Академический Документы
Профессиональный Документы
Культура Документы
Volume 10
2012
Editors
MONICA BARNES
American Museum of Natural History
and
DANIEL H. SANDWEISS
University of Maine
Associate Editor
RICHARD L. BURGER
Yale University
THOMAS F. LYNCH
Brazos Valley Museum of Natural History
MICHAEL E. MOSELEY
University of Florida
ISSN 1055-08756
The Cornell Latin American Studies Program is the publishing institution for ANDEAN PAST.
Copyright for ANDEAN PAST resides with Cornell LASP on behalf of the editors unless a specific
portion, for example, an illustration, is noted as copyrighted by another party. Authors may re-
publish their ANDEAN PAST articles, obituaries, or reports, in English or in translation, in print, or
in electronic format, provided that at least one year has elapsed since the original publication in
ANDEAN PAST, that prior publication in ANDEAN PAST is indicated in the republication, that
Cornell LASPs copyright is acknowledged, and that the editors of ANDEAN PAST are notified of the
republication. If a portion of an article is copyrighted by a third party, authors must request specific
written permission from that party to republish. This includes on-line postings in electronic format.
Orders should be addressed to: Latin American Studies Program, 190 Uris Hall, Cornell University,
Ithaca, New York 14853, U.S.A.; telephone (607) 255-3345, fax (607) 255-8989, e-mail:
lasp@Einaudi.Cornell.edu
Printed copies of Andean Past 1-4 and 6-10 are available as of 20 March 2012.
Andean Past 5 is available free on line from:
http://lasp.einaudi.cornell.edu/publications_andean past
Inquiries and manuscripts submitted for future volumes should be sent to:
Monica Barnes
377 Rector Place, Apartment 3C
New York, New York 10280
Cover: Farfan from the top of Cerro Faclo to the west of the site. Compound II is visible. The
photograph was taken by Howard Tsai on June 26, 2004.
ANDEAN PAST
Volume 10
2012
TABLE OF CONTENTS
OBITUARIES
Earl Henry Lubensky (March 31, 1921 - May 1, 2009) by Deborah M. Pearsall. . . . . . . . . . . . 1
Juan (Hans) Santiago Ren Schobinger (February 18, 1928 - July 11, 2009)
by Constanza Ceruti .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
ARTICLES
A Changing Society? Craft Specialization and Complementarity Systems during the Formative
Period in the Cochabamba Valley, Bolivia by Olga U. Gabelmann. . . . . . . . . . . . . . . . . . . . . . 27
The Well and the Huaca: Ceremony, Chronology, and Culture Change at Huaca Cao Viejo,
Chicama Valley, Peru by Jeffrey Quilter, Rgulo Franco J., Csar Glvez M., William Doonan,
Catherine Gaither, Vctor F. Vsquez S., Teresa E. Rosales T., Jaime Jimnez S., Hal Starratt,
and Michele L. Koons.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Adobe Bricks and Labor Organization on the North Coast of Peru by Howard I. Tsai. . . . 133
The Destruction of the Yurac Rumi Shrine (Vilcabamba, Cusco Department) by Brian S.
Bauer, Miriam Dayde Aroz Silva, and George S. Burr. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Regional Associations and a Ceramic Assemblage from the Fourteenth Century Llanos de
Mojos by John H. Walker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
RESEARCH REPORTS ON ANDEAN ARCHAEOLOGY
Recent Excavations at the Late Intermediate Period Village of Yanaorco in the Cajamarca
Highlands by Jason L. Toohey.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
A Colonial Human Burial Excavated in 1965 between Portals 5 and 6 at Hunuco Pampa by
Monica Barnes, Catherine Gaither, Robert A. Benfer Jr., and Daniel Shea.. . . . . . . . . . . . . . . . 267
Mitomarca: A Possible Fortification in the Upper Huallaga Basin by Yuichi Matsumoto, Jason
S. Nesbitt, and Denesy Palacios J. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
Architecture in the Coastal Desert by Simn Urbina A., Leonor Adn A., and Estefania Vidal
M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
La Bolsa 1 Site, Taf Valley, Tucumn Province: Household Mortuary Practices in a South
Andean Village (First Millennium A.D.) by Julin Salazar. . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
The colophon of Andean Past states that we contribution, Adobe Bricks and Labor
are a peer-reviewed, numbered publication Organization on the North Coast of Peru. Tsai
series dedicated to research in the archaeology examines bricks from the Moche site of Huaca
and ethnohistory of western South America. de la Luna in the Moche Valley and the Chim
Nevertheless, the interface between archaeology site of Farfn in the Jequetepeque Valley. He
and history is not always smooth. At least since builds upon earlier work by Charles Hastings
the mid twentieth century, when John Howland and Michael Moseley, by Alan Kolata, and by
Rowe published his first articles on the Inca, Raffael Cavallaro and Izumi Shimada in
when John Victor Murra launched his A Study studying adobe bricks bearing makers marks.
of Inca Provincial Life project centered on the These are diagonal slashes and other simple
great site of Hunuco Pampa (Barnes, Andean signs that permit accounting. Taking into
Past 9, Barnes et al. this volume), and when Juan consideration architectural context, Tsai
Schobinger systematized the study of Inca postulates changing labor relations as suggested
sacrificial entombments on high mountain tops, by the marked bricks. At the Moche
those who study the Andean past have construction of Huaca de la Luna, Tsai
understood that neither archaeology, nor confirmed Hastings and Moseleys observation
documents alone, can fully reconstruct life in that bricks of the same size, shape, and mark
previous centuries. Both kinds of records have were placed together in the same construction
lacunae, even when combined with the insights segments. By contrast, at Chim Farfn, bricks
of other relevant sciences such as geography, lacked makers marks and bricks of different
geology, biology, ecology, anthropology, sizes can be found in the same construction
linguistics, mathematics, physics, and chemistry. segment. Tsai suggests that Moche patterns of
In Andean Past 5, the late Catherine J. Julien labor organization differed from those in force
pointed out an excellent example. A visita or during Chim times. At Moche, each makers
official inspection tour to Pocona, in what is mark seemed to represent a distinct social
now Bolivia, provides good information on Inca group. Tsai believes that his data best fit
resettlement of the region, but does not mention Moseleys labor tax model for Huaca de la Luna.
the monumental site of Incallacta described by That is, members of a given community were
Vincent R. Lee in the same volume of Andean responsible for all stages of brick-making,
Past. Conversely, the study of the site of Inca- transportation, and construction. However, at
llacta alone would not tell us all it is possible to Farfn, it seems that a task differentiation
know about Inca control of this region. In model, as proposed by Cavallaro and Shimada,
Andean Past we strive to integrate all lines of fits the data well. That is, a sponsor had various
evidence. groups make bricks, which were then
transported and put in place by other people.
Using both physical evidence and logical
models, Howard I. Tsai makes progress in Although the contribution of Jeffrey
understanding the social means of production in Quilter, Rgulo Franco J., Csar Glvez M.,
the Moche and Chim cultures with his William Doonan, Catherine Gaither, Jaime
Jimnez S., Hal Starratt, and Michele L. Associating elements of both architectural
Koons, The Well and the Huaca: Ceremony, vessels and real buildings, she can begin to
Chronology, and Culture Change at Huaca Cao reconstruct the ceremonies performed in them.
Viejo, Chicama Valley, Peru focuses on
settlement, social class, water use, From the foundation of Andean Past in the
ceremonialism, burials, and what one could mid-1980s, until his death in 2006, Craig Morris
almost call gentrification, it, too, is relevant to served as a member of our editorial board
the study of adobe bricks. Tall adobes, which (Lynch and Barnes, Andean Past 8). It is,
have also been found in association with later therefore, particularly appropriate for us to
Moche structures at the Huacas de Moche, honor him with the publication of Inca Storage
including the Huaca del Sol and Platform 3 of and Accounting Facilities at Pachacamac by
the New Temple, were found in association with Peter Eeckhout, a paper first presented in Lima
construction above the sealed well and in in 2010 at a symposium in Craigs memory.
Montculo 2. This now provides evidence for Following on Morriss pioneering studies of Inca
the context of tall adobes at two different sites. storage, Eeckhout presents new data on Inca
collcas or storehouses. He bases his contribution
The authors calibration of the Cao Viejo on his own excavations at Pachacamac, the
dates indicates that Huaca Cao Viejo and great Peruvian central coast oracle site. He
Huaca de la Luna are more contemporaneous relates quipus and other accounting and
than previously thought. This contemporaneity recording devices found in situ to possible
suggests that Huaca Cao Viejos decorative storage facilities. He also interprets the overall
program may have been influenced by that of role of Pachacamac as a pilgrimage center and
Huaca de la Luna (or another site), and not the shares his ideas on the function of storage there
other way around. and elsewhere on the Andean coast, drawing
contrasts with Inca storage as practiced in the
Quilters excavations also provide insight on Sierra.
the non-ritual foods that lower status people
consumed in this area during the Early Another study of excavated Inca
Intermediate Period, not all of which are easily architecture, this time in the ceja de selva or
produced on the coast. I am grateful to Juliet eastern Andean slopes, is The Destruction of
Wiersema for the insights on the work of the Yurac Rumi Shrine (Vilcabamba, Cusco
Quilter and colleagues expressed above Department) by Brian S. Bauer, Miriam
(personal communication 2011). Dayde Aroz Silva, and George S. Burr. Yurac
Rumi (literally white rock) is an important
Juliet Wiersemas own research in this Inca shrine in the Vilcabamba region. It served
volume employs an entirely different as a ritual focus in the territory to which part of
methodology for studying architecture. In Mo- the Inca royal court retreated in the decades
che Architectural Vessels: Small Structures, Big after the Spanish conquest. Bauer and his co-
Implications, an article based upon her authors are able to describe and date both the
doctoral dissertation, she examines and building and the destruction of Yurac Rumi,
interprets Moche ceramics depicting buildings. taking into account archaeological and ethno-
Wiersema is able to form an architectural historical data.
typology on this basis. This she matches with
actual Moche structures, especially at Huaca Melissa S. Murphy and Mara Fernanda
Cao Viejo and Huaca de la Luna, Moche. Boza provide new information on an issue that
vii - Barnes: Editors Preface
published Spanish chronicles and local archival Regional Associations and a Ceramic
records leave unclear: did the Inca effectively Assemblage from the Fourteenth Century
limit consumption of coca leaves to the elite and Llanos de Mojos. The Llanos de Mojos,
certain restricted occupational and age groups? centered on Bolivia, also extend into Peru and
Certainly coca use is now widespread in the Brazil. Within them Walker has made
Andean countries. It was promoted, although archaeological transects at El Cerro, a large
not without criticism, by Spanish officials and island of high ground. From these transects he
entrepreneurs, both because it made work in has been able to recover stratified artifacts,
harsh conditions somewhat more bearable, and especially ceramics, and associate them with
because there was money to be made in its radiocarbon dates. On this basis he posits
trade. Is contemporary use of coca by numerous and dense connections between the
homemakers, workers, and farmers a lingering Llanos de Mojos and other regions, revealing a
result of colonialism, or did it have deeper roots? large sphere of interaction.
In A Bioarchaeological Study of Coca Use and
Coca Leaf Chewing at Puruchuco-Huaquerones, It is becoming clearer that such complex
Peru, Murphy and Boza analyze dental interactions are of long duration in western
evidence for coca chewing at two Late Horizon South America. In A Changing Society? Craft
cemeteries near Lima. They determine sex and Specialization and Complementarity Systems
age for individuals and make rough assessments During the Formative Period in the
of social class on the basis of grave goods and Cochabamba Valley, Bolivia Olga Gabelmann
mortuary treatment. Although their evidence is infers such complexity. Although the early
not unambiguous, they are able to suggest that societies of Cochabamba often have been seen
coca consumption was not restricted to the as homogeneous, non-stratified agricultural
elite, or to one sex, at least in these places. groups because of their use of monochrome
pottery and their lack of monumental
Sometimes ethnohistorical evidence seems architecture, Gabelmann argues that the site of
almost incredible, but is later supported by Santa Luca yields evidence of craft
archaeological field-work. When Gaspar de specialization and considerable economic
Carvajal described the numerous and large interaction. This involved horizontal as well as
Indian villages along the Amazonian rivers that vertical complementarity.
he claimed to have observed in the 1540s, the
forests of tropical South America appeared to be In Andean Past we have already devoted
rich lands. However, by the mid-twentieth considerable space to the life and work of John
century, opinion had shifted and was seen to be, Victor Murra. Murra (Barnes, Andean Past 9),
in Betty Meggers words, a counterfeit and his colleague Craig Morris (Lynch and
paradise, apparently lush, but actually of Barnes Andean Past 8), along with John How-
limited carrying capacity. There appeared to be land Rowe (Burger, Andean Past 8), are widely
reason to doubt Carvajals veracity. However, and appropriately credited with launching
very little archaeology had been accomplished historical archaeology in the Andean countries.
east of the Andes. As more work has been done Although we lack many sorts of emic records,
in this vast region, the role of conscious pre- we do have the next best things, information set
hispanic human activity in forging the landscape down by a variety of observers shortly after the
has become more and more apparent. In this Spanish conquest and the physical remains left
volume John H. Walker makes a fine by the Incas. In the middle of the last century
contribution to our growing knowledge with Rowe and Murra began to combine physical and
ANDEAN PAST 10 (2012) - viii
documentary evidence to produce richer and Missouri. Earls work appeared twice in
syntheses than had been possible in the past. Andean Past. He reported on Hacienda La
Less widely appreciated, at least in the United Florida, Ayalan Cemetery, Anllula Shell
States, are the contributions of Juan (Hans) Mound, and Ferdons Surface Collections in
Santiago Ren Schobinger. These are recorded Andean Past 6. In Andean Past 8 he published a
in this volume by Constanza Ceruti. Few long obituary of Edwin Nelson Ferdon, Jr.,
archaeologists have achieved the breadth of whose Ecuadorian ceramic collections were
Schobingers knowledge. Hans made fundamental to Earls own studies. In working
contributions to sub-fields as wide ranging as with Earl on his publications he and I became
human evolution, rock art research, shamanism, personal friends, and at numerous conferences,
and the Late Horizon. He was also a beloved as well as visits in New York City and in
colleague and educator and is missed by many in Lawrence, Kansas, I got to meet some of his
Argentina and beyond. relatives and to appreciate many of Earls fine
qualities. It should be said that Earl was at least
Perhaps Schobingers most outstanding as dedicated to the welfare of his family
achievement was the launching of the sub-field members and to the establishment and
of high altitude archaeology. In 1964 mountain maintenance of world peace, as he was to
climbers encountered the desiccated corpse of a archaeology, and thats saying a lot!
young man on Argentinas El Toro peak. Scho-
binger organized the recovery, preservation, and Once again we present short, tightly focused
scientific study of this body, which turned out to notes in our Research Reports section. These
be a Late Horizon sacrifice. Schobinger had are subject to editorial review only.
always been interested in the spiritual aspects of
world cultures, in contrast to Murras focus on Jason L. Toohey writes about his Recent
materialism, economics, and political Excavations at the Late Intermediate Period
organization. Hans intellectual proclivities, Village of Yanaorco. This Cajamarca site
combined with his dedication to the archaeology incorporated substantial fortifications. As has
of his country and region, led him to seek and been observed elsewhere in the Peruvian
study other high altitude burials, including those highlands, the Late Intermediate Period was a
on Mount Aconcagua and on Mount Chuscha, time when populations shifted from lower
collaborating with a number of distinguished elevations to higher, fortified residential sites.
colleagues including Constanza Ceruti, John Toohey describes Yanaorcos defenses and
Hyslop, and Johan Reinhard, among others. In houses, comments on its artifact assemblages,
so doing, Schobinger added greatly to our and argues that the site was occupied by
understanding of the Capaccocha, the members of at least two social strata.
important and elaborate Inca sacrifice of
children on high mountains, described by From the Hunuco Region we have two
Rodrigo Hernndez Prncipe in 1622. reports, one focusing on new field-work, the
other on excavation done in 1965. From 1963
In this issue we also mark the passing of until 1966, John Victor Murra directed his A
Earl H. Lubensky. Lubenksys thesis advisor, Study of Inca Provincial Life project centered
Deborah M. Pearsall shares her perspective on around the great Inca site of Hunuco Pampa.
his life. While pursuing a diplomatic career, Although Murra oversaw the first archaeological
Lubensky also developed an interest in excavations there, this aspect of his work has
archaeology, working in Ecuador, El Salvador, been overshadowed by that of his protg Craig
ix - Barnes: Editors Preface
beginning with lightly built Archaic John Victor Murras first students at Cornell. At
constructions of organic materials dating as the time of her death she was working with
early as 4030 B.C. and continuing a little more Heather to publish Murras Lewis Henry
than a millennium later with more elaborate Morgan lectures delivered at the University of
stone and mortar houses. Towards the end of Rochester in 1969. This task is nearly complete
the Archaic Period, sealed floors separate and will be seen to fruition by Heather.
mortuary remains from the living spaces above
them. By the Formative Period, more formal, ag- As this volume enters the archaeological
glutinative architecture, often with rectilinear record, readers can be assured that the next one
plans, had been developed. During the Late is in progress. We have already accepted papers
Intermediate Period, architecture continued to for it, but because we have no backlog, there is
develop with double-faced stone-and-mortar room for more. As always, we look forward to
structures appearing. hearing from our colleagues who have work to
publish.
In Household Mortuary Practices in a
South Andean Village Julin Salazar focuses Monica Barnes
on funerary customs during the first millennium in the City of New York
A.D. at the La Bolsa 1 site in the Taf Valley, 15 April 2012
Tucumn Province, northwest Argentina.
Salazars report is also based on recent field
research. To quote Salazar, his research aims to
establish relationships between daily practices
and social reproduction strategies and
household strategies, and to analyze the active
role of household material assemblages in the
construction of kinship relations and identity.
He approaches his goal through the analysis of
a household burial context.
DEBORAH M. PEARSALL
University of Missouri, Columbia
Earl Lubensky, New World archaeologist, of the National Honor Society. Henry Lubensky
charter member of the Missouri Archaeological died in 1938, and Earl supported himself and his
Society (MAS; 1935-2009), and retired family by taking over his fathers wholesale beer
diplomat, was born in Marshall, Missouri, to business. He ran the business until 1942, when
Henry Carl Lubensky and Adele Beisemeier he was drafted into the United States army,
Lubensky. His father was of German and Polish having served in the Reserve and National
heritage and his mother of German descent. Earl Guard since 1937. Earl was on active duty
spent his early years in Marshall, graduating through 1945, first as an enlisted man in Alaska,
from Marshall High School in 1937 as a member then after officers training school, running a
quartermaster battalion, and later serving as a Affairs from George Washington University,
compound commander at a prisoner of war writing the thesis Constitutionalism versus
camp in Charleston, South Carolina. He Militarism: Ecuador, A Case Study.
returned to the Army Reserve after the war,
retiring as a lieutenant colonel in 1981. Earl Lubensky began conducting
archaeological research while he was in the
Earl married Anita Ruth Price of Saline Foreign Service, and was active in archaeology
County, Missouri, on June 27, 1942. His until his death. He led excavations in Ecuador
experiences in the army instilled in him a desire at the Ayalan cemetery site (1972-1973), at the
to help prevent future wars and help humanity, Anllulla shell mound (1973), and at two sites on
and after the war he decided to become a the La Florida Hacienda (1979). In El Salvador
diplomat. He and Anita returned to Marshall, he excavated two mounds at Cihuatn (1977-
where Earl attended Missouri Valley College, 1978). He also led excavations at the Menden-
graduating in 1948 with a B.A. in History and hall sites in Missouri. These were the MAS Field
Economics. He subsequently attended the exercises from 1983 until 1986 (Lubensky
School of Foreign Service at Georgetown publications 1983c, 1983d, 1983e, 1984a,
University in Washington, D.C., and, in 1949, 1984b, 1986). In addition, he participated in
earned a B.S. in Foreign Service (International excavations at other sites in Missouri, including
Commerce). Earl and Anita were in the Foreign the Utz site, the Historical School site in Arrow
Service from 1949 to 1978. Their first overseas Rock, the Old Monroe project, the Conley
posting was to Germany (1950-1952), then to House excavations on the University of Missouri
the Philippines (1952-1954), and Spain (1955- campus, and several other digs in Boone
1956). From 1956 until 1961 they were in County.
Washington D.C. Subsequently, Earl spent most
of the rest of his career in Latin America, posted After retiring from the Foreign Service, Earl
twice to Ecuador (1961-1966; 1971-1973), and and Anita moved to Columbia, Missouri, so that
to Colombia (1967-1970) and El Salvador Earl could attend graduate school in
(1976-1978). He specialized in Latin American anthropology. In his M.A. application, Earl said
political affairs and served as Consul General in his interest was the archaeology of the Western
Guayaquil, Ecuador and Charg dAffaires ad Hemisphere, and as his professors got to know
Interim in San Salvador, El Salvador. him, it became clear this was not an
understatement. Dr. Richard Diehl directed his
During their years of diplomatic service Earl 1983 M.A. thesis, An Oneota Pit Complex
and Anita raised three sons, Tom (Professor of (Lubensky 1983a, 1983b), in which Earl
Physics, University of Pennsylvania), Gerald returned to the Utz site, where his interest in
(Professor of Art, University of Kansas), and archaeology had begun. The Utz site and his
John Christopher (computer business, Quito, early experiences interacting with members of
Ecuador) and traveled extensively, always with the Missouri Archaeological Society were
archaeological sites on their itinerary. Earl also important to him. In his M.A. application Earl
continued his education, earning a diploma in spoke of the understanding of archaeology he
cost accounting and systematizing in 1949, gained from his MAS membership.
training that served him well as treasurer of the
Missouri Archaeological Society (1981-1991). I was a member of Earls M.A. committee,
In 1967 he earned a diploma from the National and after Richard Diehl left the University of
War College and an M.S. in International Missouri, Robert A. Benfer Jr. and I directed
3- Pearsall: Earl Henry Lubensky
Earls dissertation research. Directing Earl was and 2000 Andean Past research report, Earl
a process that involved nudging him in discussed his study of Ayalan cemetery artifacts
directions we thought useful, then running to and site stratigraphy, which revealed that the
keep up. It was an adventure, a rewarding and Late Integration Period burials were dug into an
enjoyable one, which culminated in his 1991 earlier, Guangala or Jambel (Regional
dissertation, The Ferdon Collections of Prehistoric Developmental Period, ca. 500 B.C.-A.D. 500)
Ceramic Vessels and Sherds from Esmeraldas midden, which was, in turn, underlain by a
Province, Ecuador. Earl corresponded for nearly possible Chorrera (Late Formative Period, c.
thirty years with Ed Ferdon, who documented 1500-500 B.C.) occupation. Earls study of the
numerous important archaeological sites in burial urns led him to conclude that while most
Ecuador through surface collections (see urns were of Milagro-Quevedo style and shape,
Ferdons obituary by Earl Lubensky in Andean two other types were present, and that this, with
Past 8). I know it gave Earl great satisfaction to the shape of compoteras, some decorative motifs,
bring Ferdons study to fruition. and the absence of mounds, suggested
affiliations to Manteo-Huancavilca
CONTRIBUTIONS TO ANDEAN ARCHAEOLOGY (Integration Period). Based on stratigraphic
position of the extended burials and earlier C-14
Earl Lubenskys interests in the prehistory of dates (Ubelaker 1981), as well as characteristics
Ecuador ranged from the Formative to of grave goods, Earl suggested that the extended
Integration Period, from the classification of burials may have been affiliated with Manteo-
Valdivia figurines, to sources of obsidian, to Huancavilca as well. In drawing these
relationships among polities in the Gulf of conclusions, Earl supported Estradas (1957)
Guayaquil and coastal Esmeraldas. His main argument that both Milagro-Quevedo and
contributions were made through excavations at Manteo-Huancavilca populations were present
three sites, the Ayalan cemetery, the Anllulla contemporaneously in the Estero Salado region,
shell mound, and Hacienda La Florida, and and suggested that the Ayalan Cemetery
notably, through his study of the Ferdon possibly lay on the border between the polities.
ceramic collection. Neutron activation analysis of five obsidian
blades from the cemetery identified Yanaurco-
Ayalan Cemetery. The Integration Period Quiscatola as the source (Lubensky 1999).
Ayalan Cemetery site (OGES-1) is one of at
least four prehistoric cemeteries located on the Anllulla Shell Mound. In 1973 Earl undertook
Hacienda Ayalan, on the west side of the test excavations at a shell mound on the
mangrove-fringed Estero Salada, 55 kilometers Hacienda Ayalan, which he designated the
southwest of Guayaquil. It was discovered and Anllulla site (OGES-2), working to determine if
test-excavated by Earl Lubensky in 1972. it was contemporaneous with the cemeteries. A
Douglas Ubelaker of the Smithsonian summary of the ceramics and C-14 dates was
Institution joined the project in 1973, and took published in his 2000 Andean Past research
over the excavation of urn and primary report; a more comprehensive discussion and
extended burials, which are reported in detailed description and illustration of ceramics
Ubelaker (1981) and classified as the Milagro was distributed as a manuscript to colleagues in
Phase (Integration Period, ca. A.D. 500-1500). 1995. The mound, composed mostly of the
oyster Ostrea iridescens, was located on the edge
In his 1974 Boletn de la Academia Nacional of a salt flat, some three kilometers from the
de Historia publication, 1982 SAA presentation,
ANDEAN PAST 10 (2012) -4
current Estero Salada shore. It has since been on a seasonal or occasional basis by peoples from
destroyed by shrimp pond construction. different hinterland areas, and that those
populations changed during the period
A four meter deep test excavation represented by the overlap, with populations
documented a sequence that began in the producing Machalilla ceramics replacing
Valdivia Period. The earliest decorated sherds, Valdivia ceramic producers.
recovered from 280-140 centimeters below the
surface (dated at 180-170 centimeters below the Hacienda La Florida. In 1979, Earl Lubensky
surface to 2012-1741 cal BC), fit diagnostically and Allison Paulsen test excavated two multiple
into Valdivia D/Valdivia VI-VII phase (six mound sites, Estero Cecilia and Santa Marta, on
wares, following Meggers et al. 1965 typology), the Hacienda La Florida, in western Pichincha
and were underlain by strata containing non- Province 16 kilometers outside Santo Domingo
diagnostic sherds (370-280 centimeters below de los Colorados at the confluence of the Ro
the surface), and aceramic deposits (400-370 Toachi and Ro Blanco. At least ten other sites,
centimeters below the surface). Charred maize mostly consisting of single mounds, are located
kernels were recovered at depths of 210 on the hacienda. Excavations and analyses were
centimeters and 220 centimeters below the reported at the 1980 and 1999 Midwest
surface in association with Valdivia ceramics, Conferences on Andean and Amazonian
and root/tuber material occurred at 220 Archaeology and Ethnohistory, in the Current
centimeters and below (Pearsall 2003). Research section of American Antiquity in 1981
Interestingly, above a significant ash layer, with Paulsen, and in Earls 2000 Andean Past
presumably volcanic, at 140 centimeters below research report.
the surface and continuing until 70 centimeters
below the surface (dated at 110-100 centimeters A test excavation in the principle mound of
belowthe surface to 1514-1410 cal BC), two the Santa Marta mound group yielded ceramics
styles of diagnostic Machalilla sherds occurred that strongly resemble Chorrera examples. A
with four Valdivia D/VI-VII wares. Above 70 single radiocarbon date of 1260-1001 cal BC is
centimeters below the surface, and a second ash consistent with this affiliation. Test excavations
fall, deposits date to the Regional in three mounds at the Estero Cecilia site yield-
Developmental-Integration Period transition. ed pottery generally characteristic of Jama-
Coaque I and II. Three radiocarbon dates from
The excavations in the main cut in the the excavations yielded calibrated dates of A.D.
center of the mound (Cut A, 2 by 2 meters) 1023-1207, A.D. 1216-1377, and A.D. 1301-
were carried out in ten centimeters arbitrary 1408, corresponding to Integration Period Jama-
levels. However, the natural strata lay Coaque II.
practically horizontally in this cut, on the basis
of which Earl argued that materials in each level Obsidian artifacts and debitage were
of Cut A were roughly contemporaneous. Thus common at both sites. At the 1999 Midwest
while acknowledging the problems of shell Conference on Andean and Amazonian
mound stratigraphy and perturbation, he Archaeology and Ethnohistory, Earl reported
proposed that the mix of Late Valdivia and that obsidian blades from Santa Marta were pro-
Machalilla sherds from 140-70 centimeters venienced to the Mullumica and Yanaurco-
below surface was a cultural phenomenon that Quiscatola obsidian sources (as reported in
required explanation. In his 1995 manuscript, Villalba 1988), but affiliated to Chorrera (not,
Earl hypothesized that the shell mound was used as previously reported, to the Regional
5- Pearsall: Earl Henry Lubensky
Developmental Period). Earl studied the nine provinces during the 1940s (see Obituary
cutting edge/mass ratio of 29 blades from Estero by Earl Lubensky, Andean Past 8). Earl decided
Cecilia, affiliated to Integration Period Jama- to study this collection for his dissertation, and
Coaque II, not sourced, but presumed to come obtained the loan of the materials from the
also from Mullumica and/or Yanaurco- Museum of New Mexico in Santa Fe (the
Quiscatola. With a CE/M ratio of 3.42 collection is now curated at the University of
(standard deviation 1.74), values for Estero New Mexico in Albuquerque). The dissertation,
Cecilia fell between those for the El Inga (CE/M completed in 1991, focused on collections from
1.75, standard deviation 1.00) and Ayalan coastal Esmeraldas Province. Working from
cemetery (CE/M 10.54, standard deviation 6.71) Ferdons unpublished site forms and maps, Earl
sites, also studied, showing a clear relationship located and described thirty-five sites in six
between distance from source (Eastern geographic clusters, and studied the ceramic
Cordillera, c. 30-40 kilometers east of Quito) collections from sixteen sites. Ferdon collected
and cutting edge/mass ratio. A provenance representative ceramics from site surfaces and
study of clay from Estero Cecilia pottery, natural or man-made cuts, including the
conducted at the University of Missouri extensive gold-mining operations at La Tolita.
Research Reactor, indicated that some ceramics The dissertation includes topographic maps
were non-local in origin. showing site locations, reproduces Ferdons
excellent site maps, and illustrates most of the
The abundance of obsidian, presence of sherds studied.
non-local pottery, and construction of mound
complexes during both the Late Formative and Earls thesis was that one can understand
Integration Periods (and perhaps the Regional the relationship among sites by analysis of
Development Period), led Earl Lubensky and ceramic attributes. Earl developed a ceramic
Alison Paulsen to speculate that Hacienda La seriation and estimated the chronology of the
Florida was a major exchange center, possibly sites. In the study he took an attribute ap-
with ceremonial functions. Earl further proach, rather than the more common
suggested a relationship between the typological approach, in order to expand the
introduction of ceramics in the sierra and the possible features available for ceramic
beginning of obsidian use during the coastal classification. He did not attempt to fit the
Chorrera Period. He hoped to map the La Esmeraldas ceramic collection into existing
Florida mound complexes and conduct a typologies, in part for the practical reason that
detailed study of 1979 excavated materials, to the collection contained wares that were not yet
facilitate comparisons between La Florida and typed, and in part to explore the power of an
other mound complexes on the northern attribute approach.
Ecuadorian coastal plains and western Andes,
but that project was unfortunately not funded. The dissertation explains fully the process of
His 1993 research proposal (on file, University attribute analysis of ceramics, which I will not
of Missouri Department of Anthropology) notes discuss in detail here. Four major groups of
the presence of tephra deposits at the tested attribute dimensions were developed (shape,
sites, which were earlier misinterpreted as living size, manufacture, decoration), and 144
floors. attribute dimensions used. Earl developed a
nested matrix for data entry using the computer
Ferdon Collection. Edward Ferdon conducted program PARADOX. He focused on decorative
an extensive surface collection of 118 sites in technique and vessel shape for the quantitative
ANDEAN PAST 10 (2012) -6
In terms of chronology, Earl concluded that Earl Lubenskys diverse legacy of scholarship
two sites were probably Late Formative. One and service is significant for our discipline as a
site cluster was likely transitional from Late whole, as well as for Andean archaeology in
Formative to Early Regional Developmental, particular. He was an active member of both the
one cluster was Regional Developmental and Missouri Archaeological Society and the Society
strongly related to La Tolita, one cluster was for American Archaeology and in 1989 was
Integration Period, and many sites were multi- asked by the SAA to chair a task force on
component. The study documented that coastal establishing an association of archaeological
Esmeraldas was occupied at least since the Late societies. The Council of Affiliated Societies was
Formative, and that many sites were occupied activated at the 1990 meeting in Las Vegas,
for extended periods. The predominant site or with the goals of serving as a bond among
central place shifted over time, for example, individuals interested in American archaeology,
from the Esmeraldas River delta in the Late whether professional or nonprofessional, and to
Formative, to La Tolita in the Regional foster the formation and welfare of regional and
Developmental Period, to Atacames in the local societies. At the 1991 annual meeting in
Integration Period. Based on ceramic New Orleans, Earl was given the SAA
similarities, Earl concluded that contact was Presidential Recognition Award by Jeremy
intense among sites clustered together Sabloff for his service to the SAA and the
geographically, but close interactions discipline. Earl was a vocal advocate for the role
characterized the entire Esmeraldas coast. of amateurs and avocationals in archaeology,
Methodologically, attribute analysis proved to saying . . . we cannot leave archaeology alone
be a highly flexible and orderly way to classify to paid professionals, but our amateurs and
ceramics, but the system was complex to set up avocationals, especially the younger generation,
and required many subjective judgments in the must be better prepared through education and
description and delineation of attributes. The experience to work at the highest level of
study also demonstrates the value of museum professionalism among all involved (MAS
collections. Quarterly, Oct.-Dec., 1993:6). He exhorted
collectors to plan for the donation of their
After his dissertation was finished, Earl collections to museums.
completed the study of the entire Ferdon surface
collection, presenting his findings in a series of After completing his doctoral degree at the
conference papers (1992 Midwest Conference University of Missouri, Earl was granted adjunct
on Andean and Amazonian Archaeology and research associate status in the Department of
Ethnohistory, 1992 Northeast Conference on Anthropology. The department has a course
Andean Archaeology and Ethnohistory, 1993 called Research Skills in which undergraduate
Annual Meeting of the Institute of Andean students work individually with a mentor to
Studies, 1994 Midwest Conference on Andean learn what research is all about. For many years
and Amazonian Archaeology and Ethnohistory). Earl volunteered his time for this class, and
mentored several students interested in
archaeology. He taught them how to study
artifacts and write up results. More than that, he
gave generously of his time and life experiences
7- Pearsall: Earl Henry Lubensky
and enriched the student experiences at the Smithsonian Contributions to Anthropology 29.
University of Missouri. Washington, D.C.: Smithsonian Institution
Press.
Villalba, Marcelo
In recent years my intractions with Earl 1988 Cotocollao: Una aldea formativa del valle de
included gardening, as well as discussions about Quito. Miscelanea Antropolgica Ecuatoriana,
the archaeological projects he wanted to finish. Serie Monogrfia 2. Quito: Museo del Banco
Earl loved his gardens, and kept meticulous Central del Ecuador.
records about seeds he collected, sprouted, and WORKS OF EARL H. LUBENSKY1
planted. I volunteered to help him with the
basement (his study and research files), but Papers
somehow we always ended up working in the
Earl H. Lubenskys papers are at The State Historical
garden. Im glad thats how we spent our time.
Society of Missouri, Columbia, Missouri in the Western
Earl left detailed, well-organized records of his Historical Manuscript Collection.
research, now transferred to the State Historical
Society of Missouri. Among his other passions Articles, thesis, dissertation
were genealogy, ham radio, and collecting
1972 Two Polish Boys from Texas. Stirpes: Texas State
stamps and coins. Earl also had a deep desire to
Genealogical Society Quarterly, June, pp. 62-64.
understand the universe. Among the last books 1974 Los Cementerios de Anllulla: Informe preliminar
he read were ones on string theory and cosmol- sobre una excavacin arqueolgica. Boletn de la
ogy. Academia Nacional de Historia. 57(123):16-23,
plus 18 figures.
1979 Off the Job: You, too, Can e an Archaeologist,
Anita, Earls wife of over 50 years, died in the Man Says; Posting Abroad Puts You in a
1993. In 1994 he married Margot Patterson who Uniquely Advantageous Position. Department of
preceded him in death in 2008. Later that year State Newsletter 208:7-10.
he married Marion Reed. In addition to his wife 1980 Prehistoric Ceramics of Coastal Ecuador, 3 pp.
Marion and his three sons, Earl is survived by Exhibit brochure, Museum of Anthropology,
University of Missouri, Columbia.
six grandchildren. 1981a Flotation Device. Missouri Archaeological Society
Newsletter 352:3.
REFERENCES CITED 1981b Letter. 11 April 1980 to Clement W. Meighan.
In Obsidian Dates III: A Compendium of the
Estrada, Emilio Obsidian Hydration Determinations Made at the
1957 Los Huancavilcas: ltimas civilizaciones pre- UCLA Obsidian Hydration Laboratory, edited by
histricas de la costa del Guayas. Publicacin del Clement W. Meighan and Glenn S. Russell, pp.
Museo Vctor Emilio Estrada 3. 149-150. Monograph 16, Institute of
Meggers, Betty Jean, Clifford Evans, and Emilio Estrada Archaeology, University of California at Los
1965 Early Formative Period of Coastal Ecuador: The Angeles.
Valdivia and Machalilla Phases. Smithsonian 1982a Evolution of a Collector. Central States
Contributions to Knowledge 1. Washington, D.C.: Archaeological Journal 29:149-153.
Smithsonian Institution. 1982b The Fox Head Button. Missouri Archaeological
Pearsall, Deborah M. Society Newsletter 361-362:2.
2003 Plant Food Resources of the Ecuadorian Forma-
tive. In Archaeology of Formative Ecuador: A
Symposium at Dumbarton Oaks, 7 and 8 October
1995, edited by J. Scott Raymond and Richard L.
Burger, pp. 213-257. Washington, D.C.: Dum-
1
barton Oaks Research Library and Collection. Non-research chapter and membership notes published
Ubelaker, Douglas in the Missouri Archaeological Society Newsletter/Quarterly
1981 The Ayalan Cemetery: A Late Integration Period are omitted.
Burial Site on the South Coast of Ecuador.
ANDEAN PAST 10 (2012) -8
1983a An Oneota Pit Complex. M.A. Thesis, Current Research Report. Andean Past 6:372-
Department of Anthropology, University of 378.
Missouri, Columbia. 2000b Internal Security and Human RightsMilitarism
1983b An Oneota Pit Complex. The Missouri and Diplomacy. Peace Talk 7(1):41-48
Archaeologist 44:75-108. (Published by Friends of Peace Studies,
1983c 1983 MAS Field Exercise: Part I, Introduction. University of Missouri, Columbia).
Missouri Archaeological Society Newsletter 371- 2001 The Gordon Tract Archaeological Site. Country
372:7-13 (Mendenhall site, Late Woodland, Lives, November/December (published by the
23BO86). Boone County Historical Society/Walters Boone
1983d MAS Field Exercise: Part II. Missouri County Historical Museum, Columbia,
Archaeological Society Newsletter 373-374:6-9. Missouri), 2 pages.
1983e MAS Field Exercise: Part III. Missouri 2005 The Excavation of Structures P-12 and P-20 at
Archaeological Society Newsletter 375-376:3-5. Cihuatn, El Salvador/Excavacin de las Estruc-
1984a 1983 MAS field exercise: Part IV. Missouri turas P-12 y P-20 de Cihuatn, El Salvador,
Archaeological Society Quarterly 1(1):6-7. edited and with an introduction by Karen Olsen
1984b 1984 MAS Field Exercise. Missouri Archaeological Bruhns. Treganza Anthropology Museum Papers
Society Quarterly 1(3):6, 22 (Mendenhall site, 22 (San Francisco State University).
Late Woodland 23BO86). 2007 Edwin Nelson Ferdon, Jr. (1913-2002). Andean
1984c Archaeology in Missouris Public Schools. Past 8:9-32.
Missouri Archaeological Society Quarterly 1(3):15. Paulsen, Alison C. and Earl H. Lubensky
1985a Osage Petroglyphs from Vernon County. 1981 Current Research: Andean South America,
Missouri Archaeological Society Quarterly 2(3):12- Ecuador. American Antiquity 46(1):201.
13.
1985b Society Field Exercises. Missouri Archaeological Presentations
Society Quarterly 3(1):1, 4 (Mendenhall site, Late
Woodland 23BO86). 1980a Archaeology in Ecuador. Paper presented to the
1986 Missouri Archaeological Society Field Exercise Mound City Archaeological Society, Clayton,
1986. Missouri Archaeological Society Quarterly Missouri, June 3.
3(4):4-5, 15-16 (Mendenhall site mound, 1980b Exploratory Excavation at La Florida, Ecuador.
23BO87). Paper presented at the 8th Midwest Conference
1989 The Westermans of Westphalia, Missouri: Their on Andean and Amazonian Archaeology and
Ancestors and Descendants. Columbia, Missouri: Ethnohistory, Field Museum, Chicago, Illinois
Published by the author, 319 pp. (abstract and photocopies of slides in Pearsalls
1990a Avocational Archaeology. Society for American possession).
Archaeology Bulletin 8(2):17. 1980c Shell Mound Excavation at Hacienda Ayalan,
1990b Plants in Archaeology Workshop [1989]. Ecuador. Abstract of paper proposed for
Missouri Archaeological Society Quarterly 7(2):22. presentation at the 8th Midwest Conference on
1991a The Ferdon Collections of Prehistoric Ceramic Andean and Amazonian Archaeology and
Vessels and Sherds from Esmeraldas Province, Ethnohistory, Field Museum, Chicago (in Pear-
Ecuador. Ph.D. dissertation, Department of salls possession).
Anthropology, University of Missouri, Columbia. 1981a The Ayalan cemetery: An Integration Period
1991b Valdivia Figurines. In The New World Figurine Site in the Guayas Basin Area of Ecuador. Paper
Project, Volume 1, edited by Terry Stocker, pp. presented at the 9th Midwest Conference on
21-25. Provo, Utah: Research Press at Brigham Andean and Amazonian Archaeology and
Young University. Ethnohistory, Columbia, Missouri, February 28.
1991c Organization of the Council of Affiliated 1981b Historical School House Site in Arrow Rock,
Societies. Society for American Archaeology Missouri. Paper presented at the 46th Annual
Bulletin 9(1):13-14. Meeting of the Society for American
1993 Update on the Council of Affiliated Societies of Archaeology, San Diego, California, May 1. Also
the Society for American Archaeology. Missouri presented at the Annual Spring Meeting of the
Archaeological Society Quarterly 10(4):5-7. Missouri Archaeological Society, Springfield,
1999 Petroglyphs at the La Cueva del Espiritu Santo, Missouri, May 23, 1981.
El Salvador. News & Notes. Mexicon 21:71-72. 1982a Excavation of Three Prehistoric Ceramic Period
2000a Hacienda La Florida, Ayalan Cemetery, Anllulla Sites in Coastal Ecuador. Paper presented at the
Shell Mound, and Ferdons Surface Collections. Annual Fall Symposium of the Missouri
9- Pearsall: Earl Henry Lubensky
CONSTANZA CERUTI
Consejo Nacional de Investigaciones Cientficas
and Universidad Catlica de Salta (Argentina)
Hans Schobinger at the III Simposio de Arte Rupestre Peruano, Huaraz, Peru in 2008.
Photograph courtesy of Susana Barrau
It is very difficult to find the words to call to follow in his footsteps today owe him much
mind the importance of the legacy Juan Scho- more than bibliographic citations and formal
binger left archaeology, with the gratitude and tributes can indicate.
affection he merited. It is even more difficult to
try to summarize in a few pages the infinite Juan Santiago Ren Schobinger was born in
contributions he made to the development of Lucerne, Switzerland and died in Mendoza,
this discipline in Argentina. He was a re- Argentina, a city in which he resided for most of
searcher, writer, professor, holder of a doctorate, his life. He became a naturalized Argentinian
archaeologist, student of religions, but, above citizen in 1954 and was proud of his adoptive
all, a respectful colleague and a generous nationality, stressing that he had arrived in
teacher. He entered fields as varied as rock art Argentina while still a baby.
studies and high altitude archaeology, and was
a genuine pioneer in them. Those of us who
art of Argentina), written with Carlos Gradin, National University of Cuyo until 1993. He was
and published in Madrid and Milan in 1985, is the founder and director of the universitys
another. A third is Arte prehistrico de Amrica archaeology museum. Among its collections are
(Prehistoric Art of America), published in 1997 the materials Schobinger recovered high on the
in both Spanish and Italian language editions. slopes of Mt. Aconcagua. He was also editor of
In the area of mountain archaeology, his the professional journal of the Institute, the
edited volume, La momia del cerro El Toro Anales de Arqueologa y Etnologa from 1956 to
(The Mummy of El Toro Mountain) is 1993, a venue which helped to make known,
outstanding. This was first published in 1966 with admirable continuity, advances in the
by the National University of Cuyo. Scho- archaeology of western Argentina as well as
binger was also, with Clara Abal de Russo, those in high altitude archaeology.
Mnica Ampuero, and Graciela N. Arenas, an
author-editor of El santuario incaico del cerro Schobingers archaeological surveys centered
Aconcagua (The Inca Sanctuary of Mt. Acon- on western and northwestern Argentina, on the
cagua, 2001) and the motivating force behind high Andean cordillera, and on northwestern
El santuario incaico del nevado de Chuscha (The Patagonia. He made substantial contributions to
Inca Sanctuary of the Snow-capped Chuscha various aspects of research, mainly in relation to
Mountain, 2001-2003 [2004]), both published the preceramic period, to rock art, and to high
by the National University of Cuyos press, altitude Inca shrines, although he also
EDIUNC. These disseminate the results of the maintained a particular interest in shamanism
interdisciplinary studies on mummified bodies and in biblical archaeology. The wide spectrum
from high altitude sanctuaries that Schobinger of his contributions and the transcendent nature
coordinated. of many of them never cease to amaze in an era
of growing specialization in the social sciences.
Schobinger belonged to many scientific For a long time Schobingers work ran counter
associations. From 1969 to 1978 he was on the to major currents, being a humanistic view of
Advisory Commission of CONICET, the mankind and prehistory, but recent trends in
Argentine governments scientific grant-giving archaeological and anthropological thought are
body. He was a member of the International beginning to return us to his point of view.
Union of Pre- and Protohistorical Sciences,
and of the Rock Art Committee of ICOMOS, It is well known that the Cerro El Toro
the International Council of Monuments and mummy and the Aconcagua child were
Sites, from its foundation in 1982. He was a recovered thanks to Schobingers pioneering
member of its board of directors from 1984 activities in rescue archaeology during mountain
until 1994. In recent years he frequently expeditions made in 1964 and 1985 respectively.
served as a judge in archaeology teaching The work of this researcher involved ascent to
competitions held by the University of Buenos heights above 5000 and 6000 meters to recover
Aires and by the National University of Tucu- and bring to safety the mummies and grave
mn. He was also a member of doctoral thesis goods that were accidentally discovered by
committees at the University of Buenos Aires mountain climbers. For decades he sustained a
and at the National University of Cuyo. commitment to coordinate interdisciplinary
studies of these frozen bodies and their offerings,
Juan Schobinger was Director of the the publication of the scientific results, and
Institute of Archaeology and Ethnology eventual diffusion of information to the
(Instituto de Arqueologa y Etnologa) of the community via museum exhibitions of some of
ANDEAN PAST 10 (2012) - 16
the finds. These efforts are outstanding not mountain climbing, as in the cases of Antonio
only because of their pioneering nature in Beorchia Nigris and Evelio Echeverria, or
Argentina, but also for having been through comparative studies of religions, as with
accomplished with very sparse resources, yet Juan Adolfo Vzquez.2 Among the outstanding
with the considerable personal efforts of the aspects of Schobingers legacy was his
participating researchers. In addition, in 1995, equanimity and and his appreciation of the
Schobinger devoted a publication specifically problems and difficulties that so frequently color
to the history of the mummy of a little girl interactions among archaeologists, as well as his
found on Chuscha (Schobinger 1993-94), a unshakable predisposition to defend the
snow-capped mountain in northeastern profession against attacks suffered out of
Argentina, that was removed from the ignorance or motivated by special interests.
mountain by treasure hunters in the early
twentieth century. Likewise, in 2004 he edited We who had the welcome privilege of
the interdisciplinary studies that had been sharing a field or research trip with Schobinger
centered on this mummy, cooperating with know his great love of the land and his romantic
the return of the queen of the mountain to vision of nature and the archaeological record.
Salta province. He was capable of passing long minutes
contemplating a potsherd. He would celebrate
Beginning in 2000 Schobinger was an with a youthful enthusiasm (in spite of his
honorary advisor to the Institute of High seventy-odd years) his arrival at some remote
Altitude Studies (Instituto de Investigaciones Inca tambo as if it was the first in his life, even
de Alta Montaa) of the Catholic University though he had discovered and spent time at
of Salta (UCASAL). He was also an honorary dozens throughout his career. He was enchanted
member of the scientific committee for the by rock art, being the eternal lover of symbolism
preservation of the frozen mummies that as manifested by material evidence. Those of us
Johan Reinhard and I discovered and who accompanied him on mountain expeditions
recovered at the top of the Llullaillaco
volcano. In September 2006, I had the
2
privilege of teaching a course on mummy Juan Adolfo Vzquez (b. 1917 in La Plata, Argentina)
studies with Juan Schobinger under the is a Professor Emeritus of the University of Pittsburgh who
auspices of UCASAL. This center of specializes in the anthropologically and philosophically
informed study of the thought and literature of Amer-
advanced studies recognized his academic and
indians and the nature of the sacred. He edited the
teaching careers in designating him a Special journals Notas y Estudios de Filosofa (Tucumn, Argentina
Visiting Professor (Profesor Extraordinario (1949-1954); Revista del Instituto de Filosofa (Crdoba,
Visitante) by decree of its rector (resolucin Argentina 1958); Boletn de Historia de las Religiones
rectoral 912/06). (Mendoza, Argentina 1960-1966; Latin American Indian
Literatures (Pittsburgh, Pennsylvania 1977-1984); NAOS
(Pittsburgh and Bahia Blanca, Argentina 1984-1996); and
Those of us who follow Schobinger in high Textos y Contextos. An Interview of Vzquez by Mary H.
mountain archaeology know how to value and Preuss was published in Revista Iberoamericana (number
admire the pioneering character of his work. 127, 1984, pp. 571-583) and one by Graciela N.V. Cor-
Among the small family of high altitude valn is in Humanismo Siglo XX (edited by Schobinger,
1995, pp. 17-28). An essay by Norma Rusconi, Hu-
archaeologists Hans had followers and friends, manismo y educacin; Labor universitaria de Juan Adolfo
but no detractors. His contributions also were Vzquez en San Juan (1964-1966) in Humanismo Siglo XX
appreciated by those who became interested (pp. 31-39) elucidates a brief, but important period of his
in the sacred peaks through the sport of life. See also Schobinger (2004-2005) for a brief biography
of Juan Adolfo Vsquez.
17 - Ceruti: Juan Schobinger
Schobinger was a person who honored his Grabert, Hellmut and Juan [Santiago Ren] Schobinger
1971 Petroglifos a orillas del ro Madeira (N.O. Brasil).
elders. I remember that on one of my first
Anales de Arqueologa y Etnologa 24-25:93-112
visits to Mendoza, while I was preparing my (Instituto de Arqueologa y Etnologa, Facultad
masters thesis, he insisted, with the de Filosofa y Letras de la Universidad Nacional
characteristic courtesy with which he normally de Cuyo, Mendoza, Argentina).
hosted his colleagues, in driving me in his car Hyslop, John and Juan [Santiago Ren] Schobinger
1990 Establecimiento incaico en los nevados de
to the city of San Juan, so that I could get to
Aconquija, prov. de Tucumn, Argentina.
know the mummy of the El Toro mountain Gaceta Arqueolgica Andina 17:67-75.
and the archaeological museum that housed Schobinger, Juan [Santiago Ren]
it. We were accompanied by Erico Groch, the 1951 La Universidad en su historia. 45 pages. Buenos
mountain climber who, along with Antonio Aires: No publisher given.
1952 El sentido geogrfico en las civilizaciones proto-
Beorchia Nigris, discovered the famous histricas. Centro (Revista del Centro de Estu-
mummy of San Juan. Schobinger managed, dios de Filosofa y Letras, Buenos Aires) 3:15-19.
whenever the opportunity permitted, to 1953 Altamerikanische Steinkunst in Neuquen.
overcome the limitations of mobility which Sudamerika 4(1):49-51.
age and fragile health imposed on Don Erico, 1954a Arqueologa del territorio del Neuqun. Doctoral
dissertation, Universidad de Buenos Aires.
so that the old man could visit the lad, as 1954b Arqueologa del territorio del Neuqun.
well as be the guest of honor at an opulent Neuquenia 5(24):7-8.
luncheon after leaving the museum. I believe 1954-55 Ergebnisse der argentinischen urgeschichts-
that being more than ninety years of age, this forschung in den letzten zwei Jahrzehnten.
was one of the last opportunities that Don
Erico had to see the mummy, with all that it
signified for him. 3
This bibliography is based on one published by the late
Humberto Lagiglia (2005). It has been augmented by
In the Andean world, when the elders Constanza Ceruti and Monica Barnes with the assistance
depart forever, in reality they become a part of of Horacio Chiavazza, Victor Durn, David Fleming, and
Matthias Strecker. Juan Schobingers memory will be
the highest mountain peaks that are the honored at the international congress Archaeology and
protectors of the communities at their feet. Rock Art: 25 Years SIARB to be held June 25-29, 2012
The next time I raise my eyes to Aconcagua, and hosted by the Sociedad de Investigacin del Arte
to el Mercedario, to Negro Overo, to Rupestre de Bolivia (SIARB), the Programa de
Aconquija, I wont doubt that these Investigacin Estratgica en Bolivia (PIAB), the Museo
Nacional de Etnografa y Folklore (MUSEF), the Instituto
mountains, who knew how to accept the Francs de Investigacin para el Desarrollo (IRD-Bolivia),
footsteps of a dedicated archaeologist in life, and the International Federation of Rock Art
will also accept the immortal spirit of this Organizations (IFRAO), the Espacio Simn I. Patio, the
great teacher and expert. Brazilian Embassy to Bolivia, and the United States
Embassy to Bolivia.
ANDEAN PAST 10 (2012) - 18
1960a Un siglo del nacimiento de la antropologa 1962-3a Nuevos petroglifos de la provincia de Neuqun
prehistrica. Ciencia e Investigacin: Revista Anales de Arqueologa y Etnologa, 27-28:151-171
patrocinada por la Asociacin Argentina para el (Instituto de Arqueologa y Etnologa, Facultad
progreso de las ciencias 16(12):447-456. de Filosofa y Letras de la Universidad Nacional
1960b El Oreopithecus y el problema del origen del de Cuyo, Mendoza, Argentina).
hombre. Sociedad Amigos de la arqueologa, 1962-3b Un notable cntaro de la zona cordillerana del
publicacin 5 (37 pages; Mendoza, Argentina). Neuqun. Anales de Arqueologa y Etnologa 17-
Reprinted as Chapter II in Schobinger 1982b. 18:173-178 (Instituto de Arqueologa y
1961a Otra vez el hombre fsil de la Argentina: Etnologa, Facultad de Filosofa y Letras de la
Reflexiones sobre viejos problemas de la pre- Universidad Nacional de Cuyo, Mendoza,
historia pampeana. Anales de Arqueologa y Argentina).
Etnologa 16:61-102 (Instituto de Arqueologa 1962-3c El 35 Congreso Internacional de Americanistas
y Etnologa, Facultad de Filosofa y Letras de la (Informe): 19 al 25 de agosto de 1962. Anales de
Universidad Nacional de Cuyo. Mendoza, Arqueologa y Etnologa 17-18:203-207 (Instituto
Argentina). de Arqueologa y Etnologa, Facultad de Filosofa
1961b La arqueologa y el libro Mormn. Anales de y Letras de la Universidad Nacional de Cuyo
Arqueologa y Etnologa 16:259-265 (Instituto Mendoza, Argentina).
de Arqueologa y Etnologa, Facultad de 1963 Movimientos tnicos y culturales de Chile;
Filosofa y Letras de la Universidad Nacional Mendoza y Neuqun: Sus reflejos arqueolgicos.
de Cuyo. Mendoza, Argentina). Primer Congreso del rea Araucana Argentina
1961c Ameghino como impulsor de la ciencia pre- 2:225-232 (Neuqun, Argentina).
histrica. Anales de Arqueologa y Etnologa 1964a Investigaciones arqueolgicas en la provincia de
16:267-272 (Instituto de Arqueologa y San Juan Repblica Argentina (Informe
Etnologa, Facultad de Filosofa y Letras de la preliminar). Actas del XXXV Congreso
Universidad Nacional de Cuyo, Mendoza, Internacional de Americanistas (1962), Volume 1,
Argentina). pp. 615-619, plus 4 plates (Mxico).
1961d Estudios sobre el presapiens. Revista del 1964b Descubrimiento de una momia del perodo
Instituto de Antropologa 1(1-2):237-247 incaico en la cumbre del cerro El Toro (6300
(Universidad Nacional del Litoral, Rosario, m.), provincia de San Juan: Informe preliminar.
Argentina). Instituto de Arqueologa y Etnologa and
1961e A cien aos del nacimiento de la antropologa Sociedad Amigos de la Arqueologa, Publicacin
prehistrica. Revista de la Universidad de Buenos 7, Mendoza, Argentina (14 pages plus 12 plates).
Aires (5 epoca) 5(1):82-113. Also published as Informe preliminar sobre el
1961-3a Eine Pfahlbausammlung von Neuenburger See descubrimiento de una momia incaica en el cerro
in der Universitt Erlangen. Acta Praehistrica del Toro (6.300 m., provincia de San Juan, Rep.
5-7:148-161 (Buenos Aires: Centro Argentino (Argentina). In Homenaje a Fernando Mrquez-
de Estudios Prehistricos). Miranda: Arquelogo e historiador de Amrica;
1961-3b La poca de los palfitos en Europa Central ofrenda de sus amigos y admiradores, pp. 360-369.
segn investigaciones recientes. Acta Prae- Madrid: Universidad Complutense de Madrid,
histrica 5-7:162-175 (Buenos Aires). Seminario de Estudios Americanistas (1964).
1961-3c El anlisis de sedimentos: Una tcnica 1964c Discovery of an Indian Body on Cerro El Toro,
moderna al servicio de la datacin del Southern Andes (Argentine Republic). Current
Paleoltico Alpino. Acta Praehistrica 5- Anthropology 5(5):419.
7:223-239 (Buenos Aires). 1966a Nota sobre los petroglifos de Talampaya
1962a Investigaciones arqueolgicas en las tierras de (Provincia de La Rioja). Antiquitas 2:1-4 (Buenos
San Juan. Los Andes 9(7) (Mendoza, Aires). Reprinted as Chapter IV in Schobinger
Argentina). 1982b.
1962b Representaciones de mscaras en los 1966b Aportacin de las ciencias antropolgicas para
petroglifos del occidente argentino. Anthropos un nuevo humanismo. Actas de las Segundas
57: 683-699. Reprinted as Chapter III in Jornadas Universitarias de Humanidades, pp. 361-
Schobinger 1982b. 366 (Facultad de Filosofa y Letras, Universidad
1962c Consideraciones terminolgicas acerca del Nacional de Cuyo, Mendoza, Argentina).
precermico en Sudamerica y sus formas cul- 1966c Investigaciones arqueolgicas en la Sierra de
turales. Ampurias 24:165-168. Famatina (Prov. de La Rioja). Anales de
Arqueologa y Etnologa 21:139-196 (Instituto de
ANDEAN PAST 10 (2012) - 20
1980b Arte Rupestre de la regin cuyana occidental. parte, 1977). Revista del Museo de Historia Natural
Actas de las Primeras Jornadas de Arte Rupestre de San Rafael 4:129-154 (San Rafael, Argentina).
de San Luis (1978), pp. 33-41. 1983 Los comienzos del arte rupestre en el norte y
1980c Rock Art in Western Argentina: The Andean oeste de la Argentina. Homenaje a Martn
Region of Cuyo. Latin American Indian Almagro Basch, Volume 4, pp. 297-308. Madrid:
Literatures 4(1):64-69. Ed. Ministerio de Cultura.
1980d Lart rupestre de la rgion centre-occidentale 1983-5 Algunas observaciones terminolgicas sobre la
de lArgentine. Centenaire de lEnseignement de prehistoria americana. Anales de Arqueologa y
la Prhistore a Toulouse: Hommange au Etnologa 38-40:7-28 (Instituto de Arqueologa y
Professeur L. R. Nougier. Travaux de lnstitut Etnologa Facultad de Filosofa y Letras de la
dArt Prehistorique 23:441-450. Toulouse, Universidad Nacional de Cuyo, Mendoza,
France. Argentina).
1980e Los aportes del Dr. Osvaldo Menghn a la 1984a La Patagonia en el marco de la ms antigua
prehistoria del Viejo Mundo y a la teora prehistria Americana. In Culturas indgenas de la
prehistrica general. Sapiens 4:13-18 Patagnia, edited by J. Roberto Brcenas, pp 145-
(Chivilcoy, Provincia de Buenos Aires). 160. Biblioteca del Quinto Centenario. Madrid:
1980f Vikingos en Sudamerica? Ojeada sobre una Ediciones Cultura Hispnica.
reciente teora hiperdifusionista por va 1984b Pre-Columbian Art of Argentina (Review of Arte
transatlntica. In La Antropologa Americanista Precolombino de la Argetina: Introduccin a su
en la Actualidad: Homenaje a Raphael Girard, estudio cultural by Alberto Rex Gonzlez). Latin
Volume 2, pp.357-397. Mxico: Editores American Indian Literatures 8(2):99-103.
Mexicanos Unidos. 1985a Relacin entre los petroglifos del oeste de la
1981-2a Sobre las creaciones plsticas de un area Argentina y los de Chile. Estudios de Arte
marginal de la prehistoria: El arte Rupestre: Primeras jornadas de arte y arqueologa;
precolombino de la Argentina. Anales de El arte repestre en Chile, Santiago 16 al 19 de
Arqueologa y Etnologa 36-37:155-160 agosto de 1983, edited by Carlos Aldunte del
(Instituto de Arqueologa y Etnologa, Solar, Jos Berenguer R. and Victoria Castro R.,
Facultad de Filosofa y Letras de la pp. 195-203. Santiago: Museo Chileno de Arte
Universidad Nacional de Cuyo, Mendoza, Precolombino.
Argentina). 1985b Importante hallazgo arqueolgico en el Cerro
1981-2b Vikingos en Sudamrica? Una enojosa Aconcagua. Revista de Arqueologa 52:52-53
polmi- ca. Anales de Arqueologa y Etnologa (Madrid).
36-37:177-181 (Instituto de Arqueologa y 1986 La red de santuarios de alta montaa en el
Etnologa, Facultad de Filosofa y Letras de la Contisuyo y el Collasuyo: Evaluacin general,
Universidad Nacional de Cuyo, Mendoza, problemas interpretativos. In El imperio inka:
Argentina. Actualizacin y perspectiva por registros
1982a Estudios de arqueologa sudamericana: Arte arqueolgicos y etnohistricos, pp. 297-317. Special
rupestre y santuarios incaicos en el oeste de la volume, 45th International Congress of
Argentina. San Antonio de Padua, Provincia Americanists. Crdoba, Argentina, 1985.
de Buenos Aires: Editorial Castaeda. Bogot.
1982b Vikingos o extraterrestres? Estudio crtico de 1988a Prehistoria de Sudamrica: Culturas precermicas.
algunas teoras recientes sobre el origen y Amrica series 15. Madrid: Editorial Alianza.
desarrollo de las culturas precolombinas. 1988b Panorama esquemtico de la prehistoria
Coleccin Temas Bsicos. Buenos Aires: Ed. sudamericana. Comechingonia 6:7-18.
Crea-Huemul. 1988c Una nuova mummia di alta montagna.
1982c Los petroglifos del Cerro Tunduqueral, Lumana avventura, autumn issue pp. 70-83.
Uspallata, provincia de Mendoza, Repblica 1988d EL arte rupestre del rea subandina: Casos
Argentina. Ars Praehistorica: Anuario nterpretables como expresin de vivencias
Internacional de Arte Prehistorica. Edited by shamnicas. Contribuciones al estudio del arte
Eduardo Ripoll-Perell. Volume 1, pp. 123- rupestre sudamericano 2:36-53. La Paz: SIARB.
139. Sabadell (Barcelona): Editorial Ausa. Also published in LAnthropologie 92(1):361-370
1982d Paleogeografa sudamericana del pleistoceno (Pars, 1990).
tardo y holoceno: Ensayo de sntesis 1988-89 Actividades arqueolgicas en el extremo sur
actualizada. Actas y Memorias del IV Congreso americano (1988-1989): Informe (Report sent to
Nacional de Arqueologa Argentina (Segundo several foreign publications).
23 - Ceruti: Juan Schobinger
1989 200.000 aos del hombre en Amrica: que 1994c Participacin en el Panel II (La Legislacin y su
pensar? In Espacio, Tiempo y Forma, relacin con el patrimonio cultural y natural). II
Prehistrica Series 1, pp. 375-395 (Madrid: Jornadas La Cultura, fundamento de la
Universidad Nacional a Distancia). Democracia, pp. 89-93. Buenos Aires: H.
1990a Algunas observaciones terminolgicas sobre la Senado de la Nacin, Comisin de Cultura.
prehistoria americana. Clio, Serie Arqueolgica 1995a Aconcagua: Un enterratorio incaico a 5300 metros
1(6):7-29. de altura (48 pages). Mendoza: Privately
1990b La Patagonia en el marco de la ms antigua published.
prehistoria americana. In Culturas indgenas de 1995b Valcamnica Symposium 1994. Boletn 9:27-28
la Patagonia, edited by J. Roberto Brcenes, pp. (SIARB, La Paz, Bolivia).
151-168. Madrid: Sociedad Estatal Quinto 1995c El arte rupestre ms antiguo de Amrica: Una
Centenario and Turner Libros S.A. ojeada sinttica. Boletn 9:36-40 (SIARB. La Paz,
1991a Las ruinas incaicas de los Nevados de Bolivia). English-language version published as
Aconquija (Provincia de Tucumn): Informe The Earliest Rock Art in the Americas: South
preliminar. Imperio Inka 2:15-30 (special issue America. Rock Art Research 221(2):134-135.
of Comechingonia). 1996a The Earliest Rock Art in the Americas: South
1991b Sacrifices of the High Andes. Natural History, America. XIII UISPP Congress (Forli, Italy,
91/4:-68. 1996), Colloqia 8: Art in the Paleolithic and
1992a Las religiones precolombnas (76 pages). Buenos Mesolithic. Coloquium 16:127-132.
Aires: Editorial Almagesto. 1996b Le peuplement de lAmrique du Sud: Nouvelles
1992b The American Religions. Chapter 1 of The donns. XIII UISPP Congress (Forli, Italy, 1996).
Church in Latin-America, by Enrique D. Abstracts:506-508.
Dussel, pp. 23-42. A History of the Church in 1997 Arte prehistrico de Amrica. Mxico: Consejo
the Third World series, Volume 1. Tunbridge Nacional para la Cultura y las Artes. An Italian
Wells, United Kingdom: Burns & Oates and language version was published as Larte dei primi
Maryknoll, New York: Orbis Books. Also americani. Corpus Precolombiano series. Milan:
published in Italian as Le religioni amerindie Editorial Jaca Books (1997). An English
in La Chiesa in America Latina: 1492-1992; il language version was published as The Ancient
rovescio della storia, pp. 51-83. Asisi, Italy: Americans: A Reference Guide to the Art, Culture,
Citadella Editrice (1992) and in Portuguese as and History of Pre-Columbian North and South
Os religioes amerindias in Historia liberationis: America. Translated by Carys Evans-Corrales.
500 anos de histria da igreja na Amica Latina, Armonk, New York: M. E. Sharpe (2001).
pp. 35-68. So Paulo: Edies Paulinas (1992). 1998a La momia del Cerro Aconcagua (Mendoza). In
1993a La Amrica Precolombina. In Historia Arqueologa de Mendoza, by J. Roberto Brcena,
Universal, Volume 1, Prehistoria, pp. 342-379. pp. 1-18 (Primer Congreso Nacional de
Barcelona: Editorial Thema. Estudiantes de Arqueologa, Agosto, Mendoza,
1993b Arqueologa de alta montaa: Santuarios Argentina). Mendoza, Argentina: EDIUNC.
incaicos en los Andes centro-meridionales. 1998b La prehistoria como primer captulo de la Histor-
Beitrage zur Allgemeinen und Vergleichende ia Universal. Investigaciones y Ensayos 47:557-
Archologie 18:363-399 (Bonn, Germany). 574. Buenos Aires: Academia Nacional de la
1993-94 Informe sobre la relocalizacin de un hallazgo Historia.
de alta montaa del Noroeste Argentino: La 1999a Los santuarios de altura incaicos y el Aconcagua:
llamada momia de los Quilmes. Acta Aspectos generales e interpretativos. Relaciones
Archaeologica Carpathica 32: 113-124 (Cracow, de la Sociedad Argentina de Antropologa 24:7-27.
Poland). Also published in Comechingonia 1999b Las tierras cuyanas. In Nueva Historia de la
8:47-67 (1995). Nacin Argentina, Volume 1, pp. 159-180.
1994a The First Americans. Grand Rapids, Michigan: Buenos Aires: Planeta for the Academia
Eerdmans Publishing Company. Italian Nacional de Historia.
language version published as I primi americani. 1999c El territorio y la poblacin aborgen. Buenos Aires:
Origni series. Milan: Editoriale Jaca Book Academia Nacional de la Historia and Editorial
(1994). Planeta.
1994b La reunin internacional sobre el poblamiento 1999d Los poblaciones indgenas de Mendoza a travs
americano (Sao Raimundo Nonato, Brasil, del tiempo. In Guanache: Fidel Roig Matns,
1993), Boletn 8:32-36 (SIARB, La Paz, pintor del desierto by Fidel Antonio Roig, Arturo
Bolivia). Andrs Roig, Mario Roig, Virgilio Germn Roig,
ANDEAN PAST 10 (2012) - 24
and Enrique Franck Roig with M. Gmez de Juan [Santiago Ren] Schobinger and Clara Abal de
Rodrguez Brito, E. M. Abraham de Vzquez, Russo
M. R. Del Prieto, and Juan Schobinger, pp. 2009 Las Religiones Precolombinas y la Cultura de
127-129. Mendoza, Argentina: Arte Series, Chavn. Mendoza, Argentina: Ediciones La
EDIUNC for the Universidad Nacional de Aldaba and Editorial de la Facultad de Filosofa
Cuyo. y Letras. Universidad Nacional de Cuyo.
2002 Arte rupestre del Departamento de Malarge. Juan [Santiago Ren] Schobinger, Clara Abal de Russo,
In Entre montaas y desiertos: arqueologa del sur Mnica Ampuero, and Graciela N. Arenas
de Mendoza, edited by Adolfo F. Gil and 2001 El santuario incaico del cerro Aconcagua. Men-
Gustavo A. Neme, pp. 181-194. Buenos Aires: doza, Argentina: EDIUNC (Ed. de la Universi-
Sociedad Argentina de Antropologa, Buenos dad Nacional de Cuyo).
Aires. Schobinger, Juan [Santiago Ren] and Clara Abal de
2004-5a Breve curriculum vitae de Juan Adolfo Russo
Vzquez. Cuyo. Annuario de Filosofa Argentina 2009 Las religiones precolombinas y la cultura de Chavn.
y Americana 21-22:93-95. Coleccin Jarilla. Mendoza, Argentina: Edi-
2004-5b Las ltimas publicaciones de Juan Adolfo ciones La Aldaba and Editorial de la Facultad de
Vzquez: Breve comentario. Cuyo. Annuario de Filosofa y Letras, Universidad Nacional de
Filosofa Argentina y Americana 21-22:127-132. Cuyo. Mendoza, Argentina.
2008 Antropologa y Etnologa, Prehistoria y Schobinger, Juan [Santiago Ren], Mnica Ampuero, and
Arqueologa. Tres Textos sobre Aspectos Eduardo Guercio
Conceptuales y Terminolgicos. Mendoza, 1985 Descripcin de las estatuillas que conforman el
Argentina: Editorial de la Facultad de Filosofa ajuar acompaante del fardo funerario hallado
y Letras. Universidad Nacional de Cuyo. en el Cerro Aconcagua (Prov. de Mendoza).
In press Arqueologa y arte rupestre de la regin cuyana. Relaciones de la Sociedad Argentina de Antropologa
Mendoza, Argentina: Editorial de la Facultad 16:175-190 (Buenos Aires).
de Filosofa y Letras. Universidad Nacional de Schobinger, Juan [Santiago Ren] and J. Roberto Brcena
Cuyo. 1972-73 El tambo incaico de Tambillitos (Provinica de
Schobinger, Juan [Santiago Ren], Editor Mendoza, Argentina). Actas del VI Congreso de
1966 La momia del cerro El Toro: Investigaciones Arqueologa de Chile, pp. 397-404. Santiago de
arqueolgicas en la cordillera de la Provincia de Chile.
San Juan. Supplement to Volume 21 of the Schobinger, Juan [Santiago Ren], J. Roberto Brcena,
Anales de Arqueologa y Etnologa (Facultad de and Jorge Fernndez
Filosofa y Letras, Universidad Nacional de 1974 Excavacin arqueolgica del sitio precermico
Cuyo, Mendoza, Argentina). Second revised, Espinazo del Diablo, Mina Aguilar, Provincia
corrected, and expanded edition published de Jujuy. Relaciones de la Sociedad Argentina de
(2009) in Coleccin Cumbre Andina. Antropologa, n.s. 8:89-99.
Universidad Nacional de Cuyo, Mendoza, Schobinger, Juan [Santiago Ren] and M. Costanza Ceruti
Argentina: Editorial de la Facultad de Filosofa 2001 Arqueologa de alta montaa en los Andes
y Letras. argentinos. In Historia Argentina Prehispnica,
1995 Humanismo Siglo XX: Estudios dedicados a Dr. edited by Eduardo E. Berberin, Axel E. Nielsen,
Juan Adolfo Vzquez. San Juan, Argentina: and Mara Ester Albeck, et al. Volume 1:523-
Editorial Fundacin Universidad de San Juan. 559. Crdoba, Argentina: Editorial Brujas.
1997 Shamanismo Sudamericano. Buenos Aires: Schobinger, Juan [Santiago Ren] and S. Gopinathan
Editorial Almagesto-Continente. 1977 The Emergence of Man. Tangents Series 12.
2001-3 El santuario incaico del Nevado de Chuscha Singapore: Federal Publications. Also published
(zona limtrofe Salta-Catamarca). In Anales de in Indonesian as Muculnya manusia (with H.P.
Arqueologa y Etnologa 56-58 (Instituto de Idris M.T.). Jakarta: Balai Pustaka (1987).
Arqueologa y Etnologa, Facultad de Filosofa Schobinger, Juan [Santiago Ren] and Carlos J. Gradin
y Letras de la Universidad Nacional de Cuyo, 1985 Cazadores de la Patagonia y agricultores andinos:
Mendoza, Argentina). Buenos Aires and Men- Arte rupestre de la Argentina. Las Huellas del
doza: Ed. Fundacin CEPPA (Centro de Estu- Hombre Series. Madrid: Ediciones Encuentro.
dios para Polticas Pblicas Aplicadas). Also published in French and in Italian by
Editoriale Jaca Book, Miln.
25 - Ceruti: Juan Schobinger
Abal, Clara M.
2009a Juan Schobinger (1928-2009): Nuestro adis
a un caballero de la ciencia. Revista Chungar
41(2):179-182.
2009b Adis a Juan Schobinger. Diario Los Andes,
Seccin Cultura, 25 July.
Barberena, Ramiro
2008 Vida de un arquelogo, petroglifos y santuarios
de alta montaa: Entrevista con el Dr. Juan S.
Schobinger. Relaciones de la sociedad argentina
de antropologa 23:1-15.
Brcena, Roberto
2009 Permanence en nosotros. Diario Los Andes,
Seccin Culturas, 25 July.
Lagiglia, Humberto
2005 Juan Santiago Ren Schobinger, arquelogo.
In Anales de Arqueologa y Etnologa 59-60:7-
27 (Facultad de Filosofa y Letras, Universidad
Nacional de Cuyo, Mendoza, Argentina).
Podesta, Mercedes
2009 Obituario de Juan Schobinger. Revista de
Arqueologa 15 (Buenos Aires).
Rivera, Mario A., with Roberto Brcena and Mara
Rosario Prieto
2009 Juan Santiago Ren Schobinger: 1928-2009.
The Archaeological Record 9(5):7, 44 (Society
for American Archaeology, Washington,
D.C.).
Strecker, Matthias
2009a Hans (Juan) Schobinger, 1928-2009. Rock Art
Research 26(2):243.
2009b En memoriam: Hans (Juan) Schobinger (1928-
2009). Boletn 23:12-14 (SIARB, La Paz,
Bolivia).
ANDEAN PAST 10 (2012) - 26
Juan Schobinger, Keynote Speaker at a Latin American Indian Literatures Association meeting
during a reception in his honor, International Headquarters
of the Society of Woman Geographers, Washington, D.C. (2000).
From viewers left to right: Monica Barnes, Luis Arrata, Mary Preuss, and Juan Schobinger.
A CHANGING SOCIETY?
CRAFT SPECIALIZATION AND COMPLEMENTARITY SYSTEMS DURING THE FORMATIVE PERIOD
IN THE COCHABAMBA VALLEY, BOLIVIA
OLGA U. GABELMANN
Freie Universitt Berlin
at Chullpapampa. This site is a few kilometers In view of the stone tools found, he ascribed
up the valley from Colcapirhua, at the foot of the economic basis to agriculture. Because these
the Cerro Tunari (Figure 1). Contrary to people inhabited mound-like settlements, he
Bennett, Rydn was convinced that the mono- named them the Cultura de los tmulos (Mound
chrome ceramic style must be older than Culture). In agreement with scholars including
Tiwanaku, and may have influenced the later Carlos Ponce Sangins and Max Portugal Ortiz,
period (as in the case of the Grey Ware pottery he deduced a common tradition among the
in Mizque, see Rydn 1961:50). contemporary groups of the southern altiplano,
subsumed under the name Wankarani, whose
In the 1950s and 1960s, Dick Edgar Ibarra settlements also consisted of stratified mounds
Grasso, founder and first director of the and whose pottery was likewise monochrome
archaeological museum in Cochabamba (now (Ponce 1970; Portugal 1998:9-20; cf. Walter
the Instituto de Investigaciones Anthropolgias 1966). The find of a single large stone stela in
y Museo Arqueolgico Geraldine Byrne de Sierra Mokho in the Central Valley (Ibarra
Caballero or INIAM), conducted investigations 1964) slightly resembling the Wankarani llama
at numerous sites in the Cliza area, among them effigy heads (cf. Guerra 1994: figure 3-16; Portu-
Santa Luca and Chullpa Pata in the High gal 1998:11-20) supported his hypothesis. It
Valley (Figure 1) where he found pottery similar even tempted Ibarra Grasso also to use the term
to the Chullpapampa style (e.g. Ibarra 1955, Cultura megaltica. Nevertheless, Heinz Walter
1956, 1964, 1965:82; Ibarra and Querejazu (1966:118) opposed the idea of a common
1986:150). Even though his publications lack a culture shared by Cochabamba and Wankarani,
description of his methodical approach, maps, based solely on the use of monochrome ceram-
and other detailed site information, Ibarra ics. In fact, the pottery and artifacts of both
Grasso was the first to present an overview of groups seemed quite different to him.
the period of an emerging sedentary society
using monochrome pottery. He was also the first
to provide absolute radiocarbon dates from
Cliza, establishing the end of this early period at
around A.D. 280 (Ibarra 1962: Table 1:1).
Heinz Walter formed part of the German for the valleys (Brockington et al. 1985, 1987,
Mission, led by Hans Dietrich Disselhoff in 1995). Their Formative Period was somewhat
1958, who investigated four sites in Bolivia arbitrarily divided into Early, Middle, and Late
(Wankarani, Chullpa Pata, Mizque, and Icla) Phases; their ceramic types showed hardly any
(Walter 1966). With the participation of Ibarra changes throughout the period. Nevertheless,
Grasso the team investigated the site of Chullpa the beginning of their Middle Phase around 500
Pata near Cliza in the High Valley (Figure 1). B.C. (Brockington et al.: 1995) was based on the
On an alluvial terrace to the east of the village appearance of a new ceramic type in Sierra
they noticed eight mounds. They chose the Mokho (tipo 7; Brockington et al. 1987:33). In
smallest two for their excavations (ibid:122). the lower levels, which were dated to approxi-
These produced the first evidence of domestic mately 1125 B.C. (Table 1) they found altiplano
architecture (house floors) in an early settle- type pottery, probably of Wankarani origin,
ment. Walter noted a resemblance between the indicating long-distance trade between the
pottery from the Central Valley and the High valleys and the highlands.
Valley and named this style Chullpapampa
Monochrome (ibid:164). The upper layers of the Furthermore, the same team found a hint of
mounds contained sherds of the polychrome the existence of public or ceremonial architec-
Tupuraya style (cf. Rydn 1959) and were dated ture in Chullpa Pata, where they investigated a
to the beginning of the second century A.D. third mound that had not been excavated by
(Table 1). Combining the stratigraphic informa- the German Mission. According to Brocking-
tion from Chullpa Pata with the excavations of tons team, a small trench revealed part of a
Bennett and Rydn in Chullpapampa and four-step terraced construction consisting of
Colcapirhua enabled Walter to establish a local stones and clay with a sherd paving (pozo 1;
sequence that placed the Chullpapampa Mono- Brockington et al. 1985:4, 1995:11). Unfortu-
chrome Style earlier than the polychrome Tupu- nately, this construction was never the subject
raya style (Walter 1966:180). This chronology of further investigation, and, therefore, its size
is still valid. It was only refined in absolute- and function remain unknown.
chronological terms during later investigations.
An outstanding initiative of the Proyecto
In the 1980s, Donald Brockington, David Formativo is the excavation of the Conchupata
Pereira, and other staff members of the Cocha- burial site in the Mizque Valley with its abun-
bamba archaeological museum launched the dant grave goods made from sodalite and Pacific
Proyecto Formativo, dedicated to a more sys- shells, and dating to between 1180 and 800 B.C.
tematic investigation of previously known sites (Table 1). Fieldwork in the burial mound took
in the Central and High Valleys. The team place in 1986 under the leadership of the Pro-
changed Ibarra Grassos terms Megalithic or yecto Formativo. It is the first and the only
Tumulus Culture, as well as Walters comprehensive study to date of the Formative
Chullpapampa Monochrome Style, to the Period burial pattern in the greater Cochabamba
more common designation, Formative Period. area. Mizque is 140 kilometers to the southeast
Their first excavation took place in the multi- of Cochabamba, but in terms of access to pres-
component mound of Sierra Mokho in the tige goods like marine shell, the Cochabamba
Central Valley (Figure 1), which yielded the Valleys may well have played an important role
complete prehistoric occupation sequence. as a corridor for long-distance exchange.
Based on their ceramic analysis and radiocarbon
dates, they established an absolute chronology
ANDEAN PAST 10 (2012) - 30
In the late 1990s, excavations conducted by close to rivers or areas with a high ground water
the present author in Choroqollo in the table prone to periodic flooding. This includes
Santivaez Valley (Figure 1) provided the first the zone between Sipe Sipe, Quillacollo, and
evidence of pottery production in the Formative Colcapirhua in the Central Valley, where various
Period of Cochabamba (Gabelmann 2001). mountain streams cross the bottom of the valley
Tools, over-fired sherds, and clay slag attested in a north-south direction (Figure 1). Interest-
to the activity of potters. Even though it was not ingly, these mound settlements always show
possible to determine its mode of production occupation in later periods. About 48 of such
and range of distribution, it seems that the multi-component mound sites have been re-
pottery was spread over the whole valley, and ported in this area (Cspedes et al. 1994:3, map
even into the Capinota Valley as can be de- 3). In contrast, the piedmont settlements tend to
duced by rim forms (cf. Higueras 1996; be occupied only in the Formative Period.
McAndrews and Rivera 2007). Choroqollo was
dated to an Early Formative phase (Table 1) and, Multi-component mound sites are less fre-
therefore, is contemporaneous with the lower quent in the High Valley. Eight examples have
levels of Sierra Mokho (Levels 29-33; Brocking- been registered to date, and all are located on the
ton et al. 1985, 1987). In Choroqollo, a small now densely populated alluvial terrace cut by the
percentage of pottery also appeared that could be Cliza River between Chullpa Pata (Figure 2:
assigned to the Wankarani Type, indicating VA_144) and Cruz Pata. The locations of these
exchange with altiplano groups during an Early mounds share the same geographic characteris-
Formative phase between 1300 and 1000 B.C. tics. As in the Central Valley they are close to a
river flood plain. In the Cliza area also, these
Also during the 1990s, extensive surveys were mounds were occupied beyond the Formative
conducted in the valleys of Santivaez (Vetters Period, maybe even until the Middle Horizon. In
and Sanzetenea 1996), Capinota and Mizque contrast, the neighboring site of Santa Luca
(Higueras 1996, 2001) as well as in the western- shows a different settlement type, because it is
most part of the Central Valley (Gyarmati et al. not a mound site, and it is not in the foothills
1999). These investigations contributed to the (Figures 1, 3). Its location on the slightly elevated
picture of a rather common land use strategy, as edge of a large alluvial fan only three kilometers
Higueras (1996) suggests, with a settlement to the west of the Cliza River may have prevented
preference for poorer soils in the piedmont area flooding, and thus made mound construction
in all the valleys. At first sight, this may point to unnecessary. Also, Santa Luca is not a multi-
a population engaged primarily in agriculture, component site, and so there may have been
which erected their settlements on poorer soils, insufficient time for mound formation from
thus sparing the fertile soils for plant cultivation. occupation debris. It was never re-occupied after
However, a closer look at the nature of Forma- the Formative Period.
tive Period settlements reveals different types that
may vary from valley to valley. It seems that the settlement types differ from
valley to valley according to specific geographic
The elevated mound sites, depicted by Ibarra conditions. The amount of flowing water and the
Grasso as the main components of his early water table are (and always were) higher in the
Tumulus Culture, seem to characterize a Central Valley, necessitating greater mound
certain landscape management strategy during constructions in prehistory. The High Valley
the Formative Period. They are not in the pied- seems to have been drier, because mound con-
mont area, but rather appear in valley bottoms structions appear only in the river area south of
31 - Gabelmann: Craft Specialization in Formative Cochabamba
Cliza. However, the Central Valley did not yield Pacific coast; hence, Cochabamba was incorpo-
multi-component mound sites exclusively. Here rated in supra-regional trade systems. Unfortu-
single-component, flat Formative Period sites nately, the social organization of the valleys
also exist on alluvial cones in the foothill areas inhabitants themselves has never received any
(e.g. Chullpapampa; for others see Gyarmati et al. systematic assessment. Was agriculture their only
1999:27). Ibarra Grasso did not differentiate economic subsistence? Did they produce pottery
between site types, because his focus was on the only for their own use? Were they really self-
characteristic mound structures. Meanwhile, sufficient, non-stratified, agrarian societies that
even Formative Period hilltop sites have been persisted in this state until the Early Intermedi-
reported (Higueras 1996). However, the question ate Period? The Santa Luca Project sought to
of whether the different site types represent focus this blurred picture, and to conduct
different functions or ranks in terms of social research at a pottery production site, to shed
complexity has not yet been sufficiently investi- light on the socio-economic structures of the
gated, but the construction of mounds represents Formative Period, such as its specialized manu-
not only an adaptation to flooding, but also a facturing and organization, and their social and
sign of collective labor. economic interaction spheres.
There is clear evidence indicating continuous Evidence of pottery production was visible in
exchange and communication between different the remnants of more than two hundred heavily
societies in the south-central Andes. Moreover, eroded firing facilities or kilns, which mainly
there was organized access to prestige goods from consisted of a clay platforms (Figures 4, 6). The
as far as the altiplano, the lowlands, and even the facilities present circular or oval forms with
ANDEAN PAST 10 (2012) - 32
diameters between 0.80 to 3.80 meters. Because The artifacts found on the surface were made
these features have been eroded, their original from stone, semi-precious stone, shell, bone, and
forms remains unclear. Fired daub fragments ceramics. More evidence of pottery production
with imprints of cane found in the upper debris was present in the form of tools and over-fired
layer point to rather sophisticated kiln structures sherds. Shaping tools of modified sherds have
above the platforms. The existence of a closed been used to finish the surface of the raw vessels
kiln structure can be ruled out, however, because before applying the slip (Figure 7). Small stone
there was no antechamber for fuel. In addition, scrapers may have served for cutting the coils and
the pottery shows quite striking fire-clouds and taking off excess clay when shaping the pots.
cracks due to lack of control over the firing Other important finds that could be linked to
temperature and atmosphere, suggesting that the pottery production are incised, slightly conical
kilns were open (Rice 1987:155). It seems that ceramic tubes that may have been used for
the firing facilities consisted of a platform with igniting the combustible materials when firing
low walls, or a kind of modeled basin. Some the vessels (Figure 8). The diameters of the tubes
facilities presented multiple platforms indicating range from 2.8 to 6 centimeters. Unfortunately,
that they were remodeled after a certain time. only fragments were found in Santa Luca, hence
One platform even showed depressions, possibly their original lengths are unknown. However,
for the firing of large storage vessels with round examples in the INIAM museum show total
bases (Figure 6, right). This demonstrates that lengths of between18 to 38 centimeters. While
the potters used quite a sophisticated technology these lengths may not have been sufficient for a
for the open fire or bonfire method, which safe approach to the fire, they may be ceramic
was suitable for repetitive application (Gabel- tips or tuyeres, which were elongated with reed
mann 2008b). or totora for this purpose.
Additionally, circular stone foundations with The mapping of superficial artifacts was seen
diameters of 2 to 6 meters appear on the surface as problematic because of erosion, mainly due to
of the site. They consist of either ring-shaped wall heavy precipitation during rainy seasons, and the
foundations or circular pebble layers. The latter subsequent possibility of water transport. Parts of
may have formed the foundations of now-eroded the site are, indeed, heavily eroded, and with the
clay floors. The larger structures may have served drain direction towards the northeast one would
as domestic buildings, as is the case with struc- expect a greater density of artifacts in the north-
ture C5 that has a clay floor and a central post eastern sector of the site. The density of material
hole (Figure 5). Smaller ones could be inter- in that sector was high, but other sectors less
preted as huts, sheds, or animal pens. affected by possible water transport, such as the
southwestern sector, showed high densities, too
The mapping of all the structures revealed a (Table 2). Though we cannot discard the possibil-
surprising picture, with three different functional ity of a surface pattern formed by erosion, the
sectors (Figure 4). The concentration of large and fact that with every heavy rainfall new artifacts
small circular stone structures visible in the appeared on the surface implies that the pattern
center can be interpreted as the domestic sector. may not have drastically changed. Additionally,
This area is surrounded by the production sector, the inclination of the site is not steep, and in
which mainly consists of firing facilities and some areas grass may have prevented water
smaller structures. The periphery shows ash transport. Artifacts most prone to water trans-
mounds in a semi-circle, which served as a dump port are the lightweight beads, shells, and small
site and burial ground. sherds, rather than stone artifacts and the larger
33 - Gabelmann: Craft Specialization in Formative Cochabamba
sherds of storage vessels. Finally, the site was (Santa Luca III; Gabelmann 2008a:124-25).
occupied only during the Formative Period, and Nevertheless, it cannot be ruled out that material
shows no interference from later periods. of an earlier phase appeared, especially in the
lower elevations towards the north and the
The surface material belongs to a Late Forma- northeast of the site.
tive phase dating between 500 B.C. and A.D. 60
Table 2: Ceramic type distribution of five systematic collections (see Figure 15) and from total sherd count on surface.
Despite these problems, the mapped tools define a functional sector only on the basis of the
and artifacts on the surface do perfectly under- presence of tools.
line the functional spatial division between the
working and living areas (Figure 9). Tools used Furthermore, the surface showed concentra-
for pottery production, like stone scrapers and tions of raw materials and objects made of miner-
shaping tools, amount to a higher percentage in als and semi-precious stones like malachite and
the production area, or even to a total absence in sodalite, as well as shell from land snails and
the domestic area. The same is true for the marine mollusks, which provides evidence of the
decorated ceramic tubes or tuyeres, which most settlers activity in making beads and other
probably served for igniting fires. On the other adornments, in addition to pottery production.
hand, grinding and hammer stones appeared in According to the distribution of the artifacts, this
both sectors, because they could be used for activity seems to have taken place in both the
either grinding food or temper (Sillar 2000:56). production and the domestic sectors (Figure 9).
It should be mentioned that many tools could However, because lightweight artifacts are most
have had multiple functions and may have served prone to water transport during heavy rains, this
household, agricultural, and production pur- evidence should be interpreted with caution.
poses equally. Therefore, it may be misleading to
ANDEAN PAST 10 (2012) - 34
Judging from the degree of erosion on the on the site. The proximity of the finds lets us
surface, the site was not expected to have any assume that the plaster once had been part of
deeper levels of occupation. Nevertheless, excava- this structure. Painted wall plaster is otherwise
tion proved otherwise. A small trench excavated unknown in the Cochabamba Formative Period
in the production sector showed occupation to date. Thus, we can infer a special function for
levels to a depth of 3.65 meters without reaching Structure C5.
sterile soil (Figure 10). It yielded floor and fill
layers of the Middle Formative (Santa Luca II, POTTERY PRODUCTION AND ORGANIZATION
950-500 B.C.), and the Early Formative phase
(Santa Luca I, before 950 B.C.; Gabelmann et al. On the basis of the spatial division between
2009). Shaping tools for pottery production were the production and the domestic sectors, as well
present until Layer 19, suggesting that pottery as of the technological investments, it is apparent
manufacture was always the principal task in the that pottery production exceeded the needs of
settlement. In contrast with the surface distribu- the village population. Indeed, the pottery
tion, here many shaping tools were found on showed a rather wide distribution sphere; It was
house floors, associated with cooking pots, found throughout the High Valley, and appeared
weaving tools, ceramic spoons, and lithic debris, as a small percentage of total ceramics, even in
supporting the assumption that pottery produc- the Central Valley (Sierra Mokho).
tion was performed as a household activity
during the Early and Middle Formative phases. Hence, the existence of specialists may be
assumed, although considerable confusion exists
Another trench was excavated in the western about what specialization is (Rice 1987:189).
periphery in one of the ash mounds (Figure 4, The term may refer to potters, who dedicate
coordinates North 480/East 340), where ten most of their time to production, and whose
burials dating to the Late Formative or Santa income is mainly generated by production (pro-
Luca III phase (500 B.C.-A.D. 60) were exca- ducer specialization sensu Longacre 1999:44).
vated. The individuals range from about six to Alternately, it may apply to pottery production
about fifty years of age and were deposited in the that developed due to favorable ecological condi-
ash layers as primary or secondary burials. All tions (e.g. presence of clay mines), which may be
skulls are modified, and at least two different related, at the same time, to poor agricultural
deformation types are present. Evidence of potential (site specialization sensu Rice 1987:189;
arthritis is present and there are pronounced cf. Costin 1991:3). In Santa Luca, a site special-
muscle attachments in the chest and neck areas, ization is probable on the grounds of the poor
which point to carrying heavy burdens, possibly and salty soils at the edge of the alluvial fan (soil
associated with pottery production (Gabelmann class V-VI; Gabelmann 2008a) and because of its
2008a: 303). proximity to clay mines only five kilometers
distant in the piedmont zone to the south. Here
The excavation of a waste pit in the domestic a number of modern pottery producing villages
sector close to the circular structure C5 revealed (e.g. Huayculi and Vilaque; Figure 4) possibly
fragments of polychrome painted wall plaster exploit the same clay resources (Gabelmann
(Figure 11). These fragments seem to come from 2008a:191; Goins 1967:126-127, 153; Sillar
the interior of a circular house as deduced from 2000:25). The fact that these villages show signs
a slightly concave curvature. Structure C5, with of prehistoric production (see below) underlines
a diameter of 6.10 meters, and a hole for the site specialization throughout various periods
central post, is the largest circular stone structure until the present.
35 - Gabelmann: Craft Specialization in Formative Cochabamba
The degree or intensity of a producer special- lands. Thus, a more humid climate in the high-
ization is far more difficult to estimate for prehis- lands may even have led to a displacement of the
toric contexts. According to Dean Arnold (1985: sub-tropical humid montane forest belt towards
18) a full-time specialization occurs, if potters the west into the eastern Cochabamba valleys
work year round. This is only possible in favor- (Ricardo Cspedes, personal communication).
able arid and dry climate zones like the Peruvian For all the missing details on the paleoclimatic
coast, or the lower montane thorn steppes, where conditions in the valley region that need to be
precipitation is less than 500 millimeters per year determined in the future, we may cautiously
(Arnold 1993:224-231). Although precipitation deduce a more humid climate for the Late For-
in the High Valley amounts to 420 millimeters in mative Period based on the fact that eight of
the valley bottom, due to pronounced seasonal thirty-nine archaeo-botanical samples contained
changes between rainy and dry periods, such a Cyperaceae (reeds or totora) from the swampy
full-time specialization is not practiced today, valley bottom close to Santa Luca (Gabelmann
even with the existence of canopied kilns (e.g. at 2008a:176, table 3.4).
Huayculi; Gabelmann 1999: figure 5). Humid
conditions and rain complicate production Even in the case of less humid climatic
because they cause difficulties in drying and conditions, we should not assume the existence
2
firing the pots. Not only in Cochabamba, but of full time specialists in the Formative Period,
throughout the Andes, potters have adapted to but rather potters who produced only during the
3
part-time activity during the months between dry periods. Nevertheless, a complementary (or
April/May and October (Arnold 1993:54; Ch- risk-minimizing) activity of the potters in agricul-
vez 1992:68; Litto 1976:13; Sillar 2000:73, table ture during rainy periods is difficult to detect.
4.6) and may compensate for their shortfall with Tools generally associated with agriculture like
agricultural activities used as a risk minimization stone axes and clod breakers may have had
measure (Hagstrum 1989:103). multiple functions, and perhaps were also used
for crushing temper, or beating clay lumps in
Certainly, the climate in Cochabamba during pottery production. Neither can agricultural
the Formative Period may have been different activities be excluded. Until now, there are no
from that of today, even though explicit paleo- explicit data as to whether the potters were also
climate studies for the valleys are lacking to date. landholders and cultivated crops, or whether
Studies of highland lake levels and Bolivian ice they exchanged food products for pottery (and
cores suggest a warm and humid climatic opti- possibly dedicated their time during the rainy
mum between 4500 and 3200 B.P., followed by season to the fabrication of beads). Still, a pro-
an arid phase (Argollo 1991; Argollo and ducer specialization, or better, a seasonal pro-
Mourguiart 1995, 2000; Seltzer and Wright ducer specialization, in the sense of pottery
1991; Thompson et al. 1998). Thus, we may production representing the primary activity, can
assume a more humid climate until the Early be deduced from the investment made in tech-
Formative Period. However, weather conditions
in the altiplano may not necessarily coincide with 3
In the end, full time specialists can only be expected in
those of the valleys because the dominant mois-
centrally organized production centers with a high degree
ture source there comes from the eastern low- of elite sponsorship, which covers provisioning and
possible risks for the potters (cf. Brumfiel and Earle 1978).
This situation may have occurred during the Inca Empire,
2
Pronounced seasons are the case in most areas of the but cannot, as yet, be assumed for any other prehistoric
world. This is why a possible full time specialization of period in the South Central Andes (Gabelmann
prehistoric potters is questionable. 2008a:260).
ANDEAN PAST 10 (2012) - 36
bulging. These may have served as a substitute below, it seems appropriate to assume that only
for handles. The surfaces of the vessels are coated one type was produced in Santa Luca. However,
with a thin bright-to-dark red slip after the pro- interestingly, only 53 percent of the total sherds
duction of striation marks with a hard-edged on the surface consist of Type 10 (Table 2). This
tool. The striation marks sometimes form simple is a rather low figure for a production site. One
patterns. The temper consists of fine, rounded would expect the rate of the produced ware to be
river gravel with the addition of grog. higher.
Type 10, by contrast, is composed of a more Nonetheless, the distribution ratio differs
limited variety of vessel forms, mostly large oval according to sectors. There is a rather high
storage or fermentation vessels and straight-sided concentration (80 to 97 percent) of Type 10
bowls, but also small oval jars (Figures 13, 14). sherds visible in the North-Central and Western
The rim forms are very simple, angular or oval, production sectors as compared to a rate of 63
but mostly rounded down and slightly thickened. percent of Type 20 sherds in the center, the
The vessels are always coated with a thick red- residential area (Figure 15; Table 2). This corrob-
brown self-slip, and then burnished with pebbles orates the identification of the production sector,
to a silky luster, obtaining a compact, less perme- and identifies the products as belonging to Type
able, and abrasion-resistant surface at the same 10. Second, it seems that Type 20 vessels are
time. Crushed river gravel and quartz were added more common in pure household activities than
as temper; grog is less common than in Type 20. Type 10 vessels, with the ratio being almost 2:1
with Type 20 predominant. Type 10 vessels
Although forms and treatment of the two represent only about one-third of pottery in
types clearly point to a functional difference, a households. An explanation for the relatively low
distinction between serving and utility pottery for percentage of Type 10 in the residential area is
either one cannot be observed, because both that large storage jars with thick walls have a
types show vessel forms for pure household longer life span and break less often than the
activities (e.g. cooking, storage) as well as for thinner serving or cooking vessels of Type 20.
serving (Figure 13). The producers of Type 10 Additionally, if thick vessels break, they may
were primarily specialized in large storage vessels produce a lower number of fragments than a
that can hold liquids, while Type 20 mainly thin vessel, even though they may be larger.
presents cooking pots with rounded bases. Both
ceramic types present tableware like flat or deep The exclusive production of Type 10 in Santa
bowls and jars. Also, a small percentage of vessels Luca represents a specialization in storage ves-
with incised decoration or applications appear in sels, which were objects of commerce. The
both ceramic types (see below). Here, the exis- explanation that presents itself is that Type 10
tence of highly decorated ceramic tubes needs to vessels were exchanged for those of Type 20. But
be mentioned. These were made exclusively from where did Type 20 come from? The high percent-
a variety of ceramic Type 10 (Figure 8). age of Type 20 vessels in Santa Luca suggests a
valley-specific production in close proximity. This
The differences between the two types are assumption is supported by the crystallographic
considerable and raise the question of their structure (XRD) and the elemental analysis
production location. Over-fired sherds, as well as (XRF) of the pottery types (Table 3 and Figure
vessels left on the firing platforms verify the 16). Both ceramic types exhibit similar mineral
production of Type 10 pottery on the spot. and chemical components, indicating the exploi-
Because of this, and for other reasons explained
ANDEAN PAST 10 (2012) - 38
tation of vicinal clay resources (Gabelmann SETTLEMENT PATTERN IN THE HIGH VALLEY
2008a).
As part of the Santa Luca Project, a survey
Sherd No. Laboratory No. Ceramic type was conducted in the southwestern area of the
SL-01 SL-01 20
High Valley. The survey area ranged over 150
SL-208/054 SL-208-054 10
o square kilometers, and included all ecological
SL-N Bolsa 224 SL (Bolsa 224) clay slag
o
SL-N Bolsa 161 SL (Bolsa 161) clay slag
levels, such as the valley bottom, foothills, slopes,
and mountaintops within 500 vertical meters
Element Samples (Figure 2). The results basically confirm the
SL-01 SL-208-054 SL Bol 224 SL Bol 161 Formative Period settlement patterns already
Majority presence observed in the neighboring Capinota Valley
Silicon xx xx xx xxx (Higueras 1996), coinciding with a general
Potassium xx xx xx xxx preference for locations in the upper valley, pied-
Aluminium xxx xxx xx - mont levels, or alluvial fans, and mostly along-
5
Iron xxx xxx xx x side the major rivers. Among the 48 Formative
Calcium - x xxx xx Period sites we registered, some are outside the
Titanium xx xxx x - survey area like the previously known site of
Minority presence Chullpa Pata near Cliza (Figure 2: VA_144). No
Calcium xxx - - - new mound-like settlements were encountered.
Phosphorus - - - xxx Nevertheless, three multi-component sites were
Titanium - - - xxx noted on natural hills or slopes (Figure 2:
6
Aluminium - - - xxx VA_49, 64, 127), which exhibit the whole
Zirconium - xxx xx x occupation sequence from the Formative to the
Strontium - xxx - xxx Late Intermediate Period, while the Formative
Trace presence Period usually occupies slopes and bottoms.
Zirconium xxx - - - Most of the other 44 sites were reoccupied at
Rubidium xxx xx xx x least once (mostly during the Late Intermediate
Strontium xx - xxx - Period), and only a few sites are small single-
Phosphorus x x xxx - component sites. The latter may be interpreted
Magnesium xx xx x xxx
as sites dedicated to agrarian activities, because
Barium xxx xxx xx x
Zinc xxx xx xx x
Copper x x x x 5
The absence of sites in the valley bottom may be due to
Niobium x xx x xxx
the existence of an extensive marsh district until the first
Lead x x x - Spanish settlers drained it in the sixteenth century. On
Sodium x xx x xx the other hand, if there were any prehistoric sites in the
Nickel x x x - valley bottom, they probably have been destroyed by the
Gallium x x x -
extensive agrarian activities that have taken place since
colonial times. This may be the reason why sites close to
the artificial modern lake are very small (see Figure 2;
Table 3: XRF results of four samples (ceramic and clay slag) from VA_3, 68, 72).
Santa Luca (Instituto de Investigaciones Geolgicas y del Medio
Ambiente, Universidad Mayor de San Andres, La Paz). The 6
At Convento Tarata (Figure 2; VA_127) we did not see
number of xs indicates the relative abundance of the element in any Formative Period pottery, but the collection of Padre
relation to other elements in the sample with x representing a low Mauricio Valcanover includes Formative type pottery said
percentage and xxx representing a high percentage of the element. to come from the site. It is very likely that the site had a
Formative Period occupation, as this is the case with
VA_122 and 123 (see Figure 2).
39 - Gabelmann: Craft Specialization in Formative Cochabamba
they occupied the fertile soils of class I-III close functional needs of activities like cooking. In-
to the valley bottom (Gabelmann 2008a: map stead, they preferred to trade in pots from other
A.4). production sites in the valley that specialized in
heat-resistant forms, tempering, and surface
Interestingly, we encountered three addi- treatment. Thus, they created a kind of volitional
tional pottery production sites within the survey interdependency, because it does not result from
area and one lying only slightly outside it (Figure the lack of appropriate raw materials, or from the
2: VA_11, 64, 107, and 139). These sites are lack of technological skills. The thought behind
aligned in regular distances from each other in this principle is not competition but comple-
the southern piedmont area, within a distance of mentarity. But what were the reasons for an
5.8 to 12 kilometers from Santa Luca (Figure 2: artificially created complementarity operating on
VA_10). All four sites show the same homoge- a horizontal level?
nous distribution of Type 10 and Type 20 ceram-
ics on the surface. Because Type 10 was pro- The following conclusion is based on analogy
duced in Santa Luca, it is likely that one, or with ethnographic data from the High Valley,
even all four, of these sites produced Type 20 which shows that a similar specialization system
vessels. The existence of four production sites of still existed in the twentieth century, and in some
Type 20 is feasible, because it would explain the villages persists up to the present. The principal
presence of the broad variety of vessel and rim production area extends over nine neighboring
forms (Figure 13). Hypothetically, we may inter- villages on the southwestern edge of the valley.
pret specific rim form groups as characteristic of These villages specialize in certain forms, like
a certain production site. However, this assump- small cooking pots in Vilaque and serving dishes
7
tion will require further comparisons with the and toys in Huayculi. Another modern produc-
pottery from each of the four sites. tion site lies outside of this area. Paracaya in the
northeastern slopes of the valley (Figure 1)
Not only the production sites, but also specializes in giant open vessels used for chicha
almost all Formative Period sites, revealed sherds production (Litto 1976:61-63; Sillar 2000:27,
of Types 10 and 20 on the surface, a fact that is 160-161). Here we observe a complementarity
even valid for sites outside the survey zone system with a restriction on certain forms for the
(Gabelmann 2008a). Both types seem to be more different villages. In Huayculi, the number of
or less evenly spread throughout the valley. Such vessel forms produced is even divided into the
a diffusion points to a certain distributive system, number of producing households in the village.
where all settlements had unlimited access to the Each household manufactures only two or three
products from either manufacturing center to objects of the whole spectrum (flower pots,
complete their ceramic inventory. The investiga- plates, soup dishes, cups, piggy banks, etc.),
tion of the nature of and the motivation for such which add up to the complete set that is then ex-
a system is indeed challenging. changed at local markets (Gabelmann 1999).
Thus, each household has a fair chance to partic-
COMPLEMENTARITY ON THE MICRO-SCALE
LEVEL OR INTRA-VALLEY INTERACTION 7
The location of these villages coincides with the
prehistoric production sites (Figure 2). Modern Vilaque is
First, the question arises as to why the vessel close to VA_139, the prehistoric site of Huayculi is
forms of Type 20 were not produced locally in VA_11). This continuity in production is due to the
Santa Luca. No doubt, the potters had the skills presence of the best clay resources in the valley (see
to produce additional forms in order to cover the discussion about site specialization above; see also Arnold
1993).
ANDEAN PAST 10 (2012) - 40
ipate in the production as a whole. Here, the may, therefore, not have been the only interest of
complementarity system is even more complex the valley population. Rather, pottery exchange
because it operates down to the household level. points to a complementary system on a micro-
scale, or valley level, which also covered the
The similarities to the prehistoric production exchange of food products, raw materials, and
model are quite apparent, even though we other goods from different (micro) ecozones and,
should be cautious with analogies over such a therefore, worked horizontally as well as verti-
large time span. However, I have detected conti- cally.
nuity between prehistoric and modern vessel
forms (Gabelmann 2008a), so why should es- An example of such a vertical complementary
tablished economical principles not have sur- model is reported from the Vilcanota Valley in
8
vived changes during different societal orders? It southern Peru in the twentieth century. Karen
is feasible to think of the organization of todays Mohr Chvez (1984-85:183; 1992:79-83) de-
pottery production in the High Valley as a tradi- scribed a complementarity of functional pottery
tion that has been passed down from prehistoric forms that corresponds to two eco-zones within
times. The concept of keeping the product range the region. The inhabitants of the puna villages
down to only one, or a few, categories leaves specialize in the production of cooking and
space for other local production sites, and creates roasting vessels, braziers, and incense burners,
demand on a local level. Additionally, a special- while the potters in Raqchi, in the valley, pro-
ization (or rather a restriction) on particular duce storage and fermentation vessels, as well as
products of the whole range (e.g. vessel forms) for serving and carrying containers. In order to
each household would prevent competition complete the functional vessel inventory, ex-
between the potters of one production site. change between the two eco-zones is necessary.
However, in Santa Luca there is no evidence to Storage vessels (urpus) from the valleys are ex-
date for complementarity of forms on a house- changed in the highlands for quinoa. Therefore,
hold level. Nonetheless, the complementarity the interdependency of this exchange system also
system on a community level in the prehistoric provides a balanced circulation of agrarian (and
High Valley seems a feasible explanation for the other?) goods between the valley bottom and the
existence of the two ceramic types. puna region.
For a better understanding of this principle, For the prehistoric High Valley, we may
it is useful to examine the spatial arrangement of formulate a similar artificial interdependency,
the Formative Period production sites. The four where a limitation of forms in pottery produc-
sites assumed to produce Type 20 are all at the tion triggers exchange of other goods. Archaeo-
southwestern edge of the valley in the foothills of botanical analyses in Santa Luca revealed mainly
the mountain ridge, with altitudes of up to 3500 quinoa as a staple crop, which may have been
m.a.s.l. (Figure 2). There people had easier access cultivated in the foothill areas. At least, the
to resource goods from puna or pre-puna regions swampy valley bottom to the north of Santa
(e.g. camelid products like wool and meat). Santa Luca was probably not suitable for agriculture in
Luca, by contrast, is close to the swampy soils in prehistoric times, but instead for the exploitation
the center of the valley. The exchange of pottery of reed or totora. Reed was used for the roofing
of houses, as forage for animals, and probably as
combustible material for firing facilities. It may
8
The principle itself may not have changed during the well have represented a resource for an intra-
prehistoric and colonial periods, but the degree of
valley exchange system, along with the Type 10
organization and control of production may have shifted.
41 - Gabelmann: Craft Specialization in Formative Cochabamba
vessels. Thus, an artificial interdependency shows bined with product exchange, possibly related to
certain advantages. An eco-zonal comple- an ideological or ritual agrarian cycle (Hagstrum
mentarity with specialized production provides 1989; cf. Vranich 2006).
equitable participation of all valley inhabitants at
an intra-valley level, and guaranteed access to Exchange at an intra-valley or a micro-scale
resource goods from different ecological levels. level is also observable in the neighboring valleys
of Cochabamba. Choroqollo in the Santivaez
Thus, the principle of product or community Valley likewise displays pottery production of
specialization triggered exchange, contact, and large vessel types that were distributed through-
communication. Not only was this principle out the valley (Gabelmann 2001) and even seem
relevant for access to and distribution of specific to have reached the neighboring Capinota Valley
goods, but also in terms of reproduction (Higueras 1996:209; McAndrews and Rivera
(Hagstrum 1989:2-3; Sillar 2000:46). Contact 2007). Here at least two other ceramic types
between different groups and settlements was existed that were possibly produced in the valley.
necessary and it was kept up through artificial Unfortunately, there is insufficient data for
production limitations and the subsequent locating the other production sites, but it is very
9
necessity of exchange. Hence, households likely that the principle of complementarity and,
and/or settlements were neither self-sufficient, hence, specialized communities, is also applicable
nor did they aim for complete self-sufficiency. here.
This is one key mechanism for understanding
the social and economic behavior not only of the The principle seems to be valid first and
Formative Period, but also of later societies. foremost for the Central Valley, based upon the
ceramic variety in Sierra Mokho. Sierra Mokho
What is still unclear is the nature of the presents two ceramic types (Type 2, Type 6;
distributive system that was responsible for an Brockington et al. 1985, 1987) that closely resem-
even diffusion of the products. What would be ble the two types in Santa Luca, or the High
an appropriate distributive system if every settle- Valley, respectively, but are clearly of local pro-
ment acts as an independent productive agent? A duction. The production sites in the Central
redistributive system can be ruled out due to the Valley have not been located either, but in view
lack of a centralized power represented by an of the situation in the High Valley, they may
elite class living in a regional center from where coincide with todays production areas. There-
the goods were allocated. Reciprocity, in compar- fore, Colcapirhua may have been an important
ison, is usually a bilateral exchange between two prehistoric pottery production site for one of the
agents, and cannot be responsible for an even types (Figure 1). The strikingly dense distribution
distribution of goods in a whole valley. Rather, of mounds in the fertile area north of Colca-
exchange must have taken place in various pirhua (Cspedes et al. 1994: map 3) also points
locations with a broad coverage for production to settlements specialized in agriculture. Despite
centers and the population. We can only specu- the as yet incomplete settlement pattern in the
late about prehistoric market and distributive Central Valley, sites existed in the northern
systems but it would not be implausible to think (Chullpapampa) and in the western foothills
of periodical, inter-communal gatherings com- (Gyarmati et al. 1999), which may have fulfilled
special roles and responsibilities in intra-valley
9
interaction because of their easy access to puna
The question arises whether we can conceive of different resources. As a matter of course, the nature of
ethnic groups or ayllus behind the producers of the two
different ceramic types.
the described intra-valley complementarity system
ANDEAN PAST 10 (2012) - 42
in this paper should be examined in more detail Santa Luca has not revealed any highland
10
(and in more valleys). There may be community pottery to date. Conversely, a ceramic type with
specialized activities other than pottery produc- a distinct style of incised decoration appears
tion subsumable in this model. here, which is characteristic of the southeastern
valleys of Aiquile, Mizque, and Khopi (Brocking-
It is, therefore, likely that each valley main- ton et al. 1995). The decoration often shows fine-
tained a similar complementarity system, and line incisions of long narrow triangles, ribbons,
that the valleys acted as autonomous entities or other geometric designs filled with hatching or
concerning the procurement of resources for grids. These motifs are found above all on ce-
daily life. Despite a difference in altitude of 300 ramic tubes and a few sherds (Figures 8 and 17),
meters, the valleys basically provide the same which are of local production in Santa Luca and
agricultural resources. Hence, exchange between belong to a variety of Type 10 ceramics. Incision
the valleys is expected to be more frequent in the was always accomplished either on the com-
case of ritual or exotic goods like sodalite, mala- pletely dry surface after slipping and burnishing,
chite, and marine shell, which came from out- or after firing with a hard-edged tool.
side. Such goods passed the corridor through the
valleys as in the example of Conchupata, Mizque In contrast, incised pottery is rather scarce in
(see below) and may surely be found in several the Central Valley (Brockington et al. 1985,
important sites throughout the valleys. 1995), or even absent at some sites (Cspedes et
al. 1994:13). If incision appears, the motifs are
MACRO-SCALE LEVEL OR SUPRA-REGIONAL rather simple lines, dots or dashes, which also
INTERACTION appear in a small percentage of Wankarani
pottery (Ayala and Uribe 2003; Walter 1966:109,
For all the autonomy given with the procure- 114). Interestingly, Type 20 pottery in Santa
ment of basic local resources, the valleys main- Luca is sometimes decorated with sloppy trian-
tained ties to other regions for access to prestige gles filled with dots or dashes (Figure 17), indi-
or exotic goods from other ecological niches. cating a relationship with the decorative style of
Wankarani pottery was found in the layers of the the Central Valley. These incisions are always
Early Formative phase in Sierra Mokho as well as carried out on the wet or leather-hard surface
in Choroqollo, and points to a certain interac- before slipping and firing and, therefore, contrast
tion of the Central and the Santivaez Valley also in technology with the more complex style
with the southern altiplano. It seems that one of of the southeastern valleys. Furthermore, they
the three Formative Period sites discovered by underline the distinctions between the two
Gyarmati et al. (1999:27) in the western Central ceramic types in Santa Luca.
Valley also exhibits Wankarani pottery, because
the authors note mica temper in the sherds, On this basis, we can delineate two spheres
which is generally an indicator of ceramics from of long-distance interaction that divide the
the southern highlands. Cochabamba valleys. While the westernmost
valleys preferred to interact with the southern
highlands or the Wankarani region, the High
10
A recent survey directed in 2011 by the author in the Valley exhibits closer contact with the lower
Sacaba Valley corroborates the hypothesis of the intra- southeastern valleys. On the one hand, this may
valley complementary systems. Here, two ceramic types be a result of the geographical location of the
were observed that closely resemble the two types of Santa respective valleys and the shorter distance to the
Luca, as well as those of Sierra Mokho, but seem to be
local.
next eco-zone. On the other hand, there is the
43 - Gabelmann: Craft Specialization in Formative Cochabamba
pending question of an affiliation of the two Pacific coast and triggered interaction in both
ceramic types with different ethnic groups that directions. Cochabamba was not only a major
may have inhabited the valleys. The pottery of passageway to the lowlands, but possibly also the
Santa Luca shows a strong influence from the junction for pre-hispanic routes along the Cordil-
Mizque-Aiquile region, based on the particular lera. Artifacts made of gold, sodalite, and basalt
incised motifs, and the vessel forms of Type 10. indicate exchange between Santa Luca and
Can we thus conceive of a population from the Ayopaya or Independencia to the north, while
lower valleys in Santa Luca? This hypothesis may fossils from Apillabamba across the Caine River
be too audacious for the Formative Period, but point to a route to Chuquisaca and the northern
there is a strong relationship for whatever rea- part of Potos department (Norte Potos) further
sons. Certainly, this issue will require detailed south. Sodalite from Independencia, and mala-
analysis in the future. chite, were also found in Conchupata, indicating
a trade route intersection in the valleys, thus
Nevertheless, because of their strategic loca- Sierra Mokho or Santa Luca may have acted as
tion in the corridor between the highlands and a port of trade.
the lowlands, the Cochabamba valleys played an
important role as a port of trade for the long- Certainly, more investigation concerning
distance distribution of goods. Prestige goods long distance interaction and the nature of the
like malachite, sodalite, and Pacific shell entered economic participation of the Cochabamba
the valleys via the highlands to the west, or the valleys is necessary. We do not have a clear
cordillera to the north. Thus, the marine shells picture of prehistoric trade routes in the region,
found as offerings in the Conchupata burial site but there are several Inca road sections known to
dating between 1180 and 800 B.C. (Pereira et al. date. A direct route from highlands to the low-
1992, Table 1) must have passed through the lands leads from Paria to Tapacar and possibly
Cochabamba valleys on their way to Mizque. The to Incarracay (Hyslop 1984:142 ) and from Vacas
shell specimens were assigned to the family of to the east of the High Valley to Pocona, Khopi
Pectenidae coming from the cold Pacific waters and Chimboata (Pereira 1982). The Paria-Tapa-
(Brockington et al. 1986:38). Shells of this family car road probably dates back to the Middle
(Argopecten purpuratus) and others (Trachycardium Horizon (Gutirrez 2005) and the question arises
procerum) have been found in Santa Luca, origi- whether it may have already been in use during
nating from the southern Peruvian or Chilean the Formative Period, because it would have
coast. However, Santa Luca even yielded shell been the most direct route between the Wan-
fragments of Pteriidae, which come from as far as karani region and Cochabamba (cf. Condarco
11
the Peruvian north coast, indicating a consider- 2002). Nevertheless, in view of archaeological
able operating distance of Formative Period trade evidence and the extent of interactions, there
systems. may have been several routes to the highlands
(see Hyslop 1984:139,149). Fragments of Inca
Naturally, there were also goods from the roads have been found in Ayopaya (e.g. Cocapata
lowlands, mostly of an organic kind like wood, to Cotacajes and Independencia), which may
fruits, herbs, and feathers, which were most have led directly to Tapacar and Quillacollo in
interesting to people in the highlands and on the the Central Valley (Pereira 1982), indicating the
possibility of an older route for gold and sodalite
exchange in the Formative Period.
11
The analysis of mollusks from Santa Luca was done by
Victor F. Vsquez Snchez, Director of ARQUEOBIOS,
Trujillo, Per (www.arqueobios.org).
ANDEAN PAST 10 (2012) - 44
INTER-VALLEY CONSOLIDATION AND may have developed in the highlands for goods
INCREASE OF LONG-DISTANCE TRADE from the lowlands (or vice versa?), which finally
triggered an inter-valley consolidation and a
The above-described situation on the surface stronger economic position of the Cochabamba
of Santa Luca features a clear division between Valleys as an entity. At least Santa Luca seems to
the production and domestic sectors, although have benefited from this new situation. Prestige
this may not have been the case in earlier occupa- goods are far more numerous than before. Ob-
tion levels. In the layers of the Middle Formative jects of Pacific shell (Pteriidae) from as far as the
phase dating between 930 and 770 B.C. (Table Peruvian north coast and gold from Ayopaya
1:7; see Figure 10) there was evidence of an indicate the coverage of long distance trade. Raw
activity area for cooking, weaving, and pottery material and artifacts made of malachite and
production, indicating that pottery production sodalite point to a new or increasing activity of
took place in the household realm. Towards the the settlers in the fabrication of beads made of
end of the Middle Formative the site was at least non-local resources. It still remains unclear
partly intentionally restructured, and these layers whether the activity in bead fabrication in Santa
were covered and leveled with a thick stratum of Luca can be traced back to the fact of a labor
rubbish for the new architectural layout (Figure division between potters and non-potters or
10, layer 4). The end of this event dates to the whether we can conceive of a compensating rainy
beginning of the Late Formative I phase around period activity of the potters themselves. How-
540/500 B.C. (Table 1:8, 9; Gabelmann et al. ever, it illustrates an active participation of the
2009). The spatial division and the displacement population in Santa Luca in long-distance trade.
of pottery production outside the household area
is not only evidence of collective labor, but also The eco-zonal complementarity of each valley
presents a clear shift in the socio-economic was not affected by this consolidation. Quite the
pattern with a different organizational level. opposite, it increased with an extension of the
distributive system. Products like local pottery
Around 500 B.C. pottery of Type 10 from could, therefore, cross the valleys as a side effect
Santa Luca also appears in Sierra Mokho in the (e.g. packaging of goods) to the progressive partic-
Central Valley. The appearance of their Tipo 7, ipation in long-distance trade with prestige
which coincides with ceramic Type 10, prompted goods.
the excavators to determine the beginning of
their Middle Formative phase (Brockington et al. The question arises as to whether this new
1987:33). As a result of the shift in the organiza- situation arose due to a shift in the highland
tion of production, the distribution radius of interaction from the southern altiplano (Wan-
Santa Luca increased to more than fifty kilo- karani) towards the Titicaca area, which may
meters, and thus to an exchange of pottery have triggered this development in the valleys.
between the valleys. More precisely, there were Chiripa at the southern tip of Lake Titicaca
closer interactions and relational bonds among gained power around 500 B.C. and extended its
the five valleys along with a dissolution of their economic ties towards the eastern valleys (Lare-
earlier autonomies. caja, Muecas). The corridor via the Cocha-
bamba valleys may have been interesting to
This indicates a concentration of political Chiripa, as were the valleys themselves with their
and economic power in the valleys around 500 attractive climate and fertile arable soils. It would
B.C., which may be seen as an answer to dynam- not be illogical to suppose that an increased
ics created from outside. An increasing demand supra-regional trade system enhanced the socio-
45 - Gabelmann: Craft Specialization in Formative Cochabamba
13
economic environment in the valleys and, conse- ing site of Chullpa Pata. Chullpa Pata indicates
quently, the development of more dominant a longer occupation, because some of the
settlements like Santa Luca in the High Valley mounds yielded Tupuraya sherds (Walter 1966,
and Sierra Mokho in the Central Valley. Brockington et al. 1985), and thus were still
occupied during the Early Intermediate Period.
Contrary to the evidence of Wankarani However, we still lack information about the
interactions, no Chiripa sherds have been discov- transition from the Late Formative phase to the
ered to date in the Cochabamba area, which subsequent Early Intermediate Period.
would confirm such a relationship. However,
there exists a collection of small portable stone During the survey we found the site of Waka-
stelae in the INIAM museum, some of which puuni that is outstanding due to its settlement
closely resemble the Yayamama iconography (cf. type. It is situated upon a terraced mountain
Chvez and Chvez 1975) characteristic of the spur on the southern limits of the survey zone
Late Formative phase in the circum-Titicaca area (Figure 2: VA_59). The pottery is monochrome,
(Figure 18). Unfortunately, neither the sites nor but lacks both the red slip and the burnished
the original contexts of these objects is known, luster of the two ceramic types from Santa Luca
although Ibarra Grasso points out that most of described in this paper. Instead, it exhibits a
12
them came from the area of Cliza (Ibarra 1955, rough sandy surface and a higher percentage of
1958). If this is true and the stone artifacts can zoomorphic applications. Interestingly, this
be assigned to the Late Formative Period, it ceramic type was not found in any of the other
would emphasize the inter-valley consolidation Formative Period sites, making cross-dating
and the powerful position of Santa Luca and difficult. Hilltop sites are reported from the
14
Chullpa Pata after 500 B.C. Capinota Valley (Higueras 1996) and seem to
exist in the Santivaez Valley (Vetters and Sanze-
TRANSITION TO THE EARLY INTERMEDIATE tenea 1996), but the topic has not been fully
PERIOD investigated. The question arises as to whether
we can conceive of a Late Formative II phase
The end of occupation at Santa Luca is not combined with a dislocation of settlements from
clear, because the surface is completely eroded. the valley to hilltops for whatever reasons. On
The youngest radiocarbon sample dates to an the one hand, the location of Wakapuuni
interval between 200 B.C. and 60 A.D. (Table marks the entrance to a corridor to the Caine
1:10). Tupuraya pottery was not found. There- River and a possible trade route to Norte Potos
fore, occupation must have ended, for unknown and the Chuquisaca area and may have served
reasons, during the Late Formative phase. It may
be assumed that a dislocation of the settlers of
Santa Luca took place, possibly to the neighbor- 13
The ceramic material of the upper layers of the 1984
Chullpa Pata excavation (Brockington et al. 1985, 1987)
does display a ceramic type that is very close to Santa Luca
pottery and perhaps represents a posterior occupation.
14
The site of Pirque Alto (CP 11), on a prominent bluff,
12 may now appear as a strategic hilltop site, but this may not
Ibarra Grasso almost never specified finds and sites; have been the case in prehistoric times. Its geographical
therefore when he mentions Cliza, he may refer to situation today is due to post-occupational erosion by the
Chullpa Pata, Cruz Pata, or even Santa Luca, all sites Tapacar River (McAndrews and Rivera 2007). Other
lying within a distance of 1 to 3 kilometers from Cliza (see Formative Period hilltop sites mentioned are CP 15, 43,
Figure 2). and 45 (Higueras 1996:58).
ANDEAN PAST 10 (2012) - 46
economic purposes. On the other hand, we may of monochrome pottery is not necessarily a
assume climate changes forced people to leave consequence of the lack of ability. On the other
the fertile valley bottoms. As a third reason, the hand, we see that the painted Early Intermediate
occupation of strategic hilltop settlements may Period style may not have appeared so suddenly.
point to increased territorial concerns and times It was well known and employed, but on a
of crisis during this phase. For a better under- medium about which we have no knowledge as
standing of this phenomenon, we need more yet. Again, this gives room for interpretation
investigation of monochrome ceramic styles as about the existence of an affiliation with a differ-
well as radiocarbon dating of hilltop sites in the ent ethnic group, who may have slowly provided
future. access to the fashion of the new polychrome
Tupuraya style in Cochabamba. The new style
The appearance of the polychrome Tupuraya was then adopted and ultimately dissolved the
pottery, which characterizes the Early Intermedi- traits of Formative Period style. Naturally, this
ate Period, was hypothetically connected with the hypothesis needs more investigation in the
invasion of ethnic groups from the southeastern future.
valleys of Mizque and Aiquile where the style
seems to have originated (Rydn 1959:117; Another topic should be mentioned concern-
Walter 1966:269). The polychrome style enters ing the polychrome wall plaster. We can assume
the Cochabamba Valleys with no antecedents, that painted decoration, scarce as it is, had a
and appears to replace the monochrome pottery. certain importance and meaning to the inhabit-
However, the question of whether the mono- ants of Santa Luca. Painted interior rooms may
chrome pottery of the Formative Period still point to either a ceremonial structure, or the
existed contemporaneously or disappeared residence of a local chief. The latter may be
completely with the presence of the Tupuraya deduced as a consequent development from the
style has never really been investigated. This also new organization of the pottery production and
holds true with regard to social development, the increased participation in long-distance trade.
that is, the question as to whether the bearers of Though we can discard the idea of a strong
the Tupuraya style pottery replaced the Forma- hierarchy in Santa Luca, it is possible to think of
tive Period society, or whether the Formative a supra-household organization, with the potters
Period people simply adopted the new fashion. themselves acting as rotating chiefs.
goods to create demand and to be compelled to economic structure is absent (just as it does not
enter dynamic interactions with other groups, necessarily imply the existence of a state-like
not only to gain access to specific products, but society when there is evidence of a centralized
also for reasons of reproduction. We may even power). Centralized power may develop if more
detect hints of the existence of a tiered society for efficiency or security in production, or control
the organization of pottery production through over the production of specific items is needed
the evidence of painted wall plaster of a chiefly (Brumfiel and Earle 1987). Pottery, if it is not
building, or ceremonial structure. Actually, the highly symbolic or politically charged, does not
investigations revealed a rather complex socio- require production controlled by an elite. Eco-
economic system representing an innovation for nomic principles are not dependent on central-
the Formative Period of the Cochabamba region. ized power, because a shift in politica organiza-
tion would not change the economic structure.
Coming back to the title of this paper, it is Thus, it seems possible to deduce a certain
not my concern to refer to changes during the traditionalism that conserved such principles
Formative Period, if social change is seen within over a long time span.
a framework of unilinear development from
emerging sedentary to complex stratified societies However, the Formative Period develop-
(cf. Snchez 2003). The creation of a framework ments described in this article can only be pre-
for the classification of historical or prehistorical liminary, because many aspects still are not well
events may be necessary for a better overview, but investigated in the Cochabamba region, as has
at the same time, it is responsible for a simplifica- been stated. There are several open questions,
tion of social processes. This led to a picture of which should be given closer consideration in
the societies in the Cochabamba region as play- the future, but the idea of this paper is to pres-
ing only a marginal role in the Formative Period ent a different view of so-called peripheral
in the Andes. Nevertheless, socio-economic or regions.
socio-political development is not unilinear but
multifaceted. The dynamics of such processes ACKNOWLEDGMENTS
can have diverse expressions, and require a
The Santa Luca Project was possible due to grants
detailed investigation and careful interpretation awarded by the NaFG-Commission of the Free University
of the archaeological material because changes of Berlin and the DAAD (German Academic Exchange
may or may not be visible depending on the Service). I am very grateful to the staff at the INIAM
parameters one applies. museum in Cochababamba, especially David Pereira H.
and Ramn Sanzetenea R. for their great help and support
during the research. Thanks also go to the project staff in
The question does not concern the detection the field and the lab, but first and foremost, to the late
of the chronology of complex societal, economic, Marcos Veizaga Arevalo from Huasa Rancho to whom I
and political development, but an appreciation dedicate this paper. A preliminary version was presented in
of their different facets. Their expression in the 2010 at the seventy-fifth annual meeting of the Society for
American Archaeology in St. Louis, Missouri, at the
archaeological material may change according to sponsored symposium, Archaeology in the Valleys of
shifts of the socio-economic organization. A Cochabamba: A Session in Memory of Craig Morris. For
centralized production is easier to detect than an help with the text I thank Emily Schalk.
independent household production with a lower
output. Thus, wealth accumulations are not as
clearly expressed in the Formative Period as they
are in more centralized societies with elite struc-
tures. This does not mean that an organized
ANDEAN PAST 10 (2012) - 48
Figure 1: The Cochabamba Valley system indicating the Formative Period sites studied
(after Montaos map in Snchez 2008: figure 3.1 and an unpublished map by Jnos Gyarmati).
ANDEAN PAST 10 (2012) - 52
Figure 2: Formative Period settlement pattern in the south-western part of the High Valley
(after a map prepared by the Instituto Geogrfico Militar [IGM], Bolivia).
53 - Gabelmann: Craft Specialization in Formative Cochabamba
Figure 3: The Santa Luca site looking northeast over sherd clusters and eroded firing facilities.
ANDEAN PAST 10 (2012) - 54
Figure 4: The Santa Luca site: topographic map with domestic structures, household rubbish, firing facilities,
and agricultural fields indicated. The left coordinates represent north and the top coordinates represent east.
55 - Gabelmann: Craft Specialization in Formative Cochabamba
Figure 5: The Santa Luca site: circular structure C5 with entrance, clay floor, and central post.
Figure 6: The Santa Luca site. Firing facility H20 at different stages of excavation.
ANDEAN PAST 10 (2012) - 56
Figure 7: Shaping tools from the surface of the Santa Luca site.
57 - Gabelmann: Craft Specialization in Formative Cochabamba
Figure 8:. Five fragments of ceramic tubes or tuyeres from the Santa Luca site (above)
and two specimens from the INIAM museum, Cochabamba, Bolivia (below)
542-4308 from the Colleccin Guilln- Mercado and 03080101 from Cliza.
ANDEAN PAST 10 (2012) - 58
Figure 9: The Santa Luca site: plan of tools and artifacts found on the surface.
59 - Gabelmann: Craft Specialization in Formative Cochabamba
Figure 10: Santa Luca. Trench 5, East Profile. Layers 7 and 11 are Middle Formative phase floor levels;
20 and 22 are Early Formative phase floor levels. Key: (1) gray-brown compact clay; (2) yellowish-gray
compact clay; (3) sherd paving; (4) loose sand with raw and burnt clay; (5) dark brown, soft, ashy sand;
(6) brown brittle clay; (7, 7A) grayish-green clay floor; (8) yellowish-gray argil; (9, 10) brittle clay and sand
with burnt clay and carbon; (11) layered grayish-green clay floor; (12) burnt clay platform; (13, 15) reddish-
gray burnt clay = firing facility 99; (14) not visible on this profile; (16) brown sandy clay; (17) light gray-
brown brittle clay with animal bones; (18) alternating clay and sand layers; (19) compact-to-brittle clay with
carbon; (20) grayish-green clay floor with carbon; (21) grayish-green brittle clay; (22) layered grayish-green clay
floor; (23) compact-to-brittle clay; (24) brown clay with carbon; (25) reddish compact clay, black sherds.
ANDEAN PAST 10 (2012) - 60
Figure 13: Open and closed vessel and rim forms of the ceramic Types 10 and 20
from Santa Luca (not to scale).
63 - Gabelmann: Craft Specialization in Formative Cochabamba
Figure 14: Ceramic Type 10 vessels (not to scale). Note cracks from over-firing on upper left vessel.
ANDEAN PAST 10 (2012) - 64
Figure 15: Santa Luca, distribution of sherds found on the surface expressed in percentages
(for sample sizes, see Table 2).
65 - Gabelmann: Craft Specialization in Formative Cochabamba
SL-01
SL-208-054
Figure 16: XRD results of two ceramic types from Santa Luca (Instituto de Investigaciones Geolgicas y
del Medio Ambiente,Universidad Mayor de San Andres, La Paz)
ANDEAN PAST 10 (2012) - 66
JULIET WIERSEMA
University of Maryland, College Park
and The Metropolitan Museum of Art
colorful murals, expansive plazas, and decorated ongoing excavation projects on Perus north
patios connected by corridors and ramps (Muji- coast (Donnan 1978:82-83; Donnan and Cock
ca et al. 2007; Uceda 1997, 2001b; Uceda and 2002:31-35). These discoveries underscore the
Mujica, editors 1994, 2003). Many of these value that architectural vessels hold for
same architectonic features can be identified on archaeologists, as they preserve the complete
Moche architectural vessels, signaling an form of significant structures and point us to
important correlation between small scale and areas of key importance in Moche ceremonial
full scale architecture. precincts.
most of the vessels in the corpus exhibit stylistic tectural styles or types could be discerned.
traits attributable to the southern Moche sphere Within these ten styles, the three most often
(south of the Pampa de Paijn).6 The fact that discussed in Moche literature, the Architectural
architectural vessels are relatively rare in the Complex (Type 1, Figures 3-5), the Open
greater Moche fine ware vessel assemblage Gabled Type (Type 2, Figure 9), and the Open
(approximately 1 in 500) suggests these objects Shed Roof Type (Type 3, Figure 10), had
had been the prerogative of elite or high status minimal representation in the architectural
individuals.7 Corroborating this hypothesis is the vessel corpus.8 Instead, the architectonic style
fact that, when discovered archaeologically, with the greatest representation was a simple
Moche architectural vessels occur as mortuary single-room structure (Figures 1, 2), a form
offerings accompanying important Moche which, until now, has been largely overlooked in
deceased who are buried within ceremonial Moche studies. This single-room type, the
architecture. Enclosed Gabled Type (EGT) occupies over one
third (34 percent) of the total Moche
Type No. in Corpus % in Corpus architectural vessel corpus (Table 3). This
Proto-Moche (PM) 26 15%
percentage is especially significant, given that
Moche I-II (M I-II) 13 8% EGTs are largely limited to Moche phases IV
Moche III (M III) 22 13%
Moche IV (M IV) 76 45% and V, where they comprise more than half of
Moche V (M V) 28 17% both samples (Table 2). EGTs notably
Late Moche (LM) 4 2%
overshadow all other architectural types in the
Total 169 100%
Moche architectural vessel corpus, suggesting
that this particular architectural style (and its
Table 1. Distribution of vessels in the Moche full scale counterpart) held symbolic importance
architectural corpus by phase.
for Moche ideology, or at least the Moche
As part of this study, I devised a detailed ideology reflected in Moche IV and V pottery.
architectural typology, building upon Chris-
topher Donnans earlier work (1978:79-81). DIAGNOSTIC FEATURES OF THE ENCLOSED
This new architectural typology revealed some GABLED TYPE AND THE CHALLENGES
intriguing patterns (see Table 2). For example, PRESENTED BY ANCIENT MOCHE
despite the significant quantity of architectural ARCHITECTURE
vessels in the corpus, only ten distinct archi-
The Enclosed Gabled Type is easily identified by
its consistent suite of diagnostic traits. It is
high status individuals at El Castillo de Santa (Chap- enclosed and square in plan with a central
delaine et al. 2005:25-28, 32, figure 10); Huaca de la Luna entrance and a gabled roof (Figures 1, 2). Its
(Chauchat and Gutirrez 2005:120); Huaca Cao Viejo roof is supported by one or two beams, indicated
(Mujica et al. 2007:212, 215); and San Jos de Moro
(Mauricio and Castro 2008: figures 33 and 34; Rengifo et
by a raised circle or circles on the exterior sides
al. 2008:129-131, figure 38).
6 8
Examples from the northern Moche sphere have been While rare in the corpus (six percent), Architectural
excavated at the site of San Jos de Moro in the Jequete- Complex vessels offer a type of architectural key that not
peque Valley (Mauricio and Castro 2008: figures 33, 34; only corroborates the existence of different structural
Rengifo et al. 2008:129-131, figure 38). forms and features, but also pinpoints their approximate
location within Moche monumental architecture. More
7
This estimate is based on vessels held in the Museo frequently discussed in Moche literature is the Open
Larco, where, of an approximate 15,000 Moche vessels, Gabled Type, which appears often in Moche fineline, but
only 30 are architectonic. comprises a small percentage of the sculpted corpus.
ANDEAN PAST 10 (2012) - 72
Architectural Types Moche Phase # in % in viewer by the potter. In Figure 2, we see a series
(no. vessels) Corpus Corpus
of painted step motifs on the chamber which
Architectural Complex PM (3), I-II (6),
III (1) 10 6%
separate the three-dimensional building above
Open Gable Type IV (5), LM (3) 8 5% from the platform levels below, conveyed
Open Shed Roof Type I-II (1), IV (3),
V (4), LM (1) 9 5%
through horizontal bands of alternating red and
Open Three Wall, Shed Roof PM (1), I-II (1) white. How might these two-dimensional details
III (1) 3 2%
Open Three Wall, Gabled Roof PM (1), III (4) 5 3% help us to identify this particular architectural
Open Front Gable, Back Shed
Roof III (2), IV (3) 5 3%
type in Moche full scale architecture?
Enclosed Front Gable or
Gable, Back Shed Roof PM (12), III (3) 15 9%
Enclosed Front Flat, Back Shed Moche IV (76 vessels total) No. %
Roof PM (4), I-II (2) 6 3%
Enclosed Gabled Type III (3), IV (39), 57 34% With step-shaped roof comb 28 37%
V (15) With step and tumi roof comb 4 5%
Miscellaneous PM (2), I-II (1), Without roof embellishments 7 9%
III (3) 6 3%
Motif Vessels (step motif, step I-II (2), III (5), Total Moche IV 39 51%
and wave spiral throne, etc.) IV (26), V (9) 42 25%
Moche V (28 vessels total) No. %
Unclassifiable PM (3) 3 2%
With step-shaped roof comb 7 25%
Total 169 100%
With step and tumi roof comb 2 7%
Without roof embellishments 6 21%
Table 2. Distribution of types in the Moche
Total Moche V 15 53%
architectural vessel corpus.
Table 4. Breakdown of EGT variants in Moche
Architectural Moche Phase No. in % in % in % in phases IV and V.
Type (No./phase) Corpus Corpus M IV MV
be based on the traits that are preserved in the architectural vessels was manipulated by Moche
architecture, that is, those of a structure that is potters to provide a detailed picture of
enclosed, and square in plan, with a central represented architectural space. Moche potters
entrance. Such structures are, to my knowledge, worked creatively within the restrictions of their
unknown in Moche excavated architecture chosen medium, the ceramic vessel, using both
except in the upper areas of Huaca de la Luna painted imagery and sculpted forms together to
where a small structure fitting this description is communicate identifiable aspects of specific
found in Plaza 3c. We will return to this structures found in the Moche ceremonial
structure later in this article. precinct, or huaca.
Further complicating our ability to compare In order to see architectural vessels as the
Moche architectural vessels with Moche Moche likely saw them, it is necessary to devise
ceremonial architecture is that, while the vessel a two-dimensional floor plan from the vessel
reflects a single construction episode, excavated itself, incorporating both two- and three-
Moche monumental remains present a palimpsest dimensional elements. To illustrate this, let us
of architectural episodes. Today, at sites such as work from the vessel depicted in Figure 2. If we
Huaca de la Luna and Huaca Cao Viejo, translate the sculpted and painted details from
structures and features corresponding to Figure 2 into a map or plan, we end up with
different chronological phases appear side by something comparable to the drawing in Figure
side, skewing our understanding of architectural 13. This plan reveals a series of walls
space as the Moche knew it. On the other hand, surrounding the three-dimensional or modeled
the destructive nature of archaeology can deny architectural structure at center. Just outside
us architectonic details critical to our the innermost wall, in direct alignment with the
understanding of the space. For example, the structures entrance, is a small projection,
outer U-shaped wall surrounding Recinto I, possibly an altar or a ramp. The outermost wall
Plaza 3c at Huaca de la Luna is not visible in is adorned with a series of step motifs. Through
Figure 11 because it was dismantled as part of the inclusion of these walls, the potter suggests
the archaeological process. The excavation that the EGT is set apart from the rest of the
plans of this plaza, however, relay that these two complex. Horizontal bands on the chamber
features, a U-shaped wall and an enclosed suggest platform levels, indicating that this
architectural structure, were contemporary structure is additionally situated in an elevated
(Figure 12). Because of the complexity area of the complex or precinct. One final
presented by construction episodes, an consideration is the spherical shape of the
excavation plan becomes much more useful in chamber which results in circular walls
drawing comparisons than the site itself, because surrounding a square structure. Making allow-
excavation plans provide us with a view of the ances for the fact that Moche artists were
structures and features that pertained to the working on globular vessels and taking into
same construction moment. account that the Moche did not appear to build
circular structures, we must translate these
CONVERTING TWO- AND THREE- circular forms to rectilinear ones (Figure 14).
DIMENSIONAL IMAGERY INTO A SINGLE With this newly created floor plan and a more
ARCHITECTURAL SPACE nuanced understanding of the Enclosed Gabled
Types location in the complex, we can more
Remaining to be explored is how two- and easily seek a full scale correlate for this
three-dimensional imagery on Moche architectural type in the Moche archaeological
ANDEAN PAST 10 (2012) - 74
record. Before we do, however, let us first the complex.10 This information suggests that
examine another type of architectural vessel the EGT was a privileged, and somewhat
that aids us in our visual interpretation of the restricted structure within a certain type of
Enclosed Gabled Type. Moche ceremonial precinct. Its singular status is
corroborated by the plethora of EGT examples
CONTEXTUAL DETAILS PROVIDED BY in the Moche architectural vessel corpus, as we
MOCHE ARCHITECTURAL VESSELS will see.
Within the Moche architectural vessel Of the 76 vessels in the Moche IV sample,
corpus is another vessel type, referred to as the 39 of them are Enclosed Gabled Types (EGTs).
Architectural Complex (Figures 3, 4, 5). While Despite this large number, no two EGT vessels
rare in the corpus, Architectural Complex are identical.11 Each is differentiated by its
vessels are critical to our understanding of chamber shape (spherical, cylindrical, etc.), the
single-room structures, including the EGT. fineline imagery decorating the chamber, and
Architectural Complex vessels unite, in a single the pictorial details flanking the structures
object, distinct architectural types and entrance (Figures 13, 16, 17). While these
architectonic features which are conveyed elements make each vessel visually distinct, all
individually, or in isolation, on other vessels. As iterations reinforce the same message: the EGT,
such, Architectural Complex vessels provide us while depicted in isolation, is conceived as a
with information about the ways distinct free-standing structure occupying an important
structures and features likely interacted within place within the Moche ceremonial complex.
a greater, often multi-tiered space. In addition
to corroborating the existence of different To illustrate this point, let us look at the
structural forms and features, Architectural vessels in Figures 13, 16, 17, where three
Complex vessels also indicate the approximate different Moche IV Enclosed Gabled Type
locations of rooms, corridors, and architectonic vessels are presented. Each vessel exhibits the
features within Moche monumental diagnostic traits of the EGTs: an enclosed
architecture. structure, square in plan, with a central
entrance and a gabled roof decorated with step-
A handful of Architectural Complex vessels shaped roof combs. While the chamber shapes
incorporate representations of the Enclosed are distinct for each vessel, all display a series of
Gabled Type, easily identified by its enclosed painted step motifs just below the sculpted
square form, central entrance, and gabled roof structure. These demarcate the EGT from the
adorned with step shaped roof combs (Figures 4, rest of the schematically-rendered precinct.
5). In Figure 4, the EGT is taller than the
structures surrounding it and is differentiated by
its red roof combs. Its location, at back left, 10
While not visible in the photograph, the EGT in Figure
makes it the least accessible structure in the 4 has a central entrance which serves as the opening for the
miniature complex. In another Architectural whistling mechanism housed inside it.
Complex vessel, the EGTs importance is 11
suggested by its exaggerated size (Figure 5). An interesting exception is a vessel housed in the Staat-
liche Museen in Berlin (Figure 15), which exhibits
Here again, the EGT is also at the back left of identical architectonic features to the vessel in the
American Museum of Natural History. The Berlin vessel
(V A 18281) measures 13.7 by 14.2 by 20.5 centimeters.
The roof has been erroneously reconstructed (a roof comb
has been replaced by what appears to be a sculpted frog).
75 - Wiersema: Moche Architectural Vessels
Below these step motifs are horizontal bands step motifs flank a nude figure bound to a rack,
indicative of platform levels, which place the placing him within sacred architectural space.
EGT in an upper area of a tiered complex. Our understanding of the sculpted figure above
Through these vessels (that eliminate all is aided by the fineline illustration on the vessel
subsidiary structures and give primacy of place chamber, which depicts four females with short
to the EGT), potters were able to convey in a hair and long tunics who gaze upwards toward
greatly simplified form the same message found the sculpted figure. Their open palms receive
on Architectural Complex vessels. the drops of blood (a metaphor for rain?) which
fall from the bound figures severed neck. These
The appearance of a series of step motifs on examples from the Moche artistic record make
these vessels is significant (Figure 13, 16, 17). intriguing and undeniable, connections between
Step motifs in series appear prominently in other step motifs, sacred space, and ritual activity.
Moche architectural vessels. For example, the
Architectural Complex vessel in Figure 3 Interestingly enough, a connection between
exhibits a series of painted step motifs which step motifs and sacred space is also supported by
line the ramp or parapet leading from the main the archaeological record. At Huaca de la Luna,
plaza (at bottom) to the upper, more restricted a series of sculpted step motifs (made of slipped
areas of the complex. This same information is adobe) have been excavated within the
conveyed in another Architectural Complex ceremonial precinct where they line the ramp or
vessel published by Elas Mujica (Mujica et al. parapet leading from the main plaza below
2007:93). Architectural Complex vessels make (Plaza 1) to the precincts upper reaches (Plaza
explicit what is merely implied by EGT vessels, 3).12 It appears, therefore, that in both small
namely that step motifs in series, when depicted scale and full scale architecture, step motifs in
in an architectonic setting, delineate sacred or series demarcate and offset important ritual
ceremonial space. space.
This same idea is also expressed in the PAINTED IMAGERY: A CLUE TO FUNCTION?
Moche fineline scene of the Sacrifice Ceremony,
where a series of step motifs delineate and Continuing our visual analysis of the
separate two distinct but related ritual activities: Enclosed Gabled Type (EGT), the final compo-
the ritual bleeding of bound captives on the nent to be considered is the imagery on the
bottom register, and the presentation of the miniature architectonic structure itself. As we
goblet to the Warrior Priest on the upper have seen, EGT vessels communicate
register (Figure 18). Three of these figures, information through imagery and motifs on both
identified as Figure B, Figure C, and Figure D the chamber and the roof. There is additional
(Donnan 2010), process along a route marked information conveyed on the structures faade,
by step motifs in series. This illustration, as well (Figures 1, 2, 13, 16). The painted motifs
together with the examples from the Moche on either side of the structures entrance
architectural vessel corpus, suggests that step provide us with clues about the likely function
motifs in series (when associated with of this miniature structures full scale
architectonic settings) served to cordon off counterpart.
sacred space associated with propitiatory
activities. This idea is also communicated by
another vessel, housed in the Museo Larco in 12
These motifs are associated with Building C, or the
Lima. In this image (Figure 19), two sculpted middle construction phase (Uceda and Tufinio 2003:
figure 20.15).
ANDEAN PAST 10 (2012) - 76
remain today due to heavy erosion in this area, At this point, Recinto I, Plaza 3c seems to be
the northern and western perimeter walls the most likely match for the Enclosed Gabled
maintain evidence of sculpted adobe step motifs Type. If we compare the plan view of Recinto I,
(Montoya 1997:64; Tufinio 2004:102, 105, Plaza 3c (Figure 12) with the floor plan adapted
116). from the EGT (Figure 23, see also Figure 13),
striking similarities emerge. Both plans reveal
Of additional interest is Recinto Is elaborate enclosed square structures with central
decorative program, underscoring its entrances surrounded by U-shaped walls.
importance. Adorning the buildings faade is Directly in front of both entrances is a small
relief imagery depicting feline-human pairs, ramp or altar. Given these shared traits, it
where a feline attacks a human figure wearing appears Recinto I, Plaza 3c presents us with a
the short hairstyle associated with captives.15 full scale counterpart for the Enclosed Gabled
The sides and back of Recinto I were painted in Type.
three separate episodes or renovation cycles.
Imagery associated with the earliest decorative The ubiquity of EGT structures in the
phase depicted stylized catfish (pez life) (Tufinio Moche architectural vessel corpus would lead us
2004:113, figures 110 and 111; Uceda and to believe that enclosed square structures with
Tufinio 2003:192-195; lmina 20.2a and 20.2b), central entrances were common forms in Moche
a motif associated with the onset of the rainy ritual architecture. This, however, is not the
season.16 case. At Huaca Cao Viejo, Chicama Valley, and
at Guadalupito, Santa Valley, archaeologists
Archaeologists have noted the presence of have confirmed that, as yet, no square structures
two additional enclosed free-standing structures with central entrances have been found
(Recinto I; Recinto II, Plaza 3b) located in an (Antonio Murga, personal communication 2008
area adjacent to Plaza 3c. Both of these were and Victor Pimentel, personal communication
built after Recinto I, Plaza 3c fell into disuse and 2009). Much more frequent in Moche ritual
was interred. While the similarity of these architecture are rectangular structures with
structures to Recinto I, Plaza 3c might suggest lateral or off-center entrances, excavated at
that they, too, were full scale counterparts for both Huaca de la Luna and Huaca Cao Viejo
the Enclosed Gabled Type, I argue that, because (see for example Mujica et al. 2007:116-119,
Recinto I and Recinto II, Plaza 3b are more 122-129, and 162-169 and Uceda 2001b).
rectangular in plan, and have off-center entry
ways, they are not the structures referenced by THE GREATER ARCHAEOLOGICAL
EGTs in the Moche architectural vessel corpus. CONTEXTS OF EGTS IN MOCHE FULL SCALE
Furthermore, neither of these chronologically ARCHITECTURE
later structures from Plaza 3b bears the
elaborate decoration of Recinto I, Plaza 3c. Now that we have established Recinto I,
Plaza 3c, as a likely analog for the Enclosed
Gabled Type, let us briefly consider the greater
15
This structure is discussed in Tufinio (2004:113, figures archaeological contexts for this full scale
110, 111) and Uceda and Tufinio (2003:192-195; lminas structure. Recinto I, located in an upper,
20.2a, 20.2b).
restricted area of Huaca de la Luna, is situated
16
If the buildings function remained the same over time, along a circumscribed route (Figure 24) which
it may indicate that these different images, stylized catfish begins in the main plaza below (Plaza 1) and
and felines attacking human figures, are linked, or perhaps ascends toward Plaza 3 via a zig-zag corridor.
equivalent in meaning.
ANDEAN PAST 10 (2012) - 78
This corridor is adorned with a series of sculpted complexs adobe walls (Figure 24). It was in
step motifs (Figure 25), similar to those we have Plaza 3a, in the context of this rock outcrop,
identified in the Moche architectural vessel that Steve Bourget unearthed human remains of
corpus (Figure 3; see also Mujica et al. 2007:93). more than 70 individuals who had been
This trajectory is interrupted at specific points sacrificed in five distinct rituals (Bourget
by small, elaborately decorated architectural 2001:96; see also Verano 1998). The
structures which may have anchored ritual spectacular nature of this sacrifice has
activity, while also shifting the direction of the overshadowed other finds of this type, including
procession. As I have argued elsewhere, these the sacrificial human remains unearthed in Plaza
three-dimensional structures (and certainly the 3c.
one located in the southeast corner of the main
plaza) appear to interrupt and subsequently alter In Plaza 3c, just to the east of Recinto I,
the direction of the two-dimensional poly- archaeologists also discovered evidence for
chrome procession depicted on the plaza walls, human sacrificial activity. They interpreted this
suggesting that these structures acted as stops activity as distinct from that found in Plaza 3a,
along a directed route, similar to the stations of however (Tufinio 2006:63; Uceda 2000:93-95;
the cross in Christian religion (Wiersema 2010: Uceda and Tufinio 2003:192-195; Verano et al.
29, 166). 2007). While skeletons in Plaza 3a bore
evidence of torture and dismemberment,
Plaza 3c, housing Recinto I, formed part of skeletons in Plaza 3c were decapitated,
Plaza 3, the sector of Huaca de la Luna found in dismembered, and had had their flesh
closest proximity to the imposing one-peaked intentionally removed (Tufinio 2004:111;
mountain, Cerro Blanco.17 Plaza 3 has been Uceda and Tufinio 2003:195, 222; see also
subdivided by archaeologists into three units, Verano et al. 2007).18 This distinction suggests
Plaza 3a, 3b, 3c. These areas were operational at that Recinto I, Plaza 3c, was devoted to
different moments in time, with Plaza 3c in use sacrificial activity of a different nature (Tufinio
first (Uceda and Tufinio 2003:192-195). Plaza 2006:59). Additionally, these sacrificial deposits
3c, associated with the middle construction spoke of not one, but several episodes which
phase, or building C, was later interred to occurred over a long period of time (Tufinio
facilitate the construction of Plaza 3a (Tufinio 2004:112). According to the excavator, Recinto
2002:57, 2004:117), one of the last construction I, Plaza 3c was directly related to these sacrificial
phases at the site (Bourget 2001: 96). Therefore, deposits and also experienced a prolonged
Plaza 3c and Plaza 3a represent distinct period of use (ibid.). While it is not clear that all
structures from chronologically separate periods. individuals found in the context of Recinto I,
Plaza 3c is associated with Moche III ceramic Plaza 3c were sacrificed in front of this structure,
artifacts and Plaza 3a is associated with Moche we can ascertain that the characteristics of this
IV artifacts. particular ritual included decapitation and the
removal of flesh. Human remains were, at some
Plaza 3a features a natural rock outcrop point, deposited near this small enclosed free-
which was intentionally incorporated into the
17
Many Moche huacas were intentionally constructed at 18
I am grateful to an anonymous reviewer for pointing out
the foot of these cerros, visually and symbolically connecting that the skeletons in Plaza 3c formed a deposit of human
the huaca to this natural feature associated with supernatu- remains, i.e. individuals who had been sacrificed or
ral potency. processed elsewhere and were later deposited in Plaza 3c.
79 - Wiersema: Moche Architectural Vessels
standing structure which sat in the shadow of a important subset of this group depicts a one-
one-peaked mountain. peaked mountain.21 The one-peaked vessels, in
particular, contain information relevant to our
This suite of traits which have been discussion.
discovered archaeologically, decapitation in the
context of an enclosed gabled structure in the One-peaked mountain sacrifice scenes differ
shadow of a one-peaked mountain, are also from multi-peaked scenes in a few important
found as an iconographic set in Moche ceramic ways. First, the center of the mountain houses a
art. This thematic group is known as the form resembling the faade of the Enclosed
Mountain Scene of Human Sacrifice. In the Gabled Type (Figures 26, 27).22
next section, we will focus specifically on a
subset of this theme which depicts sacrifice Second, Dead Person appears directly
(specifically decapitation) in the context of an beneath this structure, his head separated from
enclosed gabled structure. Information proffered his body. In other words, Dead Person in one-
by this subset of vessels lends further support to peaked mountain sacrifice scenes lies
a connection between architectural structures of decapitated in the context of an EGT structure.
the Enclosed Gabled Type and human sacrificial Also particular to one-peaked depictions are jars
activity. (rendered in low relief) which flank the
entrance of the enclosed gabled edifice, though
THE ENCLOSED GABLED TYPE IN THE the way they are rendered suggests they sit just
CONTEXT OF A ONE-PEAKED MOUNTAIN below the structure (Figures 26, 27). Jars are
another motif associated with the Enclosed
Within the greater Moche ceramic vessel Gabled Type, as we see from an EGT vessel in
corpus is a thematic group discussed as the the Museo Larco, where ritual jars appear on
Mountain Scene of Human Sacrifice (Zighel- either side of the structures central entrance
boim 1995).19 Vessels in this thematic category (Figure 22).
share two key elements: a sculpted
representation of a mountain, and a modeled The one-peaked mountain scene provides
figure draped face-down over the central peak, valuable information as it unites, on a single
referred to as Bent Person. Also appearing in vessel, the Enclosed Gabled Type, human
these scenes is a figure lying akimbo at the base sacrifice via decapitation, and a towering one-
of the mountain, referred to as Dead Person peaked mountain. These same elements occur
(Zighelboim 1995:153-154).20 While mountain together in the archaeological record at Huaca
sacrifice scenes are most often multi-peaked, an
21
Two one-peaked mountain scene vessels are housed in
the Museo Larco (Figures 26, 27), two are published in
19
Bourget has also discussed these vessels in connection Zighelboim (1995: figures 2b, 2c), and one is housed in a
with Cerro Blanco, Huaca de la Luna, and human sacrifice private collection. Both one-peaked mountain sacrifice
(1994). The link between mountains and Moche sacrifice scenes from the Museo Larco are Moche IV, as is evident
is also noted by Donnan (1978:144-148). Zighelboim by both their long, straight spouts and their exaggerated
proposes that the one-peaked depictions reference Cerro stirrup-shaped handles.
Blanco (1995:157).
22
Zighelboim acknowledges the form, describing it as
20
One of the most frequently published examples of the terraces flanked on the top, with two half-pyramids
Mountain Scene of Human Sacrifice depicts a five-peaked facing each other . . . (1995:64, 81, figure 2.b), a reason-
vessel where Bent Person is connected to Dead Person by able description given the scarce attention given to EGTs
a river or a stream of blood (Donnan 1978: figure 2.25). in the past.
ANDEAN PAST 10 (2012) - 80
de la Luna in Plaza 3c, where an EGT (found in chronological development from early to late
association with human sacrificial activity) was (Castillo and Donnan 1994:143-144; Castillo
constructed in the shadow of a one-peaked and Uceda 2008:710-714). Nonetheless, Moche
mountain (Cerro Blanco).23 Imagery found on IV and V spouts are still viewed as reflecting the
one-peaked mountain vessels, then, provides us later phases of Moche occupation in the
with visual evidence supporting the hypothesis southern Moche sphere, from the Chicama
that the EGT represented a specific structure, Valley southwards. Decades of excavation at
most likely Recinto I, Plaza 3c. That a ceramic Huaca de la Luna have turned up substantial
architectural vessel mimicked the form of a full amounts of Moche IV ceramic ware, indicating
scale structure supports Donnans early proposal that Moche IV was an important ceramic style
that everything depicted in Moche art is for this site in its later phases of occupation. As
eventually corroborated in the archaeological such, it makes sense that fine ware ceramic
record (1978:82). This correspondence vessels mimicking the form of a symbolically
underscores the importance of architectural important structure at Huaca de la Luna would
vessels to our continued understanding of be fashioned with Moche IV spout types. What
Moche ritual architecture. Through a remains perplexing, however, is that Recinto I,
methodical visual deconstruction of Moche Plaza 3c and its related contexts were associated
sculpted vessels, used together with with Moche III rather than Moche IV ceramics.
archaeological data, scholars can arrive at Might this suggest that the EGT reflected the
su r p r i si ng l y deta i l ed a nd nu a nced memory of an earlier ritual structure which had
interpretations for Moche art and archaeology. been interred by the time Moche IV ceramics
began to appear? Because none of the seventy-
IDEOLOGICAL SIGNIFICANCE OF THE six EGT vessels in the corpus was excavated
ENCLOSED GABLED TYPE scientifically, we are left to speculate.
We are left to question why the Enclosed I have suggested here and elsewhere that
Gabled Type, one of ten architectural types in architectural vessels represent in miniature full
the vessel corpus, and an architectonic form scale structures of sacred significance to the
uncommon in the Moche archaeological record, Moche (Wiersema 2010). If true, all of the
came to dominate the Moche IV architectural architectural types documented through cera-
vessel sample. What might this architectonic mic vessels would have been ideologically
type have symbolized for Moche ideology and important. However, the EGT as a likely analog
ritual during this chronologically late Moche for a structure like Recinto I, Plaza 3c, located
phase? in a restricted area, decorated with elaborate
and explicit imagery, and associated with human
At present, our understanding of spout type sacrificial activity, would have taken on special
and its relationship to Moche chronology is status or symbolism.
being revised. As a result, Moche phases I-II, III,
IV, and V are no longer seen as reflecting a Scholars believe that the Moche undertook
human sacrifice for political, social, and
ideological reasons (Uceda 2000:92, 99). In
23
Acts of human sacrifice in the context of a sacred
mountain were likely undertaken in an effort to venerate
addition, such activity strengthened the sacred
and appease mountain deities, forces which the Moche bond between powerful individuals and deities
depended upon for water and agricultural sustenance. For (Tufinio 2008:461). Sacrifice appears to have
more on this, see Donnan (1978:148), Hocquenghem taken on particular importance in Moche IV, a
(1987:80), and Zighelboim (1995:171-173).
81 - Wiersema: Moche Architectural Vessels
phase associated with the height of Moche structure of some consequence. It is, however,
power and influence in the southern Moche the overwhelming presence of this architectonic
valleys.24 This fact is corroborated by Moche art, type in the artistic record that signals its
as well as by the archaeological record.25 For important ideological role during Moche IV.27
example, in Moche IV fineline painting, certain
themes proliferate and become codified, most CONCLUSIONS
notably images of warfare, capture, arraignment,
bloodshed, captive sacrifice, and the This study, focused on a single architectural
presentation of a goblet to an elaborately type in the Moche vessel corpus, aims to shift
dressed Moche priest or lord (Donnan and our perception of ancient Andean ceramic
McClelland 1999:136).26 Evidence of ritualized vessels from decorative objects to decoding
human sacrifice on a significant scale at Huaca objects. Here I have proposed that Moche
de la Luna is apparent in the archaeological architectural vessels, once deciphered, aid us in
record at this time, culminating in Moche IV our understanding of excavated structures and
with activity in Plaza 3a. It is in this same phase direct us to full scale structures of great ritual
that the EGT (and its associated imagery significance. Our ability to identify
referencing human blood-shed) first emerges in correspondence between architectural vessels
force in the architectural vessel corpus, and Moche architectural remains suggests that
overshadowing all other architectural types. these often abstracted vessels present us with
specific and explicit depictions of full scale
The predominance of the Enclosed Gabled Moche architectonic structures and features.
Type in the Moche IV sample suggests it Contexts provided by Moche art and
emblematized a structure of key political and corroborated in the Moche archaeological
ideological significance. As we have seen, the record suggests that architectural vessels mimic
archaeological record confirms the EGT was a key aspects of full scale structures which were
intimately tied to ideology, power, and religious
ritual. As such, these objects act as key tools in
24
While Moche chronology is presently being reconsid- our quest to interpret and identify ceremonial
ered, it is largely agreed that Moche IV reflects the height architecture built by this fascinating, but often
of Moche power and influence in the southern valleys, enigmatic, pre-Hispanic Andean society.
especially evident at Huaca de la Luna. This power was
likely demonstrated through large spectacles of ritual ACKNOWLEDGEMENTS
violence. See for example Swensons discussion of this for
the Jequetepeque Valley (2003, 2008, in press). I am grateful to Christopher Donnan for carefully and
critically reading this article in manuscript form and for
25
While the construction and ritual use of Huaca de la his feedback on an earlier version of this work. I thank
Luna spanned more than a century (450-650 A.D.; Uceda Santiago Uceda, Moises Tufinio, and Jorge Gamboa for
2000:92), evidence for human sacrifice at this site is familiarizing me with the architecture of Huaca de la
especially strong in the later phases of occupation, corre- Luna, especially Plaza 3, as well as for their generous
sponding to Moche IV ceramics. assistance and input. I am also indebted to the museums
and collections whose directors and expert staff facilitated
26
This interrelated and sequential complex of images,
referred to as the Warrior Narrative, is believed to have
occurred within circumscribed architectural space, most
likely the Moche ceremonial precinct (Donnan and 27
In Moche V, the EGT continues to have a significant
McClelland 1999:130-136). Of the many events comprising presence, but becomes more stylized in its representation,
Moche religious ritual as it is depicted in the Warrior decorated with geometric motifs rather than pictorial ones.
Narrative, human sacrifice is viewed as the most crucial As of yet, there is no evidence of mass ritualized human
(Donnan 2010). sacrifice found in Moche V contexts.
ANDEAN PAST 10 (2012) - 82
this investigation: the Museo Nacional de Antropologa, Prou. Ph.D. dissertation, Universit de Montral.
Arqueologa, e Historia del Per (Carmen Arellano, 2001 Rituals of Sacrifice: Its Practice at Huaca de la
Dante Cassaretto, Maritza Perez, Ana Flores, ngel Luna and its Representation in Moche
Ludena, Csar Crdoba, Paco Merino, Rosa Martnez, Iconography. In Moche Art and Archaeology in
Wilfredo Cordero, Milano Trejo, Carlos Mansilla, and Ancient Peru. Studies in the History of Art 63,
Dimitri Manga), the Museo Arqueolgico Rafael Larco edited by Joanne Pillsbury, pp. 89-110.
Herrera in Lima (Andrs lvarez Caldern, Carlos Vels- Washington D.C.: National Gallery of Art.
quez, Sr. Huarache, Ulla Holmquist, Rocio Aguilar, and Campana, Cristbal
Paloma Manrique), the American Museum of Natural 1983 La vivienda mochica. Trujillo: Varese, S.A.
History (Sumru Aricanli, Paul Beelitz, and John Hansen), Castillo, Luis Jaime and Christopher B. Donnan
Manuela Fischer at the Staatliche Museen zu Berlin- 1994 Los mochicas del norte y los mochicas del sur:
Preuischer Kulturbesitz, Ethnologisches Museum, and Una perspectiva desde el valle del Jequetepeque.
Ramiro Matos and Emily Kaplan at the Smithsonians In Vics, edited by Krzysztof Makowski,
National Museum of the American Indian. Generous Christopher B. Donnan, Ivan Amaro Bullon,
financial support came from the Smithsonian Institution, Luis Jaime Castillo, Magdelena Diez Canseco,
Fulbright IIE, the University of Maryland, and the Cosmos Otto Elespuru Revoredo, and Juan Antonio
Club Foundation. The Pre-Columbian Society, Morro Mena, pp. 143-181. Lima: Banco de
Washington D.C., provided an opportunity to develop Crdito del Per.
these ideas for a society lecture. I thank the society for Castillo, Luis Jaime and Santiago Uceda
their enthusiasm and interest and in particular Rosemary 2008 The Mochicas. In Handbook of South American
Lyon and Lucy Wilson. Later versions of this work were Archaeology, edited by Helaine Silverman and
presented to the Department of Anthropology, Harvard William H. Isbell, pp. 707-729. New York:
University, thanks to an invitation by Gary Urton and Springer Science and Business Media, LLC.
Adam Stack, and at the 28th Northeast Conference on Chapdelaine, Claude, Vctor Pimentel, and Jorge Gamboa
Andean Archaeology and Ethnohistory, SUNY New Paltz, 2005 Contextos funerarios moche del sitio El Castillo
organized by Ken Nystrom. I extend special thanks to de Santa: Una primera aproximacin. In Muerte
Andean Pasts anonymous reviewers who took the time to y evidencias funerarias en los Andes centrales:
read this manuscript and provide their insightful and Avances y perspectivas; Actas del III Seminario de
valuable comments. Any mistakes are, of course, my own. Arqueologa UNFV, edited by Claudio Csar
Olaya and Mariana A. Romero Bernales, pp. 13-
REFERENCES CITED 41. Corriente Arqueolgica 1. Lima: Universidad
Nacional Federico Villarreal.
Aimi, Antonio with Duccio Bonavia Chauchat, Claude and Belkys Gutirrez
2003 Per: Tremila anni di capolavori. Milan: Electa. 2005 Excavaciones en la Plataforma Uhle. In Proyecto
Alva, Walter Arqueolgico Huaca de la Luna: Informe tcnico
2001 The Royal Tombs of Sipn: Art and Power in 2004, edited by Santiago Uceda and Ricardo
Moche Society. In Moche Art and Archaeology in Morales, pp. 93-141. Trujillo, Per: Facultad de
Ancient Peru. Studies in the History of Art 63, Ciencias Sociales, Universidad Nacional de
edited by Joanne Pillsbury, pp. 223246. Trujillo.
Washington D.C.: National Gallery of Art. de Lavalle, Jos Antonio
Alva, Walter and Christopher B. Donnan 1995 Trujillo precolombino. Lima: Odebrecht.
1993 Royal Tombs of Sipn. Los Angeles: University of Donnan, Christopher B.
California, Fowler Museum of Cultural History. 1978 Moche Art of Peru: Pre-Columbian Symbolic
Benson, Elizabeth P. Communication. Los Angeles: Museum of
1972 The Mochica: A Culture of Peru. London: Thames Cultural History, University of California, Los
and Hudson. Angeles.
1975 Death-Associated Figures on Mochica Pottery. 2007 Moche Tombs at Dos Cabezas. Los Angeles:
In Death and the Afterlife in Pre-Columbian Art, Cotsen Institute of Archaeology, University of
edited by Elizabeth P. Benson, pp. 105-144. California, Los Angeles.
Washington D.C.: Dumbarton Oaks Research 2010 Moche State Religion: A Unifying Force in
Library and Collection. Moche Political Organization. In New
Bourget, Steve Perspectives on Moche Political Organization,
1994 Bestiaire sacr et flore magique: cologie rituelle de edited by Jeffrey Quilter and Luis Jaime Castillo,
liconographie de la culture Mochica, cte nord du pp. 47-69. Washington D.C.: Dumbarton Oaks
Research Library and Collection.
83 - Wiersema: Moche Architectural Vessels
Donnan, Christopher B. and Guillermo A. Cock Gregory Lockard, and Julio Rucabado, pp. 411-
2002 Informe. Excavaciones en Dos Cabezas, valle del 431. Lima: Instituto Francs de Estudios An-
Jequetepeque, Per: 8a Temporada de trabajo de dinos and Fondo Editorial de la Pontificia
campo, junio-agosto 2001. Report submitted to Universidad Catlica del Per.
the Instituto Nacional de Cultura (now the in press Dramas of the Dialectic: Sacrifice and Power in
Ministerio de Cultura) and deposited in the Ancient Polities. In Violence and Civilization:
library of the Museo Nacional de Antropologa, Interdisciplinary Sketches of a Deep History, edited
Arqueologa e Historia del Per, Lima. by Rod Campbell. Providence: Joukowsky
Donnan, Christopher B. and Donna McClelland Institute for Archaeology and the Ancient
1999 Moche Fineline Painting: Its Evolution and its World, Brown University.
Artists. Los Angeles: University of California, Los Tello, Julio C.
Angeles, Fowler Museum of Cultural History. 1938 Arte Antiguo Peruano: Primera parte; Tec-
Hocquenghem, Anne Marie nologa y Morfologa. Inca. Revista de estudios
1987 Iconografa Mochica. Lima: Pontificia Universidad antropolgicos 2 (Lima: Museo de Arqueologa de
Catlica del Per, Fondo Editorial. la Universidad Mayor de San Marcos).
Mauricio, Ana Cecilia and Jessica Castro Tufinio, Moises
2008 La ltima sacerdotisa mochica de San Jos de 2001 Plaza 3c. In Proyecto Arqueolgico Huaca de la
Moro: Excavaciones en el rea 42. In Programa Luna: Informe Tcnico 2000, edited by Santiago
Arqueolgico San Jos de Moro: Temporada 2007, Uceda and Ricardo Morales, pp. 41-59. Trujillo,
edited by Luis Jaime Castillo, pp. 67-117. Lima: Per: Facultad de Ciencias Sociales, Universidad
Pontificia Universidad Catlica del Per. Nacional de Trujillo.
Miro-Quesada, Luis 2002 Plaza 3c. In Investigaciones en la Huaca de la Luna:
1976 Arquitectura en la cermica precolombina (en Informe tcnico 2011, edited by Santiago Uceda
colaboracin con Cermica-Mosaicos, S.A.) Lima and Ricardo Morales, pp. 47-58. Trujillo, Per:
7-30 de Junio 1976. Lima: Galera Banco Facultad de Ciencas Sociales, Universidad
Continental. Nacional de Trujillo.
Montoya, Mara 2004 Excavaciones en la Plaza 3c de la Huaca de la
1997 Excavaciones en la Plaza 3B. In Proyecto Arque- Luna 1998-1999. In Proyecto Arqueolgico Huacas
olgico Huacas del Sol y de la Luna: Investigaciones del Sol y de la Luna. Investigaciones en la Huaca de
en la Huaca de la Luna 1995, edited by Santiago la Luna 1998-1999, edited by Santiago Uceda,
Uceda, Elas Mujica, and Ricardo Morales, pp. Elas Mujica, and Ricardo Morales, pp. 99-120.
61-66. Trujillo, Per: Facultad de Ciencias So- Trujillo, Per: Facultad de Ciencias Sociales,
ciales, Universidad Nacional de Trujillo. Universidad Nacional de Trujillo.
Mujica, Elas, Rgulo Franco, Csar Glvez, Jeffrey 2006 Excavaciones en la Plaza 3c y sacrificios hu-
Quilter, Antonio Murga, Carmen Gamarra de la Cruz, manos en la Huaca de la Luna. In Proyecto
Vctor Hugo Ros Cisneros, Segundo Lozado Alcalde, Arqueolgico Huacas del Sol y de la Luna:
John Verano, and Marco Aveggio Merello Investigaciones en la Huaca de la Luna 2000,
2007 El Brujo: Huaca Cao, centro ceremonial moche en edited by Santiago Uceda, Elas Mujica, and
el valle de Chicama. Lima: Fundacin Wiese. Ricardo Morales, pp. 47-63. Trujillo, Per:
Pardo, Luis A. Facultad de Ciencias Sociales, Universidad
1936 Maquetas arquitectnicas en el antiguo Per. Nacional de Trujillo.
Revista del Museo e Instituto Arqueolgico 1:6-17. 2008 Huaca de la Luna: Arquitectura y sacrificios
Rengifo, Carlos E., Daniela Zevallos, and Luis Muro humanos. In Actas de la primera conferencia
2008 Excavaciones en las reas 28, 33, 34, 40 y 43: La internacional de jvenes investigadores sobre la
ocupacin mochica en el sector norte de San cultura Moche. Lima, 2004, edited by Luis Jaime
Jos de Moro. In Programa Arqueolgico San Jos Castillo, Helne Bernier, Gregory Lockard, and
de Moro: Temporada 2007, edited by Luis Jaime Julio Rucabado, pp. 451-470. Lima: Instituto
Castillo, pp. 118-161. Lima: Pontificia Universi- Francs de Estudios Andinos and Fondo
dad Catlica del Per. Editorial de la Pontificia Universidad Catlica
Swenson, Edward del Per.
2008 San Idelfonso and the Popularizationof Moche Uceda, Santiago
Ideology in the Jequetepeque Valley. In Actas de 1997 Huaca de la Luna: La arquitectura y los espacios
la primera conferencia internacional de jvenes ceremoniales. Arkinka 20:104-112.
investigadores sobre la cultura Moche. Lima, 2004, 2000 El templo mochica: Rituales y ceremonias. In Los
edited by Luis Jaime Castillo, Helne Bernier, dioses del antiguo Per, edited by Krzysztof Ma-
ANDEAN PAST 10 (2012) - 84
Figure 1: Moche architectural vessel of the Enclosed Gabled Type (EGT), Moche IV,
Museo Nacional de Antropologa, Arqueologa, e Historia del Per (C-54620), Lima,
24.4 centimeters high.
ANDEAN PAST 10 (2012) - 86
Figure 2: Moche architectural vessel of the Enclosed Gabled Type (EGT), Moche IV,
Courtesy of the Division of Anthropology, American Museum of Natural History, New York City
(AMNH B4857), 20.3 centimeters high. Vessel from the Adolph Bandelier collection,
acquired by the AMNH in 1894.
87 - Wiersema: Moche Architectural Vessels
Figure 3: Moche architectural vessel of the Architectural Complex Type, Moche I-II,
Eugenio Nicolini Collection (12836), Lima, 14 centimeters high.
ANDEAN PAST 10 (2012) - 88
Figure 4: Front (left) and top (right) views of Architectural Complex Vessel, Moche III,
Museo Nacional de Antropologa, Arqueologa, e Historia del Per (C-61802), Lima,16.7 centimeters
high. A vessel sharing similar morphology (Phoebe Hearst Museum of Anthropology, 4-3262)
was excavated by Max Uhle from grave 26, Site F, Huacas de Moche in 1899, height unknown
(Donnan 1965: plate 1, number 10; see also Kroeber 1925: plate 561).
Figure 6: Map of archaeological sites on Perus north coast (from Mujica et al. 2007).
ANDEAN PAST 10 (2012) - 90
Figure 8: Double step motif along northern perimeter wall of the Decorated Patio, Huaca Cao Viejo
91 - Wiersema: Moche Architectural Vessels
Figure 13: Floor plan created from fineline on Enclosed Gabled Type,
Photograph courtesy of the Division of Anthropology, American Museum of Natural History,
New York City (AMNH B4857).
Figure 14: Circular structures from Figure 13 translated into rectilinear ones which better reflect forms
found in excavated Moche architecture (both plans are based on AMNH B4857, Figure 13 above).
ANDEAN PAST 10 (2012) - 94
Figure 18: Fineline illustration by Donna McClelland of the Moche Sacrifice Ceremony (from Donnan
1978: figure 239b and Donnan and McClelland 1999: figure 4.102). The bottom register maintains
pictorial elements (chisel, streams of blood, war clubs, and eared serpents) also found on
Enclosed Gabled Type vessels.
ANDEAN PAST 10 (2012) - 96
Figure 19: Front and back of Moche IV vessel depicting a sculpted figure bound to a rack and set
between two step motifs. Blood from the figures neck falls to the women painted in fineline below,
Museo Larco, Lima, Peru (ML001474), 19 centimeters high.
Figure 20: Pictorial imagery found on vessels (often flanking the entrance)
of the Enclosed Gabled Type (EGT).
97 - Wiersema: Moche Architectural Vessels
Figure 24: Artists reconstruction of Huaca de la Luna showing direction of proposed processional route,
beginning in Plaza 1 and leading up to Plaza 3c (from image courtesy of Santiago Uceda).
Circles indicate structures which likely anchored ceremonial activity at the site.
99 - Wiersema: Moche Architectural Vessels
Figure 25: Isometric image of south western section of Huaca de la Luna showing the direction of
processing captives on the south and east walls. The series of step motifs that marks the route from
Plaza 1, below, to Plaza 3c, above, can be seen at top left (after Tufinio 2004: figure 12).
Victor F. Vsquez S.
ARQUEOBIOS, Lima, Peru
THE ARCHAEOLOGICAL CONTEXT OF THE ing it with events at the nearby Huaca Cao
WELL AT EL BRUJO Viejo we infer that there was a significant reno-
vation of spaces in the entire architectural
This article presents information on the complex and changes in its use. Extending the
excavation and analysis of materials from the analysis to compare changes at Huaca Cao Viejo
Ceremonial Well (Pozo Ceremonial) and its complex with those at the Huacas de Moche
surrounding area at the El Brujo Archaeologi- also offers insights into other cultural changes
cal Complex in the Chicama Valley, Peru, and on Perus north coast.
the relationship of events there to activities
elsewhere at the site and in the region (Figure The El Brujo Archaeological Complex
1). The well was discovered in 1993 and exca- occupies a raised terrace at the northern edge of
vated by a team of Peruvian archaeologists in the lower Chicama Valley. This tableland,
the following year. In field seasons in 2002 and averaging eighteen meters above mean sea level,
2004, a combined United States and Peruvian is likely a remnant of a Pleistocene ground
team carried out research in the area surround- surface that was preserved when the rest of the
ing the well. Here, we present a summary of the land mass was washed away late in the era, or in
various field investigations, reviews of analyses the early Holocene. For the most part, the
of material remains, and interpretations of the material comprising the terrace consists of an
significance of these studies. By determining alluvium of gravel of rounded cobbles ranging in
the history of the use of the well and correlat- size from small pebbles to oval or round stones
20 to 40 centimeters in diameter, mixed with (Phases A, C, F, and G; Figure 2). If only the
finer soil. phases of total remodeling are considered, there
were four major building stages (Mujica Barreda
The Brujo terrace is a one square kilometer 2007:102) which we will refer to here as Huaca
triangle with its base on its northern end and 1 (Construction Phase G), Huaca 2 (Construc-
with the long sides meeting at its southern tip tion Phase F), Huaca 3 (Construction Phase C)
where Huaca Prieta, famed for its role in defin- and Huaca 4 (Construction Phase A). The
ing the Preceramic Period (Bird et al. 1985), is remodeling events of the front of the huaca
located. Several large archaeological complexes were considerable but fundamentally were
are on the rest of the terrace and have been continuations of the previous full-scale renova-
identified as associated with Cupisnique, tion. Thus, building phases E and D were elabo-
Moche, Lambayeque, Chimu, Inca, and Span- rations of Huaca 2 and building phase B was an
ish Colonial occupations (Franco et al. 2005a; elaboration of Huaca 3.
Mujica Barreda 2007; Tate 2006). The most
intensive and extensive occupation is associ- The front terraces of the huaca were brightly
ated with the Moche archaeological culture, painted, as were patios on its summit. Rituals
consisting of a large, relatively low complex on took place in both large, apparently public, and
the southwest, known as Paredones, Huaca El smaller, possibly private, spaces. The huaca
Brujo (also known as Huaca Cortada) near the remodelings were mostly on the front terraces
northwestern corner of the complex, and and the front of the huaca summits. These areas
Huaca Cao Viejo, in the northeastern sector of were seen from the plaza and thus received the
the terrace. most attention from viewers and ritual partici-
pants.
Some research has been carried out at
almost all of the archaeological sites of the El A high-ranking woman, now referred to as
Brujo complex, but since 1990, work has con- the Seora de Cao (Burial 3), was buried in a
centrated at Huaca Cao Viejo and its sur- patio on the northwest upper corner of the
rounding area. The huaca is a large, terraced huaca. Her remains were found in a deep shaft
adobe structure with an extensive walled plaza in the adobes of the huaca while three other
at its front, facing north. Initially, based on burials, all of males, were in shallower tombs
work mostly on the northern front of Huaca nearby (Franco et al. 2005a; Williams 2006).
Cao Viejo, Peruvian archaeologists (Franco et During the time of Huaca 2, rituals took place
al. 2003a) determined that the huaca is the in the patio above her tomb. The high status of
sum of seven construction phases, designated the Seora de Cao is demonstrated not only by
Phase A (the latest) to Phase G (the earliest). the quantity and quality of grave goods, but also
In Phases D through G the front terraces of the by the presence of a sacrificed young woman
huaca were decorated with murals, Phase B was next to her burial bundle in the tomb. One of
not completed, while Phases A and C were the other burials (Burial 1) also had a sacrificed
decorated with polychrome bas-reliefs. Each victim in his grave, suggesting that he, too, was
one of these construction phases likely repre- of high status.
sented a significant event in the history of the
huaca although some phases appear to have The term montculos is used at the El Brujo
involved only the remodeling of the front complex to refer to structures with no other
section of the structure (Phases D and E) while name. Montculos means mounds in Spanish,
others completely encased an older structure but excavations in them prior to our work at the
103 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
well revealed that they are purpose-built, and approximately 170 meters north of Mont-
substantial structures made of adobe bricks. culo 3. A primary goal of our research was to
After abandonment, either deliberate entomb- determine the associations of the well with these
ment of the structures or deterioration, various structures and the activities that took
through time, caused them to assume the shape place at and in them.
of small hills. Six montculos are at El Brujo,
three of which are in the northeastern sector of DISCOVERY OF THE WELL AND FIRST STAGE
the complex (Figure 1). Montculo 1 is relatively RESEARCH
small and has two building phases. Excavations
revealed rooms painted white, a corridor, and The well was discovered accidentally in
a patio. Domestic refuse with camelid remains 1993 when an El Brujo archaeologist fell into it
and maize cobs suggest a domestic function. while walking across the site. Dense vegetal
Remains of the second occupation included material in the well mouth supported the ar-
throne or altar-like features and fragments of chaeologist by his armpits, preventing him from
ceramic war clubs known to have served as falling into the void. Subsequently, the Peruvian
roof-top decorations on important buildings. site archaeologists decided to conduct an exca-
Although the original dimensions of the struc- vation of the well.
ture are uncertain, by its second building stage
it was 400 square meters in area, and ceremo- From its lip to the current water level, the
nial activities appear to have increased from well is 12.47 meters deep with the water 2.36
earlier times, based on changes in architectural meters in depth (Franco et al.1995, 2003b). At
detail. its mouth, the well is 3.40 meters in diameter
and the width of the shaft at the water level is
Montculo 2 is approximately seven thou- 1.33 meters in diameter (Figures 3, 4). Descent
sand square meters in area and also was found from the ground surface to the water was ac-
to have two construction phases. The earlier complished via a spiral path that combined
structure was a small, stepped pyramid. The features of both ramp and stairway. Cobbles
structure grew with the addition of platforms from the surrounding gravel were compacted
and a large surrounding wall. Again, all rooms into a series of 40 steps. The treads of some of
were painted white, and decorative ceramic these are wider than others so that two or more
club fragments were found during excavations. footsteps likely would have been necessary
before reaching the next tread. Persons de-
Montculo 3 is roughly one thousand square scending would have entered the cavity from
meters in area and underwent four construc- the west, moved counter-clockwise, and reached
tion phases. Staircases, platforms, floors, and the pool at its northern edge. The descending
ramps all were painted white, although red and spiral takes two-and-a-half turns from top to
yellow paint was found on fragments of plaster bottom.
with cane impressions on the back, indicating
that they are from either a suspended ceiling or Water enters the well through three sepa-
a column made of bunches of reeds as its struc- rate channels in the subsoil. The source of the
tural core. water is the relatively high water table on the
Peruvian coast, a phenomenon exploited by
The Ceremonial Well is 470 meters north- many people, ancient and modern, who exca-
west of the front terraces of Huaca Cao Viejo, vate to just above the water table to create
about 100 meters to the west of Montculo 2, sunken gardens, known as huachaques. In
ANDEAN PAST 10 (2012) - 104
addition, there are large wells at the Chimu site fragments of copper sheets, mollusc shells, and
of Chan Chan entered by ramps, although they non-diagnostic sherds, also in poor states of
are much larger than the Brujo well and are preservation.
rectangular in shape (Campana Delgado
2006:132-135; Moseley 1975). The water in the well was thick with soil and
materials which included a whale vertebra
The well was excavated in arbitrary levels worked to serve as a receptacle, two wooden
based on the turns of the spiral. Five levels paddle-like tools, decorated gourd fragments,
were defined, including the water level. The mollusc shells, and charcoal. The recovered
first level was the mouth of the well, levels 2-4 diagnostic ceramics include types commonly
each accounted for a turn in the spiral while termed Gallinazo and Salinar as well as
level 5 was the muddy deposit below the water Moche vessels. The latter exhibited design
surface. elements that placed most of them in the middle
of the five-phase sequence developed by Rafael
The excavated deposits consisted of great Larco Hoyle (1939; Figures 5, 6).
amounts, and a diverse array, of materials.
These included loose soil, cobbles and angular There was no clear patterning of these
stones, various plant remains, mollusc shells, remains to suggest that the well was filled in
camelid bones, utilitarian and diagnostic ce- stages. The mix of more than a ton of materials
ramic fragments, and human skeletal remains. and their general consistency suggests a single
None of the latter was a complete skeleton but and likely rapid filling of the space by people
the collection included a minimum of ten dumping trash in the well. Once the well was
individuals of different ages, mostly infants and excavated by the archaeologists, the stair-ramp
adolescents, as well as the remains of men and was stabilized and gravel was placed in the space
women. According to John Verano (1995:168) around the entry. A roofless, four-walled room
who conducted the analyses, these bones likely of adobes was built around the well in order to
were deposited in the well after having been prevent stray animals and tourists from falling in
removed from other contexts. They may repre- it.
sent a random assortment, as there is no indi-
cation that they were selected with any prefer- SECOND STAGE RESEARCH
ence to the age or sex of the individuals.
A combined U.S.-Peruvian team returned to
The last tread before the water was wider the well for further studies in 2002 and 2004.
than those above it, creating a small recessed During this work, research was concentrated in
area where the skeletal remains of an adult the area around the well to better understand its
male, between 30 and 40 years of age at time of context within the El Brujo site complex (Figure
death, was encountered (Verano 1995:166). 7).
The skeleton was on its back with the head
oriented to the east. Conservation of skeletal One to two meter wide trenches were placed
remains was poor due to high humidity and the running perpendicular to each of the four walls
skeleton was incomplete. Bone porosity and of the room surrounding the well and then
loss on vertebrae indicate a trauma in life with expanded outwards as work revealed features of
subsequent bone regrowth, and various teeth interest. In 2002, various features were found on
had caries. There was no clear sign of a cause the east and west sides of the well. These mostly
of death (ibid.: 167). Near the remains were consisted of fragments of walls that had been
105 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
robbed of adobes, and small sections of rooms densities are higher than in the upper levels.
that had been badly damaged due to looting, This layer is also between 30 and 60 centimeters
making it difficult to interpret past activities in thick and is associated with structures made of
the area. In one of the rooms, a poorly pre- adobes and small cobbles.
served adult burial was found laid on a mat. On
the north and south sides, however, few traces Capa B, Level 3, is a smooth textured,
of human occupation were clearly in evidence, brown soil. It ranges in thickness from 2 to 10
other than areas of melted adobe. centimeters. This layer rests on sterile soil
deposits and is associated with circular hearths
In 2004 work continued in the same man- and non-diagnostic sherds of bowls and plates.
ner as previously. There was a relatively large,
flat area running along the northwest to the The stratigraphic levels discussed above
northeast side of the well with relatively little correlate with major occupations near the well.
looting in evidence. Work was focused in this The Phase 1 occupation is associated with Capa
sector and uncovered features amenable to B, Level 3 and is the first evidence of human
interpretation. Excavations determined that occupation in the area. The occupation was
there are two major natural stratigraphic light, and activities included making small fires.
levels in the area around the well, Capas A and The nature and quantities of remains are such
B (Figure 8). Capa A is a single deposit, where- that we cannot assign a firm relative or chrono-
as Capa B is subdivided into three levels. Capa metric date to this occupation. Based on fea-
A includes the contemporary ground surface, tures of the utilitarian wares, this occupation is
and is between 10 and 20 centimeters in thick- likely to have occurred in the Early Intermediate
ness. In addition to light amounts of modern Period. We do not know, however, if the well
trash, this layer contains angular stones, cob- was dug or in use at this time.
bles, shells, ceramic sherds, a few bone frag-
ments, and fragmented adobes. The adobes are The Phase 2 occupation of the area around
of a tall form associated with late building the well is associated with Capa B, Level 2.
phases at Huaca Cao Viejo (Glvez Mora et al. During this phase, a number of small structures
2003). made of rectangular adobes and cobbles were
built surrounding the well. Although the recent
Capa B, Level 1, is between 30 and 60 construction of the roofless structure around the
centimeters in thickness and generally contains well prohibited investigation, it appears that a
the same kinds of materials as Capa A. The soil space of two or more meters around the circum-
color is lighter than the upper level and more ference of the well was kept clear of structures.
compact and homogenous. The primary differ-
ence between this level and Capa A is that the The utilitarian ceramics associated with
adobes found in the lower level are of the these structures include many examples of neck-
short variety, associated with earlier building ed jars and bowls, suggesting that they con-
phases at Huaca Cao Viejo. tained liquids. The diagnostic sherds are of the
same kinds as found in the well, Moche style
Capa B, Level 2, is gray in color due to ceramics and examples of Gallinazo and Salinar-
abundant ashes and decomposed organic like wares. It is worth noting that of the eight
materials. The latter include chili pepper seeds, stirrup spout fragments found in the second
cotton, fish and camelid bones, crab parts, and stage work that were large enough to identify,
bivalves, as well as other materials. Sherd four resembled the Moche Phase III style in the
ANDEAN PAST 10 (2012) - 106
relatively squat shape of the stirrups, and four gastropods, mussels, sardines, anchovies, drum-
had tall stirrups or had narrow spouts in the fish, Peruvian weakfish, and guinea pig. Plant
manner of the Moche Phase IV style. remains included willow, goosefoot, lima bean,
common bean, maize, manioc, and potato. The
We took two samples of materials from latter two plants were detected by starch grain
different areas of a large hearth in one of the analysis because no macro remains were present.
structures in Capa B. Combined, the two
unscreened samples weighed 1.63 kilograms The animal and plant materials found in the
and consisted of soil, ashes, cinders, and plant hearth are probably the remnants of meals eaten
and animal remains. Subsequently, these re- by the people who lived in the vicinity of the
mains were analyzed by Victor F. Vsquez S. well, as also indicated by partial burning of some
and Teresa E. Rosales T. (2009) of the of the large fish bones. Such meals likely were
ARQUEOBIOS center at the National Univer- common, everyday ones. No traces of festival
sity of Trujillo (Table 1). foods, such as llama bones, were encountered,
but our sample was relatively small. The overall
Common Spanish Common English impression offered by the analyzed sample,
Animals however, is that the people who ate the food
Molluscs and lived in the vicinity of the well were rela-
Tetula atra caracol negro snail
Prisogaster niger caracolito negro snail tively low status residents of the El Brujo com-
Polinices uber caracol luna snail plex.
Xanthochorus buxea snail
Thais haemastoma caracol snail
Bostryx sp. snail The Phase 3 occupation is associated with
Helisoma sp. ramshorn snail
Physa sp. tadpole snail Capa B, Level 1. This phase witnessed a sub-
Pisidium sp. pill clam, pea clam stantial renovation of the area. The Phase 2
Semimytilus algosus chorito playero mussel
structures were filled in and covered over with
Fish
Sardinops sagax sagax sardina sardine rubble, and the well was filled with trash, in-
Engraulis ringens anchoveta anchovy cluding human remains from the previous
Sciaena deliciosa lorna drum, croaker
Cynoscion analis cachema Peruvian weakfish occupation, as described above. The area was
Mammals smoothed over with puddled adobe, and a series
Cavia porcellus cuy guinea pig of small adobe platforms and a low cobble-faced
Plants wall were constructed (Figure 9). The adobes
Dicots
Salix sp. sauce willow used in these features consisted of old, short-
Chenopodium sp. white goosefoot, pigweed style adobes robbed from the earlier construc-
Prosopis sp. algarrobo carob
Crotalaria sp. rattle pod tions and new style, tall adobes. This wall ran
Phaseolus lunatus pallar lima bean over or very close to the lip of the now filled-in
Phaseolus vulgaris frijol common bean
Manihot esculenta yuca manioc well. The filling in and end of the use of the well
Solanum tuberosum papa potato were associated with a substantial reorientation
Monocots of the overall use of the site area.
Zea mays maz maize, corn
Cyperus sp. junco sedge
Twenty-five meters east of the well, the
Table 1. Animal and plant remains from a hearth near remains of an adobe wall are in evidence by a
the Ceremonial Well. rise in the ground surface formed by the melting
of the adobes from the upper levels of the wall
The contents of the hearth included (Figures 1, 10). The mound of the wall runs 18
macrofossil remains of small mollucs, especially degrees west of north. The rise in the ground
107 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
above the wall can be seen directly east of the shrines on the ceque system of Inca Cusco were
well. It runs northward for 45 meters, then fountains or other water sources (Bauer
makes a right angle, running another 38 meters 1998:62-134, 2004; Rowe 1979).
before disappearing from the ground surface.
Given that the wall is relatively close to the It is a common practice, worldwide, to carry
well, and that it separates the well area from out rituals for or about water and its importance
Montculo 2, we decided to investigate it. We and power. These involve, among other things,
excavated a total of four units on the wall. Our the first rains of the agricultural cycle, great
largest excavation, Unit 1, revealed two paral- rivers as sources of fecundity, and springs as sites
lel single-file adobe walls running with the of miracles. All peoples, especially those in
same orientation, and a large double-file wall environments where the quantity, quality, and
over them, at a different orientation (Figure dependability of water is uncertain, tend to
11). Due to the severe looting of the entire site focus ceremonial and religious activities to try to
area, we could not determine if the bases of guarantee that water is available when needed,
walls were associated with any of the phases in and is controlled.
the well area, although some reasonable infer-
ences are possible as will be discussed below. Given that water in agricultural societies
may be enmeshed in religious concepts and
EVALUATING THE CEREMONIAL NATURE activities, the extent to which we can interpret
OF THE WELL the well as ceremonial versus utilitarian is
not clear. Reserving water for only ceremonial
The feature under discussion has been use and making it off limits for consumption is a
called the Ceremonial Well because it has been course of action that is maladaptive in environ-
assumed, given its location near Huaca Cao ments where water is scarce. Small amounts of
Viejo and the ceremonial activities that are water may be off-limits for consumption, such as
known to have occurred there, that the well holy water used for non-immersion baptisms or
also was a place of ritual activity. A review of for sprinkling on the faithful in some Christian
the issues involved in evaluating ritual at the (Roman Catholic, Orthodox, Anglican) rituals.
well seems appropriate here, and several factors However, the sacrality of water often is attrib-
may be taken into account in such an assess- uted to its life-giving properties in rather funda-
ment. mental, biological ways, whether it be the Nile
or the Ganges. Washing or bathing in sacred
There are many examples, past and pres- grottoes and similar features, however, allows it
ent, of Andean people conducting ceremonies to be both holy and still of use.
and holding beliefs that celebrate and revere
water, as well as fluids in general (Carrin Ca- In the Stage I research, the well water was
chot 1955). Among the many examples are the tested, compared with water from a contempo-
role of blood in Moche sacrifices (Quilter rary well nearby, and found to be highly saline
2001) and the great Inca ceremonies of state (Franco Jordn et al. 2003b). No clear conclu-
involving irrigation canals and their waters, sions can be drawn about the potability of the
such as the sacrifice of a young woman, Tanta water in ancient times from this study, however.
Carhua, and her entombment at an important Many factors, such as surface conditions, precip-
section of an irrigation canal, as described by itation rates, intensity of land and water use,
Hernndez Principe (1923 [1621]:62-63; see and others, including the use of motorized water
Silverblatt 1987:90-101). So, too, many of the pumps, could have affected the quality of the
ANDEAN PAST 10 (2012) - 108
water in the well at El Brujo between antiquity not possible to determine sex morphologically in
and the present day. In short, we cannot say sub-adult remains.
whether the water in the well was potable or
not in the past. Many of the adults showed evidence of hard
lives including periodontal disease and extensive
It may very well be that the waters of the tooth wear. Periostitis, a non-specific indicator
Ceremonial Well at El Brujo were sacred. The of stress that manifests itself as new bone depo-
chief problem in attempting to evaluate this sition on pre-existing boney surfaces faces, was
issue is that the behavior associated with sa- common. Interestingly, all three relatively
credness is not clearly in evidence in the ar- complete crania showed evidence of cranial
chaeological record here, and it is difficult to modification, symmetrical occipital flattening,
develop testing procedures that might elucidate for both men and women.
the matter. Numerous fragments of utilitarian
jars associated with the structures of the Phase Hard lives also are indicated by the evidence
2 occupation indicate that liquids were con- of trauma on the bones. This includes cases of
sumed, but we cannot infer that the liquid in parry fractures and defensive wounds indicative
question was water from the well. of interpersonal violence in two males. A
woman (QM02-02), associated with the B1
A conclusive determination of the role of occupation, had a particularly hard life. Al-
the well at El Brujo cannot be made with though she lived to the relatively advanced age
certainty, at present. Its waters likely were used of forty to forty-five, she had a healed depres-
and perhaps they were considered to have had sion fracture on her right frontal bone and a
special qualities. The only way to further evalu- wide, perimortem cut mark on her left parietal
ate the status of the well as ceremonial, or not, bone (Figure 12). In other words, she had sur-
would be to gain a better appreciation of how vived a blow to the head on at least one occa-
it relates to other architectural features nearby. sion, and had received a cut on the back of her
There were signs of domestic debris to the head shortly before, or just after, her death. It
north and south of the well, but the great seems reasonable to infer that her death was
amount of looting in the area challenges such violent and at the hands of another human
research. being.
The relatively low status of the local resi- In sum, the evidence suggests that the
dents is in evidence through our analysis of people who were buried and, inferentially, had
human remains (Gaither 2002). Few burials lived around and near the well, were non-elites
were found with relatively complete skeletons, and lived relatively hard lives. This suggests that
and most could not be identified with a partic- if these people had any special association with
ular stratum, because they were almost all from the well it did not accord them high rank in
highly disturbed contexts. Of those with com- Moche society. Indeed, the signs point to the
plete enough remains to determine sex or age, interpretation that these people were poor
there were four adult males, three adult females townspeople.
(including one whose sex is not completely
certain due to ambiguous skeletal characteris- Ceremonial activities are in evidence in the
tics), six adults whose remains did not include Phase 3 occupation with its small platforms and
the skeletal elements necessary for the determi- clean, leveled area. These features and any
nation of sex, and five unsexed sub-adults. It is ritual activities that were associated with them,
109 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
which left no observable traces, occurred after tion of forms suggests that relatively similar
the well was sealed and no longer in use. It is activities took place during both the B1 and B2
possible that the platforms above and around occupations. Occupation B1 appears to have
the well area evoked memories of the well been somewhat more intense, based solely on
below them. This seems unlikely, however. the fact that there are more artifacts in its
Rather, there is a sense of erasure of past prac- strata.
tice in the way in which the features above the
sealed well seem to obliterate its presence. 2004 Ceramic vessel forms by levels
quantitative analysis were found for Capas B1 B1 (113) 121 (97.6) 93 (58.4) 55 (23.1) 23 292
and B2 (Table 2). A simple chi-square analysis B2 (34.0) 26 (29.4) 34 (17.6) 21 (6.95) 7 88
(Table 3) of the distribution of forms suggests,
Totals 147 127 76 30 380
however, that there is no strong patterning in
the data that might be interpreted as due to
Expected values in parentheses, observed values without.
human activities. With a probability of 0.235, Chi-square = 4.26, Degrees of freedom = 3, Probability = 0.235
the observed distribution of vessel forms is not
that different from what would be expected in Table 3. Chi-square analysis of ceramic vessel forms
a random distribution and the variations from found in area surrounding the Ceremonial Well.
that expectation, slightly more cntaros in B1
and slightly fewer in B2, slightly fewer ollas in A SECOND WELL
B1, and more in B2, do not readily yield to
interpretation that might be significant in A second well has been found recently at El
terms of human behavior. Indeed, the distribu- Brujo. It is 270 meters to the northeast of the
ANDEAN PAST 10 (2012) - 110
Ceremonial Well and 575 meters almost di- were closed at about the same time. The latter
rectly north of the front terraces of Huaca Cao is a possibility partly because the assumption
Viejo. Well 2 is 8.6 meters deep from the that the Larco phases follow in strict chronolog-
ground surface to the top of the current water ical order is being increasingly challenged by a
level, which 60 centimeters deep. The maxi- number of scholars (see Quilter and Castillo
mum diameter of the orifice is seven meters, 2010). Overlapping or even contemporaneous
with a minimum diameter of four meters. Like existence of styles seems quite likely. As in the
the Ceremonial Well, Well 2 was entered by a case of studies of the Northern Moche (Castillo
counter-clockwise ramp, although it lacks the and Donnan 1994), a chronology of Early
step-like treads of the previously discovered Moche (Larco Phases I and II), Middle Moche
feature (Figure 15). (Larco Phase III and early Phase IV), and Late
Moche (late Larco Phase IV and Phase V;
Details of the features of the upper sections Castillo 2001:308, 2003:67) seems to best
of Well 2 were destroyed in antiquity when it address current understandings and has been
was filled with soil containing relatively small used in this article.
amounts of cultural debris compared to the
Ceremonial Well. Another activity that de- Also as previously discussed, there is a large
stroyed the construction details of the upper wall running between the Ceremonial Well and
part of Well 2 was the building of an adobe Montculo 2. The wall may have enclosed the
chamber tomb on top of, or beside, the access montculo or it may have turned to form a large
ramp, now 2.5 meters below the modern compound next to it, with domestic occupations
ground surface. Burials in the chamber con- located outside of the wall. There were at least
sisted of a male between 20-35 years at time of two major wall construction phases. The earliest
death, and two females, aged 20-34 and 35-45 wall is made of single-file adobes, and our exca-
years (Gaither 2008). No sign of severe trauma vations revealed a parallel wall next to it, sug-
was evident in these remains, and a few small gesting that the two walls formed a street in the
bones, such as from the hands and feet, were town. On top of that wall, however, is a larger
missing from all three burials, and a clavicle, two-file adobe wall running at a slightly differ-
molar, and a few other small bones from a ent, more northwesterly, direction. This wall
fourth individual were found. These data seems more a perimeter wall than a house
suggest that the human remains were not compound wall, and appears to be the same wall
sacrifices, but were burials removed from an- that is observable as a low mound that makes a
other location and reburied when the well was turn to the east north of the Ceremonial Well.
closed. Among the diagnostic artifacts in the
tomb were two fineware ceramics in the Larco Although more excavation would be neces-
Phase V style (Figure 16). sary to confirm it, there is a strong suggestion
from our research that the walls just mentioned
We cannot be sure, at present, if both demarcated different spaces at the El Brujo site.
known wells were in operation simultaneously. The area east of them contains the large Mont-
We only know that the Ceremonial Well was culo 2 and the area to the west consists of resi-
filled with trash that appears to be relatively dences. The walls may have been a division
early, and Well 2 had burials with Moche V between sacred and profane spaces at the site.
ceramics. The wells may have overlapped in
dates of use with Well 2 being later than the In addition, the closing of both wells, after
Ceremonial Well or, alternatively, both wells or in late Middle Moche times, strongly implies
111 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
that whatever their ceremonial roles, their use- Huaca 3 shares the same ceramic styles and
life was deliberately ended. These changes adobe forms as Phase 2 of the well. At the end
conform to larger reorientations of the site, of the use-life of Huaca 3, the temple underwent
including the remodeling of Huaca Cao Viejo a period of decline and perhaps even temporary
itself, and to dramatic changes among the abandonment or, at least, disuse. Walls were
Moche in general. torn down on the huaca terraces and wind-
blown soils accumulated in the main plaza.
CULTURAL CONTEXT AND SOCIAL CHANGE Eventually, however, Huaca 4 was built over the
ruins of earlier construction. Work carried out
To summarize the results of field-work, in this phase reutilized some old adobes, but the
after a light occupation in the area, the Cere- new, tall form was used extensively, as in the
monial Well at El Brujo was excavated and put construction of the cobble wall and platforms of
into use. While the date of this event is uncer- occupation Phase 3 over and around the well.
tain, the well appears to have been used during
the Middle Moche period. The use of the well In addition to the correlations between the
occurred at the same time as the use of small Ceremonial Well and Huaca Cao Viejo, the same
structures nearby which may have been resi- general pattern of a substantial remodeling that
dences of relatively low status people. Eventu- included the use of the new tall adobe style oc-
ally, the well was filled in and the structures curred in Montculo 2. In short, there was a major
were abandoned. This part of the site was then break in the occupation and use of Huaca Cao
remodeled and reoriented for a use or uses that Viejo, followed by a period of disuse and then, by
included small adobe platforms, perhaps for a time of revitalization in which the huaca and
ceremonial activities. How much time elapsed many other constructions were reconfigured. This
between the closing of the well and its associ- included the end of the use of the Ceremonial
ated buildings and the remodeling is uncertain, Well and possibly also of Well 2. This suggests
but the people who changed the spaces appar- that the entire architectural program and, inferen-
ently did not identify themselves with the tially, the religious and political systems associated
human remains that they threw into the well with it, were dramatically altered at the site at the
with other refuse even if a perfunctory human end of the use of Huaca 3 and that Huaca 4
offering was placed in the recessed area near represents a major change in the nature of the
the water. occupation at the site.
Well 2 may have been used concurrently It has been known for some time there were
or, as seems more likely, somewhat later than substantial changes in Moche culture towards
the Ceremonial Well. Both wells were located the end of its history, marked by sharp changes
outside of Montculo 2 and far from Montculo 3 in fineware ceramic styles and their symbolic
and the plaza and temple of Huaca Cao Viejo. contents, changes in construction activities at
Although we cannot correlate the various re- ceremonial centers, and shifts in settlement
modelings and construction phases of all of patterns (see Bawden 1982, 1996; Lockard
these features in detail, we can note that the 2009). The combination of data and informa-
closing of the wells occurred in the same phase tion for the various areas of the El Brujo com-
and, likely, at about the same time, as a sub- plex may now contribute to an interpretation of
stantial and, apparently, dramatic reorientation how these events transpired there, and an
of the structures at the site and the activities evaluation of their importance for understand-
that occurred in them. ing Moche culture change in the greater region.
ANDEAN PAST 10 (2012) - 112
Sample # Sample Type SampleContext RC Age Uncali- 1 * Cal 2 * Cal SH 1 * SH 2 *
(BP) brated date Cal Cal
(BP)
OxA-6896 caa brava Phase A (Huaca 1480 40 550-621 441-652 599-655 556-667
4)
OxA-7005 caa brava Phase D (Huaca 1675 70 255-432 179-548 354-540 256-591
2+)
OxA-7006 caa guayaquil Phase F (Huaca 2) 1670 65 257-502 232-542 385-540 258-590
OxA-7007 caa guayaquil Phase D (Huaca 1865 80 61-241 41 BC- 90-329 29-408
2+) 343
OxA-7008 paint (vegetal Phase F (Huaca 2) 1650 65 263-532 244-546 402-546 261-606
carbon)
Beta-109132 wood wooden idol Phase 1490 80 153060 436-596 416-641 542-646 433-659
C (Huaca 3)
Beta-208269 collagen sacrifice, Huaca E 1210 40 1370 40 650-680 620-700 659-767 644-805
base
Beta-208630 collagen sacrifice, Patio C, 1340 50 1540 50 440-580 410-630 542-638 433-654
Offering 1
Beta-208631 plant sacrificial rope, 1740 40 1750 40 240-350 220-400 261-408 239-440
hild, T. 1, Sra.
group.
Beta-208632 plant sacrificial rope, girl 1619 40 1580 40 420-540 400-570 440-598 427-620
next to Sra.
Beta-212819 cotton unprocessed from 1530 40 1550 40 440-560 420-610 538-622 435-647
Sra. wrapping
Beta-212820 textile threads from 1740 40 1760 40 230-340 150-390 260-396 238-425
(cotton) Sra. wrapping
Beta-230123 charred mate- ritual fire above 1450 50 1420 50 593-659 542-680 613-761 585-773
rial Sra. Cao Tomb
Beta-230124 plant material from Sra. 1580 50 1580 40 430-536 385-597 438-600 425-636
Cao
Beta-230125 plant upper section of 1610 50 1590 50 424-535 349-583 435-584 418-631
huaca front
Beta-230126 wood upper terrace 1730 50 1730 50 249-381 140-420 260-427 243-533
All calibrations performed on the OxCal/ORAU database. The first (left-hand) two columns of calibrated dates are based on the Northern
Hemisphere correction while the two (right-hand: SH) columns are adjusted for the Southern Hemisphere correction. All dates, unless
otherwise indicated, are A.D. and are published here for the first time.
Table 4. Radiocarbon dates for major building phases at Huaca Cao Viejo.
In reviewing previously published radiocarbon Given the overall patterning of the dates,
dates for Huaca Cao Viejo, we discovered that however, the general indication is that construc-
the conversions to calendar dates had been made tion phases F-D occurred within the fourth to
using the Northern Hemisphere calibration. sixth centuries A.D., most likely in the period
Adjusting to the Southern Hemisphere calibra- between the 400s and early 500s A.D. The OxA-
tion brings the Cao Viejo dates more in line with 7007 date is anomalous because Phase F occurred
known chronologies elsewhere, such as Huaca de stratigraphically before Phase D, and there are
la Luna. As adjusted (Tables 4 and 5), the two two consistent dates for Phase F that are later
dates for Construction Phase F are close: A.D. than the out-of-sequence Phase D date. Phases F
258-590 (OxA-7006) and 261-606 (Ox-7008) for and D cannot be temporally separated with these
the 95% confidence interval. The dates for Con- radiocarbon dates, but the law of superposition
struction Phase D are of interest: OxA-7007 can establish that the Phases D and E remodeling
(A.D. 29-408) is earlier than the dates for Huaca occurred after the Phase F construction phase
F while OxA-7005 (A.D. 256-591) is contempo- (Huaca 2). The Seora de Cao burial dates, and
rary with them. other dates associated with it, are generally coeval
with the D-F dates; her burial has been associated
with Phase F, stratigraphically.
113 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Table 5: Bar chart of radiocarbon dates for the Moche occupation at Huaca Cao Viejo at the El Brujo
archaeological complex.
The date for Construction Phase A is satis- Quilter 2008), it is interesting to consider that
factory in being later than any of the previously the greatest activity, or at least the greatest
mentioned, at A.D. 556-667 (OxA-6896). The number of radiocarbon dates indicating activi-
only radiocarbon date that may be associated ties at Huaca Cao Viejo, are in the late fifth
with Constuction Phase C is from a wooden through the sixth centuries. We suggest, how-
idol found buried in a room dating to that era. ever, that these occurred before the Phase
The two-sigma date range for the idol is 433- C/Huaca 3 construction that dramatically
659 A.D. (Beta-109132). As the Phase C reworked the Ceremonial Well area and Huaca
construction is stratigraphically between Phase Cao Viejo. The establishment of the bas-relief
D and Phase A, a cautious placement of the friezes similar to those at Huaca de la Luna
time of the Phase C (Huaca 3) construction occurred after the inferred El Nio events and
would be in the first half of the seventh century. shortly before the huaca was abandoned alto-
gether.
If there were environmental disruptions on
the north coast of Peru due to severe El Nio To the south, in the Moche Valley, Santi-
events in the sixth century (Moseley et al. 1981; ago Uceda Castillo (2010) suggests that at some
Moseley et al. 2008; see also Sandweiss and time circa A.D. 600 construction at Huaca de
ANDEAN PAST 10 (2012) - 114
la Luna halted, with the temple no longer terrace wall has designs of stylized catfish,
actively in use. Huaca del Sol was then greatly imagery that was used over many years at the
expanded relatively quickly, and a new struc- site and may have been emblematic of it. An-
ture, Platform 3, was constructed on the side of other terrace, below the catfish frieze, has small
Cerro Blanco, above the plaza of the old Huaca panels depicting a deity, hero, or similar
de la Luna. Now known as the New Temple anthropomorph sacrificing another (Franco et
and famed for the Revolt of the Objects mural, al. 1994; Mujica Barreda 2007:138-139). This
this platform (Quilter 1990) became the main technique of representation, while appearing
religious structure of the site complex. Both it relatively late at the site may have developed
and Huaca del Sol were built with tall adobes. independently of specific imagery.
It has long been recognized that the times In the last major phase of the Huaca Cao
associated with the transition between Phases Viejo, Huaca 4, the artistic program on the
IV and V of the Larco sequence marked major front of the huaca, as well as the designs in
changes in Moche culture (e.g. Bawden 1982, rooms on its summit, followed the same decora-
1996; Lockard 2009) and significant changes in tive program as at Huaca de la Luna, where at
cultural patterns, including ceramic styles, are least three major building phases previously had
now well in evidence at many other sites, such been made in a distinctive style. That style
as San Jos de Moro in the northern Moche utilized bas-relief to represent life-sized prison-
region (Castillo 2010). Even though there is an ers paraded by victorious warriors on the face of
ongoing reevaluation of Moche sub-styles that the lowest terrace of the huacas, and a series of
are now identified as likely reflecting geograph- mythological beings on the walls of the terraces
ical and even site-specific styles instead of above them. It thus appears that this decorative
relatively uniform chronological changes, there program, its symbolism, and, inferentially, a
still appears to have been dramatic changes in suite of religious practices, were imported from
ceramics and in socio-political relations late in Huaca de la Luna to Huaca Cao Viejo, or that
the Moche era. there was a third, as yet undiscovered, source
for the art program of both Huacas de la Luna
The evidence gathered from excavation of and Cao Viejo. Unless another source is identi-
the wells at the El Brujo Archaeological Com- fied, however, Huaca de la Luna should be
plex and the correlation of events in these considered the origin point for this design
locales with changing construction phases in program and its associated social and political
the montculos and at Huaca Cao Viejo demon- phenomena.
strate that great and important changes oc-
curred at both places. After centuries of huaca The evidence from the excavation of the
buildingHuacas 1 and 2, Construction Phases wells, and our consideration of changes that
E and D, and Huaca 3construction stopped occurred in every structure at the site that has
and Huaca Cao Viejo was little used. Then, been investigated, indicate that the adoption of
some more work was carried out, followed by new ideas and practices at the El Brujo complex
the last major construction, Huaca 4. was a radical departure from older ways. The
adoption of an artistic program similar to that of
It is important to note that the use of bas- the Huaca de la Luna program at Huaca Cao
relief friezes begins at Huaca Cao Viejo on the Viejo thus appears to represent not just a close
front of Huaca 3 (Construction Phase C). Little political and social relationship between the
remains of this decorative program, but one two sites, but a reconfiguration of ritual practice
115 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
and political alliances as well. This change a draft of this article and offering constructive comments,
occurred while many other shifts in behaviors and to Denis Vargas of the El Brujo archaeological team
for help with some last minute illustration needs. We
and relations were occurring elsewhere in the offer special thanks to Gregory Lockard and Juliet
Moche region and was likely part of dramatic Wiersema, Andean Past reviewers, who gave up their
and significant events throughout the Andes. anonymity, and whose highly useful comments on the
draft of this paper greatly improved it.
In conclusion, our research at the Ceremo-
REFERENCES CITED
nial Well, in combination with a review of
architectural phases in other structures at El Bauer, Brian S.
Brujo, has identified a dramatic and profound 1998 The Sacred Landscape of the Inca: The Cusco
halt in activities in the Moche occupation Ceque System. Austin: University of Texas Press.
2004 Ancient Cuzco: Heartland of the Inca. Austin:
associated with Huaca Cao Viejo. After a
University of Texas Press.
period of time, a new artistic and, very likely, Bawden, Garth
religious and political system, apparently very 1982 Galindo: A Study in Cultural Transition During
similar to that of Huaca de la Luna, revitalized the Middle Horizon. In Chan Chan: Andean
Huaca Cao Viejo. These events confirm that Desert City, edited by Michael E. Moseley and
Kent Day, pp. 285-320. Albuquerque: Univer-
there was likely a widespread reorientation of
sity of New Mexico Press.
cultures on the north coast of Peru that oc- 1996 The Moche. Peoples of America series. Cam-
curred in the seventh century, resulting in what bridge, Massachusetts: Blackwell Publishers.
is seen in art and iconography as a new Moche Bird, Junius B., John Hyslop, and Milica Skinner
style, commonly seen in Larco Phase V. Further 1985 The Preceramic Excavations at the Huaca Prieta
Chicama Valley, Peru, edited by John Hyslop.
studies of the specific events that occurred at Anthropological Papers of the American Museum of
sites such as Huaca de la Luna and the El Brujo Natural History 62(1).
complex will surely help clarify the causes and Campana Delgado, Cristbal
consequences of such changes. 2006 Chan Chan del Chimo: Estudio de la ciudad de
adobe ms grande de Amrica antgua. Lima:
ACKNOWLEDGMENTS Editorial Oruss.
Carrin Cachot, Rebecca
The excavations at the Ceremonial Well were 1955 El culto al agua en el antiguo Per. Revista del
supported by Fundacin Wiese of Lima in Stage I, and Museo Nacional de Antropologa y Arqueologa
subsequently by a grant from the National Geographic 2(2):50-140.
Society and research leave from Dumbarton Oaks. The Castillo B., Luis Jaime
research also would have been impossible to carry out 2001 The Last of the Mochicas: A View from the
without the kind offices of the National Institute of Jequetepeque Valley. In Moche Art and Archaeo-
Culture of Peru (now the Ministerio de Cultura) in Lima logy in Ancient Peru, edited by Joanne Pillsbury,
and Trujillo. Analysis of the remains from the hearth pp. 307-332. Studies in the History of Art 63 and
were carried out thanks to research funds from the Symposium Papers 40. Washington, D.C.: Center
Peabody Museum, Harvard University. We thank these for Advanced Studies in the Visual Arts, Sym-
institutions. Special thanks are offered to Marco Aveggio posium Papers XL, National Gallery of Art.
of the Fundacin Wiese for his enthusiastic and generous 2003 Los ltimos Mochicas en Jequetepeque. In
support of the U.S. scholars through his provision of the Moche: Hacia el final del milenio, Actas del Segun-
use of a vehicle, house, and support staff. do Coloquio sobre la Cultura Moche (Trujillo , 1 al
7 de agosto de 1999), Volume 2, edited by San-
Thanks go to all of the project members who helped tiago Uceda and Elas Mujica, pp. 65-123.
make this research possible. We all thank Michael L. Trujillo, Per: Universidad Nacional de Trujillo
Strobel, Director of the National Water and Climate and Lima: Pontificia Universidad Catlica del
Center, United States Department of Agriculture, for his Per.
commentary on issues of water potability in the El Brujo 2010 Moche Politics in the Jequetepeque Valley, A
well. Thanks are also due to Richard L. Burger for reading Case for Political Opportunism. In New Perspec-
tives on Moche Political Organization, edited by
ANDEAN PAST 10 (2012) - 116
Jeffrey Quilter and Luis Jaime Castillo B., pp. Hernndez Principe, Rodrigo
83-109. Washington, D.C.: Dumbarton Oaks 1923 [1621] Mitologa andina. Inca 1:24-68.
Research Library and Collection. Larco Hoyle, Rafael
Castillo B., Luis Jaime and Christopher B. Donnan 1939 Los Mochicas (2 volumes). Lima: Empresa
1994 Los Mochicas del norte y los Mochicas del sur. Editorial Rmac.
In Vics, edited by Krzystof Makowski, Chris- Lockard, Gregory D.
topher B. Donnan, Ivan Amaro Bullon, Luis 2009 The Occupational History of Galindo, Moche
Jaime Castillo, Magdalena Diez Canseco, Otto Valley, Peru. Latin American Antiquity
Elspuru Revoredo, and Juan A. Murro Mena, 20(2):279-302.
pp. 143-176. Lima: Banco de Crdito del Per. Moseley, Michael E.
Franco Jordn, Rgulo, Csar Glvez Mora, and Segundo 1975 Chan Chan: Andean Alternative of the Pre-
Vsquez S. industrial City. Science 187(4173):219-225.
1994 Arquitectura y decoracin mochica en la Huaca Moseley, Michael E., Robert Feldman, and Charles
Cao Viejo, Complejo El Brujo: Resultados Ortloff
preliminares. In Moche: Propuestas y perspectivas; 1981 Living with Crises: Human Perception of Pro-
Actas del Primer Coloquio sobre la Cultura Moche cess and Time. In Biotic Crises in Ecological and
1993, edited by Santiago Uceda and Elias Muji- Evolutionary Time, edited by Matthew H. Nitec-
ca, pp. 147-180. Travaux de lInstitut Franais ki, pp. 231-267. New York: Academic Press.
dEtudes Andines 79. Moseley, Michael E., Christopher B. Donnan, and David
1995 Informe 1995: Programa Arqueolgico Comple- K. Keefer
jo El Brujo. Manuscript on file, Museo de Cao, 2008 Convergent Catastrophe and the Demise of Dos
Chicama Valley, Peru. Cabezas: Environmental Change and Regime
2003a Modelos, funcin y cronologa de la Huaca Cao Change in Ancient Peru. In The Art and Ar-
Viejo, Complejo El Brujo. In Moche: Hacia el chaeology of the Moche: An Ancient Andean
final del Milenio, edited by Santiago Uceda and Society of the Peruvian North Coast, edited by
Elas Mujica, Volume 2:125-177. Trujillo, Per: Steve Bourget and Kimberly L. Jones, pp. 81-92.
Universidad Nacional de Trujillo and Lima: Austin: University of Texas Press.
Pontificia Universidad Catlica del Per. Mujica Barreda, Elas, editor
2003b Un pozo sagrado en el Complejo Arqueolgico 2007 El Brujo: Huaca Cao, centro ceremonial moche.
del Brujo: Un caso de arquitectura mochica In El valle de Chicama/Huaca Cao: A Moche
subterrnea. Arkinka 9(97): 88-95. Ceremonial Center in the Chicama Valley. Lima:
2005a El Brujo: Pasado milenario. Trujillo, Per: Fundacin Wiese.
Ediciones Sin. Quilter, Jeffrey
Gaither, Catherine 1990 The Moche Revolt of the Objects. Latin Ameri-
2002 Summary of the Osteological Analysis of the can Antiquity 1(1):42-65.
Skeletons Excavated by Jeffrey Quilters Team 2001 Moche Mimesis: Continuity and Change in
at El Brujo 2002 Season. Report on file at the Public Art in Early Peru. In Moche: Art and
Peabody Museum, Harvard University, and in Archaeology in Ancient Peru, edited by Joanne
the archives of Andean Past. Pillsbury, pp. 21-44. Studies in the History of Art
2008 Pozo 2, Huaca Cao Viejo2008, Three Moche 63, Symposium Papers 40. Washington, D.C.:
SkeletonsInforme. Internal project document Center for Advanced Study in the Visual Arts,
on file, Museo de Cao, Chicama Valley, Peru, National Gallery of Art.
and in the archives of Andean Past. Quilter, Jeffrey and Luis Jaime Castillo B.
Glvez Mora, Csar, Antonio Murga Cruz, Denis Vargas 2010 New Perspectives on Moche Political Organization.
Salvador, and Hugo Ros Cisneros Washington, D.C.: Dumbarton Oaks Research
2003 Secuencia y cambios en los materiales y tcnicas Library and Collection.
constructivas de la Huaca Cao Viejo, Complejo Rowe, John H.
El Brujo, valle de Chicama. In Moche: Hacia el 1979 An Account of the Shrines of Ancient Cuzco.
final del Milnio, Actas del Segundo Coloquio sobre awpa Pacha 17:1-80.
la Cultura Moche (Trujillo , 1 al 7 de agosto de Sandweiss, Daniel H. and Jeffrey Quilter, editors
1999), Volume 2, edited by Santiago Uceda and 2008 El Nio, Catastrophism, and Culture Change in
Elas Mujica, pp. 79-118. Trujillo, Per: Univer- Ancient America. The 2002 Pre-Columbian
sidad Nacional de Trujillo and Lima: Pontificia Symposium at Dumbarton Oaks. Washington,
Universidad Catlica del Per. D.C.: Dumbarton Oaks Research Library and
Collection.
117 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Figure 1: Map showing the northeastern sector of the El Brujo Archaeological Complex
with features discussed in the text. M designates a montculo.
Huaca El Brujo lies to the west of the complex and is not shown in this figure.
119 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Figure 2: North-South profile of building stages of the Huaca Cao Viejo (not to scale).
The northern, front side of the huaca is to the left.
Figure 5: Body sherds found in excavations. All are red on cream except for upper right
(black and brown on white) and lower left (black on red).
ANDEAN PAST 10 (2012) - 122
Figure 6: Two sherds found in excavations around the well. Both are red on cream.
123 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Figure 8: North wall profile of excavations around the well (see Figure 6). 1. Capa A, salitre;
2. Surface in use during late occupation; 3. Capa B, Level 1; 4. Capa B, Level 2; 5.Capa B,
Level 3 showing two floors (P1 and P2); Declivity to the viewers left (west) is a looters hole.
Figure 10: Aerial photograph of the well and its surrounding area.
Note the large wall at right. North is at the top of photograph.
ANDEAN PAST 10 (2012) - 126
Figure 11: Unit 1. Possible room or compound walls with large wall superimposed. View to the North.
127 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Figure 13: Rim profiles of vessel types found in excavations around the Ceremonial Well.
Row 1: Cntaros Cuellos Rectos; Row 2: Cntaros Cara Gollete; Row 3 Cntaro Cuello Convexo;
Rows 4 and 5: Cntaros Cuellos Expandidos; Row 6: Cuencos; Row 7: Bases.
129 - Quilter et al.: Huaca Cao Viejo, Chicama Valley
Figure 14: More rim profiles of vessel types found in excavations around the Ceremonial Well.
1. (Top 3 Rows): Left Column: Ollas Sin Cuello; Middle Column: Cancheros;
Right Column: Top: Olla Cuello Expandido; Bottom (2): Ollas Cuellos Truncados.
2. (Middle 2 Rows): Left Column: Ollas Cuellos Convexos;
Middle Column: Ollas Cuellos Expandidos; Right Column: Ollas Cuellos Evertidos.
3. (Bottom 2 Rows): Left Column: Tinajas Cuellos Expandidos;
Right Column: Tinajas Bordes Reforzados.
ANDEAN PAST 10 (2012) - 130
Figure 16: Two stirrup spout vessels found with the Well 2 burials. The dimensions of the restored
vessels are: Left: Height: 21.7 centimeters; Maximum Width.: 14.6 centimeters;
Right: Height 22.3 centimeters; Maximum Width: 14.5 centimeters.
ADOBE BRICKS AND LABOR ORGANIZATION ON THE NORTH COAST OF PERU
Howard I. Tsai
University of Michigan
sions and be made from the same soil type site in the Lambayeque Valley, the adobe
(Hastings and Moseley 1975:197-198). Other mounds were built in segments similar to those
Moche period sites with marked adobes include at the Huacas of Moche (Meneses and Chero
Pampa Grande (Shimada 1994:162-166), El 1994:182-183). Adobe structures from the
Brujo (Franco et al.1994:160-161), Galindo Moche V (A.D. 550-750) site of Pampa Grande,
(Lockard 2008:287-289), Huaca Vichanzao however, differed from earlier patterns of con-
(Prez 1994:241-246), and Dos Cabezas (Don- struction. Some construction segments at
nan 2007: figures 2.3 and 2.11). Some marks Pampa Grande included bricks with different
were common and are found at various sites, makers marks. Furthermore, some segments
whereas other marks are unique and observed at had only one makers mark but the bricks were
only one site (Lockard 2008:288; Prez 1994: of different sizes (Shimada 1994:162-165).
244).
During the Middle Horizon (A.D. 750-
Moseley (1975:192) proposed that each 1000), construction principles no longer fol-
makers mark represented a distinct social group, lowed the earlier Moche scheme of one seg-
possibly a work party or a community (Figure 3). ment=one mark=one brick size. At the site of
A makers mark could not have represented a Sicn, one segment often contained adobes of
single individual because (1) some marks per- different sizes and makers marks (Shimada
sisted for more than a century and (2) one 1997:79). This pattern of brick mixing also
hundred individuals could not have produced occurred at the site of Tcume, where adobes
the 143 million bricks that made up the Huaca within a given wall segment vary in size, propor-
del Sol (Hastings and Moseley 1975:197; tions, soil, and mark, and well over half of the
Moseley 1975:192). The Moche state, therefore, adobes are marked (Sandweiss 1995:166).
must have enlisted work parties as represented Thus, unlike the Huacas of Moche, the post-
by the different makers marks to construct the Moche sites had segments built with bricks of
two pyramids. diverse sizes, soil types, and makers marks.
Some two centuries later, a different pattern Kolata (1978) and Shimada (1997) recog-
of brick production characterized Chan Chan, nized the importance of studying bricks from a
the capital of the Chim empire. Bricks from diachronic perspective. If brick size and con-
Chan Chan did not have makers marks and, struction method reflect how states organized
within any given structure, the adobes were labor, then any change in these variables could
made from the same soil and look as if they indicate a corresponding change in labor organi-
came from the same mold(s) (Kolata 1978: zation. Variations in brick sizes and makers
163). Because of this contrast between Chan marks are the result of certain decisions made
Chan and the two Huacas of Moche, Kolata during adobe production. To understand such
(ibid.:164) suggested that the principle of variations, we need to divide the production of
organizing the labor for construction [at Chan adobes into multiple stages, and then infer how
Chan] was different from that used at the patterns in the archaeological record reflect
Moche pyramids. actions taken at each stage of production.
discussion is to show how various stages of brick mixed soil into the mold, flips the mold over,
production affect the variability (size, shape, removes the mold, and leaves behind a damp
color, etc.) of adobes. The production of adobes adobe. Modern bricks made from a single mold
consists of four stages: (1) procuring the soil; (2) are very uniform. The standard deviation is less
mixing the soil with water and temper; (3) than 0.4 centimeters (n=61). Therefore if one
placing that mixture in brick molds; and (4) brick is larger than another by several centime-
drying the bricks. In the following discussion of ters, it is likely that it was made from a different
adobe production, I present my observations of mold.
modern-day adobe production from the town of
Moche (May-June 2004) and archaeological During the Gallinazo period (200 B.C.-A.D.
evidence regarding prehispanic adobe produc- 200, Early Intermediate Period), rectangular
tion. bricks were made from cane molds, as indicated
by cane imprints on the sides of the bricks
Modern-day adoberas are large rectangular (Hastings and Moseley 1975:198). In the follow-
depressions in the ground where the soil is ing Moche period, most bricks had smooth
excavated, mixed, and made into bricks. The sides, although a few bricks still retained cane
soil used to make adobes is a mixture of sand, imprints (Hastings and Moseley 1975:198;
silt, and clay (Brown and Clifton 1978:139). Moubarac 2002:59, 69). Chim bricks from
Before the introduction of trucks or donkeys, Farfn have smooth sides without any cane
adobe makers would obtain the soil close to the impressions. Instead of being flat, the top sur-
structure about to be built, generally within a faces of Chim bricks are often slightly concave,
kilometer. At Farfn, the type and color (brown sometimes displaying finger grooves. These
10YR 5/3 to red 7.5YR 5/4; Munsell Color 2000) grooves suggest that brickmakers at Farfn used
of the soil used in the bricks most resembled the their hands to smooth the upper surfaces of
type and color of the natural soil near those their bricks.
bricks. Bricks from Compound I of Farfn, for
example, contain many rock inclusions, and the After removing the molds, the brickmaker
natural soil surrounding Compound I also has might incise a makers mark on the damp adobe
many rock inclusions compared to other areas. (Hastings and Moseley 1975:198). The bricks
are then placed on their sides, and after about a
During soil mixing, water is added to moist- week, they will be dry and ready for use in
en the soil. It is likely that adobe manufacture constructions (Gillin 1947:37). Workers carry
occurred between the months of November and them to the construction site and the masons
April, when there is more water from the river. lay the bricks, using mud mortar to bind them.
Today adobe makers follow a recipe that The final touch to the construction is a layer of
dictates the ingredients of the adobe soil. I plaster that covers the wall and obscures the
recorded four such recipes at the town of brickwork.
Moche. Each recipe gives a different proportion
of fine sand, coarse sand, earth, and lime. The MODELS OF ADOBE PRODUCTION AND
mason then mixes the soil either by foot or by LABOR ORGANIZATION
shovel. Adobe makers often use straw as a
temper. The production, transportation, and laying
of adobe bricks could have been done by one
Wooden molds are used to manufacture group or different teams accomplishing separate
rectangular bricks. The adobe maker pours the tasks. For example, a community could have
137 - Tsai: Adobe Bricks and Labor Organization
made and transported a few hundred bricks to Farfn, I argue that Moseleys labor tax model
the construction site. Then a separate group of still holds for the pattern observed at Huaca de
masons, summoned from another community, la Luna, and the task-differentiation model, as
would use those adobes to build a wall. Alterna- outlined by Cavallaro and Shimada (ibid.: 90-
tively, people from the same community could 92), best explains the pattern of brick variability
be responsible for every task, including the at Farfn.
manufacture, transportation, and laying of the
adobes. Each stage of adobe production and BRICKS FROM THE HUACA DE LA LUNA
construction can be accomplished in different
ways, and by evaluating all the possibilities, we The Huaca de la Luna is the smaller of the
have several permutations of the production two pyramids at Moche, reaching at least twenty
sequence to consider. meters high and consisting of more than fifty
million adobes (Hastings and Moseley 1975:
Different models of adobe production and 197). Ongoing excavations at the Huaca de la
construction were outlined by Cavallaro and Luna, directed since 1991 by Santiago Uceda
Shimada (1988), who then tested these models and Ricardo Morales, have uncovered evidence
using brick data from the sites of Sicn and of ritual offerings and human sacrifices, as well
Batn Grande (Figure 3). At Sicn, they discov- as friezes depicting dancers, warriors, and vari-
ered that bricks of the same size often had ous supernatural figures (Morales 2003; Uceda
different makers marks, leading them to believe 1997, 2001; Uceda and Tufinio 2003; Uceda et
that the discontinuous territoriality or spon- al. 1994). Excavation of the area between the
sor model best explains their data (Model C or two huacas uncovered a densely occupied urban
D in Figure 3). In these models, the sponsor zone filled with numerous workshops and resi-
(individual or community who donated bricks to dences (Chapdelaine 2001; Uceda and Armas
a temple) or the curaca commissioned the pro- 1997; Uceda and Chapdelaine 1998).
duction of adobe bricks. The sponsor or curaca
first received bricks from different workshops, During the summer of 2004, I participated in
marked the bricks with the lords or sponsors the excavation of Unit 16 on top of the Huaca
insignia, and finally donated the adobes for the de la Luna (Figure 4). Unit 16 contained nu-
construction of a temple (Figure 3, Model C or merous burials. The ceramics associated with
D). Cavallaro and Shimada (ibid.: 98) assumed these burials date to Moche IV, ca. A.D. 400-
that different adobe workshops produced bricks 700 (Zavaleta 2007:13). Bricks from Unit 16 are
of different sizes: The task differentiation-labor rectangular and sometimes incised with a mak-
tax model would have to stipulate the use of ers mark. Most of the adobes were made from
molds of identical shape and size at numerous wooden molds with smooth sides. A few bricks
adoberas, a situation we regard as unlikely. I were made from cane molds, as indicated by
will later examine this assumption using data rows of parallel grooves on their sides.
from Huaca de la Luna. Following Cavallaro and
Shimada (ibid.), I test which model best de- I measured the length, width, and height of
scribes brick production in the states of Moche 95 bricks from Unit 16. I also recorded the
and Chim. If the two states can be character- general location (north, northwest, and south)
ized by two different models, it would indicate of the bricks within the unit (Figure 5). One
that adobe production and labor organization makers mark was observed (diagonal line) on
had changed through time. Using adobe brick 14 bricks from the north corner (Figure 6).
data I collected from Huaca de la Luna and Although the sample is very limited, especially
ANDEAN PAST 10 (2012) - 138
in contrast to the number of adobes measured at of bricks characterized different sections of the
Farfn. I have decided to present these data Moche pyramids.
because the pattern I found contrasts sharply
with any sample from Farfn. In other words, no Cluster n Length (cm) Width (cm) Height (cm)
sample of bricks at Farfn, out of the 4,826 Mean s Mean s Mean s
measured, had produced the results obtained 1 12 34.67 1.23 25.67 1.5 14.00 1.48
from Huaca de la Luna. I believe this result is 2 24 28.04 .91 19.75 .68 9.67 1.46
significant and indicates a contrast in the types
3 59 31.61 1.19 21.83 .81 11.83 1.04
of adobe production at the two sites.
Table 1. Bricks from Unit 16, Huaca de la Luna.
In addition, an examination of Moubaracs
(2002) data, collected from the Huacas of
Location Cluster 1 Cluster 2 Cluster 3 Total
Moche, led me to revise some of the assump-
tions made by Cavallaro and Shimada (1988). North 0 0 47 47
During field-work, I noticed that bricks from Moubarac (ibid.: 86-88) notes that there are
different locations within Unit 16 differed in two sizes of bricksstandard and large. A histo-
size. For example, bricks from the south side of gram of the lengths of the bricks shows two
Unit 16 tended to be smaller than bricks from peaks, with the cut-off at around 33-34 centime-
the northwest corner (Figure 5). A scatterplot of ters (Figure 8). The standard-sized bricks are
the lengths and widths corroborates my observa- shorter than 33 centimeters and on average
tion (Figure 7). Cluster analysis of the bricks, measure 30 by 19 by 13 centimeters (length,
with length and width as the variables (three width, and height). The large bricks are longer
cluster solution, Wards method), and chi- than 33 centimeters and on average measure 37
square analysis of the clusters show that the size by 23 by 15 centimeters (Table 3). It is possible
difference between bricks from different loca- that the larger bricks were used for the base of a
tions is statistically significant (Tables 1 and 2). wall or building, a phenomenon observed at
This result shows that sometimes different sizes Chan Chan (Kolata 1982:68). Indeed, a layer of
139 - Tsai: Adobe Bricks and Labor Organization
giant adobes, each measuring 96 by 60 by 21 bricks, there are smaller size classes or sub-types.
centimeters and weighing some two hundred Bricks from the north and south sides of Unit
kilograms, had been found eight meters below 16, as represented by Clusters 2 and 3, respec-
Floor 1 (Piso 1) of Plaza 2B at Huaca de la Luna tively, show a difference of a few centimeters
(Tufinio 2003:29). that was observed in the field and confirmed
through statistical analysis. Had I clumped all
Brick size n Length (cm) Width (cm) Height (cm) the bricks from Unit 16 as a single data set, this
Mean s Mean s Mean s
small difference would have gone undetected
Standard 134 30.22 1.18 19.00 1.04 13.38 1.16 and all these bricks would have been classified
Large 34 37.17 1.40 23.44 1.05 15.41 1.28 as standard. It is therefore necessary to control
for location and separate the bricks from differ-
Table 3. Comparison of standard and large bricks from ent corners of the excavation unit.
the Huacas of Moche (data from Moubarac 2002:16-
122). It has been argued that the construction of
separate segments is due to purely technical (as
In some cases both the standard and large opposed to social or administrative) reasons. For
bricks bear the same makers mark. For example, instance, Meneses and Chero (1994:183) sug-
diagonal marks occur on both standard and large gested that segmented construction made the
bricks, though certain marks, such as double structures more resistant to earthquakes (contra
vertical bars, appear only on standard bricks Moseley 1975:193). While the separation of
(Table 4). Adobes of different sizes that had the building segments might be attributed to con-
same makers mark were also present at El Brujo struction technology, there is no apparent
(Franco et al. 1994:161). This means that some technical reason for the use of different brick
communities or workshops produced both large marks in different sections, and therefore the
and standard bricks. The fact that a community use of makers marks is better explained as
made bricks of different sizes is not surprising, identifying social groups. The division of con-
because, as Moseley (1975:192) pointed out, struction projects based on social groups (ayllus
these brick marks spanned many generations, or lineages) is a common Andean practice,
and it is also possible that, within a workshop, documented ethnohistorically (Thompson and
brickmakers with different molds were working Murra 1966:636) and observed ethnographically
at the same time. as recently as the 1980s by Urton (1988) in the
highland town of Pacariqtambo, where the
Diago- Double Double Point Semi- Vertical Finger churchs perimeter wall was maintained by
nal vertical vertical circle lines on different ayllus, with each ayllu responsible for a
bars lines the side
(center) (off- certain section of the wall.
center)
the following section I present data from Farfn though the east-west extent of the site is no
to demonstrate this difference. Because my more than 0.25 kilometer. Compound II was the
sample of bricks from Unit 16 of Huaca de la first compound built by the Chim at Farfn. It
Luna is extremely limited, future investigators is identified as Chim because it contains archi-
should measure more bricks from Huaca de la tectural elements similar to those at Chan
Luna, and record their exact provenience in Chan, the Chim capital (Keatinge and Conrad
order to confirm or reject my observations. 1983:271). For example, like the Ciudadela
Rivero of Chan Chan, the only entrance to
FARFN AND THE CHIM CONQUEST OF THE Compound II is from the north. Visitors first
JEQUETEPEQUE VALLEY encounter a plaza, followed by a series of store-
rooms and audiencias, or U-shaped niched
Farfn, located in the Jequetepeque Valley, rooms. The storerooms or bins may have con-
was first excavated by Richard Keatinge and tained valuables, and someone sitting in the
Geoffrey Conrad (1983) in 1978 and then by audiencia would be able to control access to the
Carol Mackey (2004, 2006, 2009) from 1999 to bins (Andrews 1974:256), or, according to John
2004. The site was occupied by three successive Topic (2003:251-252), keep track of the num-
polities: Lambayeque, Chimor or Chim, and ber of goods in the storerooms. At the southern
Inca or Late Horizon (Table 5). Because of its end of the compound lies the burial platform
strategic location at the crossroad of two major (Figure 10), which overlooks another set of
routes (Figure 9), Farfn became the administra- plazas and storage bins. This arrangement of
tive center for the Jequetepeque Valley under plazas, storerooms, audiencias, and burial plat-
Chim and Inca rule (Mackey 2006:328). Dur- forms at Compound II is analogous to the archi-
ing each occupation new structures were built, tectural layout of the late-period ciudadelas of
eventually creating the six compounds of Farfn Chan Chan (Mackey 2009).
that can be seen today along the Pan-American
Highway (Figures 10 and 11, Table 5). The Keatinge and Conrad (1983) excavated
investigation of Farfn, therefore, represents an Compound II of Farfn to investigate Chim
important case study on how imperial conquest expansion into the Jequetepeque Valley. Based
can be documented using ethnohistory (Conrad on an ethnohistorical account by Calancha
1990) and archaeology (Mackey 2006, 2009). (1977 [1638]), Keatinge (1982:206-209) pro-
posed that Compound II was the palace of
Occupation Date Compounds built Brick forms Pacatnam, a Chim general who conquered
Lambayeque 1100-1310 III Flat rectangular
the Jequetepeque Valley during the first wave of
Chim expansion (Mackey 2009). Calancha
Chim 1310-1460 II, IV and VI Flat rectangular, some
with slightly bulging top (1977 [1638]:1228-1229) mentioned a chain of
mountains 3 leguas (fifteen kilometers) long
Inca 1460-1532 I and V Flat rectangular, plano-
convex, and oval next to the palace of Pacatnam, and the moun-
tain range next to Farfn, Cerro Faclo, is indeed
Table 5. Dates, compounds, and brick forms of the about fifteen kilometers long.
three occupations of Farfn (Moore and Mackey
2008: figure 39.3). The geographic and economic importance of
the Jequetepeque Valley may have been the
Farfns northernmost compound (Com- initial motivation for conquering that region
pound VI) is more than 3 kilometers from its (Conlee et al. 2004:216); the valley was situated
southernmost compound (Compound I), even between Chan Chan and the metal ores of
141 - Tsai: Adobe Bricks and Labor Organization
Cajamarca (Lechtman 1976:14), the Sicn as fill, a practice observed at Chan Chan (Day
polity in the Lambayeque Valley (Heyerdahl et 1982a: 340). After all the bricks were laid down
al. 1995; Tschauner 2001), and the prized and mortared, masons applied a layer of mud
Spondylus shell in Ecuador (Cordy-Collins 1990; plaster (about three centimeters thick) to both
Pillsbury 1996). Keatinge and Conrad (1983: sides of the wall. The bricks were arranged in
282) postulated that, after the conquest of the the header-stretcher or header-runner configu-
Jequetepeque Valley, Chim administrators at ration; i.e., the adobes in each layer were placed
Farfn managed the flow of information and perpendicular to the layer above and below.
goods from lower level sites in the valley One layer would have bricks parallel (length-
(Talambo, for example) to Chan Chan. wise) to the wall, and the layer above would
have bricks perpendicular to the wall, and so on.
From 1999 to 2004, Mackey excavated all All the walls and structures at Farfn were built
the compounds of Farfn (Mackey and Juregui in the header-stretcher configuration.
2004). Mackeys work showed that Compound
III was built by a local polity prior to the Chim Construction at Farfn reflects both local
invasion, and the Inca empire continued to use (Jequetepeque Valley) and Chim (Moche
the site after conquering the north coast around Valley) traditions (Table 6). Like the structures
A.D. 1460 (Mackey 2003). Located at the at Chan Chan, the perimeter walls of Farfn
crossroads of empires (Kosok 1965:118), compounds were made almost entirely of flat
Farfn hosted a heterogeneous mix of local, rectangular bricks. Structures from Pacatnam,
Chim, and Inca personnel, demonstrating its a Late Intermediate Period site in the
importance as an administrative center in the Jequetepeque Valley, have both rectangular and
lower Jequetepeque Valley. plano-convex bricks. Plano-convex bricks,
which have a bulging top surface, are common
ADOBE STRUCTURES AT FARFN in Late Intermediate Period structures from the
Lambayeque and La Leche Valleys. Some struc-
Almost all the structures at Farfn consist of tures at Farfn, especially Inca period annexes
adobes. Walls made from tapia are present, but and modifications, have a mixture of rectangular
rare. The compounds were made from adobes, and plano-convex bricks.
and some perimeter walls, like those of Com-
pound IV, still reach 4.5 meters in height (Fig- All adobe walls and structures (e.g. burial
ure 12). At its base, the width of the perimeter platforms) at Farfn were made from solid
wall ranges from 1.7 to 3 meters, with large adobes. In contrast, certain walls from Chan
rocks serving as the foundation. Some perimeter Chan exemplify the chamber-and-fill method
walls taper as they ascend, perhaps to ensure (Day 1982b: 55-57). At Chan Chan, bricks at
stability. Walls within the compounds are not as the base of the walls are larger than bricks on
high as the perimeter walls. They do not taper as top (Kolata 1982:68), a phenomenon not ob-
they ascend, nor do they have large rocks as served at Pacatnam (McClelland 1986:41) or
their foundation. Instead, a layer of gravel serves Farfn. Bricks from Farfn, Chan Chan (Kolata
as the foundation for these internal walls. 1978:163), and Pacatnam (the only exception
being some marked bricks from a ramp; see
The compound walls were made from ado- McClelland 1986:39) do not have makers
bes. In building the walls, Chim masons set marks. In contrast, makers marks are present on
down layers of bricks bound by mortar. Some- bricks from the Late Intermediate Period site of
times broken bits of mortar or adobes were used
ANDEAN PAST 10 (2012) - 142
Tcume in the La Leche Valley (Sandweiss similar to how native Andeans built churches in
1995:164-166). the Spanish canon during the colonial period
(Fraser 1990).
Chan Chan Pacatnam Farfn
1986:41). Exposed wall surfaces at Compounds III 1695 30.2 17.4 11.4 1.75 0.67
II and IV of Farfn, where wall plaster had been
IV 583 29.3 17.3 10.9 1.72 0.64
removed by rain or wind, revealed no sign of
V 116 28.5 19.1 13.2 1.53 0.70
segmentation.
VI 1505 28.7 18.1 12.4 1.61 0.70
noting the layer, row, and orientation (header terize the variability of bricks within a com-
vs. stretcher). pound or site.
Brick Change through Time Location n Length (cm) Width (cm) Height (cm)
Mean s Mean s Mean s
According to Kolata (1978, 1982), flat Storeroom 48 29.79 2.72 16.98 2.02 11.42 1.46
2
rectangular bricks are associated with the earlier
ciudadelas of Chan Chan, whereas tall rectangu- Storeroom 52 30.06 1.78 17.81 1.66 10.79 1.45
3
lar bricks are associated with later ones (see also
Storeroom 83 31.28 2.39 17.52 1.91 12.16 1.26
Cavallaro 1991:15-18). In other words, earlier 4
ciudadelas, such as Ciudadela Tello and Ciuda-
Cut 1 110 31.73 3.25 16.63 2.40 11.15 1.45
dela Uhle, have bricks with lower height-to-
width ratio than bricks from later ciudadelas, Cut 2 108 30.06 1.81 14.14 1.99 9.57 1.06
such as Ciudadela Gran Chim and Ciudadela Cur 3 131 29.47 1.57 17.70 1.22 10.31 1.01
Laberinto. No rectangular brick from Farfn, Cut 4 101 32.51 2.47 18.15 1.79 10.95 1.49
even those from late contexts (Inca period),
approach the height-to-width ratio of the tall Table 8. Bricks from Compound II, Farfn.
bricks at Chan Chan.
At Farfn, a diagnostic feature of later
Three different projects sampled the bricks buildings is the plano-convex brick. Except for
from Compound II at Farfn: Kolata (1978: some slightly plano-convex bricks (n=3) found
190), Keatinge and Conrad (1983:274), and the in Cut 1, all Chim bricks at Farfn were rect-
present study. The bricks I sampled in Com- angular. Plano-convex bricks appear in annexes
pound II (n=775) have an average height-to- and renovations that date to the Late Horizon
width ratio of 0.64; Keatinge and Conrads or Chim-Inca horizon. For example, one struc-
(1983:274) height-to-width ratio for Compound ture north of Compound II, Elite Residence C
II is 0.56 for the audiencia/storeroom complex (not shown in Figure 10), was built entirely of
(n=13) and 0.57 for the burial platform (n= oval plano-convex bricks.
140); Kolatas (1978:190) brick samples from
Farfn yielded an average height-to-width ratio Synchronic Differences in Brick Dimension
of 0.732 (sample size unknown). Tello (2004:
33) also mentioned a general dimension of To test for synchronic differences in brick
bricks from Farfn (29 by 18 by 10 centimeters, size, one must isolate an area in which the
or 0.55 height-to-width ratio), though he did structures were built within the same period.
not specify the sample size or where the bricks Once chronological contemporaneity is estab-
were measured. lished, the analysis of brick variation can be
conducted at various levelscomparing different
These differences in the measurements of layers of adobes within a sampling cut, compar-
Farfn bricks show that different parts of Com- ing different sampling cuts within a wall, com-
pound II have bricks with different dimensions paring different sampling cuts within a com-
(Table 8). For instance, bricks from Cut 2 pound, comparing different compounds within
(n=108) have a height-to-width ratio of 0.78, the site, and so forth.
whereas bricks from Cut 4 (n=101) have a
height-to-width ratio of 0.61. Hence a limited To investigate synchronic differences in
sampling of bricks would not accurately charac- brick dimension, I placed four adobe sampling
ANDEAN PAST 10 (2012) - 144
cuts on the perimeter walls of Compound II The difference in brick sizes between the
(Figure 10, Table 8). The perimeter walls of cuts, even though small, is statistically signifi-
Chan Chan prevented outsiders from looking cant. Median test (Siegel 1956:111-116) com-
inside the rectangular compounds (Moore paring the length, width, and height of bricks
1996:117-118); thus the perimeter walls of from different sampling cuts shows that there is
Farfn presumably had to have been built before less than 0.1 percent chance that the bricks
the compound could function properly. The came from the same population (1-way test,
perimeter walls were likely built in a relatively ChiSquare= 45.6776; df=2; p<.0001). In other
short time, in months or years, but not decades words, it is unlikely that all the bricks from
or centuries. It is unlikely that any administrator Compound II came from a single maker. This
(perhaps General Pacatnam) had to wait many suggests that multiple brickmakers were in-
years before a compound came into use. In volved in the construction of Compound II, and
addition, because no separate wall segments that one area of construction (a storeroom or a
were found at Farfn, I sampled adobes from section of a wall) might contain more bricks
different storerooms to test whether each store- made by, say, Brickmaker A than by Brickmaker
room was built by a separate group of workers. B. The difference between bricks of different
sampling cuts is not visually obvious. There is
While measuring the bricks from Farfn, my too much overlap in brick dimensions to distin-
first impression was that one area of the site did guish bricks from different sampling cuts (Fig-
not have bigger or smaller bricks than any other ures 14, 15). One cannot say with confidence,
area. A scatterplot of bricks sampled from three from size alone, where a brick was used. Rectan-
different storerooms seemed to confirm my gular bricks sampled from Compound IV, which
observation (Figure 14). Yet the cluster analysis dates to the Chim occupation (Mackey 2009:
of these bricks (with length and width as the 330), all tend to measure (give or take a few
variables, three cluster solution, Wards method) centimeters) 30 by 17 by 11 centimeters (Figure
shows that the differences between the store- 16). Therefore size or dimensions cannot be
rooms are statistically significant (Table 9). In used to determine where a brick should be
other words, the difference is very subtle and not placed in the compound. Both Brickmaker A
visually obvious. Such minor differences also and Brickmaker B contributed bricks to the
exist between parts of the perimeter wall. Bricks construction of Storeroom 2, though we cannot
from Cut 2, for example, are generally narrower separate the two kinds of bricks after they are
than other bricks in Compound II (Table 8), pooled and mixed.
whereas bricks from Cut 4 tend to be longer.
One section of the north wall of Compound
Storeroom Cluster 1 Cluster 2 Cluster 3 Total IV had alternating layers of bricks made from
2 12 23 13 48
different kinds of soil (Figure 17). Some layers
had greater amounts of rock inclusions than the
3 7 39 6 52
layers above and below. Alternating layers of
4 41 33 9 83 adobes with different makers marks were found
at Galindo, a Moche site (Lockard 2008:287-
Table 9. Bricks from the storerooms of Compound II, 289); these patterns suggest that certain sec-
Farfn. Cluster analysis of the bricks, with length and tions of a wall or building were built in horizon-
width of the variables (three cluster solution, Wards tal stretches with adobes from different sources
method) and chi-square analysis of the clusters show or workshops. At Farfn, bricks found within
that the size difference between bricks from different
the same cut sometimes have different dimen-
locations is statistically significant (p<.0001).
145 - Tsai: Adobe Bricks and Labor Organization
sions (Figures 18-21, Table 10). This shows that Number Layer Length Width Height Orienta-
bricks made from different molds (and by infer- (cm) (cm) (cm) tion
58 8 27 18 9 Header
ence, different brickmakers) were placed in the 59 8 28 19 10 Header
same area within the wall. 60 8 27 19 9 Header
61 8 26 18 9 Header
62 8 33 16 11 Header
63 8 33 15 9 Header
Number Layer Length Width Height Orienta- 64 8 33 15 9 Header
(cm) (cm) (cm) tion 65 8 29 15 11 Stretcher
1 5 29 18 13 Stretcher 66 8 26 17 8 Header
2 5 29 16 13 Stretcher 67 8 29 18 10 Header
3 5 29 16 11 Stretcher
4 5 27 18 10 Stretcher Table 10. Measurements of adobes from Layers 5 to 8
5 5 29 16 12 Stretcher
6 5 28 18 9 Stretcher of Cut 3, Compound II, Farfn.
7 5 31 19 10 Stretcher
8
9
5
5
28
29
18
16
11
12
Stretcher
Stretcher
DISCUSSION
10 5 28 18 9 Stretcher
11 5 28 19 10 Stretcher
12 5 27 18 9 Stretcher
My data show many important differences in
13 5 29 18 9 Stretcher brick production and construction between
14 5 29 17 10 Stretcher
15 5 30 16 10 Header Huaca de la Luna and Farfn. To account for
16
17
5
5
27
28
18
18
9
10
Stretcher
Stretcher
these differences, I suggest that at Huaca de la
18 6 27 18 10 Header Luna, Moche brickmakers dried their adobes in
19 6 29 18 10 Header
20 6 27 17 9 Header
separate areas (Figure 22 or Model A in Figure
21 6 28 15 12 Header 3). Each group of Moche workers then built the
22 6 30 18 11 Header
23 6 27 17 10 Header pyramids from their respective pool of bricks.
24
25
6
6
32
28
15
17
10
10
Header
Header
There was some exchange of bricks between
26 6 28 18 10 Header work parties, but this exchange was minimal. As
27 6 33 20 11 Header
28 6 31 17 10 Header
a result, archaeologists can still identify different
29 6 30 17 10 Header groups of bricks with unique makers marks or
30 6 28 17 9 Header
31 6 30 19 10 Header sizes at Huaca de la Luna. My data corroborate
32
33
6
6
32
29
17
19
9
10
Header
Header
earlier observations made by Hastings and
34 6 29 18 11 Header Moseley (1975:202-203).
35 6 33 20 12 Header
36 7 30 20 10 Stretcher
37 7 30 19 9 Stretcher By contrast, brickmakers at Farfn placed
38 7 29 18 11 Stretcher
39 7 27 19 10 Stretcher their bricks together to dry, and a group of
40
41
7
7
30
29
16
17
10
8
Stretcher
Stretcher
workers, separate from the brickmakers, went to
42 7 30 18 10 Stretcher this pool of bricks to construct walls for the
43 7 29 18 10 Stretcher
44 7 29 17 11 Stretcher
compounds. As a result, archaeologists find a
45 7 29 18 10 Stretcher mix of different brick sizes within a construction
46 7 29 17 10 Stretcher
47 7 30 18 10 Stretcher segment (Figure 22 or Model B in Figure 3).
48
49
7
7
30
29
18
17
10
10
Stretcher
Stretcher
Because the bricks were pooled and mixed, the
50 7 30 18 9 Stretcher difference in brick size between the walls or
51 7 29 20 10 Stretcher
52 7 29 17 11 Stretcher
storerooms at Farfn is not as strong as the
53 7 35 17 12 Stretcher difference in brick size at Huaca de la Luna
54 7 29 16 11 Stretcher
55 8 30 18 10 Header (compare Figure 7 with Figure 14).
56 8 32 18 10 Header
57 8 28 18 10 Header
Why did Moche bricklayers use separate
/Continued columns of adobes to distinguish one part of the
monument from another, while Chim bricklay-
ANDEAN PAST 10 (2012) - 146
ers did not? One explanation is that Chim labor receipt or punch card with direct
rectangular compounds at Farfn, unlike con- supervision and monitoring.
structions at Huaca de la Luna or Huaca del Sol,
could be divided into clear subunits. For exam- This difference between Moche and Chim
ple, one work group could be assigned to finish methods of supervision could be due to (1) the
all the storerooms, whereas another work group distance between the area of adobe production
could be responsible for building the perimeter and the construction site or (2) the degree of
wall. Indeed, Day (1982a: 340) suggested that at state centralization. First, if Chim brick manu-
Chan Chan smaller walls of structures within facture occurred near the site of construction,
ciudadelas usually do not have joints or sections, then Chim officials could directly observe
but the volume of these walls is a great deal less which communities were making the bricks and
than the volume of the large surrounding walls supervise construction at the same time. In
[which were segmented]. Perhaps entire small contrast, if Moche brick manufacture occurred
walls represent task units, equivalent to a sec- away from the pyramids, then Moche officials
tion or sections of the larger wall. could not witness the production of adobes.
Instead, they waited at the construction site,
Monuments such as the Huaca del Sol and received the adobes, and credited the work of
Huaca de la Luna, in contrast, required divid- each community by examining the makers
ers to differentiate one segment from another. marks.
Because the Moche pyramids are mostly solid
adobe blocks, it would be very difficult for super- It is also possible that under the Chim,
visors to know where one group of workers had specialized groups of brickmakers and bricklay-
started building and where another group had ers were directly under the command of local
stopped without some kind of segmentation. lords, who were themselves subordinates of the
Separate segments, therefore, could help distin- overarching Chim state. In this specialized
guish one task from another. system, each group of brickmakers or bricklayers
became the unit of labor tribute, and because
To explain why makers marks were used by the state was no longer keeping track of differ-
Moche but not by Chim brickmakers, I suggest ent communities, the laborers work resulted in
that Moche brickmakers were not as closely a diversity of adobe shapes and sizes within a
monitored as Chim brickmakers. With makers section of a wall or building. This contrasts with
marks, a Moche state official, instead of having the Moche system, in which the unit of tribute
to visit the adobe workshops and supervise brick was a community or ayllu. Each ayllu had its
production, was able to (1) receive the bricks at own brickmakers and bricklayers, and Moche
the site of construction, (2) identify the maker state officials had to separate the efforts of
and transporter of the bricks, and, (3) ensure different ayllus. Specialists in construction, the
that each group of laborers completed their unit albailes, were recorded in sixteenth-century
of construction using bricks from their own Spanish documents from the north coast of Peru
community (and not accidentally borrow the (Ramrez 1986: table 11.1). Although it is
bricks of others). If state officials personally unclear whether the term albailes referred to
supervised every stage of the production and brickmakers, bricklayers, or both, such special-
construction, then makers marks would not ists may have originated during the Late Inter-
have been needed to account for labor effort. mediate Period. They may have worked on state
And this is what I believe the Chim officials projects for a few months, paying a time-based
did: they replaced makers marks as a form of
147 - Tsai: Adobe Bricks and Labor Organization
labor tax similar to that imposed by the Inca guish the construction of one segment from
(LeVine 1987:24). another. In this case, we need more data on
bricks from sites dating to the time between
The Sicn pattern of adobe construction, as Huaca de la Luna and Farfn to detail the
documented by Cavallaro and Shimada (1988), relationship between monuments, construction
may represent an intermediate stage between techniques, and segmentation. I would expect
the Moche and Chim states: like the Moche, that as structures become more internally heter-
Sicn adobe production occurred away from the ogeneous, that is, having more and different
construction site, thereby requiring makers architectural featuresstorerooms, platforms,
marks. The Sicn system of construction, how- benches, ramps, corridorsthere would be less
ever, was specialized (like the Chim system) need to have unbounded segments, because
and entailed separate groups of brickmakers and each internal feature would serve as a unit of
bricklayers, and, therefore, resulted in the mix- labor taxation.
ture of bricks observed at Farfn (Table 11).
Thus I would argue that Shimadas task Second, as the site of brick production
differentiation-labor tax model (Model B in moves farther from the site of construction,
Figure 3) best describes Sicn and Chim forms state officials can either visit every adobe work-
of labor organization, whereas the Moche form shop (which is time consuming) or stay on-site
of labor organization is best described by to make sure that enough bricks had been
Moseleys labor-task model (Model A in Figure brought to the building area. In the case of the
3). Moche, makers marks were used to help offi-
cials keep track of brick production. Future
Brick Drying Construction investigations should test my interpretation by
production bricks
Moche Makers marks Separate Same brick
sourcing the soil used to make the bricks at
(A.D. 100-800) 6 production areas mark for a seg- Moche and Chim sites. I would expect Moche
not monitored. ment. Hence
brickmakers bricks to have been made from soil to be found
came from the at some distance from the monument, and
same commu-
nity as the Chim bricks to have been made from soil
bricklayers of a
single segment.
found close to the rectangular compounds.
Sicn Makers marks Pooled Mixed brick
(Middle and 6 production sizes and Finally, I expect more centralized states to
Late, A.D. 900- not monitored. makers marks.
1375) Bricklayers dif- have a different type of labor organization than
ferent from the
brickmakers.
less centralized states. To test this hypothesis,
Chim No makers Pooled Mixed brick we would need other lines of evidence to deter-
marks 6
(A.D. 900-
1470) production
sizes.
Bricklayers dif-
mine how much control the state had over its
monitored. ferent from the production. The archaeological evidence for
brickmakers.
centralized control of production and distribu-
tion would include pottery or metal workshops
Table 11. Comparison of hypothetical Moche, Sicn
and administrative features such as audiencias
and Chim processes of brick production and construc-
and storerooms. Indeed, researchers working on
tion.
the north coast of Peru have already proposed
that the Chim state was more centralized with
The scenario I have proposed can be tested
more bureaucratic control over its labor and
in various ways. First, when different communi-
resources than the preceding Moche states
ties come together to build one large monument,
(Topic 2003; Topic 1982:282-283). We could
I would expect some kind of division to distin-
ANDEAN PAST 10 (2012) - 148
extend this line of inquiry into coastal valleys to make sure that they would receive credit for
beyond the Moche heartland, such as Nepea, their labor. Moche state officials then received
Casma, or Huarmey. the bricks at the site of construction and over-
saw the completion of a building.
CONCLUSION Despite the one hundred kilometer and
seven hundred year gap between Huaca de la
Kroeber (1930:61) suggested that, during the Luna and Farfn, there is no a priori reason to
construction of Chimu edifices, each contin- assume that construction methods at the two
gent of a community had built its own wall or sites were different or the same. It is not ade-
column. He also noted the difference between quate (nor accurate; see Topic 1982:280-282)
northern and southern Peru: most Chimu to refer to the Chim as another ethnic group
edifices appear to have been largely reared as to explain its differences with the Moche, since
units rather than by the gradual accretion which we still need to account for the origin of Chim
is so often indicated in central and southern principles of adobe production and construc-
Peru (ibid.). The study of adobe construction tion, or any other Chim practices, which, like
and its relation to labor organization on the all cultural traits, did not emerge out of no-
north coast of Peru was continued by Hastings where. To address this lacuna in our data, I
and Moseley (1975), Kolata (1978), Cavallaro recommend a continuation of the present study
and Shimada (1988), and Moubarac (2002). by investigating labor organization in the valleys
from Jequetepeque to Moche, spanning the
Moseleys (1975) model is often cited in Early Intermediate Period to the Late Horizon.
overviews of Moche society (Bawden 1996:104- Such long-term investigation of adobe construc-
105; Stone-Miller 1995:90-91; Wilson 1999: tion and labor organization has already been
387), but many scholars who work in the Andes initiated in the Lambayeque region (Shimada
do not know that makers marks were not used 1997).
at later sites like Chan Chan, or that there is a
diversity of makers marks and brick sizes within Future studies of adobe variability should
one segment of a Sicn pyramid. Subsequent measure an adequate sample of bricks from
changes in adobe production and construction different parts of each site and document the
among polities such as the Sicn and Chim exact locale of the bricks. Many more bricks
indicate a restructuring of labor organization by need to be measured in situ, especially from
second- or third-generation states on the north Huaca de la Luna. Then we should collect
coast of Peru. similar data from other sites from different
valleys and periods. Tschauner (2001:787-809)
I suggest that, at Farfn, Chim officials has measured adobes at several sites in the
supervised the entire process of brick production Lambayeque region and Reindel (1993) has
and construction. Because Chim officials could surveyed the architecture of numerous sites on
witness and count the number of bricks made by the north coast of Peru, though both studies
each community, makers marks or other types lack detailed intra-site brick data. Because
of labor record were not needed. By contrast, various adobe forms coexist in time and space
Moche state officials did not supervise the (Shimada 1990:355), it is important to measure
production of bricks. Local communities made an adequate sample of bricks. Unless the bricks
bricks away from the site of construction and are perfectly standardized, brick dimension will
away from the supervision of officials. Moche vary across the site. My recommendation is to
adobe workshops therefore marked their bricks sample bricks from different parts of the site,
149 - Tsai: Adobe Bricks and Labor Organization
with at least 100 bricks from each part to assess from different sites, in different contexts (monu-
adobe variability across space. I would also ment construction, ceramic production, etc.),
suggest investigating adobe variability through and from different regions and periods. Only
time (in other words, between different cultural then can we begin to understand the relation-
periods), which would require the investigator to ship between the type of labor organization and
know when parts of a site were constructed. It the nature of the sociopolitical system.
would be interesting, for example, to study
adobes at sites dating to the Late Horizon to see ACKNOWLEDGMENTS
whether labor organization had changed under
This research was supported by a National Science
Inca administration. Foundation Graduate Fellowship with additional funding
provided by the International Institute and the Depart-
To understand adobe variability within the ment of Anthropology of the University of Michigan. I
site, adobe sampling cuts, like the ones made in thank Carol Mackey and Santiago Uceda for generously
this study, should be implemented. The investi- allowing me to collect adobe data from Farfn and Huaca
de la Luna. I am grateful to Raffael Cavallaro, Claude
gator should open sections of architecture, Chapdelaine, Robyn Cutright, Tom Dillehay, Cameron
observe the construction method, and record Gokee, Carol Mackey, Michael Moseley, John OShea, Jeff
the adobe dimensions in situ. Documenting Parsons, Kenny Sims, Charles Stanish, Henry Wright, and
adobes in situ will allow an investigator to detect anonymous reviewers for their comments and suggestions
on earlier drafts of this paper. I thank Carol Mackey,
the presence (or absence) of vertical or horizon-
Csar Jaregui, and Enrique Zavaleta for their guidance
tal segments of construction. Furthermore, one and support during my field-work at Huaca de la Luna and
should try to look for differences between bricks Farfn. My deepest thanks go to Joyce Marcus for her
in horizontal and vertical segments, what the encouragement and assistance, without which the publi-
nature of that difference is, and what that might cation of this paper would not have been possible.
tell us about labor organization.
REFERENCES CITED
Keatinge, Richard W. and Geoffrey W. Conrad edited by Joyce Marcus and Patrick Ryan Wil-
1983 Imperialistic Expansion in Peruvian Prehistory: liams, pp. 325-349. Monograph 63. Cotsen Insti-
Chimu Administration of a Conquered Territory. tute of Archaeology, University of California,
Journal of Field Archaeology 10(3):255-283. Los Angeles.
Kolata, Alan L. Mackey, Carol J. and Csar Jaregui.
1978 Chan Chan: The Form of the City in Time. Ph.D. 2001 Informe preliminar de Proyecto Arqueolgico Far-
dissertation, Harvard University, Cambridge, fn. Preliminary report submitted to the Instituto
Massachusetts. Nacional de Cultura, Lima (now the Ministerio
1982 Chronology and Settlement Growth at Chan de Cultua). Copies available from the Ministerio
Chan. In Chan Chan: Andean Desert City, edited de Cultura, Trujillo, Per.
by Michael E. Moseley and Kent C. Day, pp. 67- 2004 Informe preliminar de Proyecto Arqueolgico Far-
85. Albuquerque: University of New Mexico fn. Preliminary report submitted to the Instituto
Press. Nacional de Cultura, Lima (now the Ministerio
Kosok, Paul de Cultua). Copies available from the Ministerio
1965 Life, Land and Water in Ancient Peru. New York, de Cultura, Trujillo, Per.
New York: Long Island University Press. McClelland, Donald H.
Kroeber, Alfred L. 1986 Brick Seriation at Pacatnamu. In The Pacatnamu
1930 Archaeological Explorations in Peru, Part II: The Papers, Volume 1, edited by Christopher B.
North Coast. Chicago: Field Museum of Natural Donnan and Guillermo A. Cock, pp. 27-46.
History. Museum of Cultural History, University of
Lechtman, Heather California, Los Angeles.
1976 A Metallurgical Site Survey in the Peruvian Meneses, Susana and Luis Chero
Andes. Journal of Field Archaeology 3(1):1-42. 1994 La arquitectura. In Sipn: Descubrimiento e
LeVine, Terry Yarov investigacin, by Walter Alva, Liliana Checa de
1987 Inka Labor Service at the Regional Level: The Costa, and Susana Meneses, pp. 181-188. Lima:
Functional Reality. Ethnohistory 34(1):14-46. privately published by Walter Alva.
Lockard, Gregory D. Moore, Jerry D.
2008 A New View of Galindo: Results of the Galindo 1996 Architecture and Power in the Ancient Andes: The
Archaeological Project. In Arqueologa mochica: Archaeology of Public Buildings. Cambridge:
Nuevos enfoques, edited by Luis Jaime Castillo Cambridge University Press.
Butters, Hlne Bernier, Gregory D. Lockard, Moore, Jerry D. and Carol J. Mackey
and Julio Rucabado Yong, pp. 275-294. Lima: 2008 The Chim Empire. In The Handbook of South
Pontificia Universidad Catlica del Per. American Archaeology, edited by Helaine Silver-
Mackey, Carol J. man and William H. Isbell, pp. 783-807. New
2003 La transformacin socioeconmica de Farfn York: Springer.
bajo el gobierno inka. In Identidad y trans- Morales, Ricardo
formacin en el Tawantinsuyu y en los Andes 2003 Iconografa litrgica y contexto arquitectnico
coloniales: Perspectivas arqueolgicas y etno- en Huaca de la Luna, valle de Moche. In Moche:
histricas; Segunda parte, edited by Peter Kaulicke, Hacia el final del milenio, Volume 1, edited by
Gary Urton, and Ian Farrington, pp. 321-353. Santiago Uceda and Elas Mujica, pp. 425-476.
Boletn de Arqueologa PUCP 7. Lima: Pontificia Lima: Pontificia Universidad Catlica del Per.
Universidad Catlica del Per. Moseley, Michael E.
2004 La ocupacin de dos centros administrativos en 1975 Prehistoric Principles of Labor Organization in
el valle de Jequetepeque: El Algarrobal de Moro the Moche Valley, Peru. American Antiquity
y Farfn. In Desarrollo arqueolgico costa norte del 40(2):191-196.
Per, Volume 2, edited by Luis Valle lvarez, pp. 2001 The Incas and their Ancestors: The Archaeology of
75-88. Trujillo, Per: Ediciones Sin. Peru. Revised edition. London and New York:
2006 Elite Residences at Farfn: A Comparison of the Thames and Hudson.
Chim and Inka Occupations. In Palaces and Moubarac, Jean-Claude
Power in the Americas: From Peru to the Northwest 2002 La production dadobes dans lancienne capitale de
Coast, edited by Jessica Joyce Christie and Patri- ltat Moche, socit prcolombienne de la cte nord
cia Joan Sarro, pp. 313-352. Austin: University of du Prou. Master's thesis, Department of Anthro-
Texas Press. pology, Universit de Montral, Qubec, Can-
2009 Chim Statecraft in the Provinces. In Andean ada.
Civilization: A Tribute to Michael E. Moseley,
ANDEAN PAST 10 (2012) - 152
Figure 1: Map of the north coast of Peru. Archaeological sites are indicated by triangles.
Solid triangles indicate the sites where bricks discussed in this study were sampled.
155 - Tsai: Adobe Bricks and Labor Organization
Figure 3: Models of adobe construction and labor organization after Cavallaro and Shimada
(1988: figures 11-14): (A) Moseleys labor tax model, (B) task differentiation-labor tax model,
(C) discontinuous territoriality model, and (D) sponsor model (Models C and D produce similar
results). In Model A, the brick-makers and brick-layers come from the same community, whereas in
Model B, the brick-makers and brick-layers are two separate groups of laborers. In Models C and D,
(1) each workshop produces bricks of a unique size; (2) the patrons or sponsors then receive bricks from
various workshops and incise the bricks with their own marks or insignia;
(3) The bricks are then donated for the construction of a temple.
157 - Tsai: Adobe Bricks and Labor Organization
Figure 4: Map showing the location of Unit 16 at Huaca de la Luna. Small boxes with dashed borders
indicate areas where adobes were measured (redrawn from Zavaleta 2007: figure 2).
Figure 5: Bricks from three different locations, Northwest, South, and North,
within Unit 16 of Huaca de la Luna.
ANDEAN PAST 10 (2012) - 158
Figure 6: Group of bricks of the same size (31 by 22 by 13 cm) and makers mark from Unit 16,
Huaca de la Luna. The scale measures 20 centimeters.
Figure 7: Scatterplot of the lengths and widths of bricks sampled from Unit 16, Huaca de la Luna.
S represents bricks from the south side of Unit 16; N denotes bricks from the north;
and NW represents bricks from the northwest corner. The three ellipses with dashed borders represent
density ellipses (p=.95) drawn around the clusters as indicated by cluster analysis.
159 - Tsai: Adobe Bricks and Labor Organization
Figure 12: Cross-section of north perimeter wall, Compound IV, Farfn. Note header-stretcher
configuration of construction, stone boulder foundation, and wall tapering toward the top
(after Mackey and Jauregui 2001:Figure 8).
163 - Tsai: Adobe Bricks and Labor Organization
Figure 14: Scatterplot of the lengths and widths of adobes sampled from the storerooms of Compound II,
Farfn. The number 2 represents bricks from Storeroom 2; 3 represents bricks from Storeroom 3;
and 4 represents bricks from Storeroom 4. Ellipses with dashed borders represent density ellipses
(p=.95) drawn around the clusters as indicated by cluster analysis. Compare with Figure 7.
165 - Tsai: Adobe Bricks and Labor Organization
Figure 15: Histograms comparing length, width, and height (in centimeters)
of bricks sampled in Compound II, Farfn.
ANDEAN PAST 10 (2012) - 166
Figure 16: Northwest corner of Compound IV at Farfn, showing locations of excavation units
and the average length, width, and height of bricks recovered in the excavation units.
Figure 17: A section of the north wall of Compound IV, Farfn. Note layers of bricks
(indicated by arrows) with greater amounts of rock inclusions than adjacent layers.
167 - Tsai: Adobe Bricks and Labor Organization
Figure 19: Adobes from Layer 6 of Cut 3, Compound II, Farfn. The trowel points north.
Numbers on adobes correspond to Table 10. Note broken adobes used as fill.
ANDEAN PAST 10 (2012) - 168
Figure 20: Adobes from Layer 7 of Cut 3, Compound II, Farfn. The trowel points north.
Numbers on adobes correspond to Table 10. Note that Adobe 51 is slightly wider than adjacent bricks.
Figure 21: Adobes from Layer 8 of Cut 3, Compound II, Farfn. The trowel points north.
Numbers on adobes correspond to Table 10.
Note that Adobes 62-64 are slightly longer and narrower than other bricks.
169 - Tsai: Adobe Bricks and Labor Organization
Figure 22: Hypothetical diagram comparing the three stages of making, drying, and laying adobes by
Moche and Chim masons.
A BIOARCHAEOLOGICAL STUDY OF COCA USE AND COCA LEAF CHEWING
AT PURUCHUCO-HUAQUERONES, PERU
MELISSA S. MURPHY
(University of Wyoming)
and
MARA FERNANDA BOZA
(University of Wyoming)
specialists or shamans chewing coca, evince the Administrative and court documents also con-
restricted access of commoners to coca and coca tain references to coca consumption and pro-
consumption (Calatayud and Gonzalez 2003: duction, but they, too, should be critically ques-
1504; Martin 1970:424-426; Naranjo 1981:163- tioned because they present static perspectives
166). Coca bags, stone sculptures, lime, lime of what were arguably dynamic and dramatic
gourd containers, and coca itself have been processes of cultural change and these docu-
recovered from coastal and highland sites from ments may just as easily strive to conceal as to
Ecuador, Peru, Bolivia, and Chile.1 Several reveal (Julien 1993:178-179).
mummies from coastal sites have been recovered
with coca quids in their cheeks (Arriaza et al. Several mid- to late sixteenth century ac-
1995:39; Aufderheide 2003:157; Cartmell et counts describe the high value of coca, its re-
al.1991:161, figure 1). While the archaeological stricted access, and the role it played in religious
evidence associated with prehistoric coca leaf rites.2 In his historical examination of coca
chewing is diverse and the practice appears prohibition, Gagliano reveals that the Inca
ubiquitous before the Late Horizon, few studies nobility were not the only people who consumed
have specifically addressed the access to coca coca, because the military, chasqui (couriers),
and the frequency of coca chewing in different the elderly, and anyone who received imperial
regions during the Inca empire and the sanction were reportedly permitted to chew it
bioarchaeological evidence for coca chewing (Gagliano 1994:16). Together, these ethno-
remains largely unexplored. historic sources present a picture of coca chew-
ing as a revered practice that was restricted to
Much of what we know about coca chewing the Inca elite or those with their permission, and
during the Inca empire comes from ethno- suggest that coca may have been prohibited for
historical sources, written from both Spanish the bulk of the Andean population.
and indigenous perspectives. However, each
written source offers accounts of different expe- Information contained in regional adminis-
riences, with different agendas, and with differ- trative documents typically describes the ubiq-
ent, but pervasive, prejudices about indigenous uity of coca plants and coca fields (Murra 1986).
peoples (Silverblatt 1987:xxii-xxv). Further- As a result, coca is interpreted as having been a
more, many of these accounts of Andean culture staple in many highland communities and,
were written several decades after the Spanish therefore, accessible to all (ibid.). Citing survey
conquest, and after the devastating effects of data collected in 1539 from the Pillkumayu
depopulation, Christian conversion, and colo- Valley in what is now the Hunuco Region of
nial resettlement programs, all of which had Peru, Murra asserts that every highland village
severely altered the lives of Andean people. possessed individuals responsible for managing
Therefore, scholars caution against the literal and cultivating tropical products, including
reading of these ethnohistorical sources because coca, and that these goods were part of the
they are likely to be distorted by the experience normal peasant repertoire in the Andes (ibid.:
of conquest and invasion, as well as the biases of 51). Furthermore, in 1572, for Sonqo, north of
their authors (Julien 1993:187; Urton 1990). modern La Paz, Bolivia, ethnohistorical docu-
1 2
Cassman et al. 2003:151; Engel 1963:77; Hastorf 1987; Acosta 2002 [1590]:210-211; Betanzos 1996 [1557]:42-
Indriati 1998:17-19; Kolata 1993:146; Lathrap 1973:180- 53, 56; Cobo 1964 [1653]: 473-474, 1990 [1653]:64,
181; Lathrap et al.1975: figures 61, 62, items 434-466; 65-66, 94, 119, 128, 137; Matienzo 1967 [1567], chapter
Menzel 1977: figure 128; Murphy 2004:119. 44, p. 163; Pizarro 1969 [1571]:262, 265.
173 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
ments report that everyone had access to food This brief review of the ethnohistorical
and to coca fields (ibid.; Murra 1991:572). literature reveals different perspectives on the
However, Juan de Matienzo, a Spanish adminis- consumption of, and access to, coca during and
trator familiar with coca production around after the Inca empire, with variability by region
Cusco and nearby regions, asserts that the Incas in the type and exertion of Inca control of coca
permitted only the nobility and members of the cultivation, distribution, and access possible.
military access to coca and commoners were After the Spanish conquest, some accounts of
proscribed from using it (Julien 1998:129; Inca coca use were colored by the controversial
Matienzo 1967 [1567]:163). colonial dialogues over whether or not the
stimulant should be eradicated because of its
With variable strategies, the Inca assumed association with indigenous religious practices,
control of, and regulated, the numerous coca or promoted because of the economic opportu-
plantations that were widely distributed along nities it provided in trade, and in its role in the
some of the middle and upper reaches of the exploitation of Andean farmers and the miners
Peruvian coastal valleys (e.g. Chilln, Chancay, at Potos (Cieza de Len 1984, La crnica del
and Ica) and in the highlands (e.g. Chuquioma, Per [1553]: Chapter 96, p. 121; Gagliano 1994;
a lowland valley on the eastern slopes of the Murra 1991:566; Parkerson 1983). Coca pro-
Andes northeast of Pocona, Bolivia, and south- duction and consumption grew dramatically in
east of modern Cochabamba (Julien 1998; conjunction with the Spanish intensification of
Murra 1980:32, 90-91, 144, 156, 177; Netherly agriculture and silver mining (Stern 1982:37).
1988:263-265; Rostworowski 1973:203-204).
While Inca imperial control of coca production BACKGROUND ON COCA CHEWING
and cultivation was well organized and main-
tained in some regions (Cieza de Len 1984, Before presenting the data and analyses from
Seorio de los Incas [1553]: Chapter 18, pp. 166- Puruchuco-Huaquerones, a brief review of coca
168), restrictive access and coca use prohibition chewing and oral health is in order. According
were more difficult to enforce in the northern to modern accounts, coca is usually chewed
coca fields and in the yungas, the warm, low- three to five times per day for approximately
lying valleys on the eastern slopes of the Andes 15-45 minutes (Allen 2002:105; Fuchs 1978:
(Gagliano 1994:17; Murra 1975:61- 65; Parker- 281; Indriati 1998:12; Plowman 1984:129).
son 1983:120; Rostworowski 1999:63-67). In a Although the frequency and distribution of coca
regional administrative survey (visita) document chewing vary by age, sex, and geographic region,
from 1558-1570 (Archivo General de Indias both men and women have been observed
Justicia 413), the ownership of the Quivi coca chewing coca, and children are first exposed to
fields in the Chilln Valley on the central coast coca during family rituals (Allen 2002:112;
of Peru was the subject of a dispute among Carter and Mamani 1986:141, 143, 156-57,
several different ethnic communities after the 257; Fuchs 1978:279; Hamner and Villegas
Spanish conquest (Marcus and Silva 1988; 1969: 289). Adolescent silver miners in Potos
Rostworowski 2004:131, 286-300, 308-310). have been observed chewing coca and preparing
Shortly after Inca expansion onto the central ritual offerings of coca for the devil that
coast, the coca plantations in the middle region resides in the mine (Davidson and Ladkani
(chaupiyunga) of the Lurn Valley were annexed 2005).
by Tupac Yupanqui and presented to the Yauyos
of Huarochiri (Rostworowski 2002:174). Research in human biology has focused on
the physiological effects of coca, but little has
ANDEAN PAST 10 (2012) - 174
been reported about the effects of coca chewing mortem loss of molars, cervical-root caries on
on the oral health of regular chewers. Through the buccal surface of premolars accompanying
the examination of human dental remains, the antemortem loss of adjacent molars, the
several possible characteristics were believed to presence of molar roots only, and interproximal
be associated with the practice of coca leaf caries (Indriati 1998:117; Indriati and Buikstra
chewing including unusual types of carious 2001:245). Testing these dental characteristics
lesions, brown staining on teeth, heavy calculus on archaeological samples of human remains,
formation, and alveolar recession/periodontal Indriati identified individuals based upon the
disease (Klepinger et al. 1977; Langsjoen 1996; degree of likelihood that they were coca leaf
Leigh 1937; Turner 1993). It was recently chewers (definite, probable, possible, or non-
concluded that calculus deposition is more likely chewer; Indriati 1998). The dental method for
the result of overall poor dental health, rather the identification of coca leaf chewing is partic-
than coca leaf chewing (Ubelaker and Stothert ularly valuable because teeth are often preserved
2006). Most of these dental consequences can in archaeological samples of human remains.
occur in the absence of coca leaf chewing, and
none of these studies examined the effects of the There are slight methodological differences
practice on living chewers. Cartmell et al. (1991) between Indriati (1998) and Indriati and Buik-
pioneered an alternative technique to the exam- stra (2001). We choose to follow the latter be-
ination of dental remains by using a radio- cause it is published in a peer-reviewed journal
immunoassay of hair samples from a sample of and is a more recent statement. However, when
mummified human remains to detect cocaine applying the Indriati and Buikstra method, we
and/or its metabolic product benzoylecgonine found that individuals who possessed 1-2 strong
(BZE). Unfortunately, the radioimmunoassay indicators, lacked mild indicators, and possessed
test requires the preservation of hair, a situation weak indicators (or just 2 strong indicators and
that is not always encountered in many geo- no other indicators, as was the case with HP01-
graphic areas and across all time periods in the 116) were not classified and they fell between
Central Andes. categories. In the absence of the strongest indi-
cators, we agree with Indriati and Buikstra that
Currently, the dental technique is the only these individuals should be classified as either
other method to detect coca use in human nonchewers, or depending on the number of
remains (Indriati 1998; Indriati and Buikstra mild indicators and the number of molars, as
2001). Indriati conducted the first large scale possible chewers. We have confidence in In-
investigation of dental health of chewers by driatis assessment of modern coca chewers and
examining the teeth of modern peoples from the the correlation of the strongest indicators with
highlands of Peru, Bolivia, and Chile and inter- these individuals (as a consequence of coca leaf
viewing them about their coca leaf chewing chewing). For consistency, we were comfortable
practices (Indriati 1998). Based on her dental classifying individuals with 1-2 of these strong
exams and interviews, she concluded that the indicators in the most conservative category,
strongest indicator of coca leaf chewing among possible chewer. Because there are several
the modern day samples was the presence of individuals who possessed strong indicators, but
large and wide cervical-root caries on the buccal lacked the requisite number of mild or weak
surface of mandibular molar teeth accompanied indicators, we think that the presence of 1-2
by root exposure (ibid.; Indriati and Buikstra cervical-root carious lesions on the buccal
2001:245). Mild and weak indicators included surface of the mandibular molars (alone or with
cervical-root caries on maxillary molars, ante-
175 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
several weak indicators) is sufficient to classify administrative centers were modified and uti-
them as possible chewers. lized by the Inca state (Eeckhout 2004; Wallace
1998). It is likely that future work will contrib-
The etiology of cervical-root caries is not ute to the emerging picture of the altered social
entirely understood. However, Indriati and and political landscapes after Inca colonization,
Buikstra hypothesize that the anesthetic quali- and of the variable strategies that were applied
ties of the cocaine metabolites released during in different regions (Covey 2000; DAltroy
coca leaf chewing act to suppress the activity of 2002:248-262, 2005; Dillehay 1977, 1979; Mal-
the salivary glands around the location of the pass 1993; Menzel 1959; Morris 1998; Sand-
coca quid (Indriati and Buikstra 2001:243). The weiss 1992).
reduction of the salivary flow results in dry
mouth, or xerostomia, which is correlated with The archaeological zone of Puruchuco-
a higher frequency of carious lesions in modern Huaquerones is 11.5 kilometers east of the
clinical contexts, particularly on the cervical modern city of Lima as measured from the city
and root surfaces (Epstein and Scully 1992; center, on the southern side of the lower middle
Guggenheimer and Moore 2003; Indriati and Rmac Valley. It contains architecture and
Buikstra 2001; Nrhi et al. 1999; Neville et al. several cemeteries (Figures 1, 2; Cock Carrasco
2002:346-346). 1999, 2001, 2006; Cock Carrasco and Goy-
cochea Daz 2004). One of the structures, the
To evaluate the different perspectives on palace at Puruchuco, may have been involved in
coca access and consumption during the Inca Inca administration of the region, based on the
empire, we can turn to the evidence for coca presence of imperial Inca architectural influ-
chewing and use from Puruchuco-Huaquerones. ences (Tabo 1965; Villacorta 2004:553). How-
ever, others have suggested that it was an elite
THE ARCHAEOLOGICAL CONTEXT OF residence or that it may have served as the
PURUCHUCO-HUAQUERONES center for the local polity (Lati), which was part
of the Ychma macroethnic group (Rostworowski
Tupac Yupanqui and his troops arrived on 2002:220-221). Recent mathematical decoding
the central coast circa A.D.1470, when they of 21 khipus from Puruchuco revealed hierarchi-
assumed control of Pachacamac and the Rmac cal accounting information passing between the
and Lurn Valleys, part of the Ychma macro- local elites and the provincial Inca lord, which
ethnic community (Cobo 1964[1653]: Book bolsters the assertion that the site may have
2:299-302; Rostworowski 2002:174). Recent played a role in Inca administration of the
work in nearby sites in the Rmac and neighbor- region (Urton and Brezine 2005:1067).
ing valleys has begun to identify the archaeologi-
cal correlates of the Inca presence on the central Burials from two of the cemeteries at Puru-
coast of Peru and the sociopolitical organization chuco-Huaquerones, Huaquerones and
of this region during the Late Intermediate 57AS03, are the subject of this study (Figure 2).
Period.3 Preliminary findings suggest a pattern of Previous work uncovered an area where addi-
Inca sociopolitical control in which pre-existing tional burials may be located, but this area has
not been systematically studied or scientifically
excavated (Tabo 1965). It is believed that the
3
Cornejo Guerrero 1999; Eeckhout 2004; Farfan Lobatn majority of burials in both cemeteries are Late
2004; Feltham and Eeckhout 2004; Makowski and Vega Horizon and date from the early presence of the
Centeno 2004; Marcone Flores 2004; Shimada et al. Inca in the Rimac Valley until shortly after the
2004; Vallejo Berrios 2004; Villacorta 2004.
ANDEAN PAST 10 (2012) - 176
arrival of the Europeans, or from approximately ings) in a pristine state of preservation. The
A.D. 1475 to 1540, based on the ceramic chro- burials from 57AS03 were affected by different
nology, the presence of Inca aryballos, and activities, specifically the use of the area for
stylistic patterns and details on textile bags animal grazing (sheep), the construction of
(Cock Carrasco 2006; Cock Carrasco and Avenida Javier Prado, the use of a drainage
Goycochea Daz 2004:185). channel next to the cemetery, and the increas-
ing encroachment of the neighborhoods and
The burials range in their preservation from buildings in the area.
large textile bundles containing mummified
individuals with soft tissue preservation to Despite the potential for postdepositional
completely skeletonized individuals. Several modification, the majority of the mortuary
factors affected the degree of preservation of the contexts are extraordinarily well preserved.
burial bundles, including type of bundle fill, Cock and Goycochea (2004:187-193) classified
presence or absence of textiles, and location the burials by their fill type, whether or not they
within the cemeteries. Individuals who possessed possessed a false head, and their size, but they
considerable textiles in their bundles were more caution that there is considerable variability in
likely to be mummified because the textiles drew the burial preparation, elaboration, and quantity
and absorbed the decomposition fluids, as well and quality of the funerary offerings. For sum-
as some of the moisture that may have seeped mary purposes here, we will describe some of the
into the burial pit. Several types of burial fill common characteristics of a typical burial. The
were observed, including cotton, cotton with principal individual was usually arranged in a
cottonseeds, and cotton with other organic seated or flexed position within the bundle and
materials (ibid.). Those bundles that were filled was accompanied by internal offerings, such as
with a large amount of cotton were usually weaving implements, beans, corn, gourds, Spon-
better preserved because the raw cotton also dylus or scallop shells, textile bags, and metal
absorbed any moisture in the bundle. Bundles tweezers. A textile layer typically surrounded the
that contained bundle fill of cotton and cotton- individual and these internal offerings and,
seeds were often the most deteriorated and depending on how elaborate the burial was,
individuals in these bundles were more likely to additional offerings were sometimes placed
be skeletonized. within the bundle, but outside of this layer . The
next layers in the bundle were the bundle fill
In the Huaquerones cemetery, some areas and any additional textiles. An outer textile
were more affected by recent human activities layer surrounded the bundle fill and maintained
than others, which contributed to differential the integrity of the bundle. Additional offerings,
preservation of the bundles. For example, some such as ceramic vessels, weaving baskets,
areas were used for gray water disposal. Other wooden heddles, and other weaving implements,
areas were more protected because human were placed outside of the bundle, but within
occupation had not occurred in these areas. the burial pit (Cock 2002; Cock and Goycochea
Several areas were leveled for construction, 2004). Some of the burials f r o m
which caused the underlying burials to be closer Puruchuco-Huaquerones contain internal
to the modern surface. Those burials from offerings of coca leaves, gourds containing lime,
greatly modified areas were more likely to con- and coca bags, suggesting these individuals may
tain skeletonized individuals (versus mummified have been coca chewers during their lifetimes.
or partially mummified individuals) and fewer
organic offering (e.g. gourds, textiles, food offer-
177 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
and severity of the calculus (Dobney and Broth- burials as elite, intermediate, commoner, or
well 1987). indeterminate based on their mortuary treat-
ment to explore whether or not coca was re-
Following the methodology of Indriati and stricted to those individuals with elaborate
Buikstra (2001:246), we identified the presence, mortuary treatment. This burial typology did not
absence, and frequency of the strong, mild, and specifically identify Inca royalty, provincial lords
weak indicators of coca chewing. Accordingly, or chiefs, but the mortuary variability present
the strongest indicator of coca leaf chewing is suggests that a degree of social differentiation
the presence of cervical-root caries on the existed in the cemetery. While it is unlikely that
buccal surface of the mandibular molars with these social strata were the same as those identi-
severe root exposure equal to or more than three fied and recognized by members of the commu-
mm. from cemento-enamel junction. Mild nity interred at Puruchuco-Huaquerones or
indicators include cervical-root caries on the even those recognized by Inca nobility, the
buccal surface of the maxillary molars, differentiation between the elites and the non-
cevical-root caries on the mesial, distal, and elites (intermediate and commoner) is the best
lingual surfaces of the maxillary and mandibular attempt we can make to capture differences
molars, cervical-root caries on the premolars among high and low status individuals, if they
when adjacent molars are lost, and buccal-pit existed.
caries. Weak indicators were the presence of
interproximal caries, the presence of molar roots For the present study, these burials were
only, and antemortem tooth loss. Individuals categorized into different social status groups
were then classified as definite, probable, and (elite, intermediate, and commoner) by inspec-
possible chewer or nonchewer according to the tion of their mortuary treatment and on the
presence and frequency of the strong, mild, and basis of the presence of metal, ceramics, Spon-
weak indicators (after Indriati and Buikstra dylus, and scallop shells (Table 2). Artifacts
2001). Definite chewers possessed at least two made of perishable or organic materials, such as
strong indicators and two other mild or weak weaving baskets, textiles, and food, were ex-
indicators. Probable chewers possessed one cluded from the burial typology due to their
strong indicator and one mild indicator. Possible differential preservation.
chewers possessed one strong indicator, or at
least one mild indicator, or several weak 57AS03 Huaquerones Total % total (N=
209)
indicators. Elite 0 22 22 9.5
Intermediate 14 23 37 17.7
Dental data were compared by age, sex, and Commoner 15 60 75 35.8
Atypical 50 0 50 23.9
mortuary treatment. Because a description of all Indetermi- 8 17 25 12.0
of the variables scored and analyzed is beyond nate
tions of faces, such as those from other Andean 57AS03 Huaquerones Total
vessels. A subsample of burials from 57AS03 Table 3: Comparison of dental health indicators between
resembled the commoner burials in their mortu- 57AS03 and Huaquerones.*
*Note: this comparison includes the full adult sample listed in Table 1.
ary treatment; however, the orientation of the
burials and the position of the bodies departed 57AS03 Huaquerones Total
from the Late Horizon pattern observed, so Total adults 85 116 201
these burials were classified as atypical burials. Males 49 65 114
Females 36 47 83
Some burials could not be classified because Indeterminate 0 4 4
they were in a poor state of preservation or Number of caries 276 416 692
Teeth with caries (%) 13.2% 15.8% 14.7%
because they were fully or partially disturbed, so Total individuals
they were categorized as indeterminate. The with caries 68/85 101/116 169/201
presence of coca, gourd lime containers (cale- Individuals with
caries (%) 80% 87.1% 84.1%
ros), and specific textile bags (chuspas) accompa- Average caries per 4.0 4.2 4.1
nied by either coca or lime gourds were consid- individual
Males with caries (%) 38 55 93
ered as mortuary offerings of coca or coca para- (55.9%) (54.5%) (55.0%)
phernalia. Females with caries 30
(44.1%)
42
(41.6%)
72
(42.6%)
Indeterminate with
RESULTS caries 0 4 4
Anterior caries 48 97 145
(% total caries) (17.4%) (23.3%) (21.0%)
In the total sample, 4,705 teeth and 6,398 Posterior caries 228 320 548
(% total caries) (82.6%) (76.9%) (79.2%)
sockets were examined. Approximately 882
Table 4. Comparison of carious lesions between 57AS03
teeth were lost antemortem. Of the total teeth and Huaquerones.
examined from the two cemeteries, 14.7%
possessed carious lesions, and 84.1% of the In the total adult sample, more males than
individuals examined possessed carious lesions. females possessed antemortem tooth loss (52.2%
Individuals averaged 4.1 carious lesions each versus 46.7%) and the number of males with
(Tables 3, 4). carious lesions was higher than the number of
females with carious lesions (55% versus 42.6%;
Table 4). Approximately 87.1% individuals from
the Huaquerones cemetery exhibited at least
one carious lesion and individuals averaged 4.1
lesions each. The frequency of carious lesions
ANDEAN PAST 10 (2012) - 180
between Huaquerones males and females was Several individuals exhibited distinctive
very similar (54.5% and 41.6% respectively) and carious lesions on the buccal aspect of the
chi-square tests reveal that there are no statisti- cervical root region of the mandibular molars,
cally significant differences between males and considered the strongest indicator of coca leaf
females (p>.05). In the adults from 57AS03, chewing. The buccal cervical root carious le-
approximately 80% possessed carious lesions and sions ranged in severity from incipient lesions to
individuals averaged 4.1 carious lesions (55.9% large, wide lesions extending down the root of
of the males and 44.1% of the females), but the molars (Figures 3, 4). Typically, the cervical
these differences are not statistically significant root carious lesions were accompanied by several
(p>.05). The frequency of posterior caries is of the mild and weak indicators of coca chewing
much greater than the frequency of anterior (Indriati and Buikstra 2001). Following Indriati
caries, which is not unexpected given the greater and Buikstras framework for identifying coca
number of posterior teeth (molars and premol- chewing in skeletal samples, 32 individuals
ars) and the greater likelihood that the fissures (32/201; 15.9%) were identified as definite
and cusp morphology on the occlusal surfaces of chewers, probable chewers, and possible chewers
the posterior teeth will contribute to the growth of coca on the basis of presence and frequency of
of cariogenic bacteria and carious lesions. the strong, mild, and weak indicators of coca
leaf chewing (Tables 6, 7).
Approximately 57.5% of the individuals
from 57AS03 experienced antemortem tooth 57AS03 Huaquerones Total
loss and 71.3% of the individuals from Huaque- Total chewers (%) 11/85 21/116 32/201
(12.9%) (18.1%) (15.9%)
rones experienced antemortem tooth loss. In Male chewers 3 8 11
both cemeteries, males and females experienced Female chewers 8 12 20
antemortem tooth loss in nearly equal numbers Indeterminate chewers 0 1 1
Definite 5 4 9
(Table 3) and these differences are not statisti- Probable 3 4 7
cally significant (57AS03 p>.05, Huaquerones Possible 3 13 16
Individuals with offerings 1 8 9
p>.05). Individuals buried in 57AS03 averaged Chewers with offerings 0 2 2
3.6 teeth lost antemortem and those buried in Chewers - Elite 0 4 4
Huaquerones lost an average of 4.7 teeth ante- Chewers - Intermediate 2 1 3
Chewers - Commoner 5 13 18
mortem. As expected, due to the age-progressive Chewers - Atypical 1 0 1
nature of tooth decay, the middle adult and Chewers - Indeterminate 3 3 6
Of these 32 individuals, eight were identified majority of individuals with offerings of coca or
as definite coca leaf chewers, seven were identi- coca paraphernalia were male (six males, three
fied as probable coca leaf chewers, and 17 were females) and eight of them were interred in the
identified as possible coca chewers. Eleven of the Huaquerones cemetery and one was interred in
total coca chewers were males, 20 were females 57AS03.
and one was of indeterminate sex. These differ-
ences are not statistically significant (p>.05).
More individuals were identified as coca chew-
ers from the Huaquerones cemetery than from
57AS03 (21 versus 11; 18.1% and 12.9%, re-
spectively). Nine individuals were interred with
mortuary offerings of coca and/or coca para-
phernalia (Figure 5, Table 8). Most of these
individuals were classified as nonchewers and
only two chewers possessed coca or coca para-
phernalia among their mortuary offerings. The
ANDEAN PAST 10 (2012) - 182
Individual Age, yrs Sex Coca leaf chewer? Type of offering
HP01-122 35-49 M Definite Textile bag, lime gourd container Elite
HP02-157 35-49 F Possible Textile bag, lime gourd container Intermediate
HP01-40 35-49 F Non-chewer Textile bag, four lime gourd containers Intermediate
HP01-70 20-34 M Non-chewer Textile bag, lime gourd container, coca Elite
HP01-88 35-49 M Non-chewer Textile bag with lime Intermediate
HP01-125 35-49 M Non-chewer Lime gourd with container Intermediate
HP02-206 50+ M Non-chewer Textile bag with lime gourd container inside Elite
HP01-14 35-49 F Non-chewer Coca leaves Commoner
57AS03E 081 25-49 M Non-chewer Coca leaves Atypical
Table 8. Individuals with mortuary offerings of coca and/or coca paraphernalia
When we examine the mortuary treatment However, the majority of the samples were
of the chewers, most of the individuals with from the southern coastal and southern high-
dental evidence of coca leaf chewing were land regions (Indriati 1998). Using the radio-
classified as commoner burials (18 individuals; immunoassay technique, the individuals who
Table 7). Three intermediate burials and four tested positive for coca use in the samples ana-
elite burials possessed dental evidence of coca lyzed by Cartmell and colleagues (1991) ranged
leaf chewing. One chewer was classified as an from zero percent to close to 69%, with the
atypical burial and the poor preservation of the highest frequency of coca use detected in their
funerary contexts of six chewers did not allow provincial Inca sample (69.2%; Table 9). A
for classification (indeterminate). Only two of similar low prevalence was reported from hair
the coca chewers possessed mortuary offerings of samples of non-elite Lambayeque burials from
coca or coca paraphernalia and they were both the coastal site of El Brujo that were analyzed by
from the Huaquerones cemetery. Indriatis radioimmunoassay and only tested positive for
(1998) study found evidence of coca chewing in coca use in 17% (4/23) of the cases sampled
40-80% of her sample, which is significantly (Verano 2000).
higher than was found in this study (15.9%,
Table 9).
DISCUSSION AND CONCLUSIONS coca, which may explain the higher number of
female coca chewers. Other studies have re-
The overall prevalence of coca leaf chewing ported higher numbers of coca chewing among
found at Puruchuco-Huaquerones is low in females than males, but the results were not
comparison with that of other studies of the statistically significant (Cartmell et al. 1991:
practice. This evidence suggests limited access to 265; Indriati and Buikstra 2001:255).
coca, but we believe access was unrelated to
social status and not restricted to the elite The patterning of offerings of coca and coca
members of the community. The coca planta- paraphernalia contrasts with the dental evi-
tions in the upper Lurn Valley were annexed by dence of coca chewing. Mortuary offerings of
Tupac Yupanqui and given to the Yauyos of coca, lime gourds, and textile bags, although
Huarochiri during the Inca expansion onto the present, were uncommon (nine individuals) and
central coast (Rostworowski 2002:90, 174). were recovered more among the burials of adult
Therefore, members of the Ychma macroethnic males than females (six males, three females).
group living in the middle and lower Rimac and Most of these individuals were also classified as
Lurn Valleys and, specifically, people from elite or intermediate burials, so they also re-
Puruchuco-Huaquerones, lost control of their ceived a more elaborate mortuary treatment that
coca plantations and may have experienced a included a higher quantity and quality of offer-
reduction in access to the plant. It is unclear ings (Table 8). Only two individuals with coca
how local elites or Inca administrators in the or coca paraphernalia were identified as coca
region distributed the available coca, but it may chewers. In the absence of dental evidence of
have been distributed as part of mita labor the practice, these offerings likely highlight the
obligation (Hastorf 1991:151-152; Murra 1982). symbolic importance of coca and paraphernalia,
A high number of the identified coca chewers particularly for men in this community. Dental
were female (20/32, 62.5%) and it is conceivable evidence of the practice of coca leaf chewing at
that coca may have been distributed to these Puruchuco-Huaquerones indicates that males
women during mita labor, such as weaving or and females chewed coca, but that mortuary
chicha production. This pattern is particularly offerings were usually reserved for males. Al-
interesting because the frequency of carious though the sample size of individuals with
lesions is higher in males than in females in the offerings of coca or coca paraphernalia is small
entire sample (93 males versus 72 females), but in the present study, it is known that during
more females than males were identified as coca seventeenth and eighteenth century Andean
chewers (20 females, 11 males, one indetermi- funerary rites, mourners and family members of
nate). There is a significant difference in carbon the deceased reportedly offered coca and coca
isotope composition of dietary protein between paraphernalia (Doyle 1988), so the objects could
males and females at Puruchuco-Huaquerones, simply reflect ceremonial offerings that say little
indicating that males had differential access to about the practice of coca chewing by the de-
animal protein fed C4 fodder, or that males may ceased. Contrary to what we would expect based
have consumed higher quantities of chicha on some ethnohistorical sources (Acosta 2002
(Williams 2004: 157, 160). Differential con- [1590]; Cobo 1964 [1653]; Matienzo 1967
sumption of food and coca could have occurred [1567]; Pizarro 1969 [1571]), this study of coca
for male and female participants as a result of a leaf chewing demonstrates that access to coca
gendered division of labor (Hastorf 1991:149- was not restricted to the elite individuals from
151; Murra 1982:256), where males may have Puruchuco-Huaquerones. Only four chewers
received meat and chicha, and females received were classified as elite individuals (4/32, 12.5%).
ANDEAN PAST 10 (2012) - 184
Individuals identified as possible, probable, or infrequent chewers, may have died too young to
definite coca chewers were not correlated with develop the pathological conditions, and/or may
one particular burial type, nor were these indi- have possessed a higher resistance to the distinc-
viduals found only among the more complex tive dental pathologies. These issues will be
mortuary contexts and treatments. Rather, explored through the analysis of hair of individ-
individuals identified as coca chewers were uals from Puruchuco-Huaquerones through
interred in all of the different burial contexts radioimmunoassay tests. These findings will
and most of the identified coca leaf chewers further clarify the frequency of the practice of
were from commoner burials in the typology coca leaf chewing for members of this coastal
used in this study (18/32, 56.3%). community during the Late Horizon.
A higher frequency of coca leaf chewing has The bioarchaeological data from Puru-
been observed in archaeological samples from chuco-Huaquerones do not indicate that coca
highland regions and south coastal regions use was widespread in this region during the
(Table 9; Cartmell et al. 1991; Indriati 1998; Late Horizon or that the practice of coca chew-
Indriati and Buikstra 2001). Some of the coca ing was restricted only to elites. Despite the low
leaf chewers from Puruchuco-Huaquerones may prevalence, its association with individuals from
have possessed some highland affinities, either elite, intermediate, and commoner burials
through intermarriage, trade, or ethnicity. It is supports the view that everyone had access to
possible that the chewers were members of the leaf (Murra 1986, 1991). In comparison to
communities or an ethnic group or groups other Andean archaeological populations, the
relocated to this area of the central coast, possi- population at Puruchuco- Huaquerones exhib-
bly from the highlands, who continued to chew ited a relatively low incidence of the practice,
coca in this coastal region. Dental metric data and few individuals were interred with offerings
indicate that the community from Puruchuco- of coca or coca paraphernalia. These results may
Huaquerones is characterized by considerable reflect a change in access to coca resulting from
phenotypic variability and the samples from the loss of coca plantations to the Yauyos when
Huaquerones and 57AS03 show similar levels of the Incas arrived on the central coast. However,
phenotypic similarity (Murphy et al. 2008). In access to the leaf was sufficient for a sample of
the future, we will investigate residence and individuals to chew it frequently and for coca
migration patterns through the analysis of and coca paraphernalia to be included as offer-
oxygen and strontium isotope ratios, which may ings during the funerary rites of some individu-
help reveal if these chewers have highland als. If Inca imperial control was exercised
affinities. through diplomacy and alliance building with
local elites and/or indirect measures in the lower
Unfortunately, the results of this study middle Rimac Valley, as suggested by Villacorta
cannot address the ceremonial use of coca, its (2004) and others, then coca use at Puru-
infrequent use, or differential susceptibility to chuco-Huaquerones may reflect less restrictive
the dental consequences. Individuals who sel- strategies of control and access to coca by the
dom or rarely chewed coca, only participating in local elites in this region than what was experi-
occasional sacred rituals, may not evince the enced by other regions. Another possible expla-
dental consequences of the practice. For exam- nation is that the low frequency of coca chewers
ple, individuals who were interred with coca and and coca use may reflect a coastal pattern that
coca paraphernalia, but who lack dental evi- was less frequent among people living at sea
dence of coca leaf chewing, may have been level and some of these identified chewers may
185 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
actually represent highlanders living in the Arriaza, Bernardo T., William Salo, Arthur C. Aufder-
heterogeneous community of Puruchuco-Hua- heide, and Thomas Holcomb
1995 Pre-Columbian Tuberculosis in Northern Chile:
querones. Finally, the debate over access to coca Molecular and Skeletal Evidence. American
may have been slightly overstated due to the Journal of Physical Anthropology 98(1): 37- 45.
overemphasis on those ethnohistorical sources Aufderheide, Arthur C.
that are removed from the daily realities of 2003 The Scientific Study of Mummies. Cambridge and
provincial life. Additional data from Late Inter- New York: Cambridge University Press.
Bass, William M.
mediate and Late Horizon coastal sites and 2005 Human Osteology: A Laboratory Manual and Field
bioarchaeological samples would greatly illumi- Manual. Fifth Edition. Columbia, Missouri:
nate whether or not coca availability shifted Missouri Archaeological Society.
after the Inca established a presence on the Betanzos, Juan de
central coast, as well as the prevalence of coca 1996 [1557] Narrative of the Incas, edited and trans-
lated by Roland Hamilton and Dana Buchanan.
leaf chewing and coca access from different Austin: University of Texas Press.
regions of the north and central coasts. Buikstra, Jane and Douglas Ubelaker, editors
1994 Standards for Data Collection from Human
ACKNOWLEDGMENTS Skeletal Remains.: Arkansas Archeological Survey
Series 44 (Fayetteville Arkansas).
This research was supported in part by grants from Calatayud, Jess and ngel Gonzlez
Wenner Gren Foundation (Grant number 6791) and the 2003 History of the Development and Evolution of
National Science Foundation (number 0618192). This Local Anesthesia since the Coca Leaf. Anesthesi-
research would not have been possible without the ology 98:1503-1508.
tireless efforts of the Puruchuco-Huaquerones Project Carter, William E. and Mauricio Mamani
members (present and former), especially the Peruvian 1986 Coca en Bolivia. La Paz: Libera Editorial Juven-
researchers and scholars responsible for the long-term tud.
excavations and laboratory work: Guillermo Cock, Elena Cartmell, Larry W., Arthur C. Aufderheide, Angela
Goycochea, Violeta Chamorra, Bertha Herrera, and Springfield, Cheryl Weems, and Bernardo Arriaza
Antonio Gamonal. Jeffrey Quilter, Kate Moore, and 1991 The Frequency and Antiquity of Prehistoric
Andrew Scherer read and commented on earlier versions Coca-Leaf-Chewing Practices in Northern
of this manuscript, and despite their painstaking editorial Chile: Radioimmunoassay of a Cocaine Metabo-
labors and astute advice, all of the mistakes are our own. lite in Human-Mummy Hair. Latin American
Catherine Riihimaki and Natalia Ventocilla deserve Antiquity 2(3):260-268.
acknowledgment for their assistance with Figure 1. We Cassman, Vicky, Lawrence Cartmell, and Eliana Bel-
would also like to thank the two anonymous reviewers monte
who provided us with constructive and helpful feedback 2003 Coca as a Symbol and Labor Enhancer in the
and suggestions. Andes. In Drugs, Labor, and the Colonial Expan-
sion, edited by William Jankowiak and Daniel
REFERENCES CITED Bradburd, pp. 149-158. Tucson: University of
Arizona Press.
Acosta, Jos de Cieza de Len, Pedro
2002 [1590] Natural and Moral History of the Indies, 1984 [1553] Obras Completas, Volume 1, edited by
edited by Jane E. Mangan, with an introduction Carmelo Saenz de Santa Mara. Monumento
and commentary by Walter D. Miglolo, and Hispano-Indgena II.1. Madrid: Consejo Superior
translated by Frances M. Lpez-Morillas. Dur- de Investigaciones Cientficas, Instituto de
ham, North Carolina: Duke University Press. Gonzalo Fernndez de Oviedo.
Allen, Catherine Cobo, Bernab
2002 The Hold Life Has. Washington: D.C.: Smithso- 1964 [1653] Historia del nuevo mundo. Madrid: Biblio-
nian Institution Press (second edition; first teca de Autores Espanoles, Madrid.
edition 1988). 1990 [1653] Inca Religion and Customs, translated by
Roland Hamilton. Austin: University of Texas
Press.
ANDEAN PAST 10 (2012) - 186
Cock Carrasco, Guillermo A. Dillehay, Tom, Patricia J. Netherley, and Jack Rossen
1999 Estudio de Evaluacin arqueolgica del rea 1989 Middle Preceramic Public and Residential Sites
ocupada por el asentamiento humano Tpac on the Forested Slope of the Western Andes,
Amaru en el sitio arqueolgico de Hua- Northern Peru. American Antiquity 54(4):733-
querones-Puruchuco, Distrito de Ate-Vitarte, 759.
Prov. y Dpto. de Lima. Report on file, Instituto Dobney, Keith and Don Brothwell
Nacional de Cultura del Per, Lima (now the 1987 Method for Evaluating the Amount of Dental
Ministerio de Cultura). Calculus on Teeth from Archaeological Sites.
2001 Proyecto de rescate arqueolgica y ampliacin Journal of Archaeological Science 14(4):343-351.
de investigaciones en el Cementerio de Donnan, Christopher B.
Huaquerones-Puruchuco, en el rea ocupada 1978 Moche Art of Peru. Los Angeles: Regents of the
por el asentamiento humano Tpac Amaru. University of California.
Report on file, Instituto Nacional de Cultura del Donnan, Christopher B. and Nathan Benn
Per, Lima (now the Ministerio de Cultura). 1990 Masterworks of Art Reveal a Remarkable Pre-
2002 Inca Rescue. National Geographic 201(5):78-89. Inca World. National Geographic 177:17-35.
2006 Proyecto de recuperacin de contextos fu- Doyle, Mary
nerarios en el Cementerio 57AS03zona 1988 The Ancestor Cult and Burial Ritual in Seventeenth
arqueolgica Puruchuco-Huaquerones. Report and Eighteenth Century Central Peru. Ph.D.
on file, Instituto Nacional de Cultura del Per, dissertation, University of California, Los An-
Lima (now the Ministerio de Cultura). geles.
Cock Carrasco, Guillermo A. and Carlos E. Goycochea Eeckhout, Peter
Daz 2004 La sombra de Ychsma: Ensayo introductorio
2004 Puruchuco y el cementerio inca de la Quebrada sobre la arqueologa de la costa central del Per
de Huaquerones. In Puruchuco y la sociedad de en los perodos tardos. Bulletin de lInstitut Fran-
Lima: Un homenaje a Arturo Jimnez Borja, edited ais dtudes Andines 33(3):403-423.
by Luis Felipe Villacorta Ostolaza, Luisa Mara Engel, Fdric
Vetter Parodi, and Carlos E. Ausejo Castillo, pp. 1963 A Preceramic Settlement on the Central Coast
179-197. Lima: CONCYTEC. of Peru: Asia, Unit 1. Transactions of the Ameri-
Cornejo Guerrero, Miguel A. can Philosophical Society 53(3):3-139.
1999 An Archaeological Analysis of an Inka Province: Epstein, Joel B. and Crispian Scully
Pachacamac and the Ischma Nation of the Central 1992 The Role of Saliva in Dental Health and the
Coast of Peru. Ph.D. dissertation, Australian Causes and Effects of Xerostomia. Journal of the
National University, Canberra. Canadian Dental Association 58(3):217- 221.
Covey, R. Alan Espinoza Soriano, Waldemar
2000 Inka Administration of the Far South Coast of 1963 La Guaranga y la reduccin de Huancayo. Re-
Peru. Latin American Antiquity 11(2):119-138. vista del Museo Nacional 32:8-80.
DAltroy, Terence N. Farfn Lobatn, Carlos
2002 The Incas. Malden, U.K.:Blackwell Publishers. 2004 Aspectos simblicos de las pirmides con rampa:
2005 Remaking the Social Landscape: Colonization Ensayo interpretativo. Bulletin de lInstitut Fran-
in the Inka Empire. In The Archaeology of Colo- ais dtudes Andines 33(3):449-464.
nial Encounters: Comparative Perspectives, edited Feltham, Jane and Peter Eeckhout
by Gil J. Stein, pp. 263-297. School of American 2004 Hacia una definicin del estilo ychsma: Aportes
Research Advanced Seminar Series. Santa Fe, preliminares sobre la cermica ychsma tarda de
New Mexico: School of American Research and la pirmide III de Pachacamac. Bulletin de lInsti-
Oxford: James Currey. tut Franais dtudes Andines 33(3):634-679.
Davidson, Kief and Richard Ladkani Fleming, Stuart J., W. Miller, and J. L. Brahin
2005 The Devil's Miner. A film by Urban Landscapes 1983 The Mummies of Pachacamac, Peru. MASCA
Productions, Los Angeles, California. Journal 2(5):137-156.
Dillehay, Tom Fuchs, Andrew
1977 Tawantinsuyu Integration of the Chilln Valley, 1978 Coca Chewing and High-Altitude Stress: Possi-
Peru: A Case of Inca Geo-Political Mastery. ble Effects of Coca Alkaloids on Erythropoiesis.
Journal of Field Archaeology 4(4):397- 405. Current Anthropology 19(2):277-291.
1979 Pre-Hispanic Resource Sharing in the Central
Andes. Science 204(4388):24- 31.
187 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
Figure 1: Location of the cemeteries of Huaquerones and 57AS03 relative to the Puruchuco Museum,
the Pyramids, and the Palace within the archaeological zone of Puruchuco-Huaquerones.
Inset: location of Puruchuco-Huaquerones in the Rmac Valley.
191 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
Figure 2: Location of 57AS03 and the Huaquerones cemetery within the archaeological zone of
Puruchuco-Huaquerones and relative to the Puruchuco Site Museum, the Pyramids with Ramps,
and the Puruchuco Palace
ANDEAN PAST 10 (2012) - 192
Figure 3: An adult individual (HP01-17) exhibits a deep and wide cervical root carious lesion on the
buccal aspect (A) with root exposure of the left mandibular first molar (B). These dental characteristics
are considered the strongest indicators of coca leaf chewing.
Figure 4: Example of an adult male individual (HP01-184) with two cervical root carious lesions
affecting the buccal aspect of his right first and second mandibular molars (arrows labeled A).
Root exposure (B) due to alveolar recession is also apparent.
The crown of the adjacent premolar has been lost to caries.
193 - Murphy & Boza: Coca Chewing at Puruchuco-Huaquerones
Figure 5: Ten individuals in the present study did not possess dental evidence of coca leaf chewing,
but they were interred with coca and coca paraphernalia. This lime gourd container was one of the
mortuary offerings accompanying an adult male individual from an elaborate false head mummy bundle.
The lime gourd was recovered from inside a coca bag (chuspa) that also contained a Spondylus shell
and coca leaves. Some of the coca leaves are still adhering to the exterior of the gourd.
The coca bag and its contents were placed underneath the principal individuals left arm.
THE DESTRUCTION OF THE YURAC RUMI SHRINE (VILCABAMBA, CUSCO DEPARTMENT)
Brian S. Bauer
University of Illinois at Chicago
George S. Burr
University of Arizona
The Vilcabamba region, however, gains its House of the Sun),5 various attendants, and a
unique importance from the role that it held in head priest (Figure 2).
the final days of the Incas. In 1537, after the
Incas failed to regain control of Cusco, their In 1568, while the Inca were still in control
capital city, from the Spaniards, many native of the Vilcabamba region, the ruling Inca (Titu
loyalists withdrew into this mountainous region. Cusi Yupanqui) invited the Augustinian priest,
Marcos Garcia, to enter the area and establish a
The leader of the resistance, Manco Inca, mission. About a year later a second Augustin-
established himself at the town of Vitcos,3 as he ian, Diego Ortiz, was also given permission to
attempted to restore control over what had been live in the area.6 As Titu Cusi contemplated a
the largest indigenous state to develop in the full conversion to Christianity7 the relationship
Americas.4 For nearly eight years Manco Inca between the Incas and the Christians were
both negotiated with the Spaniards and orga- stable, and the missions run by Ortiz and Garcia
nized raids against them. After his death, strong lived in an uneasy state of truce with the Yurac
indigenous resistance continued under the Rumi shrine and its attendants. However, in
successive leadership of three of Manco Incas 1570, angered by their recent poor treatment by
sons: Sayri Tupac, Titu Cusi, and Tupac Amaru. the Inca and emboldened by a growing number
During this same period, the Spanish leaders of of converts to Christianity, the two priests and
Cusco sent diplomats, priests, miners, and, at a small group of local people burned the Yurac
times, military expeditions into the Vilcabamba Rumi shrine complex (Calancha 1981[1638]:
region, attempting to bring an end to the con- 1826, Book 4, Chapter 4]; Mura 1987[1611-
flict (Gose 2008; Hemming 1970).
1616]: 270, Book 1, Chapter 75). In the wake of there. Unlike most other known shrine centers
this provocative move, the Inca himself had to in the Andes, the Yurac Rumi site remains
race to the shrine to restore order and to save relatively well preserved (Lee 2000; Samanez
the priests from being killed by a large and angry and Zapata 1996).
mob. The next year, however, Titu Cusi sud-
denly died while visiting Vitcos,8 and soon
afterwards, in April 1572, Viceroy Toledo OVERVIEW OF THE 2008 EXCAVATIONS
ordered a massive raid on the Vilcabamba
region which resulted in the capture of the new The Yurac Rumi shrine is in a small, but
Inca (Tupac Amaru), and a general depopula- steep-sided, river valley at an altitude of approx-
tion of the area soon followed. The city of imately 3,080 m.a.s.l. Now, as in Inca times, the
Vitcos was abandoned, and the important valley slopes are covered with dense vegetation,
rituals that had been performed at the shrine of while the valley floor has been cleared and
Yurac Rumi slowly faded from local memory. terraced. The complex is relatively small, less
than 80 meters square, and is some twenty
BINGHAM AT YURAC RUMI minutes by foot from the royal palace structures
of Vitcos. The sculpted rock of Yurac Rumi is in
Over the centuries, the forests of Vilca- the center of the shrine complex, and the spring
bamba reclaimed the city of Vitcos as well as its mentioned by the chroniclers still emerges from
nearby shrine. On August 9, 1911, less than a beneath it, and flows north into a small court-
month after he discovered Machu Picchu, yard framed by a suite of buildings (Figure 3).
Hiram Bingham entered the Vilcabamba region The area immediately to the south of the sacred
looking for the final capital of the Incas. After rock contains a large open plaza (approximately
visiting the ruins of Vitcos he was told of, and 25 by 20 meters), and a series of small carved
subsequently found, a large, nearby, carved rock rocks. The area to the southwest of the Yurac
which stood above a spring. Bingham cleared Rumi contains a second set of buildings with an
the forest around the rock the following day and interior courtyard and a large looters pit.
found various Inca buildings grouped in patio-
units beside it. He also found a host of smaller During June and July of 2008, Bauer and
carved rocks and an impressive terrace and Aroz Silva conducted field-work at Yurac
canal system beside the rock. Bingham (1912a, Rumi. The project included fourteen (one meter
1912 b, 1912c, 1913, 1922) immediately recog- by one meter) test units, as well as six larger
nized that this was the shrine complex of Yurac excavations.9 The goal of the smaller units was
Rumi. He returned to the shrine site the next to gain information on the stratigraphy of the
year (1912), drained the swamp at its base, and site, while the larger units were dug to more
conducted limited excavations. Finding little, fully investigate specific areas and buildings. At
and being drawn to working at the larger site of
Machu Picchu, Bingham left the rest of the site
untouched. Remarkably, no other archaeologi- 9
Major funding for this project was provided by the
cal projects have been conducted at Yurac National Geographic Society, the University of Illinois at
Rumi, and only limited looting has taken place Chicago, the Curtiss T. Brennan and Mary G. Brennan
Foundation, the Institute for New World Archaeology,
and the Accelerator Mass Spectrometry Laboratory at the
University of Arizona. We thank Vincent Lee for
8
The exact date of Titu Cusi Yupanquis death is not providing the building drawing used in this work, as well
known. However it most likely occurred in 1571, or early as Andean Past reviewers Steven A. Wernke and Steven
in 1572. Kosiba for their comments on improving the manuscript.
ANDEAN PAST 10 (2012) - 198
the conclusion of the field-work, a series of dark earth, followed by another floor of crushed
carbon samples were submitted for dating under white granite. Under the second granite floor
the direction of Burr at the Accelerator Mass was a thin layer of earthen fill used to level this
Spectrometry Laboratory at the University of area of the site, and a then thick stratum, more
Arizona. While a report on the full excavation than thirty centimeters deep, of very dark
is currently in preparation, here we discuss the brown, clayish loam which is thought to be the
contributions of the radiocarbon dates to our natural sediment of the valley before the Inca
understanding of the shrine complex. Our occupation (Figure 4).
research suggests that the shrine was established
during the early expansion of the Inca empire. Evidence of a similar level of crushed white
Furthermore, we find overwhelming evidence granite was found in P11. This unit, measuring
that the shrine was destroyed by fire, and that 1.0 by 1.2 meters was placed in the northeast
the ritual complex was then briefly reoccupied, sector of the site, against the back wall of the
before being completely abandoned. These building facing the sacred rock. The excavations
archaeological data strongly support the narra- soon revealed an irregular, hard-packed earthen
tive accounts given by Mura (1987 [1611- floor with large fragments of classic Inca pottery,
1616], Calancha (1981 [1638]), and Cobo carbon, and some minute burnt bone fragments.
(1979 [1653]). Beneath the earthen floor was a level fill, ap-
proximately twenty centimeters thick, which
DATING THE ESTABLISHMENT OF THE SHRINE rested above a two centimeter thick floor of
crushed white granite. This floor had been well
Excavations in the plaza and courtyards of prepared and rested directly above a thin,
the shrine documented that the open areas of greater than one centimeter, level of red clay,
the complex were covered with hard-packed which may have been used as a sealant to pro-
gravel floors during its final years of use. Fur- tect the granite floor in this waterlogged area of
thermore, excavations in the buildings revealed the site.10 Beneath the red clay, was a thick
that most of the structures contained well- stratum, fifty to sixty centimeters, of multiple fill
prepared, green clay floors. Importantly, in two levels which was used to raise the interior area
different units we also found evidence of earlier of the structure (Figure 5).
floors, made with crushed white granite.
There is little doubt that the crushed white
The crushed white granite floors were first granite in the floors identified in units P9 and
identified in Unit P9; a one by one meter exca- P11 was the direct by-product of the Incas
vation in the Southwest Courtyard. The excava- carving the many granite boulders that lie
tion revealed a humus level followed by a thick within the shrine complex. A carbon sample,
stratum of disturbed soil, representing the back- (AA83415) collected from within the upper
dirt of a nearby looters pit. The third stratum floor of crushed granite in P9, yielded an AMS
was composed of colluvially redeposited materi- radiocarbon date of 60134 BP. A carbon
als (i.e. slope wash) laid down after the site was sample, (AA83416) collected from the crushed
abandoned. Next came a series of closely granite floor in Unit P11, provided an AMS
spaced, relatively thin levels, each two to three date radiocarbon age 49651 BP. The cali-
centimeters thick. The first was the pebble floor
which extends across much of the courtyard
area. This was followed by a floor of finely 10
Similar thin levels of red clay were also noted within the
crushed, white granite. Next came a lens of profiles of two excavation units (P7, P8) on the terrace
above the shrine complex.
199 - Bauer et al.: Destruction of the Yurac Rumi Shrine
brated dates from these samples suggest that the after the two priests returned from seeing the
shrine was established in the late 1300s to mid- Inca in Vilcabamba.13 No matter which chroni-
1400s (Table 1). cler is more accurate, the burning of the shrine
appears to have taken place sometime between
DATING THE DESTRUCTION OF THE SHRINE mid-February and mid-March 1570.14
The destruction of Yurac Rumi can be Our 2008 excavations at Yurac Rumi con-
narrowed down to a relatively short time period. firm that the shrine was destroyed in a massive
The two priests, Ortiz and Garcia, witnessed fire. Each of the buildings that we sampled
and signed a document dictated by Titu Cusi contained thick levels of burnt building materi-
Yupanqui (2005[1570]:139) in the Inca town of als, including large amounts of burnt plaster, as
Vilcabamba on February 6, 1570, after which well as carbonized roof remains. These excava-
they returned to the town of Puquiura near tions also revealed that each of the buildings
Vitcos.11 According to Calancha (1981[1638]: had well maintained floors at the time of the
1825-1827, Book 4, Chapter 4),12 when Ortiz destruction. A few complete ceramic vessels
and Garcia arrived in Puquiura there were many (both for storage and cooking) were found on
people from different towns waiting for them. the floors, and one structure yielded a bronze
Some were bemoaning deaths which had been halberd (yauri) head which appears to have
caused by the demon of Yurac Rumi, while been left in the rafters of one of the buildings
others were trembling in fear from the havoc when it burned (Figure 6). Furthermore, some
and destruction that the demon was causing to organic artifacts in use at the time of the burn-
their families, herds, and fields because they had ing were preserved in carbonized form. For
been baptized. Angry, and perhaps humiliated example, we recovered a carved wood staff that
by the events that had taken place in the city of had been resting on one of the floors. It should
Vilcabamba, Ortiz and Garcia announced be noted, however, that overall our excavations
through the town crier that all the converts found very few artifacts within the structures,
should gather the next day at the church on the suggesting that the priests and attendants had
plaza, and that each person was to bring fire- had time to remove the most valued items
wood, because they were going to burn the before the complex was burned.
shrine. The following day the priests and their
followers marched to Yurac Rumi, encircled it
with firewood, and, after conducting an exor-
cism, they burned the temple and the rock. 13
The Mercedarian friar Martn de Mura lived in Cusco
Mura (1987[1611-1616]:270), provides many at least from 1585 to 1588, more likely even longer, and
of the same details. However, Mura suggests he was stationed as the parish priest of Curahuasi on the
that Yurac Rumi was burned about a month southern edge of the Vilcabamba region in 1595 (Ossio
2008a). It was during these decades that he collected
information on the history of the Incas and the early
conquest of the Andes by the Spaniards. He personally
11
17 February 1570 in the Gregorian calendar. interviewed many of the Spaniards who had lead raids into
the Vilcabamba region, as well as some of the natives who
12
Antonio de la Calancha wrote a history of the had lived there (Adorno and Boserup 2008; Ossio 2008b).
Augustinian order in Peru which was published in 1638. Like Calancha, he had access to the Augustinian records
There is no evidence that he lived in the Cusco region. concerning the death of Ortiz (Aparicio 1989).
However, as an Augustinian himself, Calancha had access
14
to the records of the order, including the inquiries which The testimony of Juana Guerra states that the burning
were conducted on the death of Ortiz (Hemming 1970; took place a cabo de un mes, poco ms o menos [At the
MacCormack 2008). end of a month, more or less] (Aparicio 1989:164).
ANDEAN PAST 10 (2012) - 200
The most intriguing information on the roof support (AA83419 [1], AA83419 [2]), the
massive burning of the shrine comes from grass thatch (AA83422), and carbon associated
Excavation Unit U6, where the ceiling and roof with the reoccupation of the structure (AA
of the structure appear to have fallen relatively 83420, AA83421).16 The results of the radiocar-
quickly (Figures 7, 8). During the course of bon assays are presented in Table 1 and Figure
excavation we exposed samples of burned and 11.
unburned mountain grass (Stipa ichu) used as
thatch, carbonized mountain bamboo (Chusqu- SUMMARY AND DISCUSSION
quea scandens) used as battens, carbonized
rafters made from alder (Alnus acuminata), and The Inca empire is well known for its elabo-
carbonized rope made from local grass (Calama- rately carved rocks, many of which were wor-
grostis heterophylla) used in lashing the battens shiped as shrines. One of the most celebrated
and rafters. These charred ceiling and roofing examples of such a rock shrine is Yurac Rumi in
materials rested directly on the green clay floor the Vilcabamba region. A series of radiocarbon
of the building, and were covered with a 30 tests were run on carbon collected at the shrine
centimeter thick stratum of burnt plaster and during excavations in 2008 to help determine:
other materials (Figure 9). 1) when the shrine was first established, 2)
when it was destroyed, and 3) when it was
A hard-packed earthen floor, enhanced by reoccupied. Not surprisingly, the two carbon
a few slate paving stones, above the burnt wall samples from the reoccupation level provided
plaster stratum, but below a large scale collapse the latest dates. However they were only slightly
of the structure walls, provided evidence of a later than the majority of the ceiling and roofing
brief reoccupation of the shrine after it had been materials which burned when the shrine was
burnt by the Augustinians.15 It appears that this initially destroyed. Although we may never
reoccupation also ended in fire, because the know with certainty, it is possible that parts of
hard-packed earth floor was also covered with a the shrine were immediately re-consecrated by
lens of ash and burnt wood (Figure 10). local inhabitants following the actions of Ortiz
and Garcia, and that those more limited struc-
The recovery of various identifiable organic tures were burnt again approximately two years
ceiling and roofing materials within a building later in 1572 as Spanish forces swept through
with a known destruction date provided a the region in their successful campaign to cap-
unique investigative opportunity. We wanted to ture Tupac Amaru and to end the independent
examine how the dates of various plant species Inca state established in the Vilcabamba region.
incorporated into the different parts of the roof
correlated with each other, as well as with the The burnt ceiling and roofing materials of
known destruction date of the shrine. As a the shrine also provided intriguing results. The
result, AMS measurements were run on a rope sampled batten, roof support, and thatch all
(AA83417 [1], AA83417 [2]), a batten (AA
83418 [1], AA83418 [2]), the outer layers of a
16
All samples were processed in The National Science
Foundation-University of Arizona Accelerator Mass
15
The reoccupation of Yurac Rumi may have been limited Spectrometry facility. Calibrated dates were generated by
to only certain areas of the shrine complex. Sometime Calib 5.1 beta, courtesy of the University of Washington
after this brief reoccupation, many of the buildings of the (Stuiver and Reimer 1993). Dates were calibrated using
shrine complex appear to have been systematically software employing southern hemisphere data (SHCal04
destroyed, with major architectural features, such as the calibration curve) (McCormac et al. 2004).
stone roof pegs, pulled from the walls.
201 - Bauer et al.: Destruction of the Yurac Rumi Shrine
yielded very similar calibrated dates which Two separate test excavations found evi-
clustered in the late-1400s to the mid-1500s. dence of crushed granite floors in the Yurac
These age estimations make sense if the shrine Rumi shrine. The incorporation of the crushed
was established, as suggested by the dates from granite into the floors of the complex must have
the granite floors, during the late 1300s to mid- made them architecturally stunning, and rein-
1400s. However, the rope which was used to tie forced the ritual nature of the area surrounding
the thatch together provided surprisingly early the sacred central stone. Use of the wastage
dates. Our sample yielded calibrated probability from the carving of the Yurac Rumi stone as a
dates in the late-1200s, suggesting that the rope; paving material for the plaza space is consistent
was more than 200 years older than the rest of with Inca aesthetics. The pulverized debitage
the roof. Indeed, the results suggest that the from a huaca like this would not have been
rope may have even predated the Inca occupa- categorized as mere spoilage, but something
tion of the Vilcabamba region. Contamination more akin to a religious relic. Its use as a floor-
agents could have been introduced as a result of ing material would have bound the site together
fiber preparation before the weaving of the rope, as a single, sacred complex.
however, the Inca are not known to have used
such agents.17 While it is possible that the Incas Carbon samples were collected from two of
used older materials in the construction of the the granite floors.22 Although the radiocarbon
shrine, the dates of the rope seem excessively ages of these two samples are separated by some
early, and we currently cannot fully explain the one hundred years (60134 BP, 49651 BP),
results. their calibrated 1-sigmas do have considerable
overlap in the earliest decades of the cal AD
The precision of radiocarbon assays can be 1400s. We currently believe that these samples
increased by using pooled means of different but date the establishment of Yurac Rumi as a
related samples. In this study, the two samples shrine, and document when the Inca first ex-
from the reoccupation18 can be pooled together, panded into the Vilcabamba region. The fact
as can the samples from the batten,19 the roof that these levels were later buried, and that
support,20 and the rope.21 The results of the additional floors were established above them,
pooling, presented in Figure 12, helps to illus- indicates that a series of modifications occurred
trate how close the reoccupation dates are to at the shrine after its initial dedication. The
the roof dates, and how much earlier the rope dates are consistent with suggestions that the
dates are. region was incorporated relatively early into the
Inca empire, during its initial expansion period,
most likely during the reign of Pachacuti Inca
Yupanqui.
17
In other cases, conservation agents might also make a
rope test older than it really is, but in this case the sample
was taken in the field and sent directly to the laboratory
for analysis.
18
Samples AA83420, AA83421.
19
Samples AA83418 (1), AA83418 (2).
20
Samples AA83419 (2), AA83419 (1).
21 22
Samples AA83417 (1), AA83417 (2). Samples AA83415, AA83416.
ANDEAN PAST 10 (2012) - 202
AA83415 Granite floor Unidentified carbon -25.1 0.9280 0.0039 601 34 One-sigma ranges:
[cal AD 1326: cal AD 1341] 0.250409
[cal AD 1390: cal AD 1419] 0.749591
Two-sigma ranges:
[cal AD 1316: cal AD 1355] 0.326466
[cal AD 1382: cal AD 1435] 0.673534
Median probability: cal AD 1397
AA83416 Granite floor Unidentified carbon -27.5 0.9401 0.0060 496 51 One-sigma ranges:
[cal AD 1411: cal AD 1464] 0.955148
[cal AD 1470: cal AD 1476] 0.044852
Two-sigma ranges:
[cal AD 1394: cal AD 1509] 0.922702
[cal AD 1579: cal AD 1620] 0.077298
Median probability: cal AD 1446
AA83417 (1) Rope Calamagrostis -23.7 0.9075 0.0059 780 52 One-sigma range:
heterophylla [cal AD 1225: cal AD 1298] 1.0
Quechua: Hiro ichu Two-sigma ranges:
Reed grass [cal AD 1190: cal AD 1191] 0.001013
[cal AD 1202: cal AD 1322] 0.859046
[cal AD 1347: cal AD 1387] 0.139941
Median probability: cal AD 1271
AA83417 (2) Rope Calamagrostis -24.2 0.9089 0.0056 767 49 One-sigma ranges:
heterophylla [cal AD 1230: cal AD 1250] 0.205490
Quechua: Hiro ichu [cal AD 1261: cal AD 1304] 0.662807
Reed grass [cal AD 1363: cal AD 1377] 0.131703
Two-sigma ranges:
[cal AD 1215: cal AD 1323] 0.803562
[cal AD 1346: cal AD 1388] 0.196438
Median probability: cal AD 1283
AA83418 (1) Batten Chusququea scanden -25.9 09481 0.0061 428 52 One-sigma ranges:
Quechua: Kurkur [cal AD 1446: cal AD 1509] 0.666726
Mountain bamboo [cal AD 1580: cal AD 1620] 0.333274
Two-sigma range:
[cal AD 1438: cal AD 1628] 1.0
Median probability: cal AD 1508
AA83418 (2) Batten Chusququea scanden -23.7 0.9466 0.0058 441 49 One-sigma range:
Quechua: Kurkur [cal AD 1442: cal AD 1504] 0.763347
Mountain bamboo Two-sigma ranges:
[cal AD 1428: cal AD 1524] 0.636179
[cal AD 1529: cal AD 1531] 0.005259
[cal AD 1535: cal AD 1626] 0.358562
Median probability: cal AD 1493
AA83419 (1) Roof support Alnus acuminata -22.8 0.9484 0.0058 426 49 One-sigma ranges:
Spanish: Aliso [cal AD 1447: cal AD 1508] 0.670719
Alder [cal AD 1583: cal AD 1619] 0.329281
Two-sigma range:
[cal AD 1441: cal AD 1626] 1.0
Median probability: cal AD 1508
Continued:
AA83419 (2) Roof support Alnus acuminata -27.3 0.9524 0.0061 392 51 One-sigma ranges:
Spanish: Aliso [cal AD 1461: cal AD 1512] 0.428560
Alder [cal AD 1548: cal AD 1563] 0.113319
[cal AD 1569: cal AD 1622] 0.458120
Two-sigma range:
[cal AD 1453: cal AD 1635] 1.0
Median probability: cal AD 1547
AA83420 Reoccupation Unidentified carbon -25.6 0.9575 0.0062 349 52 One-sigma ranges:
[cal AD 1502: cal AD 1594] 0.792765
[cal AD 1613: cal AD 1638] 0.207235
Two-sigma range:
[cal AD 1457: cal AD 1658] 1.0
Median probability: cal AD 1562
AA83421 Reoccupation Unidentified carbon -25.6 0.9557 0.0062 364 52 One-sigma ranges:
[cal AD 1499: cal AD 1598] 0.835896
[cal AD 1610: cal AD 1630] 0.164104
Two-sigma range:
[cal AD 1458: cal AD 1646] 1.0
Median probability: cal AD 1557
AA83422 Grass thatch Stipa ichu -23.6 0.9470 0.0061 437 52 One-sigma ranges:
Quechua: Ichu [cal AD 1442: cal AD 1507] 0.716657
Mountain grass [cal AD 1586: cal AD 1618] 0.283343
Two-sigma range:
[cal AD 1432: cal AD 1626] 1.0
Median probability: cal AD 1498
Figure 12: Radiocarbon dates from Yurac Rumi using pooled means.
INCA STORAGE AND ACCOUNTING FACILITIES AT PACHACAMAC
Peter Eeckhout
Universit Libre de Bruxelles, Brussels, Belgium
. . . Our best data on Inca storage come from the In this article I present information on the
Peruvian Central Highlands where, as it seems finds and contexts of quipus and collcas (Quech-
clear, storage was unusually important. However, ua for storehouses, also spelled qollqas) at the
we need more intensive work in other areas in site of Pachacamac, on Perus central coast. First
order for a more complete and comparative I describe a series of collcas, focusing on the
perspective to emerge (Morris 1992a: xiii). differences between the coast and the highlands
in the design of storehouses, their number, and
INTRODUCTION capacity. Work in the northern portion of the
site has led to the discovery of a probable Inca
The work of Craig Morris on the Incas storage system, a building designated Structure
economy and storage system represents an E8, described in this article. Several quipus have
important stage in our understanding of their been found in this building, as they have been in
great Andean empire (Morris 1967, 1986, other contexts at the site. My interpretation of
1992b; Morris and Santillana 1978; Morris and these finds in light of the general framework of
Thompson 1985; among others). Morriss Inca imperial political and economic organiza-
thought and approaches do not apply only to the tion suggests that Pachacamac was controlled
sites and regions where he, himself, worked, but indirectly by the Incas, who used the site not as
also to other sites, as he stressed. Specifically, his a state administrative center, but as a local one.
insights are relevant to the Andean coast where They concentrated their efforts on developing
the evidence for storage is very different from facilities related to the god for whom the site is
that found in the highlands. There are a few named, and for large scale pilgrimage.
published studies of Inca storage at sites on the
Pacific coast, including Farfn in the Jequete- Pachacamac, at the mouth of the Lurn
peque Valley (Mackey 2006), Chiquitoy Viejo River, was one of the most important sites of the
in the Chicama Valley (Conrad 1977), Incawasi Inca empire (DAltroy 2002; Eeckhout 1998,
in the Caete Valley (Hyslop 1985), La Cen- 1999:410-417, 469-474, 2004a; Hyslop
tinela in the Chincha Valley (Morris and San- 1990:255-61; Moseley 1992:185; Rowe 1946:
tillana 2007), and Tambo Colorado in the Pisco 191, 1963; Uhle 1903). It is, thus, interesting to
Valley (Hyslop 1984:108-11), but there is see how storage was managed there, but, some-
nothing comparable to the work on the subject what surprisingly, up to now there has been no
accomplished at highland sites like Hunuco work focused specifically on that subject. Never-
Pampa (Morris and Thompson 1985), Hatun theless, one can see Inca traces in, among other
Xauxa (DAltroy 1992; DAltroy and Hastorf things, the presence of storehouse complexes, as
1984), and Pumpu (Matos 1994), among others well as of artifacts related to accounting and
(LeVine, editor 1992). recording (quipus or khipus and yupanas). In this
article I put down on paper data and thoughts
relevant to the role of storage and accounting tion of basic products extracted from the local
facilities in the Inca imperial economic organiza- economy allowed it to sustain those who worked
tion as seen in the light of the general achieve- for the state (Earle 1992:334). Naturally, this
ments and specific finds made by the Ychsma system required adequate logistics, including
Project.1 storage facilities on a large scale, as well as
widespread economic control (ibid.).
Evidence of quipus has been found during
several field seasons at Pachacamac, and we can Craig Morris, after studying the collcas of
now identify a series of collcas whose number, Hunuco Pampa, concluded that their form
design, and location allow us to understand the (circular or rectilinear in plan) may have had a
ways the Incas controlled the site. The data, relationship to their contents.2 Above all, he
although preliminary, suggest that in the eyes of demonstrated the high degree of elaboration of
the Inca, Pachacamac was a major site both these storehouses, which were equipped with
symbolically and in strategic terms, but one that ventilation systems and other features to ensure
was not as important, perhaps, in economic the best possible preservation of the perishable
terms. products they contained (Morris and Thompson
1985:97-107). He also emphasized that a crucial
STORAGE IN THE INCA IMPERIAL ECONOMIC factor for effective storage was the location of
ORGANIZATION collcas in high places with relatively low temper-
atures and constant winds (Morris 1992b: 254-
The collcas served to warehouse various 256). This later consideration explains why the
kinds of goods (c.f. DAltroy 2002:284; Huay- sites with the largest number of collcas are found
cochea 1994; LeVine, editor 1992). These goods above 3300 m.a.s.l.
supported various goals, including maintenance
of imperial armies on the march, maintenance of The Inca state incentivized the movement of
workers serving in the mita, redistribution to products to concentrate them at specific sites.
local residents during banquets and fiestas, and For example, at the site of Cotapachi, in the
provision for difficult times such as after bad Cochabamba Valleys of Bolivia, collcas occupied
harvests. According to Timothy Earle (1992: approximately 140 hectares. They stood in rows
333), the possible, mutually non-exclusive, and added up to about three thousand units for
functions of Inca storage were to support subsis- the stockpiling of maize produced in the area
tence maintenance, to serve as distribution (Pereira 2011). Nevertheless, this site is not
facilities, and to be instruments of institutional associated with other facilities except for domes-
finance. DAltroy and Earle (1985) argue that tic ones, and is integrated into the Qapaq an,
this latter aspect is the most important one for the Inca highway system, via a secondary road
understanding imperial organization, which (Snead 1992:92). Sometimes collca complexes
needed resources to maintain itself and develop, are dozens of kilometers away from the most
but which did not have an advanced market productive agricultural zones, as is the case with
system. The mobilization by the state of a por- Hunuco Pampa (Earle 1992:332).
1
These aspects of the Inca economy have
El Proyecto Ychsma (Universit Libre de Bruxelles,
Belgium and Instituto Nacional de Cultura del Per, now
been understood since the beginning of the
the Ministerio de Cultura) began in 1999 in order to
clarify the function, development, and influence of the
2
site of Pachacamac during the Late Intermediate Period This suggestion has not been confirmed at Hatun Jauja
and the Late Horizon. by DAltroy (2002:284).
215 - Eeckhout: Inca Storage at Pachacamac
Spanish conquest. In the case of Pachacamac, found within its Chimu ciudadelas (large royal
the earliest accounts mention its economic and palaces) and upper elite compounds (Kolata
political arrangements. In January 1533, Her- 1990), and not outside the city, as with numer-
nando Pizarro arrived in the area and revealed ous highland examples. The coastal storehouses
that In all the plains and beyond, tribute is not differ in their access system, which is not via low
paid to Cusco, but to the mosque3 (Pizarro doors, as in the mountains, but via their roofs,
1872 [1533]:123). This quotation reflects the using stairs or ladders.4 Coastal storehouse
decentralized organization of the Inca empire as dimensions vary greatly, and, in general, the
described above. As various authors have em- individual coastal storage systems are larger than
phasized, Cusco, in contrast to other ancient examples from the sierra. This is true of Pacha-
imperial capitals such as Rome, did not have a camac, where we have several examples of
large population (Prssinen 2003:131-139). storehouses pertaining to the Late Intermediate
Therefore, it was not necessary to bring vast Period and the Late Horizon (Table 1).
quantities of goods from the various parts of the
empire, but only that which was sufficient for a Building Period N o of Estimated Estimated storage
m2 capacity m 3*
few thousand Cusqueos, who also obtained collcas
Totals
FIELD DATA ON STORAGE FACILITIES AT
*Volumes are structure volumes to roofline, based on an estimated
PACHACAMAC average height of two meters as determined by excavation. Structure
volumes do not take into account accommodations for packaging.
The studies on storage by Morris, and others,
Table 1. Storage facilities at Pachacamac
focus mostly on the highland collcas which are
formally different from the storehouses found at
Storehouses dating from the Late Intermedi-
sites on the coast, and in the lower portions of
ate Period are systematically associated with
the coastal valleys. It is sufficient to mention the
larger buildings, such as the pyramids with ramp
example of Chan Chan, whose storehouses are
(Figures 1, 2). In formal terms, all are orthogo-
nal, generally square in plan, sunken, and closely
3
In the early sixteenth century many Spanish writers
4
referred to important Amerindian religious structures as At Chan Chan access to storehouses is through a
mosques (mezquitas), conflating Islam with other non- central door at floor level (c.f. Moseley and Mackey
Christian religions. 1974).
ANDEAN PAST 10 (2012) - 216
associated with private sections of their sites. Table 1 reveals that, during the Late Intermedi-
That is, access was controlled by the chief ate Period, each pyramid had its own stores,
inhabitants of the pyramids, as is demonstrated which varied a great deal in terms of number
by the buildings internal systems of movement and capacity. These depended on the form of
(Eeckhout 2003). This differs a great deal from political and economic control exerted by the
the Inca pattern, in which the storehouses stood Ychsma chiefdom. According to the model I
beyond their sites centers, usually on a height propose on the basis of field-work data, the
with fresh air. Highland collcas do not exhibit pyramids are the palaces of curacas, the local
evidence of access as restricted as that of coastal leaders, who divided surplus production and
ones. The capacity of the Pachacamac storage took part of it for their own benefit, storing it in
system is relatively restricted. At any rate, it does their pyramids. Not all the pyramids functioned
not compare with the hundreds of collcas at at the same time. Instead, each one was occu-
Hunuco Pampa, for example, even though both pied for one or two generations between the
sites are comparable in size and importance fourteenth and fifteenth centuries C.E. by an
(compare Snead 1992:68, table 3-2).5 elite hierarchy reflected in the sizes of the build-
ings, and defined through excavation and radio-
These two specific characteristics, the lower carbon dates as ongoing research from the
number of storehouses, and the restricted access, Ychsma Project suggests.
suggest that storage at Pachacamac represented
a select portion of goods intended for consump- What did the storehouses contain? It must
tion, possibly reserved for a specific segment of be emphasized that for Pachacamac Pyramid 3,
the population, who took precautions in looking the most complete example studied by the Ychs-
after it. This model fits a general pre-Inca pat- ma Project, excavations confirmed that the
tern, marked by a concentration of staple prod- premises had been completely emptied (Eeck-
ucts in the domestic unit, while storehouses hout 1999:128-192, 2004b). Nevertheless, in
proper were reserved for luxury objects, associ- one of the buried areas in Plaza 2 we found
ated with elite structures (Earle 1992:340). This corn-cobs offered as conopas,6 which may suggest
pattern at Pachacamac also shows that the the way the area was used (Eeckhout and Farfn
management of storehouses was not in the 2001). Rgulo Franco Jordn has had more
hands of a force extending over the entire site, success, encountering corn-cobs, gourds, and
but, instead, was divided among different groups, chilli peppers in the main storerooms of Pyramid
each based at one of the several pyramids with 2. It appears that these were roofed to protect
ramp (Eeckhout 2004b). The data presented in them from wind and insects (Franco 1993:17-
18). The stratigraphy of the contents shows that
5
the storehouses were not always used to keep the
In the highlands, many fields produce only one crop a same type of product, and it is probable that they
year (c.f. Mitchell 1991:71-73) and many must be left
fallow for long periods. This makes storage a greater
served totally different ends. For example, there
necessity in the highlands than it is on the coast. On the was a layer of camelid excrement in a buried
coast, in Inca and early colonial times, most fields could peripheral area (ibid: 72) and there were scratch
be planted twice a year with maize and other marks made by dog claws on the walls of one of
cultivars(Cieza 1962 [1553]: Chapter 66, p. 193), as they the structures near the platform (Paredes
are at present. On the coast it is easier than in the
highlands to fertilize using fish and/or guano, and fish and
shellfish are generally available all year for human
6
consumption. Another crop and another supply of These are special natural items chosen for their unusual
protein are always available soon (Monica Barnes, size or form. They were considered to be sacred and used
personal communication, August 2011). as propitiatory offerings.
217 - Eeckhout: Inca Storage at Pachacamac
1988:50), allowing us to think that these struc- their authority, and for the maintenance of their
tures served to confine animals. Paredes (ibid.: personnel.
54) concludes, nevertheless, that Pyramid 2
appears to have been dedicated mainly to the For the Late Horizon one clearly observes
storing of chilli, present in several deposits that there was a drastic change in the distribu-
associated with the building. tion of storehouses at Pachacamac. During this
time none of the pyramids with ramp was being
It is certain that Pachacamac has been used as the residences of the local elite, with two
intensively robbed since the beginning of the exceptions, P10 (Tauri Chumpi) and P13. I will
Spanish conquest and it is most probable that return to those below. The other pyramids
only minimal traces remain of the produce its served as campsites for pilgrims, and they were
storehouses may once have contained. This is quickly reused as burial chambers (Eeckhout
without considering the valuable goods (textiles, 2010a). When the site was conquered by the
various artifacts, etc.) that were probably also Incas, Topa Inka Yupanqui undertook many
kept in them. Furthermore, given that almost all transformations there (Patterson 1983). These
the pyramids were abandoned over the course of changes were motivated, among other reasons,
the Late Intermediate Period, it is logical that by the desire to make the site a pilgrimage center
one no longer encounters anything in the store- at the imperial level (Eeckhout 2008b). The
houses. In reality we are confronting a situation Plaza of the Pilgrims (Plaza de los Peregrinos),
like that of the palaces of Chan Chan, whose the Temple of the Sun (Templo del Sol), and
storehouses were also all found to be empty (Day the House of the Chosen Women (Casa de las
1982:60). The Lurn Valley has also suffered Mamaconas or Acllawasi) were built in areas
from plundering, but on a smaller scale. previously occupied by local constructions. As
for storehouses, a series of structures exists on
Certain local palaces functioned under Inca the platforms of the Temple of the Sun that,
authority, therefore forming the second hierar- from their design, could have served as collcas,
chical level in the chiefdom, and, thus, probably but their present state of conservation does not
retained a good part of their privileges (Eeck- permit much to be said about this (Franco
hout 2008a). Likewise, excavation of a pyramid 1996). There are, as well, some probable store-
with ramp at the site of Pampa de las Flores in houses associated with the Acllawasi (Tello
the lower Lurn Valley confirms indisputably the 2009), but, according to our observations, the
role of storage attributed to buried structures, Inca storage systems observable today are con-
because maize was encountered in situ (Eeck- centrated in two places: the vicinity of the Plaza
hout 1999:289-90). At the nearby site of Tijer- of the Pilgrims, especially at structures A1-A2,
ales B, Bonavia (1965:91-92) found raw cotton and in the northern sector of the site, where we
in a roofed storehouse of the South Pyramid. identified a very unusual structure, designated
This evidence suggests that perishable goods E8.
(such as food products) were stored, but, per-
haps, so were other consumable goods (such as The A1 compound is north of the Plaza of
wool, cotton, and ordinary ceramics), prestige the Pilgrims, and consists of a series of plazas,
goods (fine clothing, imported products, deco- patios, rooms, and sunken structures standing
rated ceramics, etc.) and/or products of ritual on several levels (Figure 3). It is an important
value (coca, chicha, etc.). In summary, every- building complex probably related to the con-
thing was stored that was necessary for the daily duct of ceremonial activities and to offerings
life of curacas, for their office, for the exercise of brought by pilgrims. This part of Pachacamac
ANDEAN PAST 10 (2012) - 218
has not been excavated, but surface clearance an opportunity to excavate the rows of rooms to
revealed architectural evidence suggesting check our hypothesis, but we have been able to
storage functions, such as rows of sunken rooms excavate the front room at the edge of the patio,
with means of access control. The absence of although not as carefully as we would have
excavation data from this compound makes it liked. However, during the course of cleaning
difficult to assign it a chronological position. the room in front of the rows of chambers, we
Nevertheless, there is evidence relating it to the encountered a series of finds that I will describe
Late Horizon, for example, an aryballos found in below.
the corner of one of the compounds rooms
(Figure 4). In addition, both the A1 compound The front room is small and rectangular with
and its neighbor A2 are clearly associated with a single entrance opening upon the patio (Figure
the Plaza of the Pilgrims, which excavation has 6). One notices upon entering a division of the
demonstrated to have been constructed in Inca internal space into two parts, perhaps a vestibule
times (Eeckhout and Farfn 2003; Shimada et al. with a small adjacent portion fitted with a
2004; Uhle 1903). lateral platform or bench. Although it was
already plundered, cleaning the room to Floor 1
Structure E8 is in the E section (barrio) of revealed some very instructive finds. The first is
Pachacamac, that is, at the extreme north of the a small wooden cup finely decorated with an
site (Figure 5). It is an isolated structure, but one incised maize plant motif, a very frequent design
that is associated with circulation routes of in Inca iconography (Figure 7). This design
which only traces remain, including a wall to the suggests that the object probably served for
east topped with a path, and a trace of a street drinking chicha, a practice related to fiestas and
or passage to the west. Although partially de- ceremonies for the majority of people, and to
stroyed by the old Panamerican Highway and daily consumption for certain privileged classes
the present concrete wall that delimits the such as curacas and high state functionaries
monumental portion of the site, the principal (Bray 2003).
components of E8 are preserved, and its general
form can be reconstructed, as is illustrated by Second, we encountered lying on the floor
the isometric plan we made on the basis of field and broken into several dozen pieces, a ceramic
data. It has a generally rectilinear form and object that we reconstructed. The process of
appears to have been completely surrounded by restoration was somewhat difficult because the
a two meter wide wall, with access in the south- object weighs about ten kilograms and has a
west. It consists basically of two parts: a large most unusual form. It is basically a cube with
patio, eighty by sixty meters, associated with a several appendages and a round central hole
series of buried chambers with square plans, laid (Figure 8). Two of the appendages show four
out in two parallel rows of ten units each along small holes, perhaps designed to suspend or
the south wall of the patio. A room to the front attach thin cords while quipu knots were being
and center of the units can also be observed. We tied (see below). Its dark orange paste is very
will return to this later. The two rows of buried homogeneous. The piece is painted white and
structures are arranged in steps. That is, the row does not show marks of burning, nor traces of
nearer the patio is lower than the other. Accord- any contents. We do not know its purpose.
ing to the observable evidence, access to these Perhaps it served as some sort of model. Another
rooms was achieved via paths atop walls. The possibility is that it was used to keep quipus
form and design of the rooms is very similar to (compare Flores et al. 2007:243, figure 34),
coastal type storehouses. We have not yet had something suggested by the find I describe next.
219 - Eeckhout: Inca Storage at Pachacamac
FIELD DATA ON ACCOUNTING FACILITIES The doubled and wrapped quipus resemble
AT PACHACAMAC the manner in which quipucamayoqkuna (keep-
ers of quipu records) retained their quipus
The third remarkable find from the prelimi- (Ascher and Ascher 1981:33). The quipus
nary excavations of the front room of Structure previously found at Pachacamac by Alberto
E8 is a collection of quipus found lying on the Bueno in Structure B7 are famous, and he gave
floor, spread over the platform, and on the the building the name House of the Quipus
broken floor between the room and the en- (Casa de los Quipus; Bueno 1990). Neverthe-
trance. Unfortunately, the several masses of less, the context seems different from that which
quipu have decomposed, possibly because of rain I have just described, because the quipus discov-
or because of some other liquid thrown over ered by Bueno were hidden in a pit lined by
them (Figure 9). Together with Kusi Colonna- adobes and covered with layers of earth, sand,
Preti, the conservator of the Ychsma Project, I and late archaeological debris. Bueno (ibid.:
am studying the possibility of restoring them, or, 100) explains that the 34 quipus, enclosed in a
at least, separating them, but this would be a packet formed by the hide of a young deer were
very delicate operation. Nevertheless, we hope completely mixed up, tangled, and without
to recover significant objective data (number, order. It would appear that everything was very
position, and type of knots) so that these quipus rapid, the placement within the hide, securing
can be used for scientific purposes. On the basis the packet, depositing it to hide it, and so on.7
of field observations, we can already say that we We can assume that the owners of the quipus in
found at least nine quipus, apparently all of fine Structure B7 did not want their precious objects
cotton cords, the majority a natural white color. to suffer the same fate as those of E8.
We cannot discount the possibility that there
are more quipus, because some bundles of cords Another quipu find was made in an intrusive
might incorporate more examples. Some of the collective tomb from the Late Horizon/Transi-
quipus, at least, have colored knots. tion Period at Pachacamac Pyramid 3 (Eeckhout
and Farfn 2001:40). This artifact was found in
Quipu H24A, found on the surface of the fragments, all made of white cotton (Figure 10).
platform, is one of the most deteriorated. Never- Unfortunately, the repeated plundering of its
theless it has headers and cords of different context does not allow the suggestion of more
colors, beige, violet, and possibly brown (the than what is immediately obvious; that is, that
latter perhaps due to the liquid that caused its the quipu was part of a set of funerary objects.
decomposition). Quipu H24I was found rolled
up like a hank of yarn, has white cords, and is Finally, there is a very well preserved quipu
associated with a cane and remains of a plain- from Pachacamac that is now in the collections
weave textile with some selvage visible. Quipu of the Vatican Museum in Rome (vidi 2002).
H24J is another quipu in bad condition, also Franco (1993:82; 1998) also reported the find of
rolled like a hank of yarn. It is brown, blue, and a quipu in Sector 4 of Pyramid 2, but it is not
violet. Quipu H24K is another packet of de- described or illustrated.
composing, dark brown quipus with knots. This
was found in the central part of the structure
between the west wall and the bench. These 7
The original quotation reads: [Los quipus] estaban
artifacts are, themselves, associated with the completamente revueltos, enredados y sin ordn.
bench and Floor 1. Pareceria que todo fue muy rapido: la colocacin dentro
del pellejo, su asegurado, la deposicin para ocultarlos,
etc.
ANDEAN PAST 10 (2012) - 220
The association of collcas and quipus appears camacs Pyramid 11 in 2008 (Figure 11). This
to be generally quite coherent, because the stone artifact is a parallelepiped, 37.5 centime-
reports of quipus being used as accounting ters long. It was found in a layer of fill with late
registers are numerous, with that of Guaman rubbish. Five faces have been made in a crude
Poma being the most widely distributed (Gua- manner, while the last has a series of twenty
man Poma 1989 [c. 1615]:309, 336, 338). All round depressions carved in two parallel lines of
specialists agree that the Inca registry system was ten units each. Although out of context, the
decimal (Urton 1997, 2003). This can be seen in objects design makes one think of a yupana,
quipu structure, but can also be noted in other with its depressions forming the compartments.
media. It is worth emphasizing that quipus Also, the fact that the depression number
served to register, while other implements, called twenty attracts attention because, for one thing,
yupanas, aided simple calculations (addition, that is the average number in old yupanas, and,
subtraction, and multiplication). These instru- for another, twenty corresponds to the design of
ments are based on a principle similar to that of Structure E8 described above. I do not mean to
the abacus, which consists of a certain number make a direct connection between this artifact
of beads set on rods, each one of which indicates and Structure E8, but, rather, to underline the
a number. In the case of the ancient Peruvian fact that they are both based on a similar struc-
abacus, small stones or seeds were used instead tural logic associated with the decimal system
of beads, and they were not placed on rods, but and with duality or bipartition. In this sense, I
were put into compartments carved into tablets share the thoughts of Gary Urton when he says
of wood or stone, with each compartment hav- that Dual organization and the decimal system
ing a particular assigned value. Some yupanas of numeration, which . . . were central principles
were very elaborate and sometimes have been in Inkaic (Quechua) administrative organiza-
compared or confused with architectural mod- tion, are seen to have been not only compatible
els. Others are very simple (Radicati 1990). As but complementary principles of organization
Henry Wassn emphasized (1990:218), We do (Urton 1997:216).
not need to think that a Peruvian abacus had
inevitably to be made of a more or less solid INTERPRETATION OF FIELD DATA FROM
material like stone, wood, etc. It is quite proba- PACHACAMAC
ble that sometimes they consisted of nothing
more than the form of an abacus drawn rapidly So far as Structure E8 is concerned, the
in the sand, or on a piece of cloth, etc., and the presence of a group of quipus, of an artifact
final result was knotted into a quipu.8 possibly used to keep them, of a decorated
wooden kero, and the general context, lead to
This perhaps explains why, in spite of their the conclusion that this room was occupied by a
frequent use, yupanas are relatively rare in the functionary who dedicated himself to the admin-
archaeological record. Because of this, it is istration of the goods that arrived in the build-
worth emphasizing a find made outside Pacha- ing, strategically located at the edge of the
monumental zone. The goods left in the patio,
8
perhaps by llama caravans (which would explain
The original quotation reads: No hay que imaginar que the width of the access point), were recorded,
un baco peruano tuviese que ser invariablement de un
material mas o menos slido como piedra, madera, etc. Es
and later kept in the nearby storage chambers. It
muy probable que a veces consistiese nada mas que de la would be very interesting to check this hypothe-
figura de un baco delineada rapidamente sobre la arena, ses with systematic excavations in the various
o sobre un pedazo de tela, etc., y el resultado final parts of the structure. To make a comparison I
anudado en el khipu.
221 - Eeckhout: Inca Storage at Pachacamac
should state that Valdez (1996:42) has formu- Site Compound Number of Reference
lated the same hypothesis on the basis of a quipu size collcas
find near the storehouses at Tambo Viejo in the Pachacamac-E8 80 x 60 m 20 See this paper,
passim
Acar Valley. The architectural layout of Struc-
ture E8 has many similarities with the Inca Tambo Viejo 60 x 40 m 7 Valdez 1996:40
studied by Francis A. Riddell and Dorothy Parara Depsito 40 x 20 m Undetermined Riddell 2007:184
Menzel in 1954 (Riddel 2007) and, later, by Chala Viejo Not Undetermined Riddell 2007:184
Hermann Trimborn (1988:129). Valdez (1996: Antiguo mentioned
diate Period, because it was concentrated during This brings me naturally to the portion of
a shorter time. Secondly, the distribution of the the collcas dedicated to state administration,
collcas changed completely and it is very reveal- those that are located, with good reason, in the
ing in terms of the Incas management of the site same palace of the tucrikuk (for his personal use,
and its region. The complexes and sectors in for the use of his personal, for the needs of his
which one finds storehouses are, on one hand, office, etc.), and in Structure E8, in the same
closely related to ceremonial and religious neighborhood, probably near an entrance that
aspects of the site (Temple of the Sun, House of no longer exists. There, that portion of the
the Chosen Women, Structure A1-A2, and products of the fields and other local resources
Pyramid 13), and with state management on the requisitioned by Incas were received. It is inter-
other hand (Structure E8 and Pyramid 10/Tauri esting to note that the storehouses of Structure
Chumpi). E8 were associated with a large patio, which,
perhaps, served to receive people during the
It is logical to suppose that the storehouses banquets and redistribution ceremonies that
associated with the Temple of the Sun are were part of the exercise of Inca power (Pease
dedicated to the solar cult, and that the Aclla- 1992:15-24, 1999). Perhaps this large patio also
wasi, as is well known, had its own resources for served to house the camelid caravans that
the acllas and the lunar cult (Tello 2009). Struc- brought goods to the site. No doubt these hy-
tures A1-A2 and P13 stand on the sides of the potheses can be tested with excavation.
Plaza of the Pilgrims constructed by the Incas to
receive the large processions organized through- Apart from storehouses and the presence of
out the empire to celebrate the Pachacamac official functionaries, access to the Qapaq an
oracle (Eeckhout 2008b) and heal the sick is another recurrent characteristic of the logis-
(Eeckhout 2010b). It is known that the pilgrims tics of Inca state installations. We have an eye-
brought many offerings, and also that a substan- witness account from the earliest days of the
tial infrastructure was necessary to house them. Spanish conquest concerning Pachacamac.
The number of collcas in A1-A2 corresponds Hernando Pizarro left Cajamarca towards the
perfectly to these necessities. Pyramid 13, strate- end of 1532, arriving at the site at the end of
gically located near the entrance of the Plaza, January 1533. He remained for a month, and
and with various access points to it, perhaps from there he proceeded to Jauja and Cusco
housed an important priest. Relevant to this (Pizarro 1872 [1533]). He was accompanied by
hypothesis, early Spanish accounts mention two important Inca imperial officials, who took him
supreme officials at Pachacamac during the Inca along the roads and to the tambos that were
regime: the high priest of the temple and the found all along the highway (Guilln 1974:157-
governor or tucrikuk (Xerez 1965 [1534]:87). 160). This has been amply documented by
The latter has been identified as an individual archaeology, and almost the entire length of the
named Tauri Chumpi, and for that reason the Qapaq an from Pachacamac to Jauja has been
structure excavated by Alberto Bueno in the identified. It is interesting to emphasize in this
northern portion of Pachacamac has been respect that parts of the Inca road, or at least its
named The Palace of Tauri Chumpi (Palacio route, have been reused up to the present day
de Tauri Chumpi; Bueno 1974/75). All authors (Hyslop 1984, 1990; La Torre and Caja 2005).
agree that this building can be assigned to the
Late Horizon, being, in fact, the probable resi-
dence of the secular head of the site.
223 - Eeckhout: Inca Storage at Pachacamac
THE ROLE OF COASTAL STORAGE FACILITIES adapted them to the local context. On one hand
IN THE INCA POLITICAL ECONOMY they adopted local construction techniques and
traditional architectural design at Pachacamac9
The new finds presented in this article help but, on the other hand, built structures specifi-
us to understand the specifics of Inca strategy cally for administration, that is, ones that were
and logistics in the coastal area. Above all one isolated and mostly separate from the elite
must emphasize the relative increase of storage residences. In this sense the Incas were very
capacity at Pachacamac under Inca rule. This different from the Ychsmas. The Inca arrange-
can only be explained by three possible factors. ments also demonstrate a global control over the
The first is that the Incas may have concen- site, and over regional resources, probably with
trated at Pachacamac tribute coming from a a focus on its ceremonial aspects, as the quota-
larger area and more distant territory than that tion from Pizarro cited above (tribute is not
controlled by the Ychsma curacas in the Late paid to Cusco, but to the mosque) suggests.
Intermediate Period. However, I do not believe This situation is not unique. Other authors have
that this was the case, as I explain below. commented, on the basis of ethnohistorical
sources, that storage to support the state cult
The second factor would be demographic can be considered to be separate from storage
growth of the area, for example because of mit- supporting other state functions (Snead 1992:
maq (conscripted workers) coming from other 72).
parts of the empire and paying tribute at Pacha-
camac. It is certain that there was a colony This peculiarity perhaps explains why, in
called Quilcay established next to the site during spite of its expansion, Inca storage capacity at
the Late Horizon, and its members were dedi- Pachacamac remained modest (Table 1). Cer-
cated to fishing (Eeckhout 1999:403). Perhaps tainly the greater part of Pachacamac, outside
they contributed to the increase of products that the Second Wall but within the Third Wall, has
arrived at Pachacamac. The third factor, which not been excavated.10 Nevertheless, the little
one can add to that just mentioned, would be work that has been accomplished and published
that the Inca tribute requirements were greater does not suggest collcas in this place, but, rather,
than previously and probably added to the activity centers, modest residences (Malaga
traditional obligations of the population. In this
context it is interesting to mention the work of 9
Pyramid 13 was built and occupied during the Late
Jane Feltham (2006) on the tribute of clothing Horizon. Its design conforms to local tradition, but its
during the Late Horizon. On the basis of her location and specific characteristics suggest that it
study of the textiles found in excavations at functioned both as an elite residence and as an ushnu
Pachacamac, she suggests that this tribute was (ritual platform) associated with the Plaza of the Pilgrims
onerous to the people, who found themselves (Eeckhout and Farfn 2008). The palace of Tauri
Chumpi resulted from the adaptation and transformation
obligated to save time when making their own of a Pyramid with ramp into an Inca palace, the reason it
clothing, forcing them to look for production is also designated as Pyramid 10 (Eeckhout 2010a: 429).
methods that were faster than those employed
10
previously. By contrast, clothing made for trib- Pacahacamac is divided into three major sectors or
ute followed standards established by the Incas precincts. From south to north, the Sacred Precinct
includes the major temples and is enclosed by the First
and required more time. Wall. The Second Precinct encompasses the pyramids
with ramps and other buildings and is enclosed by the
A characteristic of Inca logistics was that Second Wall. The Third Precinct is a desertic pampa
they built their own administration facilities, but with no monumental structures and is within the Third
Wall.
ANDEAN PAST 10 (2012) - 224
2008), and Late Horizon and early colonial of the chiefdom (Eeckhout [editor] 2004; Eeck-
burials (Uhle 1903:62-70). Until we have more hout 2008a). There was no necessity for the
information, we must consider the Inca store- Incas to build more, so they simply took advan-
houses to be concentrated within the First and tage of what already existed. The context was
Second Walls, that is, within the monumental not the same in those parts of the highlands
part of the site, and, more specifically, in the where they established large sites like Pumpu,
sectors related to ceremonial activities. Hatun Xauxa, and Hunuco Pampa (LeVine
1992), or those in the Cochabamba Valleys
In this sense, one cannot consider Pacha- (Pereira 2011), which concentrated the harvests
camac to be a state installation of the major and resources of immense regions. Second, the
tambo type, used to house the imperial armies population of the chiefdom of Ychsma was not
on the march, and, therefore, designed to store so numerous, nor its productivity so large, as to
large quantities of goods. This was probably the justify the expenditure of enormous energy in
case with Hunuco Pampa, Hatun Xauxa, or the construction of hundreds of storehouses or
Pumpu, each of whose storage capacity is five or structures for their development (see Eeckhout
ten times larger than that of Pachacamac. The 2001; Feltham 1983:354-360).
storage at this latter site appears to be directed
principally towards religious functions, especially With respect to this, I can summarize
those relating to pilgrimage and administration Sneads general observations (1992:82-86) on
(the palace of Tauri Chumpi). Structure E8 the highland-coastal dichotomy in state storage.
could have played a hybrid role, that is, be at the The total number of Inca collcas on the coast
same time an instrument of local administration, represents no more than three percent of the
as I have described above, but also include, in its total number of imperial storehouses in the
placement and design, the functions of a tambo sample of 71 sites that he studied. Following
for traveling Inca officials and chaskis (messen- Snead, I think that this reflects the indirect
gers), but certainly not for an army on the move. control of the state in regions where the degree
of social integration was already very high, and
In other words, Structure E8 was involved in where the type of social management of labor
imperial administration at a local level, corre- was different from the sierra, explaining the
sponding to what Snead (1992:69, table 3-4) relatively low number of pre-Inca coastal store-
classifies as storage at secondary state facili- houses (ibid.: 85). Finally, it is my opinion that
ties. If we follow this classification, Structure the form of Inca storehouses on the coast, such
E8 embodies the characteristics of a large tambo, as I have described in this article, proceeded
but not a major one, that is, one containing a from a local tradition, because there is great
limited number of storehouses associated with a consistency, both in design and in size, in the
large number of facilities, as, for example, at Late Horizon. In the future this pattern may be
Tambo Colorado or Quebrada de la Vaca considered to be an indication of indirect con-
(ibid.:79-82). trol by the Inca over coastal provinces.
tific Research Fund) and the Fonds de la Recherche Bueno Mendoza, Alberto
Fondamentale Collective (Belgian Fund for Basic Collab- 1974-75 Cajamarquilla y Pachacamac: Dos ciudades de
orative Research), and from the Curtiss T. and Mary G. la costa central del Per. Boletn Bibliogrfico de
Brennan Foundation, Santa Fe, New Mexico. We also Antropologa Americana 37(46):171-211.
received logistical support from the Museo de Sitio de 1990 Hallazgo de un kipu en Pachacamac. In Quipu y
Pachacamac (Site Museum of Pachacamac), and from yupana: Coleccin de escritos, edited by Carol
the Belgian Embassy to Peru. I thank the entire excava- Mackey, Hugo Pereyra, Carlos Radicati, Hum-
tion team, especially Carlos Farfn, Peruvian national co- berto Rodriguez, and Oscar Valverde, pp. 97-
director of the Ychsma Project; Milton Lujn and Fran- 105. Lima: CONCYTEC.
cisco Vallejo, field assistants in charge of the sites Zone Cieza de Len, Pedro
E; and Kusi Colonna-Preti who is responsible for conser- 1962 [1553] La Crnica del Per. Coleccin Austral
vation and restoration; surveyor Valrie Decart; and Axel 507, third edition. Madrid: Calpe, S.A.
Beff, who created the three-dimensional reconstructions Conrad, Geoffrey W.
of buildings at the site. The figures that illustrate this 1977 Chiquitoy Viejo: An Inca Administrative Cen-
article were prepared by Nathalie Bloch (Centre de Re- ter in the Chicama Valley, Peru. Journal of Field
cherches Archologie and Patrimoine de la ULB [Center Archaeology 4(1):1-18.
of Archaeological Research and Patrimony of the Free DAltroy, Terence N.
University of Brussels]). An earlier version of this paper 1992 Provincial Power in the Inka Empire. Washington,
was presented at the colloquium Los Inkas y la Inter- D.C.: Smithsonian Institution Press.
accin de Sociedades, Paisajes, y Territorios en los Andes: 2002 The Incas. Malden, United Kingdom: Blackwell
Homenaje a Craig Morris, held at the Universidad Publishers.
Nacional Mayor de San Marcos, Lima, Per, June 14-18, DAltroy, Terence N., and Timothy K. Earle
2010. I thank the three anonymous Andean Past review- 1985 Staple Finance, Wealth Finance, and Storage in
ers who helped me improve this version and David the Inka Political Economy. Current Anthropol-
Pereira who kindly provided me with information on the ogy 26(2):187-206.
site of Cotapachi, Bolivia. I assume full responsibility for DAltroy, Terence N. and Christine Hastorf
the content of the present article, including any errors 1984 The Distribution and Contents of Inca State
that might exist. Storehouses in the Xauxa Region of Peru.
American Antiquity 49(2):334-349.
Translated from the Spanish by Monica Barnes Day, Kent C.
1982 Ciudadelas: Their Form and Function. In Chan
REFERENCES CITED Chan: Andean Desert City, edited by Michael
Moseley and Kent C. Day, pp. 55-66. Albuquer-
Ascher, Marcia and Robert Ascher que: University of New Mexico Press.
1981 Code of the Quipu: A Study in Media, Mathemat- Earle, Timothy K.
ics, and Culture. Ann Arbor: University of Mich- 1992 Storage and the Inka Imperial Economy: Ar-
igan Press. chaeological Research. In Inka Storage Systems,
Bauer, Brian S. edited by Terry LeVine, pp. 327-42. Norman:
2004 Ancient Cuzco: Heartland of the Inca. Austin: University of Oklahoma Press.
University of Texas Press. Eeckhout, Peter, editor
Bonavia, Duccio 2004 Arqueologa de la costa central del Per en los
1965 Arqueologa de Lurn: Seis sitios de ocupacin en la perodos tardios. Boletn Temtico del Instituto
parte inferior del valle. Tesis antropolgicas 4. Lima: Francs de Estudios Andinos 33(3).
Instituto de Estudios Etnolgicos del Museo Eeckhout, Peter
Nacional de la Cultura Peruana and Departa- 1998 Le temple de Pachacamac sous lempire inca.
mento de Antropologa de la Universidad Na- Journal de la Socit des Amricanistes 84:9-44.
cional Mayor de San Marcos. 1999 Pachacamac durant lIntermdiaire rcent: tude
Bray, Tamara L. dun site monumental prhispanique de la cte
2003 To Dine Splendidly: Imperial Pottery, Com- centrale du Prou. British Archaeological Reports
mensal Politics, and the Inca State. In The International Series 747. Oxford: Hadrian Books
Archaeology and Politics of Food and Feasting in Ltd.
Early States and Empires, edited by Tamara L. 2001 Monuments, temps et pouvoir: Force de travail
Bray, pp. 93-143. New York: Springer. et structure de lautorit Pachacamac, cte
centrale du Prou. In Itinraires belges aux
ANDEAN PAST 10 (2012) - 226
La Torre, Fabriciano and Consuelo Caja Morris, and Ramiro Matos Mendieta, pp. 135-
2005 Reconocimiento y registro del entorno territorial del 164. Washington, D.C.: Dumbarton Oaks Re-
Qhapaq an (Volumen II): El Qhapaq an en la search Library and Collection.
ruta del Chinchaysuyu entre Xauxa y Pachacamac. Morris, Craig and Donald Thompson
Lima: Instituto Nacional de Cultura. 1985 Huanuco Pampa: An Inca City and Its Hinterland.
LeVine, Terry, editor London and New York: Thames and Hudson.
1992 Inka Storage Systems. Norman: University of Moseley, Michael E.
Oklahoma Press. 1992 The Incas and their Ancestors. London and New
LeVine, Terry York: Thames and Hudson.
1992 Inka State Storage in Three Highland Regions: Moseley, Michael and Carol J. Mackey
A Comparative Study. In Inka Storage Systems, 1974 Twenty-Four Architectural Plans of Chan Chan,
edited by Terry LeVine, pp. 107-150. Norman: Peru. Cambridge, Massachusetts: Peabody
University of Oklahoma Press. Museum of Archaeology and Ethnology, Har-
Mackey, Carol vard University.
2006 Elite Residences at Farfn: A Comparison of the Paredes Botoni, Ponciano
Chim and Inka Occupations. In Palaces and 1988 PachacamacPiramide con Rampa n2. Boletin
Power in the Americas: From Peru to the Northwest de Lima 55:41-58.
Coast, edited by Jessica Joyce Christie and Prssinen, Martti
Patricia Joan Sarro, pp. 313-352. Austin: Uni- 2003 Tawantinsuyu: El estado inca y su organizacin
versity of Texas Press. poltica. Lima: Instituto Francs de Estudios
Malaga Villa, Mara Beln Andinos, Pontificia Universidad Catlica del
2008 Arquitectura domestica en las pampas de Per and the Finnish Embassy to Peru.
Pachacamac durante el Horiznte Tardo: Exca- Patterson, Thomas C.
vaciones en el sector SW de las unidades A-2, 1983 Pachacamac: An Andean Oracle under Inca
A3, y A-4. Tesis de Licenciatura en Arqueo- Rule. In Recent Studies in Andean Prehistory and
loga, Pontificia Universidad Catlica del Per, Protohistory: Papers from the Second Annual
Lima. Northeast Conference on Andean Archaeology and
Matos Mendieta, Ramiro Ethnhohistory, edited by Peter D. Kvietok and
1994 Pumpu: Centro administrativo inka de la puna de Daniel H. Sandweiss, pp. 159-176. Cornell
Junin. Lima: Editorial Horizonte. University Latin American Studies Program,
Mitchell, William P. Ithaca, New York.
1991 Peasants on the Edge: Crop, Cult, and Crisis in the Pease, Franklin G.Y.
Andes. Austin: University of Texas Press. 1992 Per Hombre e Historia II: Entre el siglo XVI y el
Morris, Craig XVIII. Lima: Ediciones Edubanco.
1967 Storage in Tawantinsuyu. Ph.D. dissertation, 1999 Curacas, Reciprocidad y Riqueza. Lima: Pontificia
University of Chicago. Universidad Catlica del Per, Fondo Editorial.
1986 Storage, Supply and Redistribution in the Econ- Pereira Herrera, David M.
omy of the Inka State. In Anthropological History 2011 Las qollqas incaicas de Cotapachi, Quillacollo,
of Andean Polities, edited by John V. Murra, Cochabamaba, Bolivia: Un tesoro de incalculable
Nathan Wachtel, and Jacques Revel, pp. 59-68. valor histrico y arqueolgico. Brochure distributed
Cambridge: Cambridge University Press. by the author. On file in the archives of Andean
1992a Foreword. In Inka Storage Systems, edited by Past.
Terry LeVine, pp. ix-xiii. Norman: University of Pizarro, Hernando
Oklahoma Press. 1872 [1533] A Letter of Hernando Pizarro to the
1992b The Technology of Highland Inka Food Stor- Royal Audience of Santo Domingo, November
age. In Inka Storage Systems, edited by Terry 1533. In Reports on the Discovery of Peru, Volume
LeVine, pp. 237-258. Norman: University of 3, edited by Clements R. Markham, pp. 111-
Oklahoma Press. 127. London: Hayklut Society.
Morris, Craig and Julin Idilio Santillana Radicati di Primeglio, Carlos
1978 Perspectiva arqueolgica de la economa in- 1990 Tableros de escaques en el antiguo Per. In
caica. Histrica 2(1):63-82. Quipu y Yupana: Coleccion de escritos, edited by
2007 The Inka: Transformation of the Chincha Carol Mackey, Hugo Pereyra, Carlos Radicati,
Capital. In Variations in the Expression of Inka Humberto Rodrguez, and Oscar Valverde, pp.
Power, edited by Richard L. Burger, Craig 205-218. Lima: CONCYTEC.
ANDEAN PAST 10 (2012) - 228
Figure 1: Plan of the site of Pachacamac (after Eeckhout and Owens 2008: figure 2).
ANDEAN PAST 10 (2012) - 230
Figure 3: Panorama and plan of Structure A1-A2, Pachacamac, along the Plaza of the Pilgrims.
231 - Eeckhout: Inca Storage at Pachacamac
Figure 7: Engraved wooden drinking vessel (kero) from Structure E8, Pachacamac.
235 - Eeckhout: Inca Storage at Pachacamac
Figure 10: Cotton quipu from an intrusive tomb in Pyramid with Ramp 3, Pachacamac.
ANDEAN PAST 10 (2012) - 238
Figure 11: Stone artifact from the outskirts of Pyramid with Ramp 11, Pachacamac.
239 - Eeckhout: Inca Storage at Pachacamac
John H. Walker
University of Central Florida
Movima) that predate the entry of Arawak archaeology, similarities were noted between the
languages. ceramics recovered in these excavations and
contemporaneous ceramics from the adjacent
Neves (2008:369) identifies the Guapor- Andean highlands. In 1936 Wendell Bennett
Mamor region as a possible center of cultural made this connection explicit, and associated it
innovation, if not a heartland from which a with sociopolitical complexity, classifying Lower
language group spread across the Amazon Basin. and Upper Velarde, Maisicito, and Hernmarck
Evidence suggests that the Guapor-Mamor as Sub-Andean (Bennett 1936). More recent
was a center for the development of several key analyses suggest the importance of other
Amazonian crops, including manioc, peanuts, geographic connections. Features such as large
and chili peppers (Piperno 2006). If the upper urns and polychrome wares suggest closer
Madeira-Guapor region was a locus of associations with Amazonian ceramic traditions,
agricultural innovation, we might expect to find rather than Andean ones. Lathrap distinguished
ceramic assemblages there that are associated between two phases in Mojos with Lower Velar-
with agricultural economies and share stylistic de and Chimay as the earlier, and Upper Velar-
traits with other regions. Neves argues that de and Hernmarck as the later. The earlier
recent data do not support a Central Amazon phase he associated ultimately with both the
heartland from which language-related Barrancoid tradition of the Central Amazon,
migrations originated. Instead, conjunctions and also with bichrome painting associated with
between language, ceramics, and agricultural the Andes (although earlier than, and not
landscapes are the most reliable basis on which necessarily associated with, Tiwanaku). The
to reconstruct culture history. description of more ceramic assemblages from
Mojeo contexts will help resolve these
As records of landscape management, raised sequences and connections (Jaimes Betancourt
fields (and other earthworks) are likely to be 2010; Walker 2011b).
associated with settlements that differ
significantly from settlements unassociated with THE IRUYAEZ RIVER IN THE LLANOS DE
earthworks. Raised fields can be thought of as MOJOS
industrial infrastructure, in the sense that they
were used to produce a range of different crops, The Llanos de Mojos (or Mojos) is a
substances, and valued objects, rather than an seasonally flooded tropical savanna (Figure 1),
exclusively agricultural infrastructure, producing created in part by a tangle of hydrologic features
only food crops. Research from across Mojos including major rivers and tributaries, seasonal
suggests that links between settlement and creeks, ancient riverbeds, oxbow lakes,
landscape are distinct in three recently oriented lakes, and both permanent and
presented examples (Erickson 2010; Lombardo seasonal wetlands (Hanagarth 1993). The
and Prmers 2010; Lombardo et al. 2011). Iruyaez River is one of the main tributaries of
the Mamor, and rides in a channel originally
Excavations conducted early in the formed by the Beni River, which today forms
twentieth century in southeastern Mojos have Mojos western boundary. The annual cycle
been used to characterize Mojos ceramics, and between wet and dry seasons determines much
investigations have continued there in the past of Mojos land use. As much as half of the
decade (Nordenskild 1913, 1924a, 1924b; landscape can be underwater in February, while
Prmers 2000, 2001, 2002). During early August can pass without any rainfall. The
syntheses of Amazonian (and South American) movements of water across slight differences in
ANDEAN PAST 10 (2012) - 244
elevation determine soil conditions and land chiefs who have authority within a single
use. This landscape is a record of the long and settlement (Eder 1985 [1791]).
complex history of rivers moving across the
savanna, cutting channels, building levees, and El Cerro
erasing them. It also records more than 2,500
years of human habitation and landscape El Cerro is a large island of forested high
management (Denevan 1966, 2001; Erickson ground named for a rock formation
2000, 2006; Erickson and Bale 2006; Erickson approximately forty meters high and ten
and Walker 2009; Walker 2004, 2008a, 2008b). hectares in extent, surrounded by about 110.8
Precolumbian farmers used earthworks to hectares of forest (Figure 2). The rock formation
extend planting seasons and improve soil is the highest point for several hundred
conditions for cultivation, and perhaps for a kilometers, a landmark for both horsemen and
variety of other purposes, including airplane pilots. The surrounding high ground
transportation, controlling water and fire, or has been occupied continuously throughout
ritual processions (Walker 2011c). local memory. A permanent stream (Cerro
Creek) lies along the southwest side, and the
The community at the Cerro Site1 used Iruyaez River lies three kilometers to the
raised fields, probably both producing food and south. The entire dry forest at Cerro contains
raw materials for a wide range of industrial evidence of modern occupation, in the form of
products. Lowland societies create rich and slash-and-burn fields, both abandoned and in
varied material cultures from animal and use, trails, abandoned houses, and mango
botanical raw materials (Roth 1916). The groves.
occupied area within the forest at Cerro was
more than eight hectares and could easily have Transects
been as large as thirty hectares. This settlement
size suggests a population greater than five Two transects of shovel test excavations,
hundred, and is consistent with later ethno- each approximately fifty centimeters in diameter
historic descriptions of 1,800-2,000 person were excavated at right angles to each other in
towns (Zapata 1906 [1693]:25-26). Political order to sample a large area of the forest.
leaders from this area are in one case said to Transect 1 extended due east of a brick-lined
have held influence over several settlements. well at the modern Cerro ranch (ca. 1997), and
The ethnohistoric descriptions cited above, extended through the levee into the savanna.
which loom large in later classifications of Its total length was 615 meters (41 individual
Mojeo societies (Steward and Faron shovel tests). Beginning from where Transect 1
1959:350), are based on descriptions of the exited the forest, Transect 2 extended due south
Cayubaba, identified with north-central Mojos, through the levee and out into the savanna
near the Iruyaez River (Block 1994; Zapata (Figure 3). Its total length was 1140 meters (76
1906 [1693]). The Mojo and Baure, both shovel tests).
Arawak groups located further south and east,
are associated with smaller settlements and The high ground is covered with mixed palm
forest, punctuated by a few modern swidden
fields. The soils of the northern and western
1
To avoid confusion, in this paper I refer to the rock part of the levee (sampled in Transect 1)
formation as El Cerro and call a modern ranch and the contrasted strongly with the soils of the central
area in general Cerro, and the archaeological site the and southern parts of the levee (sampled in
Cerro Site.
245 - Walker: Fourteenth Century Llanos de Mojos
Transect 2). Reddish sandy soils with few sherds trench was two meters by two meters square,
were found in the shovel tests of Transect 1. oriented to magnetic north. The trench was
Dark soils, burned clay, and dense deposits of excavated according to natural stratigraphy, and
sherds were found in most Transect 2 shovel excavation was controlled with a transit. The
tests, and sherds were recovered on the surface trench was excavated to at least five centimeters
in some locations. Transect 2 intersected a into the subsoil throughout, about 120 centi-
deposit of anthrosol that was approximately 500 meters below the surface. All excavated soil was
meters across (Figure 3). The definition of screened using a one centimeter screen.
anthrosol is based on three criteria: dark soil,
ceramics, and burned clay soil, and is further Stratum 1, beginning from the ground
examined elsewhere (Walker 2004:63-110, surface, extended across the trench and was
2011a). between ten and twenty centimeters thick. It
was a black (10YR2/1 in the Munsell color
The total area of occupation at the Cerro system), very friable, sandy loam, and passed
Site is difficult to determine. One method used easily through the screen. The interface with
is to map a circle around the shovel tests that Stratum 2 below was very gradual and smooth.
contained sherds. These circles provide an The difference between Strata 1 and 2 was
estimate for the area of occupation intersected probably no more than a slight difference in the
by Transect 2. The smaller of the two circles in amount of organic material present. Stratum 1
Figure 2 encloses twenty-two shovel tests, had a weak and very fine-grained structure.
covering 330 meters, all of which contained Wood charcoal was present, and two sherds
sherds (translating to an area of 8.5 hectares). were recovered in excavation.
The larger circle encloses a total of thirty-four
shovel tests, covering 510 meters, all but two of Stratum 2 was immediately below Stratum 1,
which contained sherds (an area of twenty and also extended throughout the trench.
hectares). A third estimate takes the total Stratum 2 was a very dark brown (10YR2/2)
number of shovel tests (117), divides by the sandy loam. The texture of the soil was very
number of shovel tests yielding ceramics (41), loose and friable. The interface with Stratum 1
and multiplies by the total area of the forest above was very gradual and smooth. The
(110.77 hectares) to generate an estimate of interface with Stratum 3 below was gradual and
38.8 hectares of settlement. Based on this range uneven. Wood charcoal was present, and
of estimates, Transect 2 at the Cerro Site crosses seventy-five sherds, weighing 1.50 kilograms
a site with an area between 8.5 hectares and were recovered.
38.8 hectares, and most likely more than twenty
hectares. The transects document a Stratum 3 was immediately below Stratum 2,
precolumbian settlement that covered a and also extended throughout the trench. The
significant fraction of the levee surrounding the soil was a very dark grayish brown (10YR3/2)
Cerro rock formation. sandy loam. The soil texture was somewhat
friable, but compact. The interface with Stratum
Trench 3 2 above was gradual and uneven, and the
interfaces with the three strata below, Strata 4,
This trench was located adjacent to 5, and 6, were uneven and abrupt. No soil
Transect 2, near shovel tests in which sherds structure was present. Carbon was recovered
and charcoal were recovered to a depth of 120 from this stratum as well as burned clay. A total
centimeters below the surface (Figure 4). The
ANDEAN PAST 10 (2012) - 246
of 311 sherds, weighing 6.550 kilograms were total of 52 sherds were recovered, weighing
recovered. 4.750 kilograms. In some locations, large sherds
defined the lower interface of this stratum.
Stratum 4 was below Stratum 3 and above
Stratum 5, in the southwest quadrant of Trench Trench 3 Interpretation
3. It was a dark gray (10YR3/1) to black
(10YR2/1) silt loam. The interface with Stratum The color of Strata 1 and 2 was darker and
3 above was abrupt and almost level. The grayer than the soils sampled in the shovel tests
interface with Stratum 5 below was abrupt and from Transect 1. Strata 1 and 2 had high
uneven. No soil structure was present. Wood concentrations of sherds. In combination with
charcoal was present, as well as burned clay, and Strata 3, 4, and 6, they are interpreted as
large numbers of sherds. In some places, large anthrosols, representing prehispanic occupation.
sherds defined the lower interface of Stratum 4.
The large sherds of the feature totaled 323, Stratum 3 is the thickest of the five soil
weighing 16.425 kilograms, while sherds taken strata interpreted as anthrosols. Dense
from the soil within the feature totaled 45 and concentrations of sherds were found in
weighed 1.800 kilograms. association with small pieces of wood charcoal
in this stratum. In parts of the trench, Stratum
Stratum 5 extended throughout the trench, 3 has an uneven lower limit. The lower
beneath Strata 3, 4, and 6. It was a red elevations in Stratum 3 correspond to Strata 4
(2.5YR3/6) clay loam. The soil was compact and and 6. Strata 4 and 6 are associated with
hard. Some sherds were recovered near the Stratum 3. The presence of anthrosols is
interfaces with the strata above, but no sherds demonstrated by soil color, high concentrations
were recovered deeper than 2 or 3 centimeters of wood charcoal and sherds, and the presence
into the stratum. The interfaces with all of the of burned clay.
strata above (Strata 3, 4, 6) were both uneven
and abrupt. There was some wood charcoal in Strata 4 and 6 are interpreted as parts of
Stratum 5. The soil had a blocky structure, with trash pits, used for the disposal of broken
blocks of less than 5 centimeters in diameter. pottery, ashes, and other household debris. The
The soil structure was very weak. Near the recovered sherds had no discernible spatial
upper interfaces with Strata 3, 4, and 6 a total of pattern. Some sherds were oriented vertically
10 sherds were recovered, weighing 0.350 and others horizontally. In addition, the lower
kilograms. boundaries of the stratum were not even,
suggesting a secondary deposit, such as a
Stratum 6 (not represented in the Figure 4 garbage pit, rather than a primary deposit such
profile) was between Stratum 3 above and as a house floor, or an activity area. In Stratum
Stratum 5 below, in the northeastern quadrant 6, in addition to the sherds, many pieces of
of the trench. It extended from a maximum burned clay and of wood charcoal were also
elevation of 99.04 meters to a minimum recovered, but they had no clear spatial pattern.
elevation of 98.64 meters. It was a dark gray to
very dark grayish brown (10YR3/1) sandy loam. Stratum 5 is interpreted as culturally sterile
Its interface with Stratum 3 above was almost subsoil. It was very hard and compact, and with
level, and abrupt. The interface with Stratum 5 the exception of the interfaces with the strata
below was uneven and abrupt. No soil structure above, no sherds were recovered from Stratum
is present. Some wood charcoal was present. A 5. The color and texture of this soil were
247 - Walker: Fourteenth Century Llanos de Mojos
Sample Type Lab No. Stratum 14C Age BP Calibrated age 68% range Calendric Age cal AD range
BP AD
Table 1. Calibrated radiocarbon dates from the Iruyaez River, Beni, Bolivia
When calibrated, all of the dates fall into a settlement during the two centuries before the
single group, with center points ranging from Spanish arrived in the New World.
A.D. 1342 to A.D. 1465, and 68 percent
confidence intervals ranging from A.D. 1303 to Trench 4
A.D. 1570. The dates cluster in the mid-
fourteenth century A.D. (Table 1). The dates This trench was located in a raised field
suggest that Stratum 4 may represent a longer platform to the north of the modern Cerro
history of activity than the surrounding strata 2, Ranch, just outside the fenced pasture in the
3 and 6. open savanna. The trench was six meters long
and fifty centimeters wide. No artifacts were
These samples establish the date of the recovered from Trench 4, nor was any charcoal
occupation at the Cerro Site to the fourteenth obtained for radiocarbon analysis. The raised
century A.D. In combination with the evidence field platform and canal bisected by Trench 4
from Transects 1 and 2, the dates suggest that are associated with the Cerro Site by their
the forest at Cerro was occupied by a large proximity.
ANDEAN PAST 10 (2012) - 248
Large cooking pots are represented by two the body and base cannot be reconstructed. The
examples (Figure 5). The openings of the vessels opening is twelve centimeters in diameter, and
are large (100 centimeters and 104 centimeters the neck is at least eight centimeters tall. This
or larger) and the lips of the vessels are thick vessel form could be associated with brewing or
(greater than two centimeters) and broad (about storage.
five centimeters). The rim profiles suggest
outward sloping vessel walls, perhaps with a One straight sided, slightly outward sloping,
carination, and a shallow, curved base or a flat flattened rim could represent a small carinated
base. These are close analogues to the ahukugu dish with high sides (opening is twenty-eight
described from both the modern and the centimeters in diameter) or possibly the neck of
precolumbian Xingu. a jar.
Flat-bottomed dishes are represented by two Two examples of thin, outward flaring rims
examples, with outsloping sidewalls and less (both openings are eighteen centimeters in
than ten centimeters tall. The bottoms of the diameter) curve down and could represent
plates are about as thick as the walls of the necked jars, or less probably carinated dishes.
dishes, and in one case the inside of the walls is One has an adorno2 handle, less than one
decorated with rough incision, and the interior centimeter thick, and both are thin-walled
bottom of the dish has shallow grooves. Groov- (about five millimeters).
ed platters are a common vessel form further
south in Mojos (Jaimes Bentancourt 2010), The remainder of the reconstructed rims
although this example is distinctive in that the represents a range of forms between the carina-
bottom is flat and not gently curved, and the ted, lipped dishes and the large buckpots
grooves are more deep incisions than channels (Figure 5). All have pronounced lips, eight
with rectangular profiles, as in other examples. flattened and three rounded. All slope outward;
These dishes are fairly large, about sixty and some are more straight sided and some more
fifty-four centimeters in diameter. steeply sloping. As a group, they resemble the
lipped, carinated dishes, but vessel
Two vessels represent carinated, lipless reconstruction does not extend to the
dishes. They have straight, slightly outsloping carination and the base. They may represent a
sides, are less than ten centimeters tall, and large, necked vessel form, but a shallow dish
have curving bases that are thinner than the seems more likely on the basis of other vessel
walls. The rim is curved rather than flattened, forms present at the Cerro Site.
and has no lip. Rim diameters are forty-two and
forty-eight centimeters. One ceramic grinder was recovered in
excavation (and another from surface
A single vessel, with a rim diameter of collection). This fired clay object has a
eighteen centimeters, represents shallow open rectangular cross section, the result of wear on
bowls. The bowl has a rounded base and a the two opposite sides which changed a circular
rounded rim, with thick walls. The fabric of this cross-section into a rectangular cross section
vessel has a pinkish cast, perhaps from high iron (Figure 6). Ceramic grinders (moleadores, manos
content. de moler) are common to the south, along the
Apere River and in Southern Mojos (e.g. Jaimes sherds from San Juan, about twelve kilometers
Betancourt 2010:207, 226, 272, 314, 333). They to the southwest. Fine-line painted, sponge
were likely used for grinding and processing tempered sherds are found in strata 3, 4, 5 and
food, as seen in the wear pattern. Their use as 6, which are all contemporaneous, according to
bark beaters seems unlikely, because many the radiocarbon dates.
examples are worn, and few are broken (which
would presumably occur through percussive A few sherds from the Cerro Site have rough
use). Their use as fire-dogs or vessel supports incision, present on the inside of grooved
seems less likely because none of them show platters, for instance. This incision does not
carbon deposits associated with long term resemble the fine-lined incision present on
exposure to fire. However, in the stone-free ceramics in southern Mojos, or that combined
environment of Mojos, ceramic artifacts would with painting in other Amazonian contexts.
presumably have multiple functions during their
use-lives. One sherd from Stratum 5 was corrugated,
a distinctive surface treatment not otherwise
Several sherds have deep intersecting present on the Iruyaez.
grooves that can be identified as shaft
straighteners, and probably represent a reuse of Fabric and temper
broken pottery.
Three distinct pastes are represented at the
The final artifact type from the Cerro Site is Cerro Site: one dominant mode, and two
a stone axe. Two T-shaped stone axes were subordinate modes. The most common is a grog-
recovered, one from the surface, and one from tempered paste, in some cases including some
excavation. The excavated example had an sponge temper. The surface of this fabric is
uneven edge (Figure 7) suggesting that the axe smooth and abrades easily, having almost a
was broken and resharpened. Stone of any type powdery or soapy feel. The color of this fabric
is not native to Mojos, and these artifacts must ranges widely, including buff, pink, pale tan,
have either come from the Brazilian shield to light gray, and light brown. All reconstructed
the north (across the Guapor or Itnez River) vessels were made from this fabric.
or from the Andes to the south.
The second paste is a sponge tempered
Surface decoration fabric, usually better fired, lighter, and making a
clear sound when struck. Painted sherds are all
Ceramics from the Cerro Site are not usually of this fabric (although not all sponge tempered
decorated, but do present a range of surface sherds are painted). Colors are darker, with
decoration. By count (as opposed to weight), more grays and browns, and no buff or pink
14.29 percent of the sherd lots have surface examples. No vessel forms can be associated
decoration. with this fabric. Sponge tempered fabric is
associated with fine-line painting in San Juan
Fine-line painting, usually on sponge- ceramics (Walker 2011b).
tempered sherds, is known both from surface
collections and excavation at the Cerro Site. No The third paste is a distinctive reddish color,
vessel forms with this form of surface decoration with grit, sand, or rock temper. The appearance
can be reconstructed. The appearance of these of this fabric is very rough, and it looks much
sherds is very similar to the fine-line painting on like the cascajo or eroded lateritic rock of El
251 - Walker: Fourteenth Century Llanos de Mojos
Cerro. The temper of this fabric is likely taken fourteenth centuries A.D. along the Iruyaez
directly from the nearby El Cerro, and the red River.
color of the fabric indicates large amounts of
iron in the clay. No vessel forms or surface The ratio of painted to unpainted sherds
treatments can be associated with this fabric. between the two sites differs significantly. The
San Juan Site has a higher proportion of painted
DISCUSSION sherds than the Cerro Site. At San Juan 65.22
percent of the lots contain painted sherds, while
Radiocarbon dates place this assemblage in only 14.29 percent of the lots at the Cerro Site
the fourteenth and fifteenth centuries A.D. contain painted sherds. Additionally, sherds
Calibration indicates that these dates predate painted on both sides were found at San Juan,
A.D. 1492, and direct contact with Europeans, but not at the Cerro Site. It is possible that the
who arrived in Mojos to stay in the mid- Cerro Site represents a more generalized
seventeeth century. domestic context, with a cross-section of
ceramic refuse including more utilitarian vessel
Analysis of the excavated ceramic forms. Without pushing the analysis too far,
assemblage suggests that the community was Cerro ceramics represent a more complete set of
part of a regional system or ecumene that ceramic forms. The relationship between the
spanned at least several hundred kilometers. Cerro Site and San Juan is one of some
Both vessel forms and surface treatments, as continuity within a larger sequence.
well as one example of raw material (ground
stone) illustrate four clear connections and one Vessel forms suggest a link between the
potential connection with other areas. Iruyaez and the Xingu Basin. The most
common vessel form at the Cerro Site is a
The two stone axes necessarily indicate a lipped, carinated dish with a flat or slightly
link with either the highlands of the Guapor curved bottom. The range of the mouth
basin, or with the Andes to the south. No lithic diameters of these vessels includes examples
raw material is available in Mojos, and so the larger than one meter. In the largest examples,
presence of ground stone artifacts establishes this vessel form is analogous to the ahukugu or
such a link. T-shaped stone axes have long been buckpot described from the Xingu, both in
associated with the Andes, rather than the modern and precolumbian contexts (Hecken-
Amazon (e.g. Lathrap and Roys 1963). Sourcing berger 2005). Probably used for the preparation
this raw material will establish the direction of of foods in large quantities, this vessel form
this link, but both potential sources of stone suggests a similarity in economy and cuisine
should be considered. across the Southern Amazon. For example, Dole
(1964) suggested that flat bottomed and conical
Painted ceramics at the Cerro Site cannot be vessels are associated with cooking on raised
conclusively identified with the previous San platforms and dirt floors, respectively.
Juan assemblage (Walker 2011b) but the surface
decoration, fabric, and temper are very similar. Two vessel forms: ceramic cylinders
No painted vessel forms are available from the (interpreted as grinders) and grater plates
Cerro Site, such as the distinctive flaring (although with shallower grooves) demonstrate
decorated rims from San Juan. The similarities a link to Southern Mojos. Both vessel forms are
in painting nevertheless argue for a degree of well-known from the Apere River and also the
cultural continuity between the fifth and mounds near Trinidad to the south (Jaimes
ANDEAN PAST 10 (2012) - 252
Betancourt 2010; Prmers 2000, 2001, 2002). Similarities in material culture across the
Both forms are used for the preparation of food Amazon are also part of interpretations of the
and drink. This argues for continuities in food combination of archaeological data and
preparation and cuisine across precolumbian linguistic data that support an expanded role for
Mojos. In the future, trace analysis of ceramic Arawak speaking societies in Amazonian culture
contents will address this question. history (Heckenberger 2005; Hornborg 2005).
The firm dating of the Cerro Sites artifact
Jaimes Betancourts superb study of assemblage to the fourteenth and fifteenth
ceramics from Loma Salvatierra (2010) was not centuries places it at the same period as a
extensively analyzed for this article, but a variety of complex societies throughout South
preliminary examination of the assemblage does America, from Marajoara, to the Xingu, to the
not suggest a particularly strong relationship Inca empire.
between the two locations. Loma Salvatierra
ceramics include a wide range of forms, many of The evidence from test excavations suggests
which are also common along the Ro Apere. that a large community of people, probably
For example, grater plates and moleadores are numbering in the low thousands, used these
two forms that are common at Loma Salvatierra artifacts. The Cerro Site is not an isolated
and along the Ro Apere, but are rare at the settlement, but a part of a landscape of raised
Cerro Site. Future studies will help illuminate fields that stretches for at least 50 kilometers
the relationship between these assemblages. from north to south, and 50 kilometers along
the Iruyaez and Omi Rivers. In considering the
A final point of comparison is tentative deep history of this landscape, both settlement
because it is based on a single corrugated sherd and landscape may be considered together,
from Stratum 5. Nevertheless, because this form rather than as independent lines of evidence.
of surface decoration is strongly associated with Explanations of why Mojos was the setting for
Guaran speakers in the ethnohistoric and not one, but at least five distinct landscapes,
archaeological literature, from southeastern each representing a distinctive investment of
Bolivia and Argentina (Alconini 2004:412-413), labor into infrastructure, is linked to settlement.
and corrugation is distinct from other surface
treatments at the Cerro Site, this difference These five lines of evidence from the artifact
merits consideration. assemblage at the Cerro Site suggest that links
from the Xingu to the northeast, the Guapor to
CONCLUSIONS the north, southeastern Bolivia and Argentina
to the southeast, and Southern Mojos and the
The artifact assemblage at the Cerro Site Andes to the south, were all part of the
suggests connections with four or five distinct economic, political, and social universe that the
regions, both within and outside Mojos. Cerro community inhabited. The Cerro Site is
Previously, connections have been drawn within six kilometers of the confluence of the
between Mojos and the Andes, Southeastern Iruyaez and Omi Rivers, and 25 kilometers of
Bolivia, Northwestern Argentina, and the the confluence of the Iruyaez and Mamor
Central Amazon (Lathrap 1970b). The Rivers. It therefore occupies a central location
assemblage at the Cerro Site fits into this picture within the continental network of South
of an interconnected Amazonian system, or American rivers described in the Handbook of
perhaps even an integrated South American South American Indians (Lowie 1948). The
system that included the Andes (Lathrap 1971). lowlands of South America are connected across
253 - Walker: Fourteenth Century Llanos de Mojos
Ecology, edited by W. Bale and Clark Erickson. Bolivian Amazonia. Journal of Archaeological
New York: Columbia University Press. Science 37: 1875-1885.
Erickson, Clark L. and John H. Walker Lowie, Robert H.
2009 Causeways as Landscape Capital. In Landscapes 1948 The Tropical Forests: An Introduction. In
of Movement, edited by James E. Snead, Clark Handbook of South American Indians, Volume 3,
Erickson, and J. Andrew Darling. Philadelphia, The Tropical Forest Tribes, edited by Julian H.
Pennsylvania: University Museum Publications. Steward, pp. 1-57. Washington, D.C.: United
Glaser, Bruno and William I. Woods States Government Printing Office.
2004 Amazonian Dark Earths: Explorations in Space and Mann, Charles
Time. Berlin: Springer-Verlag. 2008 Ancient Earthmovers of the Amazon, Science
Hanagarth, Werner 321:1148-1152.
1993 Acerca de la geoecologa de las sabanas del Beni en Meggers, Betty J.
el noroeste de Bolivia. La Paz: Instituto de Ecolo- 1971 Amazonia: Man and Culture in a Counterfeit
ga. Paradise. New York: Aldine.
Heckenberger, Michael Murra, John Victor
2005 The Ecology of Power: Culture, Place, and Person- 1972 El control vertical de un mximo de pisos ecolo-
hood in the Southern Amazon, A.D. 1000-2000. gicos en la economa de las sociedades andinas,
New York: Routledge. In Visita de la provincia de Len de Hunuco en
Hornborg, Alf 1562, edited by John Victor Murra, 427-476.
2005 Ethnogenesis, Regional Integration, and Ecology Hunuco, Per: Universidad Nacional Hermilio
in Prehistoric Amazonia. Current Anthropology Valdizn.
46:589-620. Neves, Eduardo G.
Jaimes Betancourt, Carla 2008 Ecology, Ceramic Chronology and Distribution,
2010 Die Keramike der Loma Salvatierra/La ceramica de Long-Term History, and Political Change in the
la Loma Salvatierra. Doctoral Dissertation, Amazonian Floodplain. In Handbook of South
Universitt zu Bonn. American Archaeology, edited by Helaine Silver-
Lathrap, Donald W. man and William H. Isbell, pp. 359-379. New
1970a Review of Archologische Untersuchungen am York: Springer.
Mittlern Amazonas. American Antiquity Nordenskild, Erland
35(4):499-501. 1913 Urnengrber und Mounds im bolivianischen
1970b The Upper Amazon. New York: Praeger. Flachlande. Baessler Archiv 3(5):205-255.
1971 The Tropical Forest and the Cultural Context of 1924a Finds of Graves and Old Dwelling-Places on the
Chavn. In Dumbarton Oaks Conference on Cha- Ro Beni, Bolivia. Ymer 44:229-237.
vn, edited by Elizabeth P. Benson, pp. 73-100. 1924b The Ethnography of South-America Seen from
Washington, D.C.: Dumbarton Oaks Research Mojos in Bolivia. Gothenberg, Sweden: Erlanders
Library and Collection. Boktryckeri Aktiebolag.
Lathrap, Donald W. And Lawrence Roys Prssinen, Marti, Denise P. Schaan, and Alceu Ranzi
1963 The Archaeology of the Cave of the Owls in the 2009 Pre-Columbian Geometric Earthworks in the
Upper Montaa of Peru. American Antiquity Upper Purs: A Complex Society in Western
29(1):27-38. Amazonia. Antiquity 83:1084-1095.
Lehmann, Johannes, D. C. Kern, Bruno Glaser, and Piperno, Dolores R.
William I. Woods 2006 The Origins of Plant Cultivation and
2003 Amazonian Dark Earths: Origin, Properties and Domestication in the Neotropics: A Behavioral
Management. Dordrecht, The Netherlands and Ecological Perspective. In Behavioral Ecology and
Boston, Massachusetts: Kluwer Academic the Transition to Agriculture, edited by Douglas J.
Publishers. Kennett and Bruce Winterhalder, pp. 137-166.
Lombardo, Umberto, Elisa Canal-Beeby, Seraina Fehr, Berkeley: University of California Press.
and Heinz Veit Prmers, Heiko
2011 Raised Fields in the Bolivian Amazonia: A 2000 Informe de labores: Excavacines arqueolgicas en
Prehistoric Green Revolution or a Flood Risk la loma Mendoza (Trinidad), 1ra Temporada, 1991
Mitigation Strategy? Journal of Archaeological Bonn: Instituto Alemn de Arqueologa.
Science 38:502-512. 2001 Informe de labores: Excavacines arqueolgicas en
Lombardo, Umberto and Heiko Prmers la loma Mendoza (Trinidad), 2da Temporada,
2010 Pre-Columbian Human Occupation Patterns in 2000. Bonn: Instituto Alemn de Arqueologa.
the Eastern Plains of the Llanos de Moxos,
255 - Walker: Fourteenth Century Llanos de Mojos
2002 Informe de labores: Excavacines arqueolgicas en Woods, William I., W. G. Texeira, Johannes Lehmann, C.
la loma Mendoza (Trinidad), 3ra Temporada, Steiner, A.M.G.A. WinklerPrins, and Lilian Rebellato
2001. Bonn: Instituto Alemn de Arqueologa. 2009 Amazonian Dark Earths: Wim Sombroek's Vision.
Raymond, J. Scott, Warren R. DeBoer, and Peter G. Roe New York: Springer Dordrecht, The
1975 Cumancaya: A Peruvian Ceramic Tradition. Netherlands.
Occasional Papers 2 (Department of Zapata, Agustn
Archaeology, The University of Calgary, Calgary, 1906 [1693] Carta del Padre Agustn Zapata al
Alberta, Canada). Padre Joseph Buenda, en la que da noticias del
Roth, Walter E. Paititi. In Juicio de lmites entre el Per y Bolivia,
1916 The Arts, Crafts and Customs of the Guiana edited by V. M. Maurtua, pp. 24-28. Barcelona:
Indians. Washington, D.C.: Bureau of American Imprenta de Henrich y Compaa.
Ethnology. United States Government Printing
Office.
Snead, James E., Clark L. Erickson, and J. Andrew
Darling, editors
2009 Landscapes of Movement: Trails, Paths, and Roads
in Anthropological Perspective. Philadelphia:
University of Pennsylvania Museum of
Archaeology and Anthropology.
Stahl, Peter
2010 Adventive Vertebrates and Historical Ecology in
the Pre-Columbian Neotropics. Diversity
1(2):151-165.
Steward, Julian H. And Leslie C. Faron
1959 Native Peoples of South America. New York:
McGraw-Hill.
Troll, Carl
1966 Landscape Ecology. Delft, Netherlands: ITC-
UNESCO .
Walker, John H.
2004 Agricultural Change in the Bolivian Amazon/
Cambio agricola en la Amazonia boliviana.
University of Pittsburgh Memoirs in Latin
American Archaeology 13. Pittsburgh,
Pennsylvania: University of Pittsburgh,
Department of Anthropology.
2008a Pre-Columbian Ring Ditches Along the Yacuma
and Rapulo Rivers, Beni, Bolivia: A Preliminary
Review. Journal of Field Archaeology 33(4):413-
427.
2008b The Llanos de Mojos. In Handbook of South
American Archaeology, edited by Helaine Silver-
man and William H. Isbell, pp. 927-940. New
York: Springer.
2011a Amazonian Dark Earth and Ring Ditches in the
Central Llanos de Mojos, Bolivia. Culture,
Agriculture, Food and Environment 33(1):2-14.
2011b Ceramic Assemblages and Landscape in the
Mid-1st Millennium Llanos de Mojos, Beni,
Bolivia. Journal of Field Archaeology 36(2):119-
131.
2011c Social Implications from Agricultural Taskscapes
in the Southwestern Amazon. Latin American
Antiquity 22(3):275-295.
ANDEAN PAST 10 (2012) - 256
Figure 1: Cerro, located along the Iruyaez River, a tributary of the Mamor,
on the upper Madeira, within the Llanos de Mojos, Beni, Bolivia.
257 - Walker: Fourteenth Century Llanos de Mojos
Figure 2: Map showing the locations of test excavations at the Cerro Site,
including two transects of shovel test excavations totaling 1755 meters,
trench 3 in the forest, and trench 4 bisecting a raised fields.
ANDEAN PAST 10 (2012) - 258
Figure 3: Schematic profile of Transect 2 at the Cerro Site, juxtaposing surface vegetation and land use,
and subsurface ceramics, dark soil, and burned clay soil.
The location of trench 2 is shown. Surface features are not to scale.
259 - Walker: Fourteenth Century Llanos de Mojos
Figure 6: Ceramic cylinders recovered from excavation (top) and from surface collection (bottom).
PERU Yanaorco
Recent Excavations at the Late Intermediate Yanaorco is on the continental divide near
Period Village of Yanaorco in the Cajamarca the headwaters of the Jequetepeque River. Its
Highlands inhabitants were part of the Cajamarca culture.
The town is on a defensible ridge, with natural
Jason L. Toohey (University of Wyoming, drops of approximately 500 meters on its north,
e-mail: jtoohey2@uwyo.edu) excavated at the west, and south sides. To further improve the
site of Yanaorco in the Cajamarca region of natural defenses of the site, its inhabitants
northern Peru in 2003 and 2004 (Instituto constructed two sets of fortifications on its
Nacional de Cultura permit number C/138- gently sloping eastern edge. Each set includes
2003/DREPH) with the support of a Fulbright three large walls. The largest retaining walls
Hays Dissertation Research Abroad Grant. stand up to three or four meters high and are
two meters thick (Figure 1). Each wall has a dry
Radiocarbon dates and ceramic chronology ditch on its exterior side. Many of these walls
place the occupation of Yanaorco within the also have abutting parapets along their interior
Late Intermediate Period (LIP; A.D. 1000-A.D. faces. These works indicate that the location
1465; Toohey 2009) when it was a major popu- occupied by Yanaorco was important and con-
lation center (Sachun 1986). It stands at ap- tested.
proximately 3550 m.a.s.l. and its occupants
practiced agro-pastoralism, taking advantage of
both the upper Quechua ecological zone below
3500 meters and the Suni zone herding re-
sources to its southeast (above approximately
3500 m.a.s.l.; Pulgar Vidal 1981).
In addition to occupying a defensive loca- nity identity for residents of Yanaorco. Non-
tion, Yanaorco is situated strategically to over- domestic, public areas of architecture are also
see traffic along the primary route in both present, including a sector of possibly ceremo-
prehistoric and modern times between the coast nial spatial features. These include a large, U-
and the highland Cajamarca Basin. Yanaorco shaped plaza (Architectural Unit 18A) closely
overlooks the Gaviln Pass into the basin by 500 associated with a set of two small rooms charac-
meters. Survey uncovered the remains of a terized by extremely limited and controlled
prehispanic trail connecting the pass to one of access (Figure 2). Excavation in one of these
the villages two entrances, physically tying the rooms indicated the presence of a low bench
community to the pass. Leadership within the along one wall and multiple niches recessed into
community would thus have been in a position the walls. Excavation in the open plaza revealed
to closely monitor, and possibly control, traffic a level activity surface and several small serving
through the pass. vessels. In the future, excavation will include an
area of four possible storage rooms, spatially
The architectural organization of Yanaorco associated with a zone of elite architectural
is complex and indicates that the original com- spaces. All of these integrated architectural
munity was rapidly augmented at some point spaces indicate that the community may have
during its occupation. The site contains two been partially planned when first constructed
principal architectural zones. The older sector is and that it did not grow haphazardly over time.
on the western end of the ridge upon which the
site sits. This sector was ringed with three large
fortification walls. At a later date, the commu-
nity grew with the construction of a newer
sector of adjoined rectangular rooms to the
southeast, outside the original fortifications.
Ultimately this new set of rooms was fortified
with three additional walls and ditches, the
longest wall connecting to one of the original
walls and effectively completing the fortification
of the entire community.
were recovered from the same room. Interviews access to camelid resources more frequently, and
with local landowners revealed the presence of in larger amounts. Excavated faunal remains
two ceramic whistles that had been collected from this room include large limb bones with cut
from an area of the site now under cultivation. marks. Many fragments also exhibit evidence of
These whistles contained modeled human faces. burning or roasting. On the other hand, very
small, friable bone fragments dominate the
faunal evidence from non-elite spaces. These
remains may indicate the practice of bone
boiling in order to extract the maximum amount
of nutrition from each available bone, implying
a very conservative cuisine practiced by non-
elite households. This pattern of conservative
foodways in non-elite spaces versus the capacity
in elite spaces to discard large, partially roasted
bones is evidence for patterned variation in
faunal use within the community.
Conclusions
Figure 6: Folded tupu.
The general settlement shift in LIP Caja-
Social and Economic Stratification at Yanaorco marca from lower elevations into often fortified
and defensible higher elevation villages is impor-
Architectural survey and variability in tant. This movement in population has two
excavated materials indicate the presence of at primary implications for life in the Cajamarca
least two social strata living at Yanaorco during region during the centuries after the collapse of
the LIP. There is clear evidence for two distinct Wari influence. The shift to higher elevations,
domestic zones at the site, marked by differences in the case of Yanaorco to an elevation of 3550
in room size, organizational complexity, and wall m.a.s.l., placed many communities at the bound-
construction quality. Elite domestic spaces ary between the Quechua and Suni ecozones.
include both interior, roofed rooms and un- This shift may indicate an increased focus on
roofed patios. Terraces are connected by well- pastoralism during the LIP. At the same time,
defined staircases, and walls are made of faced the settlement shift to more defensible loca-
and cut stone. Non-elite domestic spaces are tions, in many cases into substantially fortified
less well preserved due to the lower quality of communities, suggests the threat of armed
their walls and consequent poorer structural conflict between communities during the LIP.
preservation.
Excavation at Yanaorco indicates a dynamic
While all households at Yanaorco probably community characterized by social and eco-
engaged in agro-pastoral subsistence activities, nomic differentiation. Interaction between
including camelid herding, guinea pig raising, Cajamarca villages may have involved both
and corn and potato production, spatial analysis exchange of fine ceramics and other resources,
indicates variation in foodways and deposition and armed conflict during the LIP. This investi-
patterns between elite and non-elite contexts gation seems to show that interaction was
(Toohey 2009). Elite households like those primarily local and that little material exchange
represented at room 22A seem to have had
267 - Research Reports
vincal Inca Life (1963-1966). Under Murras of the excavation remain in the possession of
supervision was a small team of graduate stu- Shea. The bones and associated artifacts have
dents, young professional archaeologists, and been destroyed.3
Peace Corps volunteers including Mahlon [Lon]
Barash, the late Luis Barreda Murillo, the late Context
Gordon Hadden, the late Peter S. Jenson, the
late E. Craig Morris, Daniel Shea, and James To make the site of Hunuco Pampa more
Stanton, the latter a civil engineer. Manual comprehensible, in his view, to visitors, John
labor was performed by a team of approximately Murra removed large amounts of what he con-
thirty local farmers. During the course of this sidered to be overburden from the monumental
work a human skeleton was uncovered. This is or central portions of Hunuco Pampa, includ-
one of the few human remains ever found at ing the ushnu, the buildings incorporating the
Hunuco Pampa (c.f. Morris et al. 2011:31).1 It series of aligned portals in the eastern portion of
was excavated on 4-5 September, 1965 by Shea the site (the palace), the North and South ka-
and Barreda assisted by Barash2 and witnessed llankas bordering the eastern edge of the site,
by Morris and Hadden. the bath, and the unfinished temple (templete).
Although this was, in effect, an excavation,
The information available on this burial there is no indication found so far that the soils
consists of a series of eight black and white removed were screened for artifacts, that sec-
medium format photographs by Barash, fifteen tions were created or drawn, or that environ-
black and white 35 millimeter photographs by mental samples were taken. In addition, John
Barreda, and one medium format black and Murras team cleared vegetation from much of
white photo by Hadden taken during the exca- the fine masonry at Hunuco Pampa, reset
vation of the human remains, and seven photo- stones that had fallen from walls, and removed
graphs of the skull by Barreda, in addition to boulders thought to interfere with foot traffic
Sheas field notes. These are on file as part of (Barnes 2010).
the John Victor Murra Archive in the Junius
Bird Laboratory of South American Archaeol- In the course of such work, Murras team
ogy, Division of Anthropology, American Mu- encountered the grave of a young person. The
seum of Natural History in the City of New skeleton lay within the building of fine Cusco-
York. The photos are reproduced courtesy of the Inca style ashlar masonry that encompasses the
AMNH. The roll and photo numbers in the gateways known as Portals 5 and 6, the eastern-
captions of this report refer to Murras original most of the sets of aligned doorways leading to
photo inventory. In addition, color photographs the Hunuco Pampa ushnu plaza (Figure 1; see
Morris and Thompson 1985: figures 11, 13 for
1 the location of this building). The individual
In May, 1964 John Murra and Peace Corps volunteer
Peter S. Jenson encountered a looted stone burial
was in a simple grave cutting a canal or drain
chamber with a circular plan still containing human long that ran at or below the floor level of the build-
bones in the Quellaycancha portion of Hunuco Pampa ing. The channel was lined with flat stones
(Murra Archive, AMNH, Roll 15, photos 7, 8), another running along its floor and sides, and capping its
tomb (Roll 15, photo 9) and an apparently unlooted cairn top. The channel parallels the interior of the
tomb nearby (Roll 15, photo 10).
2
Field notes by Daniel Shea, John V. Murra Archive,
3
Junius Bird Laboratory of South American Archaeology, Letter from Craig Morris to Terence N. DAltroy,
Anthropology Division, American Museum of Natural November 17, 1976, Terence N. DAltroy file, Junius Bird
History, p. 7. Laboratory, American Museum of Natural History.
269 - Research Reports
south wall of the building, then makes a right- other a long gun or pistol.4 Because the possible
angle corner near Portal 5. From the informa- arrowheads do not resemble those used by
tion preserved and discovered so far, it is not Spanish bowmen, at least one of the shooters
possible to follow the entire course of this fea- would have been an Indian. However, the
ture. photographs are an insufficient basis for the
identification of these artifacts. Nevertheless,
However, given its position at Hunuco the stratigraphy and the position of the body,
Pampa, where it intersects a probable proces- coupled with the iron pin, indicate that the
sional line from the eastern quadrant of the site burial is post-Inca.
to the ushnu, Hunuco Pampas sacred center,
as well as the importance of liquid offerings to No statistical treatment can be given to a
ushnus (Meddens et al. 2010:186), it is tempting single skeleton found in isolation. However, the
to suppose that this canal has some connection burial context is important, although it cannot
to the sites ushnu. A similar arrangement can be be fully interpreted. Publication of individual
observed at Inca Pachacamac on the central burials may, gradually and collectively, lead to
Peruvian coast (Barnes, vidi, September 2011). insights that cannot be anticipated at present.
Without publication, the information inherent
Because the creation of the grave would in these burials remains unavailable to scholars
have interfered with the functioning of the for analysis and synthesis. For other examples of
canal, the grave must postdate the canals use. published single burials see Gaither et al. 2009;
An iron pin was found near the skeletons neck and Quilter et al., this volume, p. 106, figure 12.
(Shea, field notes, p. 8). This was not illustrated An important caveat must, however, be stated.
by Shea and does not appear in the photo- The authors had no access to the bones them-
graphs. Whether the pin was a shroud pin, or a selves, and worked only with photographs of the
tupu, an indigenous womans shawl fastener, its skeleton in situ and with views of each of the
material is a certain indicator that places the aspects of the skull. Furthermore, distortions
burial in colonial times, probably between 15 may have been introduced by camera angle and
August 1539 when the Spanish settlement at lens. We do not know what photographic equip-
Hunuco Pampa was officially founded and 2 ment was employed, except that it included a 35
February 1541 when it was abandoned. Shea millimeter camera, probably a rangefinder, and
also noted a few animal bones and Inca sherds a medium format camera, probably a twin lens
associated with the burial but pointed out that reflex. This situation poses obvious limits to
these could have been elements of the graves analysis.
fill and not contemporaneous with the burial
itself (Shea, field notes, p. 7). Some of the Description and analysis of the osteological and
photographs show three or four small, light- dental remains
colored conical objects and one small, light-
colored round object in the center of the chest. The skeleton is extended, lying on its back,
These are not mentioned by Shea in his field facing east, with the left forearm crossing the
notes. It is tempting to think of these objects as third lumbar vertebra and then crossing the
three bone arrow heads and one oxidized lead right arm with the left hand resting on the right
ball, and the proximate cause of death. If that is
so, they represent exceptionally precise shoot-
4
ing by two assailants, one using a bow and the For an unambiguous example of an Andean Indian killed
by a firearm in the immediate post-conquest period see
Murphy et al. 2010:641-642, figure 4.
ANDEAN PAST 10 (2012) - 270
Cranium
and the need to exert more force upward and effect on the roots at the coronas of the incisors.
laterally from the mandibular right molars. The suborbital (infraorbital) canal (artery) on
the right side shows signs of osteitis. The pala-
tine suture is irregular as expressed in the frontal
view and also visible in Figure 7, an inferior
view.
The person discussed in this report is most Morris, Craig, R. Alan Covey, and Pat Stein
likely part of the associations formed during 2011 The Hunuco Pampa Archaeological Project:
Volume 1: The Plaza and Palace Complex. Ameri-
Hunuco Pampas final occupation. The grave can Museum of Natural History Anthropological
seems to have blocked the free flow of water or Papers 96.
other liquids in an important part of the site, Morris, Craig and Donald Thompson
suggesting that maintaining this flow was no 1985 Hunuco Pampa: An Inca City and its Hinterland.
longer of importance. The prone position facing London: Thames and Hudson.
Murphy, Melissa S., Catherine Gaither, Elena Goycochea,
east, crossed arms, and iron pin all indicate a John W. Verano, and Guillermo Cock
post-Inca burial. However, the cadaver was not 2010 Violence and Weapon-related Trauma at Puru-
placed in a Christian cemetery. It is not clear chuco-Huaquerones, Peru. American Journal of
when such a cemetery was established at Physical Anthropology142:646-649.
Murra, John V. and G[ordon] J. Hadden
Hunuco Pampa. Documentary evidence only
1966 Apndice: Informe presentado al Patronato
suggests that a hacienda chapel already existed Nacional de Arqueologa sobre la labor de lim-
in 1786 when it appears on a map of Hunuco pieza y consolidacin de Hunuco Viejo (20 de
Pampa (Sobreviela and Sierra 1786). In any case, julio a 23 de noviembre 1965). Cuadernos de
the presence of a single human burial in this Investigacion, Antropologa 1:129-144 plus 12
unnumbered plates.
part of the site is intriguing.
Quilter, Jeffrey, Rgulo Franco J., Csar Glvez M.,
William Doonan, Catherine Gaither, Jaime Jimnez S.,
ACKNOWLEDGEMENTS
Hal Starratt, and Michele L. Koons
2011 The Well and the Huaca: Ceremony, Chro-
We thank Jack H. Prost and Ellen FitzSimmons Steinberg
nology, and Culture Change at Huaca Cao
for suggestions incorporated into this paper. Lon Barash
Viejo, Chicama Valley, Peru. Andean Past 10:
shared his recollections of excavating the skeleton dis-
99-129.
cussed in this report. Sumru Aricanli facilitated access to
Sobreviela, Padre Manuel and Lorenzo de la Sierra
the photographs and field notes.
1786 Plan del palacio destinado para bao de los yncas en
el partido de Huamalies con el nombre de Huanuco
REFERENCES CITED:
el viexo. British Library, Add. MS. 17671.
Barnes, Monica
2010 John Victor Murras Provincial Inca Life Project Mitomarca: A Possible Fortification in the
and American Anthropology. Antiquity Volume Upper Huallaga Basin
084, issue 326, on-line only, Project Gallery.
http://www.antiquity.ac.uk/projgall/barnes326/ Yuichi Matsumoto (Dumbarton Oaks, e-mail:
(consulted6 February 2012).
ymatsumoto1976@gmail.com), Jason S. Nes-
Gaither, Catherine, Jonathan Bethard, Jonathan Kent,
Vctor Vsquez Snchez, Teresa Rosales Tham, and bitt (Yale University, e-mail: jason.s.nesbitt@
Richard Busch gmail.com), and Denesy Palacios J. (Universi-
2009 Strange Harvest: A Discussion of Sacrifice and dad Nacional Hermilio Valdizn, Hunuco,
Missing Body Parts on the North Coast of Peru. Per, e-mail: denesy@terra.com).
Andean Past 9:177-194.
Meddens, Frank M., Colin McEwan, and Cirilo Vivanco
Pomacochan The Upper Huallaga Basin is on the eastern
2010 Inca Stone Ancestors in Context at a High- slopes of the Andes and is a mountainous region
Altitude Usnu Platform. Latin American Antiquity cut by the Huallaga River (Figure 1). It is an
21(2):173-194. important ecological interface and a corridor
Morris, Craig
1979 The Spanish Occupation of an Inca Administra-
between the central highlands and Amazonia.
tive City. Actes du XLIIe Congrs International des
Amricanistes: Congrs du Centenaire, Paris, 2-9 During the 1960s, the University of Tokyo
Septembre 1976. IX-B:209-219, Paris. carried out intensive excavations at the sites of
Kotosh and Shillacoto. Although this pioneer-
275 - Research Reports
ing research established a fine-grained relative fortification dating primarily to the Late Inter-
chronology of the regions Initial Period and mediate Period.
Early Horizon, little archaeological research has
occurred since then because of widespread Site Description
violence during the 1970s and 1980s. As a
result, there are considerable gaps in our under- Mitomarca is in the district of Churubamba,
standing of the culture history of the region, close to the eastern end of the Upper Huallaga
especially for the long span of time between the Basin. The site stands on a natural hill along the
first century A.D. and the years immediately northern bank of the Huallaga River at an
prior to the incorporation of the region into the elevation of 2300 m.a.s.l. The Early Horizon
Inca empire in the late fifteenth century A.D. ceremonial center of Sajara-patac (Inokuchi et
Therefore, for the purpose of amplifying archae- al. 2003; Matsumoto and Tsurumi in press) is
ological understanding of the prehistory of the approximately 600 meters to the southeast, on
Upper Huallaga Basin, a settlement pattern the lower part of the same hill (Figure 1).
study was conducted in 2001 by Kinya Inokuchi
(University of Saitama, Japan), Yoshio Onuki
(University of Tokyo), Eisei Tsurumi (Univer-
sity of Tokyo), Yuichi Matsumoto, and Alvaro
Ruiz.
appeared as a response to the political instabili- between 2400 and 2900 m.a.s.l., while the Late
ties of this period. However, the general settle- Horizon sites were frequently in the elevations
ment pattern data clearly contradict this view. between 3000-4000 m.a.s.l. (Grosboll 1988:75).
During the Kotosh Sajara-patac Period, the As Daniel Morales (1984) points out regarding
number of sites radically increased and new sites the Late Intermediate Period and Late Horizon
were distributed throughout the yunga ecozone sites in the Upper Huallaga Basin, it is very
in the Upper Huallaga Basin. Although new probable that the sites that Grosboll registered
sites were constructed at higher elevations near as Late Horizon have a Late Intermediate Period
the border of the yunga and quechua ecozones, component as well. Therefore, the shift of
bottomland sites also increased in frequency, but settlements to higher elevations may have
some of them were first occupied in the middle- occurred during the Late Intermediate Period,
Early Horizon and continued to be occupied rather than during the Late Horizon. If this
during the Kotosh Sajara-patac Period. Further- change in settlement pattern reflected the
more, most sites dating to this time did not political disintegration of ethnic groups, it might
utilize defensive architectural features. There- be possible to interpret Mitomarca as a fort.
fore, in the Upper Huallaga Basin, it may be Alexis Mantha (2009) identified multiple site
dangerous to assume the existence of increased categories based on survey of the Rapayn
conflict with neighboring regions during the late Valley approximately 80 kilometers to the north
Early Horizon, and, thus, it is difficult to inter- of Mitomarca. He found that some of the high
pret Mitomarca as a fortified site of the late elevation sites of the Late Intermediate Period
Early Horizon or beginning of the Early Interme- display both defensive and ceremonial charac-
diate Period. teristics, and named sites of this type as Defen-
sive, Ceremonial, and Communication settle-
If we accept that the surface architecture of ment (DCC) (ibid.:168-169). If, in Mitomarca,
Mitomarca was constructed some time after the the square plaza had a ceremonial function, and
Early Intermediate Period, probably during the the perimeter wall functioned as a defensive
Late Intermediate Period, it seems more proba- wall, this might be an appropriate term to repre-
ble that Mitomarca was a fort, because of the sent Mitomarca.
known tensions among different ethnic groups
of the Huallaga region at that time. Although Conclusion
the cultural developments after the Kotosh Hi-
gueras Period, that is, from the Middle Horizon Our preliminary observation of Mitomarca
to the Late Horizon, are not well known in the generated more questions than we had at the
Upper Huallaga Basin, the survey data of Sue beginning of our field-work, rather than solving
Grosboll (1988) provide useful insights perti- issues of the later prehistory of the Upper
nent to this issue. Huallaga Basin. Only further study of the site
will make it possible to evaluate the hypotheses
Grosboll surveyed the area where the presented in this report. While the regional
Chupachu and Quero people lived, and she cultural context from the Preceramic Period to
located many of the Late Horizon sites described the Early Horizon is well studied, the period
in the visita document of 1549 and 1562 written after the Early Intermediate Period is only
by Iigo Ortiz de Zuiga (Ortiz de Zuiga (1967, poorly known. In particular, the Middle Horizon
1972 [1562]). According to Grosboll, the sites and the Late Intermediate Period of the Upper
that belong to the Middle Period (500-1300 Huallaga lack basic archaeological data. Mito-
A.D.) in her terminology, tend to be found in marca is a multi-component site that probably
279 - Research Reports
contains occupation during these periods. Thus, Luis Millones and Hiroyasu Tomoeda, pp. 211-
the site will be an appropriate research site to fill 228. Osaka, Japan: National Museum of Ethnol-
ogy.
the gap in archaeological data and extend the Ortiz de Ziga, Iigo
regional chronology to the Late Horizon. 1967 Visita de la provincia de Len de Hunuco en 1562,
volume 1, Documentos para la Historia y Etno-
REFERENCES CITED loga de Hunuco y la Selva Central, Universi-
dad Nacional Hermilio Valdizn, Hunuco,
Bonnier, Elizabeth Per.
1997 Morfologa del espacio aldeano y su expresin 1972 Visita de la Provincia de Leon de Hunuco en 1562,
cultural en los Andes centrales. In Archaeolgica Tomo 2, Documentos para la Historia y Etnolo-
peruana 2: Arquitectura y civilizacin en los Andes ga de Hunuco y la Selva Central, Universidad
prehispnicos, edited by Elizabeth Bonnier and Nacional Hermilio Valdizn.
Henning Bischof. Reiss-Museum, Mannheim, Salcedo Camacho, Luis E., Enrique Molina, Luis E.
Germany: Sociedad Arqueolgica Peruano- Cceres, Patricia Habeteler, Vctor Curay, and Juan Rofes
Alemana. 2000 Trabajos de puesta en valor en el complejo
Burger, Richard L. arqueolgico Garu. Revista Desafos 1:60-74.
1992 Chavin and the Origins of Andean Civilization. Snchez Murrugarra, Ricardo and Denesy Palacios J.
New York:Thames and Hudson. 1988 Garu: Un complejo urbano preincaico. In
Girault, Louis Arquitectura e arqueologa: Pasado y futuro de la
1981 Fouilles sur le site de Piruru en 1968 et 1970. construccin en el Per, edited by Vctor Rangel
Bulletn de lInstitut Franais dtudes Andines Flores, pp.153-162. Chiclayo, Per: Universidad
10(1-2):101-112. de Chiclayo, Museo Brning, and CONCYTEC.
Grosboll, Sue Thompson, Donald E.
1988 An Archaeological Approach to Demography of 1967 Investigaciones arqueolgicas en las aldeas chu-
Prehispanic Andean Communities. Ph.D. disserta- pachu de Ichu y Auquimarca. In Visita de la
tion, University of Wisconsin, Madison. provincia de Len de Hunuco en 1562. Tomo I:
Inokuchi, Kinya, Yosio Ounki, Eisei Turumi, Yuichi Mat- visita de las cuatro waranqa de los chupachu, edited
sumoto, and Nelly Martell Castillo. by John Victor Murra; Ortiz de Ziga, Volume
2003 Excavations at the Site of Sajara-patac in Peru. I., pp. 357-367. Universidad Nacional Hermilio
Amrica Antigua 6:3552 (written in Japanese Valdizn, Facultad de Letras y Educacin,
and published by the Sociedad Japonesa de Estu- Hunuco.
dios sobre la Amrica Antigua).
Isbell, William H. An Initial Period Domestic Occupation at
1974 Ecologa de la expansin de los quechua-hablan-
tes. Revista del Museo Nacional 40:139-155.
Huaca Cortada, Caballo Muerto Complex
Mantha, Alexis
2009 Territoriality, Social Boundaries and Ancestor Jason S. Nesbitt (Yale University, e-mail: jason.
Veneration in the Central Andes of Peru. Journal s.nesbitt@gmail.com) reports on his recent
of Anthropological Archaeology 28(2):158-176. excavations at the Caballo Muerto site in the
Matsumoto, Yuichi and Eisei Tsurumi
in press Archaeological Investigations at Sajara-patac in
lower Moche Valley.
the Upper Huallaga Basin. awpa Pacha.
Morales, Daniel Caballo Muerto is well known in the archae-
1984 Algunos sitios arqueolgicos del Reino de Gua- ological literature as a large multi-mound Cupis-
nco. Boletn de Lima 6(33):83-95. nique center in the lower Moche Valley. Pio-
Onuki, Yoshio
1972 Pottery and Clay Artifacts. In Andes 4: Excava- neering research at Caballo Muerto in the late
tions at Kotosh, Peru, 1963 and 1966, edited by 1960s and 1970s established that each of the
Seiichi Izumi and Kazuo Terada, pp. 177248. mounds dated to the Initial Period or Early
Tokyo: University of Tokyo Press. Horizon (Chauchat et al. 2006; Pozorski 1983;
1982 Una perspectiva prehistrica de la utilizacin
Watanabe 1976). Between 2007 and 2008, I
ambiental en la sierra nor-central de los Andes
centrales. In El hombre y su ambiente en los Andes initiated new excavations at Caballo Muerto
centrales. Senri Ethnological Studies 10, edited by (Nesbitt et al. 2010). The goal of the excava-
ANDEAN PAST 10 (2012) - 280
rocks and a small quantity of Chim style pot- layers. I hypothesize that these layers originated
tery. Chim ceramics and stone wall founda- from the flooding of a branch of the Ro Seco
tions were common in other test units around during an El Nio event that covered the un-
Huaca Cortada, indicating the presence of a derlying Initial Period domestic occupation
Chim occupation postdating the use of the described below.
monuments at Caballo Muerto.
Layer 7 consisted of a brown (7.5YR4/4)
Directly underlying this first layer were light silty soil with several broken conical adobe
brown (7.5YR6/3) clayey soils with large cobbles fragments, some of which also can be seen in the
and small boulders that contained Initial Period southern profile. Mixed within this layer were
pottery fragments (Layer 2). Associated with deposits of light-colored sand. Just to the east of
these materials were small pieces of charcoal, the adobe fragments was a small, circular carbon
eroded conical adobe fragments, and numerous concentration. The artifact assemblage con-
burned, orange clay fragments with cane and sisted of a small number of non-diagnostic body
cordage impressions. In some cases these clay sherds with Initial Period pastes. As will be
pieces had traces of white paint on their exteri- discussed below, this layer was not continuous,
ors. However, no occupation surface was associ- but instead seemed to be a layer that pre-dated
ated with these materials. a flood of a branch of the Ro Seco.
Beginning at a depth of 1.46 meters below Intact Initial Period occupation surfaces
the surface was a brown (7.5YR4/4) silty soil (Layers 8-11) were uncovered beginning at a
(Layer 3) that had fewer rocks, but an increas- depth of approximately 2.20 meters below the
ing abundance of broken conical adobe frag- ground surface. These layers consisted of a series
ments, some of which can be seen in the of superimposed domestic occupation layers
southern profile. Within this layer were also thin (discussed below). Each layer is described sepa-
lenses of light-colored sand and a small circular rately below.
carbon concentration. There were no diagnostic
pottery rims, but the pastes of the body sherds Layer 8 was a surface or floor made up of a
were consistent with Cortijo Phase ceramics. brown (7.5YR4/3) compacted earth mixed in
Interestingly, this layer was not continuous, but places with fine-grained yellow (10YR7/6) sand.
instead represents a cultural layer that post- The floor also had concentrations and flecks of
dated a probable flood of the Ro Seco, now a charcoal and ash that began in the western
dry quebrada that runs just to the west of Huaca profile of Unit 2 and continued to the east. The
Cortada. profile exhibited a discrete lens of charcoal and
red (2.5YR5/6) earth that formed a probable
This flood deposit was a thick series of layers hearth measuring approximately eighty centime-
that sloped to the west (Layers 4-6). These ters in length. The presence of a hearth in the
layers together range from fifty-six to ninety southern profile indicates that the occupational
centimeters in thickness and were composed of surface continues to the south.
a brownish yellow (10YR6/6) coarse sand con-
taining small pebbles and rounded cobbles. The Layer 8 floor yielded some refuse,
There were no artifactual materials recovered including mammal, fish, and charred shellfish
from these layers. In terms of color, composi- remains. Burned, orange clay fragments contin-
tion, and thickness, these layers were quite ued to be abundant and a single adobe fragment
distinct from the overlying and underlying was found resting on the surface. Pottery was
ANDEAN PAST 10 (2012) - 282
relatively abundant in this layer and I found As with other layers, there were abundant
numerous blackened, neckless olla rim frag- burned clay quincha (wattle-and-daub) frag-
ments resting horizontally on the surface. In ments, with cane impressions. Shellfish remains
addition, I found a small, flat, stone circular were also present in this layer.
ornament with two holes in the center of the
artifact (Figure 3l).
Layer 12 was a sterile deposit consisting of al. 2010). The small assemblage of pottery
loosely consolidated cobbles within a coarse associated with this phase demonstrates close
sand matrix. No artifacts were recovered from parallels with Cortijo Phase ceramics recovered
this layer. Layer 12 was first found at a depth of from Layers 8-11 in Unit 2. Cortijo Phase pot-
2.7 meters below the present ground surface and tery in Unit 2 consists of unslipped wares with
the excavations were terminated at just over oxidized pastes and coarse inclusions. The most
three meters below the surface. common form from the Cortijo Phase is a neck-
less olla. Some examples have simple incised or
I interpret Layers 8 through 11 as represent- punctate decorations (Figure 3 j-k). The most
ing a probable domestic occupation associated diagnostic vessel form from this time, however,
with Huaca Cortada. The evidence that sup- is a type of long-necked bottle, with small punc-
ports this conclusion includes the presence of tations ringing the area where the base of the
superimposed hearths, cooking vessels, and neck meets the bottle chamber. Analysis of
refuse such as animal remains. The abundance Cortijo Phase pottery at Caballo Muerto shows
of adobe fragments and burned clay fragments broad similarities to a variety of early and middle
with cane impressions and, more rarely, pieces Initial Period ceramic traditions including the
of mud plaster, throughout most of the layers in Montegrande (Ulbert 1994) and early and
Unit 2 suggest that the domestic unit was a middle Guaape (Strong and Evans 1952) styles
wattle-and-daub building. from the nearby Jequetepeque and Vir Valleys.
At present, based on cross dating with Huaca
The superposition of hearths and occupation Cortada, I tentatively date the Cortijo Phase to
surfaces characterizing the domestic occupations approximately 1500-1200 cal BC. A radiocarbon
in Unit 2 appears to represent the accumulation dating project is underway to create a more
of households, perhaps over several generations. detailed absolute chronology, which will be
As Hirth (1993:24) observed, these types of available in the near future. More relevant to
accumulated deposits create problems for inves- this report, the ceramic data clearly show that
tigating household composition synchronically. the occupation described for Unit 2 is coeval
To get around these problems, Hirth uses the with the earliest known phases of use of Huaca
term household series to define the remains Cortada.
that amass from a succession of households
occupying the same physical space (or home- Subsistence economy
stead in Hirths [ibid.: 24] terminology). Layers
8 to 11 of Unit 2 can best be interpreted as a Careful recovery techniques, including fine-
household series accumulated during domestic grained screening and flotation of Layers 8
occupations, probably over a relatively short through 11, allowed the collection of floral and
period of time during the Cortijo Phase. faunal materials that shed some light on the
subsistence economy of the Cortijo Phase occu-
Chronology pation of Caballo Muerto. Faunal materials from
Layers 8 and 9 include the remains of shellfish,
No absolute dates currently exist for the including clams (Donax sp.) and mussels (Choro-
occupation of Unit 2. However, a pilot radiocar- mytilus chorus). In addition, the zooarchaeo-
bon dating project focusing on the earliest logical analysis found one bone from a gull
phases of monumental architecture at Huaca (Larus sp.) as well one drum otolith. A single
Cortada indicates that its first architectural thoracic vertebra from a deer (Odocoileus virgini-
phase dates to around 1500 cal BC (Nesbitt et anus) was found in Layer 9. Flotation of the
ANDEAN PAST 10 (2012) - 284
In 2003-2004, twenty-three weeks of field- tion of marine resources, and trade. Popular
work at Huambacho yielded important data on plant domesticates included varieties of tubers,
Early Horizon societies in the lower Nepea fruits, cucurbits, and legumes, as well as maize,
Valley, Coastal Ancash, Peru. Chicoine directed introduced in the region at the beginning of the
the field-work as part of the Proyecto Arqueol- Early Horizon. In a continuation of Initial
gico Huambacho. Rojas carried out the first- Period (1800-900 B.C.E.) socioeconomic pro-
hand identification and classification of the cesses, irrigation farming and inland settlements
molluscan remains. Analyses of architecture, gradually gained more importance than settle-
material culture, and radiometric measurements ments dependent upon coastal adaptations.
indicate that between 800 and 200 cal B.C.E. Nevertheless, studies of marine resources are
(2F) Huambacho represented a small elite scarce, and our knowledge of coastal economies
center with strong public components (Chicoine in Ancash is still preliminary (Pozorski and
2006). Settlement patterns and more recent Pozorski 2003). The Huambacho molluscan
field-work elsewhere in Nepea suggest that the data allow a preliminary assessment of Early
Huambacho community was related to the Horizon marine economies and their associated
larger complex of Cayln located further inland, trade networks, as well as local environmental
on the north bank of the river, some fifteen conditions.
kilometers from the littoral (Chicoine and
Ikehara 2010). Huambacho is on the south bank of the
lower Nepea Valley, some eight kilometers
Significant numbers of artifacts and ecofacts from the littoral. It consists of two distinct
were recovered during field-work, including complexes of stone-walled enclosures, patios,
mollusk remains and fish bones. This report and monumental plazas originally covering an
focuses on the place of the molluscan assem- area of more than twelve hectares. Our excava-
blage within the faunal remains and implications tions have yielded a variety of cultural remains,
for understanding past diets, economies, social including more than 4,000 diagnostic pottery
organizations, and aquatic ecologies. In particu- sherds, and faunal and botanical remains, mostly
lar, taxonomic analyses reveal patterns of ma- from secondary refuse deposits. Field-work
rine exploitation and bioindicators. results suggest that an elite group permanently
Paleoenvironmental reconstructions are, in resided onsite, while most community members
general, consistent with the cold waters typical likely lived in peripheral areas.
of the Humboldt Current and little evidence
supports the occurrence of major climatic dis- The analysis of vertebrates allowed the
turbances, including those related to the El identification of a minimum number of 1,300
Nio-Southern Oscillation phenomenon or bones (NISP), including both wild and domesti-
ENSO. cated animals (Table 1). Fish and birds include
many species of marine and riverine animals.
Background and research at Huambacho Rodents include guinea pigs (Cavia porcellus)
and also rats and mice. Bone fragments of dogs
The Early Horizon (900-200 B.C.E.) is (Canis familiaris) and sea lions (Otaria flavescens)
usually seen as a time of increased interregional were also identified.
interactions tied to the Chavn phenomenon in
some regions of what is now Peru. In coastal Animal bones are dominated by camelids (c.
Ancash, economies were mainly based on irriga- 40 percent). Further analyses will be needed to
tion agriculture, animal husbandry, the exploita- assert the relative importance of each category
ANDEAN PAST 10 (2012) - 286
of the vertebrates (e.g. meat weight, quantity of came from secondary trash deposits, mainly
animal protein, identity of camelids). The high construction fill.
frequency of camelids contrasts with Initial
Period sites in coastal Ancash where these The analysis of mollusk remains was facili-
animals are usually rare or absent. As pack tated by their good preservation. Rojas was able
animals, camelids would have been useful in the to identify genus and species in more than
Early Horizon context of intensified interre- ninety-nine percent of the cases. She used
gional interactions. reference collections and the published litera-
ture (lamo and Valdivieso 1987; Ramrez et al.
At Huambacho, the virtual absence of meat 2003). The quantification procedures favored
cutting tools, such as obsidian blades, and the Minimum Number of Individuals (MNI) over
lack of intensive fracturing or burning of camelid the Number of Individual Specimens (NISP)
bones contrasts with typical patterns of meat and weight. For bivalves, MNI were evaluated
consumption. At the same time, members of the by separating and counting left and right hinges,
project collected large amounts of shell remains, and using the higher of the two counts for each
highlighting the importance of seafood. The taxon. For gastropods, MNI were defined by the
evidence from Huambacho is consistent with number of univalve apices.
comparative data from Casma, the neighboring
valley to the south, where the local subsistence The analysis identified 11,274 MNI and a
economy was mainly based on maize agriculture, total of 53 molluscan taxa pertaining to the
as well as the exploitation of marine resources classes Monoplacophora (MNI=1, 0.01 percent),
and, with less intensity, animal husbandry. Polyplacophora (MNI=528, 4.7 percent) Gastro-
poda (MNI=1,138, 10.1 percent), and Pele-
Total cypoda (bivalves); (MNI=9,551, 84.7 percent).
Category NISP percent
Camelid 517 39.77 Rojas was unable to identify 55 fragments (0.49
Rodent 464 35.69
Fish 144 11.08
percent). The high number of MNI allows for a
Bird 55 4.23 confident assessment of the taxonomic composi-
Other/Unknown 120 9.23
TOTAL 1,300 100.00
tion of the assemblage (see Evans 1972:12). The
assemblage is taxonomically rich, but not very
Table 1: Number of individual specimens (NISP) of diverse, with three species, Perumytilus purpura-
animal bones, counts, and percentages. tus, Semimytilus algosus, and Donax obesulus
accounting for eighty percent of the corpus,
Shell remains which hints at a narrow exploitation spectrum.
mellaris, MNI=1, 0.01 percent). Several exam- Los Chims, where prehistoric shell middens
ples of perforated Prunum curtum (MNI=15, 0.1 have been reported (Thompson 1962:206).
percent) have been noted. Also, a total of 115 Huambacho is only a two hours walk from Los
specimens of the thorny oyster Spondylus sp. Chims. In contrast, the northern portion of the
were collected. Most specimens came from Nepea coastline, as well as the southern por-
surficial and/or disturbed contexts, often in tion of the Baha de Samanco, is dominated by
proximity to looted burials associated with post- sandy beaches of the Nepea Delta.
Early Horizon occupations. Hence, the Spon-
dylus data are difficult to interpret and have not From a comparative perspective, the recent
been included in the taxonomic quantification recovery of more than 20,000 MNI of shell
provided in Table 2. remains during the 2009 excavations at the
Early Horizon site of Cayln brings insights into
Shellfish exploitation in coastal Nepea patterns of marine exploitation in the lower
Nepea Valley (see Chicoine and Ikehara
Taking into account the position of the site 2010). At Cayln, preliminary analyses indicate
eight kilometers inland and the virtual absence the predominance of the sand-dwelling Donax
of fishing implements at the site, it is likely that sp. (c. seventy percent), while rock-dwelling
the Huambacho mollusks were acquired mussels are represented in lesser proportions (c.
through interactions with groups that inhabited twenty-five percent). Both the Huambacho and
the littoral. In the Moche Valley, coastal La Cayln assemblages are dominated by the same
Libertad, research suggests that during the three species, but the diametrically inversed
Initial Period, the coastal, marine-oriented emphasis of sandy vs. rocky substrates is striking.
community of Gramalote was tied to the inland This observation is consistent with the location
complex of Caballo Muerto (Pozorski 1979: of Cayln, three-and-a-half hours walk from the
173). Caballo Muerto animal protein was sup- Nepea delta and the Baha de Samanco, where
plied by marine and inland resources while sandy environments are ample today.
increasing importance was given to agriculture,
including maize cultivation. Gramalotes subsis- Available data from Early Horizon sites
tence economy was also primarily marine-ori- suggest that marine resources in Nepea trav-
ented and supplemented by inland animal elled through multiple networks. Indeed, prelim-
protein, in particular deer and llama. Further inary comparisons argue against a single, cen-
south, Haas and Creamer (2006) have recently tralized redistributive or exchange system.
suggested the existence of similar coast-inland Rather, several different, probably independent,
exchange networks during the Preceramic coastal providers and satellites have to be hy-
Period in the Norte Chico region. pothesized from the Baha de Tortugas to the
Baha de Samanco. Archaeological surveys will
Based on geographical proximity, the Huam- be necessary to find and document these com-
bacho mollusks were probably collected from munities.
the southern portion of the Nepea coastline
and/or the northern portion of the Baha de Finally, preliminary comparisons point
Tortugas. Exploitation mainly focused on rocky towards multiple exploitation strategies, perhaps
cliffs where species of mussels and, to a lesser both specialized and unspecialized. As noted by
extent, limpets and gastropods were available. Rosell et al. (2001:77), the sand-dwelling clam
These species are still abundant, especially in Donax sp. is easily collectible by hand in large
the coastal area near the modern settlement of numbers. On the basis of their observations of
289 - Research Reports
modern collectors in Huanchaco, Rosell et al. longer to recolonize its habitat after a major
(ibid.) suggest that individual fisherfolks can ENSO event. In contrast, rock-perching mussels
collect up to thirty kilograms of clams in three like Perumytilus purpuratus and Semimytilus
hours of work using rudimentary rakes. Con- algosus are found six months or less following an
versely, the authors suggest that collecting rock- ENSO event (Rollins et al. 1990:471; Sandweiss
dwelling species is a more perilous enterprise et al. 2001:604).
requiring more specialized practices. From that
standpoint, Rosell et al. (ibid.) interpret the In order to reconstruct environmental
predominance of Donax sp. at the Early Interme- conditions, archaeomalacologists need to con-
diate Period site of Huacas de Moche (c. 80 sider the impact of cultural preferences, pre-
percent of the faunal assemblage) as an indica- scriptions, and proscriptions on the exploitation
tor of the unspecialized exploitation of beach of marine taxa. In other words, the archaeologi-
environments. In Nepea, more research will be cal absence of a species cannot be straightfor-
needed, but the available data suggest diverse wardly equated with its unavailability. Con-
strategies to exploit molluscan resources. versely, mollusks can be transported great
distances and their presence at a site does not
Molluscan bioindicators and paleoenvironment automatically represent local conditions (Rollins
et al. 1990:470-471). Finally, some species
Marine mollusks are adapted to specific preserve better and have a higher probability for
ecological parameters (e.g. water temperatures, archaeological recovery.
salinity, oxygen, pH level, availability of ade-
quate substrate). As such, they represent poten- Keeping these considerations in mind, the
tial bioindicators useful for the reconstruction of Huambacho sample is dominated by cold water
local environmental conditions. In coastal Peru, taxa typical of the Humboldt Current. With the
molluscan data are particularly helpful in dis- exception of two species (Ancistromesus mexica-
cerning climate fluctuations related to the nus, Argopecten circularis), the presence of
occurrence of major ENSO events (Daz and tropical species is sparse. Indeed, less than one
Ortlieb 1993; Rollins et al. 1990; Sandweiss percent of the total MNI is represented by
1996). The most salient marine manifestations species better adapted to tropical water. Also,
of ENSO are related to an increase in water some species were clearly imported for their
temperature and changes in salinity, proportions value as adornments and perhaps currencies.
of dissolved oxygen, and tidal patterns (Daz Spondylus is a good example. With the excep-
and Ortlieb 1993:161,164). These changes tion of a complete shell, Spondylus specimens at
typically cause indigenous mollusk species to Huambacho are all worked into beads and
migrate or disappear. Only a few taxa are found blanks. The same rationale applies to Prunum
in greater abundance during ENSO events curtum specimens found with drilled holes to
including Argopecten purpuratus, and Thais hang them. Overall, as viewed through the
chocolata (ibid.: 165, Table 1). mollusk evidence, the Huambacho data does
not lend weight to the occurrence of a major
In addition to the well-known tropicalization tropicalization of local oceanic conditions typi-
of the normally cold Humboldt environments, cal of ENSO.
ENSO events impact the long-term regenera-
tion of certain species. Researchers have sug-
gested that Mesodesma donacium, a sand-dwell-
ing species of clam, may take up to a decade or
ANDEAN PAST 10 (2012) - 290
CHILE
bones, and marine mammal leather are used Clusters of structures were organized around
(Muoz 2002:489). central patios (Figure 2).
The earliest known occupations with archi- Site No. Built-up Settled Density - % ground
Enclo- area (m2) area, residences/ occupied
tecture on the coast of Tarapac are docu- sures hectares hectare
mented at the Caramucho-3 site, where an Pisagua N 25 432.21 0.090 277.78 48.02
extensive shell midden, dating to 4030 B.C. Los Verdes 1 12.25 N/R N/R N/R
(Morgas 1995) and incorporating fish hooks, Sarmenia 8 66.68 0.294 27.21 2.26
cactus remains, and shells, and with collapsed Pabelln de Pica 1 87.55 N/R N/R N/R
Chomache 5 63.07 0.021 238.10 30.03
stone structures, has been recorded (Olmos and
Punta Blanca 8 78.04 0.125 64.00 6.21
Sanhueza 1984:146). According to Moragas
Caleta Hueln Alto 159 1299.82 1.354 117.43 0.95
(1995:77), settlements on the interfluvial desert Caleta Hueln 42 30 70.79 0.328 91.46 4.29
coast between Iquique and the Loa Valley occur N/R - Not Recorded
preferentially in certain sectors near water holes
or where the concentration of mists permits Table 1: Main architectural traits of coastal settlements.
human life. The time of occupation has been
determined by radiocarbon and thermo-
luminescence dating (Urbina et al. 2011).
at Patache G) with low walls and interior spaces Period settlements were recorded out of a total
of less than five square meters. At Patache G number of 1,307 sites.
the body of a young adult, covered with a tex-
tile, was buried in an extended position, directly The first architectural form is an expedient
below the living surface, without a sealed floor. architecture recorded at the Punta Blanca
The textile yielded a radiocarbon date of cal. site(390-620 cal A.D.). There, sub-rectangular
195 B.C. (Urbina et al 2011). The pattern of structures of restricted dimensions and simple,
circular structures beneath shell middens, and adjacent stone walls are irregularly dispersed at
near rock outcrops, is represented farther to the rock outcrops or beneath small rock shelters
north, in the southern part of the Sarmenia site. (Figure 3). They lack prepared mortar and
There, two contiguous structures with interior pillars with buried foundations and can be
spaces of 4.2 cubic meters and 4.6 cubic meters, defined as a type of light architecture.
respectively, were recorded.
The second architectural form was recorded
Consequently, it is established that the at the mouth of the Loa River at the boundary
coastal populations of Tarapac, from the end of of the second marine terrace and the coastal
the Archaic and the beginning of the Formative mountain range. The Caleta Hueln Alto site,
Period, were related to the Chinchorro tradition characterized by ceramic types dating between
found farther north (Arriaza and Standen 2002) A.D. 200 and 900 (Uribe 2009), presents dis-
as well as to other, more southern, architectural persed stone structures of expedient construc-
manifestations (Schaedel 1957). tion on a terraced hillside extending more than
a kilometer. There, small parapets built with low
stone walls were associated with the sites nu-
merous footpaths. Notable is the absence of
food remains and the abundance of lithic debris,
copper minerals, and granite hammer stones,
occurring in walled work areas, or in open
spaces with otherwise clear floors. Included
among the habitation units are ovoid structures
with interior spaces of less than five square
meters, and others with sub-rectangular plans
up to fifteen square meters. Caleta Hueln Alto
combines construction elements that resemble
those at the southern sector of Sarmenia, and at
Punta Blanca and the Chomache 1 sites which
exhibit widely-dispersed occupations.
Figure 3. Plan of the Punta Blanca site.
The Chomache 1 site represents a more
Formative Period architecture agglutinated form of architecture in a later
chronological context (cal. A.D. 650-1290). A
Archaeological survey along the Iquique-Loa new pattern can be observed. There are five
River coastal transect identified a significant structures with rectangular plans, and with
complex of Formative Period occupations that architectural elements absent in the Archaic
presented elements of both architectural conti- Period, such as terracing of the site, and internal
nuity and innovation. Ninety-one Formative
ANDEAN PAST 10 (2012) - 294
division of spaces by means of partitions or to a wall in Pisagua Ns Edificio 19, and that of
substructures (Figure 4). a marine mammal cranium in the northeastern
corner of Structure 1 at Chomache 1. Both sites
achieve similar levels of agglutination (Table 1),
and both are adjacent to a rocky shore lacking
nearby fresh water. Because of this it was neces-
sary to develop techniques for condensing the
coastal fog (camanchaca).
Located next to a large rock outcrop, is a Figure 5. Plan of the Pisagua N site.
structure (Edificio 5) with slightly curved cor-
ners and pillars with stone foundations, all traits The architecture of Punta Blanca, Cho-
present in earlier buildings. A more compact mache 1 and Caleta Hueln Alto reveals a form
residential unit with a greater investment of characteristic of the desert coast south of Iqui-
labor is evident. que and Tarapac, where domestic environ-
ments are now separate from funerary spaces
The architecture of Chomache 1 represents (Llagostera 1989). On the other hand, the re-
a transformation in the way the coastal area of occupation of Chomache 1 and Pisagua N
Tarapac was occupied at this time. Similar sites indicate that a pattern with an agglutinative
are Caleta Hueln Alto and Punta Blanca, as tendency was current until the Late Intermedi-
well as a more northern site, Pisagua N. This ate Period (A.D. 900-1450). This architecture
latter site was occupied from the Late Formative had greater functional variability than that of
Period until the Late Intermediate Period or earlier sites, to judge from the sizes and forms of
from A.D. 290 to 1470 (Urbina et al. 2011). The its structures.
latest ground plan consisted of 25 agglutinated
stone structures with square, rectangular, and Late coastal dwellings
trapezoidal plans. These structures were con-
structed over time on terraced surfaces (Figure Among the Late Intermediate Period sites
5). Excavations revealed rows of reeds that were including Chomache 1 and Pisagua N are other
incorporated into the stone walls, to construct forms of residences. One is represented by
faades of organic material (quincha). Structure 1 at the Pabelln de Pica-1 site (Fig-
ure 6), dated between cal A.D.1280 and 1430.
At Pisagua N, as at Chomache 1, domestic This is a large sub-circular area (85 square
spaces were made sacred by means of foundation meters) made of curved, double-faced stone and
offerings such as the burial of a dogs head next mud walls more than 50 centimeters thick and
295 - Research Reports
laid in courses, features that resemble those of of Pisagua B (Figure 8) with dates ranging from
the Caserones-1 site. The unusual height of A.D. 1496 to 1556 (Urbina et al, 2011).
Structure 1s walls, greater than 1.5 meters, is
remarkable, as is the high proportion of mortar
in its facing. In accordance with its formal
characteristics, position, and visibility, it may
have been a communal and public space near
residential spaces.
Her face has carved lines descending from her Below this level, a thin layer of burnt clay
eyes which could represent tears, face paint, covered the entire structure. After its removal I
tattoos, or scarification (Figure 4, top). The detected a more compact stratum, Level III,
lower portion of this artifact has a flake scar, where traces of another burial were present
suggesting that it was intentionally broken or (Figure 6). It was composed of an adults
killed before deposition. osteological remains, in a very bad state of
preservation, associated with three coarse fabric
vessels. The human remains were fragments of
a skull and a jawbone in the western margin of
the base of the cist, and hundreds of unidenti-
fied little pieces of bone. The skeletal remains
were accompanied by a cup (made of ordinary
red clay, with a rough surface finish, a lip handle
vertically attached, and a modeled anthropo-
morphic face decoration applied on the side
opposite the handle; Figure 6 top), a jar (tech-
nologically similar, with a uniform surface finish,
no decoration, and a thick soot layer on the
outer surface), and numerous fragments of
vessels with similar characteristics. None of the
pottery presents complex decoration, nor corre-
sponds to what is known as fine craftsmanship
for this period. Charred wood was recovered
from this level and dated to 179937 BP, cal.
Figure 5: Schematic plan of Level II burial. AD 130-260, the earliest date for LB1-U14.
Next to the Level II base (80 centimeters The disposition of material indicated an
below the occupation floor) I found human intentional disturbance of the earliest burial,
bones, in a very bad state of preservation, associ- before depositing the final one. Bone remains
ated with a gray pottery bowl (Figure 5). Identi- were disposed next to the structures wall and
fiable bone fragments were two skull pieces and the ceramic jar was between the jaw and the
twelve teeth. Studies of tooth wear traces and skull.
roots allowed us to establish that this person was
an adult. Hundreds of unidentified bone frag- Archaeological evidence recovered from the
ments were recovered from the central area of burial structure of household compound LB1-
the burial. Within these maize (Zea mays) and U14 allows me to propose that this context was
cucurbit (Cucurbita sp.) phytoliths were identi- created by many events of digging, opening,
fied. These could be interpreted as the remains depositing, firing, closing, and burying, forming
of the deceaseds final meal, or as part of the a material palimpsest. Although palimpsests are
grave goods accompanying the dead. The ce- common archaeological contexts, they are
ramic bowl, buried horizontally, is a fine, un- typically viewed as a handicap.
decorated pottery vessel which could be
typologically dated to ca. A.D. 500 to A.D. 900.
301 - Research Reports
REFERENCES CITED
Miriam Dayde Aroz Silva: Aroz Delgado Consultores Supervisores Arquelogos Asociados
S.C.R.L, Urb. Larapa Grande, Manzana B-2, L4-2, San Jernimo, Cusco, Per
miriday@yahoo.com
Brian S. Bauer: Department of Anthropology (MC 027), Room 2102, 1007 West Harrison Street,
University of Illinois at Chicago, Chicago, Illinois 60607-7139
bsb@uic.edu
Mara Fernanda Boza: Department of Anthropology, 209 Maxwell Hall, Syracuse University,
Syracuse, NY 13244-1090
mbozacua@syr.edu
George S. Burr: National Science Foundation Accelerator Mass Spectrometer Laboratory, PAS
Building 81, University of Arizona, Tucson, Arizona 85721
burr@email.arizona.edu
Peter Eeckhout: Universit Libre de Bruxelles (CP175), 50 Av. F. Roosevelt, 1050, Brussels, Belgium
peeckhou@ulb.ac.be
Rgulo Franco J.: Proyecto Arqueolgico El Brujo, Magdalena de Cao, Fundacin Wiese,
Av. E. Canaval y Moreyra, Piso 16, Lima 27, Per
franco_brujero@yahoo.es
Csar Glvez M.: Proyecto Arqueolgico El Brujo, Magdalena de Cao, Direccin Regional de
Cultura, La Libertad-Ministerio de Cultura, Independencia 572, Trujillo, Per
cgmsepam@yahoo.es
Jaime Jimnez S.: c/o Proyecto Arqueolgico El Brujo, Magdalena de Cao, Fundacin Wiese, Av.
E. Canaval y Moreyra, Piso 16, Lima 27, Per
jaime_jimenez565@hotmail.com
Melissa S. Murphy: Department of Anthropology, Building 3431, University of Wyoming, 1000 East
University Avenue, Laramie, Wyoming 82070
mmurph20@uwyo.edu
Jeffrey Quilter: Peabody Museum of Anthropology and Ethnology, Harvard University, 11 Divinity
Ave. Cambridge, Massachusetts 02138
quilter@fas.harvard.edu
Hal Starratt: Anthropology, Western Nevada College, Carson City Campus, 2201 West College
Parkway, Carson City, Nevada 89703
starratt@wnc.edu
Howard I. Tsai: Museum of Anthropology, University of Michigan, Ann Arbor, Michigan 48109-
1079
hitsai@umich.edu
John Walker: Department of Anthropology, 4000 Central Florida Boulevard, Howard Phillips Hall
309, Orlando, Florida 32816-1361
john.walker@ucf.edu
Juliet Wiersema: Metropolitan Museum of Art, 1000 Fifth Avenue, New York, New York 10028
juleswiersema@yahoo.com
305 - Research Reports
1. Andean Past seeks original, unpublished articles on Andean archaeology and ethnohistory. Papers
submitted to Andean Past should not be under consideration elsewhere. Publication on the
Internet, including self-publication, will preclude subsequent publication in Andean Past. Ethno-
historical articles must lead to an enhanced understanding of prehispanic times. Well-illustrated
papers presenting new data or interpretations are particularly welcome, as are papers with an
environmental focus. Interim reports can be considered. We frequently publish papers on the
history of Andean archaeology. We also publish papers based on research conducted some time
ago. Papers are evaluated on a revolving basis. That is, we consider unsolicited papers as they
come to us, not according to strict or arbitrary deadlines.
2. Authors should include full contact data on the title page of their submissions.
3. If the Editors agree that a submission is ready for further consideration, it will be sent to a
minimum of two external reviewers, scholars with expertise in the subject matter of the paper
and who are not members of the Andean Past Editorial Board. After study of the external
reviews, papers are accepted or declined on the basis of a vote by the two senior Editors and the
Editorial Board members. Decisions are generally by simple majority.
4. We are flexible in the length of articles that we publish, as well as in the number of illustrations.
However, we do not publish book-length monographs.
5. Submissions may be in English or Spanish and may include short passages or quotations in other
languages, including South American Indian languages. Quotations not in English should be
translated into English with the original retained as a footnote. Submissions will be read and
evaluated in their original language(s). Reviews may be written in a language other than the one
in which the submission is composed. Articles written in Spanish that are accepted for
publication will be translated into English by Andean Past at no charge to the Authors. Special
cases, including articles in languages other than English or Spanish, may be discussed with the
Editors.
7. Prospective Authors are advised to examine the latest issue of Andean Past and use it as a model
for style. If Authors follow Andean Past house style closely it greatly reduces the Editors work
and improves their moods. When matters of style seem unclear, we urge writers to consult the
latest edition of the Chicago Manual of Style and/or the Editors.
8. For articles written originally in English, the on-line or latest CD-ROM edition of the Oxford
English Dictionary (OED) should be used as a guide to spelling and word usage. In cases where
the OED differentiates between English and American spelling or usage, Andean Past
contributors should follow American practice. Foreign loan words included in the OED should
not be italicized or glossed. Foreign words not in the OED should generally be italicized and
glossed. However, if a term-of-art does not appear in the OED but appears in another standard,
ANDEAN PAST 10 (2012) - 306
widely distributed dictionary such as The Penguin Dictionary of Archaeology or the Penguin
Dictionary of Architecture, it may be used in Andean Past without italicization or gloss. Original
dates for colonial accounts and spellings of authors names will be taken from Pillsburys Guide
to Documentary Sources for Andean Studies 1580-1900.
9. References may be given exactly as listed on the title pages of the work(s) cited. This style is
particularly appropriate for works published before the twentieth century. If Authors elect to
follow this convention, they should inform the Editors. In any case, please give the full names
of Authors, not first initials, in the References Cited if possible. In the case of multiple editions
of works, please cite the most scholarly editions available.
10. Intertextual references should include specific page numbers supporting the points being made.
Only if an entire article, book, or manuscript cited supports the point may page numbers be
omitted.
11. To make the reading of microform and digital copies easier, footnotes, not endnotes, should be
used.
12. Andean Past uses Goudy Old Style BT typeface. Please use this typeface in the text and in
illustrations if possible.
13. Illustrations should normally be in the form of black and white line drawings or grayscale
photographs. Hand lettering is unacceptable. All place names, terms for features, etc. must be
in English. Avoid sans serif typefaces in illustrations if possible. When sans serif typefaces are
required, please use Univers. Very small fonts will not reproduce well and some type faces
reproduce more clearly than others. Please keep in mind that article illustrations will not be
reproduced larger than six and a half by nine inches.
14. All site names and other place names mentioned in the text should appear on a map drafted
specifically for the submission, unless those toponyms appear in the Oxford Atlas of the World or
Google Earth. For example, it is not necessary to place Lima or Quito on a map except for
purposes of orientation. Toponyms not mentioned in the text should not appear on the map(s)
accompanying a submission unless they are major geographic features needed for orientation, for
example, cities and large towns. To avoid color management problems we do not publish color
plates of sites, artifacts, or landscapes. However, with the permission of the editors, maps and
plans may be submitted in color.
15. If illustrations have been previously published elsewhere, this should be indicated in the captions.
If the illustration is reproduced exactly as it first appeared the caption should read . . . from .
. . with a Harvard-style reference to the publication or publications. If it is modified the caption
should read . . . after . . If Authors do not hold the copyright to these illustrations they are
responsible for providing Andean Past with written permission(s) to reproduce them.
16. Best quality printed photographs usually originate as good-quality glossy chemical (silver) prints
made from black and white film negatives shot using a tripod or other camera support. Medium
307 - Research Reports
format is encouraged. Alternately, uncompressed electronic files, such as tif files, or good quality
digital prints, preferably C-prints, may be submitted. Authors should prepare electronic
illustration files with printing, not PowerPoint or web posting, in mind. Files should be sent at
the highest resolution possible with the camera or scanning equipment available. Do not try to
anticipate how we may downsize the files unless you have specific instructions from the Editors.
If Authors must use electronic formats other than tif files, including jpegs, they should discuss
this with the Editors before submission. Photographs taken with hand-held digital point-and-
shoot cameras on automatic settings rarely produce good prints.
17. Because layout is the prerogative of the Editors, not of the Authors, do not embed illustrations
in Word or other programs. Do not embed captions within the illustration files. Name files so
that they can be easily recognized, for example Barnes.Figure1.Profile.tif.
18. In Andean Past articles photographs come after the text and are not embedded within it.
Therefore, if Andean Past articles are scanned, the text and line can be scanned at a black and
white setting and the photographic illustrations scanned at a grayscale setting, then combined
in a single pdf to produce an optimal pdf for scholarly uses. Please note, though, that illustrations
are embedded in the Research Reports section.
19. Submissions should be sent as a hard (printed out) copy, preferably accompanied by electronic
copy on a CD or DVD, or on a USB 2 flash drive. Copies and CDs/DVDs will not be returned.
Flash drives will be returned upon request. We can accept a variety of word processing, graphic,
and spreadsheet programs. To avoid corruption and difficulties in transmitting long files, do not
send submissions as e-mail attachments or via large file sending services such as USendIt without
prior consultation with the Editors.
20. Authors who cite unpublished papers (except masters theses or doctoral dissertations, and
manuscripts in public archives such as the Archivo General de Indias, Seville) should provide
the Editors with copies of those papers, as well as written permission for citation from the
Authors of the unpublished papers.
21. Reports for the Current Research section of the journal are subject to editorial review only.
Reports should focus on unpublished results and should be brief, normally not more than 2,000
words. Reports may be accompanied by a limited number of line drawings and/or grayscale
photographs, normally no more than eight. These will be reproduced in a small format.
Individual researchers or teams may submit more than one report. Reports will be considered for
publication as they are submitted and, if accepted by the Editors, will appear in the next volume
of Andean Past. Authors will have the opportunity to read proof and make last-minute updates.
22. Obituaries are written at the invitation of the Editors, although people wishing to write an
obituary of a particular individual are welcome to consult with the Editors. Andean Past
obituaries normally carry a complete bibliography of works by and about the deceased unless such
a list has already been published elsewhere. It is the normally the responsibility of Obituary
Authors to construct such a bibliography. Obituaries are not peer-reviewed, although the Editors
may seek the advice of selected colleagues prior to publication.
ANDEAN PAST 10 (2012) - 308
24. All article and obituary Authors, including co-Authors, whose work is published in Andean Past
will receive, without charge, a copy of the volume in which their work appears. Current Research
authors will receive a pdf of the Current Research section containing their reports.
25. Andean Past has no provision for offprints. However, Authors may photocopy or scan their own
work for distribution to colleagues. They may also request PDF files of their articles for this
purpose.
26. Copyright for Andean Past is held by the Cornell University Latin American Studies Program on
behalf of the Editors unless a specific portion, for example, an illustration, is noted as copyrighted
by another party. Authors may re-publish their Andean Past articles, obituaries, or reports, in
English or in translation, in print, or in electronic format, provided that at least one year has
elapsed since the original publication in Andean Past as defined by the date of the Editors
Preface, that prior publication in Andean Past is indicated in the re-publication, that Cornell
LASPs copyright is acknowledged, and that the Editors of Andean Past are notified of the re-
publication. If a portion of an article is copyrighted by a third party Authors must request specific
written permission from that party to republish. This includes on-line postings in electronic
format.
18 March 2012