistoric period. During the 1991 archaeological survey, seven sites containing stone-faced terraces were recorded. Crops known to have been produced on terraces in the Andes include corn, coca, tubers, and local grains, while in the montaña zone, agricultural terraces were widely associated with the cultivation of coca. Coca was a key resource in the Precolumbian world, and Bray has posited elsewhere that control over its production may have been central to the development of social hierarchy in this region. Initial analysis of soil samples from two sets of terraces by the Smithsonian's Tropical Research Institute in 1993 indicated an abundance of potentially diagnostic phytoliths. In 1997, Cesar Veintimilla of the Escuela Politécnica Superior of Guayaquil began experimental work to assess the feasibility of identifying coca through phytolith analysis under the aegis of the Pimampiro Project. Preliminary results are promising insofar as several potentially diagnostic phytoliths have been identified in two modern plant specimens of Erythroxylon, though initial counts of these phytoliths are low.

Earl H. Lubensky (University of Missouri-Columbia [UMC], email anthearl@showme.missouri.edu) reports on work at several coastal Ecuadorian sites:
The approximately 2,000-ha Hacienda La Florida, located at Km. 16 on the road Santo Domingo de los Colorados to Quininde and Esmeraldas, at the extreme westernmost part of Pichincha Province on the Ecuadorian coast, has at least 10 archaeological sites, mostly single and complex mound sites. The two most prominent and complex sites were excavated by Earl H. Lubensky joined by Dr. Allison Paulsen in January 1979 and first reported in 1981 in the Current Research section of American Antiquity (46:201). Authorization for a test excavation (prospección arqueológico) had been granted by the director of Ecuador's National Institute of Cultural Patrimony on January 3, 1979. The two sites were called Estero Cecilia and Santa Marta, according to names provided by hacienda workers on the excavation. The Estero Cecilia site contained at least 19 mounds, four were quite small but 10 were monumental in size, up to 100 m in diameter and 8 m high. Except for two deviations, they were arranged in two almost parallel rows high above and alongside the banks of the estuary (Estero Cecilia). Of these 19 mounds, excavations were made in three of them during a two-week period; family members had excavated another previously. Figurines and figurine fragments, as well as ceramic sherds, were diagnostic of the Jama-Coaque Phase of the Regional Development Period. The Santa Marta complex contained at least 13 mounds, one large central mound about 30 m in diameter and 5 m high with the other 12 surrounding it in an elliptical pattern, essentially different from the pattern at the Estero Cecilia site. A 2-x-2-m excavation 260 cm deep was cut into the center of the principal mound. Ceramic sherds indicated a likely Chorrera association. Three radiocarbon dates from two of the mounds at the Estero Cecilia site were 950 ± 60 B.P. (Beta-43345), 770 ± 90 B.P. (Beta-43346), and 620 ± 50 B.P. (Beta-43347), or calibrated between A.D. 1020 and 1400 (M. Stuiver and B. Becker, 1993, Radiocarbon 35:35B65), indicating a somewhat later time period than conventionally accepted dates for Regional Development, but compatible with Zeidler's Jama-Coaque II of the Integration Period. The date from one test from Santa Marta was 2950 ± 80 B.P. (Beta-43348) or about 1290B1010 cal B.C. (Stuiver and Becker, see above), thus in conformity with a Chorrera phase. Obsidian artifacts from both sites were submitted for X-ray florescence (XRF) and neutron-activation (NAA) testing. Six from Santa Marta only were tested, and results show five (XRF) were from Mullumica and one (NAA) from the Quiscatola-Yanaurco source sites in the Andes.

In 1993, several possible clay source samples were collected from the Hacienda La Florida. Along with a number of ceramic vessel sherds and figurine sherds these clay samples were submitted for NAA and comparison to the UMC Research Reactor Center for the NAA program under the direction of Drs. Hector Neff and Michael Glascock. Their conclusion was that it seemed "most likely that the bulk of the pottery is locally derived, but from differently weathered sources and via paste preparation practices which modified the clay's composition." "In contrast," they continued, "the figurines and some of the pottery from Estero Cecilia are so compositionally distinct from the ‘local' pottery that they probably are imported from somewhere else." Most of the detailed research for these sites is now completed and ready to be prepared for publication.

Current research on the Ayalan Cemetery project is designed to augment and supplement Ubelaker's report (1981, The Ayalan Cemetery: A Late Integration Period Burial Site on the South Coast of Ecuador, Smithsonian Contributions to Anthropology No. 29; see also E. H. Lubensky, 1974, Los cementerios de Anllulla: Informe preliminar sobre una excavación arqueológica, Boletín de la Academia Nacional de Historia LVII(123), Quito, Ecuador). The cemetery was judged to be a "Late Integration" site located on the west bank of the Estero Salado about 55 km southwest of Guayaquil in the Anllulla sector of the Hacienda Ayalan. Ubelaker analyzed principally the human skeletal remains, the urns used as repositories for secondary burials, and the burial goods. There was, however, considerable midden material (ceramic, metal, animal bone, shell, and lithic) that still needs to be reported for complete coverage of the site. It appears that the burials were either placed in an already existing midden, or midden material was deposited during or between burial episodes. Part of this material is on loan from the Smithsonian to the Museum of Anthropology at UMC, part is at the Smithsonian on loan from Ecuador, and a large amount is at the Catholic University in Guayaquil at the museum now established there under the terms of an agreement worked out at the termination of the excavation in 1973. The principal research involved now is development and refinement of a typological classification system for ceramics at the site, which would also include much of the ceramic burial goods as well. The ceramics are largely from the Integration Period, specifically Milagro-Quevedo and Manteño-Huancavilca, but it is believed that there are also deposits from the Guangala and Chorrera phases, yet to be analyzed more thoroughly.

Excavation of Anllulla Shell Mound on the Hacienda Ayalan was undertaken as an ancillary project to the Ayalan Cemetery project, some 2,000 m to its north, to determine whether the shell mound might be associated with the settlement site related to the cemetery. Radiocarbon dates and ceramics found among the shells indicate that the shell mound was about 2,000 years earlier, the cemetery principally dating to the Integration Period and the shell mound containing ceramics (with the exception of approximately the top 70 cm) of the earlier Valdivia-Machalilla phases. The excavation was a 2-x-2m "telephone booth" 4 m deep. The bottom 30 cm of the cut were aceramic. The next 90 cm contained a number of plain and unclassified sherds, not positively identifiable as to phase, but probably Valdivia. Decorated sherds in the next 140 cm were diagnostically Valdivia Periods D (in the typology of B. J. Meggers, C. Evans, and E. Estrada, 1965, Early Formative Period of Coastal Ecuador, Smithosonian Contributions to Anthropology Vol. 1, hence MEE) or VI-VIII (in Betsy Hill's typology, 1966, A Ceramic Sequence for the Valdivia Complex, Guayas Province, Ecuador, M.A. thesis, Columbia University). They were Punta Arenas Incised, Valdivia Appliqué Fillet, Broadline Incised or Carved, Nicked Broadline Incised, Pebble Polished, and Zone Incised (MEE's typology), with a few unclassified decorated sherds. In the next 70 cm (70B140 cm from the top) two types of diagnostic Machalilla sherds appeared, Double Line Incised and Incised and Punctate, along with a number of Polished Red sherds and 3 (only) Red Banded sherds, assumed also to be Machalilla. Along with these in the 70B140 cm levels, diagnostic Valdivia sherds, specifically Punta Arenas Incised, Valdivia Appliqué Fillet, Broadline Incised, and Brushed, appeared along with the Machalilla. Above 70 cm from the top there were no more diagnostic Machalilla or Valdivia sherds, but it was assumed that the continuing relatively large number of Polished Red and Red Banded sherds, along with many plain sherds, were within pure Machalilla Phase levels. Significant ash (presumably volcanic) levels were recorded precisely at 70 cm depth and at about 140 cm depth, depths of almost dramatic ceramic change. Furthermore, radiocarbon dates taken from levels below 150 cm (in stratigraphic order, 3560 ± 95 B.P. [N-2908], 3560 ± 95 B.P. [N-2908], 4020 ± 220 B.P. [P-2761], 3530 ± 100 B.P. [N-2909]) confirmed the assumption of a late Valdivia Period, transitional to Machalilla (around 3500B4000 B.P.). A single date of around 1380 ± 70 B.P. (N-2907), cal A.D. 610B800 (Stuiver and Becker, see above), at 30 cm depth was at first rejected as being too recent, but two more dates taken from charcoal samples, from 60 and 80 cm, found among bone at the Smithsonian also gave very recent dates (1220 ± 70 B.P. [BETA-21814], 1560 ± 90 B.P. [BETA-21815]), calibrated between A.D. 420 and 940 (Stuiver and Becker, see above). This left the possible Valdivia-Machalilla overlap level of 70B140 cm without confirming radiocarbon dates. A small sample of charcoal from the 100B110 cm level was then submitted for accelerator mass spectrometry (AMS) testing and the date of 3210 ± 50 BP (BETA-108164) tended to confirm the Formative period date of the deposits at that level and in the overlap levels. Current research is directed toward determining whether ceramics at the upper 70 cm levels are from much later occupations, possibly during the Guangala or Jambeli phases, in order that the data from these upper levels can be incorporated into the analysis, for a site report, already completed for all other artifacts and of the shells themselves. Problems of shell mound stratigraphy are fully recognized.

Research continues at UMC on the surface collections made by Edwin A. Ferdon, Jr., in Ecuador before and during World War II under the auspices of the School of American Research, the Archaeological Institute of America, and the University of Southern California of Los Angeles. His mission included making an archaeological survey of "the old northern Inca province" and excavation of a selected site. Although interrupted by the war in September 1943, when he began to serve with the U. S. Cinchona Mission, he surveyed 118 sites, for which he made many detailed maps. He collected artifacts from the surface of, or from cuts at, 62 of these sites. He excavated the La Carolina site at La Libertad on the Santa Elena Peninsula, about which M. Simmons wrote his doctoral dissertation in 1970 (The Ceramic Sequence from La Carolina, Santa Elena Peninsula, Ecuador, Ph.D. dissertation, University of Arizona). Lubensky wrote his doctoral dissertation in 1991 using the site surveys from the 35 sites surveyed in Esmeraldas Province and vessel ceramics from 16 of those sites from which collections were made (The Ferdon Collections of Prehistoric Ceramic Vessels and Sherds from Esmeraldas Province, Ecuador, Ph.D. dissertation, UMC). In his dissertation, Lubensky presented a computerized system for analysis and recording of ceramic attributes of vessel shape, vessel size, manufacture, and decoration. From this, ceramic classes were developed and a suggested ceramic seriation was presented for prehistoric Esmeraldas Province. Students under Lubensky's guidance have completed attributes analysis of ceramics from three sites in Manabí Province and two sites in Guayas Province. Steven Velasquez prepared a consolidated report on the 14 sites surveyed in Manabí Province, and Jessica Coats prepared a paper on the site of Jaramijo in Manabí Province; both were students at the time in the Anthropology Department at UMC. Two other Missouri students, Pamela J. Hale and Julia Anne Wagner, completed analysis of the ceramic sherds from a second site at La Libertad where Ferdon had made a surface collection. Hale and Wagner gave a paper at the Midwest Andean meeting in Ann Arbor in 1994, entitled "Lost Legacy of La Libertad," and published a version of that paper in the Missouri Anthropology Department publication, PUMA (Vol. 1, No. 1, Winter 1995). Gene Keay did an initial analysis on the collection from the Punta Carnera site in Guayas Province. Collections from 24 additional sites from Guayas Province await analysis, as do smaller collections from Imbabura, El Oro, Pichincha, Los Rios, Tungurahua, and Chimborazo Provinces. Interrelationships among the sites and the collections, and possibly additional seriations, will be possible with the computer program (PARADOX) used for the analyses. The Ferdon surface collections are on loan to the Anthropology Museum at UMC from the Museum of New Mexico in Santa Fe, to which sometime in the past the School of American Research transferred the collections (the La Carolina collection remains at the University of Arizona in Tucson, where Simmons did his research).

PERU

In collaboration with Izumi Shimada (Southern Illinois University, Carbondale), Julie Farnum (UMC) has been studying the Sicán skeletal material excavated from Batán Grande, in the Lambayeque Valley on Peru's north coast. She hopes to return this summer (1998) to study a commoner skeletal series to contrast with the occupants of the shaft tomb.

Jack Rossen (Zana-Niepos Project, email JROSSEN@ithaca.edu) conducted excavation in July and August 1997 at Cerro Guitarra, a preceramic village in the lower Zaña Valley on Peru's north coast. The first two weeks consisted of theodolite mapping of the site and its topographic surroundings. The locations of 83 houses were documented on three site levels: hill base, slopes, and summit. Also mapped were a stone-lined public plaza in front of the hill, the dry riverbed adjacent to the site, and various paths and stairways that connected different house clusters. The final three weeks was used to excavated eleven semi-subterranean houses and test excavate the public plaza area. A large quantity of lithics, fauna, plant remains, and special samples were recovered. Artifact and sample analyses are underway.

The research was successful in gaining an understanding of a previously poorly understood phenomenon: the hillside preceramic villages of northern Peru, with their expected dates of about 3000B4000 B.C. Site layout is now understood as a series of interconnected houses clusters, and the structure of individual houses as primarily two-room, semisubterranean, elliptical, thatched-roof structures including lithic raw-material storage areas and hearth niches. Preliminary analysis indicates the site contains a variety of cultivated plant remains. The research will ultimately document social structures and village formations that accompanied early plant cultivation in the region. The public plaza may also represent early public ceremony and formation of group identities.

Rick Sutter, a 1997 Ph.D. from UMC, is beginning a program of research using dental traits to establish the major outlines of the peopling of South America. He will use Peruvian central coastal materials as well as dental materials from as wide a geographical spread as possible.

Kate Pechonkina (UMC) has discovered two different groups in the human remains from the early Early Intermediate site of Villa Salvador near Pachacamac (Lurín Valley, central coast of Peru). These two groups differ in deformation, physical size, and stress markers. These are contrasted with similar groups she studied from Huaca Pucllana.

Bob Benfer's (UMC, email anthrab@showme.missouri.edu) recent work has involved taking photographs of the reconstructed house in the National Museum of Perú as well as the site of Paloma (near the Chilca Valley on Peru's central coast) from which he excavated it. The photos were needed for the virtual reality component of the PalomaWorld Project, which is an intelligent digital library that will include all the research materials from the Paloma Project. Benfer's web site at present has only a small content component, but it can be visited at http://gonk.atc.missouri.edu/paloma/.

UMC graduate students Kate Pechonkina and Julie Farnum have collected data from Cotton Preceramic Period Asia site skeletons. They will use these data to assess a hypothesis under investigation by Joe Vradenburg and Bob Benfer (UMC) that there was a new treponemal disease introduced to the central, western flanks of the Andes from the Amazonian region in the late Initial Period that caused its collapse and permitted the widespread growth of the Chavín cult. Vradenburg and Benfer, with Krystoff Makowski (Pontificia Universidad Católica del Perú) and Mercedes Delgado as junior authors, presented the current state of this research at the the 26th Annual Midwest Conference on Andean and Amazonian Archaeology, Anthropology and Ethnohistory at the University of Illinois, Urbana-Champaign.

In 1996, Dan Sandweiss (University of Maine, email dan_sandweiss@umit.maine.edu) carried out excavations and survey at Quebrada Jaguay, near Camaná on the south coast of Peru. Project members included co-director Rolando Paredes, archaeologists Bernardino Ojeda and Maricarmen Sandweiss, students Heather McInnis (University of Maine) and Trevor Ott (University of Calgary), and field technician Osvaldo Chozo. Excavations were centered at Quebrada Jaguay 280, an Early Preceramic site discovered in 1970 by Frederic Engel. A suite of radiocarbon dates on charcoal from the site now confirms that there was a Terminal Pleistocene occupation dating between about 11,000 and 10,000 B.P. (uncalibrated) and an Early Holocene occupation dating between 9500 and 7500 B.P. (uncalibrated). Fish and shellfish provided all of the animal protein during both occupations, indicating great antiquity for maritime adaptations in the region. Full cover survey of the surrounding region (5 km to the northwest, 5 km northeast, and 10 km to the southeast) identified more than 60 sites, mostly preceramic. Comparison of surface remains and dates on basal material from shovel tests show that the sites represent two preceramic periods, one between ca. 9500 and 7500 B.P. and one centered around 4000 B.P. A few ceramic-bearing sites were also found. Analyses of excavated materials are ongoing, and future seasons are planned.

BOLIVIA

During May-July of 1996 the Taraco Archaeological Project, codirected by Christine Hastorf and Matt Bandy of the University of CaliforniaBBerkeley (UCB), and Mario Montaño Aragón of Bolivia's Instituto Nacional de Cultura, conducted research at Chiripa, a site on the SW shores of Lake Titicaca in Bolivia. The core team members are Dr. Lee Steadman, Matt Bandy, Emily Dean, David Kojan, and William Whitehead, graduate students at UCB, and José Luis Paz, a graduate student at the University of San Andrés in La Paz. Additional help in the field came from Amanda Cohen of the American Museum of Natural History, Melissa Goodman, and Ian Hodder of Cambridge University. Further specialist help is coming from Dr. John Southon of the Lawrence Livermore Laboratory Dating Lab. Dr. Susan deFrance of the Corpus Chirsti Museum of Science and History, Dr. Kate Moore of the University of Pennsylvania, Dr. David Steadman of the Florida Musuem of Natural History, and Deborah Blom of the University of Chicago will be working on bone analysis, the first three on the animal bone and the latter on the human bone.

This is the second field season of this project (the first was in 1992), although Bennett in the 1930s, Kidder and Cordero in the 1950s, Portugal in the 1960s, and Browman in the 1970s have excavated there previously. From their work at Chiripa, we know about the existence of the central mound of the site. The .36-ha mound has at least three levels of building and probably dates from 1400 B.C. on up through the Tiwanaku times of A.D. 1000. The "temple" (as it is called today) visible on the surface is a Tiwanaku III monolith-lined courtyard, but shows continuity with the earlier constructions. Chiripa is especially important to Andeanists because of its proximity to the major imperial center of Tiwanaku. Tiwanaku became an important center for almost 800 years around A.D. 300, and therefore Chiripa is considered one of Tiwanaku's important precursors. Chiripa was an important center during the earlier Formative phase of Bolivian prehistory, with a ritual center overlooking the shores of Lake Titicaca as well as the string of snow-peaked mountains to the east. The Formative phase, as it is called in Bolivian archaeology, ranges between 2000 B.C. and A.D. 400, with Chiripa dating to the Middle and Upper Formative phases.

The Taraco Project's research goals are several. While trying to understand the Formative phase in the south-central Andes, including the detailed artifactual changes, they are especially interested in the early dynamics of increasing political scale and subsistence changes, particularly agriculture. At Chiripa Formative sites, these activities are clearly intertwined with intensive ritual actions, which are also under investigation. The Project has been trying to uncover evidence of domestic areas and activities, but this is proving difficult. Excavations looking for domestic houses and middens took place in two areas, above and below the mound. Instead, both areas, Llusco and Santiago, have yielded large (approximately 11 x 13 m) semisubterranean stone-walled enclosures. At this stage in the research, the investigators believe these to be ceremonial (gathering) areas rather than habitation zones. Llusco, excavated first by Claudia Rivera in 1992 and then José Luis Paz in 1996 dates to the Late Formative, ca. 8000B600 B.C. It has patches of white plaster on the floor and a drainage canal in the lower, northwest corner.

Santiago, opened first by Sonia Alconini and Sigrid Arnott, but substantially expanded in 1996 by Emily Dean and David Kojan, directly overlooks the lakeshore on the first terrace. This area of the site is a complex of use and rebuilding layers dating throughout the Formative times. There are surfaces that seem to be fairly clean and in some cases with yellow or white plaster, with burials throughout. It seems that at least ritual burial and feasting occurred in this section, but also there could have been some habitation nearby. This area on the site is extremely important yet still mysterious as to the total types of activities that occurred there. The stratigraphy is complicated and will gain from the microstratigraphic analysis by Melissa Goodman. More excavation must be done there to explain this sector.

To the west, directly adjacent to Santiago, at the end of the 1996 season, Project members discovered a 13-x-14-m stone wall enclosure thought to be an earlier semisubterranean stone-walled courts. This one, called Choquehuanca, could date as early as 1500 B.C. Because this structure was only outlined in the last few days of excavation, the details of its construction remain unknown; however, on its east wall there is a small stone niche, about 1 m in length within the wall, although it contained nothing visible. This could be what is seen in later Pukara, Tiwanaku and even Inkaic enclosures, niches within walls that held important sacred objects. Part of the inner stone wall was plastered. The Llusco and Choquehuanca enclosures are the earliest in this region to date. Other investigated, dated structures such as these fall within the 400B200 B.C. range. This evidence suggests that Chiripa had some of the earliest ceremonial sectors in the region.

The site's importance is further seen in the systematic surface collections that were completed this year. The Formative site now seems to be about 7 ha in size. Given that these ceremonial areas, Llusco, the mound, and Santiago only make up less than 1 ha., there seems to have been substantial residence surrounding this central precinct, much larger than other Formative sites found in the nearby regional surveys.

The final part of the 1996 archaeological project was to better understand the Formative portions of the mound. Matt Bandy undertook several cleaning operations on the mound. His main task was on the east side of the mound, which had been cut back by the community in the 1960s to build a football field. There he uncovered clear evidence of the "Upper and Lower House" Formative levels. The previous work on the mound suggested that there were 15 structures surrounding a sunken plaza. This has been substantiated, but sadly, no structures are untouched and complete. Ten structures, however, are probably in good condition. All the evidence suggests that these were not lived in but were more likely ceremonial structures. The bins do not seem to have held quantities of crops like Inka collca. Each structure perhaps was used by an ayllu, or an extended family associated with a territory. These structures are made of adobe and stone, with plastered surfaces on the walls and floors. Four superimposed structures were seen in the eastern profile and further, we learned that each structure has a series of yellow plaster floors. Between each re-flooring of these structures, there is evidence of ritual sealing, with sterile soil or sand laid down, often accompanied by a fire. Further evidence of such floor treatment also was seen in the cleaning of historical fill along the south face of the mound. Most floors looked clean, although the top "Lower House" floor had lots of fish remains and pottery. One hearth was also encountered.

Melissa Goodman of Cambridge University joined the project briefly to collect microstratigraphic samples of all areas. This research should show what types and intensities of activities were undertaken in these mound structures. In collaboration with John Southon and the Lawrence Livermore Laboratories, the Project also hopes to run a series of detailed AMS dates for each of these floors, to understand the timing and scope of the re-flooring events in the mound.

Lee Steadman oversaw the ceramic analysis and ran the laboratory. Not only do the ceramics hold the key to the dating and seriation of the site and its relationships to other villages within the region, but they are critical for identifying the activities on the site, such as cooking, storage, and ceremonial and burial practices. Given that a detailed ceramic sequence has not been fully formed for this part of the Titicaca Basin, Lee is completing essential work on the Formative into the Tiwanaku phases in the southern Titicaca Basin. Three ceramic phases are defined for the Chiripa occupation of the site, based on observable differences between the ceramics in the stratigraphic levels and 14 new absolute dates. The terms used by the Taraco Project have been used before by Karen Chávez based on Kidder's excavations. The Taraco Project phases are adjusted slightly earlier. These phases are: Early Chiripa 1500B1000 B.C., Middle Chiripa 1000B800 B.C., and Late Chiripa 800B100 B.C. (calibrated).

William Whitehead oversaw the collection, processing, and sorting of the soil flotation samples from the excavations. The project followed a blanket collection strategy of 10-liter bulk (point-provenienced) samples. Furthermore, in midden and surface contexts, average soil samples were also collected, to improve the representation of those contexts. In all, 384 soil samples were collected and processed. These are important for recovering plant remains but also for fish and other small animal bones as well as to provide quantitative samples of all artifacts.

Concurrent with the Taraco Project excavations, Mario Montaño Aragón worked on a toponymic map of the local catchment area in and around the modern community of Chiripa. He discovered hundreds of place names that link to past and present activities in the area.

The Taraco Project can be contacted by email through Christine Hastorf (Hastorf@qal.berkeley.edu) or at Chiripalist@uclink.berkeley.edu.

MEETINGS

The 16th Annual Northeast Conference on Andean Archaeology and Ethnohistory was held October 4 and 5, 1997, at the University of Maine in Orono, Maine. The program remains available on the web at: http://kramer.ume.maine.edu/~anthrop/AndeanConf97.html.

The 26th Annual Midwest Conference on Andean and Amazonian Archaeology, Anthropology and Ethnohistory was held at the University of Illinois, Champaign-Urbana on February 28 and March 1. The program and related information can be accessed on the web at: http://www.staff.uiuc.edu/~helaine/main.html.